ISBN: 0-8247-0581-5

This book is printed on acid-free paper.

Headquarters
Marcel Dekker, Inc.
270 Madison Avenue, New York, NY 10016
tel: 212-696-9000; fax: 212-685-4540

Eastern Hemisphere Distribution

Marcel Dekker AG
Hutgasse 4, Postfach 812, CH-4001 Basel, Switzerland
tel: 41-61-261-8482; fax: 41-61-261-8896

World Wide Web

http://www.dekker.com

The publisher offers discounts on this book when ordered in bulk quantities. For more
information, write to Special Sales/Professional Marketing at the headquarters address
above.

Copyright © 2002 by Marcel Dekker, Inc. All Rights Reserved.

Neither this book nor any part may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopying, microfilming, and recording, or
by any information storage and retrieval system, without permission in writing from the
publisher.

Current printing (last digit):

10 9 8 7 6 5 4 3 2 1

PRINTED IN THE UNITED STATES OF AMERICA

Foreword

Foreword

Erestor, in Tolkien’s The Lord of the Rings,* declares that only two possibilities
exist for dealing with the menace of the Ring: “to hide it forever, or to unmake
it. But both are beyond our power.” This analogy was used by Amory Lovins in
an unpublished review of proposed U.S. policy on nuclear waste management.
The dilemma described by Erestor is analogous not only to the world’s nuclear
waste issues but to many other concerns related to long-life hazardous waste
being generated worldwide at an alarming rate. Once nuclear or hazardous waste
is produced, we cannot “unmake” it. Congress has responded to these concerns
through a number of legislative initiatives that attempt to minimize the amount
of waste containing hazardous constituents and also place restrictions on its
disposal in underground repositories. In the latter case, the most notable legisla-
tion is the Resource Conservation and Recovery Act (RCRA), requiring that the
best available technology be used to remove the chemical constituents in both
hazardous and mixed waste before it can be permanently disposed of in under-
ground repositories. Compliance with RCRA and other environmental regulations

*The council of Elrond from The Fellowship of the Ring: Lord of the Rings—Part One by J. R. R.
Tolkien, New York: Ballantine Books edn., pp. 349–50, 1986.

iii
iv Foreword

has come at a high cost in terms of dollars and, sometimes, potential risks to
worker health and safety.
It is with these issues in mind that the editor has compiled a comprehensive
textbook that covers the broad spectrum of pollution prevention including process
design, life cycle analysis, risk and decision analysis. The information presented
will increase awareness of the need to “do it right” the first time. The amount of
waste generated in any process results in a net reduction in profits. In every
process, the ultimate goal is to produce goods and materials that can be sold or
bartered for a profit. Looking into the future, it is ideal to proceed with industrial
development that maximizes the productions of goods and materials while
minimizing or eliminating the waste produced. This includes looking into raw
materials, and production efficiencies as well as process modification and en-
hancement that would result in the ultimate goal. This text encourages future
generations to develop the policies and priorities necessary to effectively deal
with scientific and political issues associated with hazardous and radioactive
waste management.
Jim Bickel
Former Assistant Manager
Projects and Energy Program
U.S. Department of Energy

Ron Bhada
Emeritus Director
WERC
Las Cruces, New Mexico
Preface

Preface

The most significant issues facing mankind today are related to the quality of our
environment. Past decisions did not always consider environmental factors as
critical elements. However, current decisions made daily should reflect the
importance of the environment. All environment-related issues are multidiscipli-
nary, ranging from science and engineering to social, economic, and regulatory
issues. Further, these issues are not related to any one region or country, but are
global in nature, requiring multidisciplinary, multiorganizational, and multina-
tional educational efforts.
This book provides an introduction and current information to the academic
community as well as to any professional who must deal with these issues on a
day-to-day basis. My aim is to have environmental issues become a major factor
in process design consideration. Our contributors present the fundamentals of
pollution prevention: life-cycle analysis, designs for the environment, and pollu-
tion prevention in process design. Selected case studies are provided as well.
All of the contributors to this volume are, in one way or another, associated
with WERC, a Consortium for Environmental Education and Technology Devel-
opment program.
The first part of the book deals with elements required for legal, organiza-
tional, and hierarchal components of pollution prevention. Parts II–IV deal with

v
vi Preface

the basics of pollution prevention, leading with fundamentals of pollution preven-

tion, followed by methodology and life cycle cost analysis. Part V deals with risk
and decision fundamentals, as well as, utilization of pollution prevention in
process design. Part VI presents selected case studies in various fields.
As the editor, I realize that I have just begun to scratch the surface with
some of the recent advances. I would like to take this opportunity to thank the
chapter authors for their contributions to this volume. Further gratitude goes to
Dr. Franc Szidarovszky, University of Arizona, and Kay Perkins, WERC, for their
editing assistance.

Abbas Ghassemi
Contents

Contents

Foreword Jim Bickel and Ron Bhada iii

Preface v

Contributors xi

Acronyms xv

Glossary xvii

Part I Legal/Organizational/Hierarchal Requirements

1 Pollution Prevention and Waste Minimization—Back to Basics 1

Jeff Weinrach
2 Role of Pollution Prevention in Waste
Management/Environmental Restoration 9
Harish Chandra Sharma

vii
viii Contents

3 The Waste Management Hierarchy 35

W. David Constant
4 Legislative and Regulatory Issues 43
Toni K. Ristau
5 Information Systems for Proactive Environmental
Management 57
Steven P. Frysinger
6 European Policies for Waste Management 85
Ingo F. W. Romey and Marc Obladen

Part II Fundamentals

7 Energy Conservation 99
K. A. Strevett, C. Evenson, and L. Wolf
8 Fundamentals of Heat Transfer 137
René Reyes Mazzoco
9 Macroscopic Balance Equations 147
Paul K. Andersen and Sarah W. Harcum
10 Biotechnology Principles 171
Teresa J. Cutright

Part III Methodology

11 Novel Materials and Processes for Pollution Control in the

Mining Industry 215
Alan Fuchs and Shuo Peng
12 Monitoring In-Situ Electrochemical Sensors 225
Joseph Wang
13 Using Roadmaps in Pollution Prevention:
The Los Alamos Model 233
Thomas P. Starke and James H. Scott
14 Pollution Prevention and DFE 251
Terrence J. McManus

Part IV Life Cycle

15 Pollution Prevention and Life Cycle Assessment 269

Mary Ann Curran and Rita C. Schenck
Contents ix

16 Application of Life Cycle Assessment 321

W. David Constant

Part V Risk and Decision

17 Risk-Based Pollution Control and Waste Minimization

Concepts 327
Gilbert J. Gonzales
18 Elements of Multicriteria Decision Making 351
Abdollah Eskandari, Ferenc Szidarovszky, and Abbas Ghassemi
19 Environmental Considerations and Computer Process Design 389
Victor R. Vasquez

Part VI Case Studies

20 Minimization and Use of Coal Combustion By-Products (CCBs):

Concepts and Applications 421
Harold W. Walker, Panuwat Taerakul, Tarunjit Singh Butalia,
William E. Wolfe, and Warren A. Dick
21 Engineered Wetlands for Metal Mining-Impacted
Water Treatment 449
Herold J. Gerbrandt
22 Fuel Blends and Alkali Diagnostics: European Case Study 463
Ingo F. W. Romey
23 Best Practices for the Oil and Gas Exploration, Production,
and Pipeline Transportation Industry 497
Bart Sims

Index 505
Contributors

Contributors

Paul K. Andersen, Ph.D. Associate Professor, Department of Chemical Engi-

neering, New Mexico State University, Las Cruces, New Mexico
Tarunjit Singh Butalia, Ph.D. Research Scientist, Department of Civil and
Environmental Engineering and Geodetic Science, The Ohio State University,
Columbus, Ohio
W. David Constant, Ph.D., P.E. Professor of Civil and Environmental Engi-
neering and Assistant Director of U.S. Environmental Protection Agency Hazard-
ous Substance Research Center/South and Southwest, Department of Civil and
Environmental Engineering, Louisiana State University and A&M College, Baton
Rouge, Louisiana
Mary Ann Curran, M.S. Research Chemical Engineer, Office of Research &
Development, U.S. Environmental Protection Agency, Cincinnati, Ohio
Teresa J. Cutright, Ph.D. Assistant Professor, Department of Civil Engineer-
ing, The University of Akron, Akron, Ohio
Warren A. Dick, Ph.D. Professor of Soil Science, School of Natural Resources,
The Ohio State University, Wooster, Ohio

xi
xii Contributors

Abdollah Eskandari, Ph.D. Postdoctoral Researcher, Department of Systems

and Industrial Engineering, University of Arizona, Tucson, Arizona
C. Evenson Graduate Degree Candidate, BioEnvironmental Engineering and
Science Laboratory, University of Oklahoma, Norman, Oklahoma
Steven P. Frysinger, Ph.D. Professor, Department of Integrated Science and
Technology, James Madison University, Harrisonburg, Virginia
Alan Fuchs, Ph.D. Assistant Professor, Department of Chemical Engineering,
University of Nevada, Reno, Reno, Nevada
Herold J. Gerbrandt, P.E., Ph.D. Professor, Department of General Engineer-
ing, Montana Tech of the University of Montana, Butte, Montana
Abbas Ghassemi, Ph.D. Executive Director, WERC, New Mexico State Uni-
vesity, Las Cruces, New Mexico
Gilbert J. Gonzales, Ph.D. Ecologist, Ecology Group, Los Alamos National
Laboratory, Los Alamos, and Adjunct Professor, Department of Fishery and
Wildlife Sciences, New Mexico State University, Las Cruces, New Mexico
Sarah W. Harcum, Ph.D. Associate Professor, Department of Chemical Engi-
neering, New Mexico State University, Las Cruces, New Mexico
René Reyes Mazzoco, Ph.D. Professor, Department of Chemical Food and
Engineering, Universidad de las Américas–Puebla, Cholula, Puebla, Mexico
Terrence J. McManus, P.E., D.E.E. Intel Fellow and Director, EHS Technol-
ogies, Corporate Environmental Health and Safety, Intel Corporation, Chandler,
Arizona
Marc Obladen, Ph.D. Scientific Co-Worker, Institute of Technology of Energy
Supply Systems and Energy Conversion Plants, and Professor, University of
Essen, Essen, Germany
Shuo Peng, Ph.D. Research Assistant, Department of Chemical Engineering,
University of Nevada, Reno, Reno, Nevada
Toni K. Ristau, M.S., J.D. Director of Environmental Services, Department of
Power Production and Energy Services, Public Service Company of New Mexico,
Albuquerque, New Mexico
Ingo F. W. Romey, Ph.D. Head of Chair, Institute of Technology of Energy
Supply Systems and Energy Conversion Plants, University of Essen, Essen,
Germany
Rita C. Schenck, Ph.D. Executive Director, Institute for Environmental Re-
search and Education, Vashon, Washington
Contributors xiii

James H. Scott, Ph.D. President, Abaxial Technologies, Los Alamos, New

Mexico
Harish Chandra Sharma, M.S., M.B.A, REM. Site Liaison and Project
Manager, U.S. Department of Energy, Albuquerque, New Mexico
Bart Sims, B.S. Manager, Hazardous Waste and Waste Minimization, Oil and
Gas Division, Environmental Services Section, Railroad Commission of Texas,
Austin, Texas
Thomas P. Starke, Ph.D. Program Manager, Environmental Stewardship Of-
fice, Los Alamos National Laboratory, Los Alamos, New Mexico
K. A. Strevett, Ph.D. Professor of Environmental Engineering, BioEnviron-
mental Engineering and Science Laboratory, University of Oklahoma, Norman,
Oklahoma
Ferenc Szidarovszky, Ph.D. Professor, Department of Systems and Industrial
Engineering, University of Arizona, Tucson, Arizona
Panuwat Taerakul Department of Civil and Environmental Engineering and
Geodetic Science, The Ohio State University, Columbus, Ohio
Victor R. Vasquez, Ph.D. Assistant Professor, Department of Chemical Engi-
neering, University of Nevada, Reno, Reno, Nevada
Harold W. Walker, Ph.D. Assistant Professor, Department of Civil and Envi-
ronmental Engineering and Geodetic Science, The Ohio State University, Colum-
bus, Ohio
Joseph Wang, Ph.D. Professor, Department of Chemistry and Biochemistry,
New Mexico State University, Las Cruces, New Mexico
Jeff Weinrach, Ph.D. Vice President and Director of Quality Standards,
JCS/Novation, Inc., Albuquerque, New Mexico
L. Wolf Graduate Degree Candidate, BioEnvironmental Engineering and Sci-
ence Laboratory, University of Oklahoma, Norman, Oklahoma
William E. Wolfe, Ph.D. Professor, Department of Civil and Environmental
Engineering and Geodetic Science, The Ohio State University, Columbus, Ohio
Acronyms

Acronyms

ACAA American Coal Ash E&P Exploration and production

Association E2 Energy efficiency
AMD Acid mine drainage EA Environmental assessment
ASTM American Society for Testing EDDS Environmental Decision
and Materials Support System(s)
C&D Construction and demolition EMIS Environmental Management
debris Information System(s)
CCA Clean Air Act EMS Environmental Management
CCB Coal combustion by-product System(s)
CDF Code of Federal Regulation EPCRA Emergency Planning and
CERCLA Comprehensive Environmen- Community Right-to-
tal Response, Compen- Know Act
sation, and Liability ER Environmental restoration
Act ERP Enterprise resource
CLSM Controlled low-strength planning
material ESP Electrostatic precipitation
D&D Decontamination and FBC Fluidized bed combustion
decommissioning FGD Flue gas desulfurization
DfE Design for environment FHWA Federal Highway
DQO Data quality objective Administration

xv
xvi Acronyms

GIS Geographic Information PPE Personal protective

System(s) equipment
IADC International Association of PPOA Pollution Prevention
Drilling Contractors Opportunity
IEIS Integrated Environmental Assessment
Information System(s) Psi Pounds per square inch
ISO 14001 International Standards PSM Pozzolanic-stabilized
Organization’s stand- Mixture
ard for environmental RCRA Resource Conservation and
management systems Recovery Act
Lb/mm BTU Pound per million British RI/FS Remedial investigation/
thermal unit feasibility study
LFA Lime-fly ash-aggregate ROI Return on investment
LIMB Lime injection multistage RRC Railroad Commission of
burner Texas
LOI Loss on ignition SI Site investigation
LQG Large quantity generator SO2 Sulfur dioxide
MSW Municipal solid waste SPE Society of Petroleum
NAAQS Natural Ambient Air Quality Engineers
Standards SR State route
NOx Nitrogen oxide TCLP Toxicity of characteristic
NOV Notice of violation leaching procedure
NPL National Priorities List SWMU Solid waste management
NSPS New Source Performance unit
Standards TRI Toxic Release Inventory
OCDO Ohio Coal Development US DOE U.S. Department of Energy
Office US EPA U.S. Environmental
P2 Pollution prevention Protection Agency
PA Preliminary assessment WM Waste management
PC Pulverized coal Wmin Waste minimization
POTW Publicly-owned treatment
works
Glossary

Glossary

Anode The positive element of any electrical device from which electricity
flows.
Cathodic protection The negative element that draws current away from pipe
and other metal equipment to protect the pipe or equipment from corrosion.
Centrifugal filter A device that spins a fluid at high speed to separate and
remove materials from the fluid.
Closed-Loop drilling fluid system A system of tanks that contains the drilling
fluid used in drilling an oil or gas well so that drilling fluid is not placed in
conventional pits.
Coalescor panel A device used in fluid separation equipment that stabilizes
fluid flow through the device (e.g., reduced turbulence).
Construction and demolition debris (C&D) Waste building materials, pack-
aging, and rubble resulting from construction, remodeling, repair, and demoli-
tion operations on pavement, houses, commercial buildings, plants, and other
structures.

xvii
xviii Glossary

Data quality objective (DQO) Qualitative and quantitative statements derived

from the DQO process that clarify study objectives, define the appropriate type
of data, and specify the tolerable levels of potential decision errors that will be
used as the basis for establishing the quality and quantity of data needed to
support decisions. It provides a systematic procedure for defining the criteria that
a data collection design should satisfy, including when and where to collect
samples, the tolerable level of decision errors for the study, and how many
samples to collect.
Decontamination and decommissioning (D&D) The process of reducing or
eliminating and removing from operation of the process harmful substances, such
as infectious agents, so as to reduce the likelihood of disease transmission from
those substances. After the D&D operation, the process is no longer usable.
Demolition The wrecking or taking out of any load supporting structural
member and any related razing, removing, or stripping of a structure. Also called
Deconstruction.
Design for environment (DfE) It is the systematic consideration of pollution
prevention/waste minimization options during the design consideration of any
process associated with environmental safety and health over the product life
cycle.
Drilling fluid The circulating fluid used in drilling oil and gas wells. Drilling
fluid lubricates the drill bit, carries rock cuttings from the wellbore to the surface,
and controls subsurface formation pressures.
Drilling rig The collection of equipment, such as a derrick, used to drill oil and
gas wells.
Enhanced oil recovery Methods applied to oil and gas reservoirs depleted by
primary production to make them productive once again.
Environmental assessment (EA) Document that briefly provides sufficient
evidence and analysis for determining whether to prepare an environmental
impact statement or a finding of no significant impact. Includes a brief discussion
of the need for the proposal, alternatives as required by EPA regulations, the
environmental impacts of the proposed action and alternatives, and a listing of
agencies and persons consulted.
Flowline The surface pipe through which oil travels from a well to processing
equipment or storage.
Functional unit The measure of a life-cycle system used to base reference
flows in order to calculate inputs and outputs of the system.
Glossary xix

Hazardous waste Solid waste that is hazardous as defined in Title 40 of the

Code of Federal Regulations (CFR). Hazardous waste is either specifically listed
as such or exhibits a characteristic of hazardous waste as specified in 40 CFR
Part 261, Subparts D and C.
Heat-Medium oil Oil that is used to transfer heat from one medium to another
medium.
ISO International Standards Organization (or International Organization of
Standardization).
ISO 14000 International Standardization of Environmental Management Sys-
tem Standard which is “that part of the overall management system which
includes organizational structure, planning activities, responsibilities, practices,
procedures, processes and resources for developing, implementing, achieving,
reviewing and maintaining the environmental policy.”
Large quantity generator (LQG) A hazardous waste generation site classifi-
cation. In general, LQG sites generate more than 2204 pounds of hazardous waste
each month.
Life (a) Economic: period of time after which a product, machine, or facility should
be discarded because of its excessive costs due to costs or reduced profitability;
(b) Physical: that period of time after which a product, machine, or facility can no
longer be repaired in order to perform its designed function properly.
Life cycle cost Evaluation of the environmental effects associated with any
given activity from the initial gathering of raw materials from the earth to the
point at which all materials are returned to the earth; this evaluation includes all
releases to the air, water, and soil.
Life cycle impact assessment A scientifically based process or model that
characterizes projected environmental and human health impacts based on the
results of the life cycle inventory.
Life cycle inventory An objective, data-based process of quantifying energy
and raw material requirements, air emissions, waterborne effluents, solid waste,
and other environmental releases throughout the life cycle of a project, process,
or activity.
Municipal solid waste (MSW) Residential and commercial solid wastes gen-
erated within a community.
Natural gas processing plant A facility containing equipment and vessels
necessary to purify natural gas and to recover natural gas liquids, such as butane
and propane.
xx Glossary

Paraffin A heavy, wax-like hydrocarbon commonly found in produced crude

oil. Paraffin often accumulates within wells and the associated surface equipment.

pH A unit of measure of the acid or alkaline condition of a substance. On a

logarithmic scale between 1 and 14, a neutral solution has a pH of 7, acid
solutions are less than 7, and alkaline solutions are greater than 7.

Pipeline compressor station An equipment station on a natural gas pipeline

that uses a device to raise the pressure of the gas in order to move it along the
pipeline.

Pollution prevention The use of materials, processes, or practices that reduce

or eliminate the creation of pollutants or wastes at the source.

Pollution Prevention Opportunity Assessment (PPOA) The systematic pro-

cess of identifying areas, processes, and activities that generate excessive waste
streams or waste by-products for the purpose of substitution, alteration, or
elimination of the waste.

Publicly owned treatment works Any device or system used to treat (includ-
ing recycling and reclamation) municipal sewage or industrial wastes of a liquid
nature and is owned by a State, municipality, intermunicipality, or interstate
agency (defined by Section 502(4) of the Clean Water Act).

Pump-jack A surface unit that imparts a reciprocating motion to a string of rods

that operate a pump in an oil well.

Reagent A substance that, because of the chemical reactions it causes, is used

in analysis and synthesis.

Recycling of materials The use or reuse of a waste as an effective substitute

for a commercial product, as an ingredient, or as feedstock in an industrial or
energy producing process; the reclamation of useful constituent fractions within
a waste material; or removal of contaminants from a waste to allow it to be reused.
This includes recovery for recycling, including composting.

Reserve pit The pit in which a supply of drilling fluid is stored for use in
drilling an oil or gas well. A reserve pit is typically an excavated, earthen-walled
pit, which may be lined to prevent contamination of soil and water.

Return on Investment (ROI) The calculation of time within which the process
would save the initial investment amount if the suggested changes were incorpo-
rated into it. In this calculation, depreciation, project cost, and the useful life are
taken into account.
Glossary xxi

Rod-pump The pump in an oil well that lifts oil to the surface as a result of the
reciprocating action of a rod-string (see also Pump-Jack).
Sand-blasting media Abrasive material used to remove paint and other coating
material from metal surfaces. Sand-blasting media is forced onto the surface using
high air pressure.
Screen-Type filter A filter unit, typically constructed of steel, from which the
screen portion can be removed, cleaned of filtrate, and reused.
Separator A cylindrical or spherical vessel used to isolate the components in
streams of mixed fluids.
Soda ash Sodium carbonate, typically used for pH control.
Sour gas Natural gas that contains hydrogen sulfide or another sulfur com-
pound.
Specific gravity The ratio of the weight of a given volume of a substance at a
given temperature to the weight of an equal volume of a standard substance at
the same temperature.
Strata Distinct beds of rock, which are usually parallel. An individual bed of
rock is a stratum in which the subsurface oil and natural gas are contained within
certain strata.
Sulfur dioxide scrubber A device designed to remove sulfur dioxide from the
exhaust gases of engines.
Thermal destruction Destroying of waste (generally hazardous) in a device
that uses elevated temperatures as the primary means to change the chemical,
physical, or biological character or composition of the waste. Examples of the
processes could be incineration, calcinations, oxidation, and microwave dis-
charge. Commonly used for medical waste.
Toxic release inventory (TRI) Required by EPCRA, it contains information on
approximately 600 listed toxic chemicals that facilities release directly to air,
water, or land or transport off-site.
Turn-key contract For the purpose of drilling an oil or gas well, a contract that
calls for the payment of a stipulated amount to the drilling contractor on
completion of the well. A turn-key contract may be based on a set cost per foot
of well drilled.
Vitrification The process of immobilizing waste that produces a glass-like solid
that permanently captures the radioactive materials.
xxii Glossary

Waste combustion Combustion of waste through elevated temperature and

disposal of the residue so generated in the process. It also may include recovery
of heat for use.
Waste management (WM) Activities associated with the disposition of waste
products after they have been generated, as well as actions to minimize the
production of wastes. This may include storage, treatment, and disposal.
1
Pollution Prevention and Waste
Minimization—Back to Basics

Jeff Weinrach
JCS/Novation, Inc., Albuquerque, New Mexico

1 TERMINOLOGY

Prevention, n. (prevent, v.—to keep from occurring; avert; hinder)

Minimization, n. (minimize, v.—to reduce to the smallest possible
amount or degree)
Source, n. any thing or place from which something comes, arises, or is
obtained; origin
Reduction, n. (reduce, v.—to bring down to a smaller extent, size,
amount, number, etc.)
Recycle, n. to treat or process (used or waste materials) so as to make
suitable for reuse
⋅⋅
⋅
Control, n. (control, v.—to exercise restraint or direction over; dominate;
command: to hold in check; curb; to eliminate or prevent the flourishing
or spread of)
Management, n. (manage, v.—to take charge or care of: to handle, direct,
govern, or control in action or use)

1
2 Weinrach

Treatment, n. (treat, v.—to subject to some agent or action in order to

bring about a particular result)
⋅⋅
⋅
Waste, n. (waste, v.—to consume, spend, or employ uselessly or without
adequate return; use to no avail or profit; squander: useless consumption
or expenditure; use without adequate return; an act or instance of
wasting: anything unused, unproductive, or not properly utilized: any-
thing left over or superfluous, as excess material or by-products, not of
use for the work in hand)
⋅⋅
⋅
Pollution, n. the introduction of harmful substances or products into the
environment
Pollute, v. to make foul or unclean, especially with harmful chemical or
waste products
⋅⋅
⋅
Process, n. A systematic series of actions directed to some end
⋅⋅
⋅
System, v. An assemblage or combination of things or parts forming a
complex or unitary whole

Source: The Random House Dictionary of the English Language, 2nd

edition, Unabridged. New York: Random House, 1987.

2 BACKGROUND
We are a society rooted in language: both the words and their meanings. We
choose our words carefully in an attempt to convey the particular meaning that
we have in mind. Often, over time, words will develop a “life” of their own,
and their meanings (as well as their usage) may become murky and indistinct. In
the environmental arena, few words or groups of words have gone through the
constant deluge of interpretation and meaning as “pollution prevention.” Over the
last decade, pollution prevention has gone through an evolutionary process from
specific activities such as solvent substitution (CFCs versus replacement solvents,
as an example) to more systematic approaches that often provide the core element
to cost-effective environmental management systems (EMS).
Throughout this evolutionary process, due primarily to the regulatory
oversight that has placed administrative boundaries on what pollution preven-
tion is and what it is not, the practical issue has not always been the focus
of attention:
Pollution Prevention and Waste Minimization 3

What are the most cost-effective methods to reduce or eliminate envi-

ronmental impacts without sacrificing health, safety, or other related
concerns?
With the advent of EMS and environmental standards such as the ISO
14000 series, pollution prevention is now more often than not viewed as part of
a systematic approach to environmental improvement that includes planning,
information management, and process management. The last two entries in the
glossary above, “process” and “system,” reflect the conditions under which
pollution prevention is now most practically used.
Many organizations have embraced pollution prevention and have already
addressed or are in the midst of addressing the “low-hanging fruit,” the relatively
easy pollution-prevention activities that often do not require a thorough under-
standing of the processes that are generating the pollutants. A number of articles
in the literature describe success using nonhazardous solvents, for example, for
particular applications such as paint stripping or cleaning parts. As the number of
case studies grows and the nonhazardous solvents are shown to be successful for
particular processes or types of processes, the trust in using these solvents in
similar applications naturally grows as well. But with organizations now looking
for more innovative and cost-effective solutions to reducing waste and ineffi-
ciency, the application of environmental management systems where processes
and systems are key to identifying and implementing opportunities to reduce or
eliminate the waste and the inefficiency is becoming more commonplace.
Nevertheless, it is still helpful to have an understanding of what pollution
prevention is and how it can be applied to reduce or eliminate environmental impacts.
This book will provide many examples of pollution prevention as it relates to
particular processes or industries. The remainder of this chapter will provide an
overview of the various aspects to pollution prevention, with an emphasis on getting
beyond the environmental vernacular and focusing on the practical.
In a simplified way, we can think of pollution prevention in two different
contexts: we can prevent pollutants from being generated in the first place, or we
can prevent the pollutants from being introduced into the environment. In the first
case, if the pollutants are considered waste (which is most often the case), then
preventing these materials from being generated would result in a reduction,
minimization, or elimination of the waste (waste minimization). This usually
provides direct economic benefit, since reducing waste usually coincides with
increased efficiency, productivity, and profitability. In the second case, the
pollutants are still being generated but are not being released to the environment.
This should provide some environmental benefits, since the environment is not
being negatively impacted by these materials. However, the economic benefit
may not be as great, since the wastes need to be stored, treated, or disposed, which
4 Weinrach

is an added cost. Also, the management of these wastes is likely to be a temporary

measure and ultimately there will be some release to the environment unless the
wastes can be reused or recycled. This second case is often referred to as pollution
control. This book will focus on the pollution prevention practices that primarily
involve preventing pollutants and wastes from being generated in the first place.
Pollution prevention is often viewed as part of an overall environmental
management hierarchical framework (in order of preference): source reduction,
recycling, treatment, disposal. This hierarchy typically reflects the degree of
economic benefit and environmental protection that can be realized through these
efforts. However, the hierarchy is not always practical in terms of prioritizing
opportunities. For a particular process, recycling may be a more economically and
environmentally viable option to source reduction given the existing techniques
and approaches. By developing an effective environmental management system,
identifying the most cost-effective options to reducing waste and preventing
pollution will be much more likely. Even so, the hierarchy has been shown to be
quite effective in prioritizing pollution prevention and waste management pro-
jects and is still useful as a first attempt to improve environmental performance.

3 SOURCE REDUCTION
Source reduction involves the use of processes, practices, or products to reduce
or eliminate the generation and/or the toxicity of pollutants and wastes. Source
reduction includes, but is not limited to, material substitution, process substitu-
tion, and process elimination. Examples of some source reduction applications
are described below.

3.1 Material Substitution

Materials that will result in less toxic wastes can be substituted for materials that
are currently being used. Examples include the following:

Shifting from solvent-based paints to water-based paints reduces the toxic-

ity of paint wastes.
Shifting from chlorofluorocarbons (CFCs) to nonhazardous solvents will
reduce waste management costs and health risks.

3.2 Process Substitution or Elimination

Process that result in less waste and increased efficiency can be substituted for
processes that are currently being used. Also, entire processes can be eliminated
if pollution prevention is implemented effectively. Examples of some process
substitution or elimination opportunities are described below:
Pollution Prevention and Waste Minimization 5

Replacing traditional parts-cleaning processes using solvents with pro-

cesses that use supercritical fluids
Using Dry Ice pellets or other blasting techniques to remove paint, in lieu
of solvents.

3.3 Good Housekeeping and Equipment Maintenance

Good housekeeping and equipment maintenance are two environmental manage-
ment practices that are often low-cost/high-benefit approaches to pollution pre-
vention. A common example of good housekeeping practices involves the use of
drip pans to catch leaks or drips from equipment. Equipment maintenance is
important for two distinctly different reasons: (a) routine maintenance will reduce
the occurrence of leaks and drips, and (b) routine maintenance will extend the
lifetime of the equipment. When thinking about pollution prevention and waste
minimization, it is important to consider that when equipment comes to the end
of its useful life, it also becomes a waste!

3.4 Water and Energy (Resource) Conservation

Water conservation is critically important in all of our industrial and personal
activities. Non-point-source pollution (caused by water moving over and through
the ground picking up man-made and natural contaminants) can be significantly
reduced by limiting water usage. Also, the cost of treating wastewater is often
related to the volume of water that requires treatment.
Energy conservation and pollution prevention are often thought of as two
sides of the same coin. Waste management is typically an energy-intensive step
that, naturally, provides an additional incentive to pollution prevention and waste
minimization. When we use life-cycle analysis (see below), we identify additional
costs and wasteful steps associated with the energy consumed as part of managing
wastes. Also, energy production usually coincides with particular waste streams
and pollutants entering the environment. This cyclic interdependency between
energy efficiency and pollution prevention signifies the importance of using
systematic approaches to achieve significant environmental improvement.
Resource conservation, in general, is a fitting complement to pollution preven-
tion. As critical components to an integrated environmental management system,
best practices such as pollution prevention, water conservation, and energy
conservation can be more effective in tandem than as separate activities.

3.5 Pollution Prevention in Design and Planning

Designing or planning for a new process or operation is the best time to address
pollution prevention considerations. With an existing process, implementing
pollution prevention can require some possible downtime due to either equipment
6 Weinrach

reengineering or technician training. This will add greatly to the cost and,
therefore, reduce the economic benefit of the particular pollution prevention
approach. Also, in the design phase, all environmental improvement options are
open for evaluation. There are no practical reasons to dismiss any particular
option due to inability to transition from the current process to the improved
process. In the design and planning phase, there is no status quo and, therefore,
no downtime and associated costs.

3.6 Training and Awareness

Training and awareness programs are critical to ensuring that pollution prevention
is realized to its fullest potential. The best ideas will come from people who work
with machines, use materials, and generate waste. These people must be aware
that often there are alternatives and that they constantly need to be thinking about
ways to improve operations, efficiency, etc. It is always more effective to provide
pollution prevention training to people with process knowledge (often, the
implementers and stakeholders) than to provide “pollution prevention experts”
with process knowledge to develop a pollution prevention plan.

3.7 Life-Cycle Analysis

Pollution prevention often utilizes a principle known as “life-cycle analysis” to
address all associated costs and possible solutions associated with a particular
pollutant or waste. Life-cycle analysis, sometimes referred to as “cradle-to-grave”
analysis, is often used to track a particular material from its inception to its
ultimate demise. This tracking usually requires documentation from other com-
panies (both vendors and customers) in the material chain. In material substitu-
tion, for example, a possible material alternative that would drastically reduce a
particular waste stream may require a process change by the vendor first. Also, a
positive pollution prevention approach implemented by a particular company
could have negative impacts to its customers or contractors. For these reasons, it
is helpful to include vendors, customers, and contractors as part of the pollution
prevention team.

3.8 Inventory Control

Inventory control addresses the effective use of data and information to track the
procurement, use, and management of materials throughout the operation. Inven-
tory control practices include the following:
“Just-in-time” procurement—purchase only what is needed, in the amounts
needed. This is extremely important for chemicals or materials that have
relatively short shelf-lives and have to be disposed if not used in a timely
manner.
Pollution Prevention and Waste Minimization 7

Affirmative procurement—purchase only materials that have been or can

be recycled. Purchase nonhazardous chemicals and materials whenever
possible.
Barcoding—use barcodes to track material usage throughout the facility.
This is extremely helpful in limiting the amount of material purchased if
it is known how much of that material may already be stored at the
facility. Through a chemical or material exchange program, chemicals
and materials can be obtained from operations within the facility instead
of having to purchase the material.

4 RECYCLING
For the purpose of this book, recycling is addressed in two different fashions
whenever possible: (a) in-process recycling (recycling materials), and (b) end-of-
pipe recycling (recycling wastes).
In-process recycling implies that a material is recycled before it becomes
a waste. If the material is not being treated as a waste, then waste manage-
ment regulatory requirements are not applicable to these processes (no treat-
ment permit required, for example) because the recycling is in-process.
The development of these recycling activities requires knowledge of the pro-
cess itself.
End-of-pipe recycling implies that the material being recycled has already
become a waste. In many cases, waste management regulatory requirements are
applicable to these recycling processes. Because the recycling is end-of-pipe,
knowledge of the process that generated the waste is normally not necessary.
End-of-pipe recycling as a pollution prevention alternative does not, therefore,
depend on the processes that generated the waste.

5 TREATMENT (INCLUDING WASTE SEGREGATION)

Waste treatment is usually the third option after source reduction and recycling
opportunities have been exhausted. Treatment includes techniques such as precip-
itation, neutralization, stabilization, and incineration. Waste segregation is also
considered as a treatment alternative. In many cases, waste treatment is per-
formed off-site by a contracting organization. The waste-generating organization
must maintain very careful records regarding the contents of the waste so the
proper waste management procedures can be carried out. In many cases, infor-
mation regarding the process that generated the waste is maintained with the
waste information. This information is helpful in demonstrating an understanding
of how (and why) the waste was generated, and it lessens the risk to the
contracting organization that may be treating wastes it would otherwise not be
permitted to treat.
8 Weinrach

Waste segregation is an environmental best management practice de-

signed to reduce costs through storing incompatible waste separately, including
separating hazardous from nonhazardous wastes, or regulated from nonregulated
wastes. In many circumstances, mixing regulated with nonregulated wastes
renders the entire waste contents regulated and unnecessarily increases waste
management costs.

6 DISPOSAL
If there are no other practical options, disposal needs to be carried out in
an environmentally responsible manner. In the majority of cases, waste disposal
will be provided by a contractor. It is critically important that proper documenta-
tion and records are maintained regarding waste disposal, both by the parent
company and by the contractor. In many regulatory environments, for example,
liability for the disposal of waste is not totally eliminated after the waste is
removed from the site.

7 CONCLUSION
Waste minimization and pollution prevention are two components of a broader,
effective system of process improvements that often have both environmental
and economic benefit. As technologies continue to be developed and as new,
innovative approaches to improving efficiency and productivity are im-
plemented, these and other environmental best practices will likely be drawn in
to the overall operational improvement and excellence that we strive for. If we
can effect significant improvement through these types of approaches, our com-
panies will be more productive, more profitable, and more competitive in the
global marketplace.
2
Role of Pollution Prevention in Waste
Management/Environmental Restoration

Harish Chandra Sharma

U.S. Department of Energy, Albuquerque, New Mexico

1 POLLUTION PREVENTION IN PROCESS MANAGEMENT

The single most important challenge facing today’s environmental engineer of
any major industry is how to keep environmentalist organizations, regulators, and
stakeholders on one side and business managers and shareholders on the other
side satisfied simultaneously. Complying with existing regulations is a simple
matter of incorporating a few additional process changes. Unfortunately, changes
of this kind are costly. In the process, waste changes from one form to another.
The only way out of this complex situation is through incorporation of waste
minimization/pollution prevention (WMin/P2) measures, every step of the way,
wherever economically feasible. As per a National Association of Manufacturers’
survey, companies that take advantage of the U.S. Environmental Protection
Agency (EPA) 33–50 emissions reduction program for toxic chemicals, over half
the companies saved money (1). In spite of this encouraging fact, resistance to
incorporate waste minimization or pollution prevention at any level is very real.
Many industrial houses today have placed a vice president in charge of environ-
mental affairs. These companies have a vision and mission statement to go with
the position, but when it comes to changing a process or incorporating a

9
10 Sharma

significant change, it just won’t happen. This is true for an existing process in
which waste is managed as an end-of-pipe fact under the waste management
(WM) program and also in the case of site cleanup that is conducted under
the environmental restoration (ER) process or under the decontamina-
tion/decommissioning (D&D) process for an abandoned facility.

1.1 Waste Management (WM)

Waste management and minimization program is “compliant, cost effective
management of the minimum waste” (2). Waste does not contribute added value
to the products in the marketplace but rather is an enduring liability to the
company. It is a liability when it is produced, due to the cost of transportation and
disposal, and in the future, as the landfill where it has been disposed may leak,
thus harming the environment and the community. As a result, cost effectiveness
is the primary focus of commercial waste management programs.
Waste management covers newly generated waste or waste from an on-
going process. When steps to reduce or even eliminate waste are to be considered,
it is imperative that considerations should include total programmatic oversight,
technical, and management services of the total process. From raw material to the
final product, this includes technical project management expertise (for cost and
technical effectiveness), technical project review and pollution prevention tech-
nical support and advocacy. Waste management also includes handling of waste,
including treatment, storage, and disposal.
In WM, the waste, both the quantity and composition, are known. Given
these facts, disposal can be handled in a planned manner. In most cases, other
than the waste from routine process, waste that is generated from handling of
known waste can be controlled and virtually eliminated.
Minimizing of waste in WM programs is primarily due to the opportunity
it provides not only to reduce production costs but also to reduce liability at the
same time. Liability exists at the point of generation and at the off-site disposition
of that waste. As a result, “waste minimization is not a stand alone program but
an element in a corporate liability reduction initiative” (2).

1.2 Environmental Restoration (ER)

In the case of environmental restoration (ER) projects, the quantity and compo-
sition of the waste itself is an estimate, as the waste was usually “dumped” at the
site some time ago. In most cases, these sites were “discovered” during the 1980s,
when the EPA required them to be identified. Under the Resource Conservation
and Recovery Act (RCRA), they are termed as “solid waste management units”
(SWMUs). These sites, if highly contaminated, under the Comprehensive Envi-
ronmental Response, Compensation, and Liability Act (CERCLA), may be listed
Role of Pollution Prevention 11

under the National Priorities List (NPL). These are also called Superfund sites.
Most of the waste to be processed by ER activities, i.e., contaminated soil, water,
building material, is the result of past production activities or due to a spill which
had been covered up at that time. This waste is called primary waste.
While performing restoration or cleanup operations, new waste, i.e., drilling
cuttings, personal protective equipment, gloves, or cleaning equipment that are
generated is called secondary waste. In cases where the ER site is very large or
heavily disturbed or very old, the site may require extensive study to assess the
level and extent of contamination. In those cases, waste will be generated
during the preliminary assessment (PA)/site investigation (SI) phase and the
characterization phase. This phase of investigation is sometimes referred to as the
remedial investigation (RI)/feasibility study (FS) phase. During the RI/FS or PA,
generation of waste, which is mostly secondary waste, can be controlled. The
number of samples and the extent of sampling must be carefully determined
to ensure that the waste generated during this phase is minimized. However,
this is possible through meaningful negotiations with regulators, affected parties,
and stakeholders.
Waste generated during interim action or cleanup action that is conducted
quickly or with limited analysis to reduce or eliminate imminent threats to the
environment, the public, or workers may be difficult. However, these situations
arise only in cases of emergency or time-critical action, generally due to either
an accident or a spill of a hazardous substance.
Superfund sites require additional studies. Studies include prioritizing all
the hazardous substances found on site, preparing a toxicological profile of each
substance, and making a complete health assessment of all substances on site.
Potential migration of these substances off-site and its impact on the surrounding
population at risk must be assessed. These studies require extensive sampling and
analysis of the site. Data quality objective (DQO) should be considered during
on-site sampling.
During the ER process, there are four phases for the incorporation of
Wmin/P2 in the process:

1. Negotiation and planning: This is the stage when negotiations with the
regulators will be taking place.
2. Assessment: Typically during sampling, drilling, treatability, and test-
run phases of the process.
3. Evaluation and selection of process: This is the decision-making phase,
when the cleanup process is decided upon.
4. Implementation: Actual cleanup is done in this phase. Some fine-tuning
of the Wmin/P2 already documented may be essential to further
decrease generation of waste.
12 Sharma

1.3 Decontamination and Decommissioning (D&D) Activity

Waste generated from D&D activity is generally similar to ER operations but
more predictable, both in quantity and composition aspect. In this case, secondary
waste and waste during characterization can easily be controlled. However, before
a D&D action is considered for cleanup, the history of the site must be fully
investigated. A thorough investigation will significantly reduce the amount of
waste ultimately generated from site cleanup. Collecting historical data may take
a considerable amount of time and should be completed prior to site cleanup. If
the site is heavily contaminated, it may qualify as an NPL site and require the
extensive study called for by the CERCLA.
Prior to decommissioning of any radioactive contaminated site, criteria and
procedures for decontamination and cleanup specified by the Nuclear Regulatory
Commission (NRC) must be determined. The NRC suggests that proposed
D&D work incorporate simple and inexpensive methods prior to NRC approval
to begin work.
Requirements for D&D may include incorporation of long-term monitoring
and surveillance of the site. Consideration of DQO may result in cost savings as
well as the number of samples to be taken during the surveillance phase.
Construction and demolition (C&D) is a bit more complex than D&D
activity. The difference is that in C&D activity, decontamination may not be
necessary, as the site is presumed to be clean. C&D generally refer to demolition
of buildings or structures where no hazardous or toxic substance would have ever
been handled or processed. To increase pollution prevention/waste minimization
in the process, demolition is sometimes referred to as deconstruction.

2 REGULATORY REQUIREMENTS
Through the passage of numerous regulations, state and federal regulatory
agencies such as the EPA have encouraged and sometimes mandated that
WMin/P2 be incorporated into the process. Some of the regulatory requirements
include the passage of the Pollution Prevention Act of 1990. Disposal of hazard-
ous waste requires the existence of a pollution prevention plan before a mani-
fest can be signed and the waste taken over by a transporter. Those plans exist
in each and every plant that produces waste. In most cases, those plans remain
on paper.
In 1993, the EPA formally endorsed pollution prevention as a guiding
principle for all EPA programs, to encourage sustainable development while
continuing the agency’s mission to protect human health and the environment. It
stated that the “mainstream activities at EPA, such as regulatory development,
permitting, inspections, and enforcement, must reflect our commitment to reduce
pollution at the source, and minimize the cross-media transfer of waste” (3).
Role of Pollution Prevention 13

As shown in Figure 1, the EPA recommends source reduction, reuse, and

recycle as the order while handling waste. Reduction includes not only lower use
of raw material in any process but also the substitution for hazardous or toxic
materials by either lesser toxic material or a nontoxic substance. Sometimes this
is referred to as chemical or material substitution. Disposal is the final option. The
reason for this is that in 1997, the nation’s largest 21,490 industrial users of toxic
chemicals released 2.58 billion pounds of toxic chemicals into the environment
(U.S. EPA, 1997 Toxic Release Inventory) (4) and spent billions of dollars
managing pollution control technology systems to prevent that quantity from
being even higher. If one were to add in the purchase price of the raw materials
that eventually escaped as 2.58 billion pounds of chemical waste (instead of
product), the price tag grows even higher. If the cost of effects on human health
were added, the price tag would be truly phenomenal.
Consider municipal solid waste (MSW), which consists mostly of nonhaz-
ardous material: a total of 208 million tons was generated in 1995. This corre-
sponds to 4.3 pounds per person per day. Of this, 56.2 million tons or 27% was
recycled, and 33.5 million tons or 16.1% was incinerated and the energy reused.
The other 118.3 million tons or 56.9% was landfilled (5). This corresponds to a
reduction of about a million tons from the previous year, possibly due to recycling
and reuse of material.
To encourage WMin/P2 through source reduction, reuse, and recycling of
waste, regulatory agencies offer many incentives. Under the permitting, inspec-
tion, and enforcement process, the regulatory agencies have the power to encour-
age the process. Many state and federal agencies encourage this through their
actions. For example, a primary goal of the Clean Air Act (CAA) is “to encourage
or otherwise promote reasonable federal, state, and local government actions,
consistent with the provisions of this chapter, for pollution prevention.” The Clean
Water Act (CWA) seeks “to eliminate the discharges of toxic pollutants.”
Stormwater regulations specifically require a pollution prevention plan. Strategies
of the EPA and state regulators to encourage WMin/P2 are accomplished by
issuing flexible permits, on occasion including explicit pollution prevention
requirements as a condition of the permit. Many states, such as New Mexico and
Texas, have a separate division in their environmental regulatory agency office
which conducts pollution prevention walk-throughs at facilities. These agencies
guarantee that any violation observed will not be used to issue a Notice of
Violation (NOV). Some states encourage P2 in their multimedia permitting
process. For example, the Massachusetts Department of Environmental Protec-
tion (DEP) uses a voluntary P2 worksheet in its Title V (for air) permit application
process. The Ohio EPA issues an air permit and includes with the permit a cover
letter urging the permittee to investigate P2 and energy conservation alternatives.
The Michigan Department of Environmental Quality (DEQ) has amended its
 14

PRODUCT

MANUFACTURING
PLANT

RECYCLE

REDUCE REUSE
VIRGIN
MATERIAL
WASTE
PROCESS TO
FOR LANDFILL
ANOTHER
PRODUCT
FROM
WASTE

FIGURE 1 Hierarchy for pollution prevention/waste minimization.

Sharma
Role of Pollution Prevention 15

permit application to include a state that simply reads: “Pollution Prevention is

the Best Solution.”

3 BARRIERS TO WMIN/P2
There is intense pressure from stakeholders in the industry to incorporate and
integrate WMin/P2 measures as an integral element into each and every process.
They want the industries in their neighborhood to be responsible for environmen-
tal protection and stewardship. In spite of this pressure from the public and the
incentives from regulatory agencies, there are many barriers to implementing P2
in an actual situation, whether it is waste management, environmental restoration,
or a D&D activity. Some of the barriers include (6):
1. Organizational barriers
2. Communication barriers
3. Economic barriers
4. Waste Generation barriers
5. Regulatory barriers

3.1 Organizational Barriers

Organizational barriers may be considered the biggest impediment for im-
plementation of WMin/P2 in an industry. It is true that large industrial firms and
almost all federal agencies have a separate environmental branch chief at the level
of a vice president in an industrial setting or at the Assistant Secretary level in
the federal agency, but they are so far removed from the actual work or plant
setting that, unless they are very significant, ideas may never reach them. In
addition, the person at the top may wholeheartedly support the WMin/P2 con-
cept, but the mid-management level may have different views on the subject.
For management, running the operation—keeping the line of production mov-
ing—is more important than thinking about changing the process line, however
significant that may be. Policy directives coming from the top can be easily
discarded or passed over on some pretext or the other. In most cases this is done
under the guise of low priority, and WMin/P2 is the lowest priority when
operation of the plant is the consideration. In most plant staff meetings, this point
is hardly ever raised.
In many organizations, pollution prevention is an additional responsibility,
combined with other programs. As a result, P2 may have a poor infrastructure. In
addition, the lack of status and visibility makes it difficult for P2 coordinators to
develop effective programs. In many instances, the P2 responsibility has been
combined with environmental compliance. Typically, compliance with environ-
mental regulations takes all the time, as the first priority of the person is not to
receive a Notice of Violation (NOV) for the company. In these cases, if P2 roles
16 Sharma

and responsibility are not clearly defined and incorporated into a job description,
they will most likely be overlooked entirely.

3.2 Communication Barriers

Searching for ways to reduce or eliminate generation of waste requires knowledge
of the process itself. In the case of regular plant operation, the P2 coordinator may
not know the various intricacies or all the steps that are followed in any
production line. If the P2 coordinator suggests any new procedure, it could be
turned down very easily. One example of this would be the case of an environ-
mental restoration project, when the P2 coordinator may not know about the
cleanup until a very late stage, when incorporation of pollution prevention
methods or procedures becomes increasingly difficult to implement.

3.3 Economic Barriers

Funding for P2 is hard to come by. This is especially true when the operation is
proceeding smoothly. Generally, the benefits of P2 projects are realized over the
long term. After meeting basic regulatory requirements, it becomes difficult to
push P2 beyond projects required to conduct normal plant operations. In almost
all cases, there is no clear-cut budget line for P2 projects, thus making it difficult
for such projects to compete for funding. P2 projects have to compete with
compliance projects for funds that are in short supply. In addition, funding is
based on return on investment (ROI) of the project. When there is intense
competition for scarce funds, P2 projects are set aside because the return on
investment may not be high enough. In reality, life-cycle costs must be considered
before economics are worked out, but for P2 projects these hidden costs are many
times not examined, making the P2 project look unattractive and therefore not
funded. It is much easier to pick up waste and truck it off-site, rather than modify
an entire process, even though the new process would be of a permanent basis
and savings would be perpetual. Whenever a PPOA is conducted in a project, a
full ROI consideration should be a part of that study.

3.4 Waste Generation Barriers

For many industries, waste generation may be highly episodic, thus making
implementation of a WMin/P2 project uneconomical. In the case of environmen-
tal restoration projects, quantities of waste are at best an estimate. Waste will
usually consist of many different types and implementation of P2 in each case
may seem to be an uneconomical consideration and therefore not implemented.
Operational personnel do not want to add the number of steps required in handling
the waste, unless savings will be very significant. In most cases, reduction in
waste generation is not high on the list.
Role of Pollution Prevention 17

3.5 Regulatory Barriers

Differences in state laws raise barriers to pollution prevention innovations.
Practices that are permitted under one state’s laws may be prohibited in another
state. Even in a single state, interpretation of the law may be different signifi-
cantly between one inspector and another, thus complicating things further. It is
true that many states, such as California, Massachusetts, New York, and Texas,
among others, facilitate P2 through legislation and their multimedia permitting
process, but this process has its difficulties. Difficulties include initial time
investment by companies and agencies while preparing the overall permit pack-
age, which in itself is very complicated and time consuming. Additionally, there
are risks associated for companies if they are unable to undertake steps as stated
in their permit package, as this would result in regulatory penalties.

4 TOOLS FOR WMIN/P2 IN WM, ER, AND D&D ACTIVITIES

With proper steps, barriers to implementation of WMin/P2 in waste management,
environmental restoration, and decommissioning projects can be overcome. Steps
toward implementation are not significantly different in WM, ER, or D&D
projects.
With the enactment of many new regulations, minimizing environmental
risk through WMin/P2 has become imperative. The goal of legislation is to reduce
toxic emissions. For example, since the RCRA, pollution prevention is required
because of the requirement for toxic use reduction plans. The Right to Know Act
and other criminal penalties for pollution in the law make it necessary for
industries to incorporate P2 in their plans. Waste management includes source
reduction and environmentally sound recycling, thus eliminating the problem of
handling waste that should never have been generated. Pollution prevention, on
the other hand, deals with the handling of waste that comes out of a process or
from the “end of the pipe.” In waste minimization, cost of handling, treatment,
and disposal does not exist. As the quantity of handling hazardous materials as a
raw material is reduced, worker risk and accidents are substantially reduced.

4.1 Role of Stakeholders

To achieve success in WMin/P2 programs, an effective environmental leadership
must be in place. The main drivers are stakeholders. In any major entity, the
primary stakeholders are investors, the community, employees, customers, and
regulators.
Investors: Every stockholder wants the company to have a good image and
the company to be competitive. Also, the enterprise should consistently
handle its affairs in a responsible way.
18 Sharma

Communities: The enterprise must be a responsible and respected member

of the community in which it operates its plants and offices. The
community must have a positive impression of the corporation, whether
it is a private concern or a governmental agency.
Employees: Employees must enjoy working for the company, both through
the physical atmosphere and through positive support by management.
Employees must feel completely safe in their work area.
Customers: When customers receive services from a company, they should
not only feel that they have received their money’s worth of services but
also that the company they are dealing with is environmentally conscious
and its products and services are handled in a responsible way. The
packaging used in the product should contain as much recycled product
as possible, and the contents should be as environmentally friendly as
possible.
The Regulators: The company must be compliant and take voluntary
actions whenever necessary.
Each stakeholder has his or her own distinct viewpoint on enforcement,
effect, and operation of any facility. For example, the company has cost effective-
ness and profits as motivators, while the regulators may have only the laws and
regulations as their primary interest. The citizens, who suffer the consequences
of pollution emanating from a facility, are not concerned about the profits of the
company. They simply want no pollution, whatever the cost may be. The
customer, on the other hand, is looking for a product at the best price possible.

5 PROFILE OF A COMPANY
Companies that have good environmental management and play a leadership role
achieve WMin/P2 through proactive steps. Table 1 gives the difference between
a reactive and a proactive company.
The difference between a reactive and a proactive company is that while a
reactive company looks for profit today, the proactive company looks at the
long-term cycle and considers the whole product line for increased profits.
The proactive company looks for continuous improvement through business
management and the decision-making process, taking the view of personnel on
the shop floor.
To become a proactive company, one has to develop and implement
strategic environmental management tools. For this, the strategy of the manage-
ment must integrate environmental planning with the business planning of the
enterprise. It has to consider environmental factors as a multimedia concept. The
company has to focus on pending and anticipated environment concerns. Potential
problems could be foreseen through a multimedia audit as well as past regulatory
Role of Pollution Prevention 19

TABLE 1 Reactive and Proactive Management Strategies

Reactive Proactive

Forced to comply with law Eliminates compliance cost

through audit programs
Responds to law enforcement Evaluates risks of actions
Lacks resources Utilizes resources sensibly
Follows trends Trendsetters
Minimum cost approach Takes life-cycle cost approach
Non-existing environmental aware- Employees fully aware of environ-
ness among workers mental program

inspection results. As these cannot be thrust upon management, an educational

process to highlight these must be undertaken to be “accepted” by management.
Educational processes could highlight the benefits, particularly company image,
reductions in costs, and increased safety and health of the workers. It should be
pointed out that WMin/P2 is the most critical segment of environmental manage-
ment, as this can assist when considering the entire operation. Benefits to be
highlighted include:

Greater efficiency means fewer materials used and less time and energy
wasted.
Improved worker safety and health means fewer or no accidents and
reduced or no lost time.
Total compliance with all regulations.
Improved company image among stakeholders resulting in better public
relations.
Reduction in treatment, handling, and disposal cost of waste.

Senior management can demonstrate its commitment by setting quantita-

tive, process, or specific section reduction and recycling goals; institute perform-
ance measures to measure the goals; support cost-saving P2 projects; and consider
design of P2 into new projects, processes, and facilities (7). Management should
provide a comprehensive policy direction and put in a viable infrastructure so that
a meaningful P2 program can be implemented and goals realized.

6 INCORPORATION WMIN/P2 INTO THE PROCESS

Successful implementation is possible when it is handled as a team. The members
of the team should come from all segments of the company—management,
20 Sharma

engineering, environmental regulatory compliance, shop workers, health and

safety personnel, and marketing. Suggested steps are as follows (8):
1. Introduce and market WMin/P2 practices as best management prac-
tices and not as a separate philosophy or practice. This encourages
wider acceptance. One reason for this is that present acceptance may
be low.
2. Assign a WMin/P2 coordinator to the overall environmental program
and thus eliminate a separate requirement on project managers.
3. Initiate mandatory, management-supported, WMin/P2 training, geared
specifically toward personnel who may have to deal with substances
of a hazardous nature and those involved with the production of waste,
especially in large quantities. The training should be specific in nature
and geared toward those attending the session.
4. Establish source reduction and recycling goals. These may be set up
and endorsed by the highest management personnel.
5. Include WMin/P2 as an element of the budget, planning, and all the
key decision-making documents.
6. Include WMin/P2 criteria and performance measures in negotiated
agreements with regulatory agencies.
7. Focus on implementation of the projects.
8. Include cost drivers and other incentives to reduce waste. Total cost
of disposal, which should include the cost from waste generation
through disposal and also landfill replacement, should be considered
and highlighted to personnel who produce the waste. This is a life-
cycle cost of waste disposal and should be compared with the cost of
recycling the waste.
9. Also, the cost of disposal must be the responsibility of the generating
program.
10. Highlight the successes so achieved by these programs and reward the
personnel involved in those successes. Top management must be made
aware of the successful implementation of this program so that
WMin/P2 can receive wider acceptance.

6.1 Priorities for the Process

If a company has a very poor environmental record, it may be difficult to embark
upon all the steps outlined above. For this reason, management may feel it
necessary to prioritize these steps. The company may still place immediate profits
and lack of prior results above everything else, but highlighting the benefits may
move management more quickly toward the implementation of the process.
For the team, establishing goals may be difficult at the beginning. Market
standards should be used in this regard. However, goals must be in place at the
Role of Pollution Prevention 21

beginning of the project. The earlier the goals are set, the better the chances of
attaining these goals. To achieve these goals, each and every segment of the
project should have a separate performance measure that can be related to
concrete pollution prevention goals established by company management.
The focus of the group should be on the implementation of the process within
the time frame, with goals of Wmin/P2 for added savings. The team should
highlight this focus to the entire company; without their active involvement
and support, achieving the goals may become difficult. Incorporation of
WMin/P2 in the design concept of the process is essential. While working toward
these goals, care should be taken that all safety as well as environmental
regulations are followed.

7 KEY ACTIVITIES TO BE CONSIDERED FOR WASTE

MINIMIZATION/POLLUTION PREVENTION
For successful implementation of pollution prevention in a project, early negoti-
ations with regulatory agencies, at the beginning of the project cleanup phase or
design phase for a manufacturing plant, are necessary. Most states incorporate
pollution prevention into their regulations, in both permitting as well as inspection
process. Concessions may be received from the regulatory agency if it is involved
early in the project. Discussions with regulators for permitting of the facility
should be conducted as early as possible. Table 2 lists some issues that the
companies may raise with the regulator (4).
In the case of environmental restoration and decommissioning projects,
preliminary assessment and inspection must be viewed with WMin/P2 as a
significant element of the process. In a major ER or D&D project, the character-
ization phase is significant, as substantial quantities of waste may be generated
during the sampling and analysis phase.
In the case of large ER and D&D cleanup, project evaluation of cleanup
activities should be seriously considered. All preferred alternatives should be

TABLE 2 Preliminary Discussion Topics with Regulators

Pre-permit facility analysis
Preapplication meetings and permit scoping sessions
Permit application
Draft permit writing
Public comment
Final permit issuance
Permit renewal
Permit modification for process expansion or increased capacity
22 Sharma

documented and evaluated independently. This should include the amount of

waste that may be generated, both primary and secondary waste. In the case of
ER or D&D project, cleanup action should be designed with a view of reuse of
equipment to be used for the operation. In the case of WM, the design consider-
ations should include maximum utilization of raw materials so that waste gener-
ated is minimized.
Some of the places where waste minimization should be considered include
fugitive emissions and process modifications (see Table 3).

7.1 Measures to Consider WMIN/P2

After the design is optimized, the option that generates the least amount of waste
and at minimum cost to operate the process should be considered for implementa-
tion. Even the handling of waste so produced from either the cleanup process or
from manufacturing plant needs to be designed. In the case of ER and D&D
operations, if they are properly designed, generation of secondary waste can be
significantly reduced.
Some of the Wmin/P2 applications that need to be considered include the
following.
1. Segregation of waste: This includes collecting hazardous and non-
hazardous waste separately. Depending on specific regulations con-
cerning disposal of waste, even the concentration of hazardous
substance may mean savings in disposal costs, and this must also be
looked into.
2. Use of a contamination zone: A clearly marked contamination zone
may result in not generating waste unnecessarily. This will keep
separate the area of operation into which the public may not enter.

TABLE 3 Candidate Items for Waste Minimization

Fugitive emissions reduction
Process modifications
Material and product substitution
Preventive and corrective maintenance
Routine equipment inspections
Proper material handling and storage procedures
Loss-prevention practices
Employee training programs
Material tracking and inventory control
Improved documentation
Environmental audits
Role of Pollution Prevention 23

3. Reduction of equipment used for cleanup: A specific example is the use

of personal protective equipment (PPE). Recyclable PPE should be
used wherever possible. Large equipment used in cleanup areas should
be optimized. In some cases, once equipment has been used, it may be
contaminated and may eventually become a waste. This must be
minimized as much as possible.
4. In-situ waste treatment for ER cleanup: Contamination such as petro-
leum products could be cleaned up in situ, thus also eliminating
generation of waste. Bioremediation should be considered as much as
feasible.
5. Minimization uncontaminated material removal: All the material marked
for cleanup in a site may not have to be removed, as many parts of the
same may still be untouched. Care should be taken to remove only the
contaminated portion of the site.
6. Implementation of new technologies: This may become a viable option
where the cost of disposal of any waste or cleanup of a site is very high.
It is generally considered if the waste is radioactive in nature.
7. Unpacking outside contaminated areas: Packaging material should be
removed in clean areas to reduce the amount of waste generation.
8. Avoidance ground contamination through ground cover: If this is done,
then the cost of cleanup is significantly reduced in case of a spill.
9. Strict housekeeping: Cleanliness can result in significant reduction of
generation of waste.

When several ER and D&D projects are being considered simultaneously,

plans to use equipment from one project on another can be incorporated if
considered at the planning stages. This is critical in some instances: the equipment
becomes contaminated and has to be disposed of in a landfill unless it can be used
on similar projects.
Some of the tools that are available for total WMin/P2 considerations
include the “Design for the Environment” (DfE), also called “P2 by Design.”
Another is the “Pollution Prevention Opportunity Assessment (PPOA)” concept.
For effective environmental protection of a contaminated site, sufficient quantity
and quality of data needs to be collected to support defensible decision making
and at the same time meet regulatory requirements. When conducting a PPAO, in
some cases the ROI ranking system may be considered.
To minimize expenditures related to data collection by eliminating unnec-
essary, duplicative, or overly precise data, the EPA has developed the Data
Quality Objective (DQO). Use of this process depends on the complexity of the
study, but it can be used for both small as well as large sites. However, the DQO
process should be used during the planning stage of any study that requires data
collection, before the data are collected.
24 Sharma

7.2 Design for Environment (DfE)

The environmental responsibility of the corporation and the EPA’s source reduc-
tion as the first option to reduce waste makes consideration of DfE necessary.
This is a testing ground for new approaches to risk reduction through pollution
prevention. As per the EPA, a DfE program provides guidance and tools to help
companies achieve continuous environmental improvement (9). A DfE-based
environmental management system provides a company with opportunities to go
“beyond compliance” and save money. It helps companies promote the evaluation
of cleaner production alternatives. By implementing these alternatives, a com-
pany can continuously improve its environmental performance. DfE also strongly
supports pollution prevention principles, as that is its goal. This goal is accom-
plished through front-end innovations through redesign rather than relying on
end-of-pipe controls to reducing potential risks to human health and the environ-
ment. This approach uses cleaner technology substitute assessments and life-cycle
tools to evaluate the performance, cost, and environmental impacts of competing
technologies. This serves as a guide to decision makers in order to arrive at an
optimum decision. For this to be considered, the project team should recruit
partners from among the various stakeholders. This team will select the process
that will generate the least amount of waste.
During process design consideration, not only will the flow of material be
a consideration, but also the waste that will be disposed of in the atmosphere. For
this reason, a regulatory profile identifying applicable federal and state regula-
tions will be undertaken. For process consideration, the project team will look
into the availability of raw materials and the location of availability, the effects
of the use of the product by the consumer, the effect of disposal on the
environment, and potential workplace safety hazards.
The state of New Mexico is implementing the Green Zia Environmental
Excellence program. This award program has three steps that lead an organization
from the beginning of a prevention-based environmental management system,
termed commitment level, to an established system that is demonstrating results,
termed achievement level, to a fully deployed system with significant results. This
is the Excellence Award level under the Green Zia Program.

7.3 Pollution Prevention Opportunity

Assessment (PPOA) (10)
A Pollution Prevention Opportunity Assessment (PPOA) is one of the first steps
in preparing a pollution prevention plan. For each independent process, PPOA is
a tool for the company to identify the nature and amount of wastes and energy
usage, stimulate the generation of pollution prevention and energy conservation
opportunities, and evaluate those opportunities for implementation. It is a system-
atic framework that can be used by a facility’s own employees to identify waste
Role of Pollution Prevention 25

minimization opportunities. As a structured program, it provides intermediate

milestones and a step-by-step procedure to:
1. Understand the facility’s processes and wastes
2. Identify options for reducing waste
3. Determine if the options are technically and economically feasible to
justify implementation
These procedures consist of four major steps:
1. Planning and organization—organization and goal setting
2. Assessment—careful review of a facility’s operations and waste streams
and the identification and screening of potential options to minimize
waste
3. Feasibility analysis—evaluation of the technical and economic feasi-
bility of the options selected and subsequent ranking of options
4. Implementation—procurement, installation, implementation, and
evaluation.
The initial goal of a PPOA is to encourage participation in the assessment
process by management and staff at the facility. The steps, shown in Figure 2,
include corporate commitment to a waste minimization initiative and selection of
a task force or similar group to carry out the assessment. Top management must
endorse the study and issue a policy statement regarding waste minimization,
establishing tentative waste reduction goals to be achieved by the program. This
will aid in identifying waste-generating sites and processes, conducting a detailed
site inspection, developing a list of options which may lead to the waste reduction
goal, formally analyzing the feasibility of the various options, measuring the
effectiveness of the options, and continuing the assessment.
The EPA has developed very useful worksheets to conduct a PPOA for any
process (11). Suggested steps include the following.
1. Assessment overview: This summarizes the overall program. This will
include the establishment of the P2 program at the site, preliminary
assessment, establishment of a program plan, defining P2 options,
feasibility analysis, a written assessment report, implementation of the
plan, a measure of progress, and maintainenance of the P2 program.
2. Assessment phase: This includes site description, collection of process
information, input of materials inventory, products summary, waste
stream summary, options generation, and descriptions of the options.
3. Feasibility analysis phase: This is the phase which looks at the profit-
ability of the various options under considerations. It should facilitate
the ability of management to select the optimum option that a PPOA
should bring about.
26 Sharma

The need to minimize waste

PLANNING AND ORGANIZATION

• Get Management Commitment
• Set Program Goals
• Organize PPOA Team

Collect Information
About the Process

Facility Tour By
PPOA Team members

Brainstorming for Options

Estimate Waste Reductions

Cost Savings for Options

Prioritize Options
Document and Report

Successfully Implemented
Waste Minimization Projects

FIGURE 2 Planning a pollution prevention opportunity assessment for a

process.

For municipal solid waste (MSW), the EPA’s integrated waste management
includes the following components (12).
1. Source reduction: This includes changes in packaging design, backyard
composting, grass recycling, changes in purchasing habits, increased
reuse, changes in industrial practices, and changes in use patterns. This
is considered the generation of waste management.
2. Recycling of materials: This includes recovery for recycling, including
composting.
3. Waste combustion: Combustion and disposal. Combustion means re-
covery of heat for use.
Using the above means, landfilling of waste is only 56.9% of the waste generated.
Role of Pollution Prevention 27

PPOAs have been successfully used in manufacturing, sampling and anal-

ysis, ER cleanups, D&D processes, as well as in the design and construction of
new facilities.

7.4 Return on Investment (ROI)

Though similar to the study carried out under a PPOA, ROI is a tool which assists
management in deciding on the options for WMin/P2 that are put forward. The
ability to show cost savings on WMin/P2 projects is a major reason for funding
these projects, in addition to the obvious benefit of reducing waste. Projects with
high ROI will show the value of implementing the project in an atmosphere of
diminishing resources.
For ROI consideration, the project should have a defined scope. The scope
may include the following:
1. Percentage of return-on-investment (ROI)
2. Type of project
3. Type of waste that will be reduced or eliminated
4. Volume of waste that will be reduced
5. Technical approach to accomplish the reduction
Each of the above categories will have different levels. It is suggested that there
be four levels for each of the five suggested scopes. Each could be given a specific
number based on the importance attached to that category. For example, an ROI
value of greater than 100% would be rated as 4, while an ROI value of less than
25% as 1. Similarly, if the project type results in source reduction, it would be
rated as 4; while if the waste would be sent to a landfill, it would get a 1.
Depending on the type of waste generated and the cost of disposal, ratings would
be assigned. Similar is the case with the quantity of waste that will be either
reduced or eliminated. The technical approach is a consideration. If the process
under consideration is readily available, then the rating would be high, as results
would certainly be achievable. On the other hand, for a research-related project
the return is not a certainty and management may not be very inclined to fund it.
One way of calculating the percent ROI is through the following equation (13):
(B − A) − [(C + E + D)/L]
ROI = × 100
C + E + D
where A = annual costs after implementation of P2 project
B = annual costs before implementation of P2 project
C = capital investment for the P2 project
D = estimated project termination/disassembly cost
E = installation operating expenses
L = number of useful years of the project
28 Sharma

Annual costs may include cost of the equipment; raw materials and supplies;
utilities and labor costs; routine maintenance costs for the processes; PPE costs;
waste management costs such as waste containers, treatment, storage, and dis-
posal; inspection and compliance costs; recycling costs such as material collec-
tion/separation/preparation; materials and supplies; operation and maintenance
labor costs; and administrative and other costs. For projects lasting less than five
years, D could be used as is, after more than five years, D could be zero.
Care should be taken when the ROI is calculated: it is suggested that though
implementation of the project would mean a permanent reduction in the genera-
tion of waste, savings over a three-year period should be considered for ROI
calculations.

8 DATA QUALITY OBJECTIVE (DQO)

The DQO process is a strategic planning approach based on the scientific
method that is used to prepare a data collection activity. It provides a systematic
procedure for defining the criteria that a data collection design should satisfy,
including when to collect samples, where to collect samples, the tolerable level
of decision errors for the study, and how many samples to collect (14). DQO data
collection specifies the final configuration of the environmental monitoring or
measurement effort required to satisfy the specific study. It designates the types
and quantities of samples or monitoring information to be collected; where, when,
and under what conditions they should be collected; and what variables are to
be measured.
The DQO process assures that the quality of environmental data used in
decision making will be appropriate for the intended application. This process, as
shown in Figure 3, can improve the effectiveness, efficiency, and defensibility of
decisions in a resource-effective manner. The DQO process leads to the develop-
ment of a quantitative and qualitative framework for a study, thereby facilitating
an optimum selection of a cleanup for the site.
The benefits of DQO are that the data collection is most appropriate for the
site, and condition of the site is considered during the planning stage of data
collection. All the tolerable limits on decision errors, which will be used as the
basis for establishing the quantity and quality of data needed to support the
decision, are documented before sampling is undertaken. The process assures that
optimum samples are collected so that decisions reached from the data are
acceptable to all stakeholders. This will not only optimize resources for the
project but also eliminate generation of waste.
DfE or P2 by Design is mostly useful in the planning of new processes,
while PPOA can be used in any program, including ER and D&D projects. DQO
can be used in all the cases where extensive sampling is one of the requirements.
It should be noted here that results achieved by implementation of the WMin/P2
Role of Pollution Prevention 29

Assemble and Evaluate

Background and Archival
Data for the Site

Define and Refine

conceptual Model and
Data/Parameter Distributions
based on Available
information

Define/Refine Criteria for

Decision Making (Estimation
Action Levels and
Characterization on priorities
using preliminary risk
evaluation)

Define/Refine DQOs

Develop/Modify Sampling
and Analysis Plan

FIGURE 3 Data quality objective process.

concept in projects have resulted in cost savings of up to 50% in ER projects, and

reuse of all waste in D&D projects.

9 MANAGEMENT EXAMPLES OF WMIN/P2

As indicated earlier, source reduction should be the first priority to reduce the
generation of waste. Material substitution, consideration of new technologies, and
use of best-management practices should be considered. This includes replace-
ment of the use of hazardous substances with nonhazardous substances, and
optimizing sample collection and analysis. The second option is recycling, and
that includes reuse of material and reclaiming the waste produced from one
process to make that a raw material in another. The third step is treatment of the
waste. Here, options include stabilization, volume reduction through compaction,
incineration, filtration, precipitation, thermal destruction of organic material, and
vitrification. Disposal should always be considered a last option.
30 Sharma

In most cases, regulatory agencies have indicated a willingness to adopt

pollution prevention in any cleanup activity. By making pollution prevention a
part of the planning process, it will be ratified by the regulatory agency and so
could have a part in scheduling the pollution prevention consideration itself. This
would give the agencies more time to devote to PPOA or DfE studies. Regulatory
interpretation would now favor pollution prevention. If the project is an ER or
D&D cleanup, future use of the land, cleanup levels, and release criteria could be
negotiated.
WMin/P2 is a continuous improvement cycle (15). It starts with planning,
implementation, evaluation, and review. Pollution prevention assessment has to
be reviewed again with a multi-disciplinary group. If done early enough to review
optimization of the process and energy use, it will generally result in further
reduction in waste production and thus reduction of disposal costs.

10 CONCLUSION
There will be a constant struggle regarding reduction or even elimination of
pollution from any source at any site. International Organization for Standardization
(ISO) 14000 certification, a customer-driven worldwide movement, sees
WMin/P2 as a very significant activity. Individual stakeholders have unique
viewpoints regarding pollution emanating from a process. Differences in im-
plementation of waste minimization/pollution prevention programs in waste
management, environmental restoration, and decommissioning projects include
the concept of waste in each case. Environmental restoration and decommission-
ing project managers may feel that as the waste exists, it cannot be reduced.
However, once the concept of WMin/P2 is introduced and success is very visible,
a change in attitude should take place. As a first step, secondary waste can easily
be reduced, and this in itself should result in substantial savings for the project.
Additionally, the concept of reuse of material in the cleanup process is another
instance where savings are easily attainable.
It is true that the earlier the WMin/P2 concept is considered in the program
management phase of any project, the greater the savings that will result.
The concept will result in a better understanding of not only the problem being
faced but also more unanimity in the solutions that will be presented by the parties
involved. Beginning with management sponsorship of P2 activities as an import-
ant element, a successful implementation of Wmin/P2 requires continuous aug-
mentation and support from all those involved.

ACKNOWLEDGEMENT
The author would like to thank Gaeton Falance for all his help and assistance in
putting this chapter together.
Role of Pollution Prevention 31

REFERENCES
1. Chemical and Engineering News, vol. 77, no. 17, p. 10, April 26, 1999.
2. Independent Technical Review of Three Waste Minimization and Management
Programs, p. 3-2. Albuquerque, NM: U.S. Department of Energy, Albuquerque and
Oakland Office, August 1995.
3. EPA Pollution Prevention Policy Statement: New Directions for Environmental
Protection, June 15, 1993.
4. EPA Pollution Prevention Solutions During Permitting, Inspections and Enforcement.
EPA/745-F-99-001, p. 29, June 1999.
5. Characterization of Municipal Solid Waste in the United States: 1996 Update, U.S.
EPA, Office of Solid Waste, EPA530-R-97-015, p. 10. Prepared by Franklin Associ-
ates, Prarie Village, KS, June 1997.
6. EPA Federal Facility Pollution Prevention: Tools for Compliance, EPA/600/R-94/154,
p. 54, September 1994.
7. U.S. Department of Energy, Pollution Prevention Program Plan. DOE/S-01/8, p. 4.
Washington, DC, 1996.
8. Los Alamos National Laboratory, Applicability of Waste Minimization to Environ-
mental Restoration, LA-UR-96-17-21, Los Alamos, NM, pp. 9–15, June 1996.
9. EPA Environmental Management Systems Bulletin 1, EPA 744-F-98-004, July 1998.
10. U.S. EPA Waste Minimization EPA Assessment Manual, PEA/625/7-88/003, pp. 6–
10. Cincinnati, OH: Hazardous Waste Engineering Research Lab, July 1988.
11. U.S. EPA Facility Pollution Prevention Guide, EPA/600/R-92/088, Washington, DC,
May 1992.
12. Characterization of Municipal Solid Waste in the United States: 1996 Update, U.S.
EPA, Office of Solid Waste, EPA530-R-97-015, p. 89. Prepared by Franklin Associ-
ates, Prarie Village, KS, June 1997.
13. Guidance for ROI Submissions. Albuquerque, NM: U.S. Department of Energy,
1996.
14. Environmental Protection Agency, U.S. Office of Research and Development, Guid-
ance for the Data Quality Objectives Process, EPA/600/R-96/055, Washington, DC,
September 1994.
15. U.S. EPA Facility Pollution Prevention Guide, EPA/600/R-92/088, Washington, DC,
May 1992.

ABBREVIATIONS
A annual costs after implementation of P2 project
B annual costs before implementation of P2 project
C capital investment for the P2 project
CAA Clean Air Act
CERCLA Comprehensive Environmental Response, Compensation, and
Liability Act
C&D construction and demolition debris
D estimated project termination/disassembly cost
32 Sharma

D&D decontamination and decommissioning

DfE design for environment
DQO Data quality objective
E installation operating expenses
EMS environmental management system
EPCRA Emergency Planning and Community Right-to-Know Act
ER environmental restoration
ISO 14000 International Organization for Standardization 14000
L number of useful years of a project
MSW municipal solid waste
NOV Notice of Violation
NPL National Priorities List
PA preliminary assessment
PPE personal protective equipment
PPOA Pollution Prevention Opportunity Assessment
RCRA Resource Conservation and Recovery Act
RI/FS remedial investigation/feasibility study
ROI return on investment
SI site investigation
SWMU solid waste management unit
TRI toxic release inventory
WM waste management
WMin/P2 waste minimization/pollution prevention

GLOSSARY
Construction and demolition debris (C&D) The waste building materi-
als, packaging, and rubble resulting from construction, remodeling,
repair, and demolition operations on pavement, houses, commercial
buildings, plants, and other structures.
Data quality objective (DQO) Qualitative and quantitative statements
derived from the DQO process that clarify study objectives, define the
appropriate type of data, and specify the tolerable levels of potential
decision errors that will be used as the basis for establishing the quality
and quantity of data needed to support decisions. It provides a systematic
procedure for defining the criteria that a data collection design should
satisfy, including when to collect samples, where to collect samples, the
tolerable level of decision errors for the study, and how many samples to
collect.
Decontamination and decommissioning (D&D) The process of reduc-
ing or eliminating and removing from operation of the process harmful
substances, such as infectious agents, so as to reduce the likelihood of
Role of Pollution Prevention 33

disease transmission from those substances. After the D&D operation,

the process is no longer usable.
Demolition The wrecking or taking out of any load supporting structural
member and any related razing, removing, or stripping of a structure.
Also called deconstruction.
Design for environment (DfE) The systematic consideration of pollution
prevention/waste minimization options during the design consideration
of any process associated with environmental safety and health over the
product life cycle.
Environmental assessment (EA) A document that briefly provides suf-
ficient evidence and analysis for determining whether to prepare an
environmental impact statement or a finding of no significant impact.
This document will include a brief discussion of the need for the
proposal, of alternatives as required by EPA regulations, of the environ-
mental impacts of the proposed action and alternatives, and a listing of
agencies and persons consulted.
Environmental management system (EMS) A systematic approach to
ensuring that environmental activities are well managed in any organiza-
tion. It is very similar to ISO 14000.
Environmental restoration (ER) Cleaning up and restoration of sites
contaminated with hazardous substances during past production or dis-
posal activities.
ISO 14000 International Standardization of Environmental Management
System Standard which is “that part of the overall management system
which includes organizational structure, planning activities, responsibil-
ities, practices, procedures, processes and resources for developing,
implementing, achieving, reviewing and maintaining the environmental
policy.”
Municipal solid waste (MSW) Residential and commercial solid wastes
generated within a community.
Pollution prevention opportunity assessment (PPOA) A tool for a
company to identify the nature and amount of wastes and energy usage,
stimulate the generation of pollution prevention and energy conservation
opportunities, and evaluate those opportunities for implementation.
Recycling of materials The use or reuse of a waste as an effective
substitute for a commercial product, as an ingredient, or as feedstock in
an industrial or energy-producing process; the reclamation of useful
constituent fractions in a waste material; or removal of contaminants
from a waste to allow it to be reused. This includes recovery for
recycling, including composting.
Return on investment (ROI) The calculation of time within which the
process would save the initial investment amount if the suggested
34 Sharma

changes were incorporated into it. In this calculation, depreciation,

project cost, as well as useful life are taken into account.
Source reduction Any practice which: (a) reduces the amount of any
hazardous substance, pollutant, or contaminant entering any waste
stream or otherwise released into the environment prior to recycling,
treatment, or disposal; and (b) reduces the hazards to public health and
the environment associated with the release of such substances, pollu-
tants, or contaminants.
Thermal destruction Destroying of waste (generally hazardous) in a
device which uses elevated temperatures as the primary means to change
the chemical, physical, or biological character or composition of the
waste. Examples include incineration, calcination, oxidation, and micro-
wave discharge. Commonly used for medical waste.
Toxic release inventory (TRI) Required by the EPCRA, a TRI contains
information on approximately 600 listed toxic chemicals that the facili-
ties release directly to air, water, or land or transportation of waste
off-site.
Vitrification A process of immobilizing waste that produces a glasslike
solid that permanently captures radioactive materials.
Waste combustion Combustion of waste through elevated temperature
and disposal of the residue so generated in the process. It also may
include recovery of heat for use.
Waste management (WM) Activities associated with the disposition of
waste products after they have been generated, as well as actions to
minimize the production of wastes. This may include storage, treatment,
and disposal.
3
The Waste Management Hierarchy

W. David Constant
Louisiana State University and A&M College, Baton Rouge, Louisiana

1 INTRODUCTION
The management of waste can be approached from several venues, including
regulations, history, technical methods, and interpretations of past management
practices and our current methods to manage waste in what is considered the
proper approach today. This chapter will explore the above approaches to waste
management, present the Natural Laws (1) for the reader’s consideration, and then
describe a simple hierarchy for waste management based on these laws. The im-
pact of the “implementation” of natural attenuation in many remediation schemes
of today is also discussed. The objective is to raise awareness of both the
capabilities and limitations that are placed on society in the management of waste.

2 HISTORICAL PERSPECTIVE
While we have recently increased our awareness of environmental problems and
waste management, these issues have been in effect to some degree since society
began to reach beyond simple existence. Humankind for centuries has developed
and exploited available resources in useful and necessary ways, along with
wasteful approaches. However, significant problems arose once communities,
towns and cities developed into urban centers wherein contamination of water
35
36 Constant

supplies from waste and animals caused significant deaths to occur. Further
industrialization and heavy dependence on fossil fuels has in the past century
greatly increased pressure on the environment to cope with the anthropogenic
materials and methods of humankind’s development. The development of regu-
lations in the United States, described below, best illustrates the interactions for
such a heavily industrialized nation.
In earlier history the best examples of industrial pollution are found in
England (2), where factories contaminated nearby rivers and raised awareness
about the limitations of drinking water sources. Air pollution resulted from use of
coal for fuel, but it was only after many years, in the mid-1800s and later in the
1900s, that regulations and cause-and-effect mechanisms led to control of pollu-
tant levels. Most unfortunate was the episode occurring in London during
December 1952 due to stagnant conditions over the city, wherein pollutant
concentrations resulted in death of about 4000 people from particulates and SO2
buildup. This event was followed by the passage of the Clean Air Act by the
government of England, which laid the basis for pollution control in that country.
In the United States, the historical perspective can be best represented
through actions and activities in the United States and resulting regulations, to tie
two perspectives together. Initial efforts were focused on water pollution by the
River and Harbor Act of 1899, the Public Health Service Act of 1912, and the Oil
Pollution Act of 1924, all being fairly localized in action. Only after World War
II did the U.S. government take significant action to control pollution problems
with the Water Pollution Control Act of 1948 and the following Federal Water
Pollution Control Act (FWPCA) of 1956, which set funds for research and
assisted in state pollution control with construction of wastewater treatment
facilities. In 1965, the Water Quality Act provided national policy for control of
water pollution. Focusing on drinking water, the Safe Drinking Water Act
(SDWA) of 1974 directed the U.S. Environmental Protection Agency (EPA) to
establish drinking water standards, which occurred in 1975. In 1980, Congress
placed controls on underground injection of waste, requiring permits for the
method. Finally, the SDWA amendments of 1986 led to interim and permanent
drinking water standards.
It was not until the 1972 amendments were made to the FWPCA that the
nation implemented major restrictions on effluents to restore and maintain water
bodies in the United States. The Clean Water Act of 1977 added to this focus with
consideration of toxins being 65 substances or classes as a basis to reduce and
control water pollution. This action led to the initial priority pollutants list, which
included benzene, chlorinated compounds, pesticides, metals, etc. In combina-
tion, then, the FWPCA and CWA provided the National Pollution Discharge
Elimination System (NPDES) permit system in place today.
These regulatory activities, while focused on water media and abatement
of problems in rivers and other water bodies, did not directly address the other
The Waste Management Hierarchy 37

media in our ecosystem—soil (land) and air. As industry responded to the water
regulations, unengineered disposal of waste on land (unengineered pits) became
an acceptable and legal method for waste management in many industrial streams,
including petroleum wastes, petrochemical wastes and off-spec products, and
solid waste disposal (old garbage dumps). These activities led to numerous acts
to control and mitigate pollution from dumping, etc. Initial efforts involved
control of the transportation of solid food wastes for swine, for control of
trichinosis. Modern regulations began with the Solid Waste Disposal Act (SWDA)
of 1965 and the National Environmental Policy Act of 1969, which required
environmental impact statements. The Resource Recovery Act of 1970 amended
the SWDA about the time that the Environmental Protection Agency was formed.
True regulation for solid waste management did not come into effect until the
Resource Conservation and Recovery Act (RCRA) of 1976, with guidelines for
solid waste management and a legal basis for implementation of treatment,
storage, and disposal regulations. Also, hazardous wastes and solid wastes were
defined by the RCRA. With numerous amendments, the RCRA was followed by
the Comprehensive Environmental Response, Compensation and Liability Act
(CERCLA) in 1980 to deal with abandoned sites and provide the funds and
regulations to perform cleanups. CERCLA, or Superfund, has been through
numerous revisions, and its effectiveness has come under question due to the great
deal of litigation involving cleanup of old sites.
Air quality needs became apparent in the 1950s due to the Donora,
Pennsylvania, accident, and the linkage shown between automobile emissions and
photochemical smog, but it was not until the Clean Air Act of 1963, and
amendments in the 1960s, 1970s, and 1990s that true national programs were
established for pollution control in the air medium. These regulations were
focused on motor vehicle emissions, and on emissions from industrial sources.
Thus, the United States has “chased” waste management and pollution in all
media, and while regulations are now complex, they do provide for control,
management, and abatement of pollution from recognized sources to water, land
and air.
Two points develop from this brief historical–regulatory review. First,
waste is tied directly to population, and population is growing at a rapid rate, so
these growth centers must manage and direct waste properly to avoid release and
contamination problems. Second, while many countries have significant controls
in place as in the United States, many Third World countries and underdeveloped
regions are “behind the curve” in regulatory and technical development to
manage waste. Many are still dealing with “end-of-pipe” technologies while the
United States and others are dealing with remediation, mitigation, and pollution
prevention. Still others lack the fundamentals of basic treatment technologies and
have significant population growth. Thus our history, in the United States and
England, has the potential to continue to repeat itself, unless proper technology
38 Constant

is brought to these developing population areas. While the United States and
England had time to deal with waste issues, our continued use and development
of agricultural land has diminished our resources, and places high stress on those
agricultural lands to provide food for the expanding of society. Hopefully, balance
will be achieved on a global scale in time to meet the population demand
with managed resources and sufficient waste management to protect all media
and humankind.

3 TECHNICAL APPROACH
In order to manage waste properly, we must explore the geography of a process
so that appropriate engineering (and the constraints of different areas of geogra-
phy) can be applied to solve a waste management issue or problem. Let us focus
now on a chemical manufacturing process, wherein raw materials are taken to
manufacture products, such as petroleum to petrochemicals for containers. There
are three distinct areas—the process itself, the facility boundary (fence line), and
“nature” outside the fence line. Historical sites such as those covered in Super-
fund regulations also include a boundary and “nature.” Nature is defined here as
everything except humankind or society. In order to properly apply a sound
technical approach to the waste management of such a manufacturing facility,
each of these three areas must be considered from an engineering perspective.
First, in the process itself, classical chemical engineering is applied, including
reactor design, thermodynamics, unit operations, mass transfer, etc., which are
well established methods in the chemical process industry (CPI). The focus here
is on the process, products, and profit. The second area, the boundary of the
facility, is where the bulk of waste management is located, including recycle,
reuse, treatment, source control, etc. Lines of these two areas are blurred today
with optimization of processes, recycle, and substitution of chemicals to minimize
pollution. However, both of these geographic areas are engineered and controlled
in terms of materials handling, processing, and safety, as would be found in any
chemical process. The third geographic area brings us to nature—the area around
the facility or waste site, where the fate and transport of contaminants released
from the first two regions now takes control. In the realm of environmental
chemodynamics (3), the controlling factors are the transport of chemicals in the
environment, governed by the physical-chemical relationship to reaction, trans-
port, etc. Waste management in this region now involves sorption, sediment
oxygen demand, groundwater modeling, biodegradation, partition coefficients,
and other multimedia processes. The shift in understanding in this region is
significant. We no longer have a reactor vessel, a temperature controller, or a
homogeneous catalyst bed. The systems are heterogeneous, are difficult to scale,
and may not provide consistent or reproducible results when management meth-
ods or technologies are applied to a waste problem. In addition to our lack of
The Waste Management Hierarchy 39

control over these systems, problems faced are usually dealing with low levels of
contamination, which are difficult to model, predict, or treat. However, as risk
assessment and exposure assessment methods improve in accuracy and realism,
these problems are being tackled with growing frequency. It is important to
recognize in the natural environment that our efforts are usually secondary to
existing natural forces. An excellent basis to approach management of waste, both
in the CPI model and beyond, in nature, is found in the Natural Laws, as
illustrated below. Also, a significant contrast develops when we look at the
Natural Laws, especially if one compares them to the five elements in the federal
approach to management of hazardous wastes, as listed below:

1. Classification of hazardous waste

2. Cradle-to-grave manifest system
3. Federal standards for treatment, storage, and disposal (TSD) facilities
4. Enforcement with permits
5. Authorization of state programs

4 THE NATURAL LAWS

Dealing with waste falls under the Natural Laws (1,4) and it is from these laws
that the waste management hierarchy is formed:

1. I am, therefore I pollute.

2. Complete waste recycling is impossible.
3. Proper disposal entails conversion of offensive substances into environ-
mentally compatible earthenlike materials.
4. Small waste leaks are unavoidable and acceptable.
5. Nature sets the standards for what is compatible and for what are small
leaks.

Briefly, these laws state the rules we must follow to properly manage waste in
the future. Since we exist, we generate waste, and thereby pollute. This is due to
the second law, which makes complete recycling impossible, as in thermodynam-
ics, wherein no real process is completely reversible—some loss occurs. With
some waste therefore being generated, the third law requires that the material be
returned to the environment (nature) in a compatible format—that is, earth-
enlike—in either a solid, liquid, or gaseous state. When returned, small leaks will
occur, as with minor auto emissions, and these are unavoidable and acceptable,
provided we observe nature’s standards as to what is compatible and how small
(or large) the leaks can be. A logical flow of management choices follows from
these laws.
40 Constant

5 WASTE MANAGEMENT CHOICES

The following list incorporates all options available and is similar to lists
developed by the EPA and others (5). The management list also supports the
relationship presented by Reible (2) in that environmental impact is proportional
to population times per-capita resource usage divided by environmental effi-
ciency. In words, then, the environmental impact is minimized for a given
standard of living when the environmental efficiency is high or improved.
Reible’s relationship supports the third law, to minimize impact via high environ-
mental efficiency, returning material (and energy) in compatible forms. It is
important to note here that much of the waste discussion focuses on material, and
that energy pollution should not be neglected, due to problems found in changing
river temperatures due to discharge, global warming, etc. To answer the old
question, “How clean is clean?,” a material is clean when it is returned in a form,
amount, and concentration which is acceptable to that found in nature. In other
words, a material is “clean” when its concentration does not exceed the natural
limits of that material in the space established by the balances (material) that
assimilate it (6).
Clearly, then, minimization is the first choice and the optimal one from an
environmental standpoint. However, society demands a certain standard of living,
so for those wastes remaining from minimization, destruction becomes the best
alternative. Why destruction, as such a choice would support technologies such
as incineration? Because it is the molecular structure, among other things, that
provides the toxicity of the compound, and if it can be broken down (hopefully
not yielding a more toxic compound), toxicity can be reduced or eliminated in
efficient and correct incineration processes. However, not all wastes causing
toxicity problems can be destroyed, such as heavy metals passing through an
incinerator. Thus, these materials must be properly treated prior to release,
changing their chemical states or bonding for a less toxic or hazardous form.
Finally, one notes that in all processes such as those above and others, some
residuals always remain, and lead to the final option, disposal. Disposal requires
compliance with the Natural Laws—earthenlike materials acceptable to nature’s
standards for assimilation.
Thus, the hierarchy for waste management is simply:

1. Minimization
2. Destruction
3. Treatment
4. Disposal

While technologies may overlap these steps, all are contained within, which
brings us to an important concept: how does natural attenuation fit into the waste
management scheme above? Natural attenuation, or monitored natural attenuation
The Waste Management Hierarchy 41

(MNA), is at the front of waste management schemes for remediation of sites,

coming into favor in the 1990s as a method to employ risk assessment with
source, pathway, and receptor models to decrease active remediation techniques
(and associated costs) and increase passive technologies. Clearly, budgets of
governments and industry cannot support active remediation technologies in
order to return contaminated systems to pristine conditions, and this has been
realized through the use of MNA. In reality, MNA is nothing more than our
understanding of the fifth Natural Law, and the standards set by nature. What we
are observing, understanding, and utilizing in MNA, coupled with active reme-
dies, is simply our quantification of nature’s limits as to what it can assimilate.
Our regulations tie in here with acceptable drinking water or use standards, along
with artificial boundaries placed on problems, such as fence lines and our use
needs. In any case, MNA provides treatment or destruction (reduction in toxicity)
within the four choices for waste management.
Overall, choices for waste management within the hierarchy of minimiza-
tion, destruction, treatment, or disposal are best made on a risk-based approach,
such as that expressed by Watts (7). For a site, or a waste management program
at a facility or other problem, the key elements can be broken down into three
categories—sources, pathways, and receptors. In this manner, a risk-based ap-
proach may be taken by clearly identifying the sources and receptors, and then
testing the pathways for effect, which falls under the realm of chemodynamics,
as discussed earlier. We find then that while government and industry are driven
by regulation and enforcement of waste management options, as with significant
active remediation in the 1980s, the trend is turning strongly now to a risk-based
approach, within the Natural Laws, and by understanding the sources, pathways,
and receptors, and the fate and transport of low-level contaminants in the biota.

REFERENCES
1. W. D. Constant and L. J. Thibodeaux, Integrated Waste Management via the Natural
Laws. The Environmentalist, vol. 13, no. 4, pp. 245–253, 1993.
2. D. D. Reible, Fundamentals of Environmental Engineering, pp. 10–12. Boca Raton,
FL: Lewis Publishers, 1999.
3. L. J. Thibodeaux, Chemodynamics: Environmental Movement of Chemicals in Air,
Water and Soil, pp. 1–5. New York: Wiley, 1979.
4. L. J. Thibodeaux, Hazardous Material Management in the Future. Environ. Sci.
Technol., vol. 24, pp. 456–459, 1990.
5. C. A. Wentz, Hazardous Waste Management. New York: McGraw-Hill, 1989.
6. W. D. Constant, L. J. Thibodeaux, and A. R. Machen, Environmental Chemical
Engineering: Part I—Fluxion; Part II—Pathways. Trends Chem. Eng., vol. 2, pp.
525–542, 1994.
7. R. J. Watts. Hazardous Wastes: Sources, Pathways, Receptors, pp. 38–40. New York:
Wiley, 1998.
4
Legislative and Regulatory Issues

Toni K. Ristau
Public Service Company of New Mexico, Albuquerque,
New Mexico

1 OVERVIEW
In many respects, pollution prevention and waste minimization are less creatures
of legislative fiat than are many other areas related to waste management. In part,
this is due to the way that the legislative and regulatory waste management
framework developed in the United States (1). In the United States, the major
regulatory strategy for addressing wastes, particularly hazardous or toxic wastes,
is a “command-and-control” system imposed upon the regulated community from
the top down. By contrast, many pollution prevention initiatives are voluntary
efforts initiated by companies that seek to improve the “bottom line,” rather than
requirements imposed by a regulatory agency.
To understand the current emphasis on pollution prevention, one must have
an understanding of the history of the regulation of hazardous and toxic wastes.
Now that environmental management is maturing as a discipline, there is an
increasing recognition that pollution prevention and appropriate waste manage-
ment (including minimizing waste streams wherever possible) during a facility’s
operational life can greatly reduce the potential for costly remediation and
cleanup after operations are discontinued.

43
44 Ristau

2 HISTORY
Much of the early legislative effort related to waste disposal or releases of toxic
substances was engendered by incidents such as Love Canal in New York State,
or the release of toxic gas from a factory in Bhopal, India. These incidents, which
were widely reported by the media, outraged the public and caused a demand for
Congressional action (2).

2.1 Love Canal and the Enactment of the Comprehensive

Environmental Response, Compensation and Liability
Act (“Superfund”)
The Love Canal hazardous waste disposal site became the center of attention of
the media, as well as the regulatory agencies, in the late 1970s and early 1980s,
and inspired the passage of the Comprehensive Environmental Response, Com-
pensation and Liability Act (CERCLA), also known as “Superfund.”
The Love Canal site was not originally constructed to be a waste disposal
facility. Originally, the Love Canal was to be the centerpiece of a “model city,”
and the use of the canal for waste disposal was not contemplated. The canal was
originally constructed by William T. Love at the easternmost edge of the town of
Niagara Falls, New York, in 1893. The canal was to be used to supply water to
generate a cheap and essentially unlimited supply of hydroelectric power for this
model community. The discovery and adoption of the use of alternating current
in the mid-1890s, which allowed electricity to be generated at some distance from
the point where the electricity was to be used, rendered Love’s plans for the canal
uneconomic, and Love’s dream for the canal was never realized. The abandoned
canal filled with rainwater and was used as a swimming hole and for winter ice
skating by the local community. In the 1940s, Hooker Chemical Company
obtained rights to the canal, and began to use the old canal as a dump for chemical
wastes from Hooker’s chemical manufacturing operations. Hooker Chemical
drained the old canal, lined it with clay, and used the old canal as a waste dump.
Between 1942 and 1953, an estimated 22,000 tons of chemical wastes, as well as
municipal wastes, were dumped at the site. When Hooker Chemical discontinued
active use of the site, the company capped the old canal with a thick layer of clay,
and covered the entire site with sod (3).
Hooker Chemical sold the dump site to the local board of education in 1953
for a nominal sum (one dollar), on the condition that the company would not be
liable for any problems related to the wastes that were disposed at the site. Though
the board of education was aware that the site had been used for the disposal of
hazardous and toxic wastes, a school was constructed at the site, as were
numerous houses. Though Hooker Chemical tried to stop the development on the
contaminated land, local governmental authorities ignored the warnings, and
allowed the construction at and adjacent to the old disposal site.
Legislative and Regulatory Issues 45

High groundwater levels in the Love Canal area, resulting from unusually
heavy rains and snowfalls during the 1970s, caused an increasingly serious
situation at the old disposal site. Drums and other containers began to surface as
the area over the old disposal facilities subsided, and ponded areas and other
surface waters near the site exhibited high levels of contamination. Residents of
some of the nearby houses noted that the basements were oozing an oily residue,
and there were numerous complaints of noxious chemical odors. An engineering
firm was hired to perform a study of the problems noted by the residents in the
Love Canal area, and to formulate recommendations on how to address the
problems. The engineering company recommended that the canal be covered with
clay, that sump pumps used by nearby residents to prevent flooding of basement
areas be sealed off, and that a tile drainage system be installed to control the
migration of wastes. As these measures would be costly, the city elected not to
implement the engineering recommendations. However, in some homes where the
levels of chemical residues and problems related to noxious odors were found to
be very high, the city had window fans installed.
Despite these minimal efforts by the city to address residents’ complaints,
evidence was mounting that the contamination from the old Love Canal disposal
site was causing more than just inconvenience for the residents of the area. In
March 1978, the New York State Department of Health initiated the collection of
air and soil samples from the homes and other facilities located at and near the
site. The department also conducted a health study of the 239 families who lived
nearest to the old canal. Alarmed at the preliminary results from the study, in
August 1978 the department issued a health order calling for the evacuation of
pregnant women and children under the age of two, recommending that residents
minimize the amount of time spent in the basements of their homes, and also
recommending that residents not eat vegetables and fruits grown in their home
gardens. Residents found themselves in the difficult situation of being unable to
continue occupying their homes, but, because of the increasing publicity regard-
ing the contamination, they were also unable to sell or rent their homes. Shortly
after the issuance of the health order, the State of New York agreed to purchase
the 239 homes closest to the old canal.
In 1979, subsequent to the evacuation of the 239 families living closest to
the old disposal areas, the Love Canal Homeowners Association commissioned
another study (the 239 families who had already been evacuated were not
included in this study). This study indicated that there were increases in miscar-
riages, still births, crib deaths, birth defects, hyperactivity, nervous breakdowns,
epilepsy, and urinary tract disorders in families living in the area. When the
Homeowners Association presented its study findings to state health authorities,
the significance of the findings were downplayed due to potential flaws in the
study methodology. However, the resultant public outcry ultimately caused
action to be taken at the national level. In October 1980, President Jimmy Carter
46 Ristau

ordered a total evacuation of the community. The Love Canal residents had the
option of selling their houses to the government at fair market value, and moving
to a new location.
The public outrage related to the situation in which the hapless residents of
the Love Canal neighborhood found themselves resounded through the halls
of Congress, and Congress responded in 1980 by passing the Comprehensive
Environmental Response, Compensation and Liability Act (42 U.S.C. 9601
et seq.), also known as “Superfund.” Superfund, or CERCLA, provides a mech-
anism for investigating threatened or actual releases of hazardous substances
into the environment, identifying potentially responsible parties, and funding
the requisite technical and engineering studies to address the problems caused
by the hazardous substance release. In addition, the Superfund provides a
means of funding cleanup activities through the imposition of a tax upon petro-
chemical industries.
And what was the final outcome for the Love Canal neighborhood? In 1982,
the U.S. Environmental Protection Agency (EPA) completed studies of the
contamination residing in the soils, water, and air near Love Canal, and initiated
appropriate remedial action. The 239 homes that were located nearest the old
canal were demolished. The remaining homes that were purchased by the
government, and the neighborhood school, were renamed Black Creek Village,
and the sale of the decontaminated homes to new families commenced in 1990.
There are still approximately 22,000 tons of waste buried in the center of the
community; periodic testing of the air, water, and soils in the community assure
the safety of the new residents. Today, the Love Canal neighborhood has been
“recycled,” and Black Creek Village is again a vital, living neighborhood.

2.2 The Incident at Bhopal and the Emergency Planning

and Community Right-to-Know Act
In December 1984, about four years after the enactment of CERCLA, an incident
involving the release of the volatile and highly toxic gas, methyl isocyanate,
occurred in Bhopal, India. Methyl isocyanate gas was produced and used at the
Union Carbide plant there as an intermediate product in the manufacture of
pesticides. The pesticides manufactured at the facility were important in several
ways to the local and national economy; they are and were used to aid nations
such as India to increase crop yields and improve conditions for their populace
by providing a means to control insect pests (4).
When the Bhopal incident occurred in 1984, the gas, which burst from a
tank at the Union Carbide plant, spread over a large, densely populated area near
the plant. Many people died in their beds, and others died trying to escape the
foglike cloud of poison gas. There were thousands of dead and injured in the poor
and crowded neighborhoods near the plant; though local officials were unable to
Legislative and Regulatory Issues 47

immediately determine the total number of deaths, the official death toll was
ultimately computed to be almost 10,000.
Though this incident occurred in India, it garnered much attention
within the United States. An analysis of the factors that caused the inordinate
number of deaths and injuries indicated that much, if not all, of the suffering of
the local population could have been prevented had there been appropriate
emergency procedures and evacuation plans in place. Further, it appeared that
many of the local populace may have been able to protect themselves had they
but known what kinds of substances were being produced at the plant, and what
measures they themselves might be able to take in case of a leak or a release at
the plant.
As was the case when the Love Canal situation was brought to light by the
media, there was a huge public outcry, and demands for legislation to assure that
an incident like the tragedy at Bhopal would never happen in the United States.
Congress was considering amendments to CERCLA at this time, and Congress
responded by adding a new Title III to CERCLA as a part of the Superfund
Amendments and Reauthorization Act (SARA) (42 U.S.C. 11001 et seq.) This
new Title III was also separately titled as the “Emergency Planning and Commu-
nity Right-to-Know Act,” or EPCRA.
EPCRA, unlike the other portions of CERCLA, which are largely oriented
toward cleanup of abandoned hazardous or toxic waste sites, focuses on commu-
nity preparedness and reporting by industrial facilities to assure that national, state
local response authorities, as well as local communities, are aware of the
substances that are being utilized at industrial facilities within the community and
are prepared to respond if there is a release, spill, or leak from such facilities.

2.3 Solid and Hazardous Waste Management and the

Resource Conservation and Recovery Act
In contrast to legislation enacted as a reaction to environmental crises or
catastrophes, the regulation of facilities and activities related to solid wastes
(and of hazardous wastes as a subset of solid wastes), is conducted under
the provisions of the Resource Conservation and Recovery Act (RCRA) (42
U.S.C. 6901 et seq.). The RCRA is designed in substantive part to regulate the
day-to-day operation of solid and hazardous waste management facilities and
activities through a permitting and standards system. The RCRA also contains
provisions related to response from releases from active or inactive waste
management units.
Though the RCRA contains provisions related to both solid waste and
hazardous wastes, much of the regulatory attention in recent years has been on
the hazardous waste component of solid waste streams. A particularly impor-
tant set of provisions in the RCRA gave the EPA the authority to control
48 Ristau

hazardous waste from “cradle to grave.” The EPA thus has regulatory authority
and control over the generation, transportation, treatment, storage, and disposal
of hazardous waste.
In 1984 and 1986, Congress passed major amendments to the RCRA. The
1984 amendments were known as the Hazardous and Solid Waste Amendments
(HSWA). The HSWA required phasing out land disposal of untreated hazardous
wastes. The HSWA also added increased enforcement authority for the EPA,
provided for more stringent hazardous waste management standards, and pro-
vided for a comprehensive underground storage tank program. The HSWA also
provided for corrective action for releases from solid and hazardous waste
management units (both active and inactive) at operational solid and hazardous
waste management facilities.

2.4 The Pollution Prevention Act of 1990 and a New Way

of Managing Hazardous/Toxic Waste Streams
In response to many commentators, who noted that the existing RCRA
and CERCLA regulatory frameworks in many cases provided disincentives to
recycling and other waste minimization activities, Congress passed the Pollution
Prevention Act (42 U.S.C. 13101 et seq.) in 1990. Opportunities for source
reduction as a method of minimizing pollution are often not realized because
the industries responsible for compliance with RCRA necessarily focus on treat-
ment and disposal of the hazardous wastes generated by their processes,
rather than on reducing the overall use of hazardous or toxic chemicals in
their processes.
Unlike the RCRA and CERCLA, which provide at best indirect liability-
driven disincentives to the use and production of toxic and hazardous sub-
stances and wastes, the Pollution Prevention Act attempts to focus public,
governmental, and industry attention on reducing the amount of pollution
produced, by encouraging cost-effective changes in production, operation,
and raw materials use (known as “source reduction”). Source reduction requires
solid and hazardous waste generators to concentrate on fundamental process
changes to prevent waste (particularly hazardous waste) from being generated
in the first place, rather than regarding hazardous waste streams as a necessary
concomitant to industrial production and focusing on the treatment and dis-
posal of that waste. Pollution prevention, as opposed to hazardous waste treat-
ment and disposal, emphasizes the use of production practices that increase
efficiency in the use of energy, water, or other natural resources. Pollution
prevention practices include recycling and internal reuse of waste streams, source
reduction through the minimization or elimination of hazardous or toxic sub-
stances as industrial inputs, and revision of industrial processes to minimize
thermal and other energy losses.
Legislative and Regulatory Issues 49

3 APPROACHING POLLUTION PREVENTION AS A

“SYSTEMS MANAGEMENT” PROBLEM
There has been growing recognition that treating pollution prevention and energy
efficiency as fundamental process inputs, rather than as “add-on” or “end-of-
pipe” systems, is a much more effective way of minimizing the environmental
impact of industrial operations. This approach requires a refocusing of environ-
mental management efforts from a reactive, compliance-based mode to a pro-
active, preventative approach. This model, which recognizes and works within
productive industrial processes, rather than working against fundamental indus-
trial process and adding operational complexity as well as costs, is beginning to
find acceptance not only within industry, but within the EPA and equivalent state
and local regulatory agencies in the United States.
Alternative approaches (such as the systems management approach) have
been increasingly embraced by the international economic community as a
more rational method of assuring that pollution effects are minimized but that
needed industrial growth and development is not hindered. These alternative
approaches are increasingly seen as a way of assuring that industry is not only
economically efficient, but is “environmentally efficient” as well. The new
emphasis on resource conservation and waste minimization accomplishes both
goals—it minimizes the use of raw materials and energy required to maximize
production (and thus lowers production costs), and it minimizes the environmen-
tal impacts from the extraction and production of energy and other raw materials
as well as minimizing the impacts from waste disposal (and thus lowers the
overall environmental impact of industrial development and growth).

3.1 The Environmental Management System Approach

The development and use of an environmental management system within a
company, a facility, or an activity is one method of treating environmental
management as a systems management/systems optimization approach. One such
method that is gaining increasing acceptance in the United States as well as
internationally is the environmental management system development and certi-
fication process embraced by the International Standards Organization (ISO).
The environmental management system standards are codified in the ISO 14000
standards (5).
The standards are developed by internationally based technical committees.
Each nation is free to adapt the standards as appropriate to fit each country’s
unique political and resource considerations. Within the United States, the stan-
dards, once adopted, are modified and implemented through U.S.-based organi-
zations such as

American National Standards Institute (ANSI)

50 Ristau

Underwriters Laboratories (UL)

American Standard Testing Methods (ASTM)
These U.S.-based standard-setting organizations conform the international
standards with U.S. regulatory requirements and assure that the overall objectives
of the standards can be met within the constraints of the U.S. political and
regulatory system.
The environmental management systems approach is beginning to make
headway in the United States. However, the “drivers” for adoption of ISO
performance standards and/or certification are not as well developed as in
other countries, where ISO certification may be a prerequisite for doing business
in that country.
The basic principles of environmental management systems are not unique
to any one type of business or industrial activity, but have applications in all
activities where the environment may be affected. The definition of an “environ-
mental management system” (EMS) is: “ . . . that part of the overall management
system which includes organizational structure, planning activities, responsibili-
ties, practices, procedures, processes and resources for developing, implementing,
achieving, reviewing, and maintaining the environmental policy.” The basic
principles of environmental management systems include:
Integration of environmental issues with other business issues
Looking at the environmental conundrum as an interactive system, rather
than as “add-ons” of discrete activities
The ISO 14000 Environmental Management Standards include several
standards that can be applied in the management of any company’s environmental
aspects of “doing business.” The ISO 14000 substantive standards for environ-
mental management include the following:
ISO 14001, Environmental Management Systems: Specification with Guid-
ance for Use
ISO 14004, Environmental Management Systems: General Guidelines on
Principles, Systems, and Supporting Techniques
ISO 14010, Guidelines for Environmental Auditing: General Principles
ISO 14011, Guidelines for Environmental Auditing: Audit Procedures for
Auditing Environmental Management Systems
ISO 14012, Guidelines for Environmental Auditing: Qualification Criteria
for Environmental Auditors
ISO 14024, Criteria for Certification Programs: Criteria for Self-Certification
and Third-Party Certification
Each company or facility is free to develop its own environmental manage-
ment system. The company selects and develops its own environmental perform-
Legislative and Regulatory Issues 51

ance objectives. In the United States, regulatory agency participation is not

required, but is encouraged. If the regulatory agency chooses to participate, the
agency will provide advice to the company in the setting of overall environmental
performance objectives. Ideally, the company then selects how to meet those
objectives, rather than being subjected to prescriptive control technology require-
ments (however, in the United States, this interactive systematic approach is not
currently allowed for within the existing regulatory framework).
Though the company is free to identify the components of its own environ-
mental management system, certain components must always be present if the
company wishes to seek certification of its environmental management system
under ISO 14000. These required environmental management systems compo-
nents are as follows.
Environmental policy: Senior management must define the corporation’s
environmental policy and ensure that the policy includes, among other
matters, a commitment to continual improvement, to pollution preven-
tion, and to compliance with relevant regulatory requirements.
Planning: The company must establish and maintain a procedure to identify
environmental impacts of its activities, as well as the legal and other
requirements. The corporation must establish and maintain documented
environmental targets and objectives, as well as environmental manage-
ment programs for achieving its objectives.
Implementation: Roles, responsibilities, and authorities must be defined,
documented, and communicated. The corporation must provide appropri-
ate training and must establish and maintain procedures for proper
communication. The company must have proper documentation of pro-
cedures, document control, operational control, and emergency prepared-
ness and response (contingency planning).
Corrective action: The company must establish and maintain documented
procedures to monitor and measure operations and activities that impact
on the environment, and must have documented procedures for investi-
gating nonconformances and implementing appropriate corrective ac-
tion. Procedures must be in place for identifying, maintaining, and
disposing of environmental records, and for periodic audit of the EMS.
Management review: Senior corporate management must review the EMS
on a periodic basis (and document its review) to ensure that the EMS is
suitable, adequate, and effective in meeting the company’s environmental
performance goals.

3.2 Tools Available for Employing Alternative Approaches

The EPA has, in recent years, become more interested in encouraging the
voluntary development of systems approaches to environmental management,
52 Ristau

including pollution prevention versus “end-of-pipe” controls. The EPA now

provides Web-based information and tools for companies or entities that wish to
pursue pollution prevention or environmental management systems initiatives (6).
The EPA’s Office of Enforcement and Compliance Assistance has produced
“Sector Notebooks” for the following industry sectors:
Agricultural Chemical, Pesticide and Fertilizer Industry (1999)
Agricultural Crop Production Industry (1999)
Agricultural Livestock Production Industry (1999)
Aerospace Industry (1998)
Air Transportation Industry (1997)
Dry Cleaning Industry (1995)
Electronics and Computer Industry (1995)
Fossil Fuel Electric Power Generation Industry (1997)
Ground Transportation Industry (1997)
Inorganic Chemical Industry (1995)
Iron and Steel Industry (1995)
Lumber and Wood Products Industry (1995)
Metal Casting Industry (1997)
Metal Fabrication Industry (1995)
Metal Mining Industry (1995)
Motor Vehicle Assembly Industry (1995)
Nonferrous Metals Industry (1995)
Non-Fuel, Non-Metal Mining Industry (1995)
Oil and Gas Extraction Industry (1999)
Organic Chemical Industry (1995)
Petroleum Refining Industry (1995)
Pharmaceutical Industry (1997)
Plastic Resins and Man-made Fibers Industry (1997)
Printing Industry (1995)
Pulp and Paper Industry (1995)
Rubber and Plastic Industry (1995)
Shipbuilding and Repair Industry (1997)
Stone, Clay, Glass and Concrete Industry (1995)
Textiles Industry (1997)
Transportation Equipment Cleaning Industry (1995)
Water Transportation Industry (1997)
Wood Furniture and Fixtures Industry (1995)
The notebook contents, which can be viewed online and can also be downloaded
and printed out, include the following types of information for each industry
sector (7):
Legislative and Regulatory Issues 53

A comprehensive environmental profile

Industrial process information
Pollution prevention techniques
Pollutant release data
Regulatory requirements
Compliance/enforcement history
Government and industry partnerships
Innovative programs
Contact names
Bibliographic references
Description of research methodology
These sector notebooks focus on key indicators, including air, water, and land
pollutant releases typical of the industry sector, and are a good source of
information for managers who wish to implement pollution prevention or envi-
ronmental management programs at their facilities.

3.3 “Project XL” and EPA Regulatory Reinvention Efforts

The EPA has long recognized that many of the media-specific environmental
regulations (i.e., regulations applicable to the pollution of air, water, or land) can
be counterproductive in that imposing more stringent standards or increasing
regulatory scrutiny for one medium (such as air) may minimize the amount of
pollution within that medium, but can actually cause increased releases to other
media. Several years ago, the EPA initiated a regulatory reform effort that allowed
regulated companies to propose multimedia projects meeting certain criteria to
allow trade-offs between media and to enable better environmental performance
overall within a facility. This regulatory reform effort, denominated “Project XL,”
allowed companies and other entities to propose projects that substitute perform-
ance based standards for the prescriptive, one-size-fits-all, standards that are often
imposed through the EPA’s media-specific regulatory programs.
The EPA required that pilot projects proposed by facilities, sectors, and
government agency projects meet the following criteria (8).
Environmental results. Projects that are proposed should he able to achieve
environmental performance that is superior to what would be achieved
through compliance with current and reasonably anticipated future regu-
lation. “Cleaner results” can be achieved directly through the environ-
mental performance of the project or through the reinvestment of the cost
savings from the project in activities that produce greater environmental
results. Explicit definitions and measures of “cleaner results” should be
included in the project agreement negotiated among stakeholders.
Cost savings and paperwork reduction. Projects that are proposed should
54 Ristau

produce cost savings or economic opportunity, and/or result in a decrease

in paperwork burden.
Stakeholder support. The extent to which project proponents have sought
and achieved the support of parties that have a stake in the environmental
impacts of the project is an important factor in the selection of projects.
“Stakeholders” may include communities near the project, local or state
governments, businesses, environmental and other public interest groups,
or other similar entities.
Innovation/Multimedia pollution prevention. The EPA is looking for
projects that test innovative strategies for achieving environmental re-
sults. These strategies may include processes, technologies, or manage-
ment practices. Projects should embody a systematic approach to
environmental protection that tests alternatives to several regulatory
requirements and/or affects more than one environmental medium. The
EPA has a preference for protecting the environment by preventing the
generation of pollution rather than by controlling pollution once it has
been created.
Transferability. The proposed pilot projects are intended to test new ap-
proaches that could be incorporated into other agency programs or in
other industries, or other facilities in the same industry. The EPA is
therefore most interested in pilot projects that test new approaches that
could one day be applied more broadly.
Feasibility. The proposed project should be technically and administratively
feasible and the project proponents must have the financial capability to
carry it out.
Monitoring, reporting, and evaluation. The project proponents should
identify how to make information about the project, including per-
formance data, available to stakeholders in a form that is easily under-
standable. Projects should have clear objectives and requirements
that will be measurable in order to allow the EPA and the public to
evaluate the success of the project and enforce its terms. Also, the project
sponsor should about the time frame within which results will be
achievable.
Shifting of risk burden. In addition to the above criteria, the proposed
project must be consistent with Executive Order 12898 on Environmental
Justice. It must protect worker safety and ensure that no one is subjected
to unjust or disproportionate environmental impacts.
Several Project XL pilot projects were initiated by government agencies, as
well as private industry. Unfortunately, the Project XL program was subject to
heavy criticism from environmental groups as well as from government oversight
entities, and new projects are currently not being fielded.
Legislative and Regulatory Issues 55

REFERENCES
1. A good source of statutory material related to regulation of the environment can be
found on the Web at http://www4.law.cornell.edu/uscode/42/. The regulations associ-
ated with the various laws and statutes are found at http://www4.law.cornell.edu/cfr/.
In addition, the U.S. Environmental Protection Agency (EPA) Website contains
information on environmental laws and regulations that fall under EPA jurisdiction for
compliance and enforcement. The EPA home page, which contains an index and links
to other sources of information, is found at http://www.epa.gov/epahome/rules.html.
2. For a summary of major pieces of legislation related to hazardous and toxic substances
and wastes, see http://www.ehsgateway.com/legislation.html.
3. For a more detailed discussion of the history of the Love Canal, see http://www.
essential.org/orgs/CCHW/lovcanal/lcsum.html and http://enviro.gannon.edu/
EnvRegs/eli/Love%20Canal%20Project.html.
4. For additional information regarding the incident at Bhopal, see http://
www.corpwatch.org/trac/bhopal/factsheet.html and http://www.prakash.org/issues/
environment/disasters/industrial/bhopal/envreu19941204_00.html.
5. The International Standards Organization (ISO) was established in Geneva, Switzer-
land, in 1947. The purpose of the ISO was (and is) to develop uniform, worldwide
standards for the conduct of manufacturing and other industry, business, technical, and
commercial activities. At present, 92 countries belong to the ISO (representing more
than 92% of the world’s industrial production). At the international level, more than
200 technical committees have been developed. The ISO has, to date, established over
8500 standards.
6. The following EPA Website provides common-sense information, links, and tools
for pollution prevention and other environmental management systems approaches:
http://www.epa.gov/envirosense/.
7. To view the sector notebooks online, see http://es.epa.gov/oeca/sector/index.html. For
individual sector information, each listed sector is a “hot link” to more detailed
information for that sector. Also, an excellent set of Web-based resources related to
pollution prevention, waste minimization, and environmental management issues is
maintained by the U.S. Navy. These resources are available to anyone, and are found
at http://enviro.nfesc.navy.mil/p2library/.
8. See, for example, the Website for “Campus Pollution Prevention Information Resources,”
at http://esf.uvm.edu/labxl/Background%20Information/campusp2links.html.
5
Information Systems for Proactive
Environmental Management

Steven P. Frysinger
James Madison University, Harrisonburg, Virginia

1 INTRODUCTION
Environmental computing is a very broad topic, and to some extent defies
taxonomy. However, in the interest of providing an overview of this field,
environmental information systems can be described generally as either environ-
mental management information systems (EMIS) or environmental decision sup-
port systems (EDSS). The former will be defined as systems which provide access
to information, such as records and reports, while the latter include systems which
provide access to tools with which to operate on information in order to arrive at
an environmental management decision. There is clearly a great deal of overlap
between these two definitions, and many systems might straddle the definition
uncomfortably. Nonetheless, it will be useful to address this broad topic in the
context of this dichotomy.

57
58 Frysinger

2 ENVIRONMENTAL MANAGEMENT INFORMATION

SYSTEMS (EMIS)
As corporate environment, health, and safety (EH&S) organizations endeavor to
become world-class operations, they must deal with a remarkable variety of
processes which have evolved over many years, largely in response to regulatory
events. Integration of these various processes into a cohesive and efficient
management system is a necessary part of their quest to make more forward-
looking, comprehensive decisions impacting EH&S stewardship. An integrated
management system has the potential to significantly improve the effectiveness
of the EH&S organization by facilitating awareness of hazards or opportunities
across regulatory boundaries. Likewise, integration can reduce the cost of EH&S
operations by encouraging the sharing of knowledge and effort across these same
regulatory lines. For example, a single activity which maintains inventories of
chemicals stored at company facilities might serve the needs of both Occupational
Safety and Health Administration (OSHA) Hazard Communication and Super-
fund Amendments and Reauthorization Act (SARA) Title III compliance.* Thus,
an integrated environmental management system (used here to describe all
aspects of environment, health and safety management) is a prerequisite to a
world-class EH&S organization.
Integrated environmental management requires integrated environmental
information management. While there is more to environmental management than
information, there is virtually no aspect which does not depend heavily on the
availability and accessibility of correct and current information. Many, if not
most, EH&S processes are themselves essentially focused on information man-
agement, in the form of record keeping, reporting, permitting, or training.
Therefore, it is difficult to conceive of an integrated environmental management
system which does not stand on an integrated environmental information man-
agement system.
While integration of diverse software tools has been going on for a long
time in environmental decision support systems (to be discussed later in this
chapter), the integration called for by EH&S management refers to data rather
than software integration. So-called enterprise resource planning (ERP) systems
have the goal of spanning all aspects of a company’s operations and integrating
data management across organizational boundaries, but even the most widely
implemented ERP systems have only recently begun to include EH&S informa-
tion management. These systems typically require such an intrusive implementa-
tion process that many firms decline to pursue them, preferring instead to develop
their own versions based on existing business models. Unfortunately, EH&S data

*Regulatory references throughout this chapter are based on the laws of the United States.
Information Systems for Environmental Management 59

management is rarely considered in this process because the EH&S organization,

as a cost center, is not perceived to add value to the firm, and therefore rarely
attracts such an investment. The EH&S organization is then left to manage its
data on its own, even though much of the information on which it depends is in
fact owned by line organizations within the company.

2.1 The Need for Integration

The many processes of the typical EH&S organization are usually supported by
as many diverse environmental management information systems, many of them
manual (i.e., with little or no computer support). These information systems have
evolved in response to individual needs, generally without regard to inter-
dependencies between processes and their information management needs.
Apart from the obvious inefficiencies which result from such cir-
cumstances, this ad-hoc structure has resulted in redundant and inconsistent
databases—multiple databases store the same piece of information, and they
sometimes disagree on its value. For example, several EH&S information systems
may use facilities data from different databases which conflict with one another.
This sort of inconsistency ultimately threatens compliance.

2.2 An Integrated Solution

There is an approach which improves the situation by developing the framework
for an integrated environmental information system (IEIS), an important special
case of EMIS. It is important to note that the term “information system,” as
operationally defined here, is much broader than the computer hardware and
software which might support it. It includes a data model incorporating the
structure, definition, and relationships between data elements, as well as the
processes and procedures by which these data are created, modified, used, and
destroyed. While much of this can and should be supported by computer systems,
this fact has little relevance to the conceptual definition of the information system.
Once the IEIS is defined, a systems engineering activity can readily determine
the design and structure of the hardware and software systems which will support
it, about which more will be said later.

2.3 Conceptual Framework

The IEIS approach is predicated on the notion that one can usefully separate data
from the management processes that use them. That is, most or all data of use to
EH&S are descriptive of objects, while the various management processes
undertaken by EH&S professionals are focused on these objects. An object-
oriented approach to EH&S information might start with the definition of such
high-level objects as employees, customers, buildings, vehicles, services, and
60 Frysinger

products. Each of these can then be decomposed in a similar fashion, as appro-

priate, with the terminal objects described by a data structure.
The various EH&S management processes can generally be viewed as
operating on the data objects suggested above. For instance, SARA Title III
Section 312 reporting is focused (by regulation) on buildings, while OHSA
training requirements are focused on employees. Furthermore, each process may
be supported by one or more software applications. In general, the software
applications serving EH&S processes are the agents which interact with the data
required for these processes (Figure 1).
Thus, there is envisioned a clear separation between data, processes, and
applications:
1. A datum may be used by multiple processes; e.g., Building Address is
used for SARA Title III reporting and for OSHA accident reporting.
2. A process may be served by multiple applications; e.g., one software
application might support the SARA inventory maintenance activity by
site personnel, while another application is used to generate the SARA
reports.
3. In some instances, applications may be used by multiple processes;
e.g. the software used by site personnel to maintain chemical invento-
ries may serve the purposes of both SARA and OSHA compliance
processes.

Data Data … Data

Object 1 Object 2 Object L

Software Software Software

…
Application 1 Application 2 Application M

Process 1 Process 2 … Process N

FIGURE 1 An exemplary relationship between data, processes, and software

applications.
Information Systems for Environmental Management 61

In essence, this approach addresses our need to understand this relation-

ship between our information and our processes so that we may ensure the
availability of the correct data and the correct software applications to interact
with those data.

2.4 The Path to Integration

There are four essential steps to achieving an integrated environmental informa-
tion system:
1. Develop an integrated data model.
2. Map the integrated data model onto corporate databases of record.
3. Define high-level requirements for the IEIS.
4. Implement the foundation of the IEIS.
While some of these can be executed concurrently, it is imperative that we
recognize the precedence implicit in their ordering. As with any systems engi-
neering activity, in this activity the what has to lead the how, rather than the other
way around. It will be advantageous to look ahead to current and future system
implementations to help us to achieve an understanding of requirements, but
particular discipline must be applied to prevent us from erroneously finding a
requirement in what is merely a habit. This discipline will be encouraged by a
phased approach, in which we first define an IEIS for the set of processes as they
currently exist, admitting that the model will be revisited as a result (and indeed
in support of) efforts to reengineer those processes.

2.5 Model Development

The first step in the project is the development of an integrated data model which
correctly describes the firm from an EH&S point of view. The initial (baseline)
data model must include all data items required by the current set of EH&S
processes, but must be orthogonal to these processes so that data objects and fields
which are common to multiple processes occur only once in the data model, to
be shared by the processes requiring them. This is critical to the identification of
shared information and the elimination of redundant databases. Once such a
baseline data model has been developed, it can and should be refined and revised
as appropriate to reflect the ongoing reengineering of the EH&S organization’s
structure and processes.

2.6 Mapping the Model onto Databases

The integrated data model so developed will then be analyzed to determine
the appropriate owner for each of the data categories and elements. In many
cases, this will be the so-called database of record for the company, and will
not be under the control of the EH&S organization. For example, much informa-
62 Frysinger

tion about corporate facilities might be maintained by a real estate organization

within the firm but outside of EH&S. Identifying our stake in such external
databases is essential since, as customers of these databases, we will need to
be recognized and have a voice in the implementation and management of the
data. There may also be data items of importance to EH&S which should and
could readily be maintained in these external databases; we will want to be in a
position to lobby the appropriate organizations for such extensions. Furthermore,
interfaces to these data sources must be engineered so that the data truly will be
shared, rather than simply copied into yet another system, further contributing to
data redundancy.

2.7 Defining IEIS Requirements

The third step is the definition of high-level requirements for the integrated
information system. The integrated data model and analysis described above form
the foundation for this. What must be added are the functional requirements for
the integrated system. For example, if EH&S information must be globally
accessible by EH&S leadership, this requirement should be articulated clearly.

2.8 Implementing the IEIS Foundation

The fourth step addresses the implementation of the IEIS. Implementation
includes the interaction and negotiation with other organizations whose informa-
tion assets have been identified as a subset of the EH&S data model in step 2.
It also includes the planning and acquisition and/or development of software
required to realize the IEIS from the starting position of our existing information
management systems. The result of this step is not necessarily a single software
system; in fact, this outcome is highly unlikely, given that the software to be used
by individuals and groups engaged in the various processes will have to satisfy
functional requirements which may be peculiar to those processes. As long as the
ensemble of computer systems finally in use by the EH&S organization (a) im-
plements the integrated data model developed in steps 1 and 2, and (b) satisfies
the high-level requirements defined in step 3, then we will have achieved an
integrated environmental information system and will reap the benefits thereof.
This, perhaps, is the point of departure of this approach from conventional
thinking about integration—we seek to achieve the benefits of integrated infor-
mation while valuing diversity of software applications and vendors.
Once these four steps have been executed, the design and implementation
of the integrated system using an appropriate combination of existing and new
platforms can proceed through conventional information project management and
systems engineering activities. In fact, it might be hoped that through effective
communication, any ongoing procurement and development activities underway
during the execution of these steps can be appropriately guided so as to minimize
Information Systems for Environmental Management 63

changes or disruption once they are complete. For example, an early intermediate
result will be the identification of data common to the first key processes to be
evaluated. This knowledge can surely be used during the procurement of support-
ing systems to anticipate the results of the integration effort.

2.9 EMIS Summary

This approach to integrating environmental management information systems
into an integrated environmental information systems serves to illustrate the
issues attending these systems in general. Whether this approach or some other
is used, however, the critical element for proactive environmental management is
that integration be achieved in the interests of eliminating compliance-threatening
redundancy and removing substantial inefficiencies.

3 ENVIRONMENTAL DECISION SUPPORT

SYSTEMS (EDSS)
As the complexity of our environmental management problems has increased, so
has the need to apply the information management potential of computing
technology to help environmental decision makers with the difficult choices
facing them. Environmental information systems have already taken many forms,
with most based on a relational database foundation (as described in the previous
section). Such systems have helped greatly with the day-to-day operations of
environmental management, such as chemical and hazardous waste tracking and
reporting, but they have two critical shortcomings which have prevented them
from significantly improving the lot of environmental scientists and planners
tackling more strategic decisions.
Traditional environmental management information systems generally ig-
nore the crucial spatial context of virtually all environmental management
problems, and they offer little or no support for the dynamics of environmental
systems, both manufacturing and otherwise. Fortunately, a relatively new cate-
gory of system, called an environmental decision support system (EDSS), shows
real promise in both of these areas.

3.1 What are Environmental Decision Support Systems?

Environmental decision support systems are computer systems which help
humans make environmental management decisions. They facilitate “natural
intelligence” by making information available to the human in a form which
maximizes the effectiveness of their cognitive decision processes, and they can
take a number of forms (1).
As defined here, EDSSs are focused on specific problems and decision
makers. This sharp contrast with the general-purpose character of such software
64 Frysinger

systems as geographic information systems (GIS) is essential if we are to put and

keep EDSSs in the hands of real decision makers who have neither the time nor
inclination to master the operational complexities of general-purpose systems.
Indeed, it can be argued that most environmental specialists are in need of
computer support which provides everything that they need, but only what they
need. This point becomes more critical when it is understood that many important
“environmental” decisions in design and manufacturing, for example, are not
made by environmental specialists at all, but are instead made by professionals
in other disciplines.

3.2 The Need for Environmental Decision

Support Systems
The development of environmental policies and generation of environmental
management decisions is currently, to a large extent, an “over-the-counter”
operation. Technical specialists are consulted by decision makers (who may or
may not have a technical background), to assist in gathering information and
exploring scenarios. Because of the inaccessibility of data and modeling tools,
decision makers must consult their technical support personnel with each new
question, a time-consuming and inefficient process.
If the data and analytical tools could be placed within reach of decision
makers, they would be able to consult them more readily, and would therefore be
more likely to base their decisions on a technical foundation. In some instances,
the availability of environmental decision support determines whether or not a
product design or manufacturing process will indeed be “environmentally con-
scious.” This is the premier reason why environmental decision support systems,
of a sort described in part herein, are necessary if we are to achieve higher quality
in our environmental management decisions and obtain more protection with our
finite resources.

3.3 Foundations
Environmental decision support systems address a problem domain of remarkable
breadth, ranging from selection of an appropriate light switch for an automobile
to the determination of community risk associated with stored chemicals. The
character of environmental decisions and their surrounding issues is central to the
design of a successful EDSS.

3.4 The Nature of Environmental Management Decisions

To understand environmental management decisions, we must first identify the
decision makers. The stereotypical image of an environmental manager is an
environmentally trained business manager given the responsibility for avoiding
Information Systems for Environmental Management 65

fines and other sanctions, and perhaps pursuing “beyond compliance” goals, all
within the constraints of finite (and generally tight) budgets. Indeed, many
environmental decision makers fit this description.
However, these individuals also have their counterparts in the regulatory
arena (such as agency compliance officers). Furthermore, critical environmental
decisions are often made by market researchers, product designers, and manufac-
turing process developers. Naturally, the level of environmental expertise these
individuals possess is highly variable. Nonetheless, all of them can and do make
critical environmental decisions. It is therefore incumbent upon the toolbuilders—
including EDSS architects—to craft systems and processes that will help to bridge
the gap between technical expertise and the decision maker, so that the benefits
of this expertise may be realized.

3.5 Characteristics of the Problem

Environmental decision makers are clearly a diverse group of people faced with
a diverse group of problems. The breadth of their problem domain, in fact, defines
the need for eclectic individuals with tools to match.
In general, environmental decision problems are
Spatial, in that most human activities and their environmental impacts are
associated with a place having its own characteristics which influence
the decision
Multidisciplinary, requiring consideration of issues crossing such seem-
ingly disparate fields of expertise as atmospheric physics, aquatic
chemistry, civil engineering, ecology, economics, geology, hydrology,
toxicology, manufacturing, materials science, microbiology, oceanogra-
phy, radiation physics, and risk analysis
Quantitative, because the constituent disciplines themselves are highly
quantitative, and because the costs and ramifications are generally so
significant, that objective metrics are desired to help mitigate controversy
Uncertain, in that while the elements are quantitative, the sparsity of data
and nascent state of the constituent disciplines leaves many unknowns
Quasi-procedural, since many environmental decisions are tied to a regu-
latory or corporate policy framework which specifies the steps by
which a decision is to be reached, and because the threat of liability
dictates a defensible audit trail for the decision process
Political, reflecting the fact that environmental management is driven by
public policy, influenced by such considerations as economics, social
impacts, and public opinion
The diversity of these characteristics of the problem domain make effective
environmental decision support extremely challenging.
66 Frysinger

3.6 Implications for Environmental Decision Support

Because of these factors, it is not practical to contemplate a generic decision
framework for environmental management. Even if it were possible to capture all
of the elements necessary to address the great variety of decisions to be under-
taken, the system so built would be virtually unusable. Environmental managers
are already confronted with a vastly complex problem space; one of the first jobs
of the decision support system is to simplify this space, offering them everything
that they need to make the decision at hand—but only those things.
Therefore, while our definition of EDSS includes the integration of multiple
supporting technologies (such as simulation and GIS), we further restrict this
definition to stipulate that EDSSs are focused on a particular decision problem
and decision maker. Thus, they are not general-purpose tools with which anything
can be done—if only you knew how to do it. Rather, they are particularly tailored
to the problem facing the analyst, and offer a user interface which is optimized
for this problem.
The focused nature of such EDSSs improves the user’s interaction with the
computer system, allowing the user to concentrate on the problem at hand and
the information and tools needed to solve it. It also dictates a software architecture
that facilitates the development of sibling systems embracing different decision
problems with an essentially common user and data interface (2). Such a family
of focused EDSS siblings offers user interface simplicity, in that the siblings share
interaction style, organization, and fundamental approaches (where appropriate),
while maintaining the focus each sibling has on its particular decision problem.

3.7 Task Analysis of Environmental Decision Making

The focused approach to EDSS design advocated here dictates the use of a human
factors engineering technique, called task analysis, to support the specification of
a particular EDSS for a particular problem.
As defined in the human factors community, “task analysis breaks down
and evaluates a human function in terms of the abilities, skills, knowledge and
attitudes required for performance of the function” (3). The EDSS designer
must endeavor to understand the decision problem, and all of the factors
which must be considered in solving it. In addition, the “social history” of the
problem must be understood, since there will (in general) already be a number of
different approaches to solving a given environmental management problem. For
a system to support an analyst in arriving at a credible decision, the various
competing approaches must be considered, and possibly accommodated.
A major stumbling block in task analysis is the fact that very few individ-
uals can accurately explain the way in which they actually arrive at a particular
decision. They can tell you how they think they should do it, and they can often
develop a post-hoc analytical rationale for their decision, but people are generally
Information Systems for Environmental Management 67

unaware of the actual process by which they make decisions. Thus, other
instruments must be used to understand the decision process, ranging from
observation and interview up through controlled experimentation to determine the
influence of different variables on decisions.
In the environmental arena, this is further complicated by the fact that there
are often guidelines or regulations dictating the way in which decisions are
supposed to be made about a particular problem. These do indeed dictate certain
aspects of the process, but often leave a great deal unspecified. For example, the
U.S. Resource Conservation and Recovery Act (RCRA) requires that a waste
facility be monitored by a network including at least one upgradient and three
downgradient wells in order to assure that no hazard to the public health results
from the facility. However, though the legislature was specific about this detail,
it made little effort to assist the manager in deciding where or how many (above
four) wells are to be installed. Furthermore, the language of the act would suggest
that certainty is required with respect to the detection of leaks, though no
reasonable person would argue that this is either theoretically or economically
achievable. Implicit in this example is the issue of uncertainty, which, because of
its importance in environmental management, deserves further attention.

3.8 Management of Uncertainty

Uncertainty is implicit in environmental decision making. Complex technical
decisions must be made regarding events—in both the past and the present—
which depend on many different variables. Solutions to such problems often
depend on the use of various mathematical modeling techniques. These tech-
niques, in the main, attempt to predict the future performance of a complex
system on the basis of relatively sparse empirical data. The predictions drawn
from these modeling studies form the basis for the entire process to follow,
including such expensive decisions as the design of a product and its associated
manufacturing processes. Ultimately, the environmental effectiveness of the
product throughout its life cycle, in terms of protection of human health and
reduction of environmental risk, depends on these results.
However, these modeling studies are unavoidably visited by uncertainty of
various types, ranging from conceptual model uncertainty—associated with the
selection of assumptions necessary to choose the model(s)—to parameter uncer-
tainty resulting from sparse empirical data, noisy measurements, and the general
difficulty associated with measuring critical parameters.

3.9 Sources of Uncertainty

Uncertainty in such environmental management problems exists because of a lack
of empirical data, errors in the data, incorrect models, and the general non-
determinism of nature.
68 Frysinger

The first of these, a lack of empirical data, is easy to understand; we

routinely live with imperfect knowledge of the current state of systems, owing to
lack of data (in a usable form). This and the second (errors in the data) are the
ones typically addressed in scientific and engineering studies when the goal is to
reduce uncertainty. The usual approach is to collect more data, and to attempt
to reduce the measurement error in the data collected.
The third reason, the use of incorrect models, is recently receiving more
attention in environmental management. As environmental managers come to
accept that model building (whether mental or mathematical) is an essential part
of problem solving, the disagreements as to which models are correct become
more apparent. Some would argue that a model is correct to the extent that it
accurately predicts the future behavior of the system; the limiting factor for
environmental problems is the complexity of the system in question. And here is
where an interesting human factor emerges.
As mathematical models are expanded to attempt to account for more of
the fine details of the natural system under study, the mental models of the analyst
become inadequate. While humans are capable of recognizing and apprehending
in a gestalt sense the breadth of complex systems, they are ill equipped to
mentally manage the myriad simultaneous details attending such systems. It can
be argued that we build mathematical models precisely because we cannot
manage such details mentally. Yet, as we build these models, they too become
more complex than we can fully grasp, resulting in a great deal of effort and
controversy associated with the development of the mathematical models. Many
environmental modelers spend more time studying their models than studying the
natural systems they emulate.
This problem becomes especially acute when the decision maker is not the
developer of the mathematical model, because an opportunity exists for mismatch
between the analyst’s mental model and the quantitative mathematical model he
or she is attempting to use. This results in uncertainty, both subjective (i.e., lack
of confidence on the part of the analyst) and objective (i.e., a measurable
variability in the decisions made by several analysts or by one analyst on several
occasions). Ultimately, this uncertainty finds its way into public perception,
causing the public at large to wonder how to interpret the products of science and
engineering (the public’s awareness of the modeling debate surrounding global
warming is a good example of this).
Finally, the fourth cause of uncertainty in environmental problems arises
out of the nondeterministic character of the natural environment, at least as it is
currently understood. We should not expect to eliminate uncertainty entirely in
solving environmental problems. Like the other three, this cause of uncertainty
applies to both spatial and aspatial data, and some adaptive approaches have been
proposed to help analysts arrive at accurate descriptions of the uncertain natural
parameters (e.g., Ref. 4).
Information Systems for Environmental Management 69

Unfortunately, humans tend to have some difficulty in reliably making

probabilistic judgments (5). There is a tendency toward a “fish-eye” view of
uncertainty, in that perception of unfamiliar issues or events is related to familiar
ones, resulting in distortion not unlike the familiar cartoon maps showing “the
New Yorker’s view of the World.” This is evident in studies examining human
perception of risk, and applies to probabilistic judgments more generally.
Quantification of uncertainty has been widely acknowledged as a critical
issue in risk assessment (see, for example, Ref. 6). A variety of methods for
managing uncertainty have been studied (7), most of which are beyond the scope
of the present chapter. One of these, which figures prominently in EDSS, involves
the use of computer simulation methods to quantify the uncertainty associated
with a model result, conditioned on the correctness and appropriateness of the
model for the problem at hand.

3.10 Stochastic Analysis

In considering the uncertainty of quantitative models, one considers the output of
the model to be some function of one or more input coefficients. These co-
efficients become the parameters of a numerical representation of the model. The
quantitative uncertainty in the modeling solution, then, results from the combined
uncertainties of the input parameters.
Stochastic analysis of uncertainty is predicated on the ability to articulate
the probability distributions of each uncertain parameter and then iteratively solve
one or more model equations involving these parameters. To accomplish this,
samples are drawn from the parameter distributions, most often employing Monte
Carlo or Latin hypercube sampling methods.
To generate N Monte Carlo samples from a given probability distribution,
one first produces the corresponding cumulative distribution function (CDF). The
ordinate of the CDF, which ranges from zero to one, is then sampled uniformly,
and the corresponding abscissa values are taken as pseudo-random samples of the
target distribution.
Latin hypercube sampling, a variation on the Monte Carlo method, forces
the uniform samples drawn on the ordinate to cover the entire range (zero to one)
by dividing the axis into N equal-width bins. From each bin a sample is drawn,
with uniform sampling within each bin. This modification helps to ensure that the
tails of the target distribution are sampled, and therefore can result in convergence
on the target distribution in fewer samples than the unmodified Monte Carlo
method.
To solve environmental models using such stochastic methods, one solves
the model equation iteratively, each time using parameter values drawn from the
uncertain parameter distributions by the methods just described. The set of results
of these calculations form, themselves, a distribution which aggregates the
70 Frysinger

uncertainty of each of the parameters, and whose characteristics can be used to

describe the model. The moments and upper and lower quantile bounds of such
a calculated distribution can be employed directly in decision making based on
the model. For example, if one calculates individual exposure to radionuclides
using such an approach, the CDF of the distribution of results can be used to
find the probability that exposure will exceed 25 mrem/year. It has been demon-
strated (8) that the use of such methods can help to avoid the “creeping
conservatism” which often results from the use of upper-bound parameter values
alone to model risk.

4 CONTRIBUTING DISCIPLINES
Several disciplines interact with and are integrated by environmental decision
support systems as defined in this chapter. This section will introduce the most
prominent of these, with a special focus on the particular areas of intersection and
contribution. This treatment cannot be construed as a fair representation of any
of these disciplines as a whole; rather, it is intended to provide a sense of the
interdisciplinary nature of EDSS, and to illuminate some of the opportunities for
interdisciplinary research associated with EDSS.

4.1 Environmental Science

Environmental science is itself an interdisciplinary field, integrating biology,
chemistry, mathematics, and physics in the context of environmental protection
and management. There is a distinctively applied, anthropocentric orientation to
environmental science; it differs from such fields as ecology in that it approaches
the study of our environment with an eye toward human needs and use of the
environment, and therefore addresses the science, engineering, and management
practices which will help to conserve environmental resources for human benefit.
This is not to imply that environmental scientists as a whole do not place value
on nature in and of itself, but that their professional lives are more focused on
natural resource protection, where the word resource refers to human needs and
wants. This distinction is significant for the present EDSS discussion only
because, as a practical matter, nearly all environmental decisions are anthro-
pocentric. Even in the relatively rare cases where economic resources are avail-
able for “pure” ecological protection or remediation, the decisions made must
necessarily consider cost/benefit as best they can in order to justify the use of the
limited funds. Therefore, worth is an important element of virtually every
practical environmental decision, and its analysis is most definitely in need of
assistance from EDSS technology.
The contributions of environmental science to EDSS begin with the basics.
In some instances, we are interested in the basic science involved, with no
Information Systems for Environmental Management 71

particular environmental twist, such as the solubilities of chemicals in water, the

partitioning of a chemical between the vapor or aqueous phases, the chemical
equilibrium of carbon dioxide and water, or the physics of radioactive decay. In
others there is a distinctly environmental angle, such as the adsorption of
chemicals on soil particles, or the avian toxicity of a pesticide. The line between
these two cases is blurred, which is one of the reasons that the basic sciences are
so readily integrated into environmental science pedagogically.
Of special interest to EDSS are environmental science’s contributions in
mathematical modeling of environmental processes. In this context, environmen-
tal science integrates such disciplines as geography, hydrogeology, and meteorol-
ogy, along with various associated engineering disciplines, notably civil and
chemical engineering. In some fields, mathematical models are employed to help
discover the truth about the phenomenon under study, with the (usually optimis-
tic) goal of arriving at the model which describes the way the process works. In
contrast, environmental scientists develop models primarily in order to accurately
predict the future (or sometimes past) behavior of the system, without suffering
the delusion that the model works the same way the system does. Model
fidelity—the degree to which the model reflects the way the system actually
works—is usually of secondary concern in environmental science. Model robust-
ness—the degree to which the model predicts system behavior under varying
conditions consistent with the stated assumptions—is of primary concern.
The focus of environmental modeling is prediction, useful because it can
help us to understand what has happened, or what will happen. Such models are
central to environmental decision support systems, and in fact to environmental
decision making in general. Though some environmental managers would profess
to distrust models, and prefer to make predictions through some other means, they
fail to realize that these other means invariably include mental models of the
system. Mental models may not be mathematical, but they are most certainly
models, and bear all of the constraints that apply to models.
These constraints can nearly all be reduced to one axiom: a model is only
as good as the assumptions that accompany it. In the case of environmental
models, significant assumptions are always needed in order to apply a particular
model to a particular situation. Assumptions could arise in an attempt to cope with
uncertainty in future events (such as the number of inches of rain that will fall
next year), or in an attempt to simplify the problem to make it more tractable
(such as modeling groundwater contaminant transport in two dimensions rather
than three). Assumptions in environmental models are not bad; indeed, they are
necessary. However, they must be made and validated consciously during model
building, and not forgotten when the model is applied. Part of the role of EDSS
in the application of environmental models is to help the decision maker to
acknowledge, and to an appropriate extent participate in, the assumptions made
72 Frysinger

and validated. In some systems, this is accomplished by requiring analysts to

explicitly state their assumptions respecting the models to be applied.
Another multidisciplinary grouping, drawn from the health sciences, can
be included in environmental science in this context, although it is not tradition-
ally grouped together in an academic environment. Health science is here
taken to include various branches of medicine, toxicology, and epidemiology.
These disciplines provide crucial information regarding the ultimate human
health ramifications of the systems or actions under study. For example, this
would include the first phases of risk assessment, wherein the relationships
between human exposure and human health effects are explored and described.
Like other aspects of environmental science, this (collective) discipline also
contributes models to environmental decision support. These models, both ana-
lytical and empirical, assist with such tasks as dose–response calculation and
uptake prediction.

4.2 Information Systems Engineering

Information systems engineering (meant here to include computer science and
its kin) is also a multidisciplinary field. Not surprisingly, information systems
engineering and several of its associated technologies plays a key role in
environmental decision support systems. We will explore four of these which are
of particular importance to EDSS.

4.3 Geographic Information Systems

A central feature of virtually all environmental decisions is their spatial context.
Geographic information systems (GIS) are computer software systems which
directly target the management, analysis, and display of spatial information, and
which are therefore crucial in an effective EDSS.
There are many GIS packages available, differing in the details of their
design. However, some key design features are common to virtually all commer-
cial or public-domain GIS offerings. (A more complete introduction to geographic
information systems may be found in Ref. 9.) Current GISs represent spatial
information as layers of two-dimensional data encoding different spatial data
elements, analogous to (and in fact derived from) the traditional mapmaker’s
technique of drawing different map features on separate layers of transparent
material. These layers can then be overlaid in whatever combination is desired
to produce a map showing those features which are of interest. For example,
one might overlay a property (lot/block) map onto a soils map in order to evaluate
the soils present in individual lots for septic suitability analysis. These two-
dimensional layers are typically managed as one of two data types, vector and
raster. Early in the history of GIS, packages would use either one or the other of
these two data formats, but they are now both supported in common GIS products.
Information Systems for Environmental Management 73

The vector data format, as its name implies, represents spatial objects (such
as building lots or soil regions) as polygons formed by sequences of vectors, or
line segments, each of which is in turn represented by its endpoints (in whatever
reference system, such as latitude/longitude, is convenient). Some spatial objects
(such as roads or rivers) are represented simply as vector sequences which do not
close into polygons. Finally, some objects (such as drinking-water wells) may be
represented as a single point. While the structures discussed above represent the
location of the spatial objects, they do not describe the attributes of the objects.
Such attributes are typically represented in a relational database which is linked
to the spatial description by an identifier field. Thus, if one selects the polygon
representing a soil region—for example, by clicking the mouse within that
region—the GIS would first determine the identifier of the polygon which
contained the mouse pointer, and then use this identifier to extract attribute
information (in this case soil classification) from the relational database. In fact,
when the spatial objects are drawn on the computer screen, one or more of the
attribute fields can be used to determine such drawing options as line color or
type, or polygon fill color or pattern. In this way a color-coded soils map can be
displayed, at the same time that the information used to produce it is available to
other computer software. Foremost among the virtues of the vector approach
to spatial data representation is the fact that the points (which are the building
blocks of all types of spatial objects) can be expressed with a level of precision
limited only by the computer’s number representation. (Of course, this has no
bearing on the accuracy of the data so represented.)
The raster data format takes an entirely different approach to spatial data
storage. Data layers are represented as regular matrices, with the (normally
square) cell dimensions determining the resolution of the layer. The name raster
is related to the raster display of modern cathode-ray tube (CRT) displays, which
are composed of rows and columns of pixels. However, there is no actual
correspondence between a GIS raster layer and a CRT’s pixels: the data in one
cell of a GIS raster layer can be drawn using one or more CRT pixels. In a raster
representation of a soils map layer, each cell of the raster contains a value
corresponding to the soil category within that cell. If the cell dimension is, for
example, 30 m, then the soil category assigned to the cell is that of the soil which
dominates the 30 m × 30 m area represented. It is obviously quite a simple matter
to display a color-coded soils map by mapping a raster’s cell values onto the video
memory’s pixel values through a color lookup table. This results in display
operations which are somewhat faster than can be achieved with a vector
(polygon) display. Alternatively, the raster layer’s cells can contain key values
providing connectivity to a relational database, similar to vector systems, al-
though this approach is used less often. In any case, the precision of the spatial
representation using raster data structures is limited by the data storage available
for each raster. If one wanted 10-m rather than 30-m resolution (supposing one
74 Frysinger

had corresponding information resolution), the space required to store the layer
would increase by a factor of 9.
The chief advantage of a raster data structure is the ease with which one
can perform calculations oriented toward the intersection of two or more layers.
For example, if one defines septic-suitable areas as those which have a sandy
loam soil and a slope of less than 10%, one can produce a new layer by
performing a cell-by-cell comparison of the soils layer with a slope layer (which
itself could be produced by analyzing an elevation layer).
Such calculations are common in natural resource management, which has
resulted in raster-oriented GISs dominating these fields. On the other hand, in
areas where precise locations are important (such as tax maps or pipeline
location), vector-oriented GISs have dominated. Since most GIS packages have
migrated into a hybrid orientation, supporting both data structures and conver-
sions between them, one no longer has to make the choice when purchasing the
software, and can choose the structure appropriate for the problem at hand.
As was hinted during the foregoing discussion, GIS technology includes
more than the simple storage and display of map layers. A critical component of
GIS is the analytical suite which permits calculations, comparisons, and manipu-
lation of data layers to produce either new derived layers or simple answers. For
example, given a soils layer, any competitive GIS can very simply report the area
represented by a particular soil type, either as a percentage of the whole or in such
units as acres, hectares, or square meters. Likewise, most GIS packages permit
more sophisticated spatial statistics, such as the generation of rasters by interpo-
lation of contour maps, or conversely the generation of contours from rasters.
Such analytical capabilities differentiate GIS packages from more simple map-
ping packages.
These analytical capabilities have increasingly permitted GIS technology to
be the basis for decision making in many contexts, not the least of these being
environmental management. GIS capability is now a standard in nearly all
organizations undertaking environmental analyses, with the useful side effect that
many sites of interest have already compiled significant repositories of GIS data
pertinent to their problems. However, GIS largely remains an over-the-counter
operation. Because of relatively complicated user interfaces, exacerbated by
rather breathtaking secondary memory (disk) storage requirements for GIS data,
most organizations maintain something along the lines of a GIS department or
group. Decision makers, if they recognize that they have a problem which can be
addressed by GIS methods, approach this group with the problem description and
enter the group’s service queue. For some problems, this sort of specialist
attention is necessary. GIS groups tend to be staffed by individuals with consid-
erable knowledge of cartography and the tricks necessary to manipulate map data
without corrupting it. On the other hand, a good deal of GIS capability is in
principle within the grasp of workers from other fields, but the tools and/or data
Information Systems for Environmental Management 75

themselves are not available. In integrating GIS technology into environmental

decision support systems, we attempt to address the latter problem, not the former.
For the subset of GIS-tractable problems which can be approached by the
non-GIS specialist, integration into a decision tool addressing their larger prob-
lem will solve the batch-oriented, over-the-counter bottleneck which more often
than not results in GIS methods not being used where they might otherwise be
put to good effect. Another way to think of this is to consider that EDSS can bring
some elements of GIS to decision makers in such a way that they need not know
it is GIS.

4.4 Computer Data Representation

While the geographic information systems technology just described goes a
long way toward providing display capabilities for environmental management
problems, it does not satisfy all such needs. First of all, GIS displays are
overwhelmingly two-dimensional in nature, with a strong bias toward represent-
ing data in map format, or “plan view.” Many GIS packages also provide a
“2.5-dimensional” representation capability wherein map layers containing ele-
vation information in the raster cells are drawn as surfaces from a user-specified
perspective. While often useful, such displays are not by themselves adequate.
For many environmental management problems, true three-dimensional
displays are helpful. For example, when evaluating the behavior of a modeled
airborne contaminant plume, the analyst should be able to navigate about (and
through) the three-dimensional plume in order to get a better feel for its shape
and character; contour plots fail to communicate this information. Computing and
displaying such volumetric renderings rests squarely within the domain of infor-
mation systems engineering. The algorithms required to efficiently draw, shade,
and cast virtual light upon three-dimensional objects drawn on a two-dimensional
computer screen are the result of considerable research in the field of computer
graphics. Many of these algorithms have been known for quite some time, but
the ability to use them to generate very sophisticated volumetric displays in
near-real-time is relatively new, especially on common desktop computing hard-
ware. These tools have begun to play an important role in environmental decision
support, and will be integrated into EDSS platforms with increasing frequency.
More recently, however, advances in personal computing have included the
development and widespread dissemination of what has been called multimedia
technology. This suite of computer capabilities has added photographic and
motion-picture display to the more conventional computer graphics world, and
has also added high-fidelity sound-generation capability to the platform. The
ability to include photographs (such as site familiarization photos) and videos
(such as a sequence capturing the removal of a well core) has the potential to
greatly enhance the information delivery potential of environmental decision
76 Frysinger

support systems. The audio capabilities have an obvious use in delivery of speech
(such as in online help or cooperative work situations), but also support the use
of sound to represent quantitative information which cannot readily be accommo-
dated by available visual display channels (e.g., Refs. 10 and 11).

4.5 Supercomputing and Networking

A third area of information systems engineering which has the potential to
significantly impact EDSS design relates to the execution of computationally
intensive models. Historically, such computations have been relegated to a
segment of computer technology called supercomputers. Supercomputers may be
operationally defined as computers which are both fast and expensive enough that
few of them are in existence. This rather awkward definition is necessary to
account for the fact that current personal computers offer a level of computational
throughput which would have been considered supercomputing 25 years ago. It is
pointless to attempt to define supercomputers in absolute performance terms,
because the technology advances so rapidly as to render such boundaries obsolete
in very few years.
Nonetheless, it may be presumed that no matter how fast individual
workstations become, there will be still faster computers which are few in number
but which are made available to a wide population. In this work, such super-
computing technology is considered in combination with digital networks be-
cause high-bandwidth data networks have made it possible to consider linking
supercomputers with personal workstations in such a way that the interactive user
need not be aware that computations have been “contracted out.”
In some sense, this sort of approach would represent a distribution of the
EDSS architecture across multiple, remotely located machines. This view is
especially appropriate if one distributes the data or code repository functions as
well. For example, one might keep national meteorological data in a disk farm
associated with a National Oceanic and Atmospheric Administration (NOAA)
supercomputing facility, which might also store and maintain modeling codes that
have been submitted to a quality assurance process. An individual EDSS being
used to evaluate potential emissions from a factory might make use of these data
and codes, as well as the supercomputer power, to solve a local air-modeling
problem. Avoiding the need to distribute the data and codes saves a considerable
amount of space (which would have been redundantly consumed on every similar
EDSS platform), and also reduces the risk of data (or code) contamination.
In any case, the environmental models currently in use already stretch even
high-end workstation capabilities to the point that analysts might wait several
days for a single iteration of a model to execute. As computer throughput
increases, more iterations of the Monte Carlo simulation will be executed,
and more complicated models employed. Although individual workstations
Information Systems for Environmental Management 77

can satisfy many environmental management computation requirements, there

will be a need for supercomputer access in environmental decision support for
quite some time.

4.6 Expert Systems

Finally, information systems engineering offers the environmental decision sup-
port system a technology to help capture and deliver the knowledge of experts in
particular problem domains. EDSS is predicated on the notion that human
intelligence is needed to make responsible environmental management decisions.
Artificial Intelligence (AI) might therefore seem anachronistic in this work.
However, although expert systems research has indeed grown out of AI research,
the connection stops there. Expert systems offer the possibility of providing
advisors to environmental analysts, for example, to help them choose the assump-
tions and parameters of their conceptual model of the problem (4). In this sense,
the interactive user has the benefit of aggregated advice from many experts who
would otherwise be unavailable, but still has the last word. This area of research
in EDSS is the most prospective, and much work remains to be done before it
can be claimed that expert systems technology has contributed substantially.
Nonetheless, there is great potential for a productive collaboration.

4.7 Decision Science

The term decision science is used here to refer collectively to the various fields
of investigation which attempt to provide quantitative (or at least controlled
qualitative) structure to the decision-making process. This includes subdisciplines
ranging from statistics and geostatistics, through operations research and linear
programming optimization, to classical and Bayesian probability theory. While
such formal decision methods are only sparingly applied in current environmental
decision frameworks, it can be expected that this will increase in the future, if for
no other reason than they provide some accountability for the decision process
and remove some of the air of subjectivity from it.
There is a formalism associated with decision science, the terms of which
are fairly intuitive. To begin with, a decision itself is a choice between alterna-
tives. These alternatives are compared according to some criteria, the measurable
evidence on which the decision is to be based. A criterion can be a factor which
enhances or detracts from the suitability of an alternative, or a constraint which
limits the alternatives under consideration. In order to combine criteria for
evaluation and action, one employs decision rules. These include procedures for
aggregating criteria into a single index, along with an algorithm for comparing
alternatives according to this index. Decision rules can be choice functions
(sometimes called objective functions) or choice heuristics. The former provide
a mathematical method for alternative comparison, typically involving some form
78 Frysinger

of optimization. The latter provide an algorithm or procedure to be followed,

sometimes with a stopping rule to indicate when the procedure should terminate
and the solution either taken or the search abandoned. For example, if one seeks
to fit a linear equation to a set of data points, one can solve the conventional linear
regression equation which sets a derivative to zero to solve for the minimum
cumulative squared error. This would be a choice function. Alternatively, one can
solve the equation iteratively while varying the coefficients according to some
prescription, stopping either when this same error metric is “small enough”
(but not necessarily a minimum) or when the number of iterations has exceeded
one’s patience. This choice heuristic might result in the same solution as the
choice function, but in examples such as this one it probably will not. On the other
hand, there may not be unique analytical solutions to the problem at hand, leaving
heuristic approaches the only game in town.
There is usually a specific objective of the decision at hand, and the decision
rules are structured in the context of this objective. When there are multiple
criteria which must be considered in the decision, this is termed a multicriteria
evaluation, in which some method for combining the criteria must be selected.
More complicated is the multiobjective case, in which there are multiple objec-
tives which may be complementary or may conflict.
While a great many techniques are available from decision science, two are
commonly employed in environmental decision making: linear programming and
decision trees.

4.8 Linear Programming

Linear programming methods are usually associated with operations research.
They are typically applied to optimization and resource allocation problems
where there are linear relationships between problem parameters, both objectives
and constraints. The linear equations describing the constraints associated with
decision variables are solved simultaneously to define a solution space or feasible
region (in as many dimensions as there are variables). The linear objective
function is then evaluated to determine its minimal or maximal value (for cost
functions or benefit functions, respectively). If this optimal value, plotted in the
space of the decision variables, is contained within the feasible region defined by
the constraints, then an optimal, feasible solution has been found. Given this
structure, linear programming solutions strongly resemble conventional (multi-
ple) linear regression methods, solved either graphically or iteratively. These
methods are frequently used in optimization problems such as cost/benefit
analysis for monitoring or remediation systems, or allocation of monitoring wells
along a site perimeter.
Vogel (12) cites an example of this form of systems analysis applied to a
so-called conjunctive use problem in which the best balance of water supply
Information Systems for Environmental Management 79

sources (surface and groundwater) is sought, with the goal that the total system
yield under coordinated use exceeds the sum of the yields under uncoordinated
use. Given an annual water demand of K, the decision maker seeks to find optimal
values of groundwater withdrawal (G) and surface water withdrawal (S) such that
G + S ≥ K. Naturally, there are various constraints on both ground and surface
water usage (for example, there are maximum yields from each source), which
taken together form the feasible region. If the objective function of this problem
can be described linearly (i.e., there is some linear combination of G and S whose
coefficients describe the normalized benefits of each source of water), then the
family of curves (straight lines) representing this function under various coeffi-
cient values can be plotted on the decision axes overlying the feasible region.
The selection of the optimal coefficients can then be made graphically.
For more realistic problems there are many constraint equations and terms
in the objective function, preventing the use of graphical methods. However, a
variety of techniques have been developed to evaluate such systems mathemati-
cally. Even in cases where the exact optimal solution is intractable, linear
programming has the potential to identify a range of solutions in the neighbor-
hood of the optimal solution.

4.9 Decision Trees

Decision trees are associated with a decision analytic method which accounts for
both expected value and uncertain events. A hierarchical graphical structure is
used to describe the structure of the decision problem. Nodes (vertices) in the tree
are either decision nodes or chance nodes, depending on whether the branches
result from the decision maker’s choice or some uncertain event, respectively.
Every decision tree has as its root a decision node, which is the first decision
under consideration. Every branch in the tree eventually terminates in a “leaf”
representing the outcome of that particular path through the tree, with its
associated probabilities and expected value. Folding back the path probabilities
and expected values of chance nodes (by multiplication), one can arrive at
expected values for the decision nodes, and make an optimal decision based on
this value. However, to do this one requires some metrics for expected value of
each outcome, and probabilities for each branch from each chance node. Further-
more, the decision and chance alternatives must be finite: one is selecting from a
particular set of decision alternatives, rather than adjusting an operating point on
a continuum.

5 APPLICATIONS OF EDSS
The foregoing has provided a foundation for environmental decision support
systems in general. Though the technology has seen most of its application in
80 Frysinger

natural resource management and environmental remediation, there are many

opportunities to bring the power of EDSS to bear on problems in industry. Three
examples will serve to illustrate this point.

5.1 Integrated Factory Decision Support

Environmental compliance within a manufacturing environment is an information-
intensive pursuit, and can be facilitated by the integration of information systems
and repositories (13). The value of such integration can best be illustrated in the
breach. When changes are made in the configuration of a manufacturing area,
such as the movement of solvent baths from one location to another, the failure to
include environmental compliance managers in the decision process can result in
permit violations (for example, the movement of Volatile Organic Compound (VOC)
sources from one vent stack to another can require air permit modifications). In
general, layout of the factory floor can affect environmental performance, as well as
compliance, so that one must consider environmental ramifications when attempting
to develop and/or modify the manufacturing layout.
Geographic information systems, and their cousins the computer-aided
design (CAD) systems, have been involved in facilities management for some
time, but their use in support of environmental management is relatively new (14).
Combining the spatial plant design data with relational data describing such
domains as materials inventory provides the basis for integrated decision mak-
ing—integrating environmental management with overall plant management
functions. Combining these with decision tools and simulation capabilities allows
managers to make superior decisions about plant layout, and improve their
compliance record.
Beyond physical plant arrangements, this sort of information system inte-
gration can also go a long way toward reducing the cost of regulatory compliance.
For example, in the United States the Emergency Preparedness and Community
Right-to-Know Act (EPCRA) requires annual reporting of the quantities and
whereabouts of hazardous materials, with the intent of ensuring the safety of
emergency responders in the event of fire or other disaster. This simple and
arguably worthwhile requirement can result in a great deal of expense to a
company whose information management and decision tools are not integrated.
It is typical for such companies to issue annual inventory surveys to plant
personnel, who then must physically locate and record such materials so that the
regulatory reports can be completed. A far better alternative is to integrate the
purchasing, storeroom, and environmental compliance software systems so that
the flow of materials into and within the facility is generally known at any time.
This not only permits the decision support system to easily produce the annual
reports required for the EPCRA, but also allows regular review of materials
Information Systems for Environmental Management 81

movements and usage, which in turn can facilitate such other tasks as tracking of
air emissions calculated by mass balance.
This example illustrates an EDSS emphasizing GIS and relational database
technology, and especially the integration of these technologies across organiza-
tional and functional boundaries within the operation.

5.2 Risk Management Planning

In the United States, provisions of the Clean Air Act require owners or operators
of a stationary air pollution source with more than a threshold quantity of a
regulated substance to submit a Risk Management Plan (RMP). Among other
things, this plan must describe the accidental release prevention and emergency
response policies at the source, the regulated substances handled, and the worst-
case release scenarios and alternative release scenarios, including administrative
controls and mitigation measures to limit the distances for each reported scenario.
This regulation provides a natural application for environmental decision
support systems. To plan for a release of chlorine from tank cars on a siding, for
example, one must evaluate the dynamics of material movement in the siding
area, including variations in quantity of chlorine present at any one time. Then
atmospheric transport models must be used to predict, for each of a variety of
weather conditions, where the toxic gas is likely to go, and in what concentra-
tions. Since weather prediction and atmospheric models are attended by a great
deal of uncertainty, quantitative means must be employed to manage this uncer-
tainty. Monte Carlo simulation, as described previously, can help to quantify the
uncertainty given historical data on which to base probability distributions.
This example illustrates an EDSS based on modeling and simulation, with
substantial support provided by GIS technology.

5.3 Design for Environment

A third example of the use of environmental decision support systems in the
industrial context involves supporting the decision-making process engaged in by
product and process designers intending to minimize the environmental impact
of their product. Designing for environment (DFE) requires the availability of a
great deal of information regarding alternative materials, components, and pro-
cesses available for consideration by the designer. Such information is notoriously
difficult to find, and when available its applicability to different situations is quite
variable. To support the designer adequately, the system must make this informa-
tion available for ready access, but it must also help the user to select only the
information appropriate to the problem at hand, and perhaps also assist in the
actual design decisions.
By integrating an expert system with a highly descriptive relational
database, the EDSS can meet this need. Expert assistance (even if delivered by
82 Frysinger

computer) is very appropriate in this situation, especially if one considers that the
designer very likely has little training in environmental issues.

6 EDSS SUMMARY
Environmental decision support systems, defined here as a class of information
systems integrating several technologies in support of improved environmental
decision quality, have served well in a variety of applications in natural resource
management and environmental remediation. They offer similar benefits to the
industrial environmental manager prepared to invest in their deployment. While
not turnkey, off-the-shelf solutions, such systems, once developed, can earn their
keep by helping to solve problems which might otherwise be intractable.

7 CONCLUSION
The foregoing has described a wide range of information systems designed to help
improve the quality of environmental management. The integrated approach to
environmental management information systems, while based largely on conven-
tional relational database technology, still offers the prospect of real risk reduction
and performance improvement when information maintenance and management
is required. For more active decision-making processes, the described framework
for environmental decision support systems offers the opportunity to put scientif-
ically sound tools into the hands of real decision makers. The key to both of these
approaches in integration.

REFERENCES
1. G. Guariso and H. Werthner, Environmental Decision Support Systems. Chichester,
U.K.: Ellis Horwood, 1989.
2. S. P. Frysinger, An Open Architecture for Environmental Decision Support. Int. J.
Microcomput. Civil Eng., vol. 10, no. 2, pp. 119–126, 1995.
3. R. W. Bailey, Human Performance Engineering. Prentice-Hall, London, 1982.
4. A. S. Heger, F. A. Duran, S. P. Frysinger, and R. G. Cox, Treatment of Human-
Computer Interface in a Decision Support System. IEEE International Conference
on Systems, Man, and Cybernetics, pp. 837–841, 1992.
5. R. M. Hogarth, Judgement and Choice. New York: Wiley, 1987.
6. National Research Council, Issues in Risk Assessment. Washington, DC: National
Academy Press, 1993.
7. M. G. Morgan and M. Henrion, Uncertainty. Cambridge, U.K.: Cambridge Univer-
sity Press, 1990.
8. T. E. McKone and K. T. Bogen, Predicting the Uncertainties in Risk Assessment.
Environ. Sci. Technol., vol. 25, no. 10, pp. 1674–1681, 1991.
Information Systems for Environmental Management 83

9. S. Aronoff, Geographic Information Systems: A Management Perspective. Ottawa:

WDL Publications, 1989.
10. S. P. Frysinger, Applied Research in Auditory Data Representation, In E. J. Farrell
(ed.), Extracting Meaning From Complex Data—Proceedings of the SPIE/SPSE
Symposium on Electronic Imaging, 1990.
11. G. Kramer, Auditory Display: Sonification, Audification, and Auditory Interfaces.
Proceedings of the 1992 International Conference on Auditory Display. Addison-
Wesley, 1994.
12. R. M. Vogel, Resource Allocation. In R. A. Chechile and S. Carlisle (eds.), Environ-
mental Decision Making: A Multidisciplinary Perspective. New York: Van Nostrand
Reinhold, pp. 156–175, 1991.
13. S. P. Frysinger, New Approaches to Environmental Information and Decision Support
Systems. National Association for Environmental Management’s Environmental
Management Forum, Dallas, TX, October 28–31, 1997.
14. W. J. Douglas, Environmental GIS: Applications to Industrial Facilities. Lewis
Publishers, Boca Raton, FL, 1995.
6
European Policies for
Waste Management

Ingo F. W. Romey and Marc Obladen

University of Essen, Essen, Germany

1 INTRODUCTION
With the treaty establishing the European Community and with the ratification of
the Treaty of Amsterdam, the European Community receives wide legislative
rights. The following states are members of the European Community: Austria,
Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxem-
bourg, Netherlands, Portugal, Spain, Sweden, and the United Kingdom of Great
Britain and Northern Ireland.
In order to carry out their tasks and in accordance with the provisions of
this treaty, the European Parliament, acting jointly with the Council and the
Commission, make regulations and issue directives, take decisions, and make
recommendations or deliver opinions.
A regulation shall have general application. It shall be binding in its entirety
and directly applicable in all Member States.

85
86 Romey and Obladen

A directive shall be binding, as to the result to be achieved, upon each

Member State to which it is addressed, but shall leave to the national
authorities the choice of form and methods.
A decision shall be binding in its entirety upon those to whom it is
addressed. Recommendations and opinions shall have no binding force.
By the Treaty of Amsterdam the members of the European Community are
obligated to pursue a common environmental policy. Community policy on the
environment shall contribute to pursuit of the following objectives:
Preserving, protecting, and improving the quality of the environment
Protecting human health
Prudent and rational utilization of natural resources
Promoting measures at the international level to deal with regional or
worldwide environmental problems
Further Community policy on the environment shall aim at a high level of
protection, taking into account the diversity of situations in the various regions
of the Community. It shall be based on the precautionary principle and on the
principles that preventive action should be taken, that environmental damage
should as a priority be rectified at the source, and that the polluter should pay. In
this context, harmonization measures answering environmental protection re-
quirements shall include, where appropriate, a safeguard clause allowing Member
States to take provisional measures, for noneconomic environmental reasons,
subject to a Community inspection procedure.
Without prejudice to certain measures of a Community nature, the Member
States shall finance and implement the environment policy.
Without prejudice to the principle that the polluter should pay, if a measure
based on the provisions of paragraph 1 involves costs deemed disproportionate
for the public authorities of a Member State, the Council shall, in the act
adopting that measure, lay down appropriate provisions in the form of temporary
derogation, and/or financial support from the Cohesion Fund of the European
Community.
Exemplary discussed should be the newest and most important decisions
dealing with the universal treatment of wastes and the waste management
strategy.

2 EUROPEAN DIRECTIVES ON WASTE MANAGEMENT

2.1 European Parliament and Council Directive on
Packaging and Packaging Waste
The Directive on packaging and packaging waste aims to harmonize national
measures concerning the management of packaging and packaging waste in order,
European Policies for Waste Management 87

on the one hand, to prevent any impact thereof on the environment of all Member
States as well as of third countries or to reduce such impact, thus providing a high
level of environmental protection, and, on the other hand, to ensure the function-
ing of the internal market and to avoid obstacles to trade and distortion and
restriction of competition within the Community. This Directive lays down
measures aimed, as a first priority, at preventing the production of packaging
waste and, as additional fundamental principles, at reusing packaging, at recycl-
ing and other forms of recovering packaging waste and, hence, at reducing the
final disposal of such waste. It covers all packaging placed on the market in the
Community and all packaging waste, whether it is used or released at industrial,
commercial, office, shop, service, household, or any other level, regardless of the
material used. This Directive shall apply without prejudice to existing quality
requirements for packaging such as those regarding safety, the protection of health
and the hygiene of the packed products or to existing transport requirements.
In this context, “packaging” shall mean all products made of any materials
of any nature to be used for the containment, protection, handling, delivery, and
presentation of goods, from raw materials to processed goods, from the producer
to the user or the consumer. “Nonreturnable” items used for the same purposes
shall also be considered to constitute packaging. Generally recommendable is the
prevention of packaging. Therefore it shall be ensured that preventive measures
are implemented. Such other measures may consist of national programs or
similar actions adopted, if appropriate in consultation with economic operators,
and designed to collect and take advantage of the many initiatives taken within
Member States as regards prevention. Furthermore, all Member States should
encourage reuse systems of packaging, which can be reused in an environmen-
tally sound manner, in conformity with the Treaty.
In order to comply with the objectives of this Directive, Member States
shall take the necessary measures to attain the following targets covering the
whole of their territory. No later than five years between 50% as a minimum and
65% as a maximum by weight of the packaging waste will be recovered.
Moreover, within this general target, and with the same time limit, between 25%
as a minimum and 45% as a maximum by weight of the totality of packaging
materials contained in packaging waste will be recycled, with a minimum of 15%
by weight for each packaging material. At least no later than 10 years from the
date by which this Directive must be implemented in national law, a percentage
of packaging waste will be recovered and recycled, which will have to be
determined by the Council.
Member States shall, where appropriate, encourage the use of materials
obtained from recycled packaging waste for the manufacturing of packaging and
other products.
Member States which have, or will, set programs going beyond the above-
mentioned targets and which provide to this effect appropriate capacities for
88 Romey and Obladen

recycling and recovery, are permitted to pursue those targets in the interest of a
high level of environmental protection, on condition that these measures avoid
distortions of the internal market and do not hinder compliance by other Member
States with the Directive.
Member States shall take the necessary measures to ensure that systems are
set up to provide on the one hand the collection and return of used packaging and
packaging waste from the consumer, other final user, or from the waste stream in
order to channel it to the most appropriate waste management alternatives. On the
other hand, the reuse or recovery including recycling of the packaging and/or
packaging waste collected, in order to meet the objectives laid down in this
Directive, should be ensured. These systems shall be open to the participation of
the economic operators of the sectors concerned and to the participation of the
competent public authorities.
For waste management, a marking and identification system of waste
packaging is essential. Therefore, the Council shall decide no later than two years
after the entry into force of this Directive on the marking of packaging. To
facilitate collection, reuse, and recovery (including recycling), packaging shall
indicate for purposes of its identification and classification by the industry
concerned the nature of the packaging material(s) used. To that end, the Commis-
sion will specify the numbering and abbreviations on which the identification
system is based and shall specify which materials shall be subject to the
identification system in accordance with the same procedure. Packaging shall
bear the appropriate marking either on the packaging itself or on the label. It shall
be clearly visible and easily legible. The marking shall be appropriately durable
and lasting, even when the packaging is opened.
In addition, the Commission will promote, as appropriate, the preparation
of European standards relating to the essential requirements, in particular, the
preparation of European standards relating to:

Criteria and methodologies for life-cycle analysis of packaging

The methods for measuring and verifying the presence of heavy metals and
other dangerous substances in the packaging and their release into the
environment from packaging and packaging waste
Criteria for a minimum content of recycled material in packaging for
appropriate types of packaging
Criteria for recycling methods
Criteria for composting methods compost produced
Criteria for the marking of packaging

With this Directive it will be ensured that the sum of concentration levels
of lead, cadmium, mercury, and hexavalent chromium present in packaging or
packaging components shall not exceed the following:
European Policies for Waste Management 89

600 ppm by weight in 1998

250 ppm by weight in 1999
100 ppm by weight in 2001
Regarding effective packaging management, European information system
should be introduced. Thus, Member States shall take the necessary measures to
ensure that databases on packaging and packaging waste are established, where
not already in place, on a harmonized basis in order to help Member States and
the Commission to monitor the implementation of the objectives set out in this
Directive. To this effect, the databases shall provide in particular information on
the magnitude, characteristics, and evolution of the packaging and packaging
waste flows (including information on the toxicity or danger of packaging
materials and components used for their manufacture) at the level of individual
Member States. In order to harmonize the characteristics and presentation of the
data produced and to make the data of the Member States compatible, Member
States shall provide the Commission with their available data in standard formats
which shall be adopted by the Commission. Member States shall require all
economic operators involved to provide competent authorities with reliable data
on their sector.
Regarding an information system for users of packaging, Member States
shall take measures to ensure that users of packaging, including in particular
consumers, obtain the necessary information about:
The return, collection, and recovery systems available to them
Their role in contributing to reuse, recovery, and recycling of packaging
and packaging waste
The meaning of markings on packaging existing on the market
In pursuance of the objectives and measures referred to in this Directive,
Member States shall include in the waste management plans a specific chapter on
the management of packaging and packaging waste. Acting on the basis of the
relevant provisions of the Treaty, the Council shall adopt economic instruments
to promote the implementation of the objectives set by this Directive. In the
absence of such measures, the Member States may, in accordance with the
principles governing Community environmental policy, inter alia, the polluter
pays principle, and the obligations arising out of the Treaty, adopt measures to
implement those objectives.
At least, this Directive specifies essential requirements on the composition
and the reusable and recoverable, including recyclable nature of packaging.
Packaging shall be so manufactured that the packaging volume and weight
be limited to the minimum adequate amount to maintain the necessary level of
safety, hygiene, and acceptance for the packed product and for the consumer.
Furthermore, they shall be designed, produced, and commercialized in such a
90 Romey and Obladen

way as to permit reuse or recovery, including recycling, and to minimize im-

pact on the environment when packaging waste or residues from packaging
waste management operations are disposed of. The manufacturing has to avoid
the presence of noxious and other hazardous substances and materials as constit-
uents of the packaging material or of any of the packaging components or
minimized with regard to their presence in emissions, ash, or leachate when
packaging or residues from management operations or packaging waste are
incinerated or landfilled.
The physical properties and characteristics of the packaging shall enable a
number of trips or rotations in normally predictable conditions of use and offer
the possibility of processing the used packaging in order to meet health and safety
requirements for the workforce. At least, the properties should fulfil the require-
ments specific to recoverable packaging when the packaging is no longer reused
and thus becomes waste.
Furthermore, packaging must be manufactured in such a way as to enable
the recycling of a certain percentage by weight of the materials used in the
manufacture of marketable products, in compliance with current standards in
the Community. The establishment of this percentage may vary, depending on the
type of material of which the packaging is composed. Packaging waste processed
for the purpose of energy recovery shall have a minimum inferior calorific value
to allow optimization of energy recovery. Packaging waste processed for the
purpose of composting shall be of such a biodegradable nature that it should not
hinder the separate collection and the composting process or activity into which
it is introduced, whereas biodegradable packaging waste shall be of such a nature
that it is capable of undergoing physical, chemical, thermal, or biological decom-
position such that most of the finished compost ultimately decomposes into
carbon dioxide, biomass, and water.

2.2 Council Directive on Landfill of Waste

The aim of the Directive on landfill of waste is, by way of stringent operational
and technical requirements on the waste and landfills, to provide for measures,
procedures, and guidance to prevent or reduce as far as possible negative effects
on the environment, in particular the pollution of surface water, groundwater, soil,
and air, and on the global environment, including the greenhouse effect, as well
as any resulting risk to human health, from landfilling of waste, during the whole
life cycle of the landfill.
In principle, each landfill shall be classified in one of the following classes:
Landfill for hazardous waste
Landfill for nonhazardous waste
Landfill for inert waste
European Policies for Waste Management 91

Member States shall set up a national strategy for the implementation of

the reduction of biodegradable waste going to landfills. This strategy should
include measures to achieve the following targets by means of, in particular,
recycling, composting, biogas production, or materials/energy recovery. It ensures
that long-term biodegradable municipal waste going to landfills must be reduced
to 35% of the total amount (by weight) of biodegradable municipal waste
produced in 1995. Furthermore, the Directive shall take measures in order that
liquid waste, waste which, in the conditions of landfill, is explosive, corrosive,
oxidizing, highly flammable, or flammable, as well as hospital and other clini-
cal wastes arising from medical or veterinary establishments, are not accepted in
a landfill.
The landfill permit shall state at least the class of the landfill and the list of
defined types and the total quantity of waste which are authorized to be deposited
in the landfill.
Member States shall take measures to ensure that all of the costs involved
in the setting up and operation of a landfill site, including as far as possible the
cost of the financial security and the estimated costs of the closure and after-care
of the site for a period of at least 30 years shall be covered by the price to be
charged by the operator for the disposal of any type of waste in that site.
Regarding existing landfill sites, it should be ensured that landfills which have
been granted a permit may not continue to operate unless the landfill sites will
fulfil the standard of state of the art landfill sites within eight years.
The directive sets up general requirements for all classes of landfills. The
landfill can be authorized only if the characteristics of the site indicate that the
landfill does not pose a serious environmental risk.
The location of a landfill must take into consideration the existence of
groundwater, coastal water, or nature protection zones in the area, the geological
and hydrogeological conditions in the area, as well as the risk of flooding,
subsidence, landslides, or avalanches on the site. Therefore the water control and
leachate management must control water from precipitations entering into the
landfill body, prevent surface water and/or groundwater from entering into the
landfilled waste, and collect contaminated water and leachate.
Where the geological barrier does not naturally meet the conditions of
permeability and thickness, it has to be completed artificially and reinforced by
other means giving equivalent protection. An artificially established geological
barrier should be no less than 0.5 m thick.
Recommended parameters for a check study are ph, TOC, phenols, heavy
metals, fluoride, AS, and oil/hydrocarbons. In order to reduce global warming,
the methane-containing landfill gas shall be collected from all landfills receiving
biodegradable waste and the landfill gas must be treated and used. If the gas
collected cannot be used to produce energy, it must be flared.
92 Romey and Obladen

2.3 European Parliament and Council Directive on

Incineration of Waste
Having regard to the Treaty establishing the European Community and a proposal
of the Commission, the Council of the European Union and the European
Parliament adopted a new directive on the incineration of waste in July 1999.
In accordance with the principle of subsidiary, it was considered that the
objective of reducing emissions from incineration and co-incineration plants
cannot be achieved effectively by Member States acting individually. Un-
concerted action offers no guarantee of achieving the desired objective, and
regarding the need to reduce emissions across the Community, it is more effective
to take action at the level of the Community. This Directive confines itself to
minimum requirements for incineration and co-incineration plants.
The Directive is a consequence of the fifth Environment Action Program:
Towards Sustainability—a European Community program of policy and action in
relation to the environment and sustainable development which sets as an
objective “no exceedance ever of critical loads and levels” of certain pollutants
such as nitrogen oxides (NOx), sulfur dioxide (SO2), heavy metals, and dioxins,
while in terms of air quality the objective is that “all people should be effectively
protected against recognized health risks from air pollution.” That program
further sets as an objective a “90% reduction of dioxin emissions of identified
sources by 2005 (1985 level)” and “at least 70% reduction from all pathways of
cadmium (Cd), mercury (Hg), and lead (Pb) emissions in 1995.”
The purpose of the incineration plants established and operated in accor-
dance with this Directive is to reduce the pollution-related risks of waste through
a process of thermal treatment, especially oxidation, to reduce the quantity and
volume of the waste and to produce residues that can be recycled or disposed of
safely. The co-incineration of waste in plants not primarily intended to incinerate
waste should not be allowed to cause higher emissions of polluting substances in
that part of the exhaust gas volume resulting from such co-incineration and should
therefore be subject to appropriate limitations.
The aim of this Directive is to prevent or, where that is not practicable, to
reduce as far as possible negative effects on the environment, in particular the
pollution of air, soil, surface water, and groundwater, and the resulting risks to
human health, from the incineration and co-incineration of waste and, to that end,
to set up and maintain appropriate operating conditions and emission limit values
for waste incineration and co-incineration plants within the Community.
Incineration plants shall be operated in order to achieve a level of inciner-
ation such that the total organic carbon (TOC) of the slag and bottom ashes is less
than 3% or their loss upon ignition is less than 5% of the dry weight of the
material. If necessary, appropriate techniques of waste pretreatment shall be used.
All incineration plants shall be designed, equipped, built, and operated in such a
European Policies for Waste Management 93

way that the gas resulting from the process is raised, after the last injection of
combustion air, in a controlled and homogeneous fashion and even under the most
unfavorable conditions, to a temperature of at least 850˚C, as measured near the
inner wall of the combustion chamber, for at least 2 s. If hazardous wastes with
a content of more than 1% of halogenated organic substances, expressed as
chlorine, are incinerated, the temperature has to be raised to at least 1100˚C.
Any heat generated by the incineration or co-incineration process shall be
recovered as far as possible.
The directive determines following air emission limit values.
2.3.1 Air Emission Limit Values
Daily average values:

Total dust 10 mg/m3

Gaseous and vaporous organic substances, expressed as 10 mg/m3
total organic carbon
Hydrogen chloride (HCl) 10 mg/m3
Hydrogen fluoride (HF) 1 mg/m3
Sulfur dioxide (SO2) 50 mg/m3
Nitrogen monoxide (NO) and nitrogen dioxide (NO2), 200 mg/m3
expressed as nitrogen dioxide for existing incineration
plants with a capacity exceeding 3 tonnes per hour or
new incineration plants
Nitrogen monoxide (NO) and nitrogen dioxide (NO2), 400 mg/m3
expressed as nitrogen dioxide for existing incineration
plants with a capacity of 3 tonnes per hour or less

Until 1 January 2007, the emission limit value for NOx does not apply to plants
incinerating hazardous waste only.

Half-hourly average values:

Total dust 30 mg/m3

Gaseous and vaporous organic substances, expressed as 20 mg/m3
total or organic carbon
Hydrogen chloride (HCl) 60 mg/m3
Hydrogen fluoride (HF) 4 mg/m3
Sulfur dioxide (SO2) 200 mg/m3
Nitrogen monoxide (NO) and nitrogen dioxide (NO2), 400 mg/m3
expressed as nitrogen dioxide for existing incineration
plants with a capacity exceeding 3 tonnes per hour or
new incineration plants
94 Romey and Obladen

Until 1 January 2007, the emission limit value for NOx does not apply to plants
incinerating hazardous waste only.

All average values over the sample period of a minimum of 30 mins and a
maximum of 8 h.

Cadmium and its compounds, expressed Total Total

as cadmium (Cd)
Thallium and its compounds, expressed as 0.05 mg/m3 0.1 mg/m3
thallium (Tl)
Mercury and its compounds, expressed as 0.05 mg/m3 0.1 mg/m3
mercury (Hg)
Antimony and its compounds, expressed
as antimony (Sb)
Arsenic and its compounds, expressed as
arsenic (As)
Lead and its compounds, expressed as
lead (Pb)
Chromium and its compounds, expressed Total Total
as chromium (Cr)
Cobalt and its compounds, expressed as 0.5 mg/m3 1 mg/m3
cobalt (Co)
Copper and its compounds, expressed as
copper (Cu)
Manganese and its compounds, expressed
as manganese (Mn)
Nickel and its compounds, expressed as
nickel (Ni)
Vanadium and its compounds, expressed
as vanadium (V)

These average values also cover gaseous and the vapor forms of the relevant
heavy-metal emissions as well as their compounds. Until 1 January 2007 these
average values shall apply to existing plants for which the permit to operate
has been granted before 31 December 1996, and which incinerate hazardous
waste only.

Average values shall be measured over a sample period of a minimum of 6

h and a maximum of 8 h. The emission limit value refers to the total concentration
of dioxins and furans calculated using the concept of toxic equivalence in
accordance with Annex I:

Dioxins and furans 0.1 ng/m3

European Policies for Waste Management 95

The following emission limit values of carbon monoxide (CO) concentra-

tions shall not be exceeded in the combustion gases (excluding the start-up and
shut-down phases):
50 mg/m3 of combustion gas determined as daily average value
150 mg/m3 of combustion gas of at least 95% of all measurements deter-
mined as 10-min average values or 100 mg/m3 of combustion gas of all
measurements determined as half-hourly average values taken in any
24-h period.
Exemptions may be authorized by the competent authority for incineration
plants using fluidized bed technology, provided that the authorization foresees an
emission limit value for carbon monoxide (CO) of not more than 100 mg/m3 as
a hourly average value.
2.3.2 Determination of Emission Limit Values for the
Co-incineration of Waste
The limit value for each relevant pollutant and carbon monoxide in the exhaust
gas resulting from the co-incineration of waste shall be calculated as follows:
Vwaste ⋅ Cwaste + Vproc ⋅ Cproc
= C
Vwaste + Vproc

where
Vwaste is exhaust gas volume resulting from the incineration of waste only,
determined from the waste with the lowest calorific value specified in
the permit and standardized at the conditions given by this Directive.
Cwaste is emission limit value set in Annex V for plants intended to
incinerate wastes only (at least the emission limit values for the pollu-
tants and carbon monoxide).
Vproc is exhaust gas volume resulting from the plant process, including the
combustion of the authorized fuels normally used in the plant (wastes
excluded), determined on the basis of oxygen contents at which the
emissions must be standardized as laid down in Community or national
regulations. In the absence of regulations for this kind of plant, the real
oxygen content in the exhaust gas without being thinned by addition of
air unnecessary for the process must be used. The standardization at the
other conditions is given in this Directive.
Cproc is emission limit value as laid down in the tables of this Annex for
certain industrial sectors or in case of the absence of such a table or such
value, emission limit values of the relevant pollutants and carbon mon-
oxide in the flue gas of plants which comply with the national laws,
regulations, and administrative provisions for such plants while burning
96 Romey and Obladen

the normally authorized fuels (wastes excluded). In the absence of these

measures the emission limit values laid down in the permit are used. In
the absence of such permit values the real mass concentrations are used.
C is total emission limit value as laid down in the tables of this Annex for
certain industrial sectors and certain pollutants or in case of the absence
of such a table or such values total emission limit values for CO and the
relevant pollutants replacing the emission limit values as laid down in
specific Articles of this Directive. The total oxygen content to replace the
oxygen content for the standardization is calculated an the basis of the
content above respecting the partial volumes.

2.3.3 Special Provisions for Large Combustion Plants

Cproc for solid fuels expressed in mg/Nm3 (O2 content 6%):

Pollutant 50–100 MWth 100–300 MWth >300 MWth

SO2
General case 850 850–200 (linear 200
decrease
from 100 to
300 MWth)
Indigenous fuels or rate of or rate of
desulfurization desulfurization
≥ 90% or rate of ≥ 95%
desulfurization
≥ 92%
NOx 400 300 200
Dust 50 30 30

Until 1 January 2007 and without prejudice to other Community legislation, the
emission limit value for NOx does not apply to plants co-incinerating hazardous
waste only.

Cproc for biomass expressed in mg/Nm3 (O2 content 6%):

Pollutant 50–100 MWth 100–300 MWth >300 MWth

SO2 200 200 200

NOx 350 300 300
Dust 50 30 30
European Policies for Waste Management 97

Until 1 January 2007 and without prejudice to other Community legislation, the
emission limit value for NOx does not apply to plants co-incinerating hazardous
waste only.

Cproc for liquid fuels expressed in mg/Nm3 (O2 content 3%):

Pollutant 50–100 MWth 100–300 MWth >300 MWth

SO2 850 850 to 200 (linear 200

decrease from 100 to
300 MWth)
NOx 400 300 200
Dust 50 30 30

Until 1 January 2007 and without prejudice to other Community legislation, the
emission limit value for NOx does not apply to plants co-incinerating hazardous
waste only.
2.3.4 C: Total Emission Limit Values
C expressed in mg/Nm3 (O2 content 6%). All average values over the sample
period of a minimum of 30 min and a maximum of 8 h:

Pollutant C

Cd + Tl 0.05
Hg 0.05
Sb + As + Pb + Cr + Co + Cu + Mn + Ni + V 0.5

C expressed in ng/Nm3 (O2 content 6%). All average values measured over the
sample period of a minimum of 6 h and a maximum of 8 h:

Pollutant C

Dioxins and furans 0.1

C for solid fuels expressed in mg/Nm3 (O2 content 6%); C for biomass (as
defined in Council Directive 88/609/EEC as amended) expressed in mg/Nm3 (O2
content 6%); C for liquid fuels expressed in mg/Nm3 (O2 content 3%):
98 Romey and Obladen

Pollutant C

HCl 10
HF 1

3 CONCLUSION
The Council of Europe and the European Parliament legislate the general frame-
work of environment. The European Treaty binds the Member States to put
Community laws into national legislation. Regarding the precision of the guide-
lines, the elaboration of the directives will differ more or less in each Member
State. Furthermore, different time scales may exist to take into account the
different standards of environmental legislation in the Member States as well as
their specific situations. Nevertheless, the Community legislation will be harmo-
nized in the future, and separate treatment more and more abolished. Although
the principle of subsidiary guarantees the diversity of the European Union, a
common EU legislative gains more importance determined by the tempo of the
European integration.
Current European environmental legislation has already reached a very high
standard in environmental protection. Potential new member states for the Euro-
pean Union will be constrained to adopt stronger environmental legislation.

REFERENCES
Further information is available at http://europa.eu.int/pol/env/index_en.htm.
7
Energy Conservation

K. A. Strevett, C. Evenson, and L. Wolf

University of Oklahoma, Norman, Oklahoma

1 INTRODUCTION
A large proportion of our current pollution problems is the result of energy
technologies that rely on combustion of carbon-based fuels. Included in these
problems are emissions of greenhouse gases, acid rain precursors (oxides of sulfur
and nitrogen), and carbon monoxide; formation of photochemical oxidants;
releases to the biosphere of raw and refined petroleum products; and mining-
related pollution. Obviously, then, decreasing our consumption of carbon-based
energy will result in decreases in the amounts of these pollutants entering the
biosphere.
Global warming poses the threat of an environmental impact that is global
and, at least on a time scale of centuries, irreversible. Over the very long term of
two to three centuries, temperatures could rise by as much as 10 to 18˚C. While
it is impossible at this point to predict accurately all the effects of global warming,
its consequences are potentially so threatening to human and ecosystem health
that humans have an ethical obligation to do something about it (1).
It is obvious that strategies for reducing consumption of energy derived
from combustion of carbon-based fuels are among the most important means of
preventing global pollution. After a look at energy demands, this chapter dis-
cusses several such energy conservation strategies, the fuels currently being used
99
100 Strevett et al.

Natural
Gas Coal
24% 44%

Oil
32%

FIGURE 1 Percent contribution of coal, oil, and natural gas to global energy
markets.

to supply these demands, and a survey of the environmental impacts of some of

the pollutants produced by these fuels.

2 ENERGY SUPPLIES AND DEMANDS

Coal, oil, and natural gas supply about 95% of global energy. Coal dominates
energy markets, accounting for about 44% of fossil energy consumption. Oil
accounts for about 32% of fossil fuel supply, while natural gas contributes 24%
(Figure 1).
Coal is the most abundant fossil fuel worldwide, with current reserves
expected to last more than 200 years. “Conventional” oil production is expected
to peak between 2010 and 2020, resulting in a switch to “unconventional”*
sources and a possible increase in price (2). The total ultimately recoverable
natural gas resources in the world are estimated to amount to about 80% as much
energy as the recoverable reserves of crude oil. At current usage rates, gas
reserves represent approximately a 60-year supply (3).
Although developed countries account for less than 20% of the world’s
population, these countries use more than two-thirds of the commercial energy
supply, consuming 78% of the natural gas, 65% of the oil, and about 50% of
the coal produced each year (Figure 2). The United States and Canada, for
example, account for only about 5% of the world’s population, but consume
about one-quarter of the available energy (3). Carlsmith et al. (1990, as cited in
Ref. 4) estimated that 36% of U.S. energy consumption is in commercial and
residential buildings; industry accounts for another 36% and transportation for the
remaining 28%.

*Oil is considered unconventional if it is not produced from underground hydrocarbon reservoirs by

means of production wells, and/or it requires additional processing to produce synthetic crude. It
includes such sources as oil shales, oil sands-based synthetic crudes and derivative products, and liquid
supplies derived from coal, biomass, or gas (2).
Energy Conservation 101

22%
35%
50% 50%
65% 78%

Coal Oil Natural Gas

Developed Underdeveloped

FIGURE 2 Comparison of coal, oil, and natural gas consumption in developed

and less developed countries.

In November 1998, the World Energy Outlook (2) predicted 65% growth
in world energy demand and 70% growth in CO2 emissions between 1995 and
2020, without policy changes. The Outlook estimates that fossil fuels will meet
95% of additional global energy between 1995 and 2020 and that two-thirds of
the increase in energy demand and energy-related CO2 emissions over this period
could occur in China and other developing countries. The market share of gas is
expected to increase, while that of oil will decline slightly and the share of coal
will remain stable. By 2020, global electricity generation is predicted to have
increased by nearly 88% over 1995 rates. While electricity generation from
energy sources other than carbon-based fuels and hydropower is growing fast, it
is expected to represent less than 1% of world electricity generation by 2020
without policy changes.

3 NONRENEWABLE ENERGY SOURCES

3.1 Coal
Coal is fossilized plant material preserved by burial in sediments and altered by
geological forces that compact and condense it into a carbon-rich fuel. Its
advantage lies in its abundance of supply. The environmental effects of burning
all the remaining coal, however, could be catastrophic. Coal is the worst offender
among fossil fuels in terms of CO2 per unit of energy generated. The supply of
coal is enough to permit atmospheric carbon buildup of severalfold (4). In
addition, the burning of coal is a primary source of acid rain precursors. Pollution
associated with the mining of coal is discussed later.
Industrialized countries generate between 20% and 30% of their energy
from coal; in the case of China, the figure is nearly 75% (5). In the United States,
the relative contribution of coal declined from a peak of about 75% of total energy
102 Strevett et al.

supply in 1910 to about 17% in 1973 and increased again to about 23% in 1989.
In 1989, about 86% of domestic coal consumption was accounted for in electric
power production (6).

3.2 Petroleum
Petroleum, like coal, is derived from organic molecules created by living organ-
isms millions of years ago and buried in sediments where high pressures and
temperatures concentrated and transformed them into energy-rich compounds.
Petroleum has represented a relatively inexpensive source of fuel for transporta-
tion and provides the chemical industry with feedstocks, e.g., for the production
of plastics. However, its use results in emissions of carbon dioxide, carbon
monoxide, and acid rain precursors, and in the formation of photochemical
oxidants. In addition, aquatic and terrestrial systems may become polluted by
unintentional releases of raw and refined petroleum.

3.3 Natural Gas

Natural gas is a combustible mixture of methane (CH4) and other hydrocarbons
formed during the anaerobic decomposition of organic matter. It is the least
polluting of the fossil fuels, releasing only a little more than half as much CO2 as
coal. Important disadvantages of natural gas are its limited supply, difficulty of
storage in large quantities, and difficulty of transport across oceans. It can be
transported across land via pipelines; however, leaks of methane from these pipelines
represent a significant contribution to global warming. Furthermore, such pipeline
networks are prohibitively expensive for developing countries. As a result, much of
the natural gas produced in conjunction with oil pumping is simply burned (flared
off), representing a terrible waste of a valuable resource (3).

4 SOURCES AND ENVIRONMENTAL IMPACTS

OF POLLUTANTS
The production and/or consumption of carbon-derived energy result in release to
the biosphere of a variety of pollutants. These include gaseous pollutants [carbon
dioxide, acid rain precursors (nitrogen oxides and sulfur dioxide), and carbon
monoxide], photochemical oxidants, unintentional releases of raw and refined
petroleum, mining-related pollution (i.e., acid mine drainage), methane, and
thermal pollution.

4.1 Gaseous Pollutants

4.1.1 Carbon Dioxide
Carbon dioxide is responsible for 55% of global warming. The two primary
anthropogenic sources of atmospheric CO2 are fossil fuel burning (~77%) and
Energy Conservation 103

deforestation (~23%). Cline (4) has estimated that if human sources of atmo-
spheric carbon were immediately reduced by about 43%, warming could be held
to about 2.5˚C.
Atmospheric CO2 concentration was more or less stable near 280 ppm for
thousands of years until about 1850, and has increased significantly since then
(Figure 3) (Schimel et al., 1995, as cited in Ref. 7). Since the beginning of the
industrial era, about 40% of all CO2 released through the burning of fossil fuel
has been absorbed by sinks; the remainder has remained in the atmosphere (1).
The human-caused increase in atmospheric CO2 already represents nearly a 30%
change relative to the preindustrial era (7); annual global emissions of CO2 have
increased 10 times this century (8). At the current rate of increase in concentra-
tions of CO2 and other heat-trapping gases in the atmosphere, greenhouse gas
concentrations will be equivalent to double the preindustrial CO2 concentration
by 2050 (National Academy of Sciences, 1992, as cited in Ref. 1). Ultimately,
this could increase the average global temperature by about 1–5˚C, with a likely
figure of 2.5˚C. According to Cline (4), we are already committed to about 1.7˚C
of warming from the existing accumulation of greenhouse gases, and warming
could increase by 10˚C or more if nothing is done to alter likely fossil fuel
consumption patterns. The historic record suggests that the average global surface
temperature has already risen approximately 0.3–0.6˚C since the nineteenth
century (1).
Natural gas releases slightly less than half the amount of CO2 released
during the combustion of coal, with petroleum in between. Coal and natural gas
each accounts for about 27% of U.S. fossil fuel supply, but coal accounts for about

375

350
Atm. CO2 Conc.
(ppm)

325

300

275
1700 1750 1800 1850 1900 1950 2000

FIGURE 3 Historical increase in global CO2 emissions. (Sources: Refs. 35–37.)

104 Strevett et al.

one-third of U.S. CO2 emissions. In the United States, electric utilities account
for about one-third of all CO2 releases, with transportation activities adding
approximately an additional third. Globally, oil consumption accounts for nearly
half of total CO2 emissions and much of its air pollution (6).
4.1.2 Nitrogen Oxides
Nitrogen oxides (NOx) are responsible for about 35% of acid rain, and are a
precursor of O3 pollution (Figure 4). Of all U.S. air pollutants, oxides of nitrogen
have been the most difficult to control. They are formed when ambient diatomic
nitrogen (N2) is heated to temperatures > 1200˚F, and their dominant sources are
the internal combustion engine and power plants (Figure 5) (1). The 900 million
tons of coal burned annually in the United States are responsible for about
one-third of all this country’s NOx emissions (3).
2NO + O2 → 2NO2
2NO2 + H2O → HNO2 + HNO3
There are various ways of reducing nitrogen oxide emissions including
combustion control and the use of catalysts (9). Our best option for reducing this
pollutant, however, is through reduced burning of fossil fuels and forests.
4.1.3 Sulfur Dioxide
Sulfur dioxide (SO2) is responsible for about 60% of acid rain (Figure 4). At least
two-thirds of the sulfur oxides in the United States are emitted from coal-fired
power plants. Much of the coal burned in the United States has a high sulfur
content—2% or more. Most of the remaining SO2 emissions are accounted for by
industrial fuel combustion and industrial processes such as petroleum refining,
sulfuric acid manufacturing, and smelting of nonferrous metals (Figure 5) (10).
4.1.4 Carbon Monoxide
Carbon monoxide (CO) is the result of incomplete combustion. CO inhibits
respiration in animals by binding irreversibly to hemoglobin. About half the CO
released to the atmosphere each year is the result of human activities. In the
United States, two-thirds of the CO emissions are created by internal combustion
engines in transportation (3).

4.2 Photochemical Oxidants

Photochemical oxidants are products of secondary atmospheric reactions driven
by solar energy—e.g., splitting of an O2 or NO2 molecule, freeing an oxygen
atom which reacts with another O2 to form ozone (O3). O3 is the result of
atmospheric chemistry involving two precursors, nonmethane hydrocarbons
(HCs) and NOx, which react in the presence of heat and sunlight (Figure 6) (11).
 Atmospheric mixing
Energy Conservation

yields sulfuric and Acid Rain

nitric acids
NOx
SO2 Dry deposition of
acidic compounds

Vehicular emissions
Burning of fossil fuels
yields SO2 and NOx

FIGURE 4 NOx and SO2 contributions to acid rain formation.

105
106 Strevett et al.

SO 2 Emissions Indust rial Ot her

Com bust ion Com bust ion
12% 3%
Ind/Mfg
P rocesses
13%

Ut ilit ies
T ransport . 68%
4%

Ot her
NOx Emissions Industrial Combustion Ot her
Combustion 4% 1%
16% Ut ilit ies
Ind/Mfg 32%
P rocesses
5%

T ransport .
42%

FIGURE 5 Percent contribution to SO2 and NOx emissions in the United

States of various industries. (Source: Ref. 34.)

This ground-level O3 is a pollutant that can have harmful effects on human health,
while O3 present in the upper atmosphere protects the earth from harmful
ultraviolet radiation. Figure 6 demonstrates the dynamic interactions between
HCs and NOx, which are produced from combustion, and atmospheric oxygen.
In addition to forming O3, NOx can also remove ground-level O3. This removal
is often temporary, however, as O3 is re-formed through other reactions.
Ground-level O3 is a respiratory irritant that causes health concerns at very
low concentrations because its very low solubility in water means it tends not to
be removed by the mucous in the upper respiratory tract and penetrates deeper
into the lungs. There is evidence that exposure to O3 accelerates the aging of lung
tissue and increases susceptibility to respiratory pathogens. Human exposure to
O3 can produce shortness of breath and, over time, permanent lung damage (12).
Costs of the health effects of O3 in the United States are estimated to be about
$50 billion per year. In addition, O3 causes more damage to plants than any other
pollutant (1). O3 concentrations rise with temperature and are therefore expected
to be exacerbated by global warming. If cloud cover decreases as a result of global
Energy Conservation 107

OH O2

H O2
NO
H2 O
O2

O3 O NO2

hv O2
λ=0.39µ m

HO
O2 H
NO

O2

O3 O NO2
OH

hv
λ=0.39µ m

FIGURE 6 The release of hydrocarbons and NO during combustion results in

the conversion of NO to NO2. Increased formation of NO2 increases the
production of O3.

warming, thus permitting increased penetration of sunlight, O3 concentrations

will be further increased.

4.3 Raw and Refined Petroleum Spills and Leaks

Crude oil spills such as that of the Exxon Valdez are probably the most widely
known examples of this type of energy-related pollution. In addition, it has been
estimated that about 11 million gallons of gasoline are lost each year by leaking
underground storage tanks (3).

4.4 Mining-Related Pollution

Acid mine drainage is one of the most common and damaging problems in the
aquatic environment. Many waters flowing from coal mines and draining from
the waste piles that result from coal processing and washing have low microbial
108 Strevett et al.

densities due to the highly acidic nature of these waters. Acidic mine water results
from the presence of sulfuric acid produced in a series of microbially mediated
reactions that begin with the oxidation of pyrite, FeS2 (13). Often, mining
operations result in surface waters infiltrating into the subsurface voids, especially
after the mine is exhausted and pumping ceases. In some areas of Appalachia,
large underground impoundments of water have filtered into coal mines. These
waters have become very acidic and, when they are returned to the surface via
pumping or by subsurface flows, their low pH value devastates the aquatic
systems they infiltrate (14).
Another impact of underground mining is the waste materials that are a
by-product of any mining operation. Gaining access to the vein or seam of coal,
as well as transporting the coal to the surface, requires large amounts of waste
materials to be removed to the surface. These waste materials, or tailings, are
often piled up in large mounds in close proximity to the mine. The composition
of many tailings can contain toxic minerals such as mercury, lead, or iron sulfide.
Water percolating through these waste materials often produces water quality
problems downstream from the tailings similar to those associated with subsur-
face water flows from within the mines. In addition to the sterile conditions on
tailings mounds themselves, rain water running off the tailings often is so acidic
as to kill both the vegetation in the immediately affected lands and the aquatic
life in streams and rivers receiving these waters. Many lands and streams within
the Appalachian coalfield areas of western Pennsylvania, West Virginia, eastern
Kentucky, and eastern Ohio are devastated by the acidic waters resulting from
coal mining operations. The enactment of environmental legislation limits the
damage associated with active mining operations, but the land degradation
associated with past mining has left a filthy legacy of degraded landscapes (14).
Surface mining is usually favored over underground mining for primarily
economic reasons. It is virtually impossible to prevent land degradation when
surface mining occurs. First, in some operations, huge depressions result. Second,
the overburden (extracted soil, subsoil, and unconsolidated earth and rocks) must
be stored and then replaced systematically in their original order after the mineral
is removed. Even under optimal conditions, which rarely occur, restoration
usually results in a landscape that is less productive than it was prior to mining.
Subsurface groundwater flow is always disturbed, and revegetation is often slow.
Restoration is further complicated when toxic materials are leached from the
overburden during its storage. These conditions often occur in coal mining
operations, which have disturbed about 2.3 million acres in the United States (14).
The area affected by mining can be three to five times more widespread
than the area actually exploited (15). Even when increased acidity is not consid-
ered, mining-related soil erosion alone can impact natural waters significantly.
Added nutrients may increase aquatic productivity, resulting in eutrophication.
Lower levels of dissolved oxygen associated with eutrophication may render the
Energy Conservation 109

water uninhabitable by other aquatic organisms. On the other hand, suspended

sediments may reduce light penetration, reducing productivity and therefore
available fish food. Sediments may also interfere with salmon and trout spawning
and reduce survival of their eggs. Young fish may be more susceptible to
predation when sediments fill or cover hiding places (14). Species that stalk their
prey visually may be unable to survive in murky water.

4.5 Methane
Methane is responsible for about 20% of the greenhouse effect, and concentra-
tions have already risen to more than double preindustrial estimations. Con-
centrations continue to rise at about 0.9% annually (4). The majority of
anthropogenic methane is the result of non-energy-related human activities such
as ruminant livestock and cultivation of rice (from which about half the world’s
population derive about 70% of their calories), and decomposition of organic
matter in landfills. However, leaks in natural gas pipelines contribute about 21%
of anthropogenic methane, and the burning of coal adds an additional 6%. Other
energy-related sources of methane include coal mines, natural gas leaks, gas
associated with oil production, and the creation of new wetlands when forests are
flooded following construction of hydroelectric dams.

4.6 Thermal Pollution

When coal is burned to generate electricity at a power plant, some of the coal’s
energy content is lost to coolant water, which is then discharged into rivers or
lakes. Since an inverse relationship exists between water’s temperature and
its oxygen-holding capacity, the water’s dissolved oxygen concentration can
be diminished to a point below which some aquatic organisms may be able
to survive.

5 POLLUTION PREVENTION THROUGH DECREASED

FOSSIL FUEL CONSUMPTION
Carbon dioxide can be considered an inevitable product of fossil fuel combustion;
therefore, CO2 emissions can be reduced only through reduced consumption of
fossil fuels. It is important to note that emissions of every other pollutant
discussed in Section 4 will be reduced as an additional benefit of reducing fossil
fuel consumption and thereby CO2 emissions.

5.1 Imposition of a Tax on Traditional Energy Sources

Internal costs are the expenses, monetary or otherwise, that are borne by those
who actually use a resource. External costs are the expenses, monetary or
110 Strevett et al.

otherwise, borne by someone other than the individuals or groups who use a
resource (3). As an example, according to Tenenbaum (27), a 1990 study at Pace
University concluded that the true cost of an unscrubbed coal plant was 11.6 cents
per kilowatt hour (kWh), double the 5.8 cents that utilities were charging.
Cline (4) has produced an extensive analysis of the economic effects of
global warming. One strategy for reducing dependence on fossil energy sources
is the imposition on these sources of a tax large enough to “internalize” the costs
associated with fossil fuels, such as sea-level rise (estimated by Cline to amount
to about $7 billion annually in the United States*), agricultural losses ($18
billion), curtailed water supply due to reduced runoff ($7 billion), forest loss (>$3
billion, considering only lumber value), increased electricity demand for addi-
tional cooling ($11 billion), exacerbation of urban O3 problems ($4 billion),
increased mortality from heat waves ($6 billion, valued at lifetime earning
potential), losses of leisure activities associated with winter sports (ski industry
$1.5 billion), increased hurricane ($750 million) and forest fire damage, and
species loss. Cline estimates total damage from CO2-equivalent doubling the
amount to about $61 billion,† or about 1.1% of the Gross Domestic Product
(GDP). Intangible losses such as species loss and decline in human quality of life
could raise the total to about 2% of GDP. If CO2 doubling results in a temperature
increase of 4.5˚C rather than 2.5˚C, the corresponding damage could be as high
as 4% of GDP, with even greater losses in some other countries such as low-lying
island nations.
Some of the revenue derived from the tax could be channeled toward
improvements in public transportation, development and/or subsidization of more
environmentally benign energy sources, and research directed toward efficiency
improvements. Cline (4) suggests that some of the revenue be channeled to de-
veloping countries “to secure their participation in international abatement. . . .
The importance of including developing countries in international measures for
restraining and reducing emissions, and the political and equity considerations
that seriously limit the amount of growth these countries can be expected to
sacrifice to help avoid global warming, strongly point to the need to channel some
of the revenue from a carbon tax from industrial countries to assist developing
countries that are prepared to take measures to reduce deforestation and configure
future energy development along lines that minimize carbon emissions.”

*Figures are in 1990 dollars and are based on a doubling of CO2-equivalent resulting in an
approximate temperature increase of 2.5˚C; concentrations of more than double preindustrial levels
obviously would result in even higher costs.
†In contrast, Tenenbaum (27) cites a 1991 report that says the external costs of energy currently range
from $100 billion to $300 billion in the United States.
Energy Conservation 111

5.2 Establishing Emissions Caps and Trading Programs

Establishing emissions caps and trading programs would be similar to the
imposition of limits on sulfur emissions established by the 1990 Clean Air Act
Amendments (CAAA); a brief discussion of these limits is therefore warranted.
The CAAA established an absolute cap on sulfur dioxide (SO2) emissions
by electrical utilities of 8.95 million tons after an initial reduction of 10 million
tons; it is assumed that this cap is sufficient to protect ecosystem health. Under
the technology-forcing regulatory approach of the past, each utility would have
been required to install a technology that reduced emissions by an amount
sufficient to achieve the 10-million-ton reduction. Economists have argued that
this approach results in higher control costs than necessary. Different utilities are
likely to incur different control costs due to age and technological differences in
their facilities (i.e., one utility may have a much lower per-ton incremental cost
for emission reduction than another).
The 1990 Clean Air Act Amendments provides for the issuance of permits
to utilities equal to 30–50% of their emissions 10 years earlier. Utilities whose
per-ton incremental costs for emission reduction are low can reduce emissions
beyond the level required for permit compliance and then sell surplus permits. In
turn, utilities whose incremental costs are high can reduce their control costs by
purchasing permits from utilities whose incremental costs are low. The end result
is achievement of the desired level of SO2 emissions reduction without imposing
unreasonable economic burdens on utilities while, at the same time, providing an
economic incentive for industries to reduce their SO2 emissions.
A similar program could be developed and used for carbon emissions.
The cap for carbon emissions could be based on the degree to which coun-
tries would like to limit global warming. For instance, freezing global carbon
emissions at the current level of about 6 billion tons (gigatons, or GtC) would
limit warming to about 5˚C (1). Capping emissions at 4 GtC would limit
warming to 2.5˚C. Carbon emissions could by reduced by as much as 20–25%
through energy efficiency improvements and substitution of non-carbon en-
ergy sources (both of which are discussed later) at zero net economic cost
with significant economic benefit to those companies involved in this trading
program (4).
According to Cline (4), it is widely believed that a system of tradable
permits can be applied globally, on a country-by-country basis, in much the same
manner as would a carbon tax. If a country has a quota allocation that is small
relative to its demand, its firms could bid to purchase quotas from other countries.
Other countries could sell a portion of their quotas at a price that would equal or
exceed the cost of reducing their overall carbon emissions. Thus, global carbon
emissions could be reduced through an economic incentive program that would
reward countries that reduce their overall carbon emissions.
112 Strevett et al.

Booth (1), on the other hand, believes that permits issued on an individual
basis rather than by country would be more effective:
Permits could be domestically distributed annually on a per person basis
equal in amount to existing emissions initially, and then reduced by 3.6
percent of the initial amount each year over a phase-in period of
approximately 25 years to arrive at a 90 percent total reduction. Individ-
uals who don’t need the full allocation for their own energy consumption
could sell their surplus permits at the going market price. Such a system
would tend to redistribute income away from industries and high-income
families who are heavy consumers of energy to low-income families
who tend to consume less energy. Because of the potential to sell surplus
permits, the public resistance to a permit system would be less than to a
carbon tax. The rising price of permits over time would provide the
incentive needed for increased energy conservation and to shift to
non-fossil fuel energy sources. As in the case of acid rain control, a
marketable permit system for carbon emissions control results in control
being achieved at the lowest possible cost (1). p. 23
Either of the above strategies would constitute impetus for increases in
efficiency and other conservation measures. Both taxes and tradable permits
minimize overall abatement costs by allocating the cutbacks to the countries
where marginal costs of emissions reductions are the lowest. A major difference
between the two strategies is that, with tradable permits, it is possible to specify
the exact cutback in emissions (4). Cline (4) believes the best strategy to be
reliance on nationally set carbon or greenhouse gas taxes during an initial
phase-in period and then, in a subsequent phase, to set the taxes at an inter-
nationally agreed rate while each individual nation would continue to collect
them. If such taxes failed to achieve satisfactory progress toward global emis-
sion targets, it would then be appropriate to shift to an international system of
tradable permits.

5.3 Elimination of Subsidies

5.3.1 International Subsidies
For some years, the World Bank (33) has been drawing attention to the fact that
electricity is sold in developing countries at, on average, only 40% of the cost of
its production. A recent study pointed out:
Such subsidies waste capital and energy resources on a very large scale.
Subsidizing the price of electricity is both economically and environ-
mentally inefficient. Low prices give rise to excessive demands and, by
undermining the revenue base, reduce the ability of utilities to provide
Energy Conservation 113

and maintain supplies. Developing countries use about 20 percent more

electricity than they would if consumers paid the true marginal cost of
supply. Underpricing electricity also discourages investment in new,
cleaner technologies and more energy efficient processes (16). p. 12
Shah and Larsen (1991, as cited in Ref. 4) estimated that nine large
developing and Eastern European countries (China, Poland, Mexico, Czecho-
slovakia, India, Egypt, Argentina, South Africa, and Venezuela) spend a combined
$40 billion annually in subsidization of fossil fuels (with China’s* $15.7 billion
the largest). The former Soviet Union spends more than twice this amount—$89.6
billion annually—on fossil fuel subsidies. The removal of these subsidies would
eliminate an estimated 157 million tons of carbon annually from the developing
group and 233 million tons from the former Soviet Union alone. These cutbacks
would represent about 8% of global carbon emissions (or about 6% if deforesta-
tion emissions are included).
Prices that cover production costs and externalities are likely to encourage
efficiency, mitigate harmful environmental effects, and create an awareness
conducive to conservation. Subsidized energy prices, on the other hand, are one
of the principal barriers to raising energy efficiency in developing countries,
where it is only 50–65% of what would be considered best practice in the
developed world. Studies indicate that with the present state of technology a
saving of 20–25% of energy consumed would be achieved economically in many
developing countries with existing capital stock. If investments were made in
new, more energy-efficient capital equipment, a saving in the range of 30–60%
would be possible (9).
5.3.2 U.S. Subsidies
According to Ackerman (30), two studies have attempted to measure federal
energy subsidies. The Department of Energy’s Energy Information Administra-
tion identifies subsidies worth $5–$13 billion annually, while the Alliance to Save
Energy, an energy conservation advocacy group, estimates energy subsidies at
$23–$40 billion annually (in 1992 dollars). Ackerman also states that several
provisions of the tax code are, effectively, subsidies to the oil and gas industry
and that, depending on one’s view of a local tax controversy, the total subsidy to
oil and gas production alone might be as much as $255 million, almost 5% of
sales in 1990.

*China accounts for 11% of global carbon emissions, excluding emissions from deforestation. Seventy
percent of China’s energy comes from coal (4).
114 Strevett et al.

5.4 Increases in Energy Efficiency

Primary energy is defined as the energy recovered directly from the Earth in the
form of coal, crude oil, natural gas, collected biomass, hydraulic power, or heat
produced in a nuclear reactor from processed uranium. Generally, primary energy
is not used directly but is converted into secondary energy (9). The process of
energy conversion and transformation results in part of the energy being wasted
as heat. Energy efficiency considerations focus on the following factors:
The efficiency of original extraction and transportation
The primary energy conversion efficiency of central power plants, refiner-
ies, coal gasification plants, etc.
The secondary energy conversion efficiency into storage facilities, distribu-
tion systems and transport networks (e.g., of electricity grids)
Efficiency of final energy conversion into useful forms such as light and
motion (9)
For the world as a whole, the overall efficiency with which fuel energy is
currently used is only around 3–3.5% (17). According to Orr (32), a Department
of Energy study showed that U.S. energy consumption could be reduced by 50%
with present technologies with a net positive economic impact. The United States
did indeed reduce the energy intensity of its domestic product by 23% between
1973 and 1985 (18).
5.4.1 The Industrial Sector
The industrial sector in the more advanced industrial countries is the most
efficient energy user. It is easier to be efficient when operating on a larger scale
and when energy is an explicit element of operating costs. Profit margins mandate
careful cost analysis, and in industries where energy costs comprise a significant
portion of total costs, managers are more alert to opportunities for savings (9).
According to the Office of Technology Assessment (1991, as cited in Ref. 2) four
sectors—paper, chemicals, petroleum, and primary metals—account for three-
fourths of the energy used in manufacturing. More than half the energy consumed
by industry in the leading industrial countries is as fuel for process heat, and over
one-fifth (gross) is in the form of electricity for furnaces, electrolytic processes,
and electric motors. Most process heat is delivered in the form of steam, with an
overall efficiency variously estimated to be between 15% and 25%. The biggest
users of process heat are the steel, petroleum, chemicals, and paper and pulp
industries (9).
Potential for improvements does exist. In general, sensors and controls,
advanced heat-recovery systems, and friction-reducing technologies can decrease
energy consumption (5). Many efficiency measures are specific to each industry.
For instance, the World Energy Council (9) offers several options for improving
Energy Conservation 115

efficiency in the chemical industry, including the use of biotechnology and

catalysts (Table 1).
In the paper industry, automated process control, greater process speeds,
and high-pressure rollers can boost efficiencies significantly (5). According to
Carlsmith et al. (1990, as cited in Ref. 4), electric arc furnaces using scrap are
much more energy efficient for steel production than are traditional techniques
and could increase their share of output from 36% to 60%. According to Cline
(4), these authors also estimate that by 2010, direct reduction or smelting of ore
for making iron would reduce energy requirements in steelmaking by 42% with
a net cost savings. Even greater opportunities exist for improving energy effi-
ciency in developing countries: for example, China and India use four times as
much energy as Japan does to produce a ton of steel (5).
In aluminum production, energy efficiency can be increased by improved
design of electrolytic reduction cells, recycling, and direct casting. Other exam-
ples of improvements in industrial processes include low-pressure oxidation in
industrial solvents, changes in paper-drying techniques (as well as paper recycl-
ing), and shifting from the wet to the dry process in cement making (4).
Co-generation, the simultaneous production of both electricity and steam or
hot water, represents a great opportunity for improving energy efficiency in that
the net energy yield from the primary fuel is increased from 30–35% to 80–90%.
In 1900, half the United States’ electricity was generated at plants that also
provided industrial steam or district heating. However, as power plants became
larger, dirtier, and less acceptable as neighbors, they were forced to move away
from their customers. Waste heat from the turbine generators became an unwanted
pollutant to be disposed of in the environment. In addition, long transmission
lines, which are unsightly and lose up to 20% of the electricity they carry, became
necessary. By the 1970s, co-generation had fallen to less than 5% of our power

TABLE 1 Options for Improving Efficiency in Chemical Industry

Options Benefits

Biotechnology Speed reaction times

Reduce necessary temperatures
and pressures
Catalysts Improve yields and reaction times
Reduce necessary temperatures
and pressures
Separation and concentration Improve product purity
Waste heat management Reduce necessary temperatures
and pressures
116 Strevett et al.

supply, but interest in this technology is being renewed, and the capacity for
co-generation has more than doubled since the 1980s.
5.4.2 Buildings
In developed countries, buildings are the largest or second-largest consumers of
energy. In the United States, buildings account for about 75% of all electricity
consumption (19) and about 35% of total primary energy consumption (3); most
of this is for heating and cooling. Electricity generation alone produces more than
25% of energy-related carbon dioxide emissions (20). Building improvements
could therefore have a major impact on overall energy consumption and carbon
emissions.
In a “typical” North American house, the average efficiency of insulation
is about 12% compared with the ideal. As a result, the overall energy efficiency
of air cooling systems has been estimated to be barely 5%, and the overall energy
efficiency for space heating is less than 1%. These figures do not take into account
avoidable losses through heating or cooling unoccupied rooms (9).
Building design is one of the simplest yet most effective ways to take
advantage of solar energy. Buildings can incorporate either passive or active solar
technologies. Passive solar heating and cooling function with few or no mechan-
ical devices; primarily they involve designing the form of landscape and building
in relation to each other and to sun, earth, and air movement (19). In general,
passive technologies use a building’s structure to capture sunlight and store heat,
reducing the requirements for conventional heating and lighting. Heating can be
cut substantially by the use of one or several technologies in the building’s design
(Table 2). When included in a building’s initial design, these methods can save
up to 70% of heating costs (21). Orr (32) points out that it is cheaper and less
risky by far to weatherize houses than it is to maintain a military presence in the
Persian Gulf at a cost of $1 billion or more each month.*
Cooling needs also may be reduced by passive means; one strategy is the
reduction of internal heat gains. Another passive strategy for reducing cooling
needs is by reduction of external heat gains. Several technologies that can be used
to reduce internal and external heat gains are listed in Table 2. Also, it is important

*Nearly one-quarter of all jet fuel in the world, about 42 million tons per year, is used for military
purposes. The Pentagon is considered to be the largest consumer of oil in the United States and perhaps
in the world. One B-52 bomber consumes about 228 liters of fuel per minute; one F-15 jet, at peak
thrust, consumes 908 liters of fuel per minute. It has been estimated that the energy the Pentagon uses
up annually would be sufficient to run the entire U.S. urban mass transit system for almost 14 years.
Further, it has been estimated that total military-related carbon emissions could be as high as 10% of
emissions worldwide, and that between 10% and 30% of all global environmental destruction can be
attributed to military-related activities (28).
Energy Conservation 117

TABLE 2 Technologies for Increasing a Building’s Energy Efficiency

Area for improving
energy efficiency Technology

Heating Heat-circulation systems using natural convective

forces
Heat pumps
Solar-thermal collectors
Insulated windows and shutters
Special window glazings
Heat-storing masses built into structure
Building orientation
Draft proofing
Superinsulation of structure
Cooling Fluorescent lighting over incandescent
Lower-wattage bulbs
Landscaping that provides maximum shade
Window shades
Reflective or tinted window coatings
Insulated windows
Light-colored roofs
Ventilation by natural convection
Ground absorption of heat

to trade in old, wasteful for newer, more efficient ones; the payback period may
be as little as two to three years (3).
One measure proposed in several developed countries is to require all
houses to be subject to an energy efficiency survey that would lead to an energy
efficiency rating which would have to be disclosed to prospective buyers when
the house is sold (9).
5.4.3 Lighting
About 40–50% of the energy consumed in a typical house is used for heating and
cooling, with an additional 5–10% used for lighting. Lighting is the least efficient
common use of energy: about 95% of the energy used in an average lighting
system dissipates as heat (19). Incandescent bulbs have an efficiency of about 4%
in converting electricity to visible radiant energy. In contrast, the efficiencies of
fluorescent lights is typically around 20%, and can be as high as 35% (9).
According to Lovins and Lovins (1991, as cited in Ref. 4), a 15-W compact
fluorescent bulb emits the same amount of light as a 75-W incandescent bulb and
lasts 13 times as long. Further, over its lifetime, it can save enough coal-fired
118 Strevett et al.

electricity to reduce carbon emissions by 1 ton with a net savings. The National
Academy of Science (1991, as cited in Ref. 4) contends that the replacement of
an average of just 2.5 heavily used interior incandescent bulbs and one exterior
bulb by compact fluorescent lights would reduce average household lighting
energy requirements by 50%. Why, then, do we continue to use incandescents?
Lack of awareness
Easy commercial availability or promotion
High first cost
High replacement cost in the event of breakage
Cost and inconvenience of retrofitting new lighting systems to existing
domestic buildings, where rewiring and new sockets, holders, and appli-
ances may be needed (9)

5.4.4 Government’s Role

MacNeill (31) contends that, in order to make steady gains in energy efficiency,
governments must institute politically difficult changes in at least three areas:
1. Countries must consider “conservation pricing,” i.e., taxing energy
during periods of low real prices to encourage increases in efficiency.
2. Stricter regulations should demand steady improvement in the effi-
ciency of appliances and technologies, and in building design, auto-
mobiles, and transportation systems. [In the United States, efficiency
standards for appliances were adopted in 1986. For refrigerators, the
biggest users of electricity in most households, the energy efficiency of
new models almost tripled from 1973 to 1993 (22).]
3. Institutional innovation will be necessary to break utility-supply
monopolies and to reorganize the energy sector so that energy services
can be sold on a competitive, least-cost basis.
In addition, governments should excise policies that retard the development of
new and renewable energy resources, particularly those that serve as substitutes
for fuelwood.
5.4.5 Caution
As a final word on the issue of efficiency, it is worthwhile to quote Cline (4):
In reaching the overall conclusion that some 20 percent to 25 percent of
carbon emissions in the United States might be eliminated at zero cost
by a move to “best practices,” it is important that there not be a
misguided inference that dealing with the greenhouse problem will be
cheap over the longer term. . . . Serious action to curb global warming
would involve emissions restraints over a period of two to three centu-
Energy Conservation 119

ries. . . . Whether the first step is low-cost (or even no-cost) is significant
but of limited help in gauging the eventual costs.
The central point is that a one-time gain from elimination of
inefficiencies would shift the entire curve of baseline emissions down-
ward but still leave future emissions far above present levels. Consider
the period through the year 2100 . . . a central baseline estimate calls for
approximately 20 GtC of global carbon emissions by that year . . . an
aggressive program to limit global warming would mean restricting
emissions to approximately 4 GtC annually. Suppose the engineering
approach is correct that, 20 percent of emissions can be eliminated for
free. Such gains would still leave emissions at 16 GtC in the year 2100,
far above the 4-GtC ceiling needed to substantially curb the greenhouse
effect. The remaining cutbacks would have to be achieved through more
costly industrial reductions in energy availability beyond those achiev-
able through costless efficiency gains. In short, the “best practices”
school provides a basis for expecting that addressing the global warming
problem may be less costly than otherwise might be thought, but it by
no means warrants the conclusion that action will be costless over the
longer term (4).

5.5 Energy Conservation in Transportation

Transport activities account for about 30% of the energy used by final consumers,
and about 20% of the gross energy produced (9). About 98% of the total comes
from petroleum products refined into liquid fuels, and the remaining 2% is
provided by natural gas and electricity (3). Movement of people takes about 70%
of the total, and movement of freight about 30%. Within this sector, road transport
accounts for the largest proportion, over 80% in industrialized countries, with air
transport next, at 13% (9). According to the United Nations Fund for Population
Activities (29), the world car fleet increased by seven times between 1950 and
1980 while human population only doubled during that period. Fifteen percent of
the world’s oil is consumed by automobiles and light trucks in the United States
alone (Office of Technology Assessment, 1991, as cited in Ref. 4).
About 75% of all freight in the United States is carried by trains, barges,
ships, and pipelines, but because they are very efficient, they use only 12% of all
transportation fuel (3). The rapid increase in road transport in recent years is a
major contributor to the rise in oil demand. Further, motor vehicles are believed
to be responsible for 14% of all CO2 derived from fossil fuel combustion (9),
along with their contribution to acid rain and other forms of air pollution such
as O3.
The Reagan administration relaxed automobile efficiency standards that had
already been met by Chrysler. If the regulations had been left in place, the amount
120 Strevett et al.

of gasoline saved in a decade or so would have been equivalent to the entire

amount of oil estimated to underlie the Arctic National Wildlife Refuge (23).
Gasoline prices in Europe and Japan are double or triple the U.S. price
because governments there impose levies that force consumers to consider and
internalize the full costs of their behavior (5). The gradual imposition of a
significantly higher gasoline tax, until the cost of gasoline in the United States is
comparable to that in Europe, would create a powerful incentive for people to
drive smaller, more fuel-efficient cars and use energy-efficient alternative forms
of transportation. Highways and bridges would last longer, and emissions would
be reduced, attenuating global warming and acid rain. This would, of course,
necessitate improvement of public transportation to accommodate people who
could no longer afford to drive to work; some of the gas-tax funds could be set
aside for this. In the United States, mass transport accounts for only 6% of all
passenger travel; in Germany the figure is over 15% and in Japan it is 47% (9).
Another possibility for internalization of the many hidden costs of driving
would be the implementation of an insurance program based on the average
number of miles a driver travels. This would link a portion of drivers’ insurance
programs to the number of miles they drive and collect payments at the gas pump
(12). Ledbetter and Ross (11) provide the details of such an arrangement:
The price of gasoline at the pump could include a charge for basic,
driving-related automobile insurance that would be organized by state govern-
ments and auctioned in blocks to private insurance companies. All registered
drivers in the state could automatically belong. Supplementary insurance above
that provided by the base insurance purchased at the pump could be indepen-
dently arranged, as we presently do for all our insurance. For example, owners of
expensive cars, or people who desire higher levels of liability coverage, could
purchase supplemental insurance. Drivers with especially bad driving records
could be required to purchase supplemental liability insurance. Below are some
of the advantages of such an arrangement.
Insurance costs become much more closely tied to the amount of driving
alone. The more miles a person drives, the more insurance he or she pays.
Since accident exposure is closely correlated with miles driven, the
proposed system would be more fair than the present system, in which
people who drive substantially less than the average miles per year are
given only small discounts, and people who drive substantially more than
the average don’t pay any additional premium.
If insurance were part of the cost of gasoline, a person could not drive
without paying for insurance. Uninsured motorists would be brought into
the system, substantially lowering the cost of driving for insured motor-
ists: in California for example, uninsured motorists increase premiums
for insured motorists by about $150 per year.
Energy Conservation 121

The apparent cost of gasoline at the pump would rise substantially, roughly
50 cents to a dollar per gallon. Such a price rise would encourage the
purchase of more fuel-efficient vehicles and help slow the growth in
vehicle miles of travel. For consumers, the increase in the price of fuel
would be offset by a decrease in the annual insurance premium motorists
would pay directly to insurance companies, resulting in no net increase
in driving costs.
Unlike a gasoline tax, this system would not be regressive: many low-
income persons drive substantially less miles per year than their higher-
income counterparts. They would, therefore, see a substantial drop in the
money they pay for auto insurance (11).
5.5.1 Efficiency Issues in Transportation
The efficiency of a motor vehicle is a function of several factors (Table 3).
Typically, about 80% of the fuel used in a representative vehicle traveling over a
mix of urban, rural, and highway routes is unproductive energy spent in overcom-
ing internal friction in auxiliary items and in thermodynamic losses in the engine
(9). Improvements in vehicle design and alternative fuels can have a major impact
in improving efficiency and reducing emissions. However, much of the forward
momentum achieved in the decade prior to 1985 has slowed in response to
downward oil price movements and apparent consumer preferences (9).
The inherent efficiency of the internal combustion engine began to ap-
proach its limits in the 1960s. Engines built since then range from 34% efficiency
for spark-ignition automobile-type engines under optimum load/speed conditions
to about 42% for large marine-type and direct-injection diesels. The difference is
attributable to the higher compression ratios, lower throttle losses and improved
direction injection achievable in large diesels.
In practice, however, optimum load/speed conditions are never achieved.
The energy efficiency of a vehicle operating in traffic, with variable speeds and

TABLE 3 Factors Affecting a Motor Vehicle’s Fuel Efficiency

Factor Components

Design Weight
Efficiency
Frictional losses
Aerodynamics
Use Effectiveness of use in transporting
materials and people
Typical operational cycle Length of journey
Traffic conditions
122 Strevett et al.

loads, is at least 30% lower. Short journeys, when the engine is cold at start-up
and never warms up sufficiently for optimal fuel combustion, create suboptimal
fuel use and high emissions. Stop/start conditions in heavy traffic also cause
relatively high fuel use and emissions (9).
Engine efficiency is further reduced, often by an additional 30% or so,
by the carrying of oil pumps, air pumps, fuel pumps, electrical systems, heat-
ing, air conditioning, and other related equipment. Friction and viscosity losses
in the vehicle’s drive train—e.g., in automatic transmissions, which alone can
reduce engine efficiency by 10–15%, cut efficiency still further. As a result, the
average thermodynamic efficiency of the motor vehicle is only between 10%
and 17%.
Nevertheless, significant improvements in automobile fuel economy have
been achieved in recent years. The biggest gains have been made by cutting down
on excess weight in the body, improving aerodynamics, and improving tires. Still,
the “payload efficiency” of a medium-sized car is only about 10%, while that of
fully loaded commercial aircraft is around 30–35%. Heavy-duty trucks, freight
trains, and ships also achieve greater payload efficiencies than cars (9).
Raising the average fuel efficiency of the U.S. car and light truck fleet by
1 mpg would cut oil consumption about 295,000 bbl per day. In one year, this
would equal the total amount the Interior Department hopes to extract from the
Arctic National Wildlife Refuge in Alaska (3). Increased fuel efficiency can
be supplemented by savings from transportation management, including in-
creased mass transit, carpooling, and improved maintenance (including proper
tire inflation) (4).

5.6 Increased Exploitation of Natural Gas

Increased exploitation of natural gas in preference to coal or oil as an interim
measure has the potential to slow global warming as non-hydrocarbon primary
energy sources are developed and put into place. Natural gas provides about
one-fifth of global commercial energy and is our most efficient “traditional”
energy source. Only about 10% of its energy content is lost in shipping and
processing, since it moves by pipelines and usually needs very little refining.
Ordinary gas-burning furnaces are about 75% efficient, and high-economy fur-
naces can be as much as 95% efficient (3). It generates fewer pollutants than any
other traditional fuel and less CO2 as well: 42% less than coal and 30% less than
oil (5).
According to Gibbons et al. (5), some analysts feel that the most promising
future option for electric power generation is the aeroderivative turbine, which is
based on jet engine designs and burns natural gas. With additional refinement,
this technology could raise conversion efficiency from its present 33% to more
than 45%.
Energy Conservation 123

North America has a pipeline network for delivering natural gas to market.
However, most countries cannot afford a pipeline network, and much of the
natural gas that comes out of the ground in conjunction with oil pumping is
simply burned (flared off), a terrible waste of a valuable resource (3).
Natural gas is quite easy to ship through pipelines as long as it is going
from one place to another on the same continent. The problem is that much of
the gas is in Russia or the Middle East, while the markets are in Europe, Japan,
or North America. One way of shipping gas across oceans is to liquefy it by
cooling it below its condensation point (–140˚C). Liquefied natural gas (LNG)
has only 1/600 the volume of the gaseous form, and is therefore economical to
transport by tanker ship. However, if a very large LNG tanker had an accident
and blew up, it would release as much energy as several Hiroshima-sized atomic
bombs (3).

5.7 Increased Exploitation of Passive Technologies

Because most paved surfaces, and the surfaces of most buildings, tend to retain
and release more heat than is true of vegetated areas, and because heating and air
conditioning equipment releases/generates a great deal of heat, urban areas
typically are several degrees warmer than vegetated areas. For example, an early
study of this subject showed downtown St. Louis to be 13˚F warmer in the winter
and 9˚F warmer in June than the large, tree-canopied Forest Park, 5 miles away.
Tree cover can moderate this “heat island effect,” helping to control micro-
climate in three different ways:
1. Absorption and reflection of solar radiation. A tree in full leaf intercepts
between 60% and 90% of the radiation that strikes it, depending on the
density of its canopy. Clusters of trees spaced closely together can
therefore reduce ambient summer temperature significantly. Placed
directly adjacent to buildings on the east, west, and south sides, they
can reduce incoming solar radiation in the summer and, if deciduous,
allow most of it to pass through in the winter, when a deciduous tree
intercepts only 25–50%.
2. Creation of a “still zone” under the canopy. Around the edges of a tree
canopy is a band of air turbulence where the cooler air within and the
warmer outside air meet and mix. This turbulent zone appears to form
a containing frame for the still, cool air beneath the canopy.
3. Release of cooling water vapor from their leaf surfaces through evap-
oration and transpiration (19).
A study of a mobile home in Florida showed that well-placed plantings
could reduce cooling costs by more than 50% (Hutchinson et al., 1983, as cited
in Ref. 19). Calculations of electrical energy saved by tree planting suggest that
124 Strevett et al.

this is one of the most cost-effective means of reducing the heat island effect and
thus electrical energy consumption (19). According to McPherson (1990, as cited
in Ref. 19), about 97% of the total carbon conserved annually by a tree is in
reduced power-plant emissions resulting from reduction in electrical energy use
rather than in carbon dioxide absorbed.

6 POLLUTION PREVENTION VIA CHOOSING

REPLACEMENTS FOR FOSSIL FUELS

6.1 Introduction
Despite potentially significant technological improvements in efficiency and
decreases in environmental impact, some of the inefficiencies and pollutants
associated with traditional energy sources cannot be avoided. Uneven distribution
of resources can increase transportation costs, which can amount to 25% or more
of the cost of crude oil, for example (9). Indeed, about 75% of the original energy
in crude oil is lost during distillation into liquid fuels, transportation of that fuel
to market, storage, marketing, and combustion in vehicles (3). For this reason,
alternative energy sources such as solar, geothermal, and wind should receive
much more attention.
In the United States, “renewable” energy sources account for about 7.5%
of total consumption. The vast majority of this energy comes from two sources
that have reached commercial maturity: hydroelectric power and biofuels (24).
Currently, biofuels, primarily wood, account for about 4% of the U.S. energy
supply. More than 6% of all homes burn wood as their principal heating fuel. The
paper and pulp industry burns wood scraps to provide heat and electricity to run
its operations. Wood and other biofuels are also used to generate a small amount
of electricity by utilities (6).
Worldwide, potentially sustainable or renewable energy resources, includ-
ing solar, biomass, hydroelectric, and other, less developed types of power
production, currently provide less than 3% of total energy use (3). As of 1990,
traditional biomass (e.g., fuelwood, crop residues, and dung) accounted for 60%
of total available renewable energy, and large-scale hydropower for another 30%
(9). About half of all wood harvested in the world annually is used for fuelwood;
many countries use fuelwood (including charcoal) for more than 75% of their
nonmuscle energy. About 40% of the world’s total population depend on firewood
and charcoal as their primary energy source. In some African countries, such as
Rwanda and Sudan, firewood demand is already 10 times the sustainable yield of
remaining forests (3). These figures illustrate the enormity of the potential for
environmentally benign energy sources such as solar to replace not only fossil
fuels but also traditional renewables which also cause environmental harm.
Energy Conservation 125

6.2 Biomass
As recently as 1850, wood supplied 90% of the fuel used in the United States.
Wood now provides less than 1% of energy in the United States, but in many of
the world’s poorer countries, wood and other biomass fuels provide up to 95%
of all energy consumed. Approximately half of all wood harvested annually is for
fuel. About 40% of the world’s population depend on firewood and charcoal as
their primary energy source; however, about three-fourths of these lack an
adequate, affordable supply (3).
In wood-burning power plants, pollution-control equipment is easier to
install and maintain than in individual home units. Wood burning contributes less
to acid precipitation than does coal, as wood contains little sulfur and burns at
lower temperatures than coal, resulting in the production of fewer nitrogen oxides.
However, unless trees cut for fuel are replaced with seedlings, wood burning
results in a net increase in atmospheric CO2.
Inefficient and incomplete burning of wood in stoves and fireplaces pro-
duces smoke laden with fine ash and soot and hazardous amounts of carbon
monoxide and hydrocarbons. The U.S. Environmental Protection Agency (EPA)
ranks wood burners high on a list of health risks to the general population, and
standards are being considered to regulate the use of woodstoves nationwide.
Highly efficient and clean-burning woodstoves are available but expensive (3).

6.3 Hydroelectric Dams

As of 1987, hydroelectric dams in the United States provided the energy equiva-
lent of about 71 large power plants, about 10–14% of U.S. electricity, or about
3% of total energy supply, depending on year-to-year rainfall patterns. Of the
pollutants associated with fossil fuel energy, methane is the only one that results
from the damming of rivers. However, large dams have drowned out some of the
most beautiful stretches of American rivers, flooded agricultural lands, forests,
and areas of historical and geological value, and resulted in the dislocation of
communities and loss of wildlife (6). Dam failure can cause catastrophic floods
and thousands of deaths. Sedimentation often fills reservoirs rapidly and reduces
the usefulness of the dam for either irrigation or hydropower (3).

6.4 Synthetic Fuels

Methanol would provide little reduction in greenhouse gases if made from natural
gas (Office of Technology Assessment, as cited in Ref. 4). Synthetic fuels derived
from coal or oil shales would result in the release of even more CO2 than coal
because the conversion processes require so much energy (6). The use of
compressed natural gas brings the potential for leaks of methane that could largely
offset the lesser carbon content of natural gas when compared to oil.
126 Strevett et al.

Ethanol (grain alcohol) and methanol (wood alcohol) are produced by

anaerobic digestion of plant materials (Figure 7). Ethanol is unlikely ever to play
an important energy role in our transportation future: 8% of the U.S. corn crop
would replace only 1% of U.S. gasoline. Further, making ethanol from corn
requires almost as much energy as the ethanol contains; therefore, it offers little
if any global-warming benefit (6). Among biomass fuels, synthetic natural gas or
methanol produced from woody biomass hold the largest potential for reducing
greenhouse gases (a reduction of 60–70% from that emitted by vehicle fuels used
at present), so long as the feedstock were offset with replacement biomass growth
(Office of Technology Assessment, as cited in Ref. 4).

6.5 Tides
The stormy coasts where waves are strongest are usually far from major popula-
tion centers that need the power. In addition, the storms that bring this energy can
destroy the equipment intended to exploit it (3). Even if the technology for
capture of tidal energy were available, only a minute fraction could, even
theoretically, be harnessed for useful purposes (25). France operates a tidal
generating station on the Rance Estuary that is designed to produce 240 MW of
electricity but that usually only generates 62 MW (26).

R a w M a ter ia ls
E th a n o l o r
M e th a n o l
Pr o d u c tio n

A n a e r o b ic
H ot D ig e sto r
W a te r Slu d g e
Su p p ly

Gas
Su p p ly

FIGURE 7 Production of ethanol and methanol through anaerobic digestion

of plant material.
Energy Conservation 127

6.6 Nuclear Power

Nuclear power provides about 17% percent of the world’s electricity (5) and about
5% of total energy needs, led by Western Europe with about 11% reliance on
nuclear plants (3). Although the United States has the world’s largest nuclear
power program, it provides only about 7.5% of our energy needs. In the United
States, at least, the management and operation of existing plants must improve
significantly, and existing unresolved safety problems must be convincingly
solved. The design of new reactors must be simplified and incorporate more
passive shut-down safety features. Further, there must be tangible progress in
solving the problems of storing radioactive wastes (6).
However, even if all these requirements were met, the potential contribution
of nuclear energy to solving the global energy–climate problem would be limited
for several reasons:

It is unlikely that a significant number of safer new reactors can be

designed, approved, constructed, operated, and “debugged” in a rela-
tively short period of time—say, less than 20 years. They will therefore
be unable to make a significant contribution to meeting the world’s
energy needs during the next 20–40 years.
Because of their inherent cost and complexity, nuclear plants are unlikely
to be deployed in poor, developing countries. Such facilities demand a
high level of sophisticated and expensive support to be safely constructed
and operated, a condition unlikely to occur in most of the developing
world.
Unless the world suddenly embraces peaceful solutions to its age-old
ethnic and boundary problems, the prospect of nations using nuclear
materials to build weapons clandestinely will grow with nuclear plant
deployment (6).
Of the nuclear plants that have been decommissioned so far, the costs of
tearing them down have been about two to ten times the costs of building
them in the first place.
We may never reach a breakeven point where we get back more energy
from nuclear plants than we put into them, especially considering the
energy that may be required to decommission nuclear plants and guard
their waste products in secure storage for thousands of years (3).
The raw materials required for nuclear fuels result in the same disturbances
of the landscape as other mined minerals.

Denmark has never permitted atomic power plants to be constructed within

its boundaries, and Sweden has a policy of decommissioning all its existing
plants. In the United States, few plants currently are under construction because
it has become so costly due to required environmental safeguards and the
128 Strevett et al.

inevitable litigation of nuclear opponents. Further, in most parts of the United

States, it is so politically risky that no new plans for nuclear power plant
construction are currently in existence (14).

6.7 Geothermal Energy

Geothermal energy is heat contained below the earth’s surface, either in rock or
in trapped hot water or steam. Geothermal power offers a number of environmen-
tal advantages. When compared with other alternative energy sources, geothermal
plants are reliable; the Department of Energy reports that they have a 65%
“capacity factor” (the ratio of actual output to the output that would result if the
plant ran full-tilt, full time). This is comparable with the capacity factors of new
coal or gas turbine plants. In contrast, the capacity factor of wind and solar
thermal plants is about 21%. Geothermal energy produces no ash, no scrubber
waste, and no radioactive waste. Although geothermal energy sometimes pro-
duces toxic waste from the dissolved or suspended chemicals naturally found
deep in the earth, these materials tend to be more easily disposed of than those
from other energy sources; virtually all U.S. generating plants simply reinject
them into the reservoir (27).
Geothermal power, however, suffers from resource, technological, and
economic constraints. The only type of geothermal energy that has been widely
developed is hydrothermal energy, which consists of trapped hot water or steam
(24). The total geothermal energy of the world’s volcanoes and hot springs is only
about 2% of today’s global commercial energy use. This energy flux can be
utilized in hyperthermal areas such as Iceland (where most buildings are heated
by geothermal steam) or in the “Ring of Fire” surrounding the Pacific Ocean,
where 18 nations (including the western United States) currently generate geo-
thermal electricity (27). Hydrothermal cannot, however, be of more than local or
regional importance (25).
Creating a geothermal plant is expensive, because developers usually must
bore holes a mile or so deep through hard rock. Even though geothermal plants
need no fuel, making operating costs extremely low, the capital cost still amounts
to about $3000 per kilowatt, in contrast for about $824 per kilowatt for an
efficient gas turbine plant. Innovations such as new drilling technologies promise
to cut expenses (27).
Other problems associated with the use of geothermal power include the
following:

Geothermal facilities are very large-scale plumbing pipes with an abun-

dance of giant pipes, huge valves, and specialized fittings. Some plants
need mufflers and sound blankets to reduce drilling and generating noise,
and they usually emit a plume of steam.
Energy Conservation 129

The rotten-egg stench of hydrogen sulfide released from underground can

often be overpowering (27).
The geothermal heat conducted by rocks is two and one-half times today’s
commercial energy use (25). Preliminary estimates of the cost of electricity
derived from hot dry rock (HDR) suggest that it might be relatively cheap, at least
in areas where the earth becomes at least 144˚F warmer with each mile of depth
and drilling costs are thus somewhat less formidable. Conditions like these
reportedly are found under 40,000 square miles of the lower 48 states, primarily
in Nevada, Oregon, and California (27). In view of the potentially catastrophic
effects of global warming, as well as the other environmental problems associated
with traditional energy sources, HDR-derived energy deserves serious study.

6.8 Wind Farms

Wind farms are large-scale public utility efforts to take advantage of wind power.
In 1990, wind machines in California generated enough electricity to meet the
annual residential needs of a city the size of San Francisco, or more than 1% of
California’s total electrical needs. There are enough windy sites in California to
meet about 20% of existing electricity demand. Advanced wind machines could
supply energy to the United States in amounts far in excess of the nation’s total
present energy demand.
The towers, roads, and other structures on a wind farm actually take up only
about one-third as much space as would be consumed by a coal-fired power plant
or solar thermal energy system to generate the same amount of energy over a
30-year period. The land under windmills is more easily used for grazing or
farming than is a strip-mined coal field or land under solar panels. Further, wind
power generates many more jobs per unit energy produced than do most other
technologies, even though its total cost is generally lower (3).
An obvious limitation of wind farms is the necessity of locating them in
windy areas. The best sites are in the Great Plains and include North and South
Dakota, Kansas, and Montana (6). Seacoasts also offer great potential for siting
wind farms. Opponents have complained of visual and noise pollution. While
most wind farms are too far from residential areas to be heard or seen, they do
interrupt the view in remote places and destroy the sense of isolation and natural
beauty. They can also pose a hazard to birds that fly into the whirling blades.

6.9 Solar Energy

Of all the available forms of energy, renewable or nonrenewable, solar has the
greatest potential for providing clean, safe, reliable power. The supply is in-
exhaustible: the solar energy falling on the earth’s continents is more than 2000
times the total annual commercial energy currently being used by humans (25).
130 Strevett et al.

Solar technologies can be broadly grouped into two categories:

1. Active technologies—solar thermal power plants, solar ponds, wind
turbines, and photovoltaic cells (Figure 8)
2. Passive technologies—natural materials or absorptive structures with
no moving parts that simply gather and hold heat (Figure 8) (3)
Low-temperature thermal collectors can provide heat for domestic hot
water, space heating, and industrial purposes (e.g., supplying hot water for car
washes). According to Cunningham and Saigo (3), water heating consumes 15%
of the U.S. domestic energy budget.
Active solar systems generally pump a heat-absorbing, fluid medium (air,
water, or an antifreeze solution) through a relatively small collector rather than
passively collecting heat in a stationary medium such as masonry (Figure 8) (3).
There are three main types of solar-thermal collectors: the parabolic trough, the
parabolic dish, and the central receiver.
Parabolic trough and parabolic dish units are modular and relatively small,
so that systems can be sized to suit almost any application. Central-receiver
systems generally are much larger. In all three, sunlight striking reflectors is
collected and used to heat a fluid that is piped to a central location. The heat can
be used directly to produce steam for industrial processes or to drive turbines that
generate electricity (21).
Photovoltaic cells are elegantly simple devices that generate electricity
directly from sunlight without going through the process of thermal–electric
conversion. These cells are made of silicon or other semiconductor materials; they
have no moving parts and therefore are quiet and reliable. Photovoltaic cells
require no maintenance, have the potential for long life, produce no pollution, and
consume no water in generating electricity. They can convert 20% or more of the
sunlight striking them into electricity; practical efficiencies in the 30–40 range
are possible (21).
In the United States, the land area of the lower 48 states intercepts about
47,000 quadrillion BTUs of direct sunlight per year, about 600 times total U.S.
primary energy use. At a solar collection efficiency of 15%, readily achievable
using present photovoltaic cells, significantly less than 1% of the land area of the
lower 48 states would be required to meet all our energy needs. This can be
compared to the 20% of U.S. lands devoted to croplands, or the 31% to pastures.
Moreover, many of the solar cells could be placed on the walls and roofs of
existing structures, reducing the area of land needed (6).
If the entire present U.S. electrical output came from central tower solar
steam generators, 780 square miles of collectors would be needed. This is less
land, however, than would be strip-mined in a 30-year period if all our energy
came from coal or uranium. Further, we can put solar collectors wherever we
choose (such as lands unsuited for agriculture, grazing, or habitation), whereas
 summer
solar radiation
angle Passive Active
solar heating solar heating
Energy Conservation

solar
warm air collector
winter
solar radiation
heat from hot water
angle air collector wall
space

cool air

HOT
glass COLD

heat
exchanger

FIGURE. 8 Passive and active solar systems. In a passive system, the length of roof overhang is
based on the latitude of the winter and summer sun. Natural air convection circulates heated air
between outer glass wall and collector wall. The collector wall is designed to be ~40 cm thick to
collect and store solar heat. In the active system, water is passed through solar collector panels,
131

the heated water is then pumped into the house, where heat is radiated from the hot water and
then recirculated into the system.
132 Strevett et al.

strip-mining occurs wherever coal or uranium exist, regardless of other values

associated with the land (3).
What happens when the sun goes down? One solution is hybrid energy
systems that run on 75% solar energy and 25% natural gas.
Unlike nuclear power, solar presents no problems of safety, disposal of
radioactive wastes, or danger that nuclear materials will fall into the wrong hands.
However, although solar schemes probably have the smallest environmental
impact of all current forms of energy, care must be taken with size of concentra-
tors and with the films and silicon used in photovoltaics (9).
It must be remembered that net yields and overall conversion efficiencies
are not the only considerations when different energy sources are compared. The
yield/cost ratio and conversion-cycle efficiency is much higher for coal burning,
for example, than for photovoltaic electrical production, making coal appear to
be a better source of energy than solar radiation. However, solar energy is free,
renewable, and nonpolluting; therefore, if we use solar energy to obtain electrical
energy, it does not matter how efficient the process is, as long as we get more out
of it than we put in (3).

6.10 Costs of Renewables

Geothermal energy is currently the least expensive renewable energy source. It is
closely followed by wood, hydroelectric, wind, and solar energy (Table 4). Most
solar technologies have high initial costs while providing savings down the road
in the form of lower fuel costs. For example, a solar water heater may cost $2500
to purchase and install. A solar power plant may cost $2500–$3000 per kilowatt
of capacity, while a conventional power plant costs between $400 and $1200 per
kilowatt of capacity. The difference is that solar technologies cost very little to
operate, whereas the major cost associated with conventional technologies is
usually fuel, which will be paid later. Unfortunately, our tax system gives

TABLE 4 Cost Comparison of

Renewable Energy Sources

Energy source Cost (cents/kWh)

Geothermal 4.5–6.5
Hydroelectric 5
Wood 5
Wind 7.5
Solar-thermal 8
Solar (photovoltaic) 30
Energy Conservation 133

advantages to the conventional plants, which can deduct the high, ongoing fuel
costs as operating expenses (21).
Commercialization of solar technologies could be greatly accelerated with
market incentives such as solar-energy tax credits, regulations requiring that
cost-effective passive and active solar technologies be included in new buildings,
and increased federal funding for solar-energy research and development. Cur-
rently, the great majority of subsidies go to established energy sources—fossil
fuels and nuclear power (21). U.S. federal funding for renewable energy sources
fell from $1.3 billion in 1980 to $135 million in 1990 (in 1990 dollars; Office of
Technology Assessment, 1991, as cited in Ref. 4). By recognizing the environ-
mental and social costs of energy technologies, federal, state, and local govern-
ments can help provide a “level playing field” for solar technologies and play a
decisive role in influencing energy choices (21).
When a homeowner or community invests independently in solar or wind
generation, what should be done about energy storage when electricity production
exceeds use? Many private electricity producers believe the best use for excess
electricity is in cooperation with the public utility grid. When private generation
is low, the public utility runs electricity through the meter and into the house or
community. When the wind generator or photovoltaic systems overproduce, the
electricity runs back into the grid and the meter runs backward. Ideally, the utility
reimburses individuals for this electricity, for which other consumers pay the
company. The 1978 Public Utilities Regulatory Policies Act required utilities to
buy power generated by small hydro, wind, cogeneration, and other privately
owned technologies at a fair price. Not all utilities comply yet, but some—notably
in California, Oregon, Maine, and Vermont—are purchasing significant amounts
of private energy (3).

7 CONCLUSION
Significant reductions in all pollutants that result from petroleum combustion
(CO2, CO, acid rain precursors, photochemical oxidants, unintentional petroleum
releases) could be achieved by the imposition of a gasoline tax that would
encourage the use of public transportation and fuel-efficient vehicles. Reduc-
tions in motor vehicle-related urban runoff represent an additional pollution
benefit of reducing the use of fossil fuels for transportation. The benefits of
freezing carbon emissions via a permitting system could have similar benefits to
those of a carbon tax, possibly with less public resistance.
At present, the possibilities for alternative transportation fuels appear rather
limited. The primary drawback for the use of ethanol is the relatively low energy
value obtained through its use as compared to the energy required for its
production. Synthetic natural gas or methanol produced from woody biomass may
134 Strevett et al.

be our most attractive options in this area; obviously, however, employment of

these does not eliminate CO2 emissions.
Elimination of fossil fuel subsidies is vital to the creation of incentive for
increases in efficiency, a conservation measure whose potential is enormous.
A dramatic increase in exploitation of passive technologies and especially non-
carbon energy sources is essential. Of all the available forms of energy, solar
has the greatest potential for providing clean, safe, reliable power. Wind farms
also represent a significant potential means of producing energy with minimal
environmental impacts, and geothermal energy deserves thorough investigation
as well. During a period of transition to noncarbon energy sources, increased
exploitation of natural gas represents a means of reducing CO2 emissions
significantly.

REFERENCES
1. D. E. Booth, The Environmental Consequences of Growth. London and New York:
Routledge, 1998.
2. International Energy Agency, World Energy Outlook, 1998 Edition. Paris: Organiza-
tion for Economic Co-operation and Development, 1998.
3. W. P. Cunningham and B. W. Saigo, Environmental Science. Dubuque, IA: William
C. Brown, 1997, 1992.
4. W. R. Cline, The Economics of Global Warming. Washington, DC: Institute for
International Economics, 1992.
5. J. H. Gibbons, P. D. Blair, and H. L. Gwin, Strategies for Energy Use. Sci. Am., vol.
261, no. 3, pp. 136–143, 1989.
6. J. J. MacKenzie, J.J. Energy and Environment in the 21st Century: The Challenge of
Change. In J. Byrne and D. Rich (eds.), Energy and Environment: The Policy
Challenge, New Brunswick, NJ: Transaction, 1992.
7. P. M. Vitousek, H. A. Mooney, J. Lubchenco, and J. M. Melillo, Human Domination
of Earth’s Ecosystems. Science, vol. 277, pp. 494–499, 1997.
8. A. Whyte, The Human Context. In H. Coward (ed.), Population, Consumption and
the Environment, pp. 41–59. Albany: State University of New York Press, 1995.
9. World Energy Council (WEC), Energy for Tomorrow’s World. New York: St. Martin’s
Press, 1993.
10. L. W. Canter, Environmental Impact Assessment, 2nd edition, p. 480, New York:
McGraw-Hill, 1995.
11. M. Ledbetter and M. Ross, Light Vehicles: Policies for Reducing Their Energy Use
and Environmental Impacts. In New Brunswick, NJ: Transaction, 1992. Energy and
Environment: The Policy Challenge, J. Burne and D. Rich (eds.), pp. 187–233.
12. Union of Concerned Scientists, Assessing the Hidden Costs of Fossil Fuels (briefing
paper). Cambridge, MA: Union of Concerned Scientists, 1993.
13. S. E. Manahan, Environmental Chemistry. Chelsea, MI: Lewis, 1991.
14. D. L. Johnson and L. A. Lewis, Land Degradation: Creation and Destruction.
Cambridge, MA, and Oxford, U.K. Blackwell, 1995.
15. Committee on Interior and Insular Affairs to Accompany HR 11500, Surface Mining
Energy Conservation 135

Control and Reclamation Act of 1974, HR93-1072. Washington, DC: U.S. House of
Representatives, 30 May 1974.
16. World Bank, Energy Efficiency and Conservation in the Developing World. New
York: World Bank, 1993.
17. R. U. Ayres, The Energy Policy Debate: A Case of Conflicting Paradigms. WEC J.,
vol. 111, p. 57, July 1992.
18. W. D. Ruckelshaus, Toward a Sustainable World. Sci. Am., vol. 261, no. 3, pp. 166–
174, 1989.
19. J. T. Lyle, Regenerative Design for Sustainable Development. New York: Wiley,
1994.
20. M. P. C. Munasinghe, Sustainable Energy Development: Issues and Policy. In P. R.
Kleindorfer, H. C. Kunreuther, and D. S. Hong (eds.), Energy, Environment and the
Economy, pp. 3–42. Brookfield, VT/Cheltenham, U.K.: Edward Elgar, 1996.
21. Union of Concerned Scientists, Solar Power: Energy for Today and Tomorrow.
Cambridge, MA: Union of Concerned Scientists, 1992.
22. E. Hirst and J. Ito, Justification of Electric-Utility Energy-Efficiency Programs. Oak
Ridge National Laboratory Report ORNL/CON-419, August 1995.
23. P. Ehrlich, and A. Ehrlich, The Population Explosion. New York: Simon & Schuster,
1990.
24. Union of Concerned Scientists, Cool Energy: The Renewable Solution to Global
Warming. Cambridge, MA: Union of Concerned Scientists, 1991.
25. D. Abrahamson, Climatic Change and Energy Supply: A Comparison of Solar and
Nuclear Options. In J. Byrne and D. Rich (eds.), Energy and Environment: The Policy
Challenge, p. 430. New Brunswick, NJ: Transaction, 1992.
26. E. D. Enger and B. F. Smith, Environmental Science: A Study of Interrelationships.
Dubuque, IA: William C. Brown, 1995.
27. D. Tenenbaum, Tapping the Fire. Technol. Rev., vol. 2, pp. 39–47, 1995.
28. M. Renner, Assessing the Military’s War on the Environment. State of the World 1991.
New York: Norton, 1991.
29. United Nations Fund for Population Activities, State of the World Population 1990.
30. F. Ackerman, Why Do We Recycle? Washington, DC/Covelo, CA: Island Press, 1997.
31. J. MacNeill, Strategies for Sustainable Economic Development. Sci. Am., vol. 216,
no. 3, pp. 155–165, 1989.
32. D. W. Orr, Ecological Literacy. Albany, NY: SUNY Press, 1992.
33. World Bank, World Bank Development Report 1999. Hong Kong: Asia 2000, 1999.
34. U.S. Environmental Protection Agency, Acid Rain Program—Overview, EPA 430/F-
92/019. Washington, DC: EPA, 1992.
35. H. Friedli, H. Lötscher, H. Oeschger, U. Siegenthaler, and B. Stauffer, Ice Core
Record of the 13C/12C Ratio of Atmospheric CO2 in the Past Two Centuries. Nature,
vol. 324, no. 20, pp. 237–238, 1986.
36. A. Neftel, H. Oeschger, and B. Stauffer, Evidence from Polar Ice Cores for the
Increase in Atmospheric CO2 in the Past Two Centuries. Nature, vol. 315, no. 2,
pp. 45–48, 1985.
37. National Academy of Sciences, Committee on Science, Engineering, and Public
Policy, Policy Implications of Greenhouse Warming: Mitigation, Adaptation, and the
Science Base. Washington, DC: National Academy Press, 1992.
8
Fundamentals of Heat Transfer

René Reyes Mazzoco

Universidad de las Américas–Puebla, Cholula, Mexico

1 HEAT TRANSFER MECHANISMS

1.1 Conduction
Conduction heat transfer is explained through the molecular motion in the solid’s
structure. Heat is transferred from one molecule to the adjacent molecule by
means of vibrational motion. This basic description points out that heat transfer
through a solid takes place entirely by conduction, and also states that it occurs
to a limited extent in liquids and gases because of their molecular mobility.
The mathematical formulation of conduction heat transfer was proposed by
Joseph Fourier while solving heat transfer problems in metal casting and tem-
plate. The first step for this formulation is the recognition that the amount of heat
transferred, q(W), from one point of a metal piece to another point of the same
medium (continuum) is proportional to the temperature difference between those
two points. The evaluation of the temperature difference through the derivative
in any direction (s) makes the measurement independent of any two specific
points and the distance between them:
dT
q ∝ (1)
ds

137
138 Mazzoco

For most cases of conduction heat transfer in solids it is possible to change

the proportionality relation to an equality defining the proportionality constant
called thermal conductivity, k(W/m K). Thus, the calculation of the amount of
heat transferred per cross-sectional surface area, A(m2), is obtained from Fourier’s
first law:
q dt
= −k (2)
a ds
1.1.1 Measurement of Thermal Conductivity
The thermal conductivity is a physical property characteristic of the medium and
is defined similarly for the three states of the matter. The expression for the
calculation of the amount of heat transferred by conduction is also used to define
the thermal conductivity, and one procedure for its measurement, that is, k, is
calculated after measuring the linear temperature difference between two points,
the amount of heat transferred, and the cross-sectional area associated with the
heat flow trajectory:
q/a
k = (3)
−dT/ds
The procedure for calculating values of k is especially relevant for the
evaluation of this property in mixtures of contaminated media for which no
information is normally available. For this evaluation it is easier to calculate the
thermal diffusivity, α (where α = k/ρCp, m2/s) of the mixture while solving
Fourier’s second law in cylindrical coordinates:
∂T ∂T 1 ∂T 
= α 2 +
r ∂r 
(4)
∂t  ∂r
Thus, a cylinder with a substance of unknown thermal conductivity is
placed in a bath at constant temperature to obtain the initial and boundary
conditions associated with the solution of Eq. (4):

T(r, 0) = initial sample temperature

T(0, t) = measurement at the cylinder’s center
T(R, t) = constant bath temperature

The value of α can be obtained from the solution of Eq. (4) with
these boundary conditions. The values of the density, ρ, and the constant-pressure
heat capacity, Cp, are normally available in the literature, or are measured
experimentally.
Thermal conductivity prediction models are available for gases and liquids
in several references (e.g., Ref. 1).
Fundamentals of Heat Transfer 139

1.1.2 Effect of Shape on Calculation of Conduction

Heat Transfer
Evaluation of the integral between two specific points in the direction of heat
transfer allows for the calculation of the macroscopic amount of heat. For an
object with constant cross-sectional area in the direction of heat flow, integration
of Eq. (2) gives:

q T1 − T2
= k (5)
A s2 − s1

Instead, for a cylindrical object, heat flow in the direction of the radius
finds a constantly changing cross-sectional area. Thus, integration of Eq. (2) with
A = 2πrL gives
T1 − T2
q = (6)
ln(r2/r1)/2πkl

which involves a logarithmic distance (radius) difference instead of the thickness

of the medium involved.
1.1.3 Combined Resistances
A common problem in heat transfer design is the combination of several layers
of solid to provide heat insulation, or layers of solid and fluid as in heat exchanger
design. For a rectangular geometry, two combined resistances to heat transfer can
be expressed as

q T1 − T2
= (7)
A (L1/k1) + (L2/k2)

where L and k are the thickness and the thermal conductivity of components (1)
and (2), respectively, and T1 and T2 are the temperatures at the external surfaces
of the combined wall. Each additional layer of material increases by one the
resistances to heat transfer added in the denominator of Eq. (7).
The effect of geometry on the combined resistances to heat transfer can be
obtained by integrating Eq. (2) for a double-layered cylinder:
T2 − T1
q = (8)
[ln(r2/r1)/(2πk1L)] + [ln(r3/r2)/(2πk2L)]

1.1.4 Relative Magnitude of Values of Thermal

Conductivity
The values of the thermal conductivity depend on the phase of the material
considered:
140 Mazzoco

kgas < kliquid < ksolid

Thus, solid materials are good heat conductors, while for heat insulation trapped
gases are the best option. Good electric conductor metals are the best selection
for heat conductors. The design of heat insulation follows the criteria for air
entrapment in fabrics or ceramics that could be resistant to high temperatures.

1.2 Convection
Heat convection is described as heat transport in fluid eddies promoted by the
flow derived from a mechanical device, a pump or fan (forced convection), or a
density difference (natural convection). The mechanism is associated with the
definition of the convective heat transfer coefficient, h(W/m2 ˚C):
q
h = (9)
A(T2 − T1)

As the turbulent flow process carrying the heat cannot be fully described,
the temperature difference is considered at two points (1) and (2) in the direction
of heat transfer. It is not possible to describe this process through a differential
equation, and Eq. (9) is a definition for h that is related to the specific geometry
associated to the surface area, A, and the flow conditions.
The convective heat transfer coefficient can be calculated for design
purposes from experimental information gathered in the open literature. Experi-
ments have been carried out under geometry, flow range, and similar thermo-
physical properties conditions that can be encountered in process applications.
The information has been grouped in terms of flow conditions and thermophysical
properties involved.
Flow conditions are described through the Reynolds number (Re) for forced
convection. The Reynolds number relates the momentum convection associated
to the flow velocity, v, to the momentum diffusivity associated to ν, the kinematic
viscosity (ν = µ/ρ), µ is Newtonian viscosity (kg/ms). At low Reynolds numbers,
implying low flow velocity, momentum diffusivity dominates, and the fluid
displacement is in the laminar flow condition. When the flow velocity is high
relative to the kinematic viscosity, the Reynolds number is high, indicating
turbulent flow conditions.
Lν
Re = (10)
v
L is the flow characteristic length; for internal flow in circular pipes, L is the
internal diameter.
The Grashof number (Gr) describes flow conditions for natural convection
and is used instead of the Reynolds number.
Fundamentals of Heat Transfer 141

gβ(Tw − Tinfinity)L3
Gr = (11)
v2
Here g is the acceleration of gravity, Tw is the solid wall temperature, Tinfinity is
the fluid bulk temperature, L is the heat transfer characteristic length, and β is the
volume coefficient of expansion:
(ρinfinity − ρ)
β = (12)
ρ(T − Tinfinity)
ρinfinity is the fluid bulk density.
In natural and forced convection, the Prandtl number describes the influ-
ence of thermophysical properties in the calculation of the convective heat
transfer coefficient, normally to the 1⁄3 power.
v Cpµ
Pr = = (13)
α k
The Nusselt number, Nu, is the ratio of heat convection to diffusion
associated to the heat transfer characteristic length, L:
hL
Nu = (14)
k
From the exact analysis of the boundary layer between the fluid and the
solid wall transferring heat, the correlation in forced convection among Nusselt,
Reynolds, and Prandtl numbers is
Nu = 0.664 Re1/2 Pr1/3 (15)
This theoretical correlation has very limited application, and the depend-
ence of these dimensionless numbers on the geometry makes experimentation
necessary to calculate correlations for each geometry. The correlation results are
normally reported with the same mathematical formulation:
Nu = c0 Ren Prm (16)
For natural convection, the analysis of the boundary layer provides the
correlation of the important dimensionless numbers:
Nu = C(Gr Pr)m (17)

1.3 Radiation
For practical conditions, radiation emitted (or received) by surface is calculated
from an equation that involves the effect of the area, A12, the emissivity, ε1, of
the emitting surface involved, and a view factor, F12, that describes the effect
142 Mazzoco

of the relative positions of the two surfaces involved on the amount of radiation
exchanged. The formulation of the exchanged radiation is

q = σε1A12F12(T14 − T24) (18)

All practical terms in Eq. (18) are measured experimentally and are reported in
several references (e.g., Ref. 2)

2 HEAT ACCUMULATION
Heat accumulation is described through the heat capacity. The specific property
normally used to achieve this calculation is the constant-pressure heat capacity,
Cp (J/kg ˚C). The total amount of material that stores heat should be expressed in
the mass or molar terms used for the Cp. The heat stored is then a function of the
temperature change in the total mass considered:
q
Cp = (19)
m(dT/dt)
The temperature variation with time allows the evaluation of the heat flow
accumulated.

2.1 Sensible Heat

The variation of the temperature in a fluid medium defines sensible heat. The
calculation of the amount of sensible heat is obtained from Eq. (19). The heat
capacity should correlate the fluid and phase considered.

2.2 Latent Heat

The process where a change of phase takes place requires the addition of latent
heat. The latent heat is used to change phase in a fluid without a change in the
medium temperature. The evaluation of the latent heat is necessary to measure
the amount of heat required for phase change. Latent heat values and prediction
correlations are available in Ref. 1.

3 EXPERIMENTAL MEASUREMENT AND PREDICTION OF

HEAT TRANSFER THERMOPHYSICAL PROPERTIES

3.1 Constant-Pressure Heat Capacity, Cp

Measurement of the Cp requires the evaluation of temperature change in a fixed
mass of material due to a heat flow from the surroundings according to Eq. (19).
Fundamentals of Heat Transfer 143

3.2 Thermal Conductivity, k

For the measurement of k, Fourier’s first law is normally used to define the
parameters involved in the evaluation. The heat flux in Eq. (2) is determined
from the heat flow and the body geometry while the temperature gradient is
measured directly.

3.3 Convective Heat Transfer, h

The convective heat transfer coefficient is experimental measured of forced- and
natural-convection conditions. h is part of Nu, while flow conditions are repre-
sented by Re or Gr, and the thermophysical properties form Pr.
Normally, the values of h are obtained from reported correlations. If it is
necessary to evaluate h for conditions not previously studied, the information is
gathered and analyzed according to Eq. (16) or (17).

3.4 Thermophysical Properties of Mixtures in

Pollution Control
Mixtures of contaminated media normally require the experimental evaluation of
the thermophysical properties. In some cases, due to nonavailability of the
experimental data, correlations for calculating the thermophysical properties
are limited.

4 HEAT TRANSFER DESIGN

Process efficiency is defined at the design stage. Design algorithms for heat
transfer equipment can be found in several classic references (e.g., Ref. 3) and
are still used for designing heat transfer equipment. Several software options are
also available for efficient heat transfer equipment design; software the descrip-
tion can be obtained from demos downloaded from an Internet search (any search
engine) on “Heat Exchangers.”
The basic equation for heat exchange design is
q = Uo Ao ∆TLMF (20)
where Uo is the overall heat transfer coefficient and includes all the heat transfer
conductances around the solid wall transferring heat. For a flat wall transferring
heat,
1
Uo = (21)
1/hinside + G/k + 1/houtside
where hinside is the inside convective heat transfer coefficient, G is the wall
thickness, and houtside is the outside heat transfer coefficient.
144 Mazzoco

The driving force for heat transfer in a heat exchanger is the logarithmic
mean temperature difference:

[Toutlet − tinlet) − (Tinlet − toutlet)]F

∆TLMF = (22)
ln[(Toutlet − tinlet)/(Tinlet − toutlet)]

T is the hot fluid temperature, t is the cold fluid temperature, and F is the
efficiency factor adapted for each configuration of shell and tube, plate exchang-
ers, and direct-contact heat exchangers (4).
From the calculation of the amount of heat transferred, including the
temperature changes involved and the overall heat transfer conductance, the area
for heat exchange is determined. Several heat transfer equipment can be used to
accomplish the heat exchange between the media in a given process condition.

4.1 Heat Transfer Design and Good Engineering Practices

Design defines the efficiency of the operation of a process. Once the optimized
design is utilized, it is necessary to maintain good engineering practices. These
practices should include pollution control and waste minimization.
Heat transfer equipment is subjected to fouling and corrosion, which
are among the major hurdles for the operation. Fouling increases heat transfer
resistance and waste of energy. Good engineering practices include the use of
fouling suppresants in heat transfer fluids and periodic cleaning of the ex-
changer walls.
For water as the cooling or heating medium in industrial operations there
are several standard techniques for keeping fouling low. Water in cooling-water
circuits has to be treated to keep salts and dirt content low. Common treatments
include the addition of coagulants for sedimentation of some salts and particles;
addition of biocides, to prevent microbial growth that is another source of fouling;
and the addition hardness suppresants such as polyphosphates; among others.
Although the materials used in heat transfer fluids treatment are a source of solid
waste, handling its final deposition should follow normal procedures. Fouling
prevention is not considered a polluting operation. Fouling prevention by-products
can be integrated to cement kiln operations when feasible, in order to eliminate
waste generation.
Corrosion protection of heat transfer surfaces is a suggested practice
for pollution control and waste minimization. In order to prevent corrosion,
begin with the analysis of the appropriate combination of materials and flu-
ids. For the operating equipment, passive and active cathodic protection are
recommended.
Fundamentals of Heat Transfer 145

4.2 Innovations for Efficient Heat Use

Efficient energy use is a direct way to reduce pollution and minimize wastes from
industrial sources. The ongoing research in energy efficiency and resulting
innovations highlight the intensity of scientific activity in this field.
New approaches to increase heat transfer efficiency include the following.
1. Fluidized bed combustion is the choice for eliminating solids in solid-
waste management schemes. In general, direct contact between the
materials increases heat transfer efficiency. Direct contact reduces the
heat transfer resistances due to the wall in conventional equipment, and
increases the convective heat transfer coefficients due to the higher
contact velocities between the materials and fluids.
2. To increase the efficiency in steam generation, direct-contact heat
exchangers make use of residual heat from combustion gases to preheat
the feed streams to the boiler. Thermal recovery is a possibility from
direct-contact heat exchangers and heat pumps. Rotary drums recover
heat from a residual discharge in an steam generator and transport it to
preheat the inlet streams to the generator.
3. Heat pipes are a promising technology for increasing residual heat
usage as heat pumps. Heat pipes use capillary pressure as the driving
force for condensing and evaporating the working fluid, thus elim-
inating the necessity for pump and compressor in the power cycle. The
understanding of heat pipe operation is related to the evaluation of con-
vective heat transfer coefficients for change-of-phase heat transfer.
4. Plate heat exchangers are now available for almost any process condi-
tion, including high-pressure and corrosivity conditions. Enhanced heat
transfer surfaces improve energy management, reducing wastes. Im-
proved surfaces increase the convective heat transfer coefficients for
heating–cooling operations, and change-of-phase heat transfer.
5. Co-generation in chemical and petrochemical processes makes use of
the process integration gained from the use of simulation and pinch-
point techniques to increase energy usage.

5 CONCLUSIONS
The understanding of heat transfer fundamentals is a basic step toward the
proposition of improved industrial solutions in terms of energy wastes minimization.
Clear fundamental concepts make the use of design software straight-
forward. This is the approach to equipment design that produces the best results
for waste minimization.
146 Mazzoco

Heat transfer innovations are improving energy handling in industrial

processes, reducing pollution and wastes. This research field is active in funda-
mentals such as enhanced heat transfer or heat pipe development.

REFERENCES
1. R. C. Reid, J. M. Prausnitz and B. E. Poling, The Properties of Gases and Liquids, 4th
ed. New York: McGraw-Hill, 1987.
2. J. P. Holman, Heat Transfer, 8th ed. New York: McGraw-Hill, 1997.
3. D. Q. Kern, Process Heat Transfer. New York: McGraw-Hill, 1950.
4. O. Levenspiel, Engineering Flow and Heat Exchange, 2nd ed. New York: Plenum
Press, 1992.
9
Macroscopic Balance Equations

Paul K. Andersen and Sarah W. Harcum

New Mexico State University, Las Cruces, New Mexico

The prevention of waste and pollution requires an understanding of numerous

technical disciplines, including thermodynamics, heat and mass transfer, fluid
mechanics, and chemical kinetics. This chapter summarizes the basic equations
and concepts underlying these seemingly disparate fields.

1 MACROSCOPIC BALANCE EQUATIONS

A balance equation accounts for changes in an extensive quantity (such as mass
or energy) that occur in a well-defined region of space, called the control volume
(CV). The control volume is set off from its surroundings by boundaries, called
control surfaces (CS). These surfaces may coincide with real surfaces, or they
may be mathematical abstractions, chosen for convenience of analysis. If matter
can cross the control surfaces, the system is said to be open; if not, it is said to
be closed.

1.1 The General Macroscopic Balance

Balance equations have the following general form:

147
148 Andersen and Harcum

dX ⋅ ⋅
dt
= ∑(X )i + (X )gen (1)
CS,i

where X is some extensive quantity. A dot placed over a variable denotes a rate;
⋅
for example, (X )i is the flow rate of X across control surface i. The terms of Eq. (1)
can be interpreted as follows:
dX
= rate of change of X inside the control volume
dt
⋅
∑(X )i = sum of flow rates of X across the control surfaces
CS,i
⋅
(X )gen = rate of generation of X inside the control volume
Flows into the control volume are considered positive, while flows out of the
control volume are negative. Likewise, a positive generation rate indicates that X
is being a created within the control volume; a negative generation rate indicates
that X is being consumed in the control volume.
The variable X in Eq. (1) represents any extensive property, such as those
listed in Table 1. Extensive properties are additive: if the control volume is

TABLE 1 Extensive Quantities

Quantity Flow rate (CS i) Generation rate

Total mass, m m⋅ i m⋅ gen = 0 (conservation of

mass)
⋅ ⋅
Total moles, N Ni N gen
Species mass, mA (m⋅ A)i (m⋅ A)gen
⋅ ⋅
Species moles, NA (N A)i (N A)gen
⋅ ⋅ ⋅ ⋅
E i = Q i + W i + m⋅ i E i
^
Energy, E E gen = 0 (conservation of
⋅ ⋅ ⋅ ⋅ ~
E i = Q i + Wi + N i E i energy)
⋅ ⋅ ⋅
S i = Q i /Ti + m⋅ i S~ i
^
Entropy, S S gen ≥ 0 (second law of
⋅ ⋅ ⋅
S i = Q i /Ti + N i S i thermodynamics)
Momentum, p = mv p⋅ i = m⋅ i vi ⋅
p gen = F (Newton’s second
law of motion)
Notes:
⋅
Qi ≡ heat transfer rate through CS i
⋅
W i ≡ work rate (power) at CS i
^ ~
E i ≡ energy per unit mass of stream i; E i ≡ energy per unit mole of stream i
^ ~
S i ≡ entropy per unit mass of stream i; S i ≡ entropy per unit mole of stream i
Ti ≡ absolute temperature of CS i
F ≡ net force acting on control volume
Macroscopic Balance Equations 149

subdivided into smaller volumes, the total quantity of X in the control volume
is just the sum of the quantities in each of the smaller volumes. Balance equations
are not appropriate for intensive properties such as temperature and pressure,
which may be specified from point to point in the control volume but are not
additive.
It is important to note that Eq. (1) accounts for overall or gross changes in
the quantity of X that is contained in a system; it gives no information about the
distribution of X within the control volume. A differential balance equation may
be used to describe the distribution of X (see Section 2.2).
Equation (1) may be integrated from time t1 to time t2 to show the change
in X during that time period:

∆X = ∑(X)i + (X)gen (2)

CS,i

1.2 Total Mass Balance

Material is conveniently measured in terms of the mass m. According to Einstein’s
special theory of relativity, mass varies with the energy of the system:

m⋅ gen = 2
1 dE
(special relativity) (3)
c dt
where c = 3.0 × 108 m/s is the speed of light in a vacuum. In most problems of
practical interest, the variation of mass with changes in energy is not detectable,
and mass is assumed to be conserved—that is, the mass-generation rate is taken
to be zero:

m⋅ gen = 0 (conservation of mass) (4)

Hence, the mass balance becomes

∑(m⋅ )i
dm
= (5)
dt
CS,i

1.3 Total Material Balance

The quantity of material in the CV can be measured in moles N, a mole being
6.02 × 1023 elementary particles (atoms or molecules). The rate of change of
moles in the control volume is given by
dN ⋅ ⋅
dt
= ∑(N )i + (N )gen (6)
CS,i
⋅ ⋅
where (N)i is the molar flow rate through control surface i and (N)gen is the molar
150 Andersen and Harcum

generation rate. In general, the molar generation rate is not zero; the determina-
tion of its value is the object of the science of chemical kinetics (Section 4.2).

1.4 Macroscopic Species Mass Balance

A solution is a homogenous mixture of two or more chemical species. Solutions
usually cannot be separated into their components by mechanical means. Con-
sider a solution consisting of chemical species A, B, . . . . For each of the
components of the solution, a mass balance may be written:
dmA
dt
= ∑(m⋅ A )i + (m⋅ A )gen
CS,i
dmB
dt
= ∑(m⋅ B )i + (m⋅ B )gen (7)
CS,i

...

Conservation of mass requires that the sum of the constituent mass generation
rates be zero:

(m⋅ A )gen + (m⋅ B )gen + . . . = 0 (conservation of mass) (8)

1.5 Macroscopic Species Mole Balance

The macroscopic species mole balances for a solution are
dNA ⋅ ⋅
dt
= ∑(N A )i + (N A )gen
CS,i
dNB ⋅ ⋅
dt
= ∑(N B )i + (N B )gen (9)
CS,i

...

In general, moles are not conserved in chemical or nuclear reactions. Hence,

⋅ ⋅ ⋅
(N A )gen + (N B )gen + . . . = N gen (10)

1.6 Macroscopic Energy Balance

Energy may be defined as the capacity of a system to do work or exchange heat
with its surroundings. In general, the total energy E can expressed as the sum of
three contributions:
Macroscopic Balance Equations 151

E = K + Φ + U (11)
where K is the kinetic energy, Φ is the potential energy, and U is the internal energy.
Energy can be transported across the control surfaces by heat, by work, and
by the flow of material. Thus, the rate of energy transport across control surface
i is the sum of three terms:
⋅ ⋅ ⋅
E i = Q i + W i + m⋅ i E i
^
(12)
⋅ ⋅
where Q i is the heat transfer rate, W i is the working rate (or power), m⋅ i is the
^
mass flow rate, and E i is the specific energy (energy per unit mass).
Energy is conserved, meaning that the energy generation rate is zero:
⋅
E gen = 0 (conservation of energy) (13)

Therefore, the energy of the control volume varies according to

⋅ ⋅
+ W + m⋅ E )i
dE
= ∑(Q
^
(14)
dt
CS,i

The energy flow rate can also be written in terms of the molar flow rate
⋅ ~
N i and the molar energy E i. Hence, the energy balance can be written in the
equivalent form
dE ⋅ ⋅ ⋅ ~
dt
= ∑(Q + W + N E )i (15)
CS,i

1.7 Entropy Balance

Entropy is a measure of the unavailability of energy for performing useful work.
Entropy may be transported across the system boundaries by heat and by the flow of
material. Thus, the rate of entropy transport across control surface i is given by
⋅
⋅ Qi
Si = + m⋅ i Si
^
(16)
Ti
⋅
where Q i is the heat transfer rate through control surface i, Ti is the absolute
temperature of the control surface, m⋅ i is the mass flow rate through the control
^
surface, and Si is the specific entropy or entropy per unit mass. In terms of the
~
molar flow rate and the molar entropy S i, the entropy transport rate is
⋅
⋅ Qi ⋅ ~
Si = + Ni S i (17)
Ti
152 Andersen and Harcum

According to the second law of thermodynamics, entropy may be created—

but not destroyed—in the control volume. The entropy generation rate therefore
must be non-negative:
⋅
S gen ≥ 0 (second law of thermodynamics) (18)

Processes for which the entropy generation rate vanishes are said to be reversible.
Most real processes are more or less irreversible.
In terms of mass flow rates, the entropy balance is
⋅
Q ^ ⋅
∑  T + m⋅ S + Sgen
dS
= (19)
dt
CS,i i

If molar flow rates are used instead, the entropy balance is

⋅
dS Q ⋅ ~ ⋅
dt
= ∑  T + N S + S gen
 (20)
CS,i i

1.8 Macroscopic Momentum Balance

The momentum p is defined as the product of mass and velocity. Because velocity
is a vector—a quantity having both magnitude and direction—momentum is also
a vector. Momentum can be transported across the system boundaries by the flow
of mass into or out of the control volume:

p⋅ i = (m⋅ v )i (21)

According to Newton’s second law of motion, momentum is generated by the net

force F that acts on the control volume:

p⋅ gen = F (Newton’s second law) (22)

Hence, the momentum balance takes the form

m⋅ vi + F
dmv
dt
= ∑ 



(23)
CS,i

Because this is a vector equation, it can be written as three component equations.

In Cartesian coordinates, the momentum balance becomes
dmvx
∑ m⋅ vx i + Fx
 
x momentum: = 

dt
CS,i
Macroscopic Balance Equations 153

dmvy
y momentum: = ∑ 

 m⋅ vy i + Fy (24)
dt
CS,i
dmvz
z momentum: = ∑ m⋅ vz i + Fz



dt
CS,i

2 DIFFERENTIAL BALANCE EQUATIONS

As noted previously, macroscopic balance equations account only for overall or
gross changes that occur within a control volume. To obtain more detailed
information, a macroscopic control volume can be subdivided into smaller control
volumes. In the limit, this process of subdivision creates infinitesimal control
volumes described by differential balance equations.

2.1 General Differential Balance Equation

The general macroscopic balance for the extensive property X is Eq. (1):
dX ⋅ ⋅
dt
= ∑ (X )i + (X )gen (1)
CS,i

Division by the system’s volume V yields

⋅ ⋅
d X X i X gen
dt V = ∑V V+
  CS,i

The differential or microscopic balance equation results from taking the limit as
V → 0:
∂[X] ⋅
= −∇ ⋅ (X) + [X ]gen (25)
∂
where [X] is read as “the concentration of X” and X is “the flux of X.” The terms
of this equation can be interpreted as follows:
∂
[X] = rate of change of the concentration of X
∂t
−∇ ⋅ (X) = net influx of X
[X]gen = generation rate of X per unit volume

The flux X is the rate of transport per unit area, where the area is oriented
perpendicular to the direction of transport. In Cartesian (x, y, z) coordinates, X
may be defined as
154 Andersen and Harcum

⋅ ⋅ ⋅
Xx Xy Xz
X = i + j + k
Ax Ay Az
Here, Ax, Ay, and Az are the areas perpendicular to the x, y, and z directions,
respectively; (i, j, k) are the (x, y, z) unit vectors. In Cartesian coordinates, the
del operator ∇ takes the form
∂ ∂ ∂
∇ = i + j + k
∂x ∂y ∂z
The form of the del operator in other coordinate systems may be found in texts
on fluid mechanics and transport phenomena (1–4).
Table 2 shows the concentrations, fluxes, and volumetric generation terms
for the extensive quantities considered in this chapter.

2.2 Differential Total Mass Balance

Assuming conservation of mass (ρgen = 0), the differential mass balance can be
written as

TABLE 2 Concentrations, Fluxes, and Volumetric Generation

Volumetric
Quantity Flux generation rate

Total mass, [m] = ρ m = ρv ρ⋅ gen = 0 (conservation of

mass)
Total moles, [N] = c N = cv c⋅ gen
Species mass, [mA] = ρA mA = ρAv + jA (ρ⋅ A)gen
Species moles, [NA] = cA NA = cAv + JA (c⋅ A)gen
e⋅ gen = 0 (conservation
^
Energy, [E] = e E = q + σ ⋅ v + mE
~
E = q + σ ⋅ v + NE of energy)
s⋅ gen ≥ 0 (second law of
^
Entropy, [S] = s S = q/T + mS
~
S = q/T + NS thermodynamics)
Momentum, [p] = ρv P = mv = ρvv f = −∇ ⋅ σ + b (second law
of motion)
Notes:
b ≡ body force per unit volume
jA ≡ diffusive mass flux of species A
JA ≡ diffusive molar flux of species A
f ≡ total force per unit volume
q ≡ heat flux
σ ≡ material stress
v ≡ fluid velocity
Macroscopic Balance Equations 155

∂ρ
= −∇ ⋅ (m) (26)
∂t
where m is the mass flux. It is more common to write the mass flux in terms of
the density and velocity, m = ρv. Hence,
∂ρ
= −∇ ⋅ (ρv) (27)
∂t
Equation (27) is called the continuity equation; it is one of the basic equations of
fluid mechanics.

2.3 Differential Total Material Balance

The differential material balance is
∂c
= −∇ ⋅ (N) + c⋅ gen (28)
∂t
where N is the molar flux. In the absence of chemical or nuclear reactions,
c⋅ gen = 0.

2.4 Differential Species Balances

Consider a solution consisting of component species A, B, . . . . In general, a
chemical species in such a solution may be transported by convection and by
diffusion. Convection is transport by the bulk motion of the solution. The
convective flux of species A is the product of the mass concentration ρA and the
solution velocity v:
ρAv = convective (mass) flux of A

Diffusion is the transport of a species resulting from gradients of concentration,

electrical potential, temperature, pressure, and so on. The diffusive flux of species
A is denoted by jA:
jA = diffusive (mass) flux of A

The overall material flux is the sum of the convective and diffusive fluxes:
mA = ρAv + jA (29)

A differential mass balance may be written for each of the components of the
solution:
∂ρA
= −∇ ⋅ (ρAv + jA) + (ρ⋅ A)gen
∂t
156 Andersen and Harcum

∂ρB
= −∇ ⋅ (ρBv + jB) + (ρ⋅ B)gen (30)
∂t
...

Conservation of mass requires that the constituent mass generation rates sum
to zero:

(ρ⋅ A)gen + (ρ⋅ B)gen + . . . = 0 (conservation of mass ) (31)

2.5 Differential Species Material Balance

The total molar flux of species A is the sum of the convective and diffusive
molar fluxes:
NA = cAv + JA (32)

where cAv is the convective molar flux and JA is the diffusive molar flux:

cAv = convective (molar) flux of A

JA = diffusive (molar) flux of A

The differential material balances for a solution consisting of species A, B, . . . are

∂cA
= −∇ ⋅ (cAv + JA) + (c⋅ A)gen
∂t
∂cB
= −∇ ⋅ (cBv + JB) + (c⋅ B)gen (33)
∂t
...

The sum of the constituent molar generation terms is the total molar volumetric
generation rate:

(c⋅ A)gen + (c⋅ B)gen + . . . = c⋅ gen (34)

2.6 Differential Energy Balance

Energy may be transported by heat, by work, and by convection. Thus, the energy
flux can be written as the sum of three terms:

E = q + ␴ ⋅ v + mE
^
(35)
where q is the heat flux, ␴ is the stress tensor (defined as the force per unit area),
^
and mE is the convective energy flux.
Macroscopic Balance Equations 157

As used in Eq. (35), the stress tensor σ accounts for the forces exerted on
the surface of the differential control volume by the surrounding material.
Multiplying the stress ␴ by the material velocity v gives the rate of work done
(per unit area) on the surface of the control volume:
Rate of work by material stresses (per unit area) = σ ⋅ v
Conservation of energy requires that the energy generation rate be zero:

e⋅ gen = 0 (conservation of energy) (36)

The differential energy balance may be written as

∂e
= −∇ ⋅ (q + σ ⋅ v + mE )
^
(37)
∂t
It is common practice to express the energy concentration as the product of
the mass density and the specific energy: e = ρE. Hence, the energy balance
^

becomes

∂ρE
^

= −∇ ⋅ (q + σ ⋅ v + mE )
^
(38)
∂t
If the material flux is measured in moles, the energy balance may be written
in the equivalent form

∂ρE
^

= −∇ ⋅ (q + σ ⋅ v + NE )
^
(39)
∂t

2.7 Differential Entropy Balance

Entropy may be transported by heat transfer and by convection. Thus, the entropy
flux can be written as the sum of two terms:
q
S = + mS
^
(40)
T
According to the second law of thermodynamics, the entropy generation
rate must be non-negative:

s⋅ gen ≥ 0 (second law of thermodynamics) (41)

The differential entropy balance is

∂s q ^
= −∇ ⋅  + mS  + s⋅ gen (42)
∂t  T 
158 Andersen and Harcum

If the molar flux is used instead of the mass flux, the entropy balance
becomes
∂s q ~
= −∇ ⋅  + NS  + s⋅ gen (43)
∂t  T 

2.8 Differential Momentum Balance

The momentum concentration is the product of density and velocity: [p] = ρv.
The momentum flux is the product of the momentum concentration and the
velocity: P = ρvv. The differential momentum balance is
∂ρv
= −∇ ⋅ (ρvv) + f (44)
∂t
where f is the net force (per unit volume) acting on the control volume. In most
fluid systems, f is the sum of a stress term and a body-force term:
f = −∇ ⋅ σ + b (45)
The most important body force is gravity, for which b = ρg.
The momentum balance becomes
∂ρv
= −∇ ⋅ (ρvv) − ∇ ⋅ σ + b (46)
∂t

3 FLUX AND TRANSPORT EQUATIONS

In most cases, balance equations are not sufficient by themselves; additional
equations are needed to compute the fluxes and transport rates.

3.1 Diffusive Material Fluxes

As shown before, the overall material flux is the sum of the convective and
diffusive fluxes:
mA = ρAv + jA (47)
NA = cAv + JA (48)
The diffusive flux (jA or JA) is driven by gradients of concentration, electrical
potential, temperature, pressure, and so on. In the simplest situations, the rate of
diffusion can be described by some variation of Fick’s law:
jA = ρDA∇ωA (49)
JA = −cDA∇χA (50)
where ωA is the mass fraction of A, χA is the mole fraction of A, and DA is the
Macroscopic Balance Equations 159

diffusion coefficient or diffusivity of A. In some cases, other models may be

required for describing diffusion (see Refs. 4 and 5).

3.2 Heat Conduction

Heat conduction (also called heat diffusion) is the transport of thermal energy by
random molecular motion. Conduction is driven by a temperature gradient. In
most cases, the heat flux can be described by Fourier’s law:
q = −k∇T (51)
where k is the conductivity of the material. Conductivity values may be found in
many standard handbooks and heat transfer textbooks (9,10).

3.3 Radiation Heat Transfer

Radiation is the transport of energy by electromagnetic waves. Radiative heat
transfer tends to be especially important at high temperatures, such as occur in
combustion. Consider an object having a surface temperature Ts. The magnitude
of the radiation heat flux leaving the surface of the object is given by the
Stefan-Boltzmann law:
|qrad| = εσTs4 (52)
where σ is the Stefan-Boltzmann constant, a universal constant (σ = 5.67 × 10−8
W/m2 K4); and ε is the emissivity, a material property. Values of the emissivity
for many different materials are tabulated in standard handbooks (9,10).

3.4 Fluid Stress

The stress tensor ␴ appears in both the differential energy balance and the
differential momentum balance. In general, stress is defined as force per unit area.
The force and the area can be represented as vectors having magnitude and
direction:
F = (Fx,Fy,Fz) and A = (Ax, Ay, Az) (53)
Stress is a tensor having a magnitude and two directions. For example, the yx
component of the stress tensor is*
Fx
σyx = (54)
Ay
The stress tensor ␴ is often represented as a 3 × 3 matrix having nine components:

*Some texts transpose the subscripts, so that σij = Fi /Aj.

160 Andersen and Harcum

σxx σxy σxz Fx /Ax Fy /Ax Fz /Ax

   
␴ = σyx σyy σyz = Fx /Ay Fy /Ay Fz /Ay (55)
σzx σzy σzz  Fx /Az Fy /Az Fz /Az
   
If the force is directed perpendicular to the area on which it acts, the result
is a normal stress. If the force acts tangentially to the area, the result is a shear
stress. Thus, the diagonal elements of the stress tensor are normal stresses, and
the off-diagonal elements are shear stresses:

Normal stresses: σxx, σyy, σzz

Shear stresses: σxy, σxz, σyx, σyz, σzx, σzy

For a fluid, it is customary to resolve the stress tensor into two components:

␴ = P␦ + ␶ (56)
where P is the fluid pressure, ␦ is the identity tensor, and ␶ is the viscous stress
tensor. In matrix notation,
σxx σxy σxz 1 0 0 τxx τxy τxz
  0 1 0 + τ 
σ
 yx σyy σyz = −P    yx τyy τyz (57)
σzx σzy σzz  0 0 1 τzx τzy τzz 
     
The viscous stress tensor depends on the material. For an incompressible
Newtonian fluid such as water, the diagonal elements of the viscous tensor are
given by

∂νx ∂νy ∂νz

τxx = 2µ τyy = 2µ τzz = 2µ (58)
∂x ∂y ∂z

where µ is the viscosity. For the same fluid, the off-diagonal elements of τ are
given by
 ∂νx ∂νy   ∂νx ∂νz 
τxy = τyx = µ  +  τxz = τzx = µ  + 
 ∂y ∂x   ∂z ∂x 
 ∂νz ∂νy  (59)
τyz = τzy = µ  + 
 ∂y ∂z 

In vector notation, these Eqs. (58) and (59) become

␶ = µ[∇v + (∇v)t] (60)

where the superscript t denotes a transpose. Other equations for the viscous stress
are given in textbooks on fluid mechanics and transport phenomena (1–4).
Macroscopic Balance Equations 161

3.5 Interfacial Transfer Coefficients

Many situations of practical engineering importance involve the transport of
energy or material between two phases. It is common practice to compute the
interfacial heat transfer rate using a relation of the form
⋅
Q = ±hA ∆T (61)
where h is the heat transfer coefficient, A is the interfacial surface area over which
heat transfer occurs, and ∆T is the difference in temperature between the two
phases. The sign is chosen so that the heat flows from high temperature to low
temperature. Similarly, mass transfer between two phases can be described by a
relation of the form
⋅
N A = ±kAA ∆cA (62)
where kA is the mass transfer coefficient for species A, A is the interfacial surface
area over which mass transfer occurs, and ∆cA is the difference in the concentra-
tion of species A between the two phases.
Heat and mass transfer coefficients are determined experimentally; the
results of such experiments are typically presented in terms of dimensionless
parameters. Table 3 lists a number of important dimensionless parameters. Thus,
heat transfer data are often presented in terms of the Nusselt number, Nu, a
dimensionless heat transfer coefficient. In many cases, the Nusselt number is
found to be a function of the Reynolds number Re and the Prandtl number Pr:
Nu = ƒ(Re,Pr) (63)
The equivalent relation for mass transfer is
Sh = ƒ(Re,Sc) (64)
where Sh is the Sherwood number and Sc is the Schmidt number (see Table 3).
Many standard handbooks and heat transfer textbooks list dimensionless transport
relationships for engineering systems (9,10).

4 STOICHIOMETRY AND CHEMICAL KINETICS

Stoichiometry is the study of the proportions in which elements combine to form
compounds. Chemical kinetics is the science dealing with the rates of chemical
reactions.

4.1 Stoichiometry
Consider a general chemical reaction in which reactants A, B, . . . react to form
products X, Z, . . . according to the equation
162 Andersen and Harcum

TABLE 3 Some Important Dimensionless Groups

Group Definition Significance
hL
Nusselt number, Nu Nu≡ Dimensionless heat
k
transfer coefficient
Péclet number, Pe Pe ≡ Re ⋅ Pr Ratio of heat conduction
(thermal) to convection
Péclet number, PeA PeA ≡ Re ⋅ ScA Ratio of species A diffu-
(chemical) ^ sion to convection
µ/ρ µC p
Prandtl number, Pr Pr ≡ ^ = Ratio of momentum
k/(ρC p) k
transport to heat
conduction
ρLν
Reynolds number, Re Re ≡ Ratio of inertial mo-
µ
mentum transport to
viscous momentum
transport
µ/ρ
Schmidt number, ScA Sck ≡ Ratio of viscous mo-
Dk
mentum transport to
kAL species diffusion
Sherwood number, ShA ShA ≡ Dimensionless mass
DA
transfer coefficient for
species A
Notes:
^
C p ≡ specific heat
Di ≡ diffusivity of species i
h ≡ heat transfer coefficient
ki ≡ mass transfer coefficient for species i
k ≡ thermal conductivity
L ≡ characteristic length
µ ≡ viscosity
ρ ≡ density

aA + bB + . . . = . . . + xX + zZ
This can be written in the equivalent form
0 = −aA − bB − . . . + xX + zZ
or, more compactly,
n
0 = ∑ νkIk (65)
k=1
Macroscopic Balance Equations 163

where νk is the stoichiometric coefficient of species Ik. Note that the stoichio-
metric coefficient is positive for a product and negative for a reactant.

4.2 Reaction Rate

The rate of a chemical reaction can be defined in terms of the stoichiometric
coefficient:
⋅ ⋅ ⋅ ⋅ ⋅
⋅ (NA )gen (N B )gen (N k )gen (N X )gen (N Z )gen
ξ = = = ... = = ... = = (66)
νA νB νk νX νZ

Here, ξ is the extent of reaction variable or reaction coordinate. The volumetric

reaction rate r may be defined by:
⋅
ξ
r = (67)
V
The volumetric reaction rate is usually a function of the temperature and
the concentrations of the various reactants and products:
r = ±k(T)ƒ(cA,cB, . . . ,cX,cZ) (68)

where k(T) is the so-called rate constant (which is not really a constant, but rather
is a function of temperature).
In most cases, the rate constant is given by the Arrhenius equation:
 −Ea 
k = k0 exp   (69)
 RT 
In this equation, k0 is the preexponential factor and Ea is the activation energy,
both of which must be determined experimentally.
The concentration-dependent part of the rate expression, ƒ(cA,cB, . . . , cX,
cZ, must also be determined experimentally. In many cases, the reaction rate is
found to depend on the concentrations of the reacting species as follows:
α β
ƒ(cA,cB, . . .) = cAcB . . . (70)

A reaction that obeys this equation is said to be of “α order” in species A, of “β

order” in species B, and so on. The overall order of the reaction n is the sum of
the individual reaction orders:
n = α + β + ... (71)
More information on the determination of rate constants and their applica-
tions can be found in standard texts on kinetics and reactor design (6–8).
164 Andersen and Harcum

5 EQUILIBRIUM THERMODYNAMICS
Thermodynamics is the study of the relationship between heat, work, and various
forms of energy. Of particular interest to thermodynamics are the conditions for
equilibrium.

5.1 Auxiliary Thermodynamic Functions

In addition to the energy and entropy, it is common in thermodynamics to define
certain auxiliary functions. One of these is the enthalpy or heat function H:
H = U + PV (72)
Two common free-energy functions are defined. The Helmholtz free-energy
function A is defined by
A = U − TS (73)
The Gibbs free-energy function G is defined by
G = H − TS (74)
The Gibbs free-energy function is of special interest because it provides a
convenient criterion of spontaneity and equilibrium.

5.2 Spontaneity and Equilibrium

A spontaneous process is one that occurs in a closed system without the assistance
of some external agency. The Gibbs free energy decreases in a spontaneous
process:
∆G ≤ 0 (spontaneous process) (75)
At equilibrium, the system does not change with time. The Gibbs free energy
reaches a minimum at equilibrium:
dG
= 0 (equilibrium) (76)
dt

5.3 Fugacity
The change in Gibbs free energy for an ideal gas undergoing an isothermal
compression or expansion from pressure P1 to pressure P2 is given by
P2
∆G = RT ln (ideal gas) (77)
P1
Although Eq. (77) does not apply for a real gas, it is common practice to retain
the same functional form for real gases:
Macroscopic Balance Equations 165

ƒ2
∆G = RT ln (real gas) (78)
ƒ1

Here, ƒ1 is the fugacity at pressure P1. The fugacity may be considered as an

adjusted pressure. It is defined so as to coincide with the pressure at low densities:
lim ƒ
= 1 (79)
P→0 P

The ratio of fugacity to pressure is called the fugacity coefficient, φ. Thus, the
previous equation may be written
ƒ lim
ϕ ≡ ⇒ ϕ = 1 (80)
P P→0

5.4 Chemical Potential

Consider a solution consisting of n species A, B, . . . . The Gibbs free energy of
the solution is given by
n
G = ∑ Nkµk (81)
k=1

where µk is the chemical potential of species k, defined by

 ∂G 
µk =   (82)
 ∂Nk T,P,Nj≠k

5.5 Fugacity and Activity

The chemical potential is generally a function of temperature, pressure, and
composition. It is common practice to write
µA = µ˚A(T) + RT ln aA (83)

where µ˚A(T) is the standard chemical potential and aA is the activity of species
A. The activity is defined as the ratio of the fugacity ƒA to a standard-state
fugacity ƒ˚A:
ƒA
aA = (84)
ƒ˚A

Activities and standard states are discussed in greater detail in many texts on
chemical thermodynamics (11,12).
166 Andersen and Harcum

5.6 Phase Equilibrium

Consider a multicomponent system that separates into two or more phases
(denoted I, II, . . .). The criteria for phase equilibrium are

TI = TII = . . . (thermal equilibrium) (85)

PI = PII = . . . (mechanical equilibrium) (86)
(µA)I = (µA)II = . . . (equilibrium for species A) (87)
(µB)I = (µB)II = . . . (equilibrium for species B) (88)
...

5.7 Reaction Equilibrium

Consider a reaction

aA + bB + . . . = . . . + xX + zZ

which, as before, can be expressed in the shorthand notation

n
0 = ∑ νkIk
k=1

The criterion for chemical reaction equilibrium is

n
∆G = ∑ νkµk = 0 (89)
k=1

Reaction equilibrium may also be expressed in terms of the standard Gibbs

free-energy change:

∆G0 = RT ln Ka (90)

Here, Ka is the equilibrium constant, defined by

n
Ka = ∏ akν k (91)
k=1

6 ENGINEERING FLUID MECHANICS

Fluid mechanics deals with the flow of liquids and gases. For most engineering
applications, a macroscopic approach is usually taken.
Macroscopic Balance Equations 167

6.1 Engineering Bernoulli Equation

Most engineering problems in fluid mechanics can be solved using the engineer-
ing Bernoulli equation, also called the mechanical energy balance. It can be
derived from the macroscopic energy balance (see Ref. 13), subject to the
following restrictions: (a) the system is at steady state; (b) the system has a single
fluid intake and a single outlet; (c) gravity is the sole body force, with constant
|g|; (d) the flow is incompressible; (e) the system may include one or more pumps
or turbines. Under these conditions, the macroscopic energy balance becomes
⋅ ⋅
P α  Ws F
∆  + |g|z + 〈ν〉2 = ⋅ − ⋅ (92)
ρ 2  m m
where
P = the fluid pressure
〈ν〉 = the velocity averaged over the cross section of the pipe or conduit
α = average velocity correction factor (2.0 for laminar flow and 1.07 for
turbulent flow)
⋅
W 2 = rate of work done by pumps or turbines (positive for pumps,
negative for turbines)
⋅
F = frictional loss rate
Dividing by the acceleration of gravity |g| yields the so-called head form of the
Bernoulli equation:
⋅ ⋅
 P α 2 Ws F
∆ + z + 〈ν〉  = ⋅ − ⋅ (93)
 ρ|g| 2|g|  m |g| m |g|
Each of the terms in this equation has the dimensions of length.

6.2 Fluid Friction in Pipes and Conduits

⋅
The frictional loss rate F equals the rate at which useful mechanical energy is
converted to thermal energy by friction. It is usually computed from an equation
of the form
⋅
F  L  〈ν〉2
⋅ = 4ƒ   (94)
m D  2
where ƒ is the Fanning friction factor.* In general, the friction factor is a function

*This is not the only friction factor in widespread use. Some authors prefer the Darcy-Wiessbach
friction factor, ƒDW = 4ƒ.
168 Andersen and Harcum

of the pipe diameter D, the surface roughness ε, and the Reynolds number Re,
the latter being defined as
ρD|ν|
Re = (95)
µ
where µ is the fluid viscosity.
In the laminar-flow regime, ƒ = 16/Re. For turbulent flows, a number of
charts, graphs, and equations are available to compute the friction factor. The
Colebrook equation has traditionally been used, although it requires a trial-and-
error solution to find ƒ:
1  ε/D 1.255 
= −4 log  +
ƒ 
(96)
ƒ
√  3.7 Re√
Wood’s approximation (Ref. 13) gives ƒ directly, without a trial-and-error
procedure:
ƒ = a + b Re−c (97)
where
0.225
ε ε
a = 0.0235   + 0.1325   (98)
D   D
0.44
ε
b = 22   (99)
 D
0.134
ε
c = 1.62   (100)
 D
The relations presented in this section were developed for cylindrical pipes or
tubes; however, the same equations may be used for noncylindrical ducts if the
pipe diameter D is replaced in Eqs. (94–100) by the hydraulic diameter DH:
volume of fluid
DH = 4 (101)
area wetted by fluid

6.3 Minor Losses

The relations developed in the previous section apply only to straight pipes or
conduits. Most pipelines, however, include bends, valves, and other fittings which
create additional frictional losses. These additional losses are often called “minor
losses,” although they may actually exceed the friction caused by the pipe itself.
There are two common ways to account for minor losses. One is to define
an equivalent length Leq which equals the length of straight pipe that would give
the same frictional loss as the valve or fitting in question. The total equivalent
Macroscopic Balance Equations 169

length Ltotal is the sum of the true length of the pipe and the individual equivalent
lengths of the valves and fitting:

Ltotal = Lpipe + ∑ (Leq)i (102)

fitting i

The total equivalent length Ltotal is used in Eq. (94) in place of L to compute the
total frictional losses.
The second common approach to computing minor losses relies on the
concept of a loss coefficient KL, defined for each type of valve or fitting according
to the equation
⋅
F 〈ν〉2
⋅ = KL (103)
m 2
A comprehensive listing of typical equivalent lengths and loss coefficients is
published by the Crane Company (14).

6.4 Fluid Friction in Porous Media

A porous medium is a solid material containing voids through which fluids may
flow. The most important single parameter used to described porous media is the
porosity or void fraction ε:

void volume Vvoids

ε = = (104)
bulk volume Vvoids + Vsolid

Fluid friction in a porous medium of thickness L is usually described by Darcy’s law:

⋅
F µL
⋅ = 〈ν〉 (105)
m ρ k
where k is a material property called the permeability.

REFERENCES
1. R. B. Bird, W. E. Stewart, and E. N. Lightfoot, Transport Phenomena. New York:
Wiley, 1960.
2. M. M. Denn, Process Fluid Mechanics. Englewood Cliffs, NJ: Prentice-Hall, 1980.
3. R. W. Fahien, Fundamentals of Transport Phenomena. New York: McGraw-Hill,
1983.
4. W. M. Deen, Analysis of Transport Phenomena. New York: Oxford University Press,
1998.
5. E. L. Cussler, Diffusion Mass Transfer in Fluid Systems, 2nd ed. Cambridge, U.K.:
Cambridge University Press, 1997.
170 Andersen and Harcum

6. O. Levenspiel, Chemical Reaction Engineering, 2nd ed. New York: Wiley, 1972.
7. H. S. Fogler, Elements of Chemical Reaction Engineering. Englewood Cliffs, NJ:
PTR Prentice-Hall, 1992.
8. L. D. Schmidt, The Engineering of Chemical Reactions. New York: Oxford Univer-
sity Press, 1998.
9. F. P. Incropera and D. P. DeWitt, Fundamentals of Heat and Mass Transfer, 3rd ed.
New York: Wiley, 1990.
10. D. R. Lide (ed.), CRC Handbook of Chemistry and Physics, 80th ed. Cleveland, OH:
CRC Press, 1999.
11. I. M. Klotz and R. M. Rosenberg. Chemical Thermodynamics: Basic Theory and
Methods, 3rd ed. Menlo Park, CA: W. A. Benjamin, 1972.
12. K. Denbigh, The Principles of Chemical Equilibrium, 3rd ed. Cambridge, U.K.:
Cambridge University Press, 1971.
13. N. De Nevers, Fluid Mechanics for Chemical Engineers, 2nd ed. New York:
McGraw-Hill, 1991.
14. Flow of Fluids Through Valves, Fittings, and Pipe, Crane Technical Paper 410.
Chicago: The Crane Company, 1988.
10
Biotechnology Principles

Teresa J. Cutright
The University of Akron, Akron, Ohio

1 INTRODUCTION
As mentioned throughout this text, waste minimization encompasses recycling/
reuse, waste reduction (material substitution, process changes, good housekeep-
ing, etc.), and waste treatment on-site (1,2). Biotechnology has a direct impact
on, and applicability to, an engineer’s ability to achieve waste minimization goals.
It has had demonstrated success with recycling programs via the generation of
biogas as an alternative fuel. Bioremediation approaches have also been used for:
point source reduction via biopolishing (3) and individual stream treatment (4);
by-product utilization (5); material substitution (6); facilitation of new enzy-
matic/metabolic pathways to produce “cleaner” organic substances (7); and
end-of-pipe treatments (8–10). This chapter will highlight a few of the biotech-
nology approaches to waste minimization.
When utilizing any biotechnology, it is important to remember that the
primary function of a microorganism is not to destroy man’s unwanted contami-
nants. Instead, a microbe must reproduce itself and maintain its cellular functions.
To that end, as shown in Figure 1, every microorganism must: (a) protect itself
from the environment, (b) secure nutrients (catabolism), (c) produce energy in a
usable form (catabolism), (d) convert nutrients/food into cellular material (anab-
olism); (e) discard unnecessary waste products, and (f) replication genetic infor-
171
 172

FIGURE 1 Overview of general cellular functions applicable to all living cells.

Cutright
Biotechnology Principles 173

mation. It is an added benefit to mankind that the result from the microbial
metabolism of substrates (i.e., step d), that the unwanted contaminants are
degraded. Once it was realized that microorganisms could degrade unwanted
contaminants, engineers started to manipulate the surrounding environment to
ensure that the microbes would thrive and utilize the contaminant as the substrate.
Engineers currently use microorganisms to treat drinking water, municipal
wastewater, and various industrial effluents. Usually the chemical and petrochem-
ical industry is considered the only “real” contributor to industrial effluent (11).
However, as shown in Table 1, more than just the chemical industry utilizes
microorganisms for the treatment of waste on-site.
Regardless of whether the microorganisms are being used for cleaning
drinking water, or municipal or industrial wastewaters; for end-of-pipe treatment
at contaminated sites; or for waste minimization applications, certain key aspects
apply (75). The following sections will outline the key aspects applicable to
any biological treatment, provide a brief description and design criteria for
the common waste minimization technologies, as well as highlight a few of the

TABLE 1 Some of the Industries that Utilize

Biological Treatment for the Reduction of
Waste

Industry References

Coal processing 12–14

Cosmetics 15,16
Dyes 16–20
Fertilizer plants 21,22
Food 23,24
Citric acid 23,24
Dairy 25–30
Poultry 31–34
Slaughterhouses 22,35,36
Vegetables 37,38
Heavy metals processing 13,39–45
Oil processing and refineries 46–49
Paint 50–53
Paper 6,21,22,54–57
Pesticides 13,18,58–60
Pharmaceutical 61–66
Printing 19,20,67,68
Soap 46,69
Tannery 22,70,71
Textiles 3,18,72–74
174 Cutright

innovative of bioprocesses that enable the selective removal of unwanted chem-

icals in product streams.

2 KEY ELEMENTS ESSENTIAL TO ALL BIOLOGICAL

TREATMENT METHODS
Several basic biological requirements are essential for any biological treatment
process to be successful. They are based on the principles required to support all
ecosystems and include the presence of: appropriate microbes for degrading the
contaminant(s), substrate for carbon and energy source, required terminal electron
acceptor (TEA), inducer to facilitate enzyme synthesis, nutrients for supporting
microbial growth, microbes to degrade metabolic byproducts, environmental
conditions to minimize growth of competitive organisms (76–78). These factors
will be discussed below.

2.1 Adequate Microbial Population

The primary requirement for any successful biological treatment or waste mini-
mization strategy is the presence of an adequate microbial population. Luckily for
environmental engineers, Mother Nature has supplied a wide variety of microbes
to select from and to cultivate. The organisms are subdivided into different
categories based on their metabolic capabilities and/or requirements. Table 2
contains the classifications based on the microbial carbon source, energy source,
and respiration mode. If a contaminant can only be degraded in the presence of
another organic material that serves as the primary electron source, then co-
metabolism is occurring. If the interaction of the two organisms is nonobligatory,
then it is a synergistic relationship. Mutalism occurs if the interaction is beneficial
yet obligatory. Since microbes are very versatile, it is important to remember that
they may belong to more than category.
The microbe’s versatility may also enable it to treat more than one par-
ticular type of contaminant. As shown in Table 3, different species of Pseudo-
monas have demonstrated success at reducing agricultural, heavy metal, food,
and solvent wastes. In each instance, the primary requirement for successful
treatment is the presence of an adequate population. Researchers have deter-
mined that a microbial count of 103–108 cfu/liter, 104–107 cfu/g would be
adequate for groundwater and soil applications, respectively (77,79–81). There-
fore, to ensure a successful biological treatment for waste minimization, a
minimum of 108 cfu/liter would be recommended. If the contaminant concen-
tration or toxicity increases, the microbial population will have to increase as
well. If the increase in biomass concentration does not result in the desired
treatment efficiency, the microbes being utilized may have to be changed to
another source.
Biotechnology Principles 175

TABLE 2 Microbial Classification Based on Growth Requirements: Carbon

and Energy Source, Respiration Mode

Term Definition

Prototrophs Most self-sufficient. Can synthesize all

required growth compounds given
CO2 or a single organic compound.
Auxotrophs Cannot synthesize all compounds
required for growth.
Carbon source:
Autotrophs C source from CO2. Ex: algae, photosyn-
thetic bacteria.
Heterotrophs C obtained from reduced form of
organic compound.
Methanotrophs, methanogens C and energy source from methane.
Energy source:
Phototroph Derive energy from light (i.e., algae).
Chemotroph Derive energy from chemical oxidation.
Organotrophs Energy from oxidation of organic
chemicals.
Lithotrophs Energy from oxidation of inorganic
chemicals.
TEA:
Aerobe Require oxygen source for growth.
Anaerobe (oligate) Cannot grow in the presence of oxygen.
Obtain TEA from different source.
Facultative anaerobe Can utilize O2 if present; however prefer-
able growth in absence of O2 via differ-
ent TEA.
Fermentation TEA obtained from organic compound.

The species indigenous to a natural environment contain several different

microbes living together (i.e., mixed population). Therefore, it is reasonable to
presume that it is highly unlikely that one bacterial strain will be successful for a
complete waste minimization scheme. Table 3 includes bacteria that have been
successful at degrading the parent compound of the specified waste. Most of the
references did not report the achievement of complete mineralization (conversion
of contaminant into CO2, biomass, H2O, and salts). Complete mineralization
would require the use of a microbial consortium. A consortium is more than a
simple group of bacteria that can grow together. The overall net effect of the
consortium is greater than what the individual microbes can accomplish on its
own. Consortiums facilitate the degradation sequence where one microbe de-
176 Cutright

TABLE 3 Select Microbial Genus with Demonstrated Success for

Treating Industrial Contaminants Based on Waste Classification

Waste type Microbial genus

Aliphatics Achromobacter Micrococcus

Acinetobacter Mycobacterium
Arthrobacter Pseuodmonas
Bacillus Vibrio
Flavobacterium
Agricultural (pesticides, Achromobacter Flavobacterium
herbicides) Alcaligenes Methylomonas
Arthrobacter Penicillium*
Athiorocaceae Pseudomonas
Corneybacterium* Zylerion*
Dyes Aeromonas Phaneorchaete*
Micrococcus Shigella
Klebsiella Trametes*
Pseudomonas
Food, dairy, slaughter Acinetobacter Nitrosomonas
Arthrobacter Pseudomonas
Bacillus Rhodococcus
Brevibacterium Vibrio
Metals Aeromonas Pseudomonas
Alteromonas Saccharomyces†
Bacillus
Enterobacter
Pulp and paper Arthrobacter Talaromyces*
Eisenia Trichoderma*
Chromobacter Xanthomonas
Sporotrichum*
Solvents Alcaligenes Nitrosomonas
Citrobacter Nocardia
Desulfomonite Pseudomonas
Enterobacter Rhodococcus
Morganeela Xanthobacter
Mycobacterium
*Fungi; †yeast.
Biotechnology Principles 177

grades the metabolites of the first. The number and types of microbes required
for a successful consortium depends on the contaminant classification, complex-
ity, and concentration.
Natural mixed populations (i.e., consortiums) can be viewed as an interac-
tive community that require each individual presence in order to thrive. When
unmanipulated, the mixed population will contain one or two species that
dominate the culture. These species are the most adaptable to the surrounding
environment, have the most efficient energy utilization, and often facilitate the
first step in the metabolic pathway. As time progress, the population may shift to
one in which a different species dominates to continue the metabolic pathway or
adjust for changes in substrate, nutrients, or terminal electron acceptor (82–84).
The microbes used for waste minimization applications have three basic
modes of growth: attached (fixed), suspended, or free growth. Attached growth
is similar to the biofilms used in wastewater treatment (trickling filters) and air
emissions. Bioflims are surface aggregates composed of layers of bacteria that
are embedded in a polysaccharide matrix. Biofilms differ from suspended growth
in that they are fixed in a stationary place. Suspended growth systems (i.e.,
activated sludge) still have the bacteria attached to a surface, but the surface is
freely moving within the reactor. Free growth systems are similar to slurry
treatments where the bacteria can sorb and desorb from a surface.

2.2 Terminal Electron Acceptor

Without an adequate supply or specific type of TEA, biological treatment will
fail. Table 4 includes the primary electron acceptors used. Aerobic microbes
utilize O2 as the TEA. For strict aerobes, the oxygen source is typically obtained
from air or H2O2. Incorporation of alternative TEA sources provide another name
to identify the respiration–microbe interaction. For instance, denitrifying bacteria
utilize nitrate (NO3− → NO2− → N2) as the TEA. Nitrobacter sp. is one of the
microbes capable of nitrification, the conversion of NH3-N to nitrates and
nitrates. Sulfate reducers, such as Desulfovibrio, utilize SO42– and generate

TABLE 4 Typical TEAs and Their Associated Respiration Modes

TEA Form Mode

Oxygen O2 Aerobic
Nitrate NO3 Anaerobic
Sulfate SO4 Anaerobic
Carbon dioxide CO2 Methane fermentation
Organic compound Various Aerobic, anaerobic, fermentation
178 Cutright

S2– (85). Methanogens require CO2 as the TEA. Various bacteria can obtain the
TEA from organic compounds.
The specific type of electron acceptor will dictate the metabolism mode,
and thus the subsequent degradation reactions. Therefore the combination of
microbes and TEAs utilized will enable a specific degradation pathway to be
followed to facilitate cometabolism, prevent accumulation of toxic intermediates,
etc. (86,87).

2.3 Nutrients
The nutrient requirements for microbes are approximately the same as the
composition of their cells (76,88). The exception to this is carbon, which is
sometimes needed at higher quantities and can be supplied by the contaminant
for heterotrophic microorganisms. There are three categories of nutrients based
on the quantity and essential need for them by the microorganism: macro, micro,
and trace nutrients (89,90). For example, the macronutrients carbon, nitrogen, and
phosphorus are known to comprise 50%, 14%, and 3% dry weight, respectively,
of a characteristic microbial cell. Sulfur, calcium, and magnesium, which are
micronutrients, comprise only 1%, 0.5%, and 0.5%, respectively of the cells dry
weight (75,91,92). Trace nutrients, which are found in the least quantity, are not
required by all organisms. The most common trace elements are iron, manganese,
cobalt, copper, and zinc. Based on this approach, the optimal C:N:P mole ratio
recommended for bioremediation applications is 100:10:1 (77,93,94). For exam-
ple, 150 mg of nitrogen and 30 mg of phosphorous would be required to degrade
1 g of a theoretical hydrocarbon into cellular material. If the carbon source were
easily and rapidly converted into carbon dioxide, then more carbon would be
required in order to sustain microbes. For this reason, an additional carbon
supplement may need to be supplied to facilitate the degradation of the contam-
inant depending on the microbe–contaminant interaction. The limits of the carbon
(substrate) concentration will dictate the source and concentration of supplemen-
tal carbon and nutrients that are required (95).
Nitrogen is the nutrient most commonly added at bioremediation projects. It is
primarily used for cellular growth (NH 4+ or NO3−) for the synthesis of cellular
proteins and cell-wall components. It can also be used as an alternative electron
acceptor (NO3−). It is commonly added as urea or as ammonium chloride but may
also be supplied as any ammonia salt or ammonium nitrate (92). All of these forms
are readily assimilated in bacterial metabolism. However, if an ammonium ion is used
to supply nitrogen, an increased oxygen demand on the surrounding system will be
created facilitating the need for additional TEA supplements (96).
Phosphorus is the second most commonly added nutrient in bioremediation
and is supplied to serve as a source for cellular growth. Often the ability of the
microbes to secure the required phosphorous levels depends on the correct
Biotechnology Principles 179

nitrogen level to be already in place (97). Phosphorus may be added as potassium

phosphate, sodium phosphate, or orthophosphoric and polyphosphate salts. Potas-
sium phosphate (mono- and di-basic) can also serve as a buffering agent to control
pH. However, there are also cautions regarding the addition of a phosphorus
source. The addition of potassium phosphate may accelerate the cleavage of the
hydrogen peroxide added as an oxygen source. If the hydrogen peroxide cleaves
too quickly, the oxygen source may be depleted before it reaches the contami-
nated zone. The maximum orthophosphate addition to provide microbial nutrients
while avoiding significant precipitation, peroxide cleavage, or toxicity in most
environments is 10 mg/liter (81).
Currently, there are no specific methods for predetermining the exact
nutrient sources to utilize for a given situation or the role they will play. The
specific ratio will depend on the chemicals to be treated, the microbes to be
utilized, and the presence of inorganic nutrients already in the waste stream (98).
The presence of a limiting nutrient will greatly affect the extent of remedial
treatment. Therefore, researchers have conducted studies to determine some
generalizations based on the type of microbe incorporated. For instance, nitrogen
is required in smaller quantities when fungi are utilized. Magnesium is required
in greater concentrations for photosynthetic bacteria. For aerobic cultures, iron is
necessary in higher quantities than their anaerobic counterparts. Even with these
generalizations, it is always a good idea to conduct a quick feasibility experiment
to determine the exact nutrients required (77). This is particularly true for
biological waste minimization processes with “exotic” chemicals, such as syn-
thetic polymers, pesticides, and pharmaceuticals.

2.4 Environmental Conditions

There are several other environmental conditions that are critical to biological
applications. The two major conditions are temperature and pH. The majority of
microorganisms can grow only in a specific temperature range. If the microbes
can grow only over a 10˚C range, they are termed stenothermoal. Eurythermal
organisms can grow over a 40˚C range. Three cardinal temperatures are used to
further describe microbial growth, the minimum, maximum, and optimum tem-
peratures. Psychrophilic organisms have an optimum temperature of ~15˚C with
a minimum of <0˚C (99). Most of the organisms used for degrading organic
chemicals are mesophiles that have an optimum of 30˚C. Thermophilic organisms
have an optimal growth temperature above 45˚C. Some researchers report the
thermophilic optimum to be ~60˚C, while some have listed a temperature closer
to 90˚C (100–102). When the surrounding temperature is beyond the optimum,
microbial activity decreases (89). Typically, the loss of enzymatic activity is
attributed to thermal denaturation that alters the membrane structure of the
microorganism, thereby decreasing its normal cellular functions.
180 Cutright

The pH level can affect the microbe’s ability to conduct cellular functions,
cell membrane transport, and the equilibrium of enzyme-catalyzed reactions. The
minimum and maximum pH for microbial growth typically differs by 3 log units.
Most bacteria can exist at a pH between 5 and 9, but have an optimal value near
7. Some bacteria are like fungi in that they prefer a more acidic environment (pH
1–3). For instance, Thiobacillus thioxidans has an optimum of 2.5 and is therefore
classified as an acidophilic organism. Other bacteria, such as Cyanobacteria and
Bacillus sphaericus, have a more alkaline optimum. These species are referred to
as alkalophilic bacteria (103).
The pH and supplemental nutrients may effect the redox potential, which
indicates the available TEA. Redox potential defines the electron availability as it
affects the oxidation states of hydrogen, carbon, nitrogen, oxygen, sulfur, manganese,
iron, etc. (104). As the surrounding environment is reduced, the electron density is
increased and the redox potential becomes more negative. For an optimal aerobic
environment, the redox potential must be greater than 50 mV (78,105). If nutrient
additions or deviations in pH change the redox potential the necessary TEA/
respiration mode may not be present for the desired degradation reaction.

3 COMMON BIOPROCESSES USED FOR WASTE

CONTROL
3.1 Suspended Growth
The most widely studied and used suspended growth process is the activated
sludge system used for the majority of industrial pretreatment and municipal
wastewater treatment plants. The more common orientations for activated sludge
include completely mixed, plug flow, oxidation ditch, contact stabilization, and
sequencing batch reactors (106,107). Each configuration requires that the appro-
priate biomass, TEA level and source, pH, and contaminant loading be in place
in order for the treatment to be successful. Since activated sludge processes are
the most widely implemented suspended growth system, they have been the first
used for waste minimization applications (10,92,108,109). The design equations
and common operating parameters will be covered only briefly. More detailed
information pertaining to activated sludge systems can be found in any wastewa-
ter or general civil engineering textbook.
Figure 2 represents the completely mixed activated sludge model typically
found in industrial pretreatment applications. Here i, e, r, and w are used to denote
the influent, effluent, recycle, and waste conditions, respectively. Other parame-
ters include:

Q = volumetric flow rate, m3/d

Va = volume of the aeration tank, m3
Biotechnology Principles 181

FIGURE 2 Process diagram for completely mixed activated sludge model.

c = contaminant concentration, g/m3

x = biomass concentration, g/m3
The substrate (contaminant) and biomass material balances are used to obtain the
required operating design equations. In order to develop the substrate and biomass
material balances, the following several assumptions must be made:
Complete mixing occurs.
Influent flows and substrate concentrations are constant, i.e., the system is
at steady state.
Influent biomass (xi) is zero.
All of the substrate, food, is soluble.
Biological activity occurs only in the aeration tank (no substrate is removed
in the clarifier).
Sludge wasting occurs only in the clarifier.
Solids mass in clarifier and recycle line << mass solids in aeration tank (xVa).
3.1.1 Biomass Material Balance
Using the assumptions stated above, a material balance initially neglecting the
recycle stream is made:
Influent biomass + biomass production =
effluent biomass + wasted biomass
dx
Qixi + V = (Qi − Qw)xe + Qwxw (1)
dt
In order to simplify Eq. (1), relationships pertaining to the sludge age, θc, and
biomass generation must be known. Depending on the wastewater, the amount of
biomass in the aeration tank is 800–6000 mg/liter. The cell age is a measure of
the average amount of time the biological solids remain in the aeration tank and
is used to help maintain the desired active biomass concentration. The sludge age
and biomass generation terms are typically expressed as shown in Eqs. (2) and
(3), respectively:
182 Cutright

Vx xVa
θc = = (2)
(dx/dt)R Qwxr + (Qi − Qw)xe
 dx   dc 
 dt  = Y   − kdx (3)
 gen  dt u
where the kd x term is used to describe the amount of energy to keep the biomass
alive and Y is the true cell yield. Substituting Eqs. (2), (3), and the Monod
relationship for (dc/dt) into Eq. (1) will yield the common design equation. After
rearranging, Eq. (1) becomes
K(θc−1 + kd)
ce = (4)
Yk − [θc−1 + kd]
where
K = Monod saturation constant, g/m3
kd = microbial decay coefficient, d–1
k = maximum specific substrate utilization rate, d–1
Y = true cell growth yield, g cell produced per g cell removed
ce = effluent concentration, g/m3
θc = mean cell residence time, d
Using Eq. (4), the only variable is the cell age, which can be used to control the
effluent concentration. In other words, if the desired effluent is not achieved,
the cell age is increased which increases the substrate residence time, thereby
enabling the biomass to have a longer time to facilitate the substrate degradation.
3.1.2 Substrate Material Balance
One of the uses of the substrate material balance is to determine the volume of
the recycle stream required for achieving the desired effluent concentration. In
this situation, the system boundary is around the aeration tank and includes the
recycle stream.
Using the basic material balance, accumulation = in – out + generation, and
the assumptions stated previously, the substrate material balance is
 dc 
Qici + Qrce = (Qi + Qr)ce +   Va (5)
 dt u
Grouping like terms and incorporating the relationship for cell age yields
θc−1 + kd Q(ci − ce)
= (6)
Y xVa
Rearranging Eq. (6) to group process variables yields
Biotechnology Principles 183

xVa Y(ci − ce)

= −1 (7)
Q θc + k d
Equation (7) can be further simplified by introducing the hydraulic residence
time, θH:
Y(ci − ce)
xθH = (8)
θc−1 + kd

The form given in Eq. (8) is the operating substrate design equation used by most
wastewater treatment facilities. This form allows, for a fixed effluent substrate
concentration and cell age, the hydraulic residence time and required biomass
concentration to be determined.
3.1.3 Typical Design Values and Waste Applications
As stated previously, the average initial biomass concentration for activated
sludge systems is in the range of 800–6000 mg/liter. The specific value used
depends on the food substrate (F) to microbe (M) ratio. Most F/M ratios are in
the range 0.04–0.07 [107]. The higher ratio is implemented for elevated organic
loading rates that require more microbial activity to achieve the desired effluent
concentrations. For completely mixed activated sludge systems with a 5-day cell
age, the optimal loading rate of 1 kg substrate/day/m3 would require at least 25%
of the water to be recycled through the aeration tank.
Activated sludge systems have had demonstrated success with conventional
municipal wastewater. They have also been used for dairy waste, nonpesticide
agricultural waste, pharmaceutical waste, and low-strength singular solvents in
industrial pretreatment. Activated sludge is not effective for exotic organic
chemicals such as pesticides, high-strength solvents, elevated heavy metal con-
centrations, alkaline waters, or mixed wastes. Copper, nickel, and zinc at concen-
trations as low as 1 mg/liter have been shown to inhibit microbial activity (110).
As effluent constraints become more stringent, the applicability of basic activated
sludge systems for treating most industrial pretreatment waste streams will
decrease unless the waste streams are segregated and the biomass acclimated to
the specific waste present (111,112).

3.2 Attached Growth

Attached growth systems encompasses processes from biofilms to slime layers.
Biofilms are used for facilitating biooxidation reactions of contaminated gas
emissions. Slime layers are utilized for biofiltration.
3.2.1 Biofiltration
The two most common processes employed for biofiltration are trickling filters
(TFs) and rotating biological contactors (RBCs). For wastewater applications, the
184 Cutright

slime layers are comprised of a mixed population of bacteria, protozoa, and fungi.
Depending on the waste constituents, the dominant species in the population will
change. Also, the protozoa are employed as an auxiliary control on the slime
thickness. If the slime layer is too large, the treatment efficiency will be decreased
due to mass transfer limitations.
Trickling Filters. Trickling filters are not filters as their name implies, but
a means of providing a large surface area where the microbes can attach as they
“feed” on the organics. The media support is usually in the form of rocks, stones,
or ceramics, depending on the surface area and throughput required. As depicted
in Figure 3, water trickles over the top of the media to provide the contact between
the microbial population and contaminant. TFs are classified as an aerobic
treatment since aerobic cultures are responsible for over 75% of the degradation.
However, aerobic cultures are located only to a film depth of 0.1–0.2 mm; toward
the center of the media, faculative anaerobes dominate the microbial population,
as dissolved oxygen concentrations at this point are minimal (84,113,114).
TFs have had demonstrated success at both the pilot scale (for exotic
chemicals) and full scale for municipal-industrial pretreatment. At the pilot scale,
TFs in different orientations have been effective for both inorganics and organic
constituents. One study demonstrated a 94% reduction of manganese, while
another study degraded 97% of a 300-ppm styrene waste stream (115). TFs have
been used for over 30 years as a secondary or tertiary treatment for municipal
wastewater (116). Full-scale effectiveness has also be proven for some recalci-
trant organics as well. For instance, one plant was used to treat a waste stream
containing 460 mg/liter of alkyl ethoxylate sulfates and alkylbenzene sulfonates
to between 46 and 130 mg/liter (115).

FIGURE 3 General representation of a trickling filter.

Biotechnology Principles 185

The three primary design variables for TFs are the hydraulic loading rate
(HLR), organic loading rate (OLR), and recycle ratio (α). The parameters are
determined from material balances around the system shown in Figure 4 as

Qi + Qr
HLR = (9)
A
Qici + Qrce
OLR = (10)
V
Qr
α = (11)
Q

Both the HLR and OLR will directly affect and the depth of media required
for treating the contaminant. As shown in Eqs. (9) and (10), OLR and HLR are
interconnected, thus providing engineers some room to manipulate actual values
and still achieve the desired effluent constraints. For instance, changing the
medium depth will increase the residence time, thereby enabling the microbes to
have greater contact with the contaminant. Varying will assist with maintaining
the biofilm thickness and minimum flow rate, dampen the OLR variation, and
increase the contact with the organics. Recirculation ratios will typically vary
from 0.4 to 4, depending on the effluent constraints. Values greater than 4 are
used to treat higher-strength waste streams. With this approach, a system with
high-strength wastewater may have a low HLR and an elevated OLR. However,
if the OLR is sustained beyond the design capacity, the filter will become clogged
and ultimately fail (106).
The first design equations for TFs were empirical. Over the past 20 years,
these equations were modified to incorporate biochemical kinetics. The most
frequently used design equation was developed by Eckenfelder (116,117) as
n
ce   As  
= exp−k   zavm (12)
ci∗   (1 + α)Q  

growth area

Q ci Q ce

FIGURE 4 General representation of a rotating biological contactor.

186 Cutright

where

ce = effluent concentration, g/m3

ci* = influent concentration, including recycle, g/m3
k = reaction rate constant, d–1
As = cross-sectional filter area, m2
α = recycle ratio
z = filter depth, m
av = specific surface area per packing piece
m, n = emperical constants determined by lab testing

Biotowers. Biotowers are the common biological treatment implemented

for minimizing waste in the dairy, milk, and food processing industries (107).
They are similar to TFs in design, operation, and use. HLR, OLR, and are still
the key design and operating parameters. The primary difference is that biotowers
use plastic media instead of the stones or rocks used in TFs. By changing the
media material to plastic, higher hydraulic and organic loading rates are obtain-
able, thus enabling biotowers to treat high-strength industrial wastes. Some
pilot-scale trials have also investigated the use of soil, granular activated carbon,
and perlite as the support material. However, polyethylene is still the primary
support/packing for biotower applications.
Biotowers have been used to treat waste streams containing more than 200
ppm of the more recalcitrant compounds. For instance, methyl ethyl ketone,
styrene, trichloromethane, and kerosene were each degraded by over 90% (82).
These efficiencies were achieved when each compound was present as the sole
contaminant. If the contaminants were present as a composite waste stream, the
biomass would have to be altered. Even with a different microbial consortium,
the remediation efficiency may not be as high, due to the degradation by products
that would be formed. Researchers are currently addressing this situation.
Based on preliminary successes, researchers are investigating the applicability of
biotowers for treating composite streams in the pharmaceutical, textile, surface
coating, and polymer processing industries (82).

Rotating Biological Contactors. Rotating biological contactors (RBCs)

employ aerobic fixed film treatment to degrade either organic or ammonia-based
constituents in wastewater. As shown in Figure 4, a RBC is comprised of a series
of corrugated disks 3–4 mm in diameter mounted on a horizontal shaft. Spacing
between the disks is in the range of 30–40 mm. Microbial slime layers forms on
the disks’ surface as they slowly rotate through the water (~1–2 rpm). The biofilm
formed on each disk is approximately 0.3–4 mm thick. A minimum of 40% of the
Biotechnology Principles 187

disk surface area is submerged in the waste stream as the RBC is rotated, to
provide the required moisture as well as to bring the microbes in contact with the
contaminants. The remaining surface area is exposed to the atmosphere, to supply
the required oxygen as well as to facilitate the biooxidation. One RBC unit can
provide up to 10,000 m2 of surface area for biological growth.
Often, several RBCs are connected in parallel to treat the wastewater in a
timely manner. Placement of modular units in series enables the attainment of
treatment levels exceeding that of conventional secondary treatment (118).
As with TFs, Eckenfelder et al. (119) developed a relationship to determine
the removal efficiency based on contaminant concentration:
ce  (Q/A)0.558 
= 14.2   (13)
 exp(0.32Ns)ci
ci 0.684 T 0.248

where
Q = mass rate of pollutant, g/d
A = disk area, ft2
Ns = number of RBC units used
T = water temperature, ˚C
It is important to note that Eq. (13) provides a conservative estimate for the
biodegradation efficiency, since it does not incorporate the reaction kinetics.
Eckenfelder and his associates did modify the equation to include a reaction rate
term. However, pilot-plant studies would be required to obtain the actual k value
for the specific biomass–waste combination used. Currently, the kinetic data for
RBCs are extremely limited, even for municipal wastewater applications (120).
RBCs are advantageous because they have a higher biomass concentration
than most treatment systems and are easier to operate under varying load
conditions. Unfortunately, they are prone to operational problems attributed to
biomass buildup (114). Although proven effective for low-strength organic (<1%
organic contaminants) and ammonia streams, RBCs are inefficient for high heavy
metal concentrations and complex organics such as pesticides and chlorinated
solvents (106,121). RBCs also tend to be very sensitive to temperature fluctua-
tions. For these reasons, RBCs are currently used only as secondary or tertiary
treatments.
3.2.2 Bioxidation—Scrubbing
With the newest amendment to the Clean Air Act, contaminants present in the
gas/air phase are beginning to come under closer scrutiny due to the adverse
affects they pose to human health and the environment. Part of the stricter
regulation is attributed to the acute toxicity that the contaminants pose, as well as
188 Cutright

the fact that they are an unseen threat and disperse quickly (122). For instance,
the 1984 accidental spill in Bhopal resulted in a death toll of 2500 people in less
than 2 h. Although incidents such as that at Bhopal are not frequent, air emissions
normally released from processing facilities still pose a serious threat. These
gaseous contaminants can be efficiently and cost-effectively treated biologically
via biofilters, TFs, and bioscrubbers.
Biofilters and TFs for gaseous systems operate on the same principles as
their liquid counterparts, the only difference being the medium to be treated.
Biofilters are essentially packed beds through which the contaminated gas stream
is ventilated (14,50,123). The beds have depths of 3–4 ft and can be packed with
ceramic material, soils, plastic, or inovative materials such as volcanic rocks
(124,125). The gas enters the bottom of the bed with an upward gas velocity
ranging from 0.0005 to 0.5 ft/s and residence times of 15–60 s, depending on the
bed volume (126,127). TFs are also attached growth systems. Instead of water
being trickled through the support system, liquid nutrients and contaminated gas
are passed through (128).
Bioscrubbers are comprised of biological oxidation and an adsorption step
in which the pollutants are adsorbed to water and a bioreactor for treating the
water (129). The incorporation of thermophilic bacteria or coolants for the air
stream enables bioscrubbers to be modified and extends their applicability. With
such modifications, the adsorption column serves as both the contaminant sorp-
tion mechanisms as well as the biological support media, thereby combining the
two steps in conventional bioscrubbers (130).
As with the liquid-based approach, gas treatments employ a consortia of
bacteria instead of a pure culture. Consortiums are used since they are more
effective and often occur naturally as the introduction of new, unidentified
cultures that result from nonsterile operations (131). The combination of
seeded and unidentified cultures has resulted in 75–90% remediation of the
waste constituents, depending on the initial contaminant type and concentration
(132,133).
For instance, a biotrickling reactor developed in Germany has demonstrated
full-scale success at treating solvent-laden air. Over a 3-mo period, reactors
removed 200 g/m3 of solvent per hour at an OLR of 600 m3/m3 filter/h (134).
Other researchers have proven the ability of mixed consortia to remove over 90%
of a BTEX-contaminated off-gas (131,135). A different continuous biofilter
achieved a maximum elimination capacity of 195 g/m3/h for ethanol vapors (132).
Biological treatment of off-gases has also been effective for methyl tert-butyl
ether, toluene, and styrene (52,115,136). In most bioscrubbing applications, the
removal performance based on contaminant classification follows a sequence
similar to most biodegradation trends. In other words, removal rates follow the
order alcohols > esters > ketones > aromatics > alkanes (137).
Biotechnology Principles 189

4 INNOVATIVE BIOLOGICAL APPROACHES TO WASTE

MINIMIZATION
4.1 Microorganisms for Material Substitution or
By-product Utilization
Two of the key approaches to waste minimization are material substitution and
by-product utilization (138,139). If the raw material that produces the waste can
be substituted with one that either minimizes the waste or yields a viable product,
then the waste minimization goal has been realized. Furthermore, waste minimi-
zation is achievable if the intermediates or by-products of the process can also be
reused. Microorganisms have the capability to be effective vehicles for both of
these approaches, as outlined by the brief examples provided below.
4.1.1 Material Substitution
One example of a biological approach to material substitution in waste minimi-
zation is used in the formation of phenylacetylcarbinol (PAC). PAC is the key
precursor in the pharmaceutical industry for nasal decongestants. During PAC
formation, too much of a benzaldehyde intermediate is also made and disposed
of as a waste stream (140,141). Recent studies have shown that by changing the
yeast used to produce the enzyme for PAC formation, the benzaldehyde concen-
tration is lowered (142,143). Some of these studies simultaneously increased the
product yield by 140 mmol/liter (144).
p-Hydroxybenzoate (HBA) is another key raw material used in the pharma-
ceutical industry. Typically, HBA is produced by the Kolbe-Schmidt carboxyla-
tion of phenol, but this approach produces hazardous wastes (145,146).
Bioconversion of toluene via Pseudomonas putida can also produce p-HBA. In
fact, substitution with a biological approach increased the product formation
while simultaneously eliminating the hazardous intermediate(s) (147).
The two aforementioned material substitution examples are slightly differ-
ent from the nonbiological methods in that a raw material was not substituted.
Instead, the seed culture or culture for processing an intermediate was substituted
for another process. In both scenarios the bacterial substitution altered the
metabolic pathway to remove an unwanted by-product.
4.1.2 Biodegradable Polymers
Perhaps the biggest area of material substitution involves the development of
biodegradable polymers (148–150). Engineers and environmentalists know that
synthetic polymers are essentially nonbiodegradable within an acceptable
timeframe. Natural polymers, such as cellulose, proteins, starch, and poly-
dhroxyalkanoates (PHAs), can be successfully constructed from both bacteria and
plant matter (151–153). These new, natural polymers are particularly advanta-
geous to waste minimization processes since not only do they eliminate the toxic
190 Cutright

waste streams generated via synthetic polymers, they are completely degradable
by bacterial and plant enzymes. Furthermore, the end product is often “cleaner”
than the synthetic counterpart (154). For these reasons, biopolymers are showing
great promise for use in food packaging, medical delivery devices, and cosmetics
(155–158).
Cellulose-based biopolymers have been extensively studied and therefore
will not be presented here (159–162). It is important to note that the information
gained by studying cellulose-based biopolymers led to the identification of
several new approaches to natural polymers. One of the key developments was
the substitution of PHAs for synthetic materials. PHAs are attracting attention as
the most viable biopolymers due to their similarity to synthetic polymers while
maintaining their inherent biodegradability (163–165). PHAs are produced via
both plant and bacterial metabolic pathways (166–168).
Lee and Choi (169) documented the four primary metabolic pathways for
the biosynthesis of PHAs. Two pathways involve Ralstonia eutrophia and Rhodo-
spirillium rubrum. The other two pathways are found in most Pseudomonads.
Due to the prevalence of the Pseudomonad species, these two particular biosyn-
thetic pathways can yield an almost endless supply of PHAs. Other bacteria and
their associated pathways have also been studied for PHA production [170]. Since
the bacterial source and pathway used for generating the PHA is known, it is easy
to identify and employ the enzyme required for degrading the polymer (171).
For instance, cellulose-based and PHA polymers are easily degraded by
Hyphomicrobium sp., Rhodococcus rhidochrous, and Trichoderma reesei (172).
This approach has also been used for the production and eventual degradation of
other relatively new biopolymers such as polyhydroxybutyrates (173,174).
4.1.3 Conversion of a Waste Stream to a Viable Product
For every 9 lb of cheese produced, 1 lb of whey is generated as a waste stream.
Currently the worldwide production of whey is 260,000 lb per year (175). Due to
the high percentage of crude protein found in the whey, it is highly desirable as
a substrate for producing various enzymes and oligosaccharides. For instance,
lactase, the enzyme used to hydrolyze lactose into the simple sugars of glucose
and galactose, is found naturally in the digestive tract of most humans and in
several plants. Lactase can also be generated by a number of different yeast
(Saccharomyces fragilis, Trouls utilis, Kluveromyces marcianus, and Kluveromy-
ces fragilis) cultivated on the whey (176).
Other viable materials can be generated from the whey by changing the
yeast, microbial, or fungi source. Ten different fungal strains are capable of
producing proteinase enzymes from whey (177). Gellan gum is generated when
Sphingomonas paucimobilis is utilized (178). A natural surfactanct, sophorolipid,
is the end yield for processes utilizing species of Cryptococcus curvaatus,
Candida bombicola, and Mucor miehei (179). The aforementioned are just of few
Biotechnology Principles 191

of the innovations that have demonstrated the capability of biotechnology to

convert a useless waste stream into a viable product.

4.2 Isolated Enzymes and Biocatalysts

While microbial processing may offer effective decomposition of organic com-
pounds, the practice of microbial degradation on a large scale can be complicated
by a few inherent limitations (pH sensitivity, nutrient amendments, population
dynamics, etc.), depending on the waste stream to be treated. In addition,
compounds such as organic solvents often present great toxicity toward microor-
ganisms. Above all of these, most of the recalcitrant organic contaminants usually
possess minimal solubility in the aqueous phase, where the microorganisms are
hosted and considered the most active. Recent advances in biocatalysis have
demonstrated that it is feasible to carry out biotransformations in nearly pure
organic media (180–182).
Enzymes are biocatalysts, or proteins secreted by microorganisms to accel-
erate the rate of a specific biochemical reaction without being consumed in the
reaction. The key environmental parameters critical to microorganisms must be
in place for producing the enzymes from the intact microbial cells. The difference
between normal microbial degradation and biocatalysis occurs once the enzymes
have been harvested. After harvesting, the enzymes do not require subsequent
nutrient, TEA, or energy source amendments. As with intact cells, the enzymes
can either be utilized in a free suspension or immobilized on a support. Im-
mobilized enzymes often show improved stability during variations in environ-
mental conditions over free enzymes. For example, when laccase was
immobilized for treating elevated concentrations of phenol, it maintained over
80% of its activity when subjected to changes in pH, temperature, and storage
conditions (183).
Isolated enzymes can also afford much faster reactions, which are usually
several orders of magnitude higher than those resulting from traditional microbial
process (184). The increase in reaction time has been attributed to the absence of
competing processes that are present with the natural metabolism of intact
organisms. Since the competing processes are absent, enzymes possess a greater
specificity toward a particular compound. An example of an enzyme’s ability to
treat recalcitrant compounds has been demonstrated by the textile industry.
Lacasse is the primary extracellular enzyme produced by Trametes versi-
color. Lignin peroxidase is one of the many enzymes secreted by P. chrysospor-
ium. Both of these enzymes have been used for the decolorization of the synthetic
dyes contained in textile effluents (20). Utilization of these isolated enzymes
yielded greater remediation efficiencies in less time and was independent of
microbial growth rates. Decolorizations of 80% and over 90% were achieved for
anthraquinone dyes with lignin peroxidase and lacasse (20,185). Lacasse was also
192 Cutright

capable of degrading the reaction by-products present in the textile effluent.

Lacasse and lignin peroxidase were not the only enzymes that have been used for
decolorizing textile effluents. Horseradish peroxidases and manganese peroxi-
dases have also shown success for decolorizing effluents from the textile and
paper industries (186–189).
As stated earlier, enzymes can also possess greater affinity toward a
particular compound. Therefore, utilization of enzymes as biocatalysts is one of
the processes that enable the selective removal of unwanted chemicals in product
streams (190,191). For example, researchers have recently utilized enzymes to
minimize waste at a cattle dipping facility. The cattle dipping liquid must be
discarded when potasan, a dechlorination by-product, accumulates to a specified
level. Due to the high degree of chlorination, the waste cannot be reused. Grice
et al. (192) employed parathion hydrolase to selectively hydrolyze the potasan.
Since the potasan concentration was reduced, the use of the dipping liquid was
extended, thereby decreasing the amount of waste generated.
Phenol and substituted phenolics are among the constituents present in most
industrial wastewaters. Due to their prominence in the waste streams and associ-
ated toxicity, phenolics have been identified for selective removal via enzymes.
One approach is to use tyrosinase to convert phenol to o-quinones, which are then
easily adsorbed (193). This two-step process was effective even in the presence
of microbial inhibitors. When the tyrosine was not immobilized, the remediation
efficiency was decreased. Other illustrations of treating phenolic wastewaters
include the treatment of 4-chlorophenol with horseradish peroxidase (194),
pentachlorophenol with horseradish peroxidase (195), and various phenol substi-
tutions with immobilized peroxidases (196,197).

4.3 Biological Recovery and/or Treatment of Heavy Metals

4.3.1 Background
Heavy metals are among the contaminant classifications receiving the greatest
scrutiny in waste minimization programs. These compounds are present in normal
municipal wastewater and in various industrial effluents such as those of elec-
troplating and metal finishing. Not all heavy metals pose a threat to the microor-
ganisms used in treatment operations. For instance, Fe, Mo, and Mn are important
trace elements with very low toxicity, while Zn, Ni, Cu, V, Co, W, and Cr are
classified as toxic elements. Although toxic, Zn, Ni, and Cu have moderate
importance as trace elements. As, Ag, Cd, Hg, and Pb have a limited beneficial
function to most microorganisms (198).
Activated sludge consortiums and other microbial processes can tolerate
most heavy metals at very low concentrations. As heavy metal concentration
increases, the metals can become harmful to the bacteria (nontoxic), achieve
toxicity status, or pass through the system unaltered. All three of these scenarios
Biotechnology Principles 193

pose a serious threat to microbial and human health. When metal toxicity
becomes too great, the microbes that are required for degrading the organic
compounds perish. If metals such as Cu, Cr, and Zn pass through the system and
are ingested, chronic human health disorders will arise (199). Recently, scientists
have found that certain microorganisms either possess an inherent resistance to
heavy metals or can develop the resistance through changing their internal
metabolism.
Metal-resistant microbes have potential applications to facilitate a biotech-
nology process that occurs in the presence of, but does not require, heavy metals.
Bacteria that can resist toxic metal concentrations could be utilized to convert the
organic constituent present into a readily usable or disposable form. Microorga-
nisms that can do more than simply tolerate elevated metal concentration can
be used in the recovery of metals or bioremediation of contaminated media
(198,200–202). The microbial processes that remediate or recover heavy met-
als include leaching (biological solubilization), precipitation, sequestration, and
biosorption (203–206). Biosorption has been the most widely studied aspect for
waste minimization activities.
4.3.2 Biosorption
Donmez et al. (207) defined biosorption as the “accumulation and concentration
of contaminants from aqueous solutions via biological materials to facilitate
recovery and/or acceptable disposal of the target contaminant.” This definition
can be expanded to depict the difference between active and passive biosorption.
Active biosorption entails passage across the cell membrane and participation in
the metabolic cycle. Passive sorption is the entrapment of heavy metal ions in the
cellular structure and subsequent sorption onto the cell binding sites (208). Live
biomass involves both active and passive sorption, whereas inactive cells entail
only the passive mode.
Bioadsorbent materials encompass a broad range of biomass sources,
including cyanobacteria, algae, fungi, bacteria, yeasts, and filamentous microbes.
The biomass can be cultivated specifically for metal sorption, or a waste biomass
can be utilized. Table 5 includes a brief compilation of the microorganisms
studied for the biosorption of heavy metals via passive sorption with dead
biomass. Although the research contained in Table 5 focused on the sorption of
dead biomass, live cells can also be used.
Both live and inactivated (dead) biomass possess interesting metal-binding
capacities due to the high content of functional groups contained in their cell walls
(39,222). A few differences have been exhibited between the use of live and dead
biomass. In general, live biomass can accumulate more metal ions per unit cell
weight. Live cells, if processed correctly, can be reused almost indefinitely. For
instance, live Candida sp. can sorb (per gram of live biomass) 17.23 mg, 10.37
mg, and 3.2 mg of Cd, Cu, and Ni, respectively (212). The sorption capacity was
194 Cutright

TABLE 5 Examples of a Few of the Yeast, Bacteria, Algae, and Fungi Used
for the Biosorption of Heavy Metals from Waste Effluents (metals listed in
order of selectivity)

Species Heavy metal References

Yeast:
Saccharomyces cerevisiae Cu, Cr(VI), Cd, Ni, Zn 209–211
Candida sp. Cd, Cu, Ni 212
Bacteria:
Thiobacillus ferrooxidans Cd, Zn, Cu 213
Bacillus cereus Pb, Mn, Ni, Zn 214
Pseudomonas aeuriginosa Cd, Hg, Cu 200
Cr, Cu, Pb 215
Cu 216
Rhisopus arrhisus Cr(VI) 199
Cr, Cu, Pb 211,217–219
Synechocystis sp. Cu, Ni, Cr 207
Algae:
Chlorella vulgaris Cu, Ni, Cr 207
Scenedesmus obliquus Cu, Ni, Cr 207
Fungi:
Aspergillus niger Pb, Cd, Cu, Ni 208
Bacillus thuringiensis Cd, Hg, Cu 200
Fusarium oxysporum Ni, Mn, Pb, Co, Cu 214
Phanerochaete chrysosporium Cd, Zn, Cu 220,221

reduced by over 35% of each metal when inactive cells were used. Unfortunately,
active sorption with live cells usually takes longer and poses stricter environmen-
tal controls than the use of inactivated biomass (223). Moreover, the use of dead
biomass from pharmaceutical or other industrial operations is a waste minimiza-
tion technique in itself (224). It converts a previously useless waste stream into a
viable process step.
Also, as indicated in Table 5, sorption selectivity and specificity will depend
on the source of biomass used. Research conducted by Kanosh and El-Shafei
(214) showed that fungi had a metal selectivity of Ni >> Mn > Pb > Co > Cu >>
Zn >> Al, whereas bacterial strain selectivity was Pb > Mn > Ni > Zn >> Al > Li
> Cu > Co. Isotherm studies with brewer’s yeast (Saccharomyces cerevisiae )
resulted in a sorption order of Pb > Cu > Cd ≈ Zn > Ni (225). When Candida sp.
is used as the source of live biomass, selectivity order changes for three of the
metals, to Cd > Cu > Ni. The differences in degree of sorption and specificity are
attributed to the inherent differences in cellular metabolism and functional groups
adhered to the cell walls.
Biotechnology Principles 195

It is important to note that the success of metal biosorption has led

researchers to investigate this technique for other waste compounds. For example,
biosorption has also received a cursory investigation for the decolorization of dye
effluents. The yeast strain Kluyveromyces marxianus IMB3 was used to treat
effluents containing azo and diazo dyes. The biomass was able to adsorb over 68
mg dye/mg biomass during the 3-day study (226).

4.3.3 Heavy Metal Recovery

Microbial immobilization of heavy metals is used more for the recovery of
specific metal ions. The ability of the microbes to immobilize the metals is an
extension of its normal survival mechanism. Initially the microbes capable of
immobilization were studied for their ability to transform the compound into a
less toxic form (227). The first studies documented the ability of bacteria to
change the valence state of soluble chromium from 6 to 3. When the experiment
was conducted over a longer period of time, other bacteria were able to facilitate
the precipitation of the metals (228). Precipitation occurred as a result of the
microbial excretion of organic acids, enzymes, and polymeric substance. When
the metals have been precipitated, they no longer pose a threat to the microorgan-
isms and enable easy recovery.
Biosorption can also be used for the recovery of metals. However, the
procedure is often more involved and costly than microbial immobilization.
Recovery from dead biomass entails the digestion of the cellular material,
followed by the separation of the different metal ions by chemical or electrolytic
methods. Washing live cells with different electrolytic and buffer solutions can
cause the elution of the bound metal. Again, separation and purification of the
different metals in the eluted solution can be achieved by traditional chemical or
electrolytic methods.

4.4 Minimization of Biomass

As stated in the activated sludge section, part of the biomass (i.e., sludge) is
recycled and the remainder is wasted. In the past, the wasted biomass was either
used as a nutrient amendment for farm lands, compost supplement, or disposed
of in a landfill. However, depending on the waste being treated, the degradation
by-products and inorganics that have sorbed to the suspended flocs may cause the
wasted biomass to be classified as a hazardous waste. Even when not classified
as hazardous, changing legislation and rising processing and disposal costs have
led to studies that focus on biomass growth to decrease the amount of excess
microbial matter (229,230).
The U.S. Air Force developed a two-step process to decrease the volume
of biomass disposed of. First, the wasted biomass and suspended solids were
solublized via a hot acid hydrolysis step. After cooling, the solublized material
196 Cutright

was recycled through the normal activated sludge treatment area. Utilization of
this process has resulted in a 90% reduction of the hazardous biological solid
waste disposed of in controlled landfills (231).
A second innovation was to eliminate excess sludge production by treating
it in the aeration tank. The simultaneous activated sludge treatment and sludge
reduction was brought about by the addition of ozone. In some facilities, ozone
is used to supply the necessary oxygen to the microbial population. For this
application, care has to be taken to ensure that too much ozone is not added, so
that normal chemical oxidation reactions do not occur. The phenomenon led
engineers to investigate ozonation for the reduction of excess biomass. Control-
ling the ozone dosage to 0.05 g per gram of suspended solids with a recycle ration
of 0.3 reduced the excess sludge production to essentially zero (232). Although
successful, this technique is not fully utilized due to the extensive monitoring
and controls needed to ensure that the required population is kept viable and
not seared.
Other approaches have encompassed increases in process temperatures or
extending the aeration step. Unfortunately, neither avenue is economical or
particularly effective (233). Perhaps a more efficient and controlled approach
would be the direct manipulation of the microbial population. As stated earlier,
bacterial populations will shift and alter their growth based on the available
carbon, energy, and TEA sources. If the metabolic pathway is successfully
uncoupled, substrate (i.e., contaminant) catabolism will continue, while biomass
anabolism is restricted. Most microorganisms must satisfy their maintenance
energy requirements before reproduction, so excess biomass formation will be
reduced (234).
Any easy way to uncouple metabolic pathways and cause a population shift
is to either switch the respiration mode from aerobic to anaerobic or add
chemicals to deter growth rate (235,236). For most activated sludge processes,
switching the respiration mode would be detrimental to the entire process. This
approach is recommended only for the facultative anaerobes found in anaerobic
digestors (237,238). Chemical additions, however, have proven to be quite
successful at minimizing excess biomass formation without jeopardizing the
overall process efficiency.
One study utilized nitrate as the nitrogen source instead of ammonia. This
was a very cost-effective, easily adaptable method that decreased excess biomass
generation by over 70% (239). A more unique means of manipulating biomass
formation was the introduction of nickel to the raw influent. When nickel was
added at a concentration of 0.5 mg/liter, the observed cell yield was reduced by
approximately 65% (240). Unfortunately, a better biomass reduction could not be
achieved. If too high a nickel concentration was implemented, steady state could
not be attained.
Biotechnology Principles 197

4.5 Innovative Bioreactors

Manipulation of reactor design can decrease excess biomass formation, increase
oxygen transfer, decrease shear stress, enhance contact time, enable treatment of
recalcitrant compounds, and improve overall treatment efficiencies (127,229).
The incorporation of baffles to bioreactors is known to enhance mixing and
decrease shear (241). Although effective, the use of baffles is not considered a
new technique. One of the most innovative and versatile developments is the use
of membranes. Membranes can facilitate a multitude of different scenarios, one
of which employs extractive membranes.
In general, extractive membranes are permeable to organics and virtually
impermeable to water and heavy metals, thereby facilitating the biological
treatment of the organics contained in mixed waste (198,242,243). Extractive
membranes can be modified to allow the selective transfer of specific pollutants
into an area where a specialized strain of bacteria will utilize the pollutant as the
sole carbon source. Brookes and Livingston (112) used this approach to demon-
strate the selective degradation of 3,4-dichloroanaline in wastewater. A similar
methodology was used by Inguva et al. (244) to treat TCE and 1,2-dichloroethane
contaminated wastewater. A sequence of such membranes can be employed
in conjunction with the appropriate bio-areas containing different microorgan-
isms for facilitating the treatment of high-molecular-weight compounds and/or
mixed wastes.
Traditional activated sludge system efficiencies have been compared to
those obtained by membrane bioreactors for high-molecular-weight compounds.
The membrane bioreactor removed 99% of the chemicals, while the activated
sludge system treated only 94% (245). Although this may not appear to be a
significant difference, the membrane bioreactor was also highly effective in
degrading the degradation by-products, whereas the activated sludge system was
not. This was due largely to the enhancement of the biomass’ viability by
increasing the oxygen transfer when the membrane was in place (115,246).
Membrane versatility also allows for high-strength wastes to be treated in both
plug flow and completely mixed conditions (247,248).
The successful implementation of membranes is not limited to aerobic
systems. Previously, ultrastrength wastewater was considered treatable only by a
two-stage anaerobic process. Membrane anaerobic reactors enable an effective,
single-phase treatment to be employed (38). Developments with hollow-fiber
membranes have resulted in the simultaneous aerobic-anaerobic treatment of
chlorinated compounds (249,250).
Hollow-fiber membranes have also been used as the support system for
immobilized enzymes. The membranes provide a very large surface area for the
immobilized enzymes. As with other enzyme applications, the reaction times and
subsequent remediation efficiencies are greater than those typically achieved with
198 Cutright

intact cells. This approach has been used for wastewater denitrification, heavy
metal recovery, and treatment of chlorinated compounds (251–255).

5 CONCLUSIONS
As shown by the brief examples presented here, biotechnology has demonstrated
applicability in all areas of waste minimization—material substitution, selective
removal, recycle/reuse of intermediates, end-of-pipe treatment, and source reduc-
tion. These demonstrations have included traditional processes such as activated
sludge and trickling filters to innovative techniques involving heavy metal
biosorption and immobilized enzymes. With continued advances in analytical and
technological expertise, effluent constraints will promulgate to become more
stringent. Biotechnology will continue to rise to the challenge. As scientists and
engineers strive to learn more about microorganisms and their metabolic path-
ways, they will be able to convince Mother Nature to do as they wish. Natural
approaches (i.e., green chemistry) are more effective and acceptable. Further-
more, if complete mineralization of waste effluents is achieved, or successful
biosubstitutions are made, a secondary stream will not have to undergo subse-
quent treatment as with other technologies, thereby eliminating the need for waste
minimization.

6 NOMENCLATURE
av specific surface area per packing piece
A RBC disk area, ft2
As cross-sectional filter area, m2
c contaminant concentration, g/m3
cfu colony-forming unit
e effluent
HBA hydroxybenzoate
HLR hydraulic loading rate
i influent condition
k maximum specific substrate utilization rate, d–1
kd microbial decay coeffient, d–1
K Monod saturation constant, g/m3
m,n emperical constants in Eckenfelder equation
Ns number of stages (RBC units) used
OLR organic loading rate
PAC phenylacetylcarbinol
Q volumetric flow rate, m3/d
r recycle
RBC rotating biological contactor
Biotechnology Principles 199

T water temperature, ˚C
TEA terminal electron acceptor
TF trickling filter
Va volume of the aeration tank, m3
w waste volume
x biomass concentration, g/m3
Y true cell growth yield, g cell produced per g cell removed
z filter depth, m
α recycle ratio
θc mean cell residence time, d

REFERENCES
1. S. Al-Muzaini, Waste Minimization Program in Shuaiba Industrial Area. Water. Sci.
Technol., vol. 39, no. 10–11, pp. 289–295, 1999.
2. J. A. Elterman, R. S. Reimers, L. M. Young, and C. G. Simms, Waste Minimization
Study at Louisiana Oil & Gas Exploration and Production Facilities. Industrial
Wastes Technical Conference, New Orleans, LA, 12/1–12, March 2–5, 1997.
3. T. Basu and M. Chakrabarty. Recent Achievements in the Field of Eco-processing
of Textiles. Colourage, vol. 44, no. 2, pp. 17–23, 1997.
4. Y. Cohen and F. Giralt, Strategies in Pollution Prevention: Waste Minimization and
Source Reduction. AFINIDAD, vol 53, no. 462, pp. 80–92, 1996.
5. G. E. Tong, Integration of Biotechnology to Waste Minimization Programs. In G. S.
Sayler (ed.), Environmental Biotechnology for Waste Treatment, vol. 41, pp. 127–
136, New York: Plenum Press, 1991.
6. D. J. Hardman, M. Huxley, A. T. Bull, H. Slater, and R. Bates, Generation of
Environmentally Enhanced Products: Clean Technology for Paper Chemicals. J.
Chem. Tech. Biotechnol., vol. 70, pp. 60–66, 1997.
7. R. Fringuelli, P. Pellegrino, and F. Pizzo, Synthetic Pathways for a Cleaner Produc-
tion of Organic Substances. Life Chem. Rep., vol. 10, pp. 171–179, 1994.
8. E. D. Battalones and A. Castro. A Feasibility Study on the Biotreatability of
Industrial Wastewaters. SE Asian J. Trop. Med. Public Health, vol. 5, no. 2,
pp. 290–298, 1974.
9. I. Bhattacharya, Role of Biotechnology in Pollution Control. In H. S. Sohal and A.
K. Srivastava (eds.), Environment and Biotechnology, pp. 163–170. New Dehli:
Ashish Publishing, 1994.
10. B. J. D’Arcy, R. B. Todd, and A. W. Wither, Industrial Effluent Control and Waste
Minimization: Case Studies by UK Regulators. Water Sci. Technol., vol. 39,
no. 10–11, pp. 281–287, 1999.
11. A. J. Englande, Jr., and C. F. Guarino, Toxics Management in the Chemical and
Petrochemical Industries, Water Sci. Technol., vol. 26, no. 1–2, pp. 263–274, 1992.
12. A. F. Gorovoi and N. A. Gorovaya, Assessment of the Toxicity of Mining Products
and Wastes from the Processing of Donbass Anthracites. Ugol. Ukr., vol. 12,
pp. 38–39, 1997.
200 Cutright

13. C. D. Litchfield, Practices, Potential, and Pitfalls in the Application of Biotechnol-

ogy to Environmental Problems. In G. S. Sayler (ed.), Environmental Biotechnology
for Waste Treatment, vol. 41, pp. 147–157. New York: Plenum Press, 1991.
14. F. M. Naqvi, Biological Techniques in Environmental Cleanups. Indian J. Environ.
Protect., vol. 13, no. 4, pp. 256–260, 1993.
15. M. Dias and J. D’Souza, Utilization of Photosynthetic Bacteria in Recycling
Domestic Waste. Reg. J. Energy, Heat Mass Transfer, vol. 2, no. 3, pp. 187–191,
1980.
16. J. S. Knapp and P. S. Newby, The Microbiological Decholorization of an Industrial
Effluent Containing a Diazo-Linked Chromophore. Water Res., vol. 29, no. 7,
pp. 1807–1809, 1995.
17. T. L. Hu, Degradation of Azo Dye RP2B by Pseudomonas luteola. Water Sci.
Technol., 38(4–5), pp. 299–306, 1998.
18. P. P. Kanekar, Bioremediation of Chemopollutants from Industrial Effluents. Proc.
Acad. Environ. Biol., vol. 6, no. 1, pp. 1–6, 1997.
19. B. Wang, G. Li, L. Wang, J. Shi, and J. Li, Treatment of Waste Water from a
Dye-Printing Factory in China by a Physicochemical-Biological System. Eur. Water
Manage., vol. 1, no. 1, pp. 25–30, 1998.
20. Y. Wong and J. Yu, Laccase-Catalyzed Decolorization of Synthetic Dyes. Water
Res., vol. 33, no. 16, pp. 3512–3520, 1999.
21. O. Ezeronye and N. Amogy, Microbiological Studies of Effluents from the Nigerian
Fertilizer and Paper Recycling Mill Plants. Int. J. Environ. Studies, vol. 54, no. 3/4,
pp. 213–221, 1998.
22. N. L. Nemerow and F. J. Agardy, Strategies of Industrial & Hazardous Waste
Management. New York: Van Nostrand Reinhold, 1998.
23. C. L. Traviesa and E. F. Benitez, Microalgae Continuous Culture on Pretreated Cane
Sugar Mill Wastes. Leuna-Merseburg, vol. 29, no. 2, pp. 217–221, 1987.
24. D. R. Wilson, I. C. Page, A. A. Cocci, and R. C. Landine, Case History—Two Stage,
Low-Rate Anaerobic Treatment Facility for South American Alcochemical/Citric
Acid Wastewater. Water Sci. Technol., vol. 38, no. 4–5, pp. 45–52, 1998.
25. T. J. Cutright, Y. S. Hun, and S. Lee, Effects of Operating Parameters on the Growth
of Kluveromyces fragilis on Cheese Whey. Chem. Eng. Commun., 137, pp. 47–51,
1995.
26. J. R. Danalewich, T. G. Papagiannis, R. L. Gelyea, M. E. Tumbleson, and L. Raskin,
Characterization of Dairy Waste Streams, Current Treatment Processes and Potential
for Biological Nutrient Removal. Water Res., vol. 32, no. 12, pp. 3555–3568, 1998.
27. W. Dzwolak and S. Ziajka, Enzymatic Hydrolysis of Milk Proteins under Alkaline
and Acidic Conditions. J. Food Sci., vol. 64, no. 3, pp. 393–395, 1999.
28. J. Gonzalez, P. Valdes, G. Nieves, and B. Guerrero, Application of Anaerobic
Digestion to Dairy Industry Wastes. Rev. Int. Contam. Ambiental., vol. 10, no. 1,
pp. 37–41, 1994.
29. E. Papachristou and C. T. Lafazanis, Application of Membrane Technology in the
Pretreatment of Cheese Dairy Waste and Co-treatment in a Municipal Conventional
Biological Unit. Water Sci. Technol., vol. 36, no. 2–3, pp. 361–367, 1997.
30. C. C. Ross, G. E. Valentine, Jr., and J. L. Walsh, Jr., Food-Processing Wastes. Water
Environ. Res., vol. 71, no. 5, pp. 812–815, 1999.
Biotechnology Principles 201

31. T. H. Chen and W. H. Shyu, Chemical Characterization of Anaerobic Digestion

Treatment of Poultry Mortalities. Bioresources Technol., vol. 63, no. 1, pp. 37–48,
1998.
32. B. Rusten, J. G. Siljudalen, A. Wien, and D. Eidem, Biological Pretreatment of
Poultry Processing Wastewater. Water Sci. Technol., vol. 38, no. 4–5, pp. 19–28,
1998.
33. E. A. Salminen and J. A. Rintala, Anaerobic Digestion of Poultry Slaughtering
Wastes. Environ. Technol., vol. 20, no. 1, pp. 21–28, 1999.
34. M. Takemasa, Nutritional Strategies to Reduce Nutrient Waste in Livestock and
Poultry Production. Kagaku, vol. 36, no. 11, pp. 720–726, 1998.
35. B. Kherrati, M. Faid, M. Elyachioui, and A. Wahmane, Process for Recycling
Slaughterhouse Wastes and Byproducts of Fermentation. Bioresources Technol.,
vol. 63, no. 1, pp. 75–79, 1998.
36. C. Rose, T. Sastry, V. Madhavan, and R. Ranganathan, Chemical Modification of
Serum Globulins Recovered from Slaughterghouse Waste and Its Partial Character-
ization. Iran Polymer J., vol. 7, no. 1, pp. 47–52, 1998.
37. N. Garg and Y. Hang, Microbial Production of Organic Acids from Carrot Process-
ing Waste, J. Food Sci. Technol., vol. 32, no. 2, pp. 119–121, 1995.
38. P. M. Sutton, Innovative Biological Systems for Anaerobic Treatment of Grain and
Food Processing Wastewaters. Starch/Starke, vol. 38, no. 9, pp. 314–318, 1986.
39. M. N. Akthar and P. M. Mohan, Bioremediation of Toxic Metals Ions from Polluted
Lake Waters and Industrial Effluents by Fungal Biosorbent. Chem. Sci., vol. 69, no.
12, pp. 1028–1030, 1995.
40. R. Bhagat and S. Srivastava, Biorecovery of Zinc by Pseudomonas stutzeri RS34.
Biohydrometallics Technol., vol. 2, pp. 209–217, 1993.
41. M. Chen and H. J. Oliver, Microbial Chromium (VI) Reduction. Environ. Sci.
Technol., vol. 28, no. 3, pp. 219–251, 1998.
42. D. Couillard, M. Chartier, and G. Mercier, Major Factors Influencing Bacterial
Leaching of Heavy Metals (Cu and Zn) from Anaerobic Sludge. Environ. Pollut.,
vol. 85, no. 2, pp. 175–184, 1994.
43. L. Fude, B. Harris, M. Urrutia, and T. Beveridge, Reduction of Cr(VI) by a
Consortium of Sulfate-Reducing Bacteria (SRB III). Appl. Environ. Microbiol., vol.
60, no. 5, pp. 1525–1531, 1994.
44. R. Glombitza, U. Iske, M. Bullmann, and J. Ondruschka, Biotechnology Based
Opportunities for Environmental Protection in the Uranium Mining Industry. Acta
Biotechnol., vol. 12, no. 2, pp. 79–85, 1992.
45. D. P. Smith and R. Kalch, Minerals and Mine Drainage. Water Environ. Res.,
vol. 71, no. 5, pp. 822–827, 1999.
46. S. Abou-Elela and R. Zaher, Pollution Prevention in the Oil & Soap Industry: A Case
Study. Water Sci. Technol., vol. 38, no. 4–5, pp. 139–144, 1998.
47. R. Bentham, N. McClure, and D. Catcheside, Biotreatment of an Industrial Waste
Oil Condensate. Water Sci. Technol., vol. 36, no. 10, pp. 125–1259, 1997.
48. K. E. Richardson, Refinery Waste Minimization. Hazardous Material Management
Conference 12, pp. 505–513, 1995.
49. J. M. Wong, Petrochemicals. Water Environ. Res., vol. 71, no. 5, pp. 828–832, 1999.
202 Cutright

50. M. Mascarenhas, Novel Uses of Microorganisms in the Paint Industry: A Review.

Paintindia, vol. 49, no. 5, pp. 35–40, 1999.
51. N. V. Men’shutina, A. I. Shamber, V. Men’shikov, and R. Salakhetdinov, Installa-
tions for Treatment of Wastewaters and Waste Gases from Paint and Varnish
Industries. Lakokras Mater. Ikh. Primen., vol. 5, pp. 24–27, 1998.
52. T. S. Webster, A. P. Togna, W. J. Guarinik, and L. McKnight, Application of a
Biological Trickling Filter Reactor to Treat Volatile Organic Compound Emissions
from a Spray Paint Booth Operation. Metal Finishing, vol. 97, no. 3, pp. 24–26,
1999.
53. A. Yasmin, S. Afrasayab, and S. Hasnain, Mercury Resistant Bacteria from Effluents
of Paint Factory: Characterization and Mercury Uptake Ability. Sci. Int., vol. 9,
no. 3, pp. 315–320, 1997.
54. C. Elvira, L. Sampedro, E. Genitez, and R. Nogales, Vermicomposting of Sludges
from Paper Mill and Dairy Industries with Eisenia andrei: A Pilot Scale Study.
Bioresources Technol., vol. 63, no. 3, pp. 205–211, 1998.
55. K. A. Kahmark and J. P. Unwin, Pulp & Paper Management. Water Environ. Res.,
vol. 71, no. 5, pp. 836–852, 1999.
56. U. Kaluza, P. Klingelhofer, and K. Taeger, Microbial Degradation of EDTA in an
Industrial Wastewater Treatment Plant. Water Res., vol. 32, no. 9, pp. 2843–2845,
1998.
57. G. Rozmarin and V. Gazdaru, Connections between Biotechnology—Pulp and Paper
Industry: Recent Applications, Possible Options. Celul. Hirtie, vol. 43, no. 1,
pp. 29–37, 1994.
58. D. E. Mullins, S. E. Gabbert, J. E. Leland, R. W. Young, G. H. Hetzel, and D. R.
Berry, Organic Sorption/Biodegradation of Pesticides. Rev. Toxicol., vol. 2, no. 1–4,
pp. 195–201, 1998.
59. H. Poggi-Varaldo, Agricultural Wastes. Water Environ. Res., vol. 71, no. 5, pp. 737–
785, 1999.
60. X. Shichong, Research on Treating Method of Pesticide Wastewater. Water Treat-
ment, vol. 6, no. 3, pp. 343–350, 1991.
61. P. Babu, M. S. Kumar, D. G. Reddy, P. M. Kumar, and A. K. Sanhukhan,
Biodegradation of Bulk Drug Industrial Effluents by Microbial Isolates from Soil.
J. Sci. Ind. Res., vol. 58, no. 6, pp. 431–435, 1999.
62. B. Gulmez, I. Ozturk, K. Alp, and O. Arikan, Common Anaerobic Treatability of
Pharmaceutical and Yeast Industry Wastewater. Water Sci. Technol., vol. 38, no. 4–
5, pp. 37–44, 1998.
63. I. Kabdasli, M. Gurel, and O. Tunay, Pollution Prevention and Waste Treatment in
Chemical Synthesis Processes for Pharmaceutical Industry. Water Sci. Technol., vol.
39, no. 10–11, pp. 265–271, 1999.
64. B. Ruggerri and V. Specchia, Waste Control in Food and Pharmaceutical Industries.
Kem. Ind., vol. 39, no. 12, pp. 579–597, 1990.
65. H. F. Schroder, Substance Specific Detection and Pursuit of Non-eliminable Com-
pounds During Biological Treatment of Wastewater from the Pharmaceutical Indus-
try. Waste Manage., vol. 19, no. 2, pp. 111–123, 1999.
66. W. Yin, Use of Clean Processes in Industry. Huanjing Baohu, vol. 12, pp. 15–17,
1993.
Biotechnology Principles 203

67. H. Herlitzius, Solvent Preparation with Biological Trickling Filters in a Printing Ink
Factory. VDI-Ber., vol. 1034, pp. 564–576, 1993.
68. C. Kellner and E. Vitzthum. Biological Waste Gas Cleaning in a Biological Trickling
Filter. Conversion from a Pilot-Plant Scale to Industrial Scale with Toluene-Loaded
Waste Gas at Printing Ink Producer. In W. Prins and J. Van Ham. (eds.), Biological
Waste Gas Clean, pp. 6–66. Stuttgart: VDI Verlag, 1997.
69. K. Lerche, W. Kneist, H. Rohbeck, B. Hillemann, and R. Schwarz, Biological
Treatment of Surfactant Loaded Wastewaters, Such as Laundry Effluents. German
Patent 950420, 1994.
70. E. Pfleiderer and R. Reiner, Microorganisms in Processing of Leather. Biotechnol-
ogy, vol. 6, pp. 729–743, 1988.
71. G. Ummarino, R. Mora, and A. Russo, Tannery Emissions: Possibilities of Biolog-
ical Treatment. Mater. Concianti, vol. 68, no. 2, pp. 207–215, 1992.
72. D. P. Bakshi, K. G. Gupta, and P. Sharma, Enchanced Biodecolorization of Synthetic
Textile Dye Effluent by Phaneorchaete chrysosporium Under Improved Culture
Conditions. World J. Microbiol. Biotechnol., vol. 15, pp. 507–509, 1999.
73. P. Nigam, I. M. Banat, D. Oxspring, R. Marchant, D. Singh, and W. Smyth, A New
Facultative Anaerobic Filamentous Fungi Capable of Growth on Recalcitrant
Textile Dyes as Sole Carbon Source. Microbios, vol. 84, no. 340, pp. 171–185,
1995.
74. R. Ul Haq and A. R. Shakoori, Short Communication: Microbiological Treatment
of Industrial Wastes Containing Toxic Chromium Involving Successive Use of
Bacteria, Yeast & Algae. World J. Microbiol. Biotechnol., vol. 14, no. 4, pp. 583–
585, 1998.
75. M. G. Roig, M. J. Martin-Rodrogiuez, J. M. Cachaza, S. L. Mendoza, and J. F.
Kennedy, Principles of Biotechnological Treatment of Industrial Wastes. Crit. Rev.
in Biotechnol., vol. 13, no. 2, pp. 99–116, 1993.
76. K. H. Baker and D. S. Herson, Bioremediation. New York: McGraw-Hill, 1994.
77. J. R. Cookson, Jr., Bioremediation Engineering: Design & Application. New York:
McGraw-Hill, 1995.
78. R. D. Norris, R. E. Hinchee, R. Brown, P. L. McCarty, J. T. Wilson, M. Reinhard,
E. J. Bouwer, R. C. Borden, and T. M. Vogel, Handbook Bioremediation. Boca
Raton, Florida: Lewis, 1994.
79. A. C. Palmisano, D. A. Maruscki, G. J. Ritchie, B. S. Schwab, S. R. Harper, and
R. A. Rapapoprt, A Novel Bioreactor Simulating Composting of Municipal Solid
Waste. J. Microbiol. Meth. vol. 18, no. 2, pp. 99–112, 193.
80. R. Bellandi (ed.), Innovative Engineering Technologies for Hazardous Waste Reme-
diation (O’Brian & Gere Engineers, Inc.). Boston: Van Nostrand Reinhold, 1995.
81. E. Riser-Roberts, Remediation of Petroleum Contaminated Soils. Boca Raton,
Florida: Lewis, 1998.
82. J. B. Eweis, S. J. Ergas, D. P. Y. Chang, and E. D. Schroeder, Bioremediation
Principles. New York: McGraw-Hill, 1998.
83. V. Lazarova, D. Bellahcen, D. Rybacki, B. Rittmann, and J. Manem, Population
Dynamics and Biofilm Composition in a New Three-Phase Circulating Bed Reactor.
Water Sci. Technol., vol. 37, no. 4–5, pp. 149–158, 1998.
84. M. Van Loosdrecht, L. Tijhuis, A. Wijdieks, and J. Heijnen, Population Distribution
204 Cutright

in Aerobic Biofilms on Small Suspended Particles. Water Sci. Technol., vol. 31,
no. 1, pp. 163–171, 1995.
85. R. Hedderich, O. Klimmek, A. Kroger, R. Dirmeier, M. Keller, and K. Stetter,
Anaerobic Respiration with Elemental Sulfur and with Disulfides. FEMS Microbiol.
Rev., vol. 22, no. 5, pp. 353–381, 1998.
86. J. F. Stolz and R. S. Oremland, Bacterial Respiration of Arsenic and Selenium.
FEMS Microbiol. Rev., vol. 23, no. 5, pp. 615–627, 1999.
87. S. Zala, A. Nerurkar, A. Desai, J. Ayyer, and V. Akolkar, Biotreatment of Nitrate
Rich Industrial Effluent by Suspended Bacterial Growth. Biotechnol. Lett., vol. 21,
no. 6, pp. 481–485, 1999.
88. M. T. Madigan, J. M. Martinko, and J. Parker, Brock’s Biology of Microorganisms,
8th ed., Upper Saddle River, New Jersey: Prentice Hall, 1997.
89. S. S. Sutherson, Remediation Engineering: Design Concepts. Boca Raton, Florida:
Lewis, 1997.
90. E. Ward-Liebig and T. J. Cutright, The Investigation of Enhanced Bioremediation
Through the Addition of Macro and Micro Nutrients in a PAH Contaminated Soil.
Int. Biodeter. Biodegrad., vol. 44, no. 1, pp. 55–64, 1999.
91. J. E. Bailey and D. F. Ollis, Biochemical Engineering Fundamentals, 2nd ed. New
York: McGraw-Hill, 1986.
92. J. E. Burgess, J. Quarmby, and T. Stephenson, Role of Micronutrients in Activated
Sludge-Based Biotreatment of Industrial Effluents. Biotechnol. Adv., vol. 17, pp.
49–70, 1999.
93. J. R. Boulding (ed.), EPA Environmental Engineering Source Book. Ann Arbor, MI:
Ann Arbor Press, 1996.
94. D. R. Schneider and R. J. Billingsley, Bioremediation: A Desk Manual for the
Environmental Professional. Des Plaines, Iowa: Cahners, 1990.
95. O. M. El-Tayeb, S. Megahed, and M. El-Azizi, Microbial Degradation of Aromatic
Substances by Local Bacterial Isolates. III. Factors Affecting Degradation of
2,4-Dichlorophenoxyacetic Acid by Pseudomonas stutzeri. Egypt J. Biotechnol.,
vol. 4, pp. 84–90, 1998.
96. B. R. Patel, Nutrient Control in Pulp & Paper Wastewater Treatment Plants Using
Online Measurements. Environ. Conf. Exhib. 1, pp. 79–81. Atlanta, GA: TAPPI
Press, 1997.
97. K. Kouno, H. P. Lukito, and T. Ando, Minimum Available N Requirement for
Microbial Biomass Formation in a Regosol. Soil Biol. Biochem., vol. 31, no. 6,
pp. 797–802, 1999.
98. C. Chien, E. Leadbetter, and W. Godchaux, III, Rhodococcus spp. Utilize Taurine
(2-Aminoethanesulfonate) as Sole Source of Carbon, Energy, Nitrogen, and Sulfur
for Aerobic Respiratory Growth. FEMS Microbiol. Lett., vol. 176, no. 2, pp. 333–
337, 1999.
99. C. Knoblauch, K. Sahm, and B. Jorgensen, Psychrophilic Sulfate-Reducing Bacteria
Isolated from Permanently Cold Arctic Marine Sediments: Description of De-
sulfofrigus oceanense gen nov. sp., nobv., Desulfofrigus fragile sp. nov., D. gelida
gen. nov., D. psychrophilia gen. nov. and D. arctica sp. nov. Int. J. Syst. Bacteriol.,
vol. 49, no. 4, pp. 1631–1643, 1999.
100. G. Bharadwaj and R. Maheshwari, A Comparison of Thermal Characteristics and
Biotechnology Principles 205

Kinetic Parameters of Trehalases from a Thermophilic and a Mesophilic Fungus.

FEMS Microbiol. Lett., vol. 181, no. 1, pp. 187–193, 1999.
101. L. McKane and J. Kandel, Microbiology: Essentials & Applications, 2nd ed. New
York: McGraw-Hill, 1996.
102. J. Simpa, P. Lens, A. Vieira, Y. Miron, J. B. van Lier, L. W. Hulshoff-Pol, and G.
Lettinga, Thermophilic Sulfate Reduction in Upflow Anaerobic Sludge Bed Reac-
tors Under Acidifying Conditions. Process Biochem., vol. 35, pp. 509–522, 1999.
103. J. Singh, R. M. Vohra, and D. K. Sahoo, Alkaline Protease from a New Obligate
Alkalophilic Isolate of Bacillus sphaericus. Biotechnol. Lett., vol. 21, no. 10,
pp. 921–924, 1999.
104. P. L. Bishop and Y. Tong, A Microelectric Study of Redox Potential Change in
Biofilms. Water Sci. Technol., vol. 39, no. 7, pp. 179–185, 1999.
105. M. Nay, M. Snozzi, and J. Zehnder, Fate and Behavior of Organic Compounds in
an Artificial Saturated Subsoil Under Controlled Redox Conditions: The Sequential
Soil Column System. Biodegradation, vol. 10, no. 1, pp. 75–82, 1999.
106. R. L. Droste, Theory and Practice of Water and Wastewater Treatment. New York:
Wiley, 1997.
107. G. Kiely, Environmental Engineering. New York: McGraw-Hill, 1997.
108. P. N. Cheremisinoff, Biotechnology: Treating Industrial/Municipal Wastes and
Wastewater. Pollut. Eng., vol. 19, no. 9, pp. 74–87, 1987.
109. J. M. Wyatt, Biotechnological Treatment of Industrial Wastewater. Microbiol. Sci.,
vol. 5, no. 6, pp. 186–190, 1988.
110. H. Chua, P. H. F. Yu, S. N. Sin, and M. W. L. Cheung, Sub-lethal Effects of Heavy
Metals on Activated Sludge Microorganisms. Chemosphere, vol. 39, no. 15, pp.
2681–2692, 1999.
111. S. Baccella, G. Cerichelli, M. Chiarini, C. Ercole, E. Fantauzzi, A. Lepidi, L. Toro,
and F. Veglio, Biological Treatment of Alkaline Industrial Wastes. Process Biochem.,
vol. 35, no. 2, pp. 595–602, 2000.
112. P. R. Brookes and A. G. Livingston, Biotreatment of a Point-Source Industrial
Wastewater Arising in 3,4-Dichloroaniline Manufacture Using an Extractive Mem-
brane Bioreactor. Biotechnol. Prog., vol. 10, no. 1, pp. 65–73, 1994.
113. W. Viessman, Jr., and M. J. Hammer, Water Supply and Pollution Control, 5th ed.
New York: Harper Collins, 1993.
114. T. F. Yen, Environmental Chemistry: Essentials of Chemistry for Engineering
Processes. New Jersey: Prentice Hall, 1999.
115. M. W. Fitch, J. B. Murphy, and S. S. Sowell, Biological Fixed-Film Systems. Water
Environ. Res., vol. 71, no. 5, pp. 638–655, 1999.
116. T. D. Reynolds and P. A. Richards, Unit Operations & Processes in Environmental
Engineering, 2nd ed. New York: PWS, 1996.
117. W. W. Eckenfelder, Jr., Water Quality Engineering for Practicing Engineers. New
York: Barnes & Noble, 1970.
118. G. M. Masters, Introduction to Environmental Engineering & Science, 2nd ed.
Upper Saddle River, New Jersey: Prentice Hall, 1998.
119. W. W. Eckenfelder, Jr., Y. Argaman, and E. Miller, Process Selection Criteria for the
Biological Treatment of Industrial Wastewaters. Environ. Prog., vol. 8, no. 1,
pp. 40–45, 1989.
206 Cutright

120. W. W. Eckenfelder, Jr., and A. J. Englande, Jr., Innovative Biological Treatment for
Sustainable Development in the Chemical Industries. Water Sci. Technol., vol. 38,
no. 4–5, pp. 111–120, 1998.
121. P. A. Vesilind, Introduction to Environmental Engineering. Boston: PWS, 1997.
122. C. D. Cooper and F. C. Alley, Air Pollution Control: A Design Approach, 2nd ed.
Boston: Waveland Press, 1994.
123. M. S. McGrath, J. Nieuwland, and C. van Lith, Case Study: Biofiltration of Styrene
and Butylacetate at a Dashboard Manufacturer. Environ. Prog., vol. 18, no. 3:197–
204, 1999
124. C. Colella, M. Pansini, F. Alfani, M. Cantarella, and A. Gallifuoco, Selective Water
Adsorption from Aqueous Ethanol-Containing Vapors by Phillipsite-Rich Volcanic
Tufts. Microporous Mater., vol. 3, no. 3, pp. 219–226, 1994.
125. N. De Nevers, Air Pollution Control Engineering, 2nd ed. New York: McGraw-Hill,
2000.
126. J. Joyce and H. Sorensen, Bioscrubber Design. Water Environ. Technol., vol. 11,
no. 2, pp. 37–41, 1999.
127. D. Johnson, M. J. Semmens, J. S. Gulliver, A Rotating Membrane Contactor:
Application to Biologically Active Systems. Water Environ. Res., vol. 71, no. 2,
pp. 163–168, 1999.
128. T. S. Webster, H. Cox, and M. A. Deshusses, Resolving Operational Problems
Encountered in the Use of a Pilot/Full-Scale Trickling Filter Reactor. Environ.
Prog., vol. 18, no. 3, pp. 162–172, 1999.
129. J. W. Van Groenestijn and M. E. Lake, Elimination of Alkanes from Off-gases Using
Biotrickling Filters Containing Two Liquid Phases. Environ. Prog., vol. 18, no.
3:151–155, 1999.
130. F. G. Edwards and N. Nirmalakhandan, Modeling an Airlift Bioscrubber for
Removal of Air Phase BTEX. J. Environ. Eng., vol. 125, no. 11, pp. 1062–1070,
1999.
131. M. C. Veiga, M. Fraga, L. Amor, and C. Kennes, Biofilter Performance and
Characterization of a Biocatalyst Degrading Alkylbenzene Gases. Biodegradation,
vol. 19, no. 3, pp. 169–176, 1999.
132. D. Arulneyam and T. Swaminathan, Biodegradation of Ethanol Vapor in a Biofilter.
Bioprocess Eng., vol. 22, no. 1, pp. 63–67, 2000.
133. B. Hodkinson, J. B. Williams, and J. E. Butler, Development of Biological Aerated
Filters: A Review. Water Environ. Manage., vol. 13, no. 4, pp. 250–254, 1999.
134. F. Wittorf, S. Knauf, and H. E. Windberg, Biotrickling-Reactor: A New Design for
the Efficient Purification of Waste Gases. Proc. 4th Int. Biological Waste Gas
Cleaning, vol. 4, pp. 329–335. Stuttgart, VDI Verlag, 1997.
135. W. Schiettecatte, W. Leys, R. Gerards, J. Koning, H. Verachtert, and J. F. Van Impe,
Evaluation of a Biofiltration Process: Purification of a BTEX Contaminated Air-
flow. Toegepaste Biol. Wet., vol. 64, no. 5a, 197–200, 1999.
136. N. Fortin and M. A. Deshusses, Treatment of Methyl tert-Butyl Ether Vapors in
Biotrickling filters 2. Analysis of Rate-Limiting Step and Behavior Under Transient
Condtions. Environ. Sci. Technol., vol. 33, no. 17, pp. 2987–2991, 1999.
137. H. Cox and M. A. Deshusses, Chemical Removal of Biomass from Waste Air
Biotechnology Principles 207

Biotrickling Filters: Screening of Chemicals of Potential Interest. Water Res., vol.

33, no. 10, pp. 2383–2391, 1999.
138. P. T. Anastas, L. B. Bartlett, M. M. Kirchhoff, and T. C. Williamson, The Role of
Catalysts in the Design, Development, and Implementation of Green Chemistry.
Catalysis Today, vol. 55, no. 1, pp. 11–22, 2000.
139. Y. H. Huang and G. S. Hao, Green Chemistry and New Technology for Zero
Pollutant Discharge. Ziran Kexueban, vol. 22, no. 3, pp. 348–350, 1999.
140. H. S. Shin and P. L. Rogers, Production of L-Phenylacetylcarbinol from Benzalde-
hyde Using Partially Purified Pyruvate Decarboxylase (PDC). Biotechnol. Bioeng.,
vol. 49, no. 1, pp. 52–62, 1996.
141. C. M. Tripathi, S. C. Agarwal, and S. K. Basu, Production of L-Phenylacetylcarbinol
by Fermentation. J. Ferment. Bioeng., vol. 84, no. 6, pp. 487–492, 1997.
142. A. L. Oliver, F. A. Roddick, and B. N. Anderson, Cleaner Production of Phenyl-
acetylcarbinol by Yeast Through Productivity Improvements and Waste Minimiza-
tion. Pure Appl. Chem., vol. 69, no. 11, pp. 2371–2385, 1997.
143. O. Popa, A. Ionita, C. Popa, I. Paraschiv, and A. Vamanu, Studies Regarding L-PAC
Formation by Yeast Bioconversion. Biotechnol. Lett., vol. 3, no. 2, pp. 115–122,
1998.
144. Y. Yao and J. Li, Production of L-Phenylacetylcarbinol by Pyruvate Decarboxylase.
Zhongguo Yaoke Daxue Xuebao, vol. 30, no. 2, pp. 115–118, 1999.
145. J. Gao, L. Cheng, and H. He, Study of Kinetics of Solvent Carboxylation of Phenol.
Dalian Ligong Daxue Xuebao, vol. 36, no. 3, pp. 288–292, 1996.
146. Y. Kharma, G. Sergeev, Y. Gordash, V. Bludilin, and L. Lekhter, Modeling of
Equilibria in the Kolbe-Schmitt Carboxylation of Higher Alkylphenols. Neftepererab.
Neftekhim., vol. 38, pp. 47–49, 1990.
147. E. S. Miller, Jr., and S. W. Peretti, Bioconversion of Toluene to p-Hydroxybenzoate.
Green Chem., vol. 1, no. 3, pp. 143–152, 1999.
148. P. L. Bishop, Pollution Prevention: Fundamentals and Practice. New York: McGraw-
Hill, 2000.
149. F. Huang, Present Status and Development of Biodegradable Polymers. Huaxue
Shijie, vol. 40, no. 11, pp. 570–574, 1999.
150. Y. Poirier, Green Chemistry Yields a Better Plastic. Nat. Biotechnol., vol. 17, no.
10, pp. 960–961, 1999.
151. C. Nawrath, Y. Poirier, and C. Somerville, Targeting of the Polyhydroxybutyrate
Biosynthetic Pathway to the Plastids of Arabidppsis thaliana Results in High Levels
of Polymer Accumulation. Proc. Natl. Acad. of Sci. USA, vol. 91, no. 26, pp. 12760–
12764, 1994.
152. M. Matavulj, D. Radnovic, S. Gajin, O. Petrovic, Z. Svircev, I. Tamas, M. Bokorov,
V. Divjakovic, F. Gassner, and H. P. Molitoris, Microorganisms, Producers and
Degraders of Biosynthetic Plastic Materials. Mikrobiologija, vol. 32, no. 1, pp.
169–178, 1995.
153. J. Woodard and B. R. Evans, Utilization of Biocatalysts in Cellulose Waste Minimi-
zation. Biotechnol. Res., vol. 7, pp. 157–179, 1998.
154. U. J. Haenggi, Requirements on Bacterial Polyesters as Future Substitute for
Conventional Plastics for Consumer Goods. FEMS Microbiol. Rev., vol. 16, no. 2–3,
pp. 213–220, 1995.
208 Cutright

155. C. Chu, A. Lu, M. Liszkowski, and R. Sipehia, Enhanced Growth of Animal and
Human Endothelial Cells on Biodegradable Polymers. Biochim. Biophys. Acta.,
vol. 1472, no. 3, pp. 479–485, 1999.
156. A. De la Maza, L. Codech, O. Lopez, J. L. Parra, M. Sabes, and J. Guinea,
Biopolymer Excreted by Psuedoalteromonas antartica NF3, as a Coating and
Protective Agent of Liposomes Against Dodecyl Maltoside. Biopolymers, vol. 59,
no. 6, pp. 579–588, 1999.
157. J. Fu and S. Li, Biodegradable Polymers Used in the Biomedical Field (2). Wuhan
Gongye Daxue Xuebao, vol. 21, no. 5, pp. 19–22, 1999.
158. M. Schulz, T. Blunk, and A. Gopfercih, Carrier for Drug Formulations and Artificial
Living Tissues. Pharm. Z., vol. 144, no. 45, pp. 3661–3668, 1999.
159. P. Biely and L. Kremnicky, Yeasts and Their Enzyme Systems Degrading Cellulose,
Hemicelluloses, and Pectin. Food Technol. Biotechnol., vol. 36, no. 4, pp. 305–312,
1998.
160. M. Itavaara, M. Siika-Aho, and L. Viikari, Enzymatic Degradation of Cellulose
Based Materials. J. Envrion. Polymer Degradation, vol. 7, no. 2, pp. 67–73, 1999.
161. J. M. Krochta and L. C. De Mulder-Johnston, Biodegradable Polymers from
Agricultural Products. ACS Symp. Ser., vol. 64, no. 7, pp. 120–140, 1996.
162. D. B. Wilson and D. C. Irwin, Genetics and Properties of Cellulases. Adv. Biochem.
Eng. Biotechnol., vol. 65, pp. 1–21, 1999.
163. C. Nawrath, Y. Poirier, and C. Somerville. Plant Polymers for Biodegradable
Plastics: Cellulose, Starch, and Polyhdroxyalkanoates. Mol. Breed., vol. 1, no. 2,
pp. 105–122, 1995.
164. Y. Poirier, C. Nawrath, and C. Somerville, Production of Polyhdroxyalkanoates, a
Family of Biodegradable Plastics and Elastomers in Bacteria and Plants. Bio/Tech-
nology, vol. 13, no. 2, pp. 142–150, 1995.
165. T. Wang, L. Ye, and Y. Song, Progress of PHA Production in Transgenic Plants.
Chin. Sci. Bull., vol. 44, no. 19, pp. 1729–1736, 1999.
166. K. M. Elborough, A. J. White, S. Z. Hanley, and A. R. Slabas, Production of the
Biodegradable Plastic Polyhydroxyalkanoates in Plants. Portland Press Proc.,
vol. 14, pp. 125–131, 1998.
167. H. Miyasaka, H. Nakano, H. Akiyama, S. Kanai, and M. Hirano, Production of PHA
by Genetically Engineered Marine Cyanobacterium. Studies Surface Sci. Catalysis,
vol. 114, pp. 237–242, 1998.
168. R. J. Van Wegen, Y. Ling, and APJ Middelberg, Industrial Production of Poly-
hydroxyalkanoates Using Escherichia coli: An Economic Analysis. Chem. Eng.
Res. Des., vol. 76, no. A3, pp. 417–426, 1998.
169. S. Y. Lee and J. Choi, Production and Degradation of PHAs in Waste Environment.
Waste Manage., vol. 19, pp. 133–139, 1999.
170. G. Braunegg, G. Lefebvre, and K. Genser, Polyhydroxyalkanoates Biopolyesters
from Renewable Resources: Physiological and Engineering Aspects. J. Biotechnol.,
vol. 65, no. 2–3, pp. 127–161, 1998.
171. K. Sakai, T. Yamauchi, F. Nakasu, and T. Ohe, Biodegradation of Cellulose Acetate
by Neisseria sicca. Biosci. Biotechnol. Biochem., vol. 60, no. 10, pp. 1617–1622,
1996.
172. L. A. Vanderberg, T. M. Foreman, M. Attrep, Jr., J. R. Brainard, and N. N. Sauer,
Biotechnology Principles 209

Treatment of Heterogenous Mixed Wastes: Enzyme Degradation of Cellulosic

Materials Contaminated with Hazardous Organics, Toxic and Radioactive Metals.
Environ. Sci. Technol., vol. 33, no. 8, pp. 1256–1262, 1999.
173. R. Mutitakul, S. Kulpreecha, and S. Shioya, Production of PHB, a Biodegradable
Polymer form Bacillus sp. BA-019. Biotechnol. Sustainable Util. Biol. Resource
Trop., vol. 12, pp. 407–413, 1998.
174. G. D. Reyes, R. S. So, and M. M. Ulep, Isolation, Screening and Identification of
Bacteria for poly-β-Hydroxybutyrate (PHB) Production. Study Environ. Sci., vol.
66, pp. 737–748, 1997.
175. M. I. Foda and M. Lopez-Leiva, Continuous Production of Oligosaccharides from
Whey Using a Membrane Reactor. Process Biochem., vol. 35, pp. 581–587, 2000.
176. T. J. Cutright, The Production of α-Lactase. M. S. thesis, The University of Akron,
Akron, OH, 1992.
177. M. M. Saad, A. N. Saad, M. Abdel-Hadi, I. Ghany, and H. Ismail, Production and
Some Properties of Proteinase by Fungi Utilizing Whey. Egypt J. Microbiol., vol.
32, no. 3, pp. 411–421, 1997.
178. A. M. Fialho, L. Martins, M. Donval, J. Leitao, M. Ridout, A. Jay, V. Morris, and
I. Sa-Correia, Structures and Properties of Gellan Polymers Produced by Sphingo-
monas paucimobilis ATCC 31461 from Lactose Compared with Those Produced
from Glucose and from Cheese Whey. Appl. Environ. Microbiol., vol. 65, no. 6,
pp. 2485–2491, 1999
179. R. T. Otto, H. J. Daniel, G. Pekin, K. Muller-Decker, G. Furstenberger, M. Reuss,
and C. Syldatk. Production of Sophorolipids from Whey. II Whey Composition,
Surface Active Properties, Cytotoxicity, and Stability Against Hydrolases by Enzy-
matic Treatment. Appl. Microbiol. Biotechnol., vol. 52, no. 4, pp. 495–501, 1999.
180. J. S. Dordick, Enzymatic Catalysis in Monophasic Organic Solvents. Enzyme
Microb. Technol., vol. 11, pp. 194–211, 1989.
181. P. Wang, M. S. Sergueeva, L. Lim, and J. S. Dordick, Biocatalytic Plastics as Active
and Stable Materials for Biotransformations. Nature Biotechnol., vol. 15, pp.
789–793, 1997.
182. P. Wang, C. A. Woodard, and E. N. Kaufman, Poly(ethylene glycol)-Modified
Ligninase Enhances Pentachlorophenol Biodegradation in Water Solvent Mixtures.
Biotechnol. Bioeng., vol. 64, no. 3, pp. 290–297, 1999.
183. A. D’ Annibale, S. R. Stazi, V. Vinciguerra, E. Di Mattia, and G. G. Sermanni,
Characterization of Immobilized Lacasse from Lentinula edodes and Its Use in
Olive-Mill Wastewater Treatment. Process Biochem., vol. 34, pp. 697–706, 1999.
184. F. Alfani, M. Cantarella, A. Gallifuoco, and V. Romano, On the Effectiveness of
Immobilized Enzymes with Linear Mixed-Type Product Inhibition Kinetics. Chem.
Eng. J., vol. 57, no. 1, pp. B23–B29, 1995.
185. T. Nishida, Y. Tsutsumi, M. Kemi, T. Haneda, and H. Okamura, Decolorization of
Anthraquinone Dyes by White Rot Fungi and Its Related Enzymes. Mizu Kankyo
Gakkaishi, vol. 22, no. 6, pp. 465–4714, 1999.
186. S. K. Garg and D. R. Modi, Decolorization of Pulp-Paper Mill Effluents by
White-Rot Fungi. Crit. Rev. Biotechnol., vol. 19, no. 2, pp. 85–112, 1999.
187. C. J. Jaspers, G. Jimenez, and M. J. Penninck, Evidence for a Role of Manganese
Peroxidase in the Decolorization of Kraft Pulp Bleach Plant Effluent by P. chryso-
210 Cutright

sporium: Effects of Initial Culture Conditions on Enzyme Production. J. Biotechnol.,

vol. 37, no. 3, pp. 229–234, 1994.
188. P. Peralta-Zambor, S. Gomes de Moraes, E. Esposito, R. Antunes, J. Reyes, and
N. Duran, Decolorization of Pulp Mill Effluents with Immobilized Lignin and
Manganese Peroxidases from Phanerochaete chrysosporium. Environ. Technol.,
vol. 19, no. 5, pp. 521–528, 1998.
189. T. Zhang, S. Wada, T. Yamagishi, I. Hiroyasu, K. Tatsumi, and Q. X. Zhao,
Treatment of Bleaching Wastewater from Pulp-Paper Plants in China Using En-
zymes and Coagulants. J. Environ. Sci. (China), vol. 11, no. 4, pp. 480–484, 1999.
190. N. Caza, J. K. Bewtra, N. Biswas, and K. E. Taylor, Removal of Phenolic
Compounds from Synthetic Wastewater Using Soybean Peroxidase. Water Res., vol.
33, no. 13, pp. 3012–3018, 1999.
191. J. C. Rozzell, Commercial Scale Biocatalysis: Myths and Realities. Bioorg Med
Chem., vol. 7, no. 10, pp. 2253–2261, 1999.
192. K. J. Grice, G. F. Payne, and J. S. Karns, Enzymatic Approach to Waste Minimiza-
tion in a Cattle Dipping Operation: Economic Analysis. J. Agric. Food Chem., vol.
44, no. 1, pp. 351–357, 1996.
193. W. Q. Sun and G. F. Payne, Tyrosine-Containing Chitosan Gels: A Combined
Catalyst and Sorbent for Selective Phenol Removal. Biotechnol. Bioeng., vol. 51,
no. 1, pp. 79–86, 1996.
194. T. Zhang, Q. X. Zhao, S. Wilson, and M. Qi, Study of the Removal of 4-Chlorophe-
nol from Wastewater Using Horseradish Peroxidase. Toxicol. Environ. Chem., vol.
71, no. 1–2, pp. 115–123, 1999.
195. G. Zhang and J. A. Nicell, The Characteristics of Ezymatic Treatment of Pentachlo-
rophenol in Wastewater. Chin. Sci. Bull., vol. 44, no. 2, pp. 178–180, 1999.
196. S. Davis and R. G. Burns, Decolorization of Phenolic Effluents by Soluble and
Immobilized Phenol Oxidases. Appl. Microbiol. Biotechnol., vol. 32, no. 6, pp.
721–726, 1990.
197. J. P. Solyanikova and L. A. Golovleva, Phenol Hydroxylases: An Update. Biochem-
istry, vol. 64, no. 4, pp. 365–372, 1999.
198. D.H. Nies, Microbial Heavy-Metal Resistance. Appl. Microbiol. Biotechnol., vol.
51, pp. 730–750, 1999.
199. R. S. Prakasham, J. S. Merrie, R. Sheela, N. Saswathi, and S. V. Ramakrishna,
Biosorption of Chromium VI by Free and Immobilized Rhizopus arrhizus. Environ.
Pollut., vol. 104, pp. 421–427, 1999.
200. A. Hassen, N. Saidi, M. Cherif, and A. Boudabous, Effects of Heavy Metals on
Pseudomonas aeruginosa and Bacillus thuringiensis. Bioresource Technol., vol. 65,
pp. 73–82, 1998.
201. V. M. Ibeanusi and E. A. Archibold, Mechanisms for Heavy Metal Uptake in a
Mixed Microbial Ecosystem. ASTM Spec. Technol. Pub., pp. 191–203, 1995.
202. M. Mergaey, Microbial Resources for Bioremediation of Sites Polluted by Heavy
Metals. NATO ASI Ser. 3, vol. 19, pp. 65–73, 1997.
203. J. Adams, T. Pickett, and J. Montgomery, Biotechnologies for Metal and Toxic
Inorganic Mining Processes and Waste Solutions. In N. McPherson and R. Nora,
(eds.), Randol Gold Forum Conf. Proc., pp. 143–146. Golden CO: Randol Interna-
tional, 1996.
Biotechnology Principles 211

204. G. Bunke, P. Gotz, and R. Bucholz, Metal Removal by Biomass: Physio-Chemical

Elimination Methods. J. Biotechnol., vol. 11a, pp. 431–452, 1999.
205. K. M. Paknikar, P. R. Puranik, and A. V. Pethkar, Development of Microbial
Biosorbents—A Need for Standardization of Experimental Protocols. Process Metall.,
vol. 9B, pp. 363–372, 1999.
206. B. Volesky, Biosorption for the Next Century. Process Metall., vol. 9B, pp. 161–170,
1999.
207. G. C. Donmez, Z. Aksu, A. Ozturk, and T. Kutsal, A Comparative Study on Heavy
Metal Biosorption Characteristics of Some Algae. Process Biochem., vol. 34,
pp. 885–892, 1999.
208. A. Kapoor, T. Viraraghavan, and D. R. Cullimore, Removal of Heavy Metals Using
the Fungus Aspergillus niger. Bioresource Technol., vol. 70, pp. 95–104, 1999.
209. I. Bakkaloglu, T. J. Buter, L. M. Evison, F. S. Holland, and I. C. Hancock, Screening
of Various Biomass for the Removal of Heavy Metals (Zn, Cu, Ni) by Biosorption,
Sedimentation, and Desorption. Water Sci. Technol., vol. 38, no. 6, pp. 269–277,
1998.
210. A. Stoll and J. R. Duncan, Enhanced Heavy Metal Removal from Waste Water by
Viable, Glucose Pretreated Saccharomyces cerevisiae Cells. Biotechnol. Lett.,
vol. 18, no. 10, pp. 1209–1212, 1996.
211. K. Tsekova, A. Kaimaktchiev, and A. Tsekova, Bioaccumulation of Heavy Metals
by Microorgansims. Biotechnol. Biotechnol. Equip., vol. 2, pp. 94–96, 1998.
212. M. Li, H. Jiang, W. Hou, and L. Xing, Heavy Metal Biosorption of Yeasts. Junwu
Xitong, vol. 17, no. 4, pp. 367–373, 1998.
213. A. Ruiz-Manirquez, J. A. Noriega, J. H. Yeomans, L. J. Orgega, and P. I. Magana,
Biosorption of Heavy Metals by Thiobacillus ferrooxidans. Rev. Soc. Quim. Mex.,
vol. 42, no. 5, pp. 228–233, 1998.
214. A. L. Kansoh and H. A. El-Shafei, Multiple Heavy Metal Tolerance in Some Fungal
and Bacerial Strains. Afr. J. Mycol. Biotechnol., vol. 6, no. 3, pp. 31–40, 1998.
215. J. S. Chang and J. C. Huang, Selective Adsorption/Recovery of Pb, Cu, and Cd with
Multiple Fixed Beds Containing Immobilized Bacterial Biomass. Biotechnol. Prog.,
vol. 14, no. 5, pp. 735–741, 1998.
216. L. Philip, L. Iyengar, and C. Venkobacher, Immobilized Microbial Reactor for
Heavy Metal Pollution Control. Int. J. Environ. Pollut., vol. 6, no. 2/3, pp. 277–284,
1996.
217. R. Ileri and A. Akkoyunlu, Biosorption of Immobilized Dead Biomass in a Contin-
uous Sheet Bioreactor. Fresenius Environ. Bull., vol. 8, no. 5/6, pp. 397–404, 1999.
218. Y. Sag, U. Acikel, Z. Zumriye, and T. Kutsal, Competitive Biosorption of Chromium
(VI), Iron (III), and Copper (II) Ions from Binary Metal Mixtures by R. arrhizus
and C. vulgaris. Turkish J. Eng. Environ Sci., vol. 22, no. 2, pp. 145–154, 1998.
219. Y. Sag and T. Kutsal, An Overview of the Studies About Heavy Metal Adsorption
Process by Microorganisms In: A. Hassen, N. Saidi, M. Cherif, and A. Boudabous,
(eds.) Effects of Heavy Metals on Pseudomonas aeruginosa a n d Bacillus
thuringiensis. Bioresource Technol., vol. 65, pp. 73–82, 1998.
220. A. Tomasini-Campocosio, S. Escarcega-Cruz, E. Gonzalez-Iribarren, and M. Meraz-
Rodriguez, Biosorption of Heavy Metals. Inf. Technol., vol. 9, no. 6, pp. 73–77,
1998.
212 Cutright

221. U. Yetis, G. Ozcengiz, F. B. Dilek, N. Ergen, A. Erbay, and A. Dolek, Heavy Metal
Biosorption by White-Rot Fungi. Water Sci. Technol., vol. 38, no. 4–5, pp. 323–330,
1998.
222. H. Seki and A. Suzuki, Biosorption of Heavy Metal Ions to Brown Algae, Macro-
cystis pyrifera, Kjellmaniella crassifolia, and Undaria pinnatifida. J. Colloid Inter-
face Sci., vol. 206, no. 1, pp. 297–301, 1998.
223. T. J. Butter, L. M. Evison, I. C. Hancock, and F. S. Holland, The Kinetics of Metal
Uptake by Microbial Biomass: Implications for the Design of a Biosorption Reactor.
Water Sci. Technol., vol. 38, no. 6, pp. 279–286, 1998.
224. A. I. Ferra and J. A. Teixeira, The Use of Flocculating Brewer’s Yeast for Cr(III)
and Pb(II) Removal from Residual Wastewaters. Bioprocess Eng., vol. 21, pp.
431–437, 1999.
225. A. Hammaini, A. Ballester, F. Gonzlaez, M. L. Blazquez, and J. A. Munoz, Activated
Sludge as a Biosorbent of Heavy Metals. Process Metall., vol. 9B, pp. 185–192,
1999.
226. M. Bustard, G. McMullan, and A. P. McHale, Biosorption of Textile Dyes by
Biomass Derived from Kluyveromyces marxianum IMB3. Bioprocess Eng., vol. 19,
pp. 427–430, 1998.
227. R. F. Unz and K. L. Shuttleworth, Microbial Mobilization and Immobilization of
Heavy Metals. Curr. Opin. Biotechnol., vol. 7, no. 3, pp. 307–310, 1996.
228. G. M. Gadd, Microbial Control of Heavy Metal Pollution. In J. C. Fry, G. M. Gadd,
R. A. Herbert, C. W. Jones, and I. A. Watson-Clark, (eds.), Microbial Pollution
Control, pp. 59–88. Cambridge, U.K.: Cambridge University Press, 1992.
229. L. F. Stratchan, D. J. Leak, and A. G. Livingston, Minimization of Excess Biomass
Production in an Extractive Membrane Bioreactor. Eur. Conf. Young Researchers
in Chemical Engineering, vol. 2. pp. 1043–1045. New York: IChE Publishing, 1995.
230. D. P. Smith, Submerged Filter Biotreatment of Hazardous Leachate in Aerobic,
Anaerobic, and Anaerobic/Aerobic Systems. Haz. Waste Haz. Mater., vol. 12, no.
2, pp. 167–187, 1995.
231. F. E. Hall, Jr., C-ALC Hazardous Waste Minimization Strategy: Reduction of
Industrial Biological Sludge from Industrial Wastewater Treatment Facilities. Air
and Waste Management Report WP82BO1/1, 1997.
232. H. Yasui and M. Shibata, An Innnovative Approach to Reduce Excess Sludge
Production in the Activated Sludge Process. Water Sci. Technol., vol. 30, no. 9,
pp. 11–20, 1994.
233. M. Mayhew and T. Stephenson, Low Biomass Yield Activated Sludge: A Review.
Environ. Technol., vol. 18, no. 9, pp. 883–892, 1997.
234. E. Low and H. A. Chase, The Effect of Maintenance Energy Requirements on
Biomass Production During Wastewater Treatment. Water Res., vol. 33, no. 3, pp.
847–853, 1998.
235. M. Mayhew and T. Stephenson, Biomass Yield Reduction: Is Biochemical Manip-
ulation Possible Without Affecting Sludge Process Efficiency? Water Sci. Technol.,
vol. 38, no. 8–9, pp. 137–144, 1998.
236. I. Purtschert and W. Gujer, Population Dynamics by Methanol Addition in Denitri-
fying Wastewater Treatment Plants. Water Sci. Technol., vol. 39, no. 1, pp. 43–50,
1999.
Biotechnology Principles 213

237. E. Low and H. A. Chase, Reducing Production of Excess Biomass During Waste-
water Treatment. Water Res., vol. 33, no. 5, pp. 1119–1132, 1999.
238. W. Verstrate and P. Vandevivere, New and Broader Applications of Anaerobic
Digestion. Crit. Rev. Environ. Sci. Technol., vol. 29, no. 2, pp. 151–173, 1999.
239. F. L. Smith, G. A. Sorial, M. T. Suidan, A. W. Breen, P. Biswas, R. C. Brenner,
Development of Two Biomass Control Strategies for Extended, Stable Operation of
Highly Efficient Biofilters with High Toluene Loadings. Environ. Sci. Technol.,
vol. 30, no. 5, pp. 1744–1751, 1996.
240. C. F. Gokcay and U. Yetis, Effect of Nickel (II) on Biomass Yield of Activated
Sludge. Water Sci. Technol., vol. 34, no. 5–6, pp. 163–171, 1996.
241. R. Grover, S. S. Marwaha, and J. F. Kennedy, Studies on the Use of an Anaerobic
Baffled Reactor for the Continuous Anaerobic Digestion of Pulp and Paper Mill
Black Liquors. Process Biochem., vol. 34, pp. 653–657, 1999.
242. D. Brady, P. D. Rose, and J. R. Duncan, The Use of Hollow Fiber Cross-Flow
Microfiltration in Bioaccumulation and Continuous Removal of Heavy Metals
from Solution by Saccharomyces cerevisie. Biotechnol. Bioeng., vol. 44, no. 11,
pp. 1362–1366, 1994.
243. M. Reiser, K. Fischer, and D. Bardtke, Biomembrane Reactor—From Laboratory to
Industrial Use. In W. L. Prins, J. Van Ham (eds.), Proc. 4th Int. Biological Waste
Gas Clean, pp. 181–188. Duesseldorf: VDI Verlag, 1997.
244. S. Inguva, M. Boensch, and G. Shreve, Microbial Enhancement of TCE and
1,2-DCA Solute Flux in UF-Membrane Bioreactors. AIChE J., vol. 44, no. 9,
pp. 2112–2123, 1998.
245. N. Cicek, J. P. Franco, M. T. Suidaon, V. Urbain, and J. Manem, Characterization
& Comparison of a Membrane Bioreactor and a Conventional Activated Sludge
System in the Treatment of Wastewater Containing High Molecular Weight Com-
pounds. Water Environ. Res., vol. 71, no. 1, pp. 64–70, 1999.
246. J. A. Scott, D. A. Neilson, and P. N. Boon, A Dual Function Membrane Bioreactor
System for Enhanced Aerobic Remediation of High-Strength Industrial Wastewater.
Water Sci. Technol., vol. 38, no. 4–5, pp. 413–420, 1998.
247. J. S. Almeida, M. Maria, and J. G. Crespo, Development of Extractive Membrane
Bioreactors for Environmental Applications. Environ. Protect. Eng., vol. 25, no.
1–2, pp. 111–121, 1999.
248. K. Brindle, T. Stephenson, and M. J. Semmens, Pilot-Plant Treatment of a High-
Strength Brewery Wastewater Using a Membrane-Aeration Bioreactor. Water Envi-
ron. Res., vol. 71, no. 6, pp. 1197–1203.
249. M. A. Deshusses, W. Chen, A. Mulchandani, and I. J Dunn, Innovative Bioreactors.
Curr. Opin. Biotechnol., vol. 8, no. 2, pp. 165–168, 1997.
250. C. Wen, Q. Xia, and Y. Qian, Domestic Wastewater Treatment Using an Anaerobic
Bioreactor Coupled with Membrane Filtration. Process Biochem., vol. 35, no. 3,4,
pp. 335–340, 1999.
251. M. Bodzek, Membrane Techniques in Wastewater Treatment. Environ. Protect.
Eng., vol. 25, no. 1–2, pp. 153–192, 1999.
252. H. L. Bohn and K. H. Bohn, Moisture in Biofilters. Environ. Prog., vol. 18, no. 3,
pp. 156–161, 1999
214 Cutright

253. C. Hatanaka, Development of High Performance Bioreactor. Chiaki Bio Ind., vol.
16, no. 9, pp. 40–50, 1999.
254. L. Diels, R. S. Van, K. Somers, I. Willems, W. Doyen, M. Mergeay, D. Springael,
and R. Leysen, The Use of Bacteria Immobilized in Tubular Membrane Reactors
for Heavy Metal Recovery and Degradation of Chlorinated Aromatics. J. Membrane
Sci., vol. 100, no. 3, pp. 249–258, 1995.
255. D. M. F. Prazeres and J. M. S Cabral, Enzymatic Membrane Bioreactors: Current
State of the Art and Future Prospects. Recent Adv. Mar. Biotechnol., vol. 2,
pp. 181–223, 1998.
11
Novel Materials and Processes for Pollution
Control in the Mining Industry

Alan Fuchs and Shuo Peng

University of Nevada, Reno, Reno, Nevada

Tremendous opportunities exist in the area of novel materials for environmental

engineering applications. These opportunities have traditionally been in the areas
of membranes, ion exchange, and adsorbents, but new areas relating to techno-
logical advances in “nanomaterials” and “bio-applications” have spawned new
generations of designed materials for many pollution control applications. The
emphasis in this chapter will be on new technologies which have been or will be
useful for pollution control in the mining industry. This will require a review of
developments in the general areas of membranes, ion exchange, and adsorption,
and discussion of how these materials are useful in mining applications.

1 MEMBRANES MATERIALS AND PROCESSES

A great deal of work has been done on the use of membrane processes for
treatment of mine waters. Some of the typical membrane configurations of
membranes separators are shown in varied textbooks (e.g., Ref. 1). This text also
has examples of hollow fiber membranes and typical flow arrangements of these
systems. Recent examples of this include separation of rare earths using liquid
membranes (2) and copper recovery from Chilean mine waters, also using
215
216 Fuchs and Peng

liquid membranes (3). Earlier studies in this area related to the removal of
ammonium and nitrate ions from mine effluents using nanofiltration and reverse
osmosis (4,5).
A major focus in the area of membrane development during the past few
years has been in the use of liquid membranes. Valenzuela et al. (3) describe the
use of a hollow fiber-supported liquid membrane for the recovery and concentra-
tion of copper from Chilean mine water. This system works by impregnating the
porous structure of the membrane with an organic film which acts as a selective
extraction medium. The film is salicylaldoximic extractant. Utilizing a low
concentration of the extractant, a high degree of copper recovery is possible using
this technique. An organic solution, containing 5-dodecylsalicylaldoxime was
dissolved in a solvent containing 91% aliphatics and 9% aromatics. A feedstream
containing 0.8–1.4 g/liter Cu(II) was used for testing, with a pH of 2.8–3.2 and a
density of 1.05 g/ml at 20˚C. Concentrated sulfuric acid solutions were used as
metal-acceptor stripping agents. The hollow fiber membranes were microporous
PTFE fibers with a membrane pore size of 2.0 µm. Because of the hydrophobic
nature of the membrane used, the pores are rapidly and easily filled with the
solvent-containing “carrier” extractant. A similar system was also studied by
Valenzuela et al. (3). After impregnation with the the solvent, the feed solution
and acid strip solution was recirculated through the fiber bores and outside the
fibers, respectively.
Copper extraction in a membrane extractor occurs by diffusion of copper
ions from the bulk feed into the solvent. The reaction which takes place in the
membrane is as follows:

Cu+2(aq) + 2HR(org) = CuR2(org) + 2H+(aq)

where HR is the acidic extractant and CuR2 is the metal complex extracted into
the organic phase. The extractant diffuses through the membrane into the acid,
where the stripping reaction takes place. In this way the carrier is regenerated and
copper ions are free to be collected in the stripping liquor.
The results of this work indicate that copper concentrations can be reduced
from 1 g/liter to below 0.2 g/liter in 8 h. The effectiveness of copper removal is
dependent on the sulfuric acid concentration in the stripping solution.
Yang et al. (2) describe the use of a combined extraction/electrostatic
psuedo liquid membrane (ESPLIM) for extraction and separation of rare earths
in a simulated mine water. In this process, a continuous organic phase, consisting
of 20% di-(2-ethylhexyl) phosphoric acid and 80% kerosene serves as the bulk
liquid membrane. This stream contains a specific extractant for the metal ions to
be extracted and separated. A discontinuous, aqueous stream, <0.2 mm in diam-
eter, is used for phase settlement. The system consists of a grounded electrode
coated with polyethylene film mounted at the sides of extraction and stripping
Pollution Control in the Mining Industry 217

cells in a rectangular reaction tank. A high-voltage electrode coated with poly-

ethylene is wound on the perforated baffle plate separating the extraction and
stripping cell. When a high-voltage electrostatic field is applied to the reaction
tank, the aqueous drops in the organic continuous phase disintegrate into numer-
ous smaller droplets under the action of the electrostatic field. This provides a
great deal of surface area for separation. The extractant dissolved in the continu-
ous organic phase acts as a shuttle to transport metal ions from the extraction cell
to the stripping cell.
A summary of opportunities for membrane technologies in the treatment of
mining and mineral process streams was presented by Awadalla and Kumar (4).
This study indicated a variety of applications including acid mine drainage
(AMD), treatment of flotation water, copper smelting and refining wastewater,
mill wastewater, removal of ammonium and nitrate ions, membranes in the
aluminum industry, treatment of groundwater, treatment of uranium wastewater,
treatment of dilute gold cyanide solutions, recovery of zinc from pond water, rare
earth (RE) concentration, and separation of selenium from barren solution.
AMD contains pollutants such as iron, manganese, calcium, magnesium,
and sulfate ions. Although lime neutralization is considered the “best available
technology economically achievable,” it is no longer considered environmentally
acceptable because of the low-level contamination of heavy metals which cannot
be removed. Alternatively, almost complete removal of dissolved solids can be
achieved by the use of ion exchange, distillation, and reverse osmosis (RO) to
produce high-quality water which can be used by municipalities or industry. The
use of RO is best implemented as a supplement to neutralization processes.
The RO concentrate stream is neutralized and clarified prior to discharge or
recycled. Coupled RO/ion exchange can be used when high concentration of
calcium sulfate and/or iron fouling is a problem. For the case of water reuse in
which completely demineralized water is not essential, a charged ultrafiltration
process using negatively charged noncellulosic membranes was utilized. For the
case of AMD for coal conversion processes, high-ultrafiltration recovery with
high removal of calcium sulfate and iron and good flux are required. Recovery
of up to 97% is achievable by introducing an interstage settling step. Commer-
cially available charged ultrafiltration membranes by PSAL (millipore type of
noncellulosic skin on cellulosic backing) were used in this study. Cost for
treatment using UF with interstage settling are $1.33/1000 gal of AMD, including
membrane replacement cost, pumping cost, and lime cost.
In order to avoid problems with recycling wastewater from flotation mills
which contain the breakdown products of collector-frother reagents, the water
must be purified before recycling to the mining operation. The traditional method
for treatment of flotation water involves lime precipitation, ozonation, adsorption
on activated carbon, and biological treatment (4). Biological treatment requires
excessive holdup and is dependent on the climate, the presence of toxic heavy
218 Fuchs and Peng

metals, and sensitive control of the microorganisms. Reverse osmosis has been
used for the recovery of flotation reagents. Commercial RO membranes have
been used to remove 95% of organic carbon, calcium, and magnesium from the
flotation feed stream.
Scrubber blowdown from a primary copper smelting plant and acid process-
ing water from a selenium-tellurium plant have been treated using negatively
charged noncellulosic ultrafiltration membranes (4) . Removal of over 85% of As
and Se from the acid processing water was made possible when the pH was
adjusted to 10 and the solids were settled prior to ultrafiltration. Scrubber
blowdown was effectively treated without pH adjustment to a pH of 4.5. Arsenic-
containing wastewater was also pretreated with UF and polished using RO. This
method produced a permeate stream containing less than 50 ppb arsenic.
Alkaline solutions of NaCN are used to leach gold-containing ores, produc-
ing dilute gold cyanide solutions (4). The two conventional methods of recovering
gold from these solutions include the Merill-Crowc process of cementation using
zinc powder and adsorption using activated carbon. Concentrated gold solutions
are formed by elution. Reverse osmosis has been investigated as a means to
concentrate the dilute gold solutions. In the case of metal finishing operations
using gold and cyanide solutions, FilmTec FT-30 membranes have been used to
provide rejections in the range of 91–99% for free and combined cyanides (with
copper and zinc). Membrane performance was strongly pH dependent. Reverse
osmosis has also been used for silver and copper cyanide concentration (Os-
monics, Inc). This study utilized a nitrogen-containing aromatic condensation
polymer. Experiments indicated that the feed could be concentrated three times
with 70% removal of permeate, resulting in low gold content in the permeate.
Nanofiltration (NF) and RO have been used for removal of ammonium and
nitrate ions from synthetic and actual mine effluents (5). In mine and mill water,
ammonium and nitrate ions are generated from the degradation of cyanide from
gold mill effluents and ammonium nitrate-fuel oil (ANFO) blasting agents in
mines. Nitrogen-containing reagents are also used in ore processing and extrac-
tive metallurgy. The results of experiments using NF and RO membranes were
reported for testing and actual mill effluent. The results of the testing were that
good removal of ammonium (>99%) and nitrate ions (>97%) were achieved using
RO, while NF was less effective. Lower effectiveness of the NF membrane was
believed to be caused by ammonium being present in the sulfate form and not the
larger ammonium iron sulfate complex which does not form because there is no
iron in the mining effluent. No scaling or fouling problems were observed in
these studies.
Cross-flow membrane technolgies have also been applied to mineral sus-
pensions (6). In this study, using microporous filtration (0.1-µm membranes)
suspensions of CaCO3 were investigated using an intermittent cleaning approach
in order to increase the permeate flux.
Pollution Control in the Mining Industry 219

A thorough review of membrane technology for applications to industrial

wastewater treatment has been made by Caetano (7). In this review, E. Drioli
provides a broad overview in the areas of desalination, gas separation, pervapora-
tion, membrane bioreactors, enzyme membrane reactors, and hybrid systems
based on pervaporation and distillation.
In the more general area of environmental applications, significant work
has been done on the treatment of streams containing metals. There has been a
great deal of interest in the use of ion-exchange membranes in this area. Sengupta
(8) has investigated electromembrane partitioning as a means for heavy metal
decontamination. This is a unique and rather interesting new approach for the
in-situ removal of metals from contaminated soils.
A low-level direct current (DC), less than 1 V/cm, is applied to the soil
while a composite ion-exchange membrane is wrapped around the cathode. Upon
imposition of the DC potential, the cations move toward the cathode, where they
are captured by the composite membrane. By the design of the ion-exchange
membrane, the nonselective ions should pass freely through the membrane. The
membrane utilized for this work is a thin sheet prepared by grinding a cross-
linked polymer ion exchanger and suspending the ion exchanger in a PTFE
porous matrix. These membranes are 90% ion exchanger, 10% PTFE, and are
microporous with >40% voids with a pore size distribution below 0.5 µm. One
potential problem with this process is that periodically these membranes must be
removed and chemically regenerated with strong (3–5%) mineral acid solution.
Electrodialytic decontamination of soil polluted with heavy metals has been
investigated using ion-exchange membranes by Hansen et al. (9). The process for
removal of metal ions from soils using electric current and passive membranes is
known as electrokinetic soil remediation. This method involves the use of passive
membranes to separate the polluted soil from the electrodes. There are several
shortcomings to this approach, including addition of acid counterions into the soil,
return of heavy metals back into the soil, and heavy-metal precipitation at the H+
and OH– front. By introduction of ion-exchange membranes into the electro-
kinetic soil remediation process, an electrodialytic soil remediation process
results. The ion-exchange membranes are oriented in certain directions. This
orientation, with pairs of anion- and cation-exchange membranes placed on both
sides of the polluted soil, eliminates all three of the problems mentioned above.
This configuration also provides two compartments containing liquid solutions
and the heavy metals, which can be withdrawn as needed. In this situation,
heavy-metal ions pass through the cation-exchange membrane in the direction of
the cathode and are prevented from passing through the anion-exchange mem-
brane and never reach the cathode. They end up in the compartment between the
two ion-exchange membranes.
Li et al. (10) have investigated the use of a cation-selective membrane for
removal of heavy metals from soils. An improvement in the traditional elec-
220 Fuchs and Peng

troremediation approach is described. In this work a cation membrane is placed

around the cathode to prevent hydroxyl species moving toward the anode. This
prevents precipitation of metals in the soil, and the metals precipitate in the
column of water around the cathode.
Membranes are also used for removal of metals for industrial applications
(11). Bulk liquid membranes are used for facilitated transport of silver using a
rotating film pertraction device. In this process two aqueous solutions are
separated by an organic liquid. The membrane liquid is in contact with the donor
and acceptor liquids adhering to the surfaces of the rotating disks. Transport of
the solvent involves extraction from one solution and stripping in the other. This
paper describes the recovery of silver from nitrate solutions using the rotating film
pertraction method using tri-isobutylphosphine sulfide (TIBPS) in n-octane as the
liquid membrane. Aqueous silver nitrate was the donor phase and the acceptor
phase was aqueous ammonia. The results of the study indicated that because of
low rates of transport of other metals, including copper, zinc, and nickel, rotating
film pertraction can be used effectively to separate silver from solution.
Yang et al. (2) describe a unique metal extraction method using two sets of
hydrophobic microporous hollow fiber membranes for separation of metals in
solution. One set of hollow fibers carries an acidic organic extractant (LIX 84,
anti-2-hydroxy-5-nonylacetophenone oxime) in a diluent. The other set of hollow
fibers carries a basic organic extractant (TOA, tri-n-octylamine). The aqueous,
metal-containing stream is carried on the shell side of the membrane system.
Cations, including copper, zinc, and nickel, are transported into the acidic
extractant. Anions, including chromium(VI), mercury, and cadmium, are ex-
tracted into the basic stream.
Palladium has also been separated from silver in a nitric acid solution using
liquid surfactant membranes (12). The organic carrier used in these studies is LIX
860, which is a β-hydroxyoxime. The liquid surfactant membrane is Span 80, a
commercially available surfactant, and the solvent is n-heptane. The aqueous
donor phase contains silver and palladium and is acidified using nitric acid. The
receiving phase contains thiourea and is tested in hydrochloric, perchloric, nitric,
and sulfuric acids. Under optimal conditions, palladium was separated from silver
recovered in entirety.
Another liquid membrane, investigated by Fu et al. (13), is trioctylamine
(TOA) as a mobile carrier in kerosene. Precious metals, including gold,
palladium, platinum, iridium, and ruthenium in hydrochloric acid, were ex-
tracted using this membrane system. The metals were extracted into perchlorate
and nitric acid solutions. An inert PTFE polymer 80 µm thick, 74% porous,
and 0.45 µm in average pore diameter was used as a support for the liquid
membrane.
Low-pressure reverse osmosis (RO) was used by Ujang and Anderson (14)
for separation of mono- and divalent ions. Sulfonated polysulfone membranes are
Pollution Control in the Mining Industry 221

used as a low-pressure reverse osmosis process for separation of mono- and

divalent zinc ions. It was observed that the higher the operating pressure, the
greater was the permeate flux for both species. At lower operating pressure,
higher permeate fluxes were observed using divalent ions. Metal removal of
divalent ions was greater for divalent ions than for monovalent ions for all
concentrations.

2 ADSORPTION MATERIALS AND PROCESSES

Recovery of gold from cyanide has been evaluated using many different ad-
sorbent materials. Petersen and Van Deventer (15) investigated the competitive
role of gold and organics on adsorption by a variety of adsorbents, including
activated carbon, ion-exchange resin, ion-exchange fibers, and membranes. A
variety of adsorbents were investigated, including coconut shell activated carbon,
macroporous ion-exchange resin, ion-exchange membrane, and ion-exchange
fibers (polypropylene-based strong-base and weak-base fibers). Adsorbents were
evaluated after being exposed to the organic compound, sodium ethyl xanthate,
for 6 h. The absorbents were challenged with a variety of organic compounds,
including ethanol, sodium ethyl xanthate, potassium amyl xanthate, and phenol.
The two mechanisms investigated to explain the reduced adsorption of gold in
the presence of the organics were (a) blockage of the carbon pores by the organic,
and (b) competition between gold cyanide and organics for the active sites on the
carbon surface. The results of the study indicated that both the rate of adsorption
and the equilibrium loading were affected by the organic on the adsorption of
gold cyanide onto activated carbon. The resin particles were only effected by the
rate of adsorption, while the membranes and fibers experienced both kinetic and
equilibrium changes. The results of this study indicated that the long-chain
organics (xanthates) have a higher degree of inhibition of mass transfer of gold
cyanide compared to the low-molecular-weight substance (ethanol). The aromatic
substances did not affect the performance of the fibers or membrane. This is
because the small pore diameters did not permit the large aromatics to penetrate.
The results indicated that the second mechanism, a competitive effect between
gold cyanide and the organic compounds, was responsible for the results observed
for the gold-equilibrated absorbents.
Klein et al. (16) have investigated polymeric resins as adsorbents for
industrial applications. The motivation for investigation of polymeric resins
versus activated carbon is their ease of regeneration. Activated carbon systems
are typically regenerated using steam or thermal methods, while polymeric resins
can be regenerated using simple solvents such as aliphatic alcohols. The resins
used were methylene-bridged styrene divinylbenzene-based co-polymer (Dow
Chemical, Midland, MI). Some of the characteristics of these polymeric resins
which may be controlled are hydrophobicity, pore size, and surface area. These
222 Fuchs and Peng

resins were challenged with benzoic acid and chlorobenzene, and adsorption
isotherms and bed regeneration curves were generated. The results of this study
indicated that with only few bed volumes (15–25), using methanol as regenerant,
90–95% of the adsorbed solute could be recovered. The polymeric resins main-
tained good adsorptive capacity after repeated cycling.

3 ION-EXCHANGE MATERIALS AND PROCESSES

Applications of ion exchange to leaching solutions of an Algerian gold ore have
been investigated by Akretche et al. (17). In the cyanide medium, the gold and
other metals such as silver, copper, and iron attach to the anion-exchange resin.
These metals are later eluted with acid thiourea to yield a concentrated solution
which is treated by cementation or an electrolytic method. This work describes
the use of electrodialysis of copper(I), which is normally not feasible due to the
presence of formamidine disulfide. This is accomplished when the solutions are
obtained by elution of cuprocyanides by thiourea.

4 CONCLUSIONS
There have been great strides in the development of new technologies for
pollution control in the mining industry during the past five years, many in the
development of new materials and processes. Many of these developments are in
the areas of membranes, adsorbents, and ion exchange. In the area of membranes,
a great of work has been done using liquid membranes. These are generally
supported synthetic membrane systems with a variety of liquids to facilitate
transport. Electroremediation and electrodialytic membrane approaches have also
seen a great deal of attention. Activated carbon-based and other organic ab-
sorbents have been used for treatment of mining wastes. Polymeric resins have
also been used as adsorbents for industrial applications. Anionic ion-exchange
resins have also been used for treatment of leaching solutions.

REFERENCES
1. W. L. McCabe, J. C. Smith, and P. Harriott, Unit Operations of Chemical Engineer-
ing, 5th Edition, New York: McGraw-Hill, 1993.
2. Z. F. Yang, A. K. Guha, and K. Sirkar, Ind. Eng. Chem. Res., vol. 35, pp. 1383–1394,
1996.
3. F. Valenzuela, C. Basualto, C. Tapia, and J. Sapag, J. Membrane Sci., vol. 155, pp.
163–168, 1999.
4. F. T. Awadalla and A. Kumar, Separation Sci. Technol., vol. 29, no. 10, pp. 1231–
1249, 1994.
5. F. T. Awadalla, C. Striez, and K. Lamb, Separation Sci. Technol., vol. 29, no. 4,
pp. 483–495, 1994.
Pollution Control in the Mining Industry 223

6. D. Si-Hassen, A. Ould-Dris, M. Y. Jaffrin, and Y. K. Benkahla, J. Membrane Sci.,vol.

118, pp. 185–188, 1996.
7. A. Caetano (ed.), Membrane Technology: Applications to Industrial Wastewater
Treatment. Dordrecht, The Netherlands: Kluwer, 1995.
8. S. Sengupta and A. K. Sengupta, Hazardous and Industrial Wastes—Proceedings of
the Mid-Atlantic Industrial Waste Conference Proceedings of the 1997 29th Mid-
Atlantic Industrial and Hazardous Waste Conference, July 13–16, 1997, Blacksburg,
VA, Lancaster, PA: Technomic Publishing Co. Inc., pp. 174–182.
9. H. K. Hansen, L. M. Ottosen, S. Laursen, and A. Villumsen, Separation Sci. Technol.,
vol. 32, no. 15, pp. 2425–2444, 1997.
10. Z. Li, J. Yu, and I. Neretnieks, Environ. Sci. Technol., vol. 32, pp. 394–397, 1998.
11. L. Boyadzhiev and K. Dimitrov, J. Membrane Sci., vol. 68, p. 137–143, 1994.
12. T. Kakoi, M. Goto, and F. Nakashioo, Separation Sci. Technol., vol. 32, no. 8,
pp. 1415–1432, 1997.
13. J. Fu, S. Nakamura, and K. Akiba. Separation Sci. Technol., vol. 32, no. 8, pp. 1433–
1445, 1997.
14. Z. Ujang and G. K. Anderson, Water Sci. Technol., vol. 38, no. 4–5, pp. 521–528,
1998.
15. F. W. Petersen and J. S. J. Van Deventer, Separation Sci. Technol., vol. 32, no. 13,
pp. 2087–2103, 1997.
16. J. Klein, G. M. Gusler, and Y. Cohen, Removal of Organics from Aqueous Systems:
Dynamic Sorption/Regeneration Studies with Polymeric Resins. In Novel Absorbents
and Their Environmental Applications, Y. Cohen and R. W. Peters (eds.), AIChE
Symp. Ser., vol. 91, pp. 72–78.
17. D. E. Akretche, A. Gherrou, and H. Kerdjoudj, Hydrometallurgy, vol. 46, pp. 287–
301, 1997.
12
Monitoring In-Situ Electrochemical Sensors

Joseph Wang
New Mexico State University, Las Cruces, New Mexico

1 INTRODUCTION
Electroanalytical methods are concerned with the interplay between electricity
and chemistry, namely, the measurements of electrical quantities such as current
or potential, and their relationship to chemical parameters. Electroanalytical
chemistry can play a major role in pollution control and prevention. In particular,
electrochemical sensors and detectors are very attractive for on-site and in-situ
monitoring of priority pollutants. Such devices are highly sensitive, selective
toward electroactive species, fast, accurate, compact, portable, and inexpensive.
Several electrochemical devices, such as oxygen or pH electrodes, have been
widely used for years for environmental analysis. Recent advances in electro-
chemical sensor technology have expanded the scope of electrochemical devices
toward a wide range of organic and inorganic contaminants.
The present chapter reviews recent efforts at the author’s laboratory, aimed
at in-situ monitoring of priority pollutants. Continuous monitoring, effected in the
natural environment, offers a rapid return of the chemical information (with a
proper alarm in case of a sudden discharge), avoids costs and errors associated
with the collection of discrete samples, while maintaining the sample integrity.
The use of remote sensors thus has significant technical and cost benefits over

225
226 Wang

traditional sampling and analysis. Our latest developments of remote electro-

chemical probes will be covered in the following sections.

2 REMOTELY DEPLOYED ELECTROCHEMICAL SENSORS

Remotely deployable submersible sensors capable of monitoring contaminants in
both time and location are advantageous in a variety of applications. These range
from shipboard marine surveys, downhole monitoring of groundwater contami-
nation, to real-time analysis of industrial streams. The development of sub-
mersible electrochemical probes requires proper attention to various challenges,
including the effect of sample pH, ionic strength, dissolved oxygen, or natural
convection, specificity and sensitivity, surface fouling, in-situ calibration, and
miniaturization. By addressing these and other obstacles, we were able to develop
remote sensors for a wide range of inorganic and organic contaminants.

2.1 Remote Monitoring of Metal Contaminants

Metal pollution has received enormous attention due to its detrimental impact on
the environment. The need for continuous monitoring of trace metals in a variety
of matrices has led to the development of submersible sensors based on electro-
chemical stripping analysis (1,2). Stripping analysis has been established as a
powerful technique for determining toxic metals in environmental samples (3,4).
The remarkable sensitivity of stripping analysis is attributed to its unique “built-
in” preconcentration step, during which the target metals are electroplated onto
the surface. Both electrolytic and nonelectrolytic (adsorptive) accumulation
schemes have thus been employed to achieve sub-parts-per-billion detection
limits. The analytical current signal (i), obtained during the subsequent stripping
(potential scanning) step is proportional to the metal concentration (C) and
accumulation time (tacc):

i = KC tacc (1)

Remote metal monitoring has been realized by eliminating the needs for
mercury surfaces, oxygen removal, forced convection, or supporting electrolyte
(which previously prevented the direct immersion of stripping electrodes into
sample streams). This was accomplished through the development of nonmercury
electrodes, judicious coupling of potentiometric stripping operation, and the use
of advanced ultramicroelectrode technology (1). Compatibility with field opera-
tions was achieved by connecting the three-electrode housing [including a gold
fiber working electrode, in the polyvinyl chloride (PVC) tube], via environmen-
tally sealed three-pin connectors, to a 25-m-long shielded cable. Convenient and
simultaneous quantitation of several trace metal levels (e.g., Pb, Cu, Ag, Hg) has
Monitoring In-Situ Electrochemical Sensors 227

thus been realized in connection to measurement frequencies of 20–30/h (based

on deposition periods of 1–2 min).
The in-situ monitoring capability of the remote metal sensor was demon-
strated in studies of the distribution of labile copper in San Diego Bay (CA) (5).
For this purpose, the probe was floating on the side of a small U.S. Navy vessel.
The resulting map of copper distribution reflected the metal discharge and
circulation pattern in the bay. We are currently collaborating with Prof. Daniele’s
group in using the remote probes for assessing the distribution of metal contam-
inants in the canals and lagoon of Venice, Italy (6).
The extension of remote stripping electrodes to additional metals that
cannot be electroplated relies on the adaptation of adsorptive stripping protocol
for a submersible operation (7). Such procedures rely on the formation and
adsorptive accumulation of complexes of the target metals. Accordingly, remote
adsorptive stripping sensors require a new probe design based on an internal
solution chemistry. Such a renewable-reagent adsorptive stripping sensor relies
on the continuous delivery of the ligand, its complexation reaction with the metal
“collected” in a semipermeable microdialysis sampling tube, and transport of the
complex to the working electrode compartment. Such dialysis sampling also
offers extension of the linear range and protection against surface fouling (due to
its dilution and filtration actions).
The new flow-probe format was employed for monitoring trace metals such
as nickel, uranium, or chromium. As desired for effective in-situ monitoring, such
adsorptive stripping probes have the capability to detect rapidly fluctuations in
the analyte concentration continuously. Such ability is indicated from Figure 1,
which displays the response of a chromium probe (8) upon switching from the 5-
to 25 µg/l chromium solutions. Such behavior is attributed to the reversibility of
the accumulation/stripping cycle, with the stripping and subsequent 10-s “clean-
ing” steps completely removing the accumulated complex. In addition, the

FIGURE 1 Response of the remote chromium(VI) probe to alternate expo-

sures to (a) “low” (5 µg/l) and (b) “high” (25 µg/l) chromium(VI) levels.
Accumulation for 30 s at –0.9 V; square-wave voltammetric stripping scan.
228 Wang

reagent flow continuously replenishes the solution, to “erase” an internal buildup

of chromium.
Other groups have also been involved in the development of remote metal
sensors. For example, Kounaves’s team reported on probes based on mercury-
plated iridium-based microelectrode arrays and square-wave voltammetric strip-
ping detection (9). A solid-state reference electrode that eliminates leakage of
electrolyte to the surrounding low-ionic-strength aquatic environment was em-
ployed. The device developed by Buffle’s group (4,10) has been coupled to a
thick agarose-gel antifouling membrane that facilitates measurements in complex
media. There is no doubt that these and similar developments of submersible
stripping sensors will have a major impact on the surveillance of our water
resources.

2.2 Remote Modified Electrodes and Biosensors

Chemically and biologically modified electrodes (CMEs) have greatly enhanced
the power of electrochemical detectors and devices (11). The ability to deliber-
ately control and manipulate surface properties can lead to a variety of attractive
effects. Electrochemical sensors based on modified electrodes combine the re-
markable sensitivity of amperometry with new chemistries and biochemistries.
Such manipulation of the molecular architecture of the detector surface offers new
levels of reactivity that greatly expand the scope of electrochemical devices, and
enhance the power of in-situ electrochemical probes.

2.3 Biosensors
Biosensors are small devices employing biochemical molecular recognition prop-
erties as the basis for a selective analysis. The major processes involved in any
biosensor system are analyte recognition, signal transduction, and readout. The
remarkable specificity of biological recognition processes has led to the develop-
ment of highly selective electrochemical biosensors. In particular, enzyme elec-
trodes, based on amperometric or potentiometric monitoring of changes occurring
as a result of the biocatalytic process, have the longest tradition in the field of
biosensors. Such devices are usually prepared by immobilizing an enzyme onto
the electrode surface. The integration of these devices with remotely deployed
probes should add new dimensions of specificity to in-situ electrochemical
monitoring of pollutants. In the adaptation of enzyme electrodes to a submersible
operation, one must consider the influence of actual field conditions (pH, salinity,
temperature) on the biocatalytic activity.
The first remotely deployed biosensor targeted phenolic contaminants
in connection to a submersible tyrosinase enzyme electrode (12). The enzyme,
immobilized within a stabilizing carbon paste matrix, converted its phenolic
substrates to easily reducible quinone products. The sensor responded rapidly
Monitoring In-Situ Electrochemical Sensors 229

to micromolar levels of various phenol contaminants, with no carryover (mem-

ory) effects.
We also developed a remote biosensor for field monitoring of organo-
phosphate nerve agents (13). The device relied on the coupling the enzymatic
activity of organophosphate hydrolase (OPH) with the submersible amperometric
probe configuration. Low (micromolar) levels of paraxon or parathion have thus
been measured directly in untreated natural water matrices. The OPH enzyme
obviates the need for lengthy and irreversible enzyme inhibition protocols com-
mon to inhibition-based biosensors.
Finally, hydrogen peroxide and organic peroxides have been monitored at
large instrument–sample distances by incorporating a reagentless peroxidase
bioelectrode into the remote probe assembly (14). A low detection potential
(~0.0 V) accrued from the use (co-immobilization) of a ferrocene co-substrate
allowed convenient monitoring of micromolar peroxide concentrations in un-
treated samples.

2.4 Modified Electrodes

Chemical layers can also be used to enhance the performance of electrochemical
devices. The use of electrocatalytic surfaces can expand the scope of remote
electrodes to pollutants possessing slow electron-transfer kinetics. One example
of the adaptation of modified electrodes for a submersible operation is a remote
sensor for toxic hydrazine compounds, based an electropolymerized films of
3,4-dihydroxybenzadehyde (15). The low-potential detection accrued from this
catalytic action offers convenient measurements of micromolar hydrazine con-
centrations in untreated groundwater or river water samples.
We also developed a submersible probe based on a carbon-fiber working
electrode assembly, connected to a 50 ft-long shielded cable, for the continuous
monitoring of the 2,4,6-trinitortoluene (TNT) explosive in environmental matri-
ces (16). The facile reduction of the nitro moiety allowed convenient and
fast (1–2 s) square-wave voltammetric measurements of parts-per-million levels
of TNT.

3 SUBMERSIBLE ELECTROCHEMICAL ANALYZERS

The ability to perform metal–ligand complexation reactions on a cable platform,
in connection to adsorptive stripping measurements, has led to the development
of submersible electrochemical analyzers (17). As opposed to current in-situ
sensors (which lack sample preparatory steps, essential for optimal analytical
performance), the new on-cable automated microanalyzer will eventually incor-
porate all the steps of the analytical protocol into the submersible device. The new
“lab-on-cable” concept thus involves the combination of sampling, sample pre-
230 Wang

treatment, separation of components, and detection step (along with self-calibration)

into a single sealed submersible package. The first generation of this submersible
microlaboratory integrates microdialysis sampling, with reservoirs for the re-
agent, waste, and calibration/standard solution, along with the micromump and
necessary fluidic network on a cable platform (Figure 2). The sample and reagent
are thus brought together, mixed, and allowed to react in a reproducible manner.
Future generations will accommodate additional functions (e.g., preconcentration,
filtration, extraction) for addressing the needs of complex environmental samples.
Micromachining technology is being explored for further miniaturization and for
facilitating these in-situ sample manipulations. Proper attention is also being
given to the design of compact, low-powered, automated instrumentation for
unattended operation, “smart” data processing, and signal transmission (via
satellite links). Such a standalone “microlaboratory” can be submersed directly
in the environmental sample, to provide real-time continuous information on a
wide range of priority pollutants. The ability to perform in-situ all the necessary

FIGURE 2 Schematic diagram of the electrochemical “lab-on-cable” system:

(A) cable connection; (B) micropump; (C) reservoirs for reagent and waste
solutions; (D) microdialysis sampling tube and an electrochemical flow
detector.
Monitoring In-Situ Electrochemical Sensors 231

steps of the analytical protocol should have an enormous impact on pollution

control and prevention. Preliminary examples based on stripping monitoring of
trace metals or in-situ biosensing of phenols and various enzyme inhibitors are
presented below.
For example, we demonstrated the utility of the “lab-on-cable” probe for
circumventing in-situ problems common to electrochemical stripping analysis
(18). In particular, an internal delivery of an appropriate solution, containing a
ligand, third element, or a conducting salt, was used to minimize errors due to
overlapping peaks, intermetallic compounds, or ohmic distortions, respectively.
Similarly, internal delivery of a strong acid was used for on-cable release of the
metals from “collected” metal complexes, as desired for in-situ monitoring of
the total metal content (19).
We also developed a submersible phenol analyzer, based on an enzymatic
(tyrosinase) bioassay (17). The assay involved microdialysis sampling of the
phenolic compounds, their mixing with the internally delivered tyrosinase solu-
tion, and amperometric monitoring of the quinone product. The internal buffer
solution assured independence of sample conditions such as pH or ionic strength
[which commonly influence the performance of remote biosensors (12)]. Another
enzymatic assay was developed for the in-situ monitoring of cyanide (20). Such
an enzyme inhibition assay relied on the internal delivery of tyrosinase and its
catechol substrate using a flow-injection manifold. The flow probe thus addressed
the challenges to in-situ enzyme-inhibition devices (e.g., the replacement of the
inhibited enzyme and of the consumed substrate).
We envision the integration of multiple techniques and assays onto a single
cable platform, i.e., a complete submersible laboratory. Eventually, we expect to
eliminate the cable platform, and to use microlaboratories on miniaturized boats
or submarines, which would travel across the water stream and provide the
desired spatial and temporal information on target contaminants.

4 CONCLUSIONS AND FUTURE PROSPECTS

Electrochemical sensor technology is still limited in scope and cannot address all
environmental monitoring needs, yet a vast array of electrochemical devices has
been developed in recent years for in-situ monitoring numerous organic and
inorganic pollutants. By providing a fast return of the analytical information in a
timely, safe, and cost-effective fashion, the new, remotely deployed probes would
offer direct and reliable assessment of the fate and gradient of contaminants sites,
while greatly reducing the huge analytical costs. While the concept of “lab-on-
cable” is still at infancy, such a strategy should revolutionize the way of
monitoring priority pollutants, and would have a major impact on field analytical
chemistry. Ongoing commercialization efforts, coupled with regulatory accep-
232 Wang

tance, should lead to the translation of these research efforts into large-scale
environmental applications.

ACKNOWLEDGMENT
This work was supported by the U.S. Department of Energy Environmental
Managenemt Science Program (grant DE-FG07-96ER62306) and by the DOE-
WERC program.

REFERENCES
1. J. Wang, D. Larson, N. Foster, S. Armalis, J. Lu, X. Rongrong, K. Olsen, and
A. Zirino, Anal. Chem., vol. 67, pp. 1481–XXXX, 1995.
2. J. Wang, B. Tian, J. Lu, J. Wang, D. Luo, and D. MacDonald, Electroanalysis, vol.
10, pp. 399–402, 1998.
3. J. Wang, Stripping Analysis. VCH Publishers, New York, 1985.
4. M. Tercier and J. Buffle, Electroanalysis, vol. 5, pp. 187–200, 1993.
5. J. Wang, N. Foster, S. Armalis, D. Larson, A. Zirino, and K. Olsen, Anal. Chim. Acta,
vol. 310, pp. 223–231, 1995.
6. S. Daniele, J. Wang, and J. Lu, Analyst, in press.
7. J. Wang, J. Lu, D. Luo, J. Wang, M. Jian, B. Tian, and K. Olsen, Anal. Chem., vol.
69, pp. 2640–2645, 1997.
8. J. Wang, J. Wang, J. Tian, B., D. MacDonald, and K. Olsen, Analyst, vol. 124,
pp. 349–352, 1999.
9. J. Herdan, R. Feeney, S. Kounaves, A. Flannery, C. Storment, and C. Kovacs.
Environ. Sci. Technol., vol. 32, pp. 131–137, 1998.
10. M. L. Tercier and J. Buffle, Anal. Chem., vol. 68, pp. 3670–3678, 1996.
11. J. Wang, Electroanalysis, vol. 3, pp. 255–259, 1991.
12. J. Wang and Q. Chen, Anal. Chim. Acta, vol. 312, pp. 39–45, 1995.
13. J. Wang, L. Chen, A. Mulchandani, P. Mulchandani, and W. Chen, Electroanalysis,
vol. 11, pp. 866–869, 1999.
14. J. Wang, G. Cepria, and Q. Chen, Electroanalysis, vol. 8, pp. 124–127, 1996.
15. J. Wang, Q. Chen, and G. Cerpia, Talanta, vol. 43, pp. 1387–1391, 1996.
16. J. Wang, R. K. Bhada, J. Lu, and D. MacDonald, Anal. Chim. Acta, vol. 361,
pp. 85–91, 1998.
17. J. Wang , J. Lu, B. Tian, S. Ly, M. Vuki, W. Adeniyi, and R. Armennderiz, Anal.
Chem., in press.
18. J. Wang, J. Lu, D. MacDonald, and M. Augelli, Fresenius Z. Anal. Chem., vol. 364,
pp. 28–31, 1999.
19. J. Wang, J. Wang, J. Lu, B. Tian, D. MacDonald, and K. Olsen, Analyst, vol. 124,
pp. 349–352, 1999.
20. J. Wang, B. Tian, D. MacDonald, J. Wang, and D. Luo, Electroanalysis, vol. 10,
pp. 1034–1037, 1998.
13
Using Roadmaps in Pollution Prevention:
The Los Alamos Model

Thomas P. Starke
Los Alamos National Laboratory, Los Alamos, New Mexico

James H. Scott
Abaxial Technologies, Los Alamos, New Mexico

1 INTRODUCTION
Roadmapping is a powerful technique for displaying the structural relationships
among science, technology, applications, and results of applications. Because
they can incorporate complex, multiple relationships, they are used to display the
possible paths from the present state to a desired end state. Well-constructed,
comprehensive roadmaps are used for science and technology management,
including strategic planning, evaluating cost/risk, and program execution; for
enhancing communications among researchers, technologists, managers, and
stakeholders; for identifying deficiencies and opportunities in science and tech-
nology programs; and for identifying obstacles to achieving a desired end state.
There are several roadmap methodologies in use today, including forecast roadmaps,
retrospective roadmaps, and process evaluation roadmaps. Because roadmapping
methodology is so flexible, it can be used in many applications; it is frequently
used for process evaluation, technology forecasting, and for defining investment

233
234 Starke and Scott

strategies. Roadmaps have been used successfully by the U.S. Department of

Defense (DoD), the semiconductor industry, and various manufacturing concerns.
The starting point for all roadmapping methodologies consist of a defined
current state and a very well-defined desired end state; in general, one cannot
have a high-quality map without a carefully and comprehensively defined end
state. A complex project or process will have a number of intermediate states or
goals between the current state and the desired end state. The roadmap itself
consists of a network of nodes representing activities, events, or processes. Nodes
can contain a variety of information, depending on the purpose of the node. Nodes
are linked by actions. The network of nodes and links ideally represents all
pathways from the current state to the desired end state in such a way that
schedule, cost, and technical risk can be evaluated along each pathway. Analysis
of high-quality maps can help evaluate options relative to risk, cost, and schedule;
define deficiencies in current programs; and identify opportunities.

2 ROADMAP METHODOLOGY
In 1997 the Environmental Stewardship Office (ESO) at Los Alamos National
Laboratory decided to prepare a roadmap for reaching the laboratory pollution
prevention goal of substantially eliminating waste generation and pollutant re-
lease by the year 2010. The purpose of the roadmap was to identify:
Areas in which waste minimization and pollution prevention would have
the greatest impact
Options for preventing pollution or minimizing waste in those areas
Costs, technical risk, time, and return on investment associated with im-
plementing those options
The most cost-effective strategies for reaching the goal of substantially
eliminating pollution and waste resulting from laboratory operations
In order to prepare this roadmap, ESO chose a methodology that is based
on technology roadmap principles developed by the Office of Naval Research and
widely used in the DoD community (1). This methodology was modified by Los
Alamos to incorporate the principals of process mapping developed by Robert
Pojasek (2). The resulting methodology produces a roadmap with very broad
scope but sufficient detail to allow identification of specific sources of pollution
and waste and, consequently, specific remedial action options.

3 ROADMAP CONSTRUCTION
The DoD roadmap methodology is hierarchical and proceeds through a series of
submaps or map elements from general to specific. Thus, the roadmap is made
up of several levels, with the higher levels being more general and less detailed.
Using Roadmaps in Pollution Prevention 235

The highest level contains only the definition of the desired goal or end state, the
overall strategy for achieving that goal, and the definition of the waste types to
be considered in the roadmap. Normally, the highest-level map element is called
the zero level, or the mission-level map element. This element is comprehensive
in that it identifies the current condition and lays the foundation for the succeed-
ing map elements. In the zero level map element the waste types from any
particular set of operations are defined.
Level one maps take the waste types defined at level zero and develop
process flow diagrams for each waste type. A process flow diagram is an overview
of the process that generates the waste. Process flow diagrams provide a summary
of the processes and activities that result in the generation of waste. These
diagrams are used to decompose each waste type into specific waste streams. For
example, a waste type may be sanitary waste, and waste streams within that type
may be food waste, paper, and glass.
At level two, process diagrams are developed for each waste steam within
a waste type. These diagrams depict the process flow at a greater level of detail.
In addition to these waste stream process diagrams, new or modified procedures,
processes, or technologies are identified which may reduce or eliminate the waste
stream. The point in the process flow where the new technology can be deployed
is identified, along with the likely impact of deployment.
For some high-priority waste streams, further detail is provided in a third
level, including assessment of various options. The hierarchical structure de-
scribed above is shown schematically in Figure 1.
As an example of roadmap structure and how the roadmap can be used,
consider a path through Figure 1. The mission-level map element defines the N
waste types. These waste types could be sanitary, hazardous, liquid effluent, or
many others, depending on the nature of the operations at level zero.
At level one we define each waste stream within a waste type. In the
example, we have associated five waste streams with the second waste type.
The other waste types also have associated streams, not shown here for
simplicity’s sake.
At level two, a process map element is constructed to describe the processes
that produce each waste stream. An adjunct to the process flow map element is
the definition of procedure, process, or technology options for treating the object.
The likely impact of each option is then described. Technical risk, schedule risk,
cost, and health and safety impacts are assessed.
For high-priority or complex waste streams the options identified in the
process flow map element are broken down in further detail, and a series of issues
and attributes is developed to aid in comparing options. To clarify the construction
process, we will show how each of the map elements at various levels is
constructed.
236 Starke and Scott

Level Zero Mission Level Mission Level Map Element

Map Element

Type 1 Type 2 Type 2 ….. Type N Waste Defined

Waste Waste Waste Waste

Level One Waste Stream Defined

Map Element For each Waste Type

Waste Type 2 Waste Type 2 Waste Type 2

Stream 1 Stream 2 Stream 3
Waste Type 2 Waste Type 2
Stream 4 Stream 5

Process flow defined for each Process Flow

Level Two Waste stream Possible Sol'tns
Map Element
For each process flow define waste minimization or 1 2 3 4
pollution prevention actions defined

Option 1 Option 2 Option 3 Option 4

Level Three Element Element Element Element
Map Element

In some cases, solution options detailed with issues or further mapped

FIGURE 1 Roadmap hierarchy example.

3.1 Example: Los Alamos Environmental

Stewardship Roadmap
To illustrate the techniques used in construction of the roadmap elements we will
follow the specific path for sanitary waste through the Los Alamos Environmental
Stewardship roadmap. The techniques can and should be generalized to other
applications. We will start with a conceptual mission-level map for Los Alamos
National Laboratory.
3.1.1 Level Zero or Mission-Level Map Element
Construct mission-level map element.
Define waste types.
Using Roadmaps in Pollution Prevention 237

This mission level map is constructed to represent the laboratory as a system with
a series of material and energy flows, both into and out of the system. The first
step in constructing the mission-level map is to decide the scope of the initial
system. In this case, the system is the entire laboratory site. We can also choose to
examine a smaller subset if we wish to focus on a particular area. Figure 2 shows the
laboratory process map, which is a view of the laboratory from the local environmen-
tal perspective. The perspective can be important. If we had chosen a regional
perspective, the resulting roadmap would have been quite different.
The map element is constructed by identifying inflows of materials and
energy to the system, identifying the operations that use the materials and energy,
and identifying system outflows, including all the products of the operations
including wastes and pollutants. The wastes are accrued into a number of broad
waste types. This is a critical step since it will form, in many cases, the foundation
for all subsequent analysis. The waste types must be comprehensive and include
all wastes generated from operations.
The laboratory performs work for government sponsors and private indus-
try. In performing this work, the laboratory procures services, materials, equip-
ment, new facilities, and commodities (electricity and natural gas). The laboratory
also takes in water from the regional aquifer and air from the surrounding
atmosphere. This series of inflows is shown at the left in Figure 2. Once in the
laboratory, the inflows are used in the six different kinds of operations listed in
Figure 2.
Most person-hours are spent conducting office operations. These involve
office space, furniture, information processing equipment, paper, and office

Emissions

Products
Los Alamos National Laboratory
Materials
• Office Operations
• Experimental Operations TRU Waste
Power MLLW
• Production Operations
Hazardous Waste
• Maintenance and
Infrastructure Operations Excess
Property
Water • Construction
• Environmental Restoration

Effluents Ecosystem Impact,

LLW & Sanitary Disposal

FIGURE 2 Laboratory process map.

238 Starke and Scott

supplies. Energy is expended to operate equipment and provide climate control.

Water is used in evaporative cooling to transfer waste heat to the atmosphere.
Experimental operation includes bench-scale and large-scale research. En-
ergy is expended to operate the equipment and provide climate control. Water is
used in evaporative cooling to remove waste heat. Experimental operations
typically procure large amounts of equipment but small amounts of chemicals and
other materials.
Production operations include all the site production operations. Production
operations consume material, water, and energy, but in this particular case energy
and water usage is modest.
Maintenance and infrastructure operations include all maintenance activi-
ties, facility management activities, and site-wide infrastructure systems, such as
the sanitary wastewater plant, on-site power plant, water influent system, and
highway system. These operations consume large quantities of chemicals and
produce most of the site’s hazardous waste. They also consume significant
amounts of energy and water.
Construction includes both smaller construction projects and major con-
struction projects. Construction operations are important not only as a source of
immediate environmental impact during construction activities; design decisions
made during the construction process can lock in environmental impact for the
lifetime of the facility.
Environmental remediation includes all remediation activities on the site.
For purposes of this roadmap, only newly generated wastes and pollutants were
considered, but that need not have been the case.
Because the products of the laboratory are mostly information, most
material inflows become by-product or waste outflows. Identified outflows of
waste and pollutants are divided into the eight categories shown in Figure 2.
These include transuranic waste (TRU), mixed low-level waste (MLLW), low-
level waste (LLW), hazardous waste, solid sanitary waste, excess property
entering the salvage system for reuse or recycle, gaseous emissions, and liq-
uid outfalls.
Another result of operation also occurs. The presence of laboratory facili-
ties, infrastructure, operations, and land management affects local ecosystems.
Much of this is unavoidable, and much of it is not necessarily harmful to the local
environment. This local ecosystem impact can be minimized through wise
operational choices.
Once the operations and outflows have been identified, a fundamental
choice must be made. The subsequent lower-level maps can be organized and
broken down according to either operation or waste type, depending on the
specific goals of the mapping activity. Roadmaps based on operations are
particularly good if one wishes to focus on organizational structure and its impact
on pollution and waste generation and may include issues such as structure,
Using Roadmaps in Pollution Prevention 239

funding, and customer base. Roadmaps based on waste type are generally more
useful for devising pollution prevention and waste minimization strategies and
choosing among technological alternatives. As an example, we will construct a
map based on waste type and follow a specific waste type—sanitary waste—
through the succeeding map elements. Janet Watson constructed the complete
ESO sanitary roadmap, from which this was abstracted.
3.1.2 Level One or Waste Stream Definition Level
Construct waste type process flow diagram.
Define waste streams.
Define issues and constraints.
Prioritize waste streams.
Since we have chosen to follow the development of roadmap elements through
the sanitary waste type, we first construct a sanitary waste process flow diagram.
This diagram is constructed using the same principles that are used in all process
flow diagrams: the inflows of materials and energy are identified, the process or
operations are identified, and the outflows of waste material are identified. At this
level, quantification becomes important, since it will be used to prioritize waste
streams for waste reduction activities. The first-level map element that emerges
from examination of the data is shown in Figure 3.
Nonhazardous, nonradioactive materials enter the laboratory as procured
items, mail, food, and various other substances such as glass, brush, and construc-
tion materials. These items are used by the laboratory and are either recycled,
reused, or salvaged, or are disposed in the county landfill. Materials disposed
include such items as construction waste, food and food-contaminated wastes,
paper products, glass, Styrofoam, and various other substances.
Material outflow pathways are shown at the right of the diagram. The
composition of materials in those pathways is broken down in the pie charts at
the bottom. It is important to carefully and completely identify the constituents
of the waste and to quantify the volumes, since that information will form the
foundation for the succeeding process flow diagrams.
In this case, we will examine the dumpster waste in greater detail. There
are a number of waste streams to be considered. As a normal part of the
roadmapping process, it is important to prioritize these waste streams for action
based on some criteria. The basic question here is: Which waste streams should
we attempt to minimize first? Other questions that need to be addressed are the
cost of minimization and the return on investment for minimization activities.
Before choosing criteria, it is necessary to examine all issues and con-
straints associated with the waste streams. Issues might involve such considera-
tions as lifetime of the landfill. If the landfill has a short lifetime and there are no
easy alternatives to disposal at the current landfill, an overriding consideration
240 Starke and Scott

Internal recycle/reuse

External
Procurement Laboratory Recycle
Mail
Laboratory Processes
Disposal
Food at landfill
Construction
Substances
External
Salvage

Internal Salvage

Recycle/reuse
Cardboard

Paper
Construction Glass
Metal
Wood
Styrofoam
Food Waste
Dumpsters Plastic Office Supplies

Sanitary Waste Dumpster Waste

FIGURE 3 Sanitary waste streams.

might be to minimize the volume of material sent to the landfill, even if the cost
savings for that action are small compared to other possible actions. Constraints
might involve such things as regulatory requirements or operating policy. Actions
designed to meet regulatory requirements will probably be placed high on the
priority list even if the costs are high and the return on investment is low. In many
pollution prevention and waste minimization programs, actions are prioritized
based on Pareto analysis. The underlying assumption of Pareto analysis, generally
surprisingly good, is that 80% of the waste comes from 20% of the operations
and that 20% is where you should concentrate your efforts at prevention. If the
operations are fully compliant with regulation and there are no overriding local
issues, Pareto analysis is a very good way to prioritize activities. In a classic
Pareto chart, the volumes of waste are plotted for each waste stream in a bar chart.
The streams that contain 80% of the waste are then identified. If there are no
overriding constraints, these are the streams that are selected for intervention,
usually either in the order of total waste quantity or total waste cost. Robert
Pojasek, in “Prioritizing P2 Alternatives” (Pollution Prevention Review, vol. 7,
Using Roadmaps in Pollution Prevention 241

no. 1, pp. 105–112, 1997), discusses Pareto analysis and its major variants
in detail.
At this point, it is usual to construct a table listing the waste streams, issues,
constraints, the cost of treatment, handling, and disposal, the cost of regulatory
support, and quantities for each stream. If the costs can be obtained on a unit
quantity basis, calculating the cost associated with each stream is straight-
forward. These factors all influence the final prioritization of the waste stream.
As stated previously, issues and constraints may outweigh the more obvious cost
arguments.
For this exercise we will not prioritize the sanitary waste streams but will
choose to examine a subset of the paper waste stream—mail—in detail.
3.1.3 Level Two or Waste Stream Process Definition
Construct waste stream process flow diagram.
Identify intervention points.
Identify prevention options.
The process flow diagram for the “mail” waste stream is constructed like the
previous process flow diagrams. However, at this level the operations are dia-
gramed explicitly and in some detail rather than being simply listed. This is
necessary because at this level we are trying to identify sources of waste and
potential intervention points. Often, physical inspection of the operation is
required to develop the necessary detail. For manufacturing or processing opera-
tions, these diagrams can become very complex. In this diagram, we have
diagramed all the major pathways for mail between receiving and final disposi-
tion. Each box or node identifies a process or handling step.
Every year the laboratory receives and distributes 714 MT of mail. This
mail includes junk mail, catalogs, telephone directories, and various documents,
as well as business mail. The mail received by the laboratory includes a small
amount of classified mail. The process flow diagram for the mail waste stream is
shown in Figure 4.
Mail, including internally generated mail, is received by the laboratory and
distributed. Any unwanted mail can be sent by the recipient to Mail Stop A1000
for sorting and recycle. Documents such as catalogs and directories that are
glue-bound must first have the bindings sheared off before the paper is recycled.
The bindings are sent to the landfill for disposal. Mail is also disposed by
discarding in green desk-side containers or trash bins. The contents of the green
containers are sent to recycle, while the contents of the trash bins are sorted for
recyclable materials at the Material Recovery Facility (MRF). Classified material
may not be disposed unless it has been security (crosscut) shredded. The strip-
shredded material can be recycled, but crosscut shredded material currently goes
to the landfill.
 242 Starke and Scott

LAB 3
Discard Green Recycle Sandia
- Distribute Bins Collection
1 internally
Mail 4
generated
Trash MRF- Dumpster Landfill
mail &
Bins Sort
Publications 2 publications
-Distribute Glue
Mail Shearing
External Bindings
Stop
A1000 Paper
Recycle Recycle

FIGURE 4 Mail and document distribution and disposal.

With the advent of MRF operations, the opportunity to recover nearly all
the discarded recyclable mail is realized. The emphasis will then be on reducing
the source of unwanted mail.
At this point, it is usual to identify possible intervention points. Examina-
tion of the process flow diagram shows limited opportunities to reduce the
quantity of mail going to the landfill. The strategy has to be either to prevent mail
from entering the laboratory or to increase the fraction of discarded mail that is
recycled. Eliminating all incoming mail is, of course, not an option, but there may
be ways to reduce particular incoming mail streams. Most notably, it is possible
to reduce the volume of “junk mail” and certain print documents such as paper
telephone books. This upstream approach has the advantage of preventing waste
so that it never has to be handled. At the disposal end of the diagram, the only
destinations for waste mail are recycle and the landfill, so if the quantity of
discarded mail going to the landfill is to be reduced, the recycle fraction must be
increased. Planning meetings with the personnel involved in recycle revealed two
promising avenues to pursue.
With the intervention points identified, a set of initiatives are formulated,
with the numbers keyed to process flow diagram.

Initiative 1: Reduction of “junk mail.” A substantial fraction of the mail

consists of recurring, unwanted “junk” mail. A centralized stop-mail
service for “junk mail” is currently in the pilot phase. Any laboratory
employee who wishes to request removal of his or her name from a
mailing list can use this service.
Initiative 2: Eliminate paper phonebooks. Paper phonebooks are widely
used and are difficult to recycle. US West directories, which are routinely
distributed to all employees, will be eliminated as a source of waste by
restricting delivery and asking employees to use the “on-line” directory
Using Roadmaps in Pollution Prevention 243

instead. Approximately 22 MT of waste per year can be avoided in this

way.
Initiative 3: Additional items in paper recycle system. The current paper
recycle program is limited to white and pastel paper; options for includ-
ing other types of paper products (mail items) in this mix are being
evaluated.
Initiative 4: Increase use of MS A1000. Although MS A1000 is widely used
as a means of recycling various materials, many employees are still
unaware of its existence. A publicity campaign will be developed to
increase awareness; self-inking stamps (with the A1000 logo) will also
be distributed to each mail stop within the laboratory to encourage use
of this program.
These initiatives must be analyzed, of course, according to a number of
relevant variables such as potential cost and effect on the waste stream. The
variables that describe each of the initiatives are called factors and are sometimes
tabulated in a spreadsheet. A very simple version of such a spreadsheet is shown
in Table 1. In most cases, a cost is assigned to each of the initiatives. In this case,
costs were not readily available or were being developed through pilot programs
at the time the roadmap was constructed.

3.2 Differentiating Among Technology Solutions at

Level Two
In the above example from the ESO roadmap, all the initiatives proposed were
programs that did not require new technologies. Frequently, this is not the case
and, in fact, some sets of initiatives feature competing technologies. Conceptually,
the process flow diagram for such a project might look like Figure 5. This chart
is representative of a many manufacturing processes, particularly those involving
distillation or refining. Following an initial process step, the product stream is
separated into two streams for further processing. Both of the streams have
subsequent processing steps. Assume that at point 1 there is an opportunity to
intervene and reduce the path B waste stream. Often in manufacturing processes,
this intervention will involve deployment of a new technology, and there may be
more than one candidate technology. In that case, the evaluation of the potential
initiatives will be very different from the previous case involving mail. Now we
are not dealing with administrative controls or modified processes alone, but must
consider the relative merits of the competing technologies.
Some of the factors that must be considered might include the relative
maturity of the technologies, which is usually understood to mean the scale of
prior operation—i.e., bench, pilot, or full scale—schedule for deployment, cost
to develop, cost to deploy, cost to operate, efficiency of the process or throughput,
quantity of waste avoided, nature of any secondary waste streams from a
 244
TABLE 1 Waste Minimization Initiatives
Initiative or project Action/milestone Status Funding source Waste avoided

Reduce “junk mail.” Develop a Evaluate pilot results; Ongoing Base program Source reduction,
centralized stop-mail service determine if results Pilot test: 4.4 MT/year
for “junk mail.” justify expense. completed
Eliminate paper phonebooks. Continue restricted de- Ongoing Base program Source reduction,
Delivery of US West Tele- livery in future years. 22 MT/year
phone directories is restricted;
employees are requested to
use the “on-line” directory
instead. Approximately 22 MT
of waste per year can be
avoided in this way.
Include additional items in This option is being Not funded Increased recycle
paper recycle system. Include evaluated.
other paper products (mail
items) in the program.
Increase use of MS A1000. A publicity campaign Ongoing Base program Increased recycle
Although MS A1000 is widely will be developed to
used as a means of recycling increase awareness.
various materials, many em- Self-inking stamps
ployees are still unaware of (with the A1000 logo)
its existence. This program will also be distrib-
within the laboratory will en- uted to each mail
courage use of this program. stop.
Starke and Scott
Using Roadmaps in Pollution Prevention 245

Path A
Path A Waste
Inflows
Process 2 Stream
Process 1 Process A1
Segregate

1 Path B Waste
Stream
Process B1 Process B2

Path B

FIGURE 5 Conceptual technology process map.

particular process, and safety of the processes. A matrix similar to the one shown
previously can be constructed and weights assigned to each of the factors. Issues
frequently arise when trying to determine weights for the particular factors, and
it is best to agree early about any constraints that must be applied. Typical
constraints include the stipulation that the chosen process must be at least as safe
and efficient as the process it will replace. Once weights are assigned to the
various factors, the roadmapping team must meet with the technology advocates
and the operations personnel to quantify the factors.
Since each technology is likely to have advocates and detractors, it is
important to gather information on each technology from all concerned parties,
including operators. Even then, it may be impossible to reach a consensus view
with respect to all the relevant factors. For this reason, it is important to decide
in advance how conflicts will be resolved. Normally, the roadmapping team
resolves conflicts after gathering information from the technology advocates.
After the factors have been quantified, one of several algorithms can be used to
evaluate each of the competitive technologies. In this way technologies can be
differentiated with regard to deployment in a particular process step and a basis
for an action decision is established.

4 USING THE COMPLETED ROADMAP

To review briefly:
At level zero, the overall system operation was mapped and waste types
were identified. Frequently this step is left out if waste types are well known or
if only one waste type is of interest.
At level one, each of the waste types was broken down into waste streams.
The size and nature of the waste streams was quantified and the waste streams
were prioritized for minimization or prevention action.
246 Starke and Scott

At level two, detailed process maps for the waste streams were prepared,
points of intervention were defined, and initiatives for minimization or prevention
at these points were identified. Data were prepared for each of the initiatives to
form a foundation for decision making.
At this point, a number of paths forward are possible. The zero level and
level one maps are useful for many purposes, including education, training, and
monitoring. The level two maps are normally used to enhance decision making
and monitoring progress.
Part of the decision-making process involves developing an investment
strategy. An investment strategy involves four items:
1. A decision about priorities and which waste streams should be ad-
dressed first with respect to minimization or prevention
2. A decision about which initiatives should be pursued first for the
high-priority waste streams
3. An allocation of resources against the selected initiatives
4. Development of a fallback or contingency position for the initiatives,
particularly those that require development and/or deployment of new
technologies
Finally, a schedule for implementing the initiatives is developed and
overlaid on the process map.
The schedule is normally prepared by redrawing the process map to
represent the end state that will result from the implementation of selected
initiatives. The redrawn map element includes an earliest start/latest finish date
in the appropriate process nodes. A project control chart is frequently included as
part of the revised process flow chart. The project control chart can include many
or few schedule and control parameters such as start date, finish date, cost, and
any other desired parameters. The redrawn process flow chart shown in Figure 5
would then look like Figure 6.
Clearly, if there are several initiatives in the same waste stream, the
roadmap element can become complicated. In that case, it is usually easier to
redraw a revised map element for each initiative so that the complete data on each
initiative in a particular waste stream are located on its own map element. The
redrawn map elements can be retained in one location for ease of review.
In addition, some roadmap developers include risk as part of the revised
map element. The risk may be technical risk, programmatic risk, cost risk, or
funding risk. The risk is usually specified as the risk of not being able to move
successfully from one process node to the next. The risk is then associated with
the link between nodes and aggregated along all pathways in the revised map
element. In this way, risk to the project can be assessed, the sources of greatest
risk can be identified, and contingency plans can be developed for those areas.
 Path A
Inflows Path A Waste
Process 2 Stream
Process 1 Process A1
Segregate

Path B Waste
Earliest start Stream

Modified Process B2
Process B1
Using Roadmaps in Pollution Prevention

Latest finish
Path B

Activity A Start Finish Man-hours Cost ETC…

Activity B
.
.
.
Activity N

FIGURE 6 Redrawn process map element for project control.

247
248 Starke and Scott

Estimation of risk is necessarily subjective and cannot be taken too literally.

The risk estimates serve simply as a guide to controlling risk.
Planning is a dynamic activity. Since pollution prevention operations
change, hopefully in response to good planning, it is necessary to update the
roadmaps periodically. The usual period for updates is yearly, but this can be
adjusted to reflect the actual rate of changes in the system.

5 CONCLUSIONS
Roadmaps are useful tools for systematically evaluating the generation of waste
and pollution in virtually any type of operation, large or small. For large systems
like Los Alamos National Laboratory, the roadmap can be extensive. The Los
Alamos ESO roadmap can be found online at http://emso.lanl.gov/publications.
Roadmaps provide a mechanism for evaluating the current state in detail,
for deciding how to move toward a desired end state, for assessing the effective-
ness of alternative options in moving toward the end state, for making investment
decisions, and for controlling risk. More detailed information on the various
aspects of roadmapping, as applied by a variety of institutions and industries, can
be found in the bibliography that follows.

SELECTED BIBLIOGRAPHY
The following bibliography presents further information on roadmap construction
and use and contains examples of different types of roadmaps. The Kostoff
citation contains an exhaustive bibliography.
Aerospace Industries Association of America, Detailed Technology Road-
map for Superconductivity. Washington, DC: AIAA, Superconductivity
Committee, 1992.
D. Barker, and D. Smith, Technology Foresight Using Roadmaps. Long
Range Planning, vol. 28, no. 2, pp. 21–29, 1995.
Electronic Industry Environmental Roadmap, available from MCC Corpo-
ration, 3500 West Balcones Center Drive, Austin, TX 78759, 1998.
M. P. Espenschied, Graphical Status Monitoring System for Project Man-
agers. Pretoria, South Africa: National Institute for Aeronautics and
Systems Technology, Funder: National Aeronautics and Space Adminis-
tration, Washington, DC, Report CSIRNIAST817, 1981.
J. H. Gurtcheff, US Strategic Nuclear Strategy and Forces: A Roadmap for
the Year 2000. Study Project. Carlisle Barracks, PA: Army War College,
1991.
R. N. Kostoff, Science and Technology Roadmaps, http://www.dtic.mil/
dtic/kostoff/Mapweb2I.html.
Using Roadmaps in Pollution Prevention 249

ORNL, Oak Ridge National Laboratory Technology Logic Diagram. Vol-

ume 1, Technology Evaluation: Part A, Decontamination and Decom-
missioning. Oak Ridge K-25 Site, TN, Report ORNLM2751V1PTA,
1993.
R. B. Pojasek, P2 Programs, Plans and Projects: Some Thoughts on Making
Them Work. Pollution Prevention Review, vol. 9, no. 2, 1999.
U.S. Department of Energy, National TRU Waste Management Plan, DOE/
NTP-96-1204, Revision 1, 1997.

REFERENCES
1. R. N. Kostoff, Science and Technology Roadmaps, http://www.dtic.mil/dtic/kostoff/
Mapweb2I.html.
2. R. B. Pojasek, P2 programs, Plans and Projects: Some Thoughts on Making Them
Work. Pollution Prevention Review, vol. 9, no. 2, 1999.
14
Pollution Prevention and DFE

Terrence J. McManus
Intel Corporation, Chandler, Arizona

1 BASIC PRINCIPLES OF POLLUTION PREVENTION AND

WASTE MINIMIZATION
Beginning in the mid-1970s, environmental management of industrial air emis-
sions and wastewater discharges focused on end-of-the-pipe or end-of-the-stack
treatment technologies. Both the Clean Air Act of 1970 and the Federal Water
Pollution Control Act of 1972 (now called the “Clean Water Act”), as well as the
parallel regulatory structures set up at state and local levels, required new
treatment technologies to be developed to manage air emissions and wastewater
discharges. But none of these early statutes and regulations mandated that
corporations minimize the amount of waste generated or prevent pollution during
manufacturing.
With the passage of the Resource Conservation and Recovery Act (RCRA)
in 1976, the government for the first time defined “hazardous waste” and began
to focus on waste minimization, rather than just waste treatment. Large-quantity
generators [producing more than 1000 kg (2200 lb) per month of hazardous
waste] were required to ship waste to an approved treatment, storage, and disposal
facility (TSDF), using a formal document known as a waste manifest. Because
the new regulations were very strict, however, many off-site TSDFs had to close
down, resulting in a sharp decrease in the supply of such facilities.
251
252 McManus

To reduce demand for the facilities, beyond the sharp rise in costs for TSDF
services, Section 3000(b) of the RCRA requires that large-quantity generators
who transport waste off-site must certify on the manifest that they have estab-
lished a “program in place” to reduce the volume or quantity and toxicity of
hazardous waste generated—to the extent economically practicable. For owners/
operators who manage hazardous waste on-site in a permitted TSDF, Section
3005(h) similarly requires annual certification that a waste minimization program
be in place and maintained in the facility’s operating records. These two require-
ments put the burden of proof on generators or an owner/operators of TSDFs to
show that they are implementing waste minimization strategies.
Small-quantity generators, who produce between 100 and 1000 kg per
month of hazardous waste, are required to certify on their hazardous waste
manifests that they have also “made a good faith effort to minimize” their waste
generators (51 FR 35190; October 1, 1986).
Together, the large- and small-quantity generator requirements for waste
minimization affect more than 95% of the hazardous waste generated in the
United States. The primary mechanism for achieving such minimization is to
identify the various hazardous waste streams and determine if it is possible to
reduce the volume and/or toxicity of each (1).
The U.S. Environmental Protection Agency (EPA) also collects data, annu-
ally, on the emissions and disposal of a specific list of chemical compounds.
Manufacturers who exceed certain thresholds have to inform the EPA as to
whether the chemicals were released into the environment (air, water, or land) or
transferred to another facility for management. The EPA, in turn, maintains a
database known as the Toxics Release Inventory (TRI), which is one of the best
data sources to review emissions performance on an industry sector basis. The
first year for data reporting to EPA’s TRI inventory was 1987. The database tends
to be about two years behind in its reporting, however, as the reports are not due
until July, and loading and analyzing the data takes about a year.

2 ROLE OF POLLUTION PREVENTION AND DESIGN FOR

THE ENVIRONMENT
As methodologies for waste minimization improved in the 1980s, industries
looked to more comprehensive approaches, such as pollution prevention (P2)
and design for the environment (DFE). In 1990, the U.S. Congress passed
the Pollution Prevention Act, which specifically required the evaluation of
new opportunities and approaches to eliminate the generation of emissions
and waste.
Under Section 6602(b) of the Pollution Prevention Act of 1990, Congress
established a policy that:
Pollution Prevention and DFE 253

Pollution should be prevented or reduced at the source wherever feasible.

Pollution that cannot be prevented should be recycled in an environmentally
safe manner, wherever feasible.
Disposal and/or release into the environment should be employed only as
a last resort and should be conducted in an environmentally safe manner.

The EPA established an operating definition for P2 as part of the agency’s

1991 Pollution Prevention Strategy. That definition makes clear that prevention
is the first priority within an environmental management hierarchy, which
includes:

1. Prevention
2. Recycling
3. Treatment
4. Disposal or release

The EPA also recognized that any P2 strategy needs to be flexible. Any P2
option today, in fact, depends on three factors: legal requirements, levels of risk
or toxicity reduction that can be achieved, and cost.
As with waste minimization, P2 typically focuses on existing manufactur-
ing processes, by applying the prevention hierarchy to the various waste streams.
When new manufacturing processes are developed, some corporations apply new
environmental management techniques to reduce/eliminate waste generation as
part of their manufacturing process design. This approach has been commonly
called design for the environment (DFE).
Different people have defined DFE in different ways. For instance, the EPA
defines a DFE program as “a voluntary partnership-based program that works
directly with companies to integrate health and environmental consideration in
business decisions (2). Intel Corporation has defined it as “a methodology to
develop environmentally compatible products and processes, while maintaining
desirable product price/performance and quality characteristics.”

3 ENVIRONMENTAL FRAMEWORK
How do all these environmental components or programs work together to form
a unified environmental management system (EMS)? Figure 1 presents a concep-
tual model of the environmental framework. This framework also demonstrates
the evolution of environmental management over time, with waste treatment
beginning at the center, as the earliest management technique, and current and
future management approaches extending from there.
Indeed, waste treatment is the fundamental environmental management
technology applied over many decades. The progression to each succeeding
 254
SUSTAINABLE DEVELOPMENT

POLLUTION
PREVENTION

DESIGN
FOR THE WASTE
ENVIRONMENT MINIMIZATION

WASTE
TREATMENT

Raw Materials
WASTE
Process Design

Product Design
McManus

FIGURE 1 Environmental management evolution.

Pollution Prevention and DFE 255

environmental management approach highlights the fact that both the number of
choices and their scope and complexity are continually increasing.

4 SUSTAINABLE DEVELOPMENT
At the current outer edge of the framework is sustainable development. One way
to achieve sustainable development is for many companies and their local
communities to each adopt DFE strategies. In other words, sustainable develop-
ment is the integration of many DFE programs, from many different entities, over
a large geographic area or region. Such a unified approach is necessary because
a single entity cannot provide all the components necessary to prevent pollution
and recycle all recyclable materials. For example, a semiconductor manufacturing
facility can establish programs to collect and recycle aluminum cans, paper
materials, and chemicals. However, the company must rely on the aluminum
industry, the paper recycle industry, and the chemical producers also to implement
recycling in order to use the necessary recycle technology and/or effective and
efficient economies of scale. When a region can develop such an integrated
approach to the environment, the details can be presented in a “green plan.”
In June 1993, President Clinton formed the President’s Council on Sustain-
able Development (PCSD) to develop and recommend a national strategy for
implementing sustainable development. This council consisted of leaders from
industry, government, nonprofit organizations, and Native American groups. In
1996, the PCSD published the report, Sustainable America, which contained the
following vision statement:
Our vision is of a life sustaining earth. We are committed to the
achievement of a dignified, peaceful and equitable existence. A sustain-
able United States will have a growing economy that provides equitable
opportunities for satisfying livelihoods and a safe, healthy, high quality
of life for current and future generations. Our nation will protect its
environment, its natural resource base and the functions and viability of
natural systems upon which all life depends (3, p. IV).
In support of this vision, the Council also recorded 16 beliefs that set the
basis for implementing the strategy. The following four beliefs (3, pp. v–vi) refer
specifically to industrial development:
To achieve our vision of sustainable development, some things must
grow—jobs, productivity, wages, capital and savings, profits, informa-
tion, knowledge, and education—and others—pollution, waste, and pov-
erty must not.
The United States made great progress in protecting the environment in the
last 25 years and must continue to make progress in the next 25 years.
256 McManus

We can achieve that goal because market incentives and the power of the
consumers can lead to significant improvements in environmental per-
formance at less cost.
Environmental progress will depend on individual, institutional, and corpo-
rate responsibility, commitment and stewardship.
Steady advances in science and technology are essential to help improve
economic efficiency, protect and restore natural systems, and modify
consumption patterns.
In May 1999, the PCSD published a second progress report, entitled
Towards a Sustainable America. As part of this effort, the council focused
specifically on the following issues (4, p. 3).
1. Policies to reduce greenhouse gas emissions
2. The next steps in building the new environmental management system
for the twenty-first century
3. Policies and approaches to build partnerships to strengthen communi-
ties
4. Policies to foster U.S. leadership in international sustainable develop-
ment policy, particularly in international capital flow
In the area of environmental management, it was very clear to the PCSD
that most recent environmental reforms have not really focused on the objective
of promoting sustainable development. This is partly because current definitions
of environmental pollution, management, and protection are too narrowly scoped,
with significant emphasis on point source emissions. Therefore, the solutions tend
to be focused on single pollutants within a single media from a single source.
Very little effort has been focused on aggregating and understanding environmen-
tal risks and impacts across a broader ecosystem basis. A similar conclusion was
reached by the EPA Science Advisory Board’s (SAB) Integrated Risk Project. In
essence, the SAB believes that environmental management efforts to date have
typically worked on targeted pollutants from single sources, and resulted in
improvements in environmental performance over very localized areas. Specific-
ally, SAB stated that the effort must be more holistic:
Concern for the environment has become an important part of the
American value system. We care about the environment as it relates to
human health, the viability of ecosystems, and our children’s future. We
care about the quality of life, today and in the future, and in the
interconnected environmental conditions that play such an important
role in determining life’s quality (5).
Ultimately, the vision and environmental goals must be to protect the
overall health of all people and the long-term viability of whole ecosystems. That
 Pollution Prevention and DFE 257

means that risk reduction must be designed to control more than one pollutant at
a time, to protect more than one human or ecological receptor, and thus to realize
broader benefits of environmental improvement, at a lower cost.
The model for such integrated environmental decision making is presented
in Figure 2. The three major components for the decision-making framework are
Problem formulation
Analysis and decisionmaking
Implementation and performance evaluation
Each of these steps requires constant feedback of information to both improve
and optimize the specific measures implemented, yet some of the advantages of
using an integrated decision-making model include:
An increase in the probability of focusing on the highest risks/most impor-
tant issues
A methodology that includes both human health and ecological risks

Report Card
Has Problem
PROBLEM FORMULATION Changed?
Information
Expert Judgement Risk Comparison Goal Setting
Public Values Preliminary Options Analysis
Deliberations

Report Card
Meeting Objectives?

ANALYSIS & DECISION-MAKING

Legal and
Institutional Milieu Risk Assessment Screening/Selection
Options Analysis Deliberation
Performance Measures

IMPLEMENTATION and
PERFORMANCE EVALUATION

Implementation Monitoring
Reporting and Evaluation

Source: Integrated Environmental Decision-making

in the 21st Century - EPA/SAB

FIGURE 2 Integrated environmental decision-making framework.

258 McManus

Improvements in public accountability (because the framework is applied

typically over a larger geographic area)
The inclusion of specific performance measurements
One of the biggest stumbling blocks to integrated environmental decision
making is the fact that this approach is a marked departure from the current
methodology for environmental management. That means it takes more time, in
the beginning, to communicate with and educate the public so that they are
willing to participate.

5 COMMON FOCUS AREAS AND DRIVERS FOR P2

AND DFE
Most industrial operations are committed to improvements in environmental
performance and natural resource protection. The broad categories that industries
focus on and try to manage are
Solid and hazardous waste generation
Chemical use
Air emissions
Water use
Wastewater discharge quantity and quality
Electrical use
Each category includes specific manufacturing operations that use natural re-
sources or produce emissions and that therefore need the corporation’s attention.
Some of the reasons that corporations manage these processes in environmentally
efficient ways are that they:
Reduce operating costs
Ensure compliance with environmental permits
Satisfy certification requirements of RCRA on waste manifests
Achieve a specific threshold, which eliminates reporting requirements or
allows the corporation to achieve minor source status under an air
emissions permit
Satisfy corporate commitments established in an environmental policy or a
specific commitment to a local community
Ensure that emissions reporting under TRI demonstrates reduction from
previous year’s emission reporting data
Help the corporation develop a reputation as a “good corporate citizen”
The following example for waste solvent generation illustrates this ap-
proach for an individual manufacturing facility. The basic methodology for
implementing either pollution prevention or design for the environment includes:
Pollution Prevention and DFE 259

1. Identify which areas/manufacturing operations generate emissions and/

or consume natural resources (waste solvent generation is selected for
this example).
2. Within the specific operation under evaluation, identify the various
sources of waste solvent that are generated and identify the quantities
per unit of time, quantities per unit of production, and characteristics
of the waste solvent source(s).
3. Develop a list of options to recycle, reduce the volume, eliminate, or
reduce the toxicity of the waste solvent that is generated.
4. Evaluate each potential option, using the following criteria as a minimum:
Capital cost to implement
Operating costs per year
Quantities of waste solvent reduced or eliminated
Reduction in toxicity
Environmental impacts of the remaining materials/waste
Health and safety advantages and disadvantages for implementing the
proposed option
5. Select the best option for each solvent waste stream and implement.
6. Track performance of the implemented option and compare the data
with original projections.
7. Reassess the option implemented and determine whether further im-
provements can be implemented, using the same basic methodology
again.

6 INDUSTRY SECTOR APPROACH TO P2 AND DFE

Inevitably, each industrial manufacturing operation must determine how it will
specifically implement its environmental management system and determine the
proper leverage point for applying pollution prevention or design for the environ-
ment strategies. Within a specific industrial manufacturing sector, however,
similarities in the manufacturing processes and chemistries frequently result in
similarities in types of waste or emissions generated. Evaluations of P2 and/or
DFE opportunities could be realized more efficiently by establishing specific
design criteria or environmental performance goals at a sector level, rather than
on a company or individual manufacturing facility basis. Both the semiconductor
and metal finishing industries have embarked upon sector-wide approaches.

6.1 Semiconductor Manufacturing

More than three decades ago, Gordon Moore of Intel Corporation predicted that
the number of transistors in a defined area of silicon would double every 18
months. This prediction, known as Moore’s law, is presented in Figure 3. To keep
260 McManus

Transistors
Per Die
108 256M
Memory
Microprocessor 64M
107 16M Pentium® III
4M Pentium® II
106 1M Pentium® Pro
256K Pentium®
105 64K i486™
i386™
104 4K 16K 80286
1K 8086
103 8080
4004
102

101

100
’70 ’73 ’76 ’79 ’82 ’85 ’88 ’91 ’94 '97 2000
Source: Intel

FIGURE 3 Moore’s law.

pace with the increasing demand for higher performance and more electronic
features, the semiconductor manufacturing process must continue to shrink the
feature size on a chip. As the chip shrinks, the distance between transistors
decreases, which in turn increases the speed of the device. Figure 4 presents a
graphical picture of the rate of change of the semiconductor manufacturing process.
Essentially, a new manufacturing process comes on line every two years. Currently,
most of the manufacturing technology in the semiconductor industry is operating at
a feature size of 0.25 µm and moving to 0.18 µm and on to 0.13 µm in 2002.
The rapid rate of change in manufacturing provides many opportunities for
applying DFE strategies. Each new manufacturing tool set that is developed can
be reviewed, and specific sources of air/water emissions or waste generation can
be targeted for continuous improvement in the next generation.
To assist in DFE approaches, semiconductor companies leverage the trade
association known as the Semiconductor Equipment and Materials International
(SEMI) to establish environmental, health, and safety (EHS) design guidelines for
new manufacturing equipment. The two primary guidelines are SEMI S2, “Safety
Guidelines for Semiconductor Manufacturing Equipment,” and SEMI S8, “Safety
Guidelines for Ergonomic Engineering of Semiconductor Manufacturing Equip-
ment.” These two documents provide the basis for establishing design for EHS.
Table 1 summarizes the environmental requirements contained in the SEMI S2
document.
 Pollution Prevention and DFE 261

CPU Units
0.35um 0.25um 0.18 um
Shipped (%)
100%
90%
80%
70%
60%
50% TWO YEARS
40%
30%
20%
10%
0%
8

0
9

0
19

29

39

49

29

39

49

10

20

30
19

40
Q

Q
Q

Q
Source: Intel
Forecast Data
FIGURE 4 Semiconductor technology conversion (0.35 to 0.18 µm).

Since these are industry-wide EHS guidelines for developing new manu-
facturing tools, individual semiconductor manufacturers do not have to develop
their own unique set of requirements. Nor do they suffer from the imposition of
environmental design guidelines. It should be noted that this approach does not
prevent individual semiconductor manufacturers from imposing additional stand-
ards that are stricter than those set forth in SEMI S2 or S8, on specific tools.
Implementation of the SEMI S2 guidelines by equipment manufacturers
increased in acceptance globally in the early to mid-1990s. Through the im-
plementation of the S2 guidelines and other approaches taken by the semiconduc-
tor industry, significant success in reducing emissions has been achieved. An
example of this improvement is demonstrated in Table 2 and Figure 5, which
presents the release of hazardous air pollutants (HAPs) by the U.S. semiconductor
industry.
Using the TRI database, HAPs emissions data were generated for the
semiconductor industry. For the period 1987–1990 (four years), the average
annual HAPs emissions by the U.S. semiconductor industry was 4.09 million
pounds. For the period 1994–1997, average annual HAPs emissions were 0.83
million pounds, which represents a fivefold decrease in HAPs emissions.
Figure 5 presents the HAPs emission data in pounds per million square
inches of silicon wafers produced by U.S. semiconductor manufacturers. Table 2
262 McManus

TABLE 1 Summary of Semiconductor Manufacturing Equipment

International (SEMI) Environmental Design Considerations
The Environmental Design Guidelines apply across the full life of the equip-
ment, including decommissioning and disposal. The equipment manufac-
turer should consider resource conservation, including:
Water reuse/recycling
Reduce consumption of chemicals, energy, and water
Reduce resource requirements for equipment maintenance and reduc-
tion in packaging requirements
The use of chemicals for processing, maintenance, and utilities must con-
sider chemical use effectiveness, environmental impacts, toxicity, waste
generation, and decommissioning.
Where practicable, the following chemicals should not be part of use or
operation:
Ozone-depleting substances as specified by the Montreal Protocol
Perfluorocarbons, including CF4, C2F6, NF3, C3F6, SF6, and CHF3
The equipment design must also reduce potential for unintended releases,
and include such features as overfill detectors and alarms, secondary
containment, chemical compatibility, and automatic shutoff of chemical
feed systems.
Segregation of effluents, wastes, and emissions needs to be provided to
prevent issues with chemical incompatibility, facilitation of recycle and
reuse, and to facilitate effective treatment technologies.
Equipment must also be designed to facilitate decommissioning.

provides the absolute quantity of HAPs emissions; Figure 5 reflects the fact that
production quantities of semiconductor wafers have increased each year since
1987. Both the absolute quantity of HAPs emissions and the quantity per unit of
production have continued to decrease.
Table 3 presents the total quantity of TRI chemical releases or transfers for
both the U.S. semiconductor industry and all U.S. industry. The data demonstrate
that the semiconductor industry produces less than 1% of the total TRI and that,
based on 1997 data, that sector has reduced TRI releases and transfers by a factor
of 3.5, compared to the first reporting year of 1987. During that same period, U.S.
industry as a whole reduced TRI releases by a factor of 2.7.
In summary, with the rapid change of manufacturing technology that
occurs within semiconductor manufacturing, the opportunities for engaging in
design for the environment are substantial. Use of SEMI S2 environmental
guidelines provides clear design guidelines for equipment suppliers on the types
of environmental improvements desired by the semiconductor industry. The key
to implementation is early engagement in the manufacturing development pro-
Pollution Prevention and DFE 263

TABLE 2 Hazardous Air Pollutants

(HAPs) Emissions by U.S. Semiconductor
Industry (1987–1997)

Year HAPs emissions (million pounds)

1987 3.85
1988 4.76
1989 4.30
1990 3.43
1991 2.82
1992 2.21
1993 1.39
1994 1.12
1995 0.88
1996 0.77
1997 0.55
Source: EPA Toxins Release Inventory (TRI) Data-
base (1987–1997).

cess. Figure 6 presents a model for early engagement to achieve DFE within
semiconductor manufacturing.

6.2 Metal Finishing

The metal finishing industry is characterized typically as a small manufac-
turing operation with about 10,000 job-shop or captive facilities across the
United States. Under the Clean Water Act, the EPA defines metal finishing as
follows:
Plants which perform any of the following six metal finishing operations
on any basis material: Electroplating, Electroless Plating, Anodizing,
Coating (chromating, phosphating and coloring), Chemical Etching and
Milling and Printed Circuit Board Manufacture. If any of those six
operations are present, then this part applies to discharges from those
operations and also to discharges from any of the following 40 process
operations: Cleaning, Machining, Grinding, Polishing, Tumbling, Bur-
nishing, Impact Deformation, Pressure Deformation, Shearing, Heat
Treating, Thermal Cutting, Welding, Brazing, Soldering, lame Spraying,
Sand Blasting, Other Abrasive Jet Machining, Electric Discharge Ma-
chining, Electrochemical Machining, Electron Beam Machining, Laser
Beam Machining, Plasma Arc Machining, Ultrasonic Machining, Sinter-
ing, Laminating, Hot Dip Coating, Sputtering, Vapor Plating, Thermal
 264 McManus

0
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997

SOURCE: EPA TRI Database (1987-97); Dataquest (June 1999)

FIGURE 5 HAPs release by U.S. semiconductor manufactures as a function of

silicon substrate production (1987–1997).

Infusion, Salt Bath Descaling, Solvent Degreasing, Paint Stripping,

Painting, Electrostatic Painting, Electropainting, Vacuum Metalizing,
Assembly, Calibration, Testing and Mechanical Plating (6).
The primary pollutants generated by metal finishing are wastewaters con-
taining heavy metals, waste solvent, and heavy metal-containing debris. The
manufacturing technology changes less rapidly than that in the semiconductor
industry, but is, in fact, more representative of the majority of U.S. manufacturing
sectors. Typically, changes in metal finishing manufacturing are upgrades of
specific tools within an existing line. One of the biggest improvements, which
began about two decades ago, was the use of countercurrent rinsing techniques
to reduce the volumes and heavy metal concentrations of wastewater generated
during metal plating.
One of the concerns for this industry is that the heavy metals contained in
the wastewater frequently discharge to a publicly owned treatment works
(POTW) and accumulate in the municipal treatment plant’s sludge or pass through
the facility and discharge with the treated effluent. Therefore, significant effort
has been undertaken under the Clean Water Act to reduce the emissions of metals
from this industry sector.
Pollution Prevention and DFE 265

TABLE 3 Total Quantity of TRI Chemical

Releases or Transfers for U.S. Semiconductor
Manufacturing and All U.S. Industry (1987–1997)

U.S. semiconductor All U.S. industry

Year (million pounds) (million pounds)

1987 70.4 7,011.2

1988 38.7 6,489.0
1989 32.5 5,705.7
1990 24.5 4,999.2
1991 28.7 4,411.4
1992 26.0 4,299.2
1993 20.9 3,754.6
1994 15.3 3,134.9
1995 22.3 2,637.6
1996 18.1 2,545.3
1997 18.7 2,586.9
Source: EPA Toxics Release Inventory (TRI) Database,
1987–1997.

In 1993, EPA Administrator Carole Browner embarked on a new program,

known as the Common Sense Initiative (CSI), to determine if a more effective
approach to environmental regulations and management could be achieved through
an industry sector-by-sector approach. The metal finishing industry was one of
six selected for the CSI, with the goal to promote “cleaner, cheaper, smarter”
environmental performance. One major outcome was a commitment by the metal
finishing industry to reduce its environmental footprint. Table 4 presents the
environmental commitments of the metal finishing industry developed under CSI.
The environmental goals set forth on both a facility and sector-wide basis
in Table 4 provide the necessary direction for using quantitative performance
goals to achieve design for the environment within the metal finishing industry.
If successful, this industry will significantly reduce its emissions of metals into
the environment and achieve significant reductions in consumption of natural
resources, with specific emphasis on reducing water and energy use. Through the
reductions in water and energy use, the industry should also reduce its operating
costs and make it more competitive. Since this was a cooperative effort between
the industry and the EPA, the agency has also agreed to reduce by 50% the
administrative burden. This, in turn, reduces the operating costs of these types of
metal finishing facilities.
In summary, P2 and DFE represent opportunities for industrial facilities to
improve their environmental performance and, in many instances, reduce operat-
 266

Optimum Time
to effect change
Early in technology Technology
engage- frozen, major
ment changes
Very early require much
through
engagement more effort -
tool
through use
supplier
universities continuous
targets
and improvement
government
labs

Ability to Effect Change

External Research Supplier R&D Process Development Manufacturing Ramp

6 4 2 0

Research Commercialization Phases Demonstration Ramp to HVM

YEARS to High Volume Manufacturing

FIGURE 6 EHS technology engagement model for semiconductor manufacturing.

McManus
Pollution Prevention and DFE 267

TABLE 4 Metal Finishing National Performance Goals: A Vision for a

Cleaner, Cheaper, Smarter Future
Facility-based performance goals Sector-wide performance goals
(by 2002) (by 2002)
(1) Improved Resource Utilization (4) Industry-Wide Achievement of Facil-
(“Smarter”) ity Goals.
(a) 98% of metals ultimately utilized (a) 80% of facilities nationwide
on product. achieve Goals 1–3.
(b) 50% reduction in water pur- (5) Industry-Wide Compliance with
chased/used (from 1992 levels). Environmental Performance
(c) 25% reduction in facility-wide Requirements.
energy use (from 1992 levels). (a) All operating facilities achieve
(2) Reduction in Hazardous Emissions compliance with Federal, State
and Exposures (“Cleaner”) and Local environmental perfor-
(a) 90% reduction in organic TRI mance requirements.
emissions and 50% reduction in (b) All metal finishers wishing to
metals emissions to air and cease operations have access to
water (from 1992 levels). a government sponsored “exit
(b) 50% reduction in land disposal strategy” for environmentally
of hazardous sludges and a re- responsible site transition.
duction in sludge generation (c) All enforcement activities involv-
(from 1992 levels). ing metal finishing facilities are
(c) Reduction in human exposure to conducted in a consistent man-
toxic materials in the facility and ner to achieve a level playing
the surrounding community, field, with a primary focus on
clearly demonstrated by actions those facilities that knowingly
selected and taken by the facil- disregard environmental
ity. Such actions may include, requirements.
for example, pollution preven- Note: At facilities where outstanding
tion, use of state-of-the-art emis- performance levels were reached prior
sion controls and protective to 1992, the percentage-reduction tar-
equipment, use of best recog- gets for Goals 1(b) and (c) and 2(a) and
nized industrial hygiene prac- (b) may not be fully achievable, or the
tices, worker training in effort to achieve them may not be the
environmental hazards, and par- best use of available resources. In these
ticipation in a Local Emergency instances, a target should be adjusted
Planning Committtee. as necessary to make it both meaning-
(3) Increased Economic Payback and ful and achievable.
Decreased Costs (“Cheaper”)
(a) Long term economic benefit to
facilities achieving Goals 1 and 2.
(b) 50% reduction in costs of unnec-
essary permitting, reporting,
monitoring and related activities
(from 1992 levels), to be im-
plemented through burden re-
duction programs to the extent
that such efforts do not ad-
versely impact environmental
outcomes.

Source: Ref. 8.
268 McManus

ing costs. Several industry sectors, such as metal finishing and semiconductor
manufacturing, have moved to an industry sector-wide approach to help facilitate
the implementation of P2 and DFE. It is anticipated that sector-wide approaches
will provide additional encouragement for individual facilities to move forward
in implementing new environmental management techniques.

REFERENCES
1. “Waste Minimization Requirements,” http://yosemite.epa.gov/OSW/rcra.nsf/Documents/
F68B964A59054380852565DA006F08E1.
2. “Design for the Environment,” http://www.epa.gov.
3. President’s Council on Sustainable Development, Sustainable America: A New Con-
sensus for Prosperity, Opportunity and a Healthy Environment for the Future. Wash-
ington, DC: U.S. Government Printing Office, 1996.
4. President’s Council on Sustainable Development, Towards a Sustainable America—
Advancing Prosperity, Opportunity and Health Environment for the 21st Century.
Washington, DC: U.S. Government Printing Office, May 1999.
5. U.S. Environmental Protection Agency, Science Advisory Board, Integrated Environ-
mental Decision Making in the 21st Century. Draft Document, May 3, 1999, p. 1.
6. Code of Federal Regulations, 1998 Version, 40 CFR Section 433.10.
7. Safety Guidelines for Semiconductor Manufacturing Equipment, Semiconductor
Equipment and Materials International (SEMI) S2, 2000 Version, pp. 23–24.
8. EPA Common Sense Initiative, Metal Finishing Sector, “Strategic Goals Program—
National Performance Goals and Action Plan,” www.strategicgoals.org/plan/part1.htm.
15
Pollution Prevention and
Life Cycle Assessment

Mary Ann Curran

U.S. Environmental Protection Agency, Cincinnati, Ohio

Rita C. Schenck
Institute for Environmental Research and Education,
Vashon, Washington

1 INTRODUCTION
Over the past 20 years, environmental management strategies in the United States,
as well as in many other countries, have evolved through the development of laws
and regulations that limit pollutant releases to the environment. For example,
since its inception in 1970, the U.S. Environmental Protection Agency (EPA) has
made important progress toward improving the environment in every major
category of environmental impact caused by pollutant releases. Levels of emis-
sions across the nation have stayed constant or declined, hundreds of primary and
secondary wastewater treatment facilities have been built; land disposal of
untreated hazardous waste has largely stopped, hundreds of hazardous waste sites
have been identified and targeted for cleanup, and the use of many toxic
substances has been banned. Together, these actions have had a positive effect on
the nation’s environmental quality and have set an example for nations every-

269
270 Curran and Schenck

where. However, despite the combined achievements of the federal government,

states, and industry in controlling waste emissions which have resulted in a
healthier environment, the further improvement of the environment has slowed.
This led to the realization that a new paradigm was needed for environmental
protection.
Starting in the mid-1980s, pollution prevention was seen by visionaries as
the way to go beyond such command and control approaches. Pollution preven-
tion has received widespread emphasis internationally within multinational organ-
izations and within the governments of both developing as well as developed
nations. The European Union has designed some of its rules and programs based
on the concepts behind pollution prevention. The United Nations Environmental
Programme (UNEP) has a clean technologies program and has supported many
workshops and meetings on various topics on pollution prevention as well as
sustainable development.
Worldwide, the advancement of environmental protection strategies mov-
ing from end-of-pipe to pollution prevention and beyond has been steady.
This evolution can be summarized by the chronology shown in Table 1. This
chapter provides an overview of industrial pollution prevention, beginning with
a brief definition of a pollution prevention opportunity assessment, followed by
a discussion of life cycle assessment (LCA).

2 UNDERSTANDING OF INDUSTRIAL
POLLUTION PREVENTION
Briefly stated, industrial pollution prevention is a term that is used to describe
technologies and strategies that result in eliminating or reducing waste streams
from industrial operations. The EPA defines pollution prevention as “the use of
materials, processes, or practices that reduce or eliminate the creation of pollu-
tants or wastes at the source.” It includes practices that reduce the use of
hazardous materials, energy, water, or other resources and practices that protect
natural resources through conservation or more efficient use. The basic idea of

TABLE 1 Evolution of Environmental Protection

Chronology Strategy

1970s to 1980s End-of-pipe treatment

mid-1980s Waste minimization
early 1990s Pollution prevention
mid-1990s ISO Certification/life cycle assessment
2000 and beyond Agenda 21 for Sustainable Development
Pollution Prevention and Life Cycle Assessment 271

pollution prevention follows the axiom, “an ounce of prevention is worth a pound
of cure.” The U.S. Pollution Prevention Act of 1990 and pollution prevention
experts conclude that it makes far more sense for a waste generator not to produce
waste in the first place, rather than developing extensive, never-ending treatment
schemes (1).
For industrial pollution prevention, two general approaches are used to
characterize processes and waste generation. The first approach involves gather-
ing information on releases to all media (air, water, and land) by looking at the
output end of each process, then backtracking the material flows to determine the
various waste sources. The other approach tracks materials from the point where
they enter a facility, or plant, until they exit as wastes or products. Both
approaches provide a baseline for understanding where and why wastes are
generated, as well as a basis for measuring waste reduction progress. The steps
involved in these characterizations are similar and include gathering background
information, defining a production unit, general process characterization, under-
standing unit processes, and completing a material balance.
These steps, when performed systematically, provide the basis for a pollu-
tion prevention opportunity assessment. It begins with a complete understanding
of the various unit processes and points in these processes where waste is being
generated and ends with the implementation of the most economically and
technically viable options. It may be necessary to gather information to demon-
strate that pollution prevention opportunities exist and should be explored. Often,
an assessment team is established to perform the steps along the way (2).
A preliminary assessment of a facility is conducted before beginning a more
detailed assessment. The preliminary assessment consists of a review of data that
are already available in order to establish priorities and procedures. The goal of
this exercise is to target the more important waste problems, moving on to
lower-priority problems as resources permit. The preliminary assessment phase
provides information that is needed to accomplish this prioritization and to
assemble the appropriate assessment team (3).
A subsequent detailed assessment focuses on the specific areas targeted by
the preliminary assessment. Analyzing process information involves preparing a
material and energy balance as a means of analyzing pollution sources and
opportunities for eliminating them. Such a balance is an organized system of
accounting for flow, generation, consumption, and accumulation of mass and
energy in a process. In its simplest form, a material balance is drawn up according
to the mass conservation principle:
Mass in = mass out – (generation + consumption + accumulation)
If no chemical or nuclear reactions occur and the process progresses in a
steady state, the material balance for any specific compound or constituent is as
follows:
272 Curran and Schenck

Mass out = mass in

A process flow diagram may be helpful by providing a visual means of
organizing data on the material and energy flows and on the composition of
streams entering and leaving the system (see Figure 1). Such a diagram shows the
system boundaries, all stream flows, and points where wastes are generated.
Boundaries should be selected according to the factors that are important
for measuring the type and quantity of pollution prevented, the quality of the
product, and the economics of the process. The amount of material input should
equal the amount exiting, corrected for accumulation and creation or destruction.

FIGURE 1 Example flow diagram (3).

Pollution Prevention and Life Cycle Assessment 273

A material balance should be calculated for each component entering and leaving
the process, or other system being studied. A suggested approach for making these
calculations is offered in Section 3.
Once the sources and nature of wastes generated have been described, the
assessment team enters the creative phase. Pollution prevention options are
proposed and then screened for feasibility. In this environmental evaluation step,
pollution prevention options are assessed for their advantages and disadvantages
with regard to the environment. Often the environmental advantage is obvious—
the toxicity of a waste stream will be reduced without generating a new waste
stream. Most housekeeping and direct efficiency improvements have this advan-
tage. With such options, the environmental situation in the company improves
without new environmental problems arising (3).
Along with assessing the technical and environmental effectiveness in
preventing pollution, options are evaluated for the estimated cost of purchasing,
installing, and operating the system. Pollution prevention can save a company
money, often substantial amounts, through more efficient use of valuable resource
materials and reduced waste treatment and disposal costs.
Estimating the costs and benefits of some options is straightforward, while
for others it is more complex. If a project has no significant capital costs, the
decision is relatively simple. Its profitability can be judged by determining
whether it reduces operating costs or prevents pollution. Installation of flow
controls and improvement of operating practices will not require extensive
analysis before implementation. However, projects with significant capital costs
require detailed analysis. Several techniques are available, such as payback
period, net present value, or return on investment. These approaches are also
described in Section 3.
At times, the environmental evaluation of pollution prevention options
is not always straightforward. Some options require a thorough environmental
evaluation, especially if they involve product or process changes or the sub-
stitution of raw materials. For example, the engine rebuilding industry is no
longer using chlorinated solvents and alkaline cleaners to remove grease and
dirt from engines before disassembly. Instead, high-temperature baking followed
by shot blasting is being used. This shift eliminates waste cleaner but requires
additional energy use for the shot blasting. It also presents a risk of atmo-
spheric release because small quantities of components from the grease can
vaporize. (3)
Others are moving toward the use of aqueous cleaners as substitutes for
solvents in an attempt to avoid using toxic materials. However, while the less
toxic aqueous cleaners offer a suitable substitution for chlorinated cleaning
solvents from a performance standpoint, their use may be resulting in increased
environmental impacts in other areas. Most obvious is the increased energy use
274 Curran and Schenck

that occurs from needing to heat the parts to be cleaned in order to get a
satisfactory level of cleaning performance.
To make a sound evaluation, the team should gather information on all the
environmental aspects of the product or process being assessed. This information
would consider the environmental effects not only of the production phase but of
the acquisition of raw materials, transportation, product use, and final disposal as
well. This type of holistic evaluation is called a life cycle assessment (LCA). The
stages that are included within the boundary of an LCA are shown in Figure 2.
LCA’s origins in mass and energy balance sheets have led to several important
accounting conventions, including the following.

A system-wide perspective embodied in the term “cradle-to-grave” that

implies efforts to assess the multiple operations and activities involved
in providing a product or service. This includes, for example, resource
extraction, manufacturing and assembly, energy supplies and transporta-
tion for all operations, use, and disposal.
A multimedia perspective that suggests that the account balance include
resource inputs as well as wastes and emissions to most common
environmental media, e.g., air, water, and land.
A functional unit accounting normalizes energy, materials, emissions, and
wastes across the system and media to the service or product provided.
Notably, this calculation allows the analysis of different ways to provide
a function or service, for example, one can compare sending a letter via
e-mail or via regular mail. Additionally, this approach entails allocation
procedures so that only those portions or percentages of an operation
specifically used to produce a particular product are included in the final
balance sheet (4).

The functional unit approach of LCA takes the assessment beyond looking
at the environmental impacts associated with a specific location or operation.
The value of LCA lies in its broad, relative approach to analyzing a system and
factoring in global as well as regional and local environmental impacts. This
general, macro approach makes it theoretically feasible to frame numerous
potential issues and environmental considerations, identify possible trade-offs
between different parts of the life cycle, and make these possible issues and
trade-offs apparent to decision makers. These attributes enable the user to
understand complex and previously hidden relationships among the many system
operations in the life cycle and the potential repercussions of changes in an
operation on distant operations and other media. This is particularly true where
unanticipated or unrecognized issues on the life cycle of a product or service are
revealed to decision makers. This leads to a more complete and thorough
evaluation for making decisions, including applications in strategic planning,
 Pollution Prevention and Life Cycle Assessment

FIGURE 2 Input/output flows in a product life cycle.

275
276 Curran and Schenck

environmental management, product development and R&D, and liability assess-

ment, as well as pollution prevention.

3 CALCULATION

3.1 Measurement of Pollution Prevention

Accurate and meaningful measurement systems are vital to ensure long-term
successful implementation of pollution prevention (5). To implement pollu-
tion prevention, industrial facilities must first measure the environmental im-
pacts of their facilities, beginning with the accounting of the inputs and outputs
across the facility’s boundaries. This process is captured in a material and
energy balance.

3.1.1 Material and Energy Balance

Analyzing process information involves preparing material and energy bal-
ances as a means of analyzing pollution sources and opportunities for eliminating
them. Such a balance is an organized system of accounting for the flow,
generation, consumption, and accumulation of mass and energy in a process. In
its simplest form, a material balance is drawn up according to the mass conser-
vation principle:

Mass in = mass out – (generation + consumption + accumulation)

The first step in preparing a balance is to draw a process diagram, which is

a visual means of organizing data on the energy and material flows and on the
composition of the streams entering and leaving the system. A flow diagram, such
as Figure 1, shows the system boundaries, all streams entering and leaving the
process, and points at which wastes are generated. The goal is to account for all
streams so that the the mass equation balances.
The boundaries around the flow diagram should be based on what is
important for measuring the type and quality of pollution prevented, the quality
of the product, and the economics of the process. Again, the amount of material
input should equal the amount exiting, corrected for accumulation and creation
or destruction.
In addition to an overall balance, a material balance should be calculated
for each individual component entering and leaving the process. When chemical
reactions take place in a system, there is an advantage to performing the material
balance on the elements involved.
Material and energy balances do have limitations. They are useful for
organizing and extending pollution prevention data and should be used whenever
Pollution Prevention and Life Cycle Assessment 277

possible. However, the user should recognize that most balance diagrams will be
incomplete, approximate, or both (6).
Most processes have numerous process streams, many of which affect
various environmental media.
The exact composition of many streams is unknown and cannot be easily
analyzed.
Phase changes occur within the process, requiring multimedia analysis and
correlation.
Plant operations or product mix change frequently, so the material and energy
flows cannot be accurately characterized by a single balance diagram.
Many sites lack sufficient historical data to characterize all streams.
These are examples of the complexities that will recur in the analysis of
real-world processes.
Despite the limitations, material balances are essential for organizing data
and identifying data gaps and other missing information. They can help calculate
concentrations of waste constituents where quantitative composition data are
limited. They are particularly useful if there are points in the production process
where it is difficult or uneconomical to collect or analyze samples. Data gaps,
such as an unmeasured release, can also indicate that fugitive emissions are
occurring. For example, solvent evaporation from a parts cleaning tank can be
estimated as the difference between solvent added to the tank and solvent that is
removed by disposal, recycling, or dragout (6). It is an essential characteristic of
a mass balance that unmeasured flows are used to balance the equation.
3.1.2 Industrial Production and Waste Generation
Tracking System
The Industrial Production and Waste Generation Tracking System shown in
Figure 3 (7) establishes a framework for the determination of the main parameters
for industrial production and waste generation. It is based on the following main
production process variables:
1. Raw materials (rm)
2. Other materials entering production process (v)
3. Produced products (P)
4. Generated waste (y)
The generated waste may be:
1. Managed (g) by applying waste management
2. Released (z) into the environment, causing environmental pollution
Managed waste (g) may be further processed to be
278 Curran and Schenck

FIGURE 3 Industrial production and waste generation tracking system (7).

3. Used as secondary raw material and/or energy (s)

4. Finally disposed of as processed waste (residues) in a special (or
secure) landfill (d).
The development of the Industrial Production and Waste Generation Track-
ing System model was based on the work of Baetz et al. (8). A model enabling
calculation of quantities of waste generated in an industrial production was
developed and defined as shown in Table 2.
During an industrial process at time t, a production factor U, correlating
quantity of raw and other materials r and products P, has a value of 0 ≤ U ≤ 1
and is defined as
P
U =
r
Note that raw materials r includes “other materials” not typically defined as raw
materials entering a production process. For example, paints and lacquers in
“white goods” and furniture manufacture are usually not defined as raw materials
but are still input materials.
Converting the last expression, the quantity of products may then be
expressed as
P = Ur
Pollution Prevention and Life Cycle Assessment 279

TABLE 2 Definition of IPWGTS Model Parameters (7)

Parameter Definition

t Industrial production time in which system was

observed
rm Quantity of raw material entering industrial pro-
duction in time t
v Quantity of other materials (not defined as raw
materials) entering industrial production in time t
r = rm + v Quantity of raw and other materials entering indus-
trial production in time t
U = P/r Production factor after time t; U = 1 represents
total production, while U = 0 represents zero
level of production and therefore total waste
generation
P = Ur Quantity of products produced in time t
y = (1 − U)r Quantity of solid, liquid, and/or gaseous waste
generated in time t
M = g/y = g/(1 − U)r Waste management factor after time t; M = 1 rep-
resents total waste management, while M = 0
represents zero level of waste management and
therefore total release (emission, spill, and/or
discharge) into the environment
g = M(1 − U)r Quantity of solid, liquid, and/or gaseous waste
managed by waste management (temporary stor-
age, collection, transportation, processing, and
final disposal)
z = (1 − M)(1 − U)r Quantity of solid, liquid, and/or gaseous waste
released to air, water, and/or soil/land, causing
environmental pollution
R = s/g = s/M(1 − U)r Waste recycling factor after time t; R = 1 repre-
sents total waste recycling by physical, chemi-
cal, thermal, and/or biological process as to
recover secondary materials and/or energy,
while R = 0 represents total waste processing by
physical, chemical, thermal, and/or biological
process for final disposal in the environment
s = MR(1 −U)r Quantity of secondary raw materials and/or energy
recovered from solid, liquid, and/or gaseous
waste by waste recycling
d = M(1 − R)(1 − U)r Quantity of processed waste for final disposal
280 Curran and Schenck

while quantity of waste generated is

y = (1 − U)r
Managed waste is further quantified by a waste management factor M,
which is defined as the ratio between waste managed and waste generated:
g
M =
y
and can have a value 0 ≤ M ≤ 1. The quantity of waste managed by storage,
collection, transportation, processing, and final disposition is then
g = M(1 − U)r
while the quantity of waste released (emitted, discharged, and/or spilled) into the
environment is
z = (1 − M)(1 − U)r
If waste is further processed by physical, chemical, thermal, and/or biolog-
ical processing to recover secondary raw materials and/or energy, then processed
waste is determined by the waste recycling factor R,
s
R =
g
having the value 0 ≤ R ≤1. The quantity of waste recycled into secondary raw
materials and/or energy by waste processing is
S = MR(1 − U)r
while the quantity of waste to be finally disposed is
d = M(1 − R)(1 − U)r
Knowing quantities of raw and other materials (rm + v) entering the
observed system and quantities of products (P) produced, quantities of waste
generated (y) can be calculated. If the quantity of waste managed (g) by the waste
generator is known, it is possible to predict quantities of material lost (z) through
release (emission, discharge, and/or spill). Finally, if the waste generator recycles
managed waste into secondary raw materials and/or energy (s), then the quantity
of waste to be disposed (d) can be determined.
3.1.3 Production-Adjusted Pollution Prevention
After a pollution prevention activity has been implemented, adjusted figures from
the process flow diagram should show a decrease in waste generation. This
decrease is often approached in one of two ways (5). The first way is to look at
the change in the quantity of chemicals or raw materials that are purchased or
Pollution Prevention and Life Cycle Assessment 281

used for a process. This approach is based on the idea that materials that do not
enter a production process in the first place cannot leave the process in the form
of wastes or emissions, given that production is at a steady rate. The second, more
common pollution prevention approach looks at how much the quantity of waste
streams or emissions has been reduced over a given period of time. This results
in a statement such as “Pollution prevention was achieved in a 20% reduction in
the amount of spent chromium discharged compared to last year.”
However, in both approaches, accounting for the varying levels of produc-
tion is a key issue in more accurately capturing pollution prevention progress. If
production drops off, a decrease in waste streams or emissions may be attributable
to the decreased activity instead of to any particular pollution prevention effort.
Production-adjusted measures of pollution prevention account for changes result-
ing from pollution prevention efforts. For production-adjusted pollution preven-
tion measures, a unit of product is the factor used to adjust gross quantities of
waste or chemical use to infer the amount of pollution prevention progress
(see Table 3).
Using units of product to calculate pollution prevention improvements can
filter out effects of change in production activity. For example, a firm generated
22,000 lb (1000 kg) of trichlorethylene (TCE) waste from a vapor degreasing
operation used to remove oil for the 16,000 metal circuit boxes it manufactured.
In 1994, the company implemented several pollution prevention activities which
resulted in the generation of 15,000 lb (6800 kg) of TCE to manufacture 20,000
circuit boxes. At first, it would seem logical to express the achieved pollution
prevention as the difference in the amount of TCE waste from 1993 to 1994, or
7000 lb (i.e., 20,000 lb – 15,000 lb). However, this way of measurement does not
filter out the effect of increased production. Factoring in increased output, the
pollution prevention progress can be calculated as follows:

TABLE 3 Typical Ways to Measure P2

Not production-adjusted
Change in quantity of emissions: “Reduced discharge of chromium by
20% last year.”
Change in quantity of chemical or raw materials used: “Reduced plating
solution purchases by 10% last year.”
Production-adjusted
Change in quantity of chemical used per unit produced: “10% reduction
in quantity of plating solution used per part shipped last year.”
Change in quantity of chemical used per unit activity: “Reduced solvent
use by 15% for every hour the degreaser ran last year.”
282 Curran and Schenck

boxes in 1994 20,000

Unit−of−product ratio: = = 1.25
boxes in 1993 16,000

The unit-of-product ratio is used to calculate the expected waste generation, given
this year’s level of production, if no pollution prevention changes had been made
during the past year.
Expected waste generation in 1994 is calculated as follows:

(Production ratio) ⋅ (1993 waste generation) = (1.25) ⋅ (22,000 lb) = 27,500 lb

or = (1.25) ⋅ (10,000 kg) = 12,500 kg

Therefore,

(1994 adjusted waste generation) − (1994 actual waste generation)

= 27,500 lb (12,500 kg) − 15,000 lb (6,804 kg)
= 12,500 lb (5,680 kg) waste reduction

Another way to examine the effects of pollution prevention is to assess

whether the amount of waste per “widget” produced has changed. Using the data
from the preceding example, the calculations are as follows:
TCE waste generated in 1993 22,000
=
number of widgets produced in 1993 16,000
= 1.38 lb (0.626 kg) TCE per circuit box
TCE waste generated in 1994 15,000
=
number of widgets produced in 1994 20,000
= 0.75 lb (0.34 kg) TCE per circuit box
The two waste efficiencies can then be compared to conclude that the company
made substantial waste reductions of:

1.38 lb − 0.75 lb = 0.63 lb of TCE per product

(0.626 kg − 0.34 kg = 0.29 kg of TCE per product)

These examples show the importance of finding and using a unit of product
that is closely related to the waste or chemical usage being targeted. Suppose,
however, that this facility has modified its degreasing operations and reduced
solvent loss, but the loss of solvent is more related to the number of hours the
degreaser was running than to the number of parts that were cleaned. In that case,
calculating “solvent savings per part cleaned” has less meaning for indicating
pollution prevention progress. Solvent saved per hour of degreasing operation,
however, would provide a better picture of actual savings resulting from the
change.
Pollution Prevention and Life Cycle Assessment 283

3.1.4 Calculating the Cost of Pollution Prevention

The “usual costs” of any process are commonly considered to be the costs
associated directly with the polluting practice or proposed alternatives. Costs
may then be categorized as either capital expenses that must be depreciated
for tax purposes or other expenses that can be deducted from taxes in a single
year. Other expenses are commonly calculated as “capital” costs because they
are one-time costs that are needed before the process can be used. To collect
the cost needed for determining the economic feasibility, the following cost items
are needed.

Estimate Capital Cost Items

EQUIPMENT. This cost item represents the investment in new equipment
needed to implement the pollution prevention option. The cost element should
include the price (f.o.b. factory), taxes, freight, and insurance on delivery, and the
cost for the initial spare parts inventory. Any additional equipment needed to
support the pollution prevention alternative should be included, such as additional
laboratory equipment.
MATERIALS. Materials costs include piping, electrical equipment, new
instrumentation, and changes in the structure. These costs are those incurred in
purchasing the materials needed to connect the new process equipment (or revise
the use of existing equipment).
UTILITY CONNECTIONS. This item includes the costs for connecting new
equipment (or for making new connections to existing equipment) as part of
implementing the pollution prevention option. Typical utilities include electricity,
steam, cooling water, process water, refrigeration, fuel (gas or oil), plant air
(e.g., for process control), and inert gas.
SITE PREPARATIONS. This item includes the costs for any necessary site
preparation, such as demolition, site clearing, paving, etc.
INSTALLATION. This item includes the costs incurred during the installa-
tion of the process equipment or process change, as well as charges by the vendor
as well as by in-house staff.
ENGINEERING AND PROCUREMENT. This item includes the costs incurred
to design the process equipment or process change and to purchase any new
equipment. Charges for consultants used in designing and procuring equipment
are also included.

Estimate Expenses. The costs in this category include both one-time costs
and ongoing costs that are deductible for income tax purposes.
START-UP COSTS. Start-up costs include labor and material costs incurred
during the start-up phase.
PERMITTING COSTS. These costs include both fees and the costs incurred
by in-house staff in documenting the process change to meet permit requirements.
284 Curran and Schenck

SALVAGE VALUE. Estimate the net amount (in today’s dollars) that used
equipment will be worth at the end of it useful lifetime. Include the value of
working capital and catalysts and chemicals that will remain at the completion of
the equipment’s life.
TRAINING COSTS. Training costs include the costs for on-site and off-site
training related to the use of the new equipment or for making sure the process
change achieves its goal.
INITIAL CHEMICALS. The initial charges for chemicals and catalysts can
be considered a capital item.
WORKING CAPITAL. This category includes all elements of working
capital (required inventories of raw materials, in-process inventories, materi-
als and supplies) not already included as charges for chemicals and catalysts for
spare parts. Working capital may also include personnel costs for operations
start-up.
DISPOSAL COSTS. The disposal cost includes all the direct costs associated
with waste disposal, including solid waste disposal, hazardous waste disposal, and
off-site recycling.
RAW MATERIAL COSTS. This category includes both the raw materials
directly affected (e.g., chemicals for which more effective or less toxic substitutes
are being found) and other raw materials affected by the change in the process
(e.g., a change in a cleaning agent reduces the rejection rate of metal parts,
thereby reducing total material costs).
UTILITIES COSTS. Utilities include electricity, process steam, water, com-
pressed air, and heating oil or natural gas. It is important to consider whether a
change causes downstream effects as well, e.g., recycling an aqueous waste
stream may require energy to adjust the temperature of the stream to meet process
requirements.
OPERATING COSTS. This cost element includes the labor needed to run the
process or alternative.
Operation and maintenance (O&M) costs. This cost element includes
supplies needed on a regular basis, such as glassware, buckets, cleaners,
filters, protective equipment, etc.
Insurance and liability costs. In some cases, insurance rates may be
adjusted accordingly, e.g., switching to a process that is know to be safer
may lower insurance rates.
Other operating costs. This cost element includes other operating costs that
have not been specifically mentioned above.
Estimate Operating Revenues. In some cases, implementing a pollution
prevention option may lead to a change in the revenue from operations. The two
main categories are revenues from products and revenues from marketable
by-products.
Pollution Prevention and Life Cycle Assessment 285

PRIMARY PRODUCTS. If the process or procedural change changes the

production rate of the process, then the revenues before and after the change
should be included.
MARKETABLE BY-PRODUCTS AND RECOVERED MATERIAL. An increase in
the amount of marketable by-products and materials that are recovered and reused
should be included.
According to Humphreys and Wellman (9), the most widely used methods
for calculating potential savings are
Payback (or payout) time
Payback time with interest
Return on investment
Return on average investment
Discounted cash flow (also known as interest rate of return)
Venture worth analysis (also known as incremental present worth)
Each of these techniques offers both advantages and disadvantages, possibly
resulting in different order of profitability. All these methods are used to compare
alternatives. In that sense they give several alternatives with relative position of
preference to each other. For this reason, when pollution prevention alternatives
are being evaluated, more than one method of calculation should be used.
The following section evaluates an example investment of $1 million using
two methods: payback time and rate-of-return using discounted cash flow.
Payback time is the time required for all cash flows to equal the original
investment. In other words, it is the time it takes to recover the original
investment. The estimated cash flow and savings for a proposed $1 million project
with a projected life of 5 years is shown in Table 4. Through interpolation between
the accumulated savings, we see that the savings will reach $1 million between
the second and third year:
1,000,000 − 825,000
+ 2 = 2.6
1,105,000 − 825,000

TABLE 4 Estimated Cash Flow for a $1 Million

Project with a Projected Life of 5 Years

Year Savings per year Cumulative savings

1 $525,000 $ 525,000
2 300,000 825,000
3 280,000 1,105,000
4 200,000 1,305,000
5 125,000 1,430,000
286 Curran and Schenck

Therefore the payback period is 2.6 years.

While it is very simple to calculate, the shortcoming of this method is that
it does not consider the value of money, i.e., interest. The discounted cash flow
(DCF), or interest-rate-of-return, method is the most widely used of all these types
of calculations. It is also probably the most valid technique, since it considers all
cash flows in and out of a project as well as the time value of money. DCF is
based on the realization that a dollar in the future is not worth as much as today’s
dollar, which is available for reinvestment. In effect, DCF applies the expected
rate of return on an investment made today over the life of the investment.
The equation to calculate DCF is
n
CFn
0 = −I + ∑ (1 + i)n
1

where

CFn = cash flow in year n

i = interest rate of return
I = initial capital investment

Again using the previous example, DCF is calculated as follows:

525,000 300,000 280,000 200,000 125,000
0 = −1,000,000 + + + + +
(1 + i) 1 (1 + i) 2 (1 + i)3 (1 + i) 4 (1 + i)5
At i = 20%, the equation yields –45,449; and at i = 15%, the equation yields
+43,980. Interpolating between the two values and setting the equation equal to
zero, i = 17.5%. Therefore the DCF rate of return is 17.5% per year.

3.2 Life Cycle Assessment

Every day, both individual consumers and industry make choices that affect the
environment. Manufacturers choose from among different materials, suppliers, or
production methods. Consumers decide on the need for a product and make
purchasing choices. Environmentally responsible choices need reliable informa-
tion based on the life cycle characteristics of the alternative products or processes
being considered.
LCA considers the environmental aspects and the potential impacts of a
product or service system throughout its life—from raw material acquisition
through production, use, and disposal. This information has many potential uses:
it can help identify ways to improve environmental aspects of a product at various
stages in its life cycle; it can support decision making in industry, governmental,
or nongovernmental organizations; it can aid in the selection of relevant indicators
Pollution Prevention and Life Cycle Assessment 287

of environmental performance; and (with proper precautions) it can support the

marketing of products or services.
The environment is complex, with many interrelationships, and a major
challenge in any LCA study is to isolate the impacts of a single product or service
system. Comparability between LCA studies is also an issue, as products that
perform the same function may be made of widely varying materials. There is a
clear need for neutral, scientifically oriented, consensus-based guidance on the
conduct of LCA. Toward this end, various research organizations and practition-
ers are working to develop LCA methodology and tools. Especially, the Interna-
tional Standards Organization (ISO) has developed a series of standards and
technical reports that cover the various stages involved in LCA, from scoping
through impact assessment and interpretation (10–13). Using agreed-upon prin-
ciples, an LCA study can be done responsibly, transparently, and consistently.
Before the ISO efforts in LCA, early research conducted by the U.S. EPA
in LCA methodology along with efforts by the Society of Toxicology and
Chemistry (SETAC) led to the four-part approach to LCA that is widely accepted
today (14):
Goal and scope definition: identifying the purpose for conducting the LCA,
the boundaries of the study, assumptions, and expected output
Life cycle inventory: quantifying the energy use and raw material inputs and
environmental releases associated with each stage of the life cycle
Life cycle impact assessment: assessing the impacts on human health and
the environment associated with the life cycle inventory results
Improvement analysis/interpretation: evaluating opportunities to reduce
energy, material inputs, or environmental impacts along the life cycle.
LCA is not strictly a technical process. Various simplifying assumptions and
value-based judgments must be used throughout the process. The key is to keep
these to a minimum and be explicit in the reporting phase about what assumptions
and values were used. Readers of the study can then recognize the judgments and
decide to accept, qualify, or reject them and the study as a whole.
Finally, it should be recognized that the results of an LCA can provide much
of the information needed to make a decision. In most, if not all, cases LCA
should be integrated with other assessment tools and techniques, such as the
financial tools described above, to make sound decisions.
In addition to comparing the environmental soundness of products, LCA is
also being used to assess applications within industrial processes, such as pollu-
tion prevention activities. The following two examples demonstrate how LCA has
been used to evaluate options for material substitution and raw material sourcing.
Aqueous cleaners. While aqueous cleaners offer a suitable substitution for
chlorinated cleaning solvents, they generally require pretreatment prior
288 Curran and Schenck

to discharge to a POTW to adjust the pH, remove oil, grease, and solids,
and to precipitate phosphates and inactive chelating agents. Another
impact is that energy use is often higher than that required for chlorinated
solvents (15).
Butanediol (BDO). An alternative to natural gas-derived feedstock to
produce 1,4-butanediol (BDO) is a feedstock process that is based on the
fermentation of corn-derived glucose to succinic acid, followed by
catalytic reduction to BDO. The higher energy use of the alternative
process indicates that the overall environmental consequences would
be greater than the conventional process. Because electricity generation
is inefficient, and energy production in the United States is mostly
coal-based, the alternative process was projected to have a greater
potential for impact in multiple impact categories, including global
warming, acid rain, smog, water use, particulates, and solid waste (coal
ash) disposal (16).

LCA, while comprehensive in theory, encounters practical limitations and

barriers that are slowing its widespread adoption. A major barrier is the lack of
knowledge of the life cycle concept. Producers need to be made aware of the life
cycle impacts that their activities carry and the importance of going beyond
meeting compliance. More important, government offices that issue media-based
or industry-focused regulations and policies need to begin using life cycle
thinking. There are numerous instances where life cycle thinking is potentially
beneficial in making public policy. Introducing LCA concepts into the rule-
making process extends the regulatory analysis upstream and downstream and
across all media to account for the effects of the proposed standard, which may
otherwise escape a traditional regulatory impact analysis.
Another key barrier is the lack of reliable data. Lack of data has hindered,
perhaps prevented, many applications. Several efforts are underway in North
America and Europe to make data more easily accessible. Another fundamental
barrier to performing LCAs at this time is the lack of a generally agreed-on impact
assessment method. This seems to be more of a barrier in the United States than
in Europe, where several attempts at life cycle impact assessment have been
published. Figure 4 depicts the historical development of LCA practice.

3.2.1 Goal and Scope Definition

The goal and scope of an LCA study should be clearly defined and consistent with
the intended application of the results. The goal should be stated unambiguously,
together with the reasons for carrying out the study. In defining the scope of the
study, the following items should be considered and clearly described: the
system(s) being studied, the methodology and interpretation approach to be used,
Pollution Prevention and Life Cycle Assessment 289

FIGURE 4 Emerging LCA practice (4).

290 Curran and Schenck

and data needs. Specifically, the decisions made during the planning phase
include the following:
The purpose of the study
The audience, and whether publication will occur
The system function and functional unit
The study boundaries
What types of study to perform
Allocation methods
Environmental impact categories, and model and impact indicators
Data quality requirements
Scoping decisions make for the success or failure of the study. For example,
drawing the systems boundaries tends to be a balance between time and money
resources and practical limitations. Sometimes, in their attempt to save costs,
practitioners have drawn the boundaries of the study too narrowly, thus yielding
a study that leaves important questions unanswered. On the other hand, LCA
boundary conditions sometimes expand to the point that a full data set is never
accomplished (17).
Scoping an LCA is the most critical part of the LCA Study. Studies that are
not appropriately scoped are rarely successful.
3.2.2 Inventory—Calculating Energy and Material Inputs
and Environmental Releases
The second activity of a life cycle assessment is the identification and quantifi-
cation of energy and resource use and environmental releases to air, water, and
land (18). This inventory component is a technical, data-based process with a goal
of achieving a mass and energy balance for the life cycle system being studied.
In the broadest sense, inventory begins with raw material extraction and continues
through final product consumption and disposal. The boundaries for the system
being studied are determined in the goal definition and scoping phase and should
be broad enough to allow the study to quantify resource use and environmental
releases.
The quality of a life cycle inventory depends on an accurate description of
the system to be analyzed. The necessary data collection and interpretation is
contingent upon proper understanding of where each stage life cycle begins and
ends. The general scope of each stage can be described as follows.
Raw materials acquisition. This stage of the life cycle of a product includes
the removal of raw materials and energy sources from the earth, such as
the harvesting of trees or the extraction of crude oil. Land disturbance as
well as transport of the raw materials from the point of acquisition to the
point of raw materials processing are considered part of this stage.
Pollution Prevention and Life Cycle Assessment 291

Manufacturing. The manufacturing stage produces the product from the raw
materials and delivers it to consumers. Three substages or steps are
involved in this transformation: materials manufacture, product fabrica-
tion, and filling/packaging/distribution.
Materials manufacture. This step involves converting raw material into a
form that can be used to fabricate a finished product. For example,
several manufacturing activities are required to produce a polyethylene
resin from crude oil: the crude oil must be refined; ethylene must be
produced in an olefins plant and then polymerized to produce poly-
ethylene. Transportation between manufacturing activities and to the
point of product fabrication should also be accounted for in the inventory,
either as part of materials manufacture or separately.
Product fabrication. This step involves processing the manufactured mate-
rial to create a product ready to be filled, or packaged—for example,
blow molding a bottle, forming an aluminum can, or producing a cloth
diaper.
Filling/packaging/distribution. This step includes all manufacturing pro-
cesses and transportation required to fill, package, and distribute a
finished product. Energy and environmental wastes caused by transport-
ing the product to retail outlets or to the consumer are accounted for in
this step of a product’s life cycle.
Use/reuse/maintenance. This is the stage consumers are most familiar with,
the actual use, reuse, and maintenance of the product. Energy require-
ments and environmental wastes associated with product storage and
consumption are included in this stage.
Recycle/waste management. Energy requirements and environmental
wastes associated with product disposition are included in this stage, as
well as postconsumer waste management options such as recycling,
composting, and incineration.
The following general issues apply across all life cycle stages.
Energy and transportation. Process and transportation energy requirements
are determined for each stage of a product’s life cycle. Some products
are made from raw materials, such as crude oil, which are also used as
sources for fuel. Use of these raw materials as inputs to products
represents a decision to forego their fuel value. The energy value of such
raw materials that are incorporated into products typically is included as
part of the energy requirements in an inventory. Energy required to
acquire and process the fuels burned for process and transportation use
is also included.
Environmental waste aspects. Three categories of environmental wastes are
generated from each stage of a product’s life cycle: atmospheric emis-
292 Curran and Schenck

sions, waterborne wastes, and solid wastes. These environmental wastes

are generated by both the actual manufacturing processes and the use of
fuels in transport vehicles or process operations.
Waste management practices. Depending on the nature of the product, a
variety of waste management alternatives may be considered: landfilling,
incineration, recycling, and composting.
Allocation of waste or energy among primary and co-products. Some
processes in a product’s life cycle may produce more than one product.
In this event, energy and resources entering a particular process and all
wastes resulting from it are allocated among the product and co-products.
Allocation is described below in more detail.

3.2.3 Function and Functional Unit Determination

Since any system is a collection of processes connected by flows of intermediate
products, the system is defined by the function that has been selected for study.
This in turn determines the scope of the study, which sets the boundaries and
determines which unit processes will be included. And, because the system is a
physical system, it obeys the law of conservation of mass and energy. Mass and
energy balances are the goal of a life cycle inventory and perform a useful check
on the completeness and validity of the data.
The selection of inputs and outputs to model the system are based on
the functional unit. Comparisons between systems are made on the basis of the
same function, quantified by the same functional unit. For example, one might
compare sending a letter via e-mail and via regular mail. The system function
is the transfer of the document. The system function is measured in func-
tional units, such as thousands of pages transferred and the distance of their
transfer. Functional units permit direct comparisons between disparate systems
and processes.
As another example, the function of drying hands in a public facility can
be evaluated comparing a paper towel system to an air-dryer system. The
functional unit may be expressed in terms of the identical number of pairs of
hands dried for both systems. For each system, it is necessary to determine the
mass and energy flows through the product system based on the functional unit,
e.g., the average mass of paper used or the average volume of hot air required to
dry hands each time. In cases like this, where the use is small, it is common to
multiply the product use in order to get to numbers that are easier to work with
(for example, 1000 times to dry hands).
3.2.4 Co-Product Allocation Procedures
Allocation procedures are needed when dealing with systems involving multiple
products (i.e., co-products and by-products). The materials and energy flows as
Pollution Prevention and Life Cycle Assessment 293

well as associated environmental releases and resource uses must be allocated to

the different products in the inventory process. A basis for co-product allocation
needs to be selected, with careful attention paid to the specific items calculated.
Allocation can be made based on mass, or energy content, or economic consid-
erations (the market value of the different products). Allocation on a mass basis
is most often used and is appropriate in many circumstances.
A mass-based allocation calculation begins with the identification of the
functional unit that is being used to assess the product system. As an example,
1000 units (e.g., pounds or kilograms) of a final product are being produced in a
process that generates 30 units of waste as depicted in Figure 5.
The final product is produced in a two-step process. In the first step, 600
units of raw material A and 420 units of raw material B are reacted to form the
intermediate product. The process results in the production of a co-product along
with 20 units of waste. In the second step, the intermediate product is mixed with
150 units of raw material C and 60 units of additive to form the final product.
This step results in the formation of 10 units of waste. The question then is how
much of the 20 units of waste from the first step should be allocated to the release
inventory for the production of the final product? First completing the mass
balance, we see that 800 units of the intermediate product is being produced along
with 200 units of the co-product. Using these output numbers to calculate the
mass ratio, 80% of the 20 units of waste (or 16 units) in the first step should be
allocated to the final product. This gives a total amount of waste of 26 units
(16 + 10 units) for this two-step process based on the functional unit of 1000 units
of final product.

Amount allocated to: Final product Intermediate product

Raw material A 480 120

Raw material B 336 84
Process waste 1 16 4
Raw material C 150
Additive 60
Process waste 2 10

There are exceptions to a mass-based calculations which may be appropri-

ate at times, such as the use of the market value of the products and co-products.
For example, in the process of mining a high-value ore, e.g., gold, a large quantity
of mined waste material is produced. Often this waste can be sold as a by-product
to be used as roadbed material. Using a mass-based allocation rule would result
in the majority of the environmental releases and resource use being allocated to
 294

FIGURE 5 Example allocation among mass flows.

Curran and Schenck
Pollution Prevention and Life Cycle Assessment 295

the roadbed material. However, this allocation rule is inappropriate, because the
intent of mining is to produce gold, not roadbed material. In this case, since gold
is sold at a much higher cost per weight than roadbed material, the market value
of the co-products would result in a better allocation of environmental releases
and resource use.
For policy analysis, the functional unit should be set at a level that is
politically significant, e.g., the number of products produced within a state or
within an airshed. A “higher”-level functional unit such as these allows the analyst
to include geographic and time considerations which may be significant to the
interpretation of the results. However the functional unit is set, it must be
consistent with the goal that was defined for the study.
3.2.5 Life Cycle Impact Assessment
The impact assessment component of LCA is a technical, quantitative and/or
semiqualitative process which characterizes and classifies the effects of the
resource requirements and environmental loadings that are identified in the
inventory phase (19). LCIA has three steps:
1. Identification of the impact categories, category indicators, and the
models pertaining to them
2. Classification of inventory data into impact categories
3. Characterization of the inventory to yield numerical values of indica-
tors for each impact category.
Impact categories represent the issues of concern for a particular product
system, and indicators are readily calculated entities that are believed to be well
correlated to the environmental endpoints. Thus, for example, global climate
change might be an impact category, and one classifies all greenhouse gas
emissions into this category. Then, typically one calculates the category indicator
called radiative forcing. The radiative forcing indicator is a mathematical combi-
nation of the atmospheric behavior and optical properties of the separate green-
house gases. Radiative forcing is believed to be well correlated to the endpoints
of global climate change, e.g., droughts and floods, or changes in growing season
or increased storm activity. However, the links between radiative forcing and
these endpoints are not well understood, while the link between the emissions of
greenhouse gases and radiative forcing is very well understood and thoroughly
modeled by the Intergovernmental Panel on Climate Change.
A uniform method for impact assessment has not yet been agreed to, but
the analysis should address all relevant environmental impacts: ecological and
human health impacts as well as resource use. Some effects, such as low-level
radiation or heat and noise pollution that are not easily modeled, because the level
of understanding of the environmental mechanism is low, or because there is poor
 296

FIGURE 6 Impact categories and indicators.

Curran and Schenck
Pollution Prevention and Life Cycle Assessment 297

linkage between the inventory results and effects in the environment. Others
(e.g., habitat modification) are known to play a critical role in environmental
impacts of products (e.g., agricultural products), but are difficult to model
quantitatively. Life cycle impact assessment practice is moving more and more
toward using sophisticated fate and transport models to evaluate indicators of
environmental impacts.
The choice of impact categories and category indicators and models can
drive the collection of inventory data. For example, one might choose to evaluate
only minerals whose reserves are predicted to be depleted within 100 years, or
some other reasonable time frame. This would eliminate the need to gather data
on such materials as bauxite, clay, or iron ore, and would decrease the cost of
inventory collection and management.
To date, no “standardized” listing of impact categories to be used in LCA
has been established, but several categories are employed in common practice, as
shown in Table 5.
The Classification Step. Inventory data need to be classified into the
relevant impact categories for modeling. Some emissions have influence on more
than one environmental mechanism and must be classified into more than one
category. The classic example oif this is oxides of nitrogen, or NOx, which acts
as catalyst in the formation of ground-level ozone (smog), but also is a source of
acid precipitation. These substances must be characterized into both categories.
One form of NOx (nitrous oxide, N2O) is also active as a greenhouse gas. The
classification rules for any LCIA must be clearly reported, so that readers of a
study understand what exactly was done to the inventory data.
The Characterization Step The goal of life cycle impact assessment is
to convert collected inventory inputs and outputs into indicators for each cate-
gory (aggregates can be system-wide, by life cycle stage, or by unit operation).

TABLE 5 Typical Impact Categories

1. Stratospheric ozone depletion
2. Global warming
3. Human health
4. Ecological health
5. Smog formation
6. Nonrenewable resource depletion
7. Land use/habitat alteration
8. Acidification
9. Eutrophication
10. Energy: processing/transportation
298 Curran and Schenck

These indicators do not represent actual impacts, because the indicator does not
measure actual damage, such as loss of biodiversity. However, together, they do
constitute an ecoprofile for a product or service.
While there is no universally accepted “right” list of impact categories or
indicators, basic objectives have been set by the Society of Toxicology and
Chemistry (SETAC) that help define categories:
1. Category definition begins with a specific relevant endpoint. Ideally,
the endpoint can actually be observed or measured in the natural
environment.
2. Inventory data are correctly identified for collection. In principle, those
inventory inputs and outputs which relate to the particular impact are
identified.
3. An indicator describes the aggregated loading or resource use for each
individual category. The indicator is then a representation of the
aggregation of the inventory data.
Figure 7 compares the real-world causes and effects (the environmen-
tal mechanism) with the modeled world of LCIA. There are many differences
between the two. In an LCI, for example, the inventory information is typi-
cally modeled as a constant and continuous flow, while in the real world,
emissions typically occur in a discontinuous fashion, varying from minute
to minute.

FIGURE 7 Comparison of “real-world” endpoints to LCIA indicators.

Pollution Prevention and Life Cycle Assessment 299

Both natural and anthropogenic flows act physically, chemically, and

biologically to produce real impacts on the biota (see Figure 8). This series of
events is called the environmental mechanism.
In the virtual reality of the environmental model, many assumptions and
simplifications are made to yield indicators. Even the best current air dispersion
models are accurate only within a factor of two to three, but the level of accuracy
is getting better all the time. The principle methodological issue in life cycle
impact assessment is the modeling management of often very complex, extended
environmental mechanisms. A listing of all possible endpoint impacts is quite
long and can look like the following suggested list.
I. Toxicity issues
A. Human health considerations
1. Acute human occupational
2. Chronic human by consumer
3. Chronic human by local population
4. Chronic human by occupational
5. Human health
6. Human toxicity by ingestion
7. Human toxicity by inhalation/dermal exposure
8. Inhalation toxicity
B. Ecological considerations
1. Aquatic toxicity
2. Biodiversity decrease
3. Endangered species extinction
4. Environmental toxicity
5. Landfill leachate (aquatic) toxicity
6. Species change
7. Terrestrial toxicity
8. Eutrophication (aquatic and terrestrial)
II. Global issues
A. Atmospheric considerations
1. Acid deposition
2. Acidification potential
3. Global warming potential
4. Stratospheric ozone depletion potential
5. Photochemical oxidation potential
6. Tropospheric ozone
B. Resource considerations
1. Energy use
2. Net water consumption
3. Nonrenewable resource depletion
 300
Curran and Schenck

FIGURE 8 Midpoints versus endpoints (20).

Pollution Prevention and Life Cycle Assessment 301

4. Preconsumer waste recycle percent

5. Product disassembly potential
6. Product reuse
7. Recycle content
8. Recycle potential for postconsumer
9. Renewable resource depletion
10. Resource depletion
11. Resource renewability
12. Source reduction potential
13. Surrogate for energy/emissions to transport materials to recycler
14. Waste-to-energy value
III. Local issues
A. Waste considerations
1. Airborne emissions
2. Hazardous waste
3. Incineration ash residue
4. Material persistence
5. Particulates
6. Toxic content
7. Toxic material mobility after disposal
8. Solid waste generation rate
9. Solid waste landfill space
10. Waterborne effluents
B. Public relation considerations
1. Esthetic (e.g., odor)
2. Habitat alteration
3. Heat
4. Industrial accidents
5. Noise
6. Radiation
C. Environment considerations
1. Local land
2. Local water quality
3. Physical change to soil
4. Physical change to water
5. Regional climate change
6. Regional land
7. Regional water quality

Clarifying the environmental mechanism can help determine when impacts

may be additive or when they are independent and non-additive. Two illustrative
examples are global climate change and stratospheric ozone depletion.
302 Curran and Schenck

Example 1: Global climate change. The conversion of various greenhouse

gases into radiative equivalents is universally applicable based on a
scientifically supported mechanism (once a judgment has been made to
select a time frame for analysis.)
Example 2: Stratospheric ozone depletion. Stratospheric ozone depletion is
caused by the interaction of halogenated free radicals in the upper
atmosphere directly reducing concentrations of ozone. However, many
ozone-depleting agents are effective greenhouse gases as well. In addi-
tion, recent research indicates that greenhouse effects on the lower
atmosphere have led to trapping of energy near the earth, and consequent
cooling of the upper atmosphere. The stratospheric cooling tends to
exacerbate the effects of ozone depleters.
Nevertheless, for the purposes of LCIA models, these two mecha-
nisms are treated separately. This simplification helps develop an overall
view of the environmental impacts of industrial systems at a first-order
level. In fact, although LCIA modeling tends to be technically complex,
one can view LCIAs as extended back-of-the-envelope calculations of
realistic worst-case potential impacts.

The goal in assigning LCI results to the impact indicator categories is to

highlight environmental issues associated with each. Assignment of LCI results
should:

First assign results which are exclusive to an impact category and

Then identify LCI results that relate to more than one impact category,
including
Distinguishing between parallel mechanisms (where a given molecule
is “used up” in its actions), and serial mechanisms, where a molecule
can act in one mechanism, and then in a second mechanism without
losing its potency. SOx acts in parallel mechanisms of allocated be-
tween human health and acidification, while NOx acts in a serial mech-
anism as a catalyst in photochemical smog formation and then in
acidification.

Typically, in impact assessment a “nonthreshold” assumption is used. That

is, inventory releases are modeled for their potential impact regardless of the total
load to the receiving environment from all sources or consideration of the
assimilation capacity of the environment. However, there is a trend, particularly
in Europe, to consider thresholds in evaluating indicators. For example, ground-
level ozone formation is often calculated as an indicator for photochemical smog.
Background levels of ozone are about 20 ppb, while some vegetative damage has
been observed at 40 ppb, and human health effects at 80 ppb. All these levels, as
Pollution Prevention and Life Cycle Assessment 303

well as intermediate levels, have been used in determining indicators for photo-
chemical smog.
If LCI results are unavailable or of insufficient quality to achieve the goal
of the study, then either iterative data collection or adjustment of the goal is
required.
The following sections offer descriptions of current approaches that are
being applied to model some of the impact category indicators listed in Table 4.
The most simplistic models are described in order to offer insight into the types
of approaches that are being considered useful from both a practical aspect as
well as least cost.
Stratospheric Ozone Depletion. Ozone depletion is suspected to be the
result of the release of man-made halocarbons, e.g., chlorofluorocarbons, that
migrate to the stratosphere. For a substance to be considered as contributing to
ozone depletion, it must (a) be a gas at normal atmospheric temperatures,
(b) contain chlorine or bromine, and (c) be stable within the atmosphere for
several years (21).
The most important groups of ozone-depleting compounds (ODCs) are the
CFCs (chlorofluorocarbons), HCFCs (hydrochlorofluorocarbons), halons, and
methyl bromide. HFCs (hydroflourocarbons) are also halocarbons but contain
fluorine instead of chlorine or bromine, and are therefore not regarded as
contributors to ozone depletion.
The ozone depletion potential (ODP) is calculated by multiplying the
amount of the emission (Q) by the equivalency factor (EF)
ODP = Q ⋅ EF
Current status on reporting equivalency factors uses CFC11 as the reference
substance. The equivalency factor is defined as
contribution to stratospheric ozone depletion from n over # years
EFODP =
contribution to stratospheric ozone depletion from CFC11 # years
General LCA practice uses values that represent ODC’s full contribution, but
Table 6 also shows factors for 5, 20, and 100 years for some gases. The ozone
depletion potential (ODP) is calculated by multiplying a substance’s mass emis-
sion (Q) by its equivalency factor. These individual potentials can then be
summed to give an indication of projected total ODP for substances 1 through n
in the life cycle inventory that contribute to ozone depletion:
1
ODP = ∑ (Q ⋅ EFODP)
n

Global Warming. The most significant impact on global warming has

been attributed to the burning of fossil fuels, such as coal, oil, and natural gas.
304 Curran and Schenck

TABLE 6 Equivalency Factors for Ozone Depletion (21)

ODP
g CFC11/g substance

5 20 100
Substance Formula years years years ∞

CFC11 CFCl3 1 1 1 1
CFC12 CF2Cl 0.82
CFC113 CF2ClCFCl2 0.55 0.59 0.78 0.90
CFC114 CF2ClCF2Cl 0.85
CFC115 CF2ClCF3 0.40
Tetrachloromethane CCl4 1.26 1.23 1.14 1.20
HCFC22 CHF2Cl 0.19 0.14 0.07 0.04
HCFC123 CF3CHCl2 0.014
HCFC124 CF3CHFCl 0.03
HCFC141b CFCl2CH3 0.54 0.33 0.13 0.10
HCFC142b CF2ClCH3 0.17 0.14 0.08 0.05
HCFC225ca CF3CF2CHCl2 0.02
HCFC225cb CF2ClCF2CHFCl 0.02
1,1,1,-Trichlorethane CH3CCl3 1.03 0.45 0.15 0.12
Methyl chloride CH3Cl 0.02
Halon 1301 CF3Br 10.3 10.5 11.5 12
Halon 1211 CF2ClBr 11.3 9.0 4.9 5.1
Methyl bromide CH3Br 15.3 2.3 0.69 0.64

Several compounds, such as carbon dioxide (CO2), nitrous oxide (N2O), methane
(CH4), and halocarbons, have been identified as substances that accumulate in the
atmosphere, leading to an increased global warming effect.
For a substance to be regarded as a global warmer, it must (a) be a gas at
normal atmospheric temperatures, and (b) either be able to absorb infrared
radiation and be stable in the atmosphere with a long residence time (in years) or
be of fossil origin and converted to CO2 in the atmosphere (21).
Table 7 is a list of substances that are considered to contribute to global
warming. Equivalency factors, based on carbon dioxide as 1, are shown for each
substance over 20-, 100-, and 500-year spans. The choice of time scale can have
considerable effect on how global warming potential is calculated. The 100-year
time frame is often selected, unless reasons exist that indicate otherwise.
contribution from n to global warming over # years
EFGWP =
contribution from CO2 to global warming over # years
Pollution Prevention and Life Cycle Assessment 305

TABLE 7 Equivalency Factors for Global Warming (21)

GWP
g CO2/g substance

20 100 500
Substance Formula years years years

Carbon dioxide CO2 1 1 1

Methane CH4 62 25 8
Nitrous oxide N2O 290 320 180
CFC11 CFCl3 5000 4000 1400
CFC12 CF2Cl2 7900 8500 4200
CFC113 CF2ClCFCl2 5000 5000 2300
CFC114 CF2ClCF2Cl 6900 9300 8300
CFC115 CF2ClCF3 6200 9300 13000
Tetrachloromethane CCl4 2000 1400 500
HCFC22 CHF2Cl 4300 1700 520
HCFC123 CF3CHCl2 300 93 29
HCFC124 CF3CHFCl 1500 480 150
HCFC141b CFCl2CH3 1800 630 200
HCFC142b CF2ClCH3 4200 2000 630
HCFC225ca CF3CF2CHCl2 550 170 52
HCFC225cb CF2ClCF2CHFCl 1700 530 170
1,1,1-Trichloroethane CH3CCl3 360 110 35
Chloroform CH3Cl 15 5 1
Methylene chloride CH2Cl2 28 9 3
HFC 134a CH2FCF3 3300 1300 420
HFC 152a CHF2CH3 460 140 44
Halon 1301 CF3Br 6200 5600 2200
Carbon monoxidea CO 2 2 2
Hydrocarbons (NMHC)a Various 3 3 3
Partly oxidized Various 2 2 2
hydrocarbonsa
Partly halogenated Various 1 1 1
hydrocarbonsa
aContributes indirectly due to conversion into CO2. Only compounds of petrochemical
origin.
306 Curran and Schenck

The global warming potential (GWP) is calculated by multiplying a

substance’s mass emission (Q) by its equivalency factor. These individual poten-
tials can then be summed to give an indication of projected total GWP for
substances 1 through n in the life cycle inventory that contribute to global
warming:
1
GWP = ∑ (Q ⋅ EFGWP)
n

Nonrenewable Resource Depletion. This impact category models resources

that are nonrenewable, or depletable. The subcategories include:
Fossil fuels
Net non-fuel oil and gas
Net mineral resources
Net metal resources
Some models also include the energy that is inherent in a product that is made
from a petroleum feedstock in order to reflect the amount of stock that was
diverted and is no longer available for use as an energy source.
This category can also reflect land use as a resource. Land that has been
disturbed directly due to physical or mechanical disturbance can be accounted for
as a resource that is no longer available either for human use or for ecological
benefit (such as providing habitat for a certain species). Other subcategories under
the resource category include:
Net marine resources depleted
Net land area
Net water resources
Net wood resources
Scientific Certification Systems (SCS) proposes the following approach in their
Life-Cycle Stressor Effects Assessment (LCSEA) model for calculating net
resource depletion (22). The LCSEA model is based on (a) the relative rates of
depletion of the various resources and (b) the relative degree of sustainability of
the resources.
The model considers the key factors that affect resource depletion and
includes consideration of recycled material as supplementing raw material inputs.
It also takes into account materials that are part of the standing reserve base, i.e.
materials, such as steel in a bridge, that will become available as a recovered
reserve at some future time. Recycling of metals has great significance for the
depletion calculation (see Figure 9).
The elements to be considered in factoring resource depletion include:
Pollution Prevention and Life Cycle Assessment 307

Current world reserves

Raw material input (i.e., the amount used)
Amount recycled (both direct and standing stock)
Waste generation
Natural accretion
The reserve base-to-use ratio can be calculated as follows:
Reserve base (R)
= number of years of remaining use left (at current
Use (U)
use rate)
Use (U)
= % of reserve base used
Reserve base (R)
The recycled resource is linked to the original virgin material use and correspond-
ing reserve base. Emissions are not spatially or temporally lined to the original
virgin unit operation. Accounting for all reserve bases:

Waste (∑W)
Reserve base (R) + recyclable stock (∑S)

The current assumption is that only one iteration of recycling and material
integrity is sustained. If natural accretion is accounted for, the following formula
results:

waste (∑W) − natural accretion (N)

reserve base (R) + subsequent uses (∑S)

Including the time period in the equation, we get:

FIGURE 9 Flow of metals, including standing reserve.

308 Curran and Schenck

(∑W − N) ∆T
Current assumption: ∆T = 50 years
R + (∑S) ∆T

And accounting for baseline reserve bases,

(∑W − N) ∆T + (Rb − R)
Rb = a reserve base baseline
R + (∑S) ∆T

Therefore,

(∑W − N) ∆T + (Rb − R)
Resource depletion factor (RDF) =
R + (∑S) ∆T

Resource depletion of fossil fuels represents a simple application. Accretion is

zero and recycling is nil. Thus, wasted resource equals resource used, or
(W) ∗ T
RDF =
R
The impact for the resource depletion category can then be calculated according
to the formula:
Resource depletion indicator (RD) = resource use ×
resource depletion factor (RDF)
For net resources depleted (or accreted), the units of measure express the
equivalent depletion (or accretion) of the identified resource. All of the net
resource calculations are based on RDFs.

Indicator—net resource Units of measure

Water Equivalent cubic meters

Wood Equivalent cubic meters
Fossil fuels Tons of oil equivalents
Non-fuel oil and gas Tons of oil equivalents
Metals Tons of (metal) equivalents
Minerals Tons of (mineral) equivalents
Land area Equivalent hectares

Acidification. For acidification, an equivalency approach is typically ap-

plied and the stressor flows are converted into SO2 or H+ equivalents. For
example, NO2 is multiplied by 64/(2 ∗ 46) = 0.70, since this is the molar proton
Pollution Prevention and Life Cycle Assessment 309

release potency of NO2 compared to SO2. Table 8 shows sample calculations

using potency factors for an inventory with SO2, NO2, and HCl releases. The
LCSEA approach takes the calculation one step further and includes an emission
loading factor to reflect how much of the inventory release is expected to reach
the receiving environment.
Eutrophication. Eutrophication occurs in aquatic systems when the limit-
ing nutrient in the water is supplied, thus causing algal blooms. In fresh water, it
is generally phosphate which is the limiting nutrient, while in salt waters it is
generally nitrogen which is limiting. In general, addition of nitrogen alone to fresh
waters will not cause algal growth, and addition of phosphate alone to salt waters
will not cause significant effects. In brackish waters, either nutrient can cause
algal growth, depending on the local conditions at the time of the emissions.
Eutrophication is generally measured using the concentration of chloro-
phyll-a in the water. Waters with less than 2 mg of chlorophyll-a per cubic
meter (2 mg chla m–3) are considered “oligotrophic,” while those with 2–10 mg
chla m–3 are considered “mesotrophic,” and those with more than 10 mg
chla m–3 are termed “eutrophic.” Waters over 20 mg chla m–3 are considered
“hypereutrophic.”
As waters become mesotrophic, their species assemblages change, favoring
species that grow rapidly in the presence of nutrients (“weed” species) over those
which grow more slowly. There is some indication that eutrophication in salt
waters is the source of the red tides that are a worldwide problem.
Under eutrophic conditions, the algae in the water significantly block light
passage, while in hypereutrophic conditions the amount of biomass produced is
so high that anoxic conditions occur, leading to fish kills. There are some
indications that similar sorts of effects occur in terrestrial systems as well.
The ratio of carbon to nitrogen to phosphorus in aquatic biomass is 106:16:1
(23), on an atomic basis. This ratio is the basis of combining nitrogen and
phosphorus in calculating the eutrophication potential of emissions.
(Molar quantity of nitrate + nitrite + ammonia) × Redfield ratio
+ molar quantity of phosphate × [endpoint characterization factor
(fresh, salt water)] = eutrophication indicator
Eutrophication is typically measured in PO4 equivalents. The EPA has set a
concentration of 25 µg PO4 L–1 as the level needed to protect fresh-water aquatic
ecosystems from eutrophication.
Energy. While inventory analyses involves the collection of data to quan-
tify the relevant inputs and outputs of a product system, the accounting of
electricity as a flow presents a unique challenge. The use of energy audits makes
the idea of balancing energy flows around a process a familiar one. However, in
LCA the reporting of energy flows is in itself insufficient to perform a subsequent
 310

TABLE 8 Calculating Acidification “Emission Loading” (22)

Inventory LCI result Potency Molar equivalent Characterization Emission loading
Unit operation emission (ton/30a) factor (ton/30a) factor (ton/30a)

Coal mining/transport SO2 31,620 1 31,620 0.5 15,810

NO2 9,660 0.7 6,762 0.3 2,029
HCl 270 0.88 238 0.5 119
CaO product/transport SO2 240 1 240 0.15 36
NO2 1,260 0.7 882 0.075 66
Coal use SO2 50,190 1 50,190 0.15 7,529
NO2 36,480 0.7 25,536 0.075 1,915
HCl 15,210 0.88 13,385 0.15 2,008
Total 128,853 29,512
Curran and Schenck
Pollution Prevention and Life Cycle Assessment 311

TABLE 9 Equivalency Factors for Acidifiers (21)

Mw EF kg SO2/
Formula Conversion g ⋅ mol n kg substance

SO2 SO2 + H2O → H2SO3 → 2H+ + SO32− 64.06 2 1

SO3 SO3 + H2O → H2SO4 → 2H+ + SO42− 80.06 2 0.8
NO2 NO2 + 1⁄2H2O + 1⁄4O2 → 2H+ + SO32− 46.01 1 0.7
NO NO + O3 + 1⁄2H2O → H+ + NO3− + 3⁄4O2 30.01 1 1.07
HCl HCl → H+ + Cl− 36.46 1 0.88
HNO3 HNO3 → H+ + NO3− 63.01 1 0.51
H2SO4 H2SO4 → 2H + SO42− 98.07 2 0.65
H3PO4 H3PO4 → 3H+ + PO43− 98 3 0.98
HF HF → H+ + F− 20.01 1 1.6
H2S H2S + ⁄2O2 + H2O → 2H+ + SO32−
3 34.03 2 1.88
NH3 NH3 + 2O2 → H+ + NO3− + H2O 17.03 1 1.88

impact assessment. Ideally, the environmental impacts associated with energy

generation should be captured in the approach. That is, the generation of electric-
ity from fossil fuels should also show the contribution to the emission of global
warming gases, solid waste (especially coal ash), etc. This type of detail also
allows for the consideration of the use of waste materials in energy recovery
operations. Also, the calculation of energy flow should take into account the
different fuels and electricity sources used, the efficiency of conversion and
distribution of energy flows, as well as the inputs and outputs associated with
generation and use of that energy flow. In addition, a more robust assessment may
consider an evaluation of the specific sources of electrical power that are
contributed to the national energy grid on a more regional approach. This type of
consideration is important in determining local impacts. For example, electricity
that is produced in Maine is not used in California. Therefore, the impacts of
electricity generation based on a national average may not be appropriate.
In the absence of a readily available model that can convert energy-related
inventory data into potential impacts based on the fuel source, a fallback position
can be to look at the source of the total energy used and identify what percentage
is obtained from the national energy grid (which is mainly fossil fuels) and what
percentage comes from other sources, such as the burning of waste materials. At
this high-level decision point, this information is appropriate and the approach
fits the indicator-by-indicator comparison framework.
3.2.6 Weighting
Weighting, also called valuation, assigns relative weights to the different impact
indicator categories based on their perceived importance. Since there are various
312 Curran and Schenck

ways in which different individuals consider things to be important, formal

valuation methods should make this process explicit and be representative of the
individual or group making the final decision.
ISO 14042 requires that weighting of individual categories only be done
after fully disclosing unweighted indicators. When comparing two systems, the
trade-offs between impacts often require a judgment call to be made in order to
arrive at a decision.
Table 10 shows the partial results of an evaluation that was conducted at
Fort Eustis, Virginia, as part of ongoing efforts to reduce waste generation from
chemical agent-resistant coating (CARC) depainting/painting operations (24).
This example focuses on a portion of the evaluation that compared the baseline
CARC system with an alternative system using a different primer and thinner
combination. The proposed switch to the alternative primer/thinner system was
identified as a possible way to reduce the facility’s air releases and potential
contribution to global climate change.

TABLE 10 Environmental Impact Scores for Baseline and

Alternative CARC Systems (24)

Spatial scale Impact categorya Baseline Alternative

Global ODP 1.090 0.367

GLBLWRM 1.013 0.984
FSLFUELS 1.263 1.180
Regional ACIDDEP 1.198 1.175
SMOG 1.114 0.992
WTRUSE b b

Local Toxicity:
HUMAN 2.150 1.793
ENVTERR 3.799 2.862
ENVAQ 1.280 3.540
LANDUSE 1.577 1.585
aODP = ozone depletion potential; GLBLWRM = global warming
potential; FSLFUELS = fossil fuel & mineral depletion potential;
ACIDDEP = acid deposition potential; SMOG = smog creation potential;
WTRUSE = water use; HUMAN = human health toxicity potential;
ENVTERR = terrestrial wildlife toxicity potential; ENVAQ = aquatic biota
potential; LANDUSE = land use for waste disposal.
bWater use was not reported as an impact because water availability

is plentiful where CARC operations are located, and because water is

typically treated and reused or released to the environment.
Pollution Prevention and Life Cycle Assessment 313

After life cycle inventory data for the raw materials, painting, and disposal
of the baseline CARC and alternative system were collected, additional impact
information was then included to complete the LCA. A valuation process was
conducted on nine selected impact categories using the analytical hierarchy
process (AHP) in order to assign weights to the categories. AHP is a recognized
methodology for supporting decisions based on relative preferences of pertinent
factors.
It should be recognized that valuation is inherently a subjective process. In
the CARC study, the results of the valuation process indicated that relative to this
particular group, the greatest potential environmental concern is ozone depletion
(weight = .332). Water use was included in the valuation process, but it was not
included in the impact assessment since water is plentiful near CARC operations,
and because water is treated and reused or released to the environment. The
weights of all the impact categories (in order of decreasing importance) were
determined to be as follows:

Ozone depletion .332

Acidification .189
Global warming potential .124
Human health .099
Photochemical smog formation .097
Land use .058
Fossil fuel use .037
Water use .025
Terrestrial toxicity .020
Aquatic toxicity .020

These weights were multiplied by the normalized inventory data to arrive

at the scores shown in Table 10. For most of the impact categories, the difference
is not great enough to conclude that there is a preference between these systems.
However, for ozone depletion (ODP) and aquatic toxicity (ENVAQ), some
differences can be noted. While the ozone depletion score appears to decrease
(1.090 to 0.367), showing potential improvement, the environmental aquatic
toxicity score appears to increase (1.280 compared to 3.540). Looking back at the
inventory data, it is noted that the increased aquatic toxicity is due to increased
cadmium and chlorine releases to the wastewater associated with manufacturing
the ingredients for the alternative primer.
If the decision is made in favor of selecting the alternative system because
of its potentially lower impact on the ozone layer, it is now clear that this decision
may result in an increased burden on the wastewater system. The benefit of using
life cycle data to support the decision-making process is that the decision is being
314 Curran and Schenck

made in a broader context and with recognition of how the production of the
alternative product can be factored in. If, on the other hand, concerns are more
immediate and focused on the local aquatic environment, with a higher weight
being assigned to aquatic toxicity, the final decision could go the other way, with
a preference for the baseline system, depending on whether the inventory data are
sufficient to influence the results in addition to an increased weight being placed
on aquatic toxicity. In either case, such weighting schemes should be made very
explicit in the final analysis.
3.2.7 Interpretation
In the interpretation step of LCA, the results of the inventory and impact modeling
are analyzed, conclusions are reached, and findings are presented in a transparent
manner. It is critical that the report that results from this activity is clear, complete,
and consistent with the goal and scope of the study. ISO 14043 lists key features
of life cycle interpretation as follows:
The use of a systematic procedure to identify, qualify, check, evaluate, and
present the conclusions based on the results of an LCA or life cycle
interpretation (LCI), in order to meet the requirements of the application
as described in the goal and scope of the study;
The use of an iterative procedure both within the interpretative phase and
with the other phases of an LCA or LCI
The provision of links between LCA and other techniques for environmen-
tal management by emphasizing the strengths and limits of an LCA study
in relation to its defined goal and scope
Transparency throughout the interpretation phase is essential. Whenever
preferences, assumptions, or value choices are used in the assessment or in
reporting, these need to be clearly stated in the final report. The goal of life cycle
interpretation is to give credibility to the results of the LCA in a way that is useful
to the decision maker.

3.3 Life Cycle Costing

Over 30 years ago, the U.S. Department of Defense recognized that operation and
maintenance (O&M) costs were substantial components of the total costs of
owning equipment and systems. In fact, ownership costs can far outweigh the costs
of procurement. By considering the full costs over the life cycle of the system and
the time value of money (e.g., discounting), better choices can be made.
The broader practice of environmental accounting now uses words such as
total cost analysis/assessment and life cycle costing to emphasize that traditional
approaches overlook important environmental costs (and potential cost savings
and revenues). A firm’s cost accounting system traditionally serves as a way to
Pollution Prevention and Life Cycle Assessment 315

track and allocate costs to a product or process for operational budgeting, cost
control, and pricing. In life cycle costing, accurate allocation serves to identify
environmental impacts in order to achieve pollution prevention across the entire
life cycle.
Life cycle costing has not yet achieved a single functional definition and
has been used to mean different things. However, the concept behind it refers to
the management application of environmental accounting (e.g., cost accounting,
capital budgeting, process/product design) across the life span of a product or
process. It is difficult to discern life cycle costing from total cost assessment
(TCA), because TCA is sometimes used to refer to a specific application of
environmental accounting, such as the life span of a technology or process. TCA
is often used to refer to the act of adding environmental costs into capital
budgeting, whereas life cycle costing is used more frequently when incorporating
environmental accounting into the entire design of a process or product (25).
It is essential to determine the scope of environmental costs to be included
in a life cycle costing evaluation, including not only a firm’s private costs only
(i.e., those that directly affect the firm’s bottom a line), but also private and
societal costs, some of which do not show up directly or even indirectly in the
firm’s bottom line. An expanded accounting approach is described in the EPA’s
Pollution Prevention Benefits Manual (26). The manual distinguishes among four
levels of costs:
Usual costs (Tier 0): Equipment, materials, labor, etc.
Hidden costs (Tier 1): Monitoring, paperwork, permit requirements, etc.
Liability costs (Tier 2): Future liabilities, penalties, fines, etc.
Less tangible costs (Tier 3): Corporate image, community relations, con-
sumer response, etc.
Further, there is an important distinction between costs for which a firm is
accountable and costs resulting from a firm’s activities that do not directly affect
the firm’s bottom line:
Private costs are the costs incurred by a business or costs for which a
business can be held responsible. These are the costs that directly affect
a firm’s bottom line. Private costs are sometimes termed internal costs.
Societal costs are the costs of activities, anywhere within the life cycle,
which impact on the environment and on society for which the product
manufacturer is not directly held financially responsible. These costs do
not directly affect the company’s bottom line. Societal costs are also
referred to as external costs or externalities. They may be expressed
qualitatively, in physical terms (e.g., tons of releases, exposed receptors),
or quantitatively, in dollars and cents. Societal costs can be divided as
being either environmental costs or social costs.
316 Curran and Schenck

Life cycle costing includes all internal plus external costs incurred through-
out the life cycle of a product or process. External costs are not borne directly by
the company (or the ultimate consumer of the company’s goods or services) and
do not typically enter the company’s decision-making process. The use of
electricity can be used to demonstrate the difference between internal and external
costs. The generation of electrical power imposes various environmental impacts
and costs. Facility construction, operation, and maintenance are costs that are
incurred by the electrical generators, who recover the costs through the prices
they set to sell their electricity. Other impacts are not borne by the generator and
are not reflected in the price. For example, fossil fuel plants emit sulfur dioxide
and nitrogen oxides, precursors to acid rain. Life cycle costing would attempt to
describe qualitatively or place a dollar value on those impacts to reflect the overall
cost to society and the environment, such as human health risk, damage to
buildings and other structures (e.g., statues), damage and loss of trees and other
plant life, alteration of habitat and resulting animal species loss, etc.
Uncovering and recognizing environmental costs associated with a product,
process, system, or facility is an important goal for making good management
decisions. Attaining such goals as reducing environmental expenses, increasing
revenues, and improving future environmental performance requires paying
attention to current and potential future environmental costs. Whether or not a
cost is “environmental” is not critical; the goal is to ensure that relevant costs
receive appropriate attention.
Inherent in life cycle costing are the same considerations that were dis-
cussed in conducting a life cycle inventory: costs that are omitted may skew the
results. Also, life cycle costing cannot be used to compare disparate products, but
it is a tool for assessing comparable products or processes. Further, the function
of the products being compared should be equivalent.

4 CONCLUSIONS
Pollution prevention is a valuable concept for facility managers tasked with
environmental protection. It is a method that allows them to think about their
operations and identify opportunities to improve their operations. The main goal
of pollution prevention is to reduce or eliminate the creation of pollutants and
wastes at the source in order to reduce costs and to meet or exceed federal and
state regulations on environmental discharges and emissions. Over the years,
significant work has been done by various government offices, universities, and
industry to demonstrate pollution prevention techniques and effectively transfer
this information to wider audiences for implementation. A wealth of material on
case studies for many different industrial sectors can be found in the open
literature on this subject.
Pollution Prevention and Life Cycle Assessment 317

A life cycle perspective in combination with pollution prevention elevates

the concept by looking beyond a single process or facility to encompass the
environmental aspects that may be affected somewhere else within the entire
system. This type of holistic approach to identifying secondary consequences
leads the thought process toward sustainability rather than simple environmental
protection. It is LCA’s key message and the reason why LCA is becoming widely
accepted as the basis for approaches to environmental management. The system-
atic application of life cycle thinking in all aspects of decision making, including
process improvement, product selection, and end-of-life management, provides a
stronger model for environmental management than does simple pollution pre-
vention. The information that an LCA provides allows for better-informed deci-
sion making to occur. As a result, LCA is an environmental management tool and
model that is quickly being adopted at the international level. The LCA provides
information that is useful not only to the individual facility or corporation but to
environmental policy makers in governments.
LCA is a relatively recent technique in environmental protection and
sustainability, but much has been learned in the relatively short period of time
that has been dedicated to this subject. It is increasingly obvious that life
cycle-based approaches are needed to fully evaluate environmental impacts in all
our decisions and choices. The wide spectrum of activities involved in a product
or process requires that practitioners and method developers test new ways to
model LCA and exchange information among themselves and share it with
potential users (i.e., environmental decision makers) in order to advance the
understanding and application of LCA. LCA is an evolving tool that continues to
improve as better site-specific models become coupled with simpler ways to com-
municate results to the users of life cycle data.

REFERENCES
1. Harry Freeman, Pollution Prevention. In Harry M. Freeman (ed.), Industrial Pollution
Prevention Handbook. New York: McGraw-Hill, 1995.
2. L. Case, L. Mendicino, and D. Thomas, Developing and Maintaining a Pollution
Prevention Program. In Harry M. Freeman (ed.), Industrial Pollution Prevention
Handbook. New York: McGraw-Hill, 1995.
3. U.S. Environmental Protection Agency, Facility Pollution Prevention Guide, EPA/
600/R-92/088. Cincinnati, OH: Risk Reduction Engineering Laboratory, May 1992.
4. The Society of Environmental Toxicology and Chemistry, Life-Cycle Impact Assess-
ment: The State-of-the-Art, Larry Barnthouse, Jim Fava, Ken Humphreys, Robert
Hunt, Larry Laibson, Scott Noesen, James Owens, Joel Todd, Bruce Vigon, Keith
Weitz, John Young (eds.), Pensacola, FL: SETAC Foundation, 1997.
5. U.S. Environmental Protection Agency, Developing and Using Production-Adjusted
318 Curran and Schenck

Measurements of Pollution Prevention, EPA/600/R-97/048. Cincinnati, OH: National

Risk Management Research Laboratory, September 1997.
6. David P. Evers, Facility Pollution Prevention Planning. In Harry M. Freeman (ed.),
Industrial Pollution Prevention Handbook. New York: McGraw-Hill, 1995.
7. U.S. Environmental Protection Agency, Development of Computer Supported Infor-
mation System Shell for Measuring Pollution Prevention Progress, EPA/600/R-
95/130, NRMRL, Cincinnati, OH: National Risk Management Research Laboratory,
August 1995.
8. B. W. Baetz, E. I. Pas, and P. A. Vesiland, Planning Hazardous Waste Reduction and
Treatment Strategies: An Optimization Approach. Waste Manage. Res., vol.7, no. 2,
pp. 153–163, 1989.
9. Kenneth Humphreys and Paul Wellman, Basic Cost Engineering. New York: Marcel
Dekker, 1996.
10. International Standards Organization, Environmental Management—Life Cycle
Assessment—Principles and Framework, ISO 14040, 1997.
11. International Standards Organization, Environmental Management—Life Cycle
Assessment—Goal and Scope Definition and Inventory Analysis, ISO 14041, 1998.
12. International Standards Organization, Environmental Management—Life Cycle
Assessment—Life Cycle Impact Assessment, ISO 14042, 2000.
13. International Standards Organization, Environmental Management—Life Cycle
Assessment—Life Cycle Interpretation, ISO 14043, 2000.
14. J.A. Fava, R. Denison, B. Jones, M. A. Curran, B. W. Vigon, S. Selke, and J. Barnum
(eds.), A Technical Framework for Life Cycle Assessments. Pensacola, FL: The
Society of Environmental Toxicology and Chemistry, 1991.
15. K. Stone and J. Springer, Review of Solvent Cleaning in Aerospace Operations and
Pollution Prevention Alternatives, Environ. Prog., vol. 14, no. 4, pp. 261–272, 1995.
16. U.S. Environmental Protection Agency, Streamlined Life-Cycle Assessment of 1,4-
Butanediol Produced from Petroleum Feedstocks versus Bio-Derived Feedstocks, in
Cincinnati, OH: National Risk Management Research Laboratory, September 1997.
17. The Society of Environmental Toxicology and Chemistry, Streamlined Life Cycle
Assessment, Joel Ann Todd and Mary Ann Curran (eds.), Pensacola, FL, June 1999.
18. U.S. Environmental Protection Agency, Life Cycle Assessment: Inventory Guidelines
and Principles, EPA/600/R-92/245. Cincinnati, OH: Risk Reduction Engineering
Laboratory, February 1993.
19. U.S. Environmental Protection Agency, Life-Cycle Impact Assessment: A Conceptual
Framework, Key Issues, and Summary of Existing Methods. prepared by the Research
Triangle Institute (RTI), July 1995.
20. U. de Haes, O. Jolliet, G. Finnveden, M. Hauschild, W. Krewitt, and R. Mueller-
Wenk, Best Available Practice Regarding Impact Categories and Category Indicators
in Life Cycle Impact Assessment, SETAC Life Cycle Impact Assessment Workgroup
discussion paper, February 1999.
21. Henrik Wenzel, Michael Hauschild, and Leo Alting, Environmental Assessment of
Products, London, UK: Chapman & Hall, 1997.
22. S. Rhodes, F. Kommonen, and R. Schenck, Evolution of Life-Cycle Assessment as
an Environmental Decision-Making Tool: ISO 14042 and Life Cycle Stressor Efforts
Assessment (LCSEA), Workshop booklet, February 1998.
Pollution Prevention and Life Cycle Assessment 319

23. A. C. Redfield, The Process of Determining the Concentration of Oxygen, Phosphate,

and Other Organic Derivatives within the Depths of the Atlantic Ocean. Pap. Phys.
Ocean. Meteor. 9, 1942.
24. U.S. Environmental Protection Agency, Life Cycle Assessment for Chemical Agent
Resistant Coating, EPA/600/R-96/104, prepared by Battelle and Lockheed-Martin for
the National Risk Management Research Laboratory, Cincinnati, OH, 1996.
25. Allen White, D. Savage, and K. Shapiro, Life Cycle Costing: Concepts and Applica-
tions. In M. A. Curran (ed.), Environmental Life Cycle Assessment. New York:
McGraw-Hill, 1996.
26. U.S. Environmental Protection Agency, Pollution Prevention Benefits Manual, EPA230/
R-98/100, October 1989.
27. U.S. Environmental Protection Agency, Pathway to Product Stewardship: Life-Cycle
Design as a Business Decision-Support Tool, EPA/742/R-97/008. Office of Pollution
Prevention and Toxics, December 1997.

GLOSSARY
Functional unit The measure of a life cycle system used to base reference
flows in order to calculate inputs and outputs of the system
Inventory See Life cycle inventory.
ISO International Standards Organization (or International Organization
of Standardization).
Life (1) Economic: that period of time after which a product, machine, or
facility should be discarded because of its excessive costs or reduced
profitability. (2) Physical: that period of time after which a product,
machine, or facility can no longer be repaired in order to perform its
designed function properly.
Life cycle assessment Evaluation of the environmental effects associated
with any given activity from the initial gathering of raw materials from
the earth to the point at which all materials are returned to the earth; this
evaluation includes all releases to the air, water, and soil.
Life cycle cost The sum of all discounted costs of acquiring, owning,
operating, and maintaining a project over the study period (i.e., the life
of the product or process). Comparing life cycle costs among mutually
exclusive projects of equal performance has been used as a way to
determine relative costs.
Life cycle impact assessment A scientifically based process or model
which characterizes projected environmental and human health impacts
based on the results of the life cycle inventory.
Life cycle inventory An objective, data-based process of quantifying
energy and raw material requirements, air emissions, waterborne efflu-
ents, solid waste, and other environmental releases throughout the life
cycle of a product, process, or activity.
320 Curran and Schenck

Pollution prevention The use of materials, processes, or practices that

reduce or eliminate the creation of pollutants or wastes at the source.
Pollution prevention opportunity assessment The systematic process of
identifying areas, processes, and activities which generate excessive
waste streams or waste by-products for the purpose of substitution,
alteration, or elimination of the waste.
POTW (Publicy Owned Treatment Works) Any device or system used
to treat (including recycling and reclamation) municipal sewage or
industrial wastes of a liquid nature that is owned by a state, municipality,
intermunicipality, or interstate agency [defined by Section 502(4) of the
Clean Water Act].
RCRA (Resource Conservation and Recovery Act of 1976) Amending
the Solid Waste Disposal Act (SWDA), the RCRA established a regula-
tory system to track the generation of hazardous substances from the time
of generation to disposal. The U.S. Congress declares it to be the national
policy of the country that, whenever feasible, the generation of hazardous
waste is to be reduced or eliminated as expeditiously as possible. Waste
that is nevertheless generated should be treated, stored, or disposed of so
as to minimize the present and future threat to human health and the
environment (40 USC 6902).
Waste minimization Approaches or techniques that reduce the amount of
RCRA-regulated wastes generated during industrial production pro-
cesses; the term applies to recycling and other efforts to reduce waste
volume.
16
Application of Life Cycle Assessment

W. David Constant
Louisiana State University and A&M College, Baton Rouge, Louisiana

1 INTRODUCTION
Application of life cycle assessment (LCA) ensures that environmental impact is
explicitly included in the design process, yielding the “best” alternative. The
“best” choice can be difficult in the final analysis to assess, as many factors may
cause an objective LCA to fall into a gray or subjective area. Graedel (1) presents
an excellent methodology for streamlined LCA with a matrix approach, yielding
objective results, and explores the approaches used by major manufacturers in
obtaining data for the matrices. Others (2,3) present basic studies and applica-
tions, and there are many articles and texts available between basics and detailed
methods. The objective of this chapter is to explore the application of LCA for
waste site remediation, as an example of the extension of LCA to areas beyond
manufacturing goods and consumer products.

2 BASICS FROM CHEMICAL ENGINEERING

Application of LCA to assess a waste site remedy is making use of the basics of
chemical engineering and related fields, the material and energy balances, with a
few other topics included, such as economics, eco- and/or health risk assessment,

321
322 Constant

and regulatory constraints. Key components for assessment in the material and
energy balance approach are (a) the problem basis, (b) boundaries, (c) generation
and accumulation, and (d) time. These terms and their components must be
consistent in analysis of, for example, technology comparisons in remediation or
waste management methods for treatment of a process stream or contaminant-
media matrix. In terms of a general equation, we may write for any LCA
assessment:

Input = output + generation + accumulation (1)

It should be noted that the generation term may be positive or negative

(if material is consumed). It is critical that the terms are handled in a consistent
fashion for use in LCA. The basis of “measurement” is key—should we look at,
as with paper cups versus ceramic cups, the number of cups or the number of
cups per use? The decision depends on what we are investigating in the LCA.
Also, in extension of LCA to site remediation, the value of the source term, for
example, the mass of waste within the source of a groundwater contaminant
plume, may be very difficult to obtain without significant uncertainty associated,
making the LCA an estimate with significant error. The accumulation term is
critical for understanding the burden with time for unsteady-state processes. Since
time is included in the accumulation term, the length of the LCA, or time line,
must be selected initially to satisfy stakeholders in the application (regulators,
landowners, industry, public, etc.). One should determine what interval for study
is appropriate, as regulations use years, decades, and in some cases centuries as
windows into model predictions of the fate and transport. Generation components
will include transformation, reaction, and other terms that change the physical or
chemical properties of the material of interest within the boundaries being
explored. Boundaries must be carefully defined in order to ease the solution of
the mathematical portion of the problem, be representative of the system under
study, and also to reflect the focus of the LCA being performed. Equations may
be written from this basic form for any process wherein boundaries can be
determined, in either differential form or integral manner. Further, those without
accumulation terms are considered steady state, and of course, many processes
have numerous streams to consider. The matrix approach is typically utilized in
such process balances, due to the complexity found when processes are linked for
a final solution, to ease algebra. Herein we obtain the overall balances and
individual component balances for overall matrix solutions. Energy balances are
handled in a similar fashion, incorporating thermodynamic properties and allow-
ing for both open and closed systems, determined by the transfer of mass across
a boundary. A batch process would be considered closed by definition and
continuous systems are defined as open for energy balance purposes. Basics of
Application of Life Cycle Assessment 323

material and energy balances many be found in a number of chemical engineering

texts, such as Felder and Rousseau (4) and others.

3 ADDITION OF CHEMODYNAMICS
In order to take LCA beyond its usual scope to assess a waste site remedy, we
need not only to include the mass and energy balances for environmental
compartments, but also the models and mechanisms of inter- and intramedia
transport. The concepts and models of chemodynamics are provided in significant
detail by Thibodeaux (5), and by Reible and Choi (6). For the purposes of this
discussion, it is assumed that appropriate models of the fate and transport of
contaminants for a case are known, estimated, or available in some fashion. Since
the objective of the LCA in remediation is to assess the burden being placed on
the environment by various remedies, the application of realistic and appropriate
models is essential and found in references above (5,6) for certain cases.

4 APPLICATION
With the objective being assessment of the environmental (eco- and human
health) burden from various remedies in a waste site problem, we may employ,
as with LCA, risk assessment in the reverse, following the methodology of
Hwang (7) as extended by Constant et al. (8). The exposure is summed from
different routes to get total exposure, Et, which, to have an acceptable risk, should
not exceed the reference dose level (RL). Exposures are calculated as in any other
assessment, based on concentration, time, body weight, etc. Key here is the
concentration at the exposure point (receptor), which is a function of the concen-
tration in the soil or other medium. This function is the subject of much modeling
work as described above, as it is essential to have accurate representations of the
fate and transport of the contaminants in and across media under investigation.
Thus, one may apply the above exposure method for an acceptable risk, combined
with LCA, as follows (8). First, land use or other resource needs are established,
followed by assuming a chemical management or remediation scheme or treat-
ment train concept. Then, incorporating regulations, laws, and liabilities, chemi-
cals are traced from “cradle to grave” throughout the process as in a typical LCA.
Next, risk is assessed for the scheme (7) with all material and energy balances
incorporated, along with eco-risk as appropriate. With these balances completed
including appropriate fate and transport models, economics of the process(es)
being considered are established. If the economics are acceptable, then the
technology or waste processing is appropriate for the present level of available
technology considered (or assumed) in the first step. If the economics are
unfavorable, another treatment, manufacturing, or remediation scheme should be
324 Constant

applied to this integrated management methodology for comparison to the first

attempt. Thus, methods can be evaluated for risk-based remediation or manufac-
turing as in LCA for the most economical approach. It should be noted here that
technology is on a moving time line, and what is successful and economical today
will need to be reassessed several times in the future as chemical fate and
transport modeling, our understanding of the environment, technologies, and
regulations all change. In this manner, “new” technologies such as monitored
natural attenuation (MNA) can be assessed alongside and with active remedies as
described in the example below.

5 EXAMPLE OF ASSESSMENT
The Petro Processors, Inc. (PPI), site is one of the most significant Superfund sites
in the United States. While it is not directly under Superfund, due to agreement
among the parties and existing consent decree in the U.S. District Court, Middle
District of LA, it provides an excellent example, via hindsight, of the utility of
LCA for risk-based remediation methodology as described by Constant et al. (8).
Details of the site(s), being named Brooklawn and Scenic, for the nearby
roadways, may be found at EPA’s Region VI Web site (9), and are not described
here. The site(s) contain approximately 400,000 tons of chlorinated hydrocarbon
wastes, including hexachlorobenzene, hexachlorobutadiene, TCE, DCE, lower
chlorinated products, and numerous other petrochemical wastes of similar nature.
The mixture forms a dense, nonaqueous, viscous phase, which is currently
covered and maintained by hydraulic containment of the groundwater and recov-
ery of the organic phase (both of which are treated on-site) under strict regulatory
requirements. Louisiana State University (LSU) has served as the Court’s Expert
for the last 10 years regarding cleanup of this site, with the role being research
into advanced technologies and assisting the court in monitoring and assessing
the remediation.
Just prior to LSU’s involvement, the remedy was to remove the waste,
stabilize the material, and place it in a lined 1 × 106 yd3 vault, with some
hydraulic containment and recovery of site areas found difficult to remove due to
the high water table. However, due to volatile emissions exceeding fence-line
limits, the removal method was abandoned. The next remedy to be put in place
was active hydraulic containment (expansion of the previous pumping layout),
with several hundred wells to remove organics and contaminated water over the
long term in order to protect the underlying aquifer and prevent migration. To
date, only about 1% of the free-phase material has been removed, but containment
appears successful, at the expense of treating millions of gallons of water
containing only trace contamination levels. As this pumping method was being
installed at Brooklawn, via ongoing research by LSU and site personnel for
Application of Life Cycle Assessment 325

augmentation of the pumping operation by both passive and active technologies,

intrinsic biodegradation and hence natural attenuation was studied and then
incorporated into the remedy. The addition of biodegradation, sorption, disper-
sion, etc. as required within the lines of evidence of MNA as presented by EPA
(10), has had a significant impact on the remedy of the PPI site.
It appears at this time that active pumping of significant water volumes may
be greatly reduced, while still focusing on source removal (NAPL), with the
understanding that significant residuals remain bound in these systems. The MNA
component is proposed to contain and reduce the size of contaminated ground-
water over time, with acceptable risk. Thus, using the same receptors and risk
issues, incorporation of MNA into the PPI remedy promises to reduce signifi-
cantly the cost of the remedy, without increasing eco- or human health risk. In a
recent comparison, starting with the same source and endpoints, in the same time
frame, MNA with limited hydraulic containment (a few source removal wells)
was evaluated against active hydraulic containment and recovery (many pumping
wells across the site) for the Scenic site of PPI. This assessment included
monitoring and other costs associated with these technologies, and a 70% cost
savings was found by incorporation of MNA into the remedy. This was due
mainly to the reduced number of wells to be installed and the significant reduction
in treatment of produced water. While this example case of remedy comparison
has occurred during the remediation of PPI, it clearly shows that we have the tools
of LCA, risk-based remediation, and MNA at our disposal. When properly
integrated, one can provide acceptable waste management schemes prior to
initiation of a manufacturing project or site remediation. However, as technology
changes, as stated earlier, this assessment must be ongoing.

REFERENCES
1. T. E. Graedel, Streamlined Life-Cycle Assessment, pp. 97–98. Englewood Cliffs, NJ:
Prentice Hall, 1998.
2. P. C. Schulze (ed.), Measures of Environmental Performance and Ecosystem Condi-
tion. Washington, DC: National Academy Press, 1999.
3. R. J. Walter, Practical Compliance with the EPA Risk Management Program. New
York: Center for Chemical Process Safety of the AIChE, 1999.
4. R. M. Felder and R. W. Rousseau, Elementary Principles of Chemical Processes,
pp. 81–86. New York: Wiley, 1978.
5. L. J. Thibodeaux, Chemodynamics: Environmental Movement of Chemicals in Air,
Water and Soil. New York: Wiley, 1979.
6. B. Choy and D. D. Reible, Diffusion Models of Environmental Transport. Boca
Raton, FL: Lewis Pub., CRC Press, 2000.
7. S. T. Hwang. J. Environ. Sci. Health, A, vol. A27, no. 3, pp. 843–861, 1992.
8. W. D. Constant, L. J. Thibodeaux, and A. R. Machen, Environmental Chemical
326 Constant

Engineering: Part I—Fluxion; Part II—Pathways. Trends Chem. Eng., vol. 2,

pp. 525–542, 1994.
9. U.S. Environmental Protection Agency (EPA) Region VI Website, Information on
Superfund sites, Louisiana, Petro Processors, Inc., http://www.epa.gov/earth1r6/
6sf/6sf-la.htm.
10. Robert S. Kerr Environmental Research Center (U.S. EPA), Natural attenuation short
course materials, Ada, OK, December 2–4, 1997.
17
Risk-Based Pollution Control and
Waste Minimization Concepts

Gilbert J. Gonzales
Los Alamos National Laboratory, Los Alamos, New Mexico and
New Mexico State University, Las Cruces, New Mexico

1 INTRODUCTION
Ecological risk assessment is defined as “the qualitative or quantitative appraisal
of impact, potential or real, of one or more stressors (such as pollution) on flora,
fauna, or the encompassing ecosystem.” The underlying principles behind risk
reduction and integrated decision making that are detailed in U.S. Environmental
Protection Agency (EPA) strategic initiatives and guiding principles include
pollution prevention (1). Pollution control (PC) and waste minimization (WM)
are probably the most effective means of reducing risk to humans and the
environment from hazardous and radioactive waste. Pollution control can be
defined as any activity that reduces the release to the environment of substances
that can cause adverse effects to humans or other biological organisms. This
includes pollution prevention and waste minimization. Waste minimization is
defined as pollution prevention measures that reduce Resource Conservation and
Recovery Act (RCRA) hazardous waste (2). Reduced risk is one benefit of these
practices, and it results most directly from lower concentrations of contaminants
entering the environment from both planned and accidental releases.

327
328 Gonzales

Although it is difficult to quantify the reductions in the release of contam-

inants to the environment that have resulted from reductions in waste, the
reductions have most assuredly reduced risk to humans and the environment
posed by toxicants. Using examples from Los Alamos National Laboratory
(LANL), we will discuss the interrelatedness of pollution prevention/waste min-
imization with risk reduction and how risk assessment can be generally applied
to the field of pollution prevention and waste minimization. While emphasis in
this chapter is on “ecological risk assessment,” the concepts and principles can
also be applied to human health risk assessment. For purposes of this chapter,
distinction is not made between pollution prevention, waste minimization, re-
cycling or other waste management techniques, and other related terms. Rather,
emphasis is on source reduction, which includes any practice that reduces the
amount of contaminant entering a waste stream or the environment.
It is evident that the EPA’s hierarchy of general pollution prevention and
waste minimization methods is implicitly related to risk reduction (2). The
preferred method, source reduction, results in the greatest reduction in human and
ecological risk from contaminants. Source reduction is followed in effectiveness
by recycling, treatment, and disposal. While it is possible, depending on the
technology used, that recycling and treatment may increase risk to worker health
because of increases in contact handling of waste, the prioritized list (source
reduction → recycling → treatment → disposal) generally results in a decrease
in risk to the public and the environment as one progresses from the least
preferred to the most preferred method. The reason for this is simple. The
preferred method results in little, if any, release of contaminants into the environ-
ment compared to the less preferred methods; with the less preferred methods,
not only does the quantity of contaminants potentially released into the environ-
ment increase, the potential to release them increases. With this premise, it is then
important to realize the magnitude of risk reduction achieved by employing
pollution prevention and waste minimization at facilities such as LANL.

2 SITE DESCRIPTION AND CHARACTERIZATION

LANL is located in north-central New Mexico, approximately 60 miles northwest
of Santa Fe (Figure 1). LANL is a U.S. Department of Energy-owned complex
managed by the University of California that was founded in 1943 as part of the
Manhattan Project to create the first nuclear weapon. Since then, LANL’s mission
to design, develop, and test nuclear weapons has expanded to other areas of
nuclear science and energy research.
The Laboratory comprises dozens of individual technical areas located on
43 square miles of land area; about 1400 major buildings and other facilities are
part of the Laboratory. The Laboratory is situated on the Pajarito Plateau, which
consists of a series of fingerlike mesas separated by deep east-to-west–oriented
Risk-Based Pollution Control 329

Los Alamos National Laboratory

Technical Area boundaries
County boundaries
S T
R E Other political boundaries
F O

LOS ALAMOS
0 0.5 1 2 mi L Major paved roads
A

COUNTY
N
0 0.5 1 2 km I O RIO ARRIBA COUNTY
T
LOS ALAMOS COUNTY
SANDOVAL COUNTY

SANTA FE COUNTY
A SANTA FE COUNTY
N
E SANTA FE
F
NATIONAL N
To
A

FOREST
T

Espanola
N
A

30
S

LOS ALAMOS

502
Pa 502 502 To
jar East
ito Jem
Ro ez R 4 BANDELIER Santa Fe
501 ad oad NAT. MON.
SANDO-
VAL CO.
S A N I L D E F O N S O
P U E B L O L A N D S

LO 4 WHITE
SA 4 ROCK
LA
SAN MO
SC
OU DO
NT VA
LC
BANDELIER Y
OU
NT
NATIONAL MONUMENT Y
e
nd
ra
G
io
R

U. S. A.
cARTography by A. Kron 4/5/00
UT CO
Tierra Amarilla TAOS
AZ NM OK COUNTY

TX RIO ARRIBA COUNTY Taos

LOS ALAMOS
COUNTY
Taos
Los Alamos Los Alamos
Grants ★ Santa Fe Santa Fe
Albuquerque SANDOVAL
N E W M E X I C O COUNTY SANTA
FE
Socorro Bernalillo
COUNTY
Albuquerque

BERNALILLO
Las Cruces
COUNTY

FIGURE 1 Location of the Los Alamos National Laboratory.

330 Gonzales

canyons cut by intermittent streams (Figure 2). Mesa tops range in elevation from
approximately 7800 ft on the flanks of the Jemez Mountains to about 6200 ft at
their eastern termination above the Rio Grande.
Researchers at Los Alamos work on initiatives related to the Laboratory’s
central mission of enhancing global security as well as on basic research in a
variety of disciplines related to advanced and nuclear materials research, devel-
opment, and applications; experimental science and engineering; and theory,
modeling, analysis, and computation. As a fully functional institution, LANL also
engages in a number of related activities including waste management; infrastruc-
ture and central services; facility maintenance and refurbishment; environmental,

FIGURE 2 Overhead view of the topography in and around the Los Alamos
National Laboratory.
Risk-Based Pollution Control 331

ecological, cultural, and natural resource management; and environmental resto-

ration, including decontamination and decommissioning. As result of the scien-
tific and technical work conducted at Los Alamos, the Laboratory generates,
treats, and stores hazardous, mixed, and radioactive wastes.
About 2120 contaminant potential release sites (PRSs) have been identified
at LANL. The LANL PRSs are diverse and include past material disposal areas
(landfills), canyons, drain lines, firing sites, outfalls, and other random sites such
as spill locations. Categorizing contaminants into three types—organics, metals,
and radionuclides—Los Alamos has all three present. The contaminants include
volatile and semivolatile organics, polychlorinated biphenyls (PCBs), asbestos,
pesticides, herbicides, heavy metals, beryllium, radionuclides, petroleum prod-
ucts, and high explosives (3). The primary mechanisms for potential contaminant
release from the site is surface-water runoff carrying potentially contaminated
sediments and soil erosion exposing buried contaminants. The main pathways by
which released contaminants can reach off-site residents are through infiltration
into alluvial aquifers, airborne dispersion of particulate matter, and sediment
migration from surface-water runoff. Like many other sites, the predominant
pathway by which contaminants enter terrestrial biological systems is the inges-
tion of soil, intentional or not.
Diverse topography, ecology, and other factors make the consideration of
issues related to contamination in the LANL environment complex. “Since 1990,
LANL’s environmental restoration project has conducted over 100 cleanups. The
environmental restoration project has also decommissioned over 30 structures and
conducted three RCRA closure actions during this period. Schedules have been
published for the planned cleanup of approximately 700 to 750 additional sites.
This schedule encompasses a period of about 10 years, beginning with fiscal year
1998. The number of cleanups per year varies from approximately 100 in fiscal
year 2002 to 18 in fiscal year 2008. An important and integral part of this
pollution prevention technology and of identifying interim protection measures is
ecological risk assessment” (3).

3 POLLUTION PREVENTION AND WASTE MINIMIZATION

AT LANL
The pollution prevention program at the Laboratory has been successful in
reducing overall LANL wastes requiring disposal by 30% over the last 5 years.
The program is site wide but has facility-specific components, especially for the
larger generators of radioactive and hazardous chemical wastes. Past reductions
indicate that waste generation in the future should be less than that projected. The
Site Pollution Prevention Plan for Los Alamos National Laboratory (4) describes
the LANL Pollution Prevention and Waste Minimization Programs, including a
general program description, recently implemented actions, specific volume
332 Gonzales

reductions resulting from recent actions, and current development/demonstration

efforts that have not yet been implemented.
More specifically, LANL has achieved reductions in the generation of
hazardous waste, low-level radioactive waste (LLW), and mixed LLW. These and
two other waste types are defined as follows:
Hazardous waste—Any substance containing waste that is regulated by
RCRA, the Toxic Substances Control Act, and New Mexico as a Special
Waste.
Low-level radioactive waste (LLW)—Radionuclide-containing substances
with a radionuclide activity (sometimes referred to as concentration) of
less than 100 nCi/g.
Mixed LLW—Substances containing both RCRA constituents and LLW.
Transuranic radioactive (TRU) waste—Radionuclide-containing substances
with a radionuclide activity (concentration) equal to or greater than
100 nCi/g.
Mixed TRU waste—Substances containing both RCRA and TRU waste.
Estimated reduction rates of chronically generated waste, by waste type,
over a 6-year period are (D. Wilburn, personal communication, 2000)
Hazardous waste: 11%/yr (65% from 1993 to 1999)
LLW: 11%/yr (67% from 1993 to 1999)
Mixed LLW: 12%/yr (72% from 1993 to 1999)
Production of TRU and mixed TRU waste combined has increased by an
average of 38%/yr (228% from 1993 to 1999).
The Laboratory has dozens of pollution prevention projects ongoing and
planned. Some of the efforts are pollution prevention and waste minimization in
the strictest sense and some (e.g., separation of waste types or satellite treatment
followed by centralized treatment) are pollution control in the broadest sense,
including efforts where cost savings is the primary goal and pollution control is
a secondary benefit. Three examples of pollution control/waste minimization at
LANL follow.
Generator Set-Aside Fee-Funded (GSAF) Plutonium Ingot Storage Cubicle
Project. “An aliquot casting and blending technique is under imple-
mentation at LANLs Plutonium Facility. The aliquot process allows
out-of-specification plutonium to be blended with other plutonium so that
the final mixed batch meets specifications and is uniform. This avoids
the cost and waste generation related to reprocessing out-of-specification
plutonium ingots through the nitric acid line. In addition the more
uniform product will reduce the reject rate and will avoid reprocessing
and remanufacturing wastes. This project will fabricate a storage system
Risk-Based Pollution Control 333

with 20 cubicles. It is expected that 12.5 m3 of TRU waste can be avoided

over the 25 year life of the cubicle system. This project has an estimated
100% return on investment when prior investments are included in the
base project cost. This project is the final step necessary to achieve
aliquot blending. The GSAF program is funding purchase of the storage
cubicle and the operating group has programmatic support for installing
and qualifying the cubicle.”
GSAF Reduction of Acid Waste and Emissions. “The Laboratory’s Analyti-
cal Chemistry Sciences Group’s performance of analytical services at one
of the Laboratory’s technical areas requires the dissolution of up to
20,000 samples per year. The current process is hot plate digestion which
requires large quantities of chemicals, mostly acids, and results in the
volatilization and release to the atmosphere of 90% of those chemicals.
The yearly consumption of HNO3, HCl, HF, HClO4 and NH4OH is
estimated to be 3395 kgs. About 3095 kgs of these chemicals are
volatilized and become air emissions from the stacks; the balance is
diluted and discharged to the Laboratory’s Radioactive Liquid Waste
Treatment Facility via the LLW acid line. The stack emissions constitute
about 30% of the Laboratory’s annual hazardous air pollutant discharges.
By switching to a microwave and muffle furnace oven process, annual
consumption of the chemicals listed above can be reduced to about 370
kgs, an 89% reduction. It is estimated that implementing the new process
will reduce the Laboratory’s hazardous air pollution discharges by three
tons. The estimated return on investment for this project is 82%.”
Waste Minimization and Microconcentric Nebulization for Inductively Cou-
pled Plasma-Atomic Emission Spectroscopy. “One of the most popular
techniques for multi-element analysis is inductively coupled plasma
atomic emission spectroscopy (ICP-AES). The most common method of
introducing samples into the ICP torch is pneumatic nebulization. The
standard nebulizers, in conjunction with typical spray chambers, exhibit
very low transport efficiency. Because of this inefficiency the ICP-AES
generates almost a liter per day of rinse water that consists primarily of
dilute nitric acid mixed with other contaminants. The other contaminants
can be TRU waste, LLW, mixed LLW or hazardous waste depending on
the samples analyzed. Under this project an existing microconcentric
nebulizer will be deployed with the ICP-AES and optimized for analysis
of trace elements in a variety of plutonium-containing matrices. It is
expected that using an optimized microconcentric nebulizer will reduce
the daily rinse water from the spray chamber drain to about 50 ml. This
represents a 95% or 200 l/yr reduction in the volume of waste requiring
disposal. A 50 liter per year reduction in LLW plastic sample containers
is expected. Reducing the volume of samples in the glovebox will
334 Gonzales

simplify the ICP-AES operation, reducing various risk parameters. The

return on investment ranges from 10–50%.”

4 ECOLOGICAL RISK ASSESSMENT

Ecological risk assessment is defined as “the qualitative or quantitative appraisal
of impact, potential or real, of one or more stressors (such as pollution) on flora,
fauna, or the encompassing ecosystem.” The Laboratory is employing the EPA’s
iterative and tiered approach to ecological risk assessment whereby less complex
assessments with greater conservatism and uncertainty are employed first, fol-
lowed by the advancement of potential problem areas and contaminants to
progressively more complex and realistic assessments with lower uncertainty. The
purpose of this section is not to present ecological risk assessment methods, as
there are many excellent sources on this subject (5–8). Rather, the purpose is to
introduce concepts in ecological risk assessment that could be considered in
designing pollution control and waste minimization programs.

4.1 History at the Los Alamos National Laboratory

Ecological risk assessment (“ecorisk”) at the Laboratory is a work in progress.
Methods for ecorisk screening and “tier 2” assessments have been in development
and implementation since approximately 1993. Most screenings and assessments
completed at the Laboratory thus far are based on the U.S. EPA hazard quotient
method, whereby hazard quotient values are calculated for receptors for each
contaminant by area and may be thought of as a ratio of a receptor’s exposure at
the site to a safe limit or benchmark:
n
HIi = ∑ HQij (1)
j=1

where
HIi is the hazard index for receptor I to n contaminants of potential concern
(COPCs), and
exposureij
HQij = (2)
safe limitij
where
HQij is the hazard quotient for receptor I to COPC j
exposureij is the dose to receptor i for COPC j
safe limit is an effects or no effects level, such as a chronic no-observed-
adverse-effects level (NOAEL)
Risk-Based Pollution Control 335

4.1.1 Screening
In 1995 a very conservative method for screening contaminants and contaminated
areas was developed (9). The method involved the selection of ecological
endpoints that focused on animal feeding guilds, a listing of candidate contami-
nants, exposure/dose–response estimation, estimation of food and soil ingestion,
and risk characterization. A comparison value or safe limit was based on foraging
mode, behaviors, types of food consumed, the amount consumed, and NOAELs
or radiation dose limits when available. The safe limits for radionuclides were
largely laboratory-derived rat-based values used in human risk assessments.
As such, this and the use of other conservative factors or assumptions resulted in
a method that likely overestimated potential risk by orders of magnitude, espe-
cially for radionuclides. At least one application of the method is known to have
occurred (10).
Currently at LANL, ecorisk screening encompasses qualitative “scoping
evaluation” as the basis of problem formulation and “screening evaluation” to
identify contaminants of potential concern (COPCs) by exposure media. The
screening evaluation focuses on identifying sites that require further investigation
and risk characterization (11). A key component to the screening evaluation, and
one of interest to PC/WM, is the ecological screening level (ESL) concept. An
ESL is basically a contaminant safe limit, or acceptable effects level, below which
measurable effects are not expected. ESLs are most useful in units of contaminant
amount per quantity of soil, so that measurements of contaminant levels in soil
can be quickly and directly compared. The ESLs are developed for each ecolog-
ical receptor of interest, each chemical, and are media specific. They are deter-
mined so that if an area has levels of a chemical above the ESL in any medium,
then the area is deemed to pose a potentially unacceptable risk to ecological
receptors. Calculations of ESLs require toxicity information, including toxicity
reference values (TRVs), preferably chronic NOAELs, and knowledge of transfer
coefficients including bioconcentration and bioaccumulation factors. Details on
this information and on the process for calculating and selecting ESLs are
documented in a LANL report (11); however a summary is provided here.
Nonradionuclides. “Although soil ESLs are based on exposure of terres-
trial receptors—plants, invertebrates (earthworms), and wildlife—they are deter-
mined differently for each receptor. The different approaches are required because
of the different ways that toxicological experiments are performed for these
organisms. For plants, earthworms and other soil-dwelling invertebrates, effects
are based on the concentration of a COPC in soil. Therefore, ESL values are
directly based on effects concentrations and modeling is not required. For plants
and invertebrates the soil ESL was ‘back-calculated’ from No-Observed-Effect-
Concentrations. Exposure to wildlife, however, is dependent on exposure of the
organism to a chemical constituent from a given medium (such as soil or
336 Gonzales

foodstuff) through direct and indirect means (i.e., ingestion, inhalation, and
dermal). Wildlife ESL values were based on the dietary regimen of the receptor,
including consumption of plants, invertebrates, and vertebrate flesh, with some
incidental soil ingestion (11).” The soil ESL was “back-calculated” as the soil
concentration of a COPC resulting from exposure to the NOAEL. Starting with
the EPA’s general terrestrial wildlife exposure model (12), inverting to convert
soil concentration to dose, relating food intake to soil intake, and solving for
COPC- and receptor-specific ESLs, models such as Eq. (3) for omnivores were
used to compute ESLs for nonradionuclides:
NOAELij
ESLij = (3)
Ii ⋅ (fsi + fpi ⋅ TFplant,j + fii ⋅ TFinvert,j
where
ESLij is the soil ESL for omnivore i and COPC j (mg/kg)
NOAELij is the NOAEL for omnivore i and COPC j (mg/kg/day)
fsi is the fraction of soil ingested by omnivore i, expressed as a fraction of
the dietary intake
Ii is the normalized daily dietary ingestion rate for omnivore i (kg/kg/day)
fpi is the fraction of plants in diet for omnivore i
TFplant,j is a unitless transfer factor from soil to plants for COPC j
fii is the fraction of invertebrates in diet for omnivore i
TFinsect,j is a unitless transfer factor from soil to insects for COPC j
For any given COPC, the soil ESL used for initial screening was often the lowest
receptor-specific soil ESL value among plants, invertebrates, robin, kestrel,
shrew, mouse, cottontail, and fox. Models were developed for sediment and water.
More detail can be found in Ref. 11.
Radionuclides. Radionuclide dose limits are the equivalent of the
NOAELs used to develop nonradionuclide ESLs (11). For screening at LANL,
radionuclide ESLs were based on the International Atomic Energy Agency
(IAEA)-recommended dose limit of 0.1 rad per day (13). At LANL the radionu-
clide dose to terrestrial biota was taken as the sum of the dose from internally
deposited radionuclides and external dose from gamma emitting radionuclides in
soil [Eq. (4)].
Total acceptable dose = 0.1 (rad/day) = internal dose
(4)
+ external dose
Detailed formulas can be found in Ref. 11. Conservative assumptions about the
size of the organism, its diet, the geometry of the contaminated source, and the
location of the receptor relative to the contaminated source are used by LANL for
estimating internal and external doses for screening purposes. The internal dose
Risk-Based Pollution Control 337

to wildlife is calculated by multiplying the effective energy of a radionuclide by

the body burden of that radionuclide in an organism. The external dose is the dose
coefficient for skin exposed to contaminated water or soil. Calculations for
estimating internal and external doses from radionuclides in soil can be found in
literature by Higley and Kuperman (14). At LANL, ESLs are developed sepa-
rately for plants and invertebrates versus wildlife, and decay and metabolic
elimination rates are incorporated (11).
4.1.2 Integration with Safety Analyses
In 1994 the Laboratory began a pilot project to develop an ecological risk
assessment method that could be used to augment hazards and safety analyses
(15). Specifically, the Laboratory explored the possiblity of using an ecological
risk assessment approach to explicitly incorporate environmental consequence
analysis into hazards analyses and probabilistic risk assessment processes. Inves-
tigators used a two-phase approach: (a) Using five COPCs, conservative eco-
logical screening was interacted with worst-case accident analysis results to
determine whether ecological impacts could be excluded from further investiga-
tion or would require more detailed assessment; (b) using a dynamic ecological
transport model (BIOTRAN), a more detailed and realistic assessment was
conducted of the land area that the screening process identified as most suscepti-
ble to potential adverse impacts from the five contaminants. BIOTRAN (now
BIOTRAN.2) is an ecological transport model used for predicting the flow of
organic and inorganic contaminants, including radionuclides, through simulated
ecosystems—which can include plant, animal, and human populations—using
biomass vectors (16). Twenty-two subroutines include variables of meteorol-
ogy, soil physics, hydrology, geology, plant and animal physiology, limnology,
radiation biology, ecology, epidemiology, risk analysis, and health physics.
“The results of the detailed assessment were consistent with the findings of the
screening assessment, thus validating the screening assessment approach in
the context of hazards analysis.”
4.1.3 Tier 2 Assessments
From 1997 to 1999, Gonzales et al. completed tier 2 preliminary ecological risk
assessments of federally protected threatened and endangered species (17–19).
Risk assessments of the Mexican spotted owl (Strix occidentalis lucida),
the American peregrine falcon (Falco peregrinus), the bald eagle (Haliaeetus
leucocephalus), and the Southwestern willow flycatcher (Empidonax traillii
extimus) were performed using a custom FORTRAN model, ECORSK.5, and the
geographic information system (GIS) (17–19). Using Eq. (5), estimated doses
resulting from soil ingestion and food consumption were compared against TRVs
for nonradionuclides and radiation dose limits for radionuclides. This generated
hazard indices (HIs) that included a measure of additive effects from multi-
338 Gonzales

ple contaminants (radionuclides, metals, and organic chemicals) and included

bioaccumulation and biomagnification factors.
ncs ncoc
If /BW∑ Oi ENHi ∑ [Fs + (1 − Fs) BMFj] Csi,j
i=1 j=1
HI = (5)
TRV
where
HI =
hazard index, cumulative for all summed COPC hazard quotients
If =
food consumption, kgdwt/day
BW =
body weight of organism, kgfwt
ncs =
total number of nest sites
Oi =
occupancy factor for the ith feeding site
ENHi =
enhancement factor for the ith feeding site
ncoc =
total number of different COPCs found over all nesting sites
Fs =
% soil in diet
BMFj =
biomagnification factor via food chain transport for the jth
contaminant
Csi,j = COPC concentration of soil/sediment, mg/kgdwt, for the ith
feeding site and the jth COPC
TRVj = toxicity reference value (“safe limit”), mg/kg-body weight/day,
for the jth COPC
Other terms were sometimes included, such as calculating sediment intake from
prey data and inferring BMFs through differential equations.
ECORSK.5 was used to integrate environmental contaminant data, species-
specific biological, ecological, and toxicological information, and GIS spatial
information as depicted in Figure 3. Using ECORSK.5, the risk assessor is able
to scale home ranges (HRs) so that a rectangular HR simulates foraging patterns
based on natural features such as topography or prey distribution. ECORSK.5 is
also coded to allow the option of weighting the simulated foraging process on an
exponential basis as based on the distance from randomly selected nest sites such
that a species forages more in the vicinity of its nest (or any other focal point)
than it does at the extremity of its HR. A second manner in which the simulated
foraging process is weighted in the tier 2 process is based on animal distribution
and other population data. For example, a GIS elk habitat use model based on
positions fed from radio collars via a Global Positioning System was used to
create an occupancy scheme whereby the amount of time spent foraging on any
one area is specified. Coding schemes identify home ranges, habitat core areas,
species, and occupancy weights.
ECORSK.5 generates information on risk by specific geographic location
for use in management of contaminated areas, species habitat, facility siting,
Risk-Based Pollution Control 339

GIS

EEU and NEST

coordinates FIMAD

Contaminant
CONVRT location and
concentration
LOAEL/
MAPCDE.DAT NOAEL
toxicology

Database
development
Species-
specific INRSK.DAT
input
EEUINP.DAT

OUTRSK.DAT
ENTRSK.DAT
ECORSK.5 GIS.DAT
HQ.DAT
HQP.DAT
HQPC.DAT
HQPO.DAT HABIT.DAT
HQPCO.DAT GRIDXY.DAT

Select

PLTRSK

2-D and 3-D plots

FIGURE 3 Schematic respresenting the integration of ecological risk assess-

ment parameters by ECORSK.5.
340 Gonzales

and/or facility operations in order to maintain risk from contaminants at accept-

ably low levels. Several different levels of risk breakdown, from very summarized
values to very detailed values, are generated by ECORSK.5. Figure 4 shows one
level of breakout, where the risk contribution by Aroclor-1254 from different
discrete locations in a HR is presented graphically. Thus, the strength of this
custom FORTRAN model is for animals with large individual home ranges or
populations with large distributions. ECORSK.5 is documented in an internal
report (20).

4.2 Ecorisk Assessments Applied

Results of the tier 2 assessments described above indicated a small potential for
impact to the peregrine falcon, but the studies concluded no appreciable potential
impact was expected to the Mexican spotted owl, the bald eagle, or the South-
western willow flycatcher. With all the hazardous substances used at the Labora-
tory over the years, including radionuclides, an application of a pesticide
unrelated to Laboratory operations 37 years ago resulted in the contaminant
that dominated the risk contribution for the peregrine falcon. In 1963,
dichlorodiphenyltrichloroethane (DDT) was applied to one-half million acres of
forest west of the Rio Grande at an application rate of about 140,000 ppm (21).
DDT and dichlorodiphenylethelyne (DDE), a DDT breakdown product, were two
of the three top contaminants in several modeling scenarios for the peregrine
falcon. Since this is currently a widely diffuse source that largely originated in
1963, PC/WM cannot be applied to this chemical, raising the point that not all
risk issues can be addressed in the design of a PC/WM program. Nevertheless,
the risk assessments summarized above also provide an example of how risk
information can be used to augment PC/WM activities as follows. The top three
contaminants, DDT, DDE, and a PCB mixture—Aroclor-1254—contributed a
total of 81% of the risk under conditions of one of several modeled foraging
scenarios. In part this resulted in a charter at the Laboratory to give added
emphasis to PCB pollution control (A. Jackson, personal communication, 2000).
The increased effort aims to
Adopt a PCB-free program for authorized sources as an institutional
sponsored program (consistent with Laboratory policy of zero environ-
mental incident)
Incorporate into the yearly budgetary exercise a schedule to replace PCB
items
Place greater division-level organizational emphasis on PCB-free facilities
Gain senior management endorsement of an approach to address PCB
vulnerabilities based on an ecological-risk concept as a coordinated
institutional program, i.e., as a Laboratory taking a leadership and
proactive role in working with regulators and the Department of Energy
 0.5 0.45-0.5
0.4-0.45
0.45
0.35-0.4
0.3-0.35
0.4
0.25-0.3

0.35 0.2-0.25
0.15-0.2
Risk-Based Pollution Control

0.3 0.1-0.15
0.05-0.1
0.25 0-0.05

0.2

Hazard Quotient
0.15

0.1
125
0.05 85
70
0 61 Grid cell row

0
30
60
50

75
Nesting Habitat

81
87
93
99
25

105
111
117
129
150
165
191
197

0
203

Grid cell column

209
215
221
227
233
255
341

285

FIGURE 4 Three-dimensional plot of ecological risk hazard quotient contributions for Aroclor-1254 by discrete location.
342 Gonzales

in arriving at a cost-effective and scientifically sound risk management

framework for decision making that meets performance goals and is
protective of the environment).
Implementing pollution controls such as those listed above is no small task
at a sprawling 43-mi2 industrial site, and costs can be substantial. In the example
above, risk assessments helped focus field studies such that in the first year
following the completion of the assessments approximately $122,000 was saved
by targeting specific contaminants and omitting the analysis of those deemed
inconsequential from a risk perspective. Thus, the consideration of risk issues in
the design of PC/WM programs can improve the effectiveness and efficiency of
the programs.
As mentioned earlier, the Laboratory is employing the iterative and tiered
approach to ecological risk assessment whereby less complex assessments with
greater conservatism and uncertainty are employed first, followed by the advance-
ment of potential problem areas and COPCs to progressively more complex and
realistic assessments with lower uncertainty.

4.3 Risk-Based Prioritization of PC/WM COPC Targets

As discussed in the previous section, development of the ecorisk screening
method at LANL included the calculation of screening values termed ESLs.
Table 1 is a list of soil ESLs by contaminant in decreasing order (22). For each
contaminant ESLs were developed for eight terrestrial species. The values in
Table 1 are for the species that usually had the most stringent (lowest) ESL. ESL
values are still undergoing revision at LANL, but the relative rank is not expected
to change much.
Ecological screening levels are related directly to toxicity and expo-
sure, therefore ranking contaminants on the basis of ESL value gives some
indication of the contaminants, which if eliminated or reduced by PC/WM
techniques, would most effectively reduce risk to flora and fauna. However,
before prioritizing contaminants from a PC/WM perspective, one must also
consider other factors, some practical, such as the ubiquity of contaminants in
the environment because if a contaminant were quite toxic but was absent or
present at very low levels in the environment relative to its ESL, elimination
or reduction of the contaminant would gain little in terms of risk reduction.
For example, since 2,3,7,8-tetrachlorodibenzodioxin is ranked first in Table
1, thus can be considered relatively very toxic, a large degree of risk reduction
could be achieved by PC/WM efforts that minimize emissions or releases of
it into the environment. However, its presence in the environmental contaminant
database used in the threatened and endangered species assessments discussed
above was sparse, therefore it is not likely a good candidate for PC/WM
consideration.
Risk-Based Pollution Control 343

TABLE 1 Examples of Ecological Screening Levels (ESLs) in Soil Used

for Screening-Level Ecological Risk Assessments at the Los Alamos
National Laboratory

Typea Name CAS no. ESLb Rank

dioxin/furan Tetrachlorodibenzodioxin[2,3,7,8-] 1746-01-6 1.80E-06 1

pest DDE[4,4′-] 72-55-9 1.80E-03 2
pest DDT[4,4′-] 50-29-3 2.80E-03 3
pcbs Aroclor-1248 12672-29-6 3.00E-03 4
pest Endrin 72-20-8 3.40E-03 5
inorg Mercury (methyl) 22967-92-6 1.60E-02 6
inorg Thallium 7440-28-0 3.30E-02 7
pest BHC[gamma-] 58-89-9 3.40E-02 8
pest Dieldrin 60-57-1 4.10E-02 9
inorg Mercury (inorganic) 7439-97-6 5.00E-02 10
pcbs Aroclor-1260 11096-82-5 5.00E-02 11
pcbs Aroclor-1254 11097-69-1 1.20E-01 12
svocs Di-n-butylphthalate 84-74-2 1.30E-01 13
he m-Dinitrobenzene 99-65-0 1.60E-01 14
pcbs Aroclor-1242 53469-21-9 1.60E-01 15
pest Kepone 143-50-0 1.60E-01 16
pest Heptachlor 76-44-8 1.80E-01 17
Inorg Silver 7440-22-4 2.00E-01 18
pahs Naphthalene 91-20-3 2.00E-01 19
svocs Bis(2-ethylhexyl)phthalate 117-81-7 2.40E-01 20
inorg Vanadium 7440-62-2 2.50E-01 21
he Trinitrotoluene[2,4,6-] 118-96-7 3.00E-01 22
svocs Pentachlorophenol 87-86-5 3.20E-01 23
inorg Chromium VI 18540-29-9 3.50E-01 24
pest Endosulfan 115-29-7 3.50E-01 25
inorg Antimony 7440-36-0 5.00E-01 26
inorg Selenium 7782-49-2 5.00E-01 27
inorg Cobalt 7440-48-4 5.10E-01 28
inorg Arsenic 7440-38-2 5.70E-01 29
he dinitrotoluene[2,6-] 606-20-2 7.40E-01 30
pest BHC[beta-] 319-85-7 9.70E-01 31
inorg Beryllium 7440-41-7 1.00E+00 32
inorg Cadmium 7440-43-9 1.00E+00 33
inorg Cyanide 57-12-5 1.00E+00 34
he HMX 2691-41-0 1.00E+00 35
vocs Xylenes 1330-20-7 1.10E+00 36
he dinitrotoluene[2,4-] 121-14-2 1.20E+00 37
svocs Chlorophenol[2-] 95-57-8 1.20E+00 38
pahs Benzo(a)pyrene 50-32-8 1.80E+00 39
344 Gonzales

TABLE 1 Continued
Typea Name CAS no. ESLb Rank

vocs Acetone 67-64-1 1.80E+00 40

vocs Trichloroethene 79-01-6 1.90E+00 41
pest Chlordane[alpha-] 5103-71-9 2.10E+00 42
pest Chlordane[gamma-] 5103-74-2 2.10E+00 43
he Tetryl 479-45-8 2.20E+00 44
pahs Dibenz(a,h)anthracene 53-70-3 2.30E+00 45
pahs Acenaphthene 83-32-9 2.50E+00 46
inorg Chromium, total 7440-47-3 3.10E+00 47
pcbs Aroclor-1016 12674-11-2 3.20E+00 48
pahs Benzo(a)anthracene 56-55-3 3.30E+00 49
pahs Chrysene 218-01-9 3.30E+00 50
vocs Trichlorobenzene[1,2,4-] 120-82-1 3.50E+00 51
svocs Chloro-3-methylphenol[4-] 59-50-7 3.60E+00 52
vocs Tetrachloroethene 127-18-4 3.80E+00 53
he 4-amino-2,6-dinitrotoluene 19406-51-0 4.20E+00 54
inorg Aluminum 7429-90-5 5.00E+00 55
inorg Manganese 7439-96-5 5.00E+00 56
inorg Uranium 7440-61-1 5.00E+00 57
he o-Nitrotoluene 88-72-2 5.00E+00 58
he RDX 121-82-4 5.10E+00 59
inorg Boron 7440-42-8 5.70E+00 60
pahs Methylnaphthalene[2-] 91-57-6 5.90E+00 61
he 2-amino-4,6-dinitrotoluene 35572-78-2 6.10E+00 62
he m-Nitrotoluene 99-08-1 6.10E+00 63
vocs Methylene chloride 75-09-2 6.60E+00 64
he sym-Trinitrobenzene 99-35-4 7.00E+00 65
pahs Benzo(b)fluoranthene 205-99-2 7.40E+00 66
svocs Phenol 108-95-2 7.90E+00 67
svocs Benzoic acid 65-85-0 8.20E+00 68
pest Methoxychlor[4,4′-] 72-43-5 8.40E+00 69
svocs Pentachloronitrobenzene 82-68-8 8.80E+00 70
inorg Zinc 7440-66-6 1.00E+01 71
he p-Nitrotoluene 99-99-0 1.10E+01 72
pest Toxaphene 8001-35-2 1.10E+01 73
pahs Benzo(g,h,i)perylene 191-24-2 1.20E+01 74
pahs Indeno(1,2,3-cd)pyrene 193-39-5 1.20E+01 75
vocs Dichlorobenzene[1,4-] 106-46-7 1.20E+01 76
inorg Copper 7440-50-8 1.30E+01 77
pahs Benzo(k)fluoranthene 207-08-9 1.30E+01 78
svocs Nitrobenzene 98-95-3 1.40E+01 79
pahs Pyrene 129-00-0 1.50E+01 80
rad Americium-241 86954-36-1 1.70E+01 81
Risk-Based Pollution Control 345

Typea Name CAS no. ESLb Rank

rad Plutonium-238 13981-16-3 1.70E+01 82

rad Thorium-228 14274-82-9 1.70E+01 83
rad Plutonium-239/240 15117-48-3 1.80E+01 84
rad Neptunium-237 13994-20-2 1.90E+01 85
rad Thorium-229 15594-54-4 1.90E+01 86
inorg Fluoride 16984-48-8 2.00E+01 87
inorg Lead 7439-92-1 2.00E+01 88
inorg Nickel 7440-02-0 2.00E+01 89
rad Radium-226 13982-63-3 2.00E+01 90
rad Thorium-230 14269-63-7 2.00E+01 91
rad Uranium-233 13968-55-3 2.00E+01 92
rad Uranium-234 13966-29-5 2.00E+01 93
rad Uranium-235 15117-96-1 2.10E+01 94
rad Uranium-236 13982-70-2 2.10E+01 95
rad Uranium-238 7440-61-1 2.20E+01 96
inorg Barium 7440-39-3 2.30E+01 97
rad Thorium-232 7440-29-1 2.30E+01 98
svocs Chlorobenzene 108-90-7 2.40E+01 99
pahs Fluoranthene 206-44-0 2.60E+01 100
pahs Fluorene 86-73-7 2.90E+01 101
vocs Chloroform 67-66-3 3.20E+01 102
vocs Tetrachloroethane[1,1,2,2-] 79-34-5 3.90E+01 103
svocs Dibenzofuran 132-64-9 6.10E+01 104
vocs Benzene 71-43-2 6.60E+01 105
vocs Toluene 108-88-3 7.10E+01 106
inorg Titanium 7440-32-6 7.20E+01 107
vocs Dichloroethylene[1,1-] 75-35-4 7.60E+01 108
vocs Dichloroethylene[1,2-] 540-59-0 9.20E+01 109
pahs Phenanthrene 85-01-8 1.10E+02 110
svocs Dimethylphthalate 131-11-3 1.30E+02 111
he Nitrogylcerin 53-63-0 1.50E+02 112
pahs Acenaphthylene 208-96-8 1.60E+02 113
svocs Di-n-octylphthalate 117-84-0 1.60E+02 114
inorg Strontium 7440-24-6 1.90E+02 115
pahs Anthracene 120-12-7 2.20E+02 116
rad Strontium-90 + Yttrium-90 10098-97-2 2.80E+02 117
svocs Butyl benzyl phthalate 85-68-7 3.40E+02 118
rad Plutonium-241 14119-32-5 3.70E+02 119
rad Cobalt-60 10198-40-0 4.10E+02 120
rad Europium-152 14683-23-9 4.20E+02 121
rad Cesium-134 13967-70-9 4.90E+02 122
rad Lead-210 14255-04-0 5.10E+02 123
346 Gonzales

TABLE 1 Continued
Typea Name CAS no. ESLb Rank

rad Cesium-137 + Barium-137 5.50E+02 124

vocs Dichloroethane[1,1-] 75-34-3 7.30E+02 125
vocs Butanone[2-] 78-93-3 9.50E+02 126
rad Sodium-22 13966-32-0 1.00E+03 127
vocs Trichloroethane[1,1,1-] 71-55-6 2.80E+03 128
he PETN 78-11-5 1.40E+04 129
rad Tritium 10028-17-8 2.10E+04 130
rad Radium-228 15262-20-1 3.50E+04 131
inorg Calcium 7440-70-2 n/a 132
inorg Chloride 16887-00-6 n/a 133
inorg Iron 7439-89-6 n/a 134
inorg Lithium 7439-93-2 n/a 135
inorg Magnesium 7439-95-4 n/a 136
inorg Molybdenum 7439-98-7 n/a 137
inorg Potassium 7440-09-7 n/a 138
inorg Sodium 7440-23-5 n/a 139
svocs Carbazole 86-74-8 n/a 140
vocs Dichloroethane[1,2-] 107-06-2 n/a 141
ahe,high explosives; inorg, inorganic; pahs, polycyclic aromatic hydrocarbons; pcbs,
polychlorobiphenyls; pest, pesticide; rad, radionuclide; vocs, volatile organic carbons;
svocs, semivolatile organic carbons.

Another practical consideration is the source type of a PC/WM candidate

constituent, i.e., a point source versus diffuse. Obviously, point-source contami-
nants are subject to PC/WM and diffuse sources typically are not. For example,
DDT and DDE rank high in terms of toxicity as measured by its safe limit, or
ESL (Table 1), and as measured in potential risk to threatened and endangered
species at Los Alamos (17); however, the predominant source at the Laboratory
was aerial spraying in the 1960s (20).
Seven of the 10 most stringent ESLs in Table 1 are for organic contami-
nants. This is consistent with the results of ecorisk screening and ecological risk
assessments and related studies that have been completed to date at LANL. The
assessments and studies have begun to demonstrate that organic contaminants
dominate risk to nonhuman organisms, and radionuclides are of less concern
(17–19, 23). Thus, for purposes of reducing ecological risk from contaminants, it
may be that the largest gains in risk reduction (ignoring other factors) can be
achieved by targeting a category or group of contaminants as a priority.
Risk-Based Pollution Control 347

In summary, risk-related factors that should be considered in designing a

PC/WM program include
Contaminant risk rank
Contaminant ubiquity
Contaminant source and date of origin
There are factors additional to the three above that can influence the effectiveness
of prioritization of PC/WM efforts but that are inherent in the above three. For
example, in addition to toxicity, contaminant risk ranks that are based on ESLs
inherently consider the potential for a contaminant to bioaccumulate (soil-to-first-
level transfer) or biomagnify (food-chain transfer). This is true because ESLs are
derived from safe limits using transfer coefficients that include bioaccumulation
and biomagnification factors for those contaminants that exhibit this tendency.
Another example of a factor inherently included in ESL-based risk ranking is the
amount of soil ingested by an organism. Soil ingestion is an important route of
exposure to environmental contaminants for wildlife (24).

5 CONCLUSION
Pollution control and waste minimization can be used to effectively minimize or
eliminate risk to humans and the environment. Using the Los Alamos National
Laboratory as a case study, concepts have been introduced at the rudimentary
level that may serve as “seeds” that enable practitioners to further increase the
effectiveness of pollution control and waste minimization programs in minimiz-
ing or eliminating risk. Numerous factors, some practical and some technical,
must be considered when attempting to design holistic pollution control and waste
minimization programs. Indices of risk used to rank contaminants that are targeted
for elimination or reduction by pollution control/waste minimization techniques
can be used effectively to reduce risk to flora and fauna.

REFERENCES
1. CENR (Committee on Environment and Natural Resources of the National Science
and Technology Council), Ecological Risk Assessment in the Federal Government,
CENR/5-99/001, Washington, DC, 1999.
2. P. A. Reinhardt, K. L. Leonard, and P. C. Ashbrook, Pollution Prevention and Waste
Minimization in Laboratories. Boca Raton, FL: CRC Press, Lewis Publishers, 1996.
3. DOE (U.S. Department of Energy), Site-Wide Environmental Impact Statement for
Continued Operation of the Los Alamos National Laboratory, DOE/EIS-0238 Albu-
querque, NM: Albuquerque Operations Office, 1999.
4. LANL (Los Alamos National Laboratory), Site Pollution Prevention Plan for Los
Alamos National Laboratory, Los Alamos National Laboratory Report LA-UR-97-
1726, Los Alamos, NM, 1997.
348 Gonzales

5. EPA (U.S. Environmental Protection Agency), Guidelines for Ecological Risk Assess-
ment; Final, U.S. Environmental Protection Agency Report EPA/630/R-95/002F,
Washington, DC, 1998.
6. E. J. Calabrese and L. A. Baldwin, Performing Ecological Risk Assessments. Chelsea,
MI: Lewis Publishers, 1993.
7. S. M. Bartell, R. H. Gadner, and R. V. O’Neill, Ecological Risk Estimation. Chelsea,
MI: Lewis Publishers, 1992.
8. G. W. Suter II, Ecological Risk Assessment. Chelsea, MI: Lewis Publishers, 1993.
9. M. H. Ebinger, R. W. Ferenbaugh, A. F. Gallegos, W. R. Hansen, O. B. Myers, and
W. J. Wenzel, Screening Methodology for Ecotoxicological Risk Assessment, Los
Alamos National Laboratory Report LA-UR-95-3272, Los Alamos, NM, 1995.
10. G. J. Gonzales and P. G. Newell, Ecotoxicological Screen of Potential Release Site
50-006(d) of Operable Unit 1147 of Mortandad Canyon and Relationship to the
Radioactive Liquid Waste Treatment Facilities Project, Los Alamos National Labo-
ratory Report LA-13148-MS, Los Alamos, NM, 1996.
11. LANL (Los Alamos National Laboratory), Screening Level Ecological Risk Assess-
ment Methods for the Los National Laboratory’s Environmental Restoration Pro-
gram, R. Ryti, E. Kelly, M. Hooten, G. Gonzales, G. McDermott, L. Soholt, Los
Alamos National Laboratory Report LA-UR-99-1405, Rev. 1, Los Alamos, NM,
December, 1999.
12. EPA (U.S. Environmental Protection Agency). Wildlife Exposure Handbook; Final,
U.S. Environmental Protection Agency Report EPA/630/R-95/002F, Washington,
DC, 1998.
13. IAEA (International Atomic Energy Agency), Effects of Ionizing Radiation on Plants
and Animals at Levels Implied by Current Radiation Standards, Technical Report
Series No. 332, Vienna, Austria, 1992.
14. K. A. Higley and R. Kuperman, Ecotoxicological Benchmarks for Radionuclide
Contaminants at Rocky Flats Environmental Technology Site. Appendix C, Argonne
National Laboratory Report RF/ER-96-0039, Argonne, IL, 1996.
15. A. Gallegos, E. Kelly, M. Kramer, C. McDaniel, P. Newell, K. Sasser, and D. Stack,
Integrating Ecological Risk Assessment and Hazards Analysis, A Pilot for the
Proposed Los Alamos Waste Treatment and Storage Facilities, Los Alamos National
Laboratory Report LA-UR-95-2555, Los Alamos, NM, 1995.
16. A. F. Gallegos, Documentation and Utilization of the Ecological Transport Model
BIOTRAN.2. Los Alamos National Laboratory Report LA-UR-97-1469, Los Ala-
mos, NM, 1997.
17. G. J. Gonzales, A. F. Gallegos, and T. S. Foxx, Update Summary of Preliminary Risk
Assessments of Threatened and Endangered Species at the Los Alamos National
Laboratory, Los Alamos National Laboratory Report LA-UR-97-4732, Los Alamos,
NM, 1997.
18. G. J. Gonzales, A. F. Gallegos, T. S. Foxx, P. R. Fresquez, M. A. Mullen, L. E. Pratt,
and P. E. Gomez, Preliminary Risk Assessment of the Bald Eagle at the Los Alamos
National Laboratory, Los Alamos National Laboratory Report LA-13399-MS, Los
Alamos, NM, 1998.
19. G. J. Gonzales, A. F. Gallegos, M. A. Mullen, K. D. Bennett, and T. S. Foxx,
Preliminary Risk Assessment of the Southwestern Willow Flycatcher (Empidonax
Risk-Based Pollution Control 349

traillii extimus) at the Los Alamos National Laboratory, Los Alamos National
Laboratory Report LA-13508-MS, Los Alamos, NM, 1998.
20. A. F. Gallegos and G. J. Gonzales, Documentation of the Ecological Risk Assessment
Computer Model ECORSK.5, Los Alamos National Laboratory Report LA-13571-
MS, Los Alamos, NM, 1999.
21. G. J. Gonzales, P. R. Fresquez, and M. A. Mullen, Organic Contaminant Levels in
Three Fish Species Down-channel from the Los Alamos National Laboratory, Los
Alamos National Laboratory Report LA-13612-MS, Los Alamos, NM, 1999.
22. LANL (Los Alamos National Laboratory), ECORISK Database (version 1.1), Los
Alamos National Laboratory Environmental Restoration Project, Records Processing
Facility Package ID 186, April 2000.
23. J. K. Ferenbaugh, P. R. Fresquez, M. H. Ebinger, G. J. Gonzales, and P. A. Jordan,
Elk and Deer Study, Material Disposal Area G, Technical Area 54, Los Alamos
National Laboratory Report LA-13596-MS, Los Alamos, NM, 1999.
24. W. N. Byer, E. E. Connor, and S. G. Gerould, Estimates of Soil Ingestion by Wildlife.
J. Wildlife Manage., vol. 58, no. 2, pp. 375–382, 1994.
18
Elements of Multicriteria Decision Making

Abdollah Eskandari and

Ferenc Szidarovszky
University of Arizona, Tucson, Arizona

Abbas Ghassemi
New Mexico State University, Las Cruces, New Mexico

1 INTRODUCTION
Our life is filled with situations when decisions have to be made. Buying a car,
selecting a wife, choosing the best route to work, etc., show such situations.
In engineering design we always face decisions. The consequences of our deci-
sions are usually complex; they can by measured and evaluated only by several
criteria. In environmental engineering these criteria include water and air quality,
and their effects on people, livestock, plants, and wildlife, among others. There-
fore all of these measures should by incorporated into the decision process, that
is, we have to take multiple criteria into account. This chapter will give a brief
introduction into modeling and solving such engineering decision problems.

2 MODELING DECISION PROBLEMS

Every decision-making problem is based on choice. We need to have options to
select from. Options, or in other words, decision alternatives, may be the types of

351
352 Eskandari et al.

cars available on the market, or the different technology variants to treat a region
or to build a water treatment plant. The first step in modeling decision problems
is to see which set of alternatives we may choose from. This set can be finite or
infinite. If the number of decision alternatives is finite, then the decision problem
is called discrete. For example, if finitely many technical variants are available
for an engineering project, then the decision problem is discrete. In this case we
have to prepare a list of the alternatives. If the number of alternatives is infinite,
then no such list is possible. In such cases decision variables have to be
introduced, and the decision alternatives are identified by the different values of
the decision variables. If x denotes the capacity of a wastewater treatment
plant, then the different values of this variable represent the different variants
for the capacity. The decision variables are usually collected in the compo-
nents of a decision vector. If x1, x2, . . . , xm are the decision variables, then
x = (x1, x2, . . . , xm) is the decision vector. If the decision alternatives are repre-
sented by the different continuous values of a decision vector, then the decision
problem is called continuous. Sometimes discrete problems with a large number
of alternatives are approximated by continuous models.
The second step of the modeling process is known as the feasibility check.
There might be many different reasons why some alternatives should be dropped
from the list. For example, our budget poses a constraint on which cars could be
purchased. In the case of a discrete problem, the infeasible alternatives have to
be dropped from the list. The result of the feasibility check is therefore a reduced
list of alternatives. In the case of a continuous problem, we are unable to check
all options one by one, since there are usually too many of them or we might even
have infinitely many possibilities. In such cases feasibility has to be expressed by
certain constraints, which model the additional requirements that determine
feasibility. As an example, assume that a function go(x1, . . . , xm) represents the
cost of a project, and we cannot spend more than an amount of Qo dollars. Then
this budgetary constraint can be formulated as the inequality
go(x1, . . . , xm) <= Qo (1)

In most practical problems, several similar constraints should be satisfied

for feasibility. These constraints are either <=, or =>, or = types. Without restricting
generality, we may assume mathematically that all constraints are given by
inequalities of => type. If an inequality is defined by a <= relation, then it must have
the form
g1(x1, . . . , xm) <= Q1 (2)

which is equivalent to the relation

−g1(x1, . . . , xm) => −Q1 (3)
Elements of Multicriteria Decision Making 353

which is a => type of inequality. If a constraint is given by an equality, then it can

be rewritten as a pair of => type inequalities in the following way. Assume that
g2(x1, . . . , xm) = Q2 (4)
in a constraint of equality type. It is equivalent to the pair of inequalities
g2(x1, . . . , xm) <= Q2
(5)
g2(x1, . . . , xm) => Q2

By multiplying the first inequality by (–1) we have the pair

−g2(x1, . . . , xm) => −Q2
(6)
> Q2
g2(x1, . . . , xm) =

Therefore in our further discussions we will always assume that the feasible
alternative set of continuous decision problems is given by a system of =>
inequalities
g1(x1, . . . , xm) => Q1
g2(x1, . . . , xm) => Q2
(7)
...

gn(x1, . . . , xm) => Qn

This inequality system can be formulated in a more simple and compact form by
introducing the vectors
g1(x1, . . . , xm) Q1
   
g (x , . . . , xm) Q 
g(x) =  2 1 .  and Q =  .. 2 (8)
 ..   . 
gn(x1, . . . , xm) Qn
   
Then system (7) can be rewritten as
g(x) => Q (9)

We will use the notation X for the set of the feasible decision alternatives
of continuous problems. By using the above notation we may write

X = {x|x ∈ Rm, g(x) => Q} (10)

where Rm denotes the set of all m-element real vectors. The above discussion
shows that the result of the feasibility check of continuous problems is the
354 Eskandari et al.

formulation of the system of constraint (7) or (9), and the definition of the feasible
set (10). It is worthwhile to mention that even discrete problems can be demon-
strated in this way. If there are r alternatives to select from, then X can be defined
as the discrete set X = {1,2, . . . , r}.
The third step of modeling decision problems is the formulation of the
evaluation criteria. Each criterion evaluates the goodness of the different alterna-
tives from a specific point of view. If only one decision maker is present, then the
criteria reflect his or her complicated evaluation system. In the case of multiple
decision makers, the evaluation systems of all decision makers are combined and
summarized in the criteria. In the first case a suitable alternative is determined
that gives an acceptable trade-off among the criteria, and in the second case an
appropriate compromise solution is determined that is acceptable by all parties
involved in the decision-making process.
Some criteria are easy to quantify. For example, cost, profit, volume,
capacity, etc., have specific values for each decision alternative. Some other
criteria might not be easily quantifiable. For example, the esthetic consequences
of a forestry treatment technology can be judged only subjectively. Subjective
judgments can be made verbally, such as very good, good, fair, bad, unacceptable,
or in any other similar way. These subjective measures can be then quantified.
For example, on a [0, 100] scale, the above evaluation categories can be identified
by 100, 75, 50, 25, and 0, respectively. In the case of a discrete problem, a payoff
matrix is hence constructed, where the rows correspond to the decision alterna-
tives, the columns correspond to the evaluation criteria, and the (i, j) element aij
of this matrix is the evaluation of alternative i with respect to criterion j as shown
in Figure 1.
After the payoff matrix is constructed, the decision problem is considered
to be mathematically well defined. In the case of a continuous problem, let
ƒ1, ƒ2, . . . , ƒs denote the evaluation criteria as earlier, let x be the decision vector,
and assume that the feasibility of the decision vector is given by the system of
inequalities (7) or (9). Then the decision problem is mathematically well defined
by the multicriteria optimization problem

Maximize ƒj(x1, . . . , xm) (j = 1, 2, . . . , s)

(11)
Subject to gl(x1, x2, . . . , xm) => Ql (l = 1, 2, . . . , n)

or, using vector notation,

Maximize ƒ(x)
(12)
Subject to g(x) => Q

where
Elements of Multicriteria Decision Making 355

Criteria

1 2 3 … s
Alternatives

1 a11 a12 a13 … a1s

2 a21 a22 a23 … a2s

3 a31 a32 a33 … a3s

r ar1 ar 2 ar 3 … ars

FIGURE 1 Illustration of a payoff matrix.

ƒ1(x1, . . . , xm)
 
ƒ (x , . . . , xm)
ƒ(x) =  2 1 . 
 .. 
 ƒs(x1, . . . , xm) 
 
In most applications the problem formulation is itself a complicated,
lengthy project, and needs a continuous interaction with all parties involved and
interested in the decision problem and its solution. The process should start with
identifying decision makers, or in other words, the stakeholders. Next we have to
find out what they can decide on in order to find the set of decision alternatives.
There are very often regulations, requirements, public opinion, etc., which
influence and restrict choices. These restrictions determine the set of feasible
alternatives. The formulation of the evaluation criteria is also itself a complicated
process. In every engineering project we wish to accomplish certain goals, such
as cleaning wastewater, providing certain products to the public, etc. At the same
time, we want to avoid certain negative effects such as worsening air quality,
water quality, paying too much for the project, etc. After the goals and the
negative effects have been identified verbally, we have to find those quantities
which can be used to characterize the goodness of the different alternatives with
respect to the goals and additional consequences. These quantities should measure
how the goals are satisfied if the different alternatives are selected and carried
out, how certain it is that the project will be successful under the different
alternatives, and how severe the additional negative consequences will be. These
measurable or at least subjectively quantifiable quantities are usually selected as
356 Eskandari et al.

the evaluation criteria. Without restricting generality, we may assume that all
criteria are maximized. If smaller values are better for a criterion, then by
multiplying this criterion by (–1) we obtain a maximizing criterion. If a target
value has to be reached in the case of a certain criterion, then the distance between
the actual and target values has to be minimized, or the negative of this distance
has to be maximized. In the final problem formulation in both the discrete and
continuous cases, we will always assume therefore that all criteria are to be
maximized.
In the decision science literature problem (11) or (12) and the one given in
Figure 1 are called multiple–criteria decision problems, or multiobjective pro-
gramming problems. In the next section the solution of such problems will be
defined and examined, and in the later parts of this chapter we will give the
fundamentals of the most popular solution algorithms. A comprehensive summary
and additional details are presented in Ref. 1.

3 MULTIOBJECTIVE PROGRAMMING
As a simple illustration, consider the problem of designing a wastewater treatment
plant. Assume that there are two design alternatives which cost $2,000,000 and
$2,500,000, respectively. Assume that the capacities are 1500 m3/day and 2000
m3/day, respectively. It is also assumed that both versions result in the same
output quality. Since lower cost but higher capacity is better, the first criterion has
to be transformed into a maximizing criterion by multiplying it by (–1). The
resulting payoff matrix is given in Table 1. Notice that this problem is a discrete
multiobjective programming problem.
As a simple continuous example, consider the following problem. Assume
that a combination of three alternative technologies can be used in a wastewater
treatment plant. Let x1,x2 denote the proportion (in percent) of the application of
technology variants 1 and 2; then 1 − x1 − x2 is the proportion of the third
technology. Therefore we have two decision variables, x1 and x2, and they have
to satisfy the conditions x1 => 0, x2 => 0, and x1 + x2 <= 1. The feasible set is
illustrated in Figure 2.

TABLE 1 Payoff Matrix for Wastewater

Treatment Plan

Capacity
Alternatives/Criteria –Cost (–$) (m3/day)

1 –2,000,000 1,500
2 –2,500,000 2,000
Elements of Multicriteria Decision Making 357

x2

(0,1)

x1
(1,0)

FIGURE 2 Illustration of the feasible set for the continuous example.

Assume that there are two major pollutants that have to be removed from
the wastewater, and the three technology variants remove 3, 2, 1 mg/m3, respec-
tively, of the first kind of the pollutant, and 2, 3, 1 mg/m3 of the second kind of
pollutant. Then the total removed amount of the two pollutants can be obtained
as [3x1 + 2x2 + 1(1 − x1 − x2)]V and [2x1 + 3x2 + 1(1 − x1 − x2)]V, where V is the
total amount of treated wastewater. Maximizing those objectives is equivalent to
maximizing 2x1 + x2 and x1 + 2x2. Therefore we have the following multiobjec-
tive programming problem:
Maximize 2x1 + x2 and x1 + 2x2
Subject to x1 => 0, x2 => 0 (13)
x1 + x2 <= 1
Notice that (13) is a continuous problem with two decision variables and two
objective functions.
In many decision problems we are also interested in knowing the set of all
feasible objective values. For each x ∈ X, the vector
ƒ(x) = (ƒ1(x1, . . . , xm), ƒ2(x1, . . . , xm), . . . , ƒs(x1, . . . , xm))
is called the objective vector at x, and the set of all feasible objective vectors,
H = {u|u = ƒ(x) with some x ∈ X}
is called the objective space. If there are s objectives, then H ⊂ Rs.
In our discrete problem we have two objectives, therefore the objective
space is two-dimensional. Since there are only two alternatives, the objective
space has only two points: (–2,000,000, 1,500) and (–2,500,000, 2,000). In every
358 Eskandari et al.

discrete problem, every point of the objective space corresponds to a decision

alternative, but different alternatives may have the same point, if the objective
function values are identical. For our discrete example, Figure 3 shows these
points when the first objective is given on the horizontal axis and the second
objective is identified by the vertical axis. Point A1 shows the simultaneous
objective function values for the first alternative, and point A2 shows the same
values for the second alternative.
The question that arises now is the choice among the two alternatives,
which is the same as selecting the “better” point among A1 and A2. If cost is our
only concern, then point A1 (that is, alternative 1) must be our choice. If capacity
is our only concern, then point A2 with alternative 2 is our choice. The dilemma
we face here is the conflict between the two objectives, since higher capacity can
be obtained by selecting A2 but then the cost becomes worse. If both objectives
are important to us, then neither of the alternatives is better than the other and
hence either alternative seems to be a reasonable choice. With decision science
terminology we might say that no alternative is dominated by another alternative
and therefore both alternatives are nondominated. In the economic literature,
nondominated alternatives are called Pareto optimal. Let us define these terms
mathematically. In a discrete problem, let p and q be two alternatives, and let
ƒ(p) and ƒ(q) denote the corresponding objective vectors. We say that alternatives
p and q are equivalent if ƒ(p) = ƒ(q). Here the equality of vectors is defined by
component-wise equality. That is, two alternatives are equivalent to each other if
they result in the same values in all objective functions. Similarly, we say that
alternative p dominates alternative q, if ƒ(p) => ƒ(q) and there is a strict inequality
in at least one of the objectives. Here the notation ƒ(p) => ƒ(q) means that each

Objective 2

A2
2,000
A1
1,500

-2,500,000 -2,000,000

FIGURE 3 Objective space for a discrete problem.

Elements of Multicriteria Decision Making 359

component of vector ƒ(p) is greater than or equal to the corresponding component

of ƒ(q). In our example, ƒ(1) = (–2,000,000, 1,500) and ƒ(2) = (–2,500,000, 2,000)
and none of the alternatives dominates the other one. If no other alternatives
dominate an alternative p, then it is called nondominated. In our example,
therefore, both alternatives 1 and 2 are nondominated.
In the case of continuous problems, the objective space can be constructed
by representing the set H = {u|u = (u1, . . . , us)|uj = ƒj(x) with some x ∈ X for
j = 1,2, . . . , s} in the s-dimensional space. In our continuous example
u1 = 2x1 + x2 and u2 = x1 + 2x2
From these equations we see that
2u2 − u1 2u1 − u2
x1 = and x2 =
3 3
The constraints of problem (13) have now the form
2u2 − u1 2u1 − u2
3 => 0 3 => 0

and
2u2 − u1 2u1 − u2
+ >
3 3 = 1

These inequalities can be simplified as

u1
u2 => u2 <= 2u1 u1 + u2 <= 3 (14)
2
The set of feasible payoff vectors (u1,u2) satisfying these conditions are illustrated
in Figure 4. The vertices A, B, C of the triangle can be obtained as the intercepts
of the lines u2 = u1/2 and u2 = 2u1, u2 = 2u1 and u1 + u2 = 3, u2 = u1/2 and
u1 + u2 = 3, respectively. Simple calculation shows that A = (0, 0), B = (1, 2) and
C = (2, 1). We will next show that the nondominated objective vectors form the
linear segment connecting points B and C including the endpoints.
All points between A and B including point A are dominated by the
objective vector B, since both coordinates of B are larger. Similarly, all points
between A and C are dominated by C. Let now u be a point inside the triangle.
The continuation of the linear segment connecting points A and u intercepts the
linear segment between B and C in a point D. Since the slope of the line
connecting A and u is positive (between 1⁄2 and 2), the objective vector D
dominates u. Hence u is dominated. If D is any point of the linear segment
connecting B and C (including the endpoints), then it is impossible to increase
any objective value without worsening the value of the other objective function.
360 Eskandari et al.

FIGURE 4 Objective space for a continuous problem.

Thus, the points of the linear segment between B and C are the nondominated
objective vectors.
Any nondominated point and the corresponding decision alternative is a
reasonable choice. The nondominated decision alternatives can be obtained in the
following way:
Xo = {x|x ∈ X, ƒ(x) ∈ Ho} (15)
where Ho is the set of nondominated objective vectors. In the case of our example,
the decision vector x = (x1,x2) is nondominated if and only if
1 <= 2x1 + x2 <= 2 and (2x1 + x2) + (x1 + 2x2) = 3

These relations can be rewritten as

x1 + x2 = 1 and 0 <= x1 <= 1 (16)

and the set of the nondominated decision vectors is the linear segment connecting
the points (1, 0) and (0, 1) in Figure 2. Hence any point of this linear segment is
a reasonable selection. These nondominated alternatives do not use the third
technology variants.
In the case of our discrete example both alternatives are nondominated, and
in the case of the continuous example there are infinitely many nondominated
alternatives. Notice that the different nondominated solutions give different
objective function values, contrary to single-objective optimization problems,
where all alternative optimal solutions give the same optimal objective function
value. In other words, in single-objective optimization problems the different
optimal solutions are equivalent to each other, but in the presence of multiple
Elements of Multicriteria Decision Making 361

objectives this is not true anymore. If we consider two different nondominated

objective vectors, then the first is better in one objective and the other is better in
another objective. Therefore we cannot say which objective vector is better than
the other. In decision problems, however, we have to choose one alternative that
could be considered as the best one overall. In order to answer the question of
whether an alternative is better than another alternative overall, we need addi-
tional preference or trade-off information from the decision makers and/or from
the stakeholders. Depending on the nature of this additional information, different
solution algorithms are available. In the next sections the most popular methods
will be outlined. In order to guarantee the existence of the solution, we will
always assume that the continuous problems have continuous bounded objective
functions and that the feasible set is closed.

3.1 Sequential Optimization

The application of the sequential optimization method, which is sometimes called
the lexicographic method, is based on an ordinal preference order of the evalua-
tion criteria (or objectives). This kind of information tells us which criterion is
the most important, which is the second most important, and so on, to which one
is the least important. Without losing generality, we may assume that the criteria
are numbered in decreasing importance order; that is, the first criterion is the most
important, the second is the second most important, and so on, and the sth
criterion is the least important. The solution algorithm is then the following.
Identify first the alternatives which give the highest value of the first objective
function. If a unique alternative is found, then it is the solution of the problem.
Otherwise, delete all nonoptimal alternatives, and optimize the second most
important objective function on the set of the remaining alternatives. If a unique
alternatives is found, it is accepted as the solution of the problem; otherwise,
delete again all nonoptimal alternatives, and optimize the third objective on the
remaining feasible set. Continue the process until a unique alternative is found or
until the least important objective is being optimized. It can be proved that this
method always results in nondominated solutions, and if there are multiple
solutions, then they are equivalent to each other in the sense that they result in
the same objective function values.
The above procedure can be summarized mathematically in the following
way. Let X denote the (discrete or continuous) feasible alternative set and let
ƒ1,ƒ2, . . . , ƒs denote the objectives.
Initialization: Set j = 1, Xj = X
Optimization: Solve problem
(17)
Maximize ƒj(x)
Subject to x ∈ Xj
362 Eskandari et al.

If the optimal solution is unique, then it is accepted as the solution of the problem.
Otherwise, go to the next step.

Adjustments: Set j ⇐ j + 1, and define

Xj = {x|x ∈ Xj−1, ƒj−1(x) = optimal value},
and go back to optimization
In the case of a discrete problem, X has finitely many elements, X = 1,2, . . . ,r,
and the objective function values are represented by the aij elements of the payoff
matrix (see Figure 1). The optimization step in this case can be reformulated as

max aij
i∈Xj

and the adjustment step is

Xj = {i|i ∈ Xj−1, ai,j−1 = optimal}

In the case of continuous problems the optimization step is always a single-

objective optimization problem. The objective function is the jth objective of the
original problem and at each optimization a new constraint has to be added to
the constraints of the previous optimization step. This new constraint requires the
optimality of the previous objective function. Hence the jth optimization problem
can be written as
Maximize ƒj(x1, . . . , xm)
subject to gl(x1, . . . , xm) => Ql
ƒ1(x1, . . . , xm) = optimal value (18)
...
ƒj−1(x1, . . . , xm) = optimal value

where we assumed that the original problem is given as it was presented in the
form (11). It is worthwhile to mention that if all objectives and original constraints
are linear, then each optimization step requires the solution of a linear program-
ming problem. If j is increased by 1, then an additional constraint is added to the
constraints of the previous optimization problem. This makes the solution of
the new linear programming problem easy based on the basic optimal solution
of the previous optimization problem, since in creating a new basic solution we
have to eliminate some matrix elements only from the last constraint.
Consider first the discrete problem given earlier in Table 1. If the first
objective is more important than the second, then the alternative giving the largest
negative cost has to be selected. In this case the first alternative is the choice. If
capacity is more important than cost, then the second objective is maximized,
Elements of Multicriteria Decision Making 363

leading to the choice of the second alternative. In both cases a unique optimal
alternative is found.
Consider next the continuous problem (13). Assume first that the first
objective function is the most important; then it has to be first maximized. Thus
we have to solve the single-objective optimization problem
Maximize 2x1 + x2
Subject to x1 => 0,x2 => 0 (19)
x1 + x2 <= 1
The application of the bi-variable simplex method, or elementary graphical
approach, gives the unique optimal solution
x∗1 = 1 and x∗2 = 0
Because of uniqueness, the procedure terminates. Assume next that the second
objective is more important. Then we solve the single-objective optimization
problem
Maximize x1 + 2x2
Subject to x1 => 0,x2 => 0 (20)
x1 + x2 <= 1
There is again a unique solution,
x∗∗
1 = 0 and x∗∗
2 = 1

and the algorithm terminates.

Notice that in the case of the discrete problem both nondominated alterna-
tives could be obtained as solutions, but in the case of the continuous problem
only the two endpoints of the nondominated alternative set could be obtained. The
method selection has excluded all other points between (1,0) and (0,1) (see Figure
2) from the possibility to be obtained as a solution. In order to overcome this
problem, the method can be modified in the following way. After maximizing the
most important objective function, in later steps we do not require that all more
important objectives are at their maximal levels. The optimality constraints are
relaxed by requiring that in optimizing any later objective, the more important
objective functions are only in the neighborhood of their optimal values. For
example, we may require that their values are at least 90% of the optimal values.
In the case of our continuous problem (13), assume again that the first objective
is the more important and in maximizing the second objective only at least (90%)
of the optimal value of the first objective is required. From the earlier discussions
we know that the maximal value of the first objective is 2, which occurs at
x∗1 = 1 and x∗2 = 0. Therefore in the second step we have the following problem:
364 Eskandari et al.

Maximize x1 + 2x2
Subject to x1 => 0,x2 => 0
(21)
x1 + x2 <= 1
2x1 + x2 => 1.8
The feasible set of this problem is shown in Figure 5, and the optimal solution is
given as x∗1 = 0.8 and x∗2 = 0.2. It can be shown that in the case of a unique
solution, it is always nondominated; and if multiple solutions are present, then
there is at least one nondominated alternative among the optimal solutions.

3.2 The ⑀ -Constraint Method

In the application of the sequential optimization method we may have a unique
optimal alternative in an early step. This was in fact the case in both the discrete
and continuous examples. Since the procedure terminates, no later objective is
taken into account, therefore they might have very unfavorable values. In order
to avoid very “bad” values for some less important objectives, we might use the
ε-constraint method. We need to know which objective is the most important, and
in addition, we have to be supplied with minimal acceptable levels for all other
objectives. Without losing generality, we may assume that the first objective is
the most important and ƒ2∗, ƒ3∗, . . . , ƒs∗ are the minimal acceptable levels for the
other objectives. The method simply maximizes the first objective function
subject to the original feasibility constraint and the additional conditions that the
values of all other objectives must be at least as high as the minimal acceptable
levels. It can be proven that in the case of a unique optimal alternative, the
solution is nondominated; and in the case of multiple optimal solutions, there is
at least one nondominated alternative among the optimal solutions.

x2

1.8

0.9 1 x1

FIGURE 5 Illustration of the modified sequential optimization procedure.

Elements of Multicriteria Decision Making 365

In the case of the discrete problem given in Table 1, this method can be
formulated as

Maximize ai1
i∈X∗

where
X∗ = {i|i ∈ {1,2, . . . , r},aij => ƒj∗ for j = 2,3, . . . , s} (22)

In the case of the continuous problem (11), the ε-constraint method is the solution
of the single-objective optimization problem:
Maximize ƒ1(x1, . . . , xm)
Subject to g1(x1, . . . , xm) => Q1
..
.
gn(x1, . . . , xm) >
= Qn (23)
ƒ2(x1, . . . , xm) >
= ƒ2∗
..
.
ƒs(x1, . . . , xm) => ƒs∗
Note that if all objectives and constraints of the original problem are linear, then
problem (23) is a usual linear programming problem that can be solved by the
simplex method.
Consider first the discrete problem of Table 1, and assume that the cost is
the more important objective. Therefore the additional constraint has to be given
to the capacity. If the minimal capacity requirement is 1,500 or less, then both
alternatives are feasible and the less expensive alternative is selected, which is
the first alternative. If the minimal capacity requirement is more than 1,500, then
this additional constraint excludes the first alternative and hence the only and
therefore the optimal alternative is the second one.
In the case of the continuous example (13), assume that the first objective
is more important than the other, and we require that the value of the second
objective is at least 1.5 (which is 75% of its maximal value shown in Figure 4).
Then problem (23) can be written now as
Maximize x1 + 2x2
Subject to x1 => 0,x2 => 0
(24)
x1 + x2 <= 1
2x1 + x2 => 1.5
The feasible set of this linear programming problem is shown in Figure 6,
and the optimal solution is x∗1 = 0.5 and x∗2 = 0.5. Notice that this solution gives
366 Eskandari et al.

x2

1.5

0.75 1 x1

FIGURE 6 Feasible set for the ε-constraint method applied to a continuous

problem.

equal proportions to the first two technologies without using the third technology
at all.

3.3 The Weighting Method

In the case of the weighting method, it is assumed that the 100% overall
importance of the objectives is divided up among the different objectives. That
is, a positive weight cj is assigned to each objective j( j = 1,2, . . . , s) such that
c1 + c2 + . . . + cs = 1. The problem of multiobjective optimization is then reduced
to optimize the weighted average of the objectives, where weights are selected as
the importance factors cj. In the case of a discrete problem (such as the one
presented in Figure 1), we have to compute first the weighted averages,
Ai = c1ai1 + c2ai2 + . . . + csais
for each alternative, and then the alternative i* with the largest Ai value is selected
as the solution:

Ai∗ = max{Ai} (25)

1< <
=i =r
In the case of a continuous problem (11), the single-objective optimization
problem
s
Maximize ∑ cjƒj(x1, . . . , xm)
j=1
Subject to g1(x1, . . . , xm) => Q1
.. (26)
.
Elements of Multicriteria Decision Making 367

gn(x1, . . . , xm) => Qn

is solved, and the optimal solution is accepted as the solution of the multiobjective
optimization problem. If all of the weights are possible, then the optimal solution
of problem (26) is always nondominated.
Consider first the discrete problem of Table 1 and assume that the objectives
(cost and capacity) are equally important: c1 = c2 = 0.5. Since
A1 = 0.5(−2,000,000) + 0.5(1,500) = −999,250

and
A2 = 0.5(−2,500,000) + 0.5(2,000) = −1,249,000

A1 gives the larger value and hence the first alternative has to be selected as the
solution.
Assume next that the two objectives are also equally important in the case
of the continuous problem (13). Then the optimization problem (26) can be
rewritten as

Maximize 0.5(2x1 + x2) + 0.5(x1 + 2x2) = 1.5x1 + 1.5x2

Subject to x1 => 0, x2 => 0 (27)
x1 + x2 <= 1

It is easy to see that there are infinitely many optimal solutions: x∗1 ∈ [0,1] is
arbitrary, and x∗2 = 1 − x∗1. Assume next that the first objective is twice as import-
ant as the second one, then we many chose c1 = 2⁄3 and c2 = 1⁄3. Then problem
(27) has to be modified is the following way:

Maximize 2⁄3(2x1 + x2) + 1⁄3(x1 + 2x2) = 5⁄3x1 + 4⁄3x2

Subject to x1 => 0, x2 => 0 (28)
x1 + x2 <= 1

The feasible set of this problem was shown earlier in Figure 2. The optimal
solution is x∗1 = 1 and x∗2 = 0. It can be proven that this is the solution if c1 > c2,
and the solution is x∗1 = 0 and x∗2 = 1 if c1 < c2. The case of c1 = c2 has been
examined earlier.
The weighting method is easy to apply. If all objectives and constraints of
the original problem are linear, then problem (26) becomes a linear programming
problem which can be solved by the simplex method. Notice in addition that by
selecting a different set of weights, only the objective function changes. Therefore
the solution procedure becomes very easy if an optimal basic solution is known
for a different weight set. The weighting method has a large disadvantage,
however, since the solution might change if the objective functions are given in
368 Eskandari et al.

different units. As an illustration, consider again the discrete problem given in

Table 1 and assume now that the cost is given in $1,000 and the capacity is
presented in dm3/day. Then the payoff matrix has to be changed accordingly. The
new matrix is given in Table 2.
If equal weights are assumed again, then
A1 = 0.5(−2,000) + 0.5(1,500,000) = 749,000
and
A2 = 0.5(−2,500) + 0.5(2,000,000) = 998,750
showing that the second alternative should be selected. This result conflicts with
the selection based on the original payoff matrix. In order to overcome this
difficulty, the objective functions are usually normalized. This procedure requires
knowledge of the worst and best possible values of the different objectives. These
values can be assessed subjectively, or can be computed by minimizing and
maximizing the objectives. If ƒj∗ and ƒ∗j denote the lowest and highest values for
objective j, then the jth normalized objective is obtained by simple transformation:

− ƒj − ƒj∗
ƒj = (29)
ƒ∗j − ƒj∗

In the discrete example of Table 1 assume that the lowest cost is 0 and the largest
acceptable cost is 3,000,000. Then the worst and best values for the first objective
are ƒ1∗ = −3,000,000 and ƒ∗1 = 0. The normalized elements of the first column of
the payoff matrix become
a11 − ƒ1∗ −2,000,000 + 3,000,000 1
a−11 = = =
ƒ∗1 − ƒ1∗ 0 + 3,000,000 3

and
a21 − ƒ1∗ −2,500,000 + 3,000,000 1
a−21 = = =
ƒ∗1 − ƒ1∗ 0 + 3,000,000 6

TABLE 2 Modified Matrix for Wastewater

Treatment Plant

Capacity
Alternatives/Criteria –Cost (–1000$) (dm3/day)

1 –2,000 1,500,000
2 –2,500 2,000,000
Elements of Multicriteria Decision Making 369

Assume that the smallest acceptable capacity is 1,000 (m3/day) and the largest
possibility is assessed as 3,000 (m3/day). Then ƒ2∗ = 1,000 and ƒ∗2 = 3,000, so the
normalized elements of the second column become
a12 − ƒ2∗ 1,500 − 1,000 1
a−12 = = =
ƒ∗2 − ƒ2∗ 3,000 − 1,000 4

and
a22 − ƒ2∗ 2,000 − 1,000 1
a−22 = = =
ƒ∗2 − ƒ2∗ 3,000 − 1,000 2

The normalized payoff table is given in Table 3. If equal weights are assumed,
then
A1 = 0.5(1⁄3) + 0.5(1⁄4) = 1⁄6 + 1⁄8 = 7⁄24 ≈ 0.292

and
A2 = 0.5(1⁄6) + 0.5(1⁄2) = 1⁄12 + 1⁄4 = 1⁄3 ≈ 0.333

Since A1 < A2, the second alternative is the choice.

Consider again the continuous problem (13). Figure 4 illustrates the possi-
ble values of the objective functions, from which we see that the smallest and
largest values of both objective functions are 0 and 2, respectively. Therefore we
may select

ƒ1∗ = ƒ2∗ = 0 and ƒ∗1 = ƒ∗2 = 2

The normalized objective functions are the following:

− (2x1 + x2) − 0 1
ƒ1(x1, x2) = = x1 + x2
2 − 0 2
and

TABLE 3 Normalized Payoff Matrix for Wastewater

Treatment Plan

– Normalized Normalized
Alternatives/Criteria cost capacity

1 1⁄3 1⁄4
2 1⁄6 1⁄2
370 Eskandari et al.

− (x1 + 2x2) − 0 1
ƒ2(x1, x2) = = x1 + x2
2 − 0 2
Assume again that the first objective is twice as important as the second objective,
then we may chose the weights c1 = 2⁄3 and c2 = 1⁄3. Therefore the weighted
average of the normalized objectives becomes
2⁄3(x1 + 1⁄2x2) + 1⁄3(1⁄2x1 + x2) = 5⁄6x1 + 4⁄6x2
and the optimal solution is x∗1 = 1 and x∗2 = 0.

3.4 Direction-Based Methods

Two direction-based methods are used in practical applications. The more fre-
quently applied variant is based on the knowledge of a point of the objective space
which is not considered good enough and therefore cannot be accepted by the
decision makers (or by the stakeholders) as the solution, and a direction of
improvement is specified. The point might reflect the current situation that has to
be improved, or it can be selected as the “ideally worst” point, with components
being the minimal values of the objective functions. Sometimes the components
of this point are the result of the subjective judgement of the decision makers (or
stakeholders) as their feelings about worst possibilities. If this point is constructed
by the current objective function values, then it is called the status quo point;
otherwise, it is called the nadir. Starting from the status quo point or from the
nadir, we try to improve the objective function values in the given direction as
much as possible. This idea is illustrated in Figure 7. The straight line starting
from the status quo point or from the nadir pointing in the given direction has to
be continued until its last intercept with the objective space.
This method is used only for solving continuous problems, since in the case
of finitely many alternatives the line may not contain feasible objective vectors
at all. The concept of the method can be summarized by solving the following
optimization problem:
Maximize t
(30)
Subject to u∗ + tv ∈ H
where u∗ is the given point, v is the given vector defining the direction of
improvement, and H is the objective space. In the case of the continuous example
(13), assume that u∗ = (0,0) and the direction of improvement is given by vector
v = (3,2); that is, it is required that the first objective should increase 50% faster
than the second objective. In this case,
u∗ + tv = (0,0) + t(3,2) = (3t,2t)
therefore problem (30) can be rewritten as
Elements of Multicriteria Decision Making 371

u2

Given
direction

Status quo
point of nadir u1
objective space

FIGURE 7 Illustration of direction-based methods starting from the status

quo point or from the nadir.

Maximize t
Subject to t => 0
3t + 2t <= 1

which has the obvious optimal solution t∗ = 1⁄5. Then the optimal alternative is
x∗1 = 3t∗ = 3⁄5 and 2t∗ = 2⁄5.
Another variant of this method is based on the knowledge of an “ideally
best” point that can be either assessed subjectively by the decision makers (or by
the stakeholders) or can be constructed by the components being the maximal
values of the different objective functions, and a vector defining the direction of
relaxation. Similarly to problem (30), this concept can be modeled as the
following optimization problem:

Minimize t
(31)
Subject to u∗ − tv ∈ H

where u∗ is the ideally best point, and vector v gives the direction of relaxation.
In the case of the continuous example (13), assume that u∗ = (3,3) and
v = (2,3). The definition of vector v implies that in this case, the second objective
372 Eskandari et al.

has to be relaxed 50% faster than the first objective. In this case, problem (31)
simplifies to the following:
Minimize t
Subject to 3 − 2t => 0, 3 − 3t => 0
(3 − 2t) + (3 − 3t) <= 1
since u∗ − tv = (3,3) − t(2,3) = (3 − 2t,3 − 3t). It is easy to see that the only
feasible and therefore the optimal value of t is t∗ = 1, and the corresponding
solution is x∗1 = 1 and x∗2 = 0.
There is no guarantee that the solution obtained by either variant of this
method is nondominated. Such examples are shown in Figure 8, where A denotes
the nadir and B is the ideal point. The directions for improvement and relaxation
are shown as well. The solutions are the points S1 and S2; both are dominated by
point C.

3.5 Distance-Based Methods

Similarly to direction-based methods, two major distance-based methods are
applied in practical cases. The more frequently used variant is based on the
knowledge of an “ideally best” point and a distance function of s-dimensional
vectors. The solution that gives the closest objective vector to the ideal point is
selected as the solution. Depending on the selected type of distance, different
particular methods are available.

u2

C
direction vector of relaxation
B
ideal point
S2

S1

A
direction vector of improvement u1

FIGURE 8 Example of dominated solutions of direction-based method.

Elements of Multicriteria Decision Making 373

In the case of the discrete problem of Table 1, let ƒ∗1, . . . , ƒ∗s denote the
coordinates of the ideally best point. The objective vector of alternative i is given
by the ith row of the payoff matrix: ai = (ai1, . . . , ais). Similarly to the weighting
method, let c1, . . . , cs denote the importance weights of the different objectives.
Three major distance (or in other words, metric) types are selected in most
practical applications:

d∞(i) = max{cj|aij − ƒ∗j |}

j
s
d1(i) = ∑ cj|aij − ƒ∗j |
j=1

and

The d∞ distance is sometimes called the Tchebyshev metric, d1 is known as

the first Minkowski distance, and d2 is referred to as the Euclidean metric or the
least-squares distance. In applying the method for a discrete problem, first we
have to compute the selected distances dp(i) for all alternatives i = 1,2, . . . , r
(where p = 1,2, or ∞), and the alternative i∗ with the smallest distance is accepted
as the solution:

dp(i∗) = min{dp(i)} (32)

In the case of the continuous problem (11), we may select one of the following
distances:

d∞(x) = max {cj|ƒj(x) − ƒ∗j |}

j
s
d1(x) = ∑ cj|ƒj(x) − ƒ∗j |
j=1

and

The method then solves the single objective optimization problem

374 Eskandari et al.

Minimize dp(x)
(33)
Subject to g(x) => Q

It can be proved that the selection of distances d1 and d2 always results in

nondominated solutions. If distance d∞ is chosen and a unique optimal solution
is found, then it is also nondominated. If the optimal solution is not unique, then
there is at least one nondominated solution among them.
In the case of the weighting method, we have seen that the solution might
depend on the selected units of the objectives. Since the distance functions also
contain weights, we face a similar situation in applying distance-based methods.
Therefore we have to normalize the objective functions before using the method.
Consider first the normalized payoff matrix of the discrete problem given
in Table 3. Assume that c1 = c2 = 1⁄2, and the normalized ideally best values of
the objectives are 1 for both objectives. If the d∞ distance is selected, then
d∞(1) = max{1⁄2|1⁄3 − 1|; 1⁄2|1⁄4 − 1|} = max{1⁄3; 3⁄8} = 3⁄8

and
d∞(2) = max{1⁄2|1⁄6 − 1|; 1⁄2|1⁄2 − 1|} = max{5⁄12; 1⁄4} = 5⁄12

Since the first objective gives the smaller distance, it is considered as the better
choice. If the d1 distance is selected, then
d1(1) = 1⁄2 ⋅ |1⁄3 − 1| + 1⁄2 ⋅ |1⁄4 − 1| = 1⁄2 ⋅ 2⁄3 + 1⁄2 ⋅ 3⁄4 = 17⁄24

and
d2(2) = 1⁄2 ⋅ |1⁄6 − 1| + 1⁄2 ⋅ |1⁄2 − 1| = 1⁄2 ⋅ 5⁄6 + 1⁄2 ⋅ 1⁄2 = 8⁄12

Notice that the second alternative gives now the smaller distance, therefore the
second alternative has to be selected as the solution.
In the case of the least-squares distance,

d2(1) = 
√
1⁄2 ⋅ |1⁄3 − 1|2 + 1⁄2 ⋅ |1⁄4 − 1|2 = 
√
1⁄2 ⋅ 4⁄9 + 1⁄2 ⋅ 9⁄16

= √

145⁄288
 ≈ 0.710
and

d2(2) = √

1⁄2 ⋅ |1⁄6 − 1|2 + 1⁄2 ⋅ |1⁄2 − 1|2 = √

1⁄2 ⋅ 25⁄36 + 1⁄2 ⋅ 1⁄4

= √

34⁄72 ≈ 0.687

therefore the second alternative is the better one.

Consider next the technology selection continuous problem (13). In Section
3.3 we have shown that the normalized objective functions are
Elements of Multicriteria Decision Making 375

− −
ƒ1(x1, x2) = x1 + 1⁄2x2 and f2(x1, x2) = 1⁄2x1 + x2

Select again equal weights, c1 = c2 = 1⁄2, and assume that the normalized ideal
values are ƒ∗1 = 1 and ƒ∗2 = 1. If the d∞ distance is used, then problem (33) can be
rewritten as

Minimize max {1⁄2 ⋅ |x1 + 1⁄2x2 − 1|, 1⁄2 ⋅ |1⁄2x1 + x2 − 1|}

Subject to x1 => 0, x2 => 0 (34)
x1 + x2 <= 1

The choice of distance d1 leads to the optimization problem

Minimize 1⁄2 ⋅ |x1 + 1⁄2x2 − 1| + 1⁄2 ⋅ |1⁄2x1 + x2 − 1|

Subject to x1 => 0, x2 => 0 (35)
x1 + x2 <= 1

If the d2 distance is selected, then these optimization problems are modified as

follows:

Minimize 1⁄2 ⋅ |x1 + 1⁄2x2 − 1|2 + 1⁄2 ⋅ |1⁄2x1 + x2 − 1|2

Subject to x1 => 0, x2 => 0 (36)
x1 + x2 <= 1

where we deleted the square root in the objective function. The solutions of these
problems will be computed later in this section.
We will next show that in the linear case (that is, when all objective
functions and constraints are linear), problem (33) with distances d∞ and d1 can
be solved by the simplex method, and with distance d2 it can be solved as a
convex quadratic programming problem. Assume that for j = 1,2, . . . , s,

ƒj(x) = ƒjTx

with ƒjT being a row vector of the coefficients of the jth objective function, and
g(x) = Gx
where G is a constant matrix.
With the choice of distance d∞, problem (33) becomes

Minimize max {cj|ƒjTx − ƒ∗j |}

j
(37)
Subject to Gx => Q

Introduce the new variable F for the objective function. Then for all j,
376 Eskandari et al.

cj|ƒjTx − ƒ∗j | <= F

which can be rewritten as the pair of inequalities

1 < ∗
ƒjTx − F f
cj = j
1 > ∗
ƒjTx + F ƒ
cj = j
Hence problem (37) is equivalent to
Minimize F
Subject to Gx => Q (38)
1 
< ƒ∗j 
ƒjTx − F =
cj 
1  (j = 1, 2, . . . , s)
ƒjTx + F > ∗
cj = ƒj 

which is a linear programming problem and can be solved by the simplex method.
In the case of problem (34), we have the following linear programming problem:
Minimize F
Subject to x1 => 0, x2 >
= 0
x1 + x2 <= 1
x1 + 1⁄2x2 − 2F <= 1 (39)
x1 + 1⁄2x2 + 2F => 1
1⁄2x1 + x2 − 2F <= 1
1⁄2x1 + x2 + 2F => 1
The application of the simplex method gives the unique solution
x∗1 = x∗2 = 1⁄2 and F∗ = 1⁄8
If distance d1 is chosen, then problem (33) becomes
s
Minimize ∑cj |ƒjTx − ƒ∗j |
j=1
(40)
Subject to Gx => Q
Introduce the new variables
Fj = cj |ƒjTx − ƒ∗j |
for j = 1,2, . . . , s. Then, similar to problem (38), we have again a linear program-
ming problem,
Elements of Multicriteria Decision Making 377

s
Minimize ∑Fj
j=1

Subject to Gx => Q (41)

1 
<= ƒ∗j 
ƒjTx − F
cj j 
1  (j = 1, 2, . . . , s)
ƒjTx + Fj > ƒ∗j 
cj = 

which can be also solved by using the simplex method. In the case of problem
(35), we have to solve the problem

Minimize F1 + F2
Subject to x1 => 0, x2 => 0
x1 + x2 <= 1
x1 + 1⁄2x2 − 2F1 <= 1 (42)
x1 + 1⁄2x2 + 2F1 => 1
1⁄2x1 + x2 − 2F2 < 1
=
1⁄2x1 + x2 + 2F2 > 1
=

when we have infinitely many optimal solutions:

x∗1 ∈ [0, 1] is arbitrary, x∗2 = 1 − x∗1

F∗1 = 1⁄4(1 − x∗1), and F∗2 = 1⁄4x∗1

And finally, with the selection of the d2 distance problem (33) becomes
s
Minimize ∑cj |ƒjTx − ƒ∗j |2
j=1
(43)
Subject to Gx => Q

where we deleted the square root in the objective function. This is a convex
quadratic programming problem with linear constraints and therefore it can be
solved by routine methods (such as gradient-type iteration methods). In the
special case of problem (36), we have to following:

Minimize 1⁄2(x1 + 1⁄2x2 − 1)2 + 1⁄2(1⁄2x1 + x2 − 1)2

Subject to x1 => 0, x2 => 0 (44)
x1 + x2 <= 1

which has a unique optimal solution x∗1 = x∗2 = 1⁄2.

378 Eskandari et al.

Another variant of the method is based on the knowledge of the nadir, and
the alternative with objective vector having the largest distance from the nadir is
selected as the solution. The two versions of the method are illustrated in Figure 9.
In the discrete case, the problem can be written as

dp(i∗) = max{dp(i)} (45)

i

where dp(i) is the same as before. In the continuous case, problem (33) has to be
modified in the following way:

Maximize dp(x)
(46)
Subject to g(x) => Q

Consider next the normalized discrete problem given in Table 3 and assume
that the normalized nadir is given as ƒ1∗ = ƒ2∗ = 0 and equal weights are selected:
c1 = c2 = 1⁄2. If the d∞ distance is used, then
d∞(1) = max{1⁄2 ⋅ |1⁄3 − 0|; 1⁄2 ⋅ |1⁄4 − 0|} = max{1⁄6; 1⁄8} = 1⁄6

and
d∞(2) = max{1⁄2 ⋅ |1⁄6 − 0|; 1⁄2 ⋅ |1⁄2 − 0|} = max{1⁄12; 1⁄4} = 1⁄4

showing that the second alternative is better since it has the larger distance.
If the d1 distance is selected, then
d1(1) = 1⁄2 ⋅ |1⁄3 − 0| + 1⁄2 ⋅ |1⁄4 − 0| = 1⁄2 ⋅ 1⁄3 + 1⁄2 ⋅ 1⁄4 = 7⁄24

u2
Solution with
largest distance

Ideally best point

Solution with
smallest distance

u1
Nadir

FIGURE 9 Illustration of distance-based methods.

Elements of Multicriteria Decision Making 379

and

d1(2) = 1⁄2 ⋅ |1⁄6 − 0| + 1⁄2 ⋅ |1⁄2 − 0| = 1⁄2 ⋅ 1⁄6 + 1⁄2 ⋅ 1⁄2 = 1⁄3

The larger distance is obtained for the second alternative, so it is selected as the
solution.
Using the d2 distance, we have

d2(1) = √

1⁄2 ⋅ |1⁄3 − 0|2 + 1⁄2 ⋅ |1⁄4 − 0|2 = √

1⁄2 ⋅ 1⁄9 + 1⁄2 ⋅ 1⁄16

= √
25⁄288 ≈ 0.295

and

d2(2) = 
√
1⁄2 ⋅ |1⁄6 − 0|2 + 1⁄2 ⋅ |1⁄2 − 0|2 = 
√
1⁄2 ⋅ 1⁄36 + 1⁄2 ⋅ 1⁄4

= √

10⁄72 ≈ 0.373

and the second alternative becomes the better choice again.

The technology selection example will be next solved by this method
variant. Assume equal weights and that the normalized nadir is again
ƒ1∗ = ƒ2∗ = 0. By applying the d∞ distance, problem (33) reduces to

Maximize max {1⁄2 ⋅ |x1 + 1⁄2x2 − 0|, 1⁄2 ⋅ |1⁄2x1 + x2 − 0|}

Subject to x1 => 0, x2 => 0 (47)
x1 + x2 <= 1

Similarly to the previous case, one may verify that there are two optimal
solutions: x∗1 = 1, x∗2 = 0 and x∗1 = 0, x∗2 = 1. In the case of the d1 distance, the
optimization problem can be written as

Maximize 1⁄2 ⋅ |x1 + 1⁄2x2 − 0| + 1⁄2 ⋅ |1⁄2x1 + x2 − 0|

Subject to x1 => 0, x2 => 0 (48)
x1 + x2 <= 1

which has infinitely many optimal solutions: x∗1 ∈[0,1] is arbitrary and x∗2 = 1 − x∗1.
Selecting the d2 distance, problem (33) reduces to

Maximize 1⁄2 ⋅ |x1 + 1⁄2x2 − 0|2 + 1⁄2 ⋅ |1⁄2x1 + x2 − 0|2

Subject to x1 => 0, x2 => 0 (49)
x1 + x2 <= 1

which has two optimal solutions: x∗1 = 1, x∗2 = 0 and x∗1 = 0, x∗2 = 1.
380 Eskandari et al.

4 SOCIAL CHOICE
In many practical cases we have finitely many alternatives to select from, and the
objective functions cannot be or are difficult to quantify. In such cases we apply
methods that require ranking the alternatives with respect to each criterion. If we
imagine the situation as that each criterion is associated with a decision maker,
then a compromise selection can be found. This solution may not be the best for
any of the decision makers, but it is a solution that can be accepted by all of them.
This kind of methodology is known as social choice. A comprehensive summary
of social choice procedures can be found in Refs. 2 and 3. This section introduces
the fundamentals of this methodology.
The data, or mathematical model, consists of the sets of alternatives and
criteria, the importance weights of the different criteria, and, furthermore, the
rankings of the alternatives with respect to all criteria. These data can be
summarized in the way shown in Figure 10. The rows correspond to the
alternatives, the columns to the criteria. The last row shows the weights of the
different criteria.
Each column of the matrix has the ranking of the alternatives with respect
to the criterion associated to that column. That is, each column is a permutation
of the numbers 1,2,3, . . . , r. Here 1 indicates the most preferred alternative, 2
indicates the second most preferred one, and so on; r indicates the least preferred
alternative.
In this section, four particular methods will be introduced for finding the
best “common” choice.

4.1 Plurality Voting

Under plurality voting, for each alternative we find the sum of the weights of all
criteria which give best ranking for the given alternative. Then the alternative
with the highest sum is the social choice.

Criteria
Alternatives 1 2 … s
1 an a12 … a1S
2 a21 a22 … a2S
3 a31 a32 … a3S





r ar1 ar 2 … ars
w1 w2 … ws

FIGURE 10 Data for social choice.

Elements of Multicriteria Decision Making 381

The mathematical formulation of this method is the following. Introduce

the function
1 if x = 1
ƒ(x) = 
0 otherwise

and for each alternative i define

s
Ai = ∑ wjƒ(aij) (50)
j=1

The social choice is alternative k if

Ak = max{Ai} (51)
As an illustration of this method, consider the problem given in Table 4. Simple
calculations show that A1 = 0.4, A2 = 0.3, A3 = 0.1, A4 = 0.2. Since A1 is the
largest among the Ai numbers, alternative 1 is the social choice.

4.2 Borda Counts

The previous method considers the number and weight of first rankings only. The
Borda count takes all rankings into account by computing the quantities
s
Bi = ∑ wj aij
j=1

and alternative k is the social choice if

Bk = min{Bi}
As an illustration, consider again the problem given in Table 4. In this case,
B1 = 2(0.3) + 4(0.2) + 1(0.4) + 4(0.1) = 2.2

TABLE 4 Data for Social Choice Example

Alternatives/Criteria 1 2 3 4

1 2 4 1 4
2 1 2 3 2
3 4 3 2 1
4 3 1 4 3
0.3 0.2 0.4 0.1
382 Eskandari et al.

B2 = 1(0.3) + 2(0.2) + 3(0.4) + 2(0.1) = 2.1

B3 = 4(0.3) + 3(0.2) + 2(0.4) + 1(0.1) = 2.7
B4 = 3(0.3) + 1(0.2) + 4(0.4) + 3(0.1) = 3.0
Since B2 is the smallest, alternative 2 is the social choice.

4.3 Hare System

The Hare method is based on successive deletion of less desirable alternatives. If
there is an alternative that is selected as best by criteria with total weights at least
0.5, then it is the social choice. Otherwise, the alternative with the smallest such
total weight is deleted from the table, and the table is adjusted. If more than one
alternative has the smallest such total weight, then all of them are deleted. After
this step the number of alternatives becomes less. Then we go back to the
beginning of the procedure by looking for an alternative that is selected the best
(in the new table) by criteria with total weight at least 0.5. If we can find such an
alternative, then it is the social choice. Otherwise we continue the deletion.
In the case of the example of Table 4, there is no alternative that is
considered the best by criteria with total weights at least 0.5, since the total
weights are A1 = 0.4, A2 = 0.3, A3 = 0.1, and A4 = 0.2, as was seen in applying
plurality voting. Since A3 is the smallest Ai number, alternative 3 is deleted from
the table. The modified table is shown in Table 5. The modification is
 aij if aij < akj
aij =  (52)
aij − 1 if aij > akj

where k is the deleted alternative.

In the new table A1 = 0.4, A2 = 0.4, A4 = 0.2. There is no Ai value that is at
least 0.5, so no social choice can be made immediately. Since A4 is the smallest
Ai number, alternative 4 is deleted from the table, which is modified as shown in
Table 6. Now we have A1 = 0.4 and A2 = 0.6, so alternative 2 is the social choice.

TABLE 5 Modified Table for Hare System Made

According to the Formula

Alternatives/Criteria 1 2 3 4

1 2 3 1 3
2 1 2 2 1
4 3 1 3 2
0.3 0.2 0.4 0.1
Elements of Multicriteria Decision Making 383

TABLE 6 Further Modified Table for Hare System

Alternatives/Criteria 1 2 3 4

1 2 2 1 2
2 1 1 2 1
0.3 0.2 0.4 0.1

4.4 Pairwise Comparison

To apply the pairwise comparison method, first we have to agree on an order of
pairwise comparisons. Mathematically, an order of the alternatives has to be
specified in the following way. Let i1,i2, . . . , ir be the order. First alternatives i1
and i2 are compared. The preferred alternative is then compared to i3. The winner
is then compared to i4, and so on; the winner of the last comparison is finally
accepted as the social choice.
If i1 and i2 are any two alternatives, then they are compared as follows.
Define

N(i1,i2) = ∑ {wj|ai j
1
< ai2j} (53)
j

which is the sum of the weights of all criteria which prefer alternative i1 against
i2. Similarly,

N(i2,i1) = ∑ {wj|ai j
2
< ai1j} (54)
j

and alternative i1 is considered better than i2 if and only if

N(i1,i2) > N(i2,i1) (55)
Consider again the example of Table 4, and assume that comparisons are
made in the order 1, 2, 3, 4. We first compare alternatives 1 and 2 by computing
N(1,2) = 0.4
and
N(2,1) = 0.3 + 0.2 + 0.1 = 0.6
Since N(2,1) > N(1,2), we drop alternative 1 and the comparison continues with
alternative 2. In comparing the winner to alternative 3, we have
N(2,3) = 0.3 + 0.2 = 0.5
384 Eskandari et al.

and
N(3,2) = 0.4 + 0.1 = 0.5
So alternatives 2 and 3 are equivalent, and we have to continue the comparison
with both of them. Simple calculations show that
N (2, 4) = 0.3 + 0.4 + 0.1 = 0.8
and
N (4, 2) = 0.2
Furthermore,
N (3, 4) = 0.4 + 0.1 = 0.5
and
N (4, 3) = 0.3 + 0.2 = 0.5
Therefore alternatives 3 and 4 are equivalent, but alternative 2 is better than 4.
Hence alternative 4 is dropped and both alternatives 2 and 3 are considered as the
social choice solution.

5 CONFLICT RESOLUTION
In this section we assume that there are s decision makers, and each has a
well-defined objective function on the set of all alternatives. Let X denote the set
of feasible alternatives and let ƒi : X |→ R denote the real-valued objective func-
tion of decision maker j. In formulating a conflict we need a reference point, the
coordinates of which give the objective function values if no agreement is
reached. The existence of an unfavorable reference point gives the incentive to
negotiate and to reach a compromise.
In the conflict resolution literature there are many different solution con-
cepts; they are discussed in detail in Ref. 4. In this section the most commonly
used concept of Nash will be introduced.
It is assumed that the solution is given in the objective space H and it
depends on only the objective space
H = {u|u = ƒ(x) with some x ∈ X}
and the reference point
d = (d1, d2, . . . , ds)

so we may denote it as ␸ (H,d). It is also assumed that function ␸ satisfies the

following properties:
Elements of Multicriteria Decision Making 385

1. ␸ (H,d) ∈ H (that is, the solution has to be feasible).

2. ␸ (H,d) => d (for all decision makers, any agreement has to be at least as
good as the payoff without agreement).
3. ␸ (H,d) is a Pareto solution in H (that is, the payoff of any decision
maker can be improved only by worsening the payoff of another
decision maker).
4. If H1 ⊂ H and ␸(H,d) ∈ H1, then ␸ (H,d) = ␸(H1,d) (that is, if the
feasible objective space is reduced and the solution remains feasible,
then the solution does not change).
5. Define the linear transformation L(ƒ) = Dƒ + b on H with a diagonal
matrix D having positive diagonal elements and with an arbitrary
s-element vector b. Then it is required that ␸(L(H), L(d)) = L( ␸ (H,d))
(that is, scaling and shifting do not alter the solution).
6. If for a conflict (H,d) there are two indices i and j such that di = dj, and
− − − −
(ƒ1, . . . , ƒs) ∈ H if and only if (ƒ1, . . . , ƒs) ∈ H with ƒi = ƒj, ƒj = fi and
−
ƒk = ƒk for k ∉ {i,j}, then at the solution, ϕi(H,d) = ϕj(H,d) (that is, if
two decision makers are equivalent in the model, then they have to get
the same payoff at the solution).
Nash (5) has proven that if H is convex, closed, and bounded, and there is
at least one ƒ ∈ H such that ƒ > d, then there is a unique solution ␸ (H,d) = ƒ∗
which is the solution of the optimization problem

Maximize (ƒ1 − d1)(ƒ2 − d2) . . . (ƒs − ds)

Subject to ƒ ∈ H (56)
ƒ => d

Notice that this problem has a quasi-concave objective function, so routine

methods can be used to solve the above problem (see Ref. 6).
As an illustration, consider the continuous bi-objective problem (13). We
have seen that the feasible payoff set is characterized by inequalities
ƒ1
ƒ2 => ƒ2 <= 2ƒ1 ƒ1 + ƒ2 <= 3
2
and is illustrated in Figure 4. The worst point of this set is (0, 0), so it is considered
as the reference point. Thus, in this case, problem (56) has the special form
Maximize ƒ1 ⋅ ƒ2
Subject to ƒ1,ƒ2 => 0
ƒ1
ƒ2 =>
2
ƒ2 <= 2ƒ1
386 Eskandari et al.

ƒ1 + ƒ2 <= 3

Since the solution has to be Pareto optimal, it is on the linear segment connecting
points (1, 2) and (2, 1). This segment has the equation
ƒ1 + ƒ2 = 3,

so ƒ2 = 3 − ƒ1, and the objective function is therefore

ƒ1ƒ2 = ƒ1(3 − ƒ1) = −ƒ12 + 3ƒ1 = −(ƒ1 − 3⁄2)2 + 9⁄4

The global maximum occurs at ƒ1 = 3⁄2 and ƒ2 = 3 − 3⁄2 = 3⁄2, which is the middle
point of the segment. Notice that the solution of equations
ƒ1 = 2x1 + x2 = 3⁄2
ƒ2 = x1 + 2x2 = 3⁄2

gives the corresponding decision variable values:

x1 = x2 = 1⁄2

6 SUPPLEMENTARY NOTES
In this chapter the most popular multiobjective programming methods were
discussed. Each method is based on a certain way of expressing preference orders.
Ordinal preference order was the basis of the sequential optimization method, and
the specification of the most important objective and defining minimal acceptable
levels for all other objectives were needed in applying the ε-constraint method.
The weighting method is the appropriate method choice if importance weights
are assigned to the objectives. The knowledge of the status quo point or a nadir
and the suggested direction of improvement or the specification of an ideally best
point and the required direction of relaxation makes the use of direction-based
methods possible. Distance-based methods are used when the directions of
improvement or relaxation are replaced by distance functions. This diversity of
types of preference information makes the application of this methodology
complicated, but also gives users several ways and viewpoints by which to
examine the decision problems in hand. Applying only one method gives the
answer from only one viewpoint, and therefore the answer cannot be considered
convincing. If different methods are used to solve the same problem, in most cases
the results will be different. The reason for this inconsistency is the fact that even
an experienced decision maker is unable to express his or her preferences
accurately via a small number of quantities (such as minimal acceptable levels,
weights, ideal values, etc.). If multiple decision makers (or stakeholders) are
present, a certain compromise is needed in order to find a commonly acceptable
Elements of Multicriteria Decision Making 387

preference order. Compromises via different quantities are usually different.

Therefore we strongly believe that multiobjective decision making has to be a
repeated interactive process with the decision makers (or stakeholders). The
preference orders have to be assessed by several different ways, and the corre-
sponding methods have to be applied simultaneously. It is very seldom the case
that the same answer is found by using different methods. If this is the case, then
a satisfactory solution has been obtained. Otherwise the results have to be
analyzed in order to recover and understand the reasons for the discrepancies of
the results. The preference orders then have to be modified and the computations
redone. This process has to be repeated until a satisfactory solution is obtained.
No mathematical algorithms are available to show how preference information
has to be adjusted. These adjustments are usually based on subjective judgments,
and intuitions.
If only rankings by the different criteria or decision makers are available,
then social choice is applied. Each method is based on a certain interpretation of
fairness. Different methods result in different solutions. The simultaneous appli-
cation of several methods and repeated adjustments of weights can lead to a
satisfactory solution.
In conflict resolution we require that the solution satisfies a system of
reasonable properties. We discussed here the solution concept of Nash, but
modified axiom systems usually give different solutions. The most suitable axiom
system should be selected, and the corresponding solution then accepted as the
final solution.

REFERENCES
1. F. Szidarovszky, M. Gershon, and L. Duckstein, Techniques for Multi-objective
Decision Making in Systems Management. Amsterdam: Elsevier, 1986.
2. A. D. Taylor, Mathematics and Politics. New York: Springer-Verlag, 1995.
3. J. Bonner, Introduction to the Theory of Social Choice. Baltimore: Johns Hopkins
University Press, 1986.
4. F. Forgo, J. Szep, and F. Szidarovszky, Introduction to the Theory of Games.
Dordrecht: Kluwer, 1999.
5. J. Nash, The Bargaining Problem. Econometrica, vol. 18, pp. 155–162, 1956.
6. O. L. Mangasarian, Nonlinear Programming. New York: McGraw-Hill, 1962.
19
Environmental Considerations and
Computer Process Design

Victor R. Vasquez
University of Nevada, Reno, Reno, Nevada

1 INTRODUCTION
The goal of this chapter is to summarize the main computational methods used
in computer simulation and design of chemical processes that can be used to
identify and analyze opportunities in pollution prevention and control. The
material covered is by no means exhaustive. Therefore, the reader is encouraged
to review the references for more specific details. The philosophy behind pollu-
tion prevention involves finding technological pathways and operations to make
products in such way that the environmental impact of the process is minimum.
We can see that such a framework or approach will necessarily create design and
analysis of different alternatives. The total number of alternatives will depend on
the type of application and objectives to be accomplished and can vary from a
few to hundreds. The evaluation of the different alternatives can be greatly
accelerated by the use of appropriate computational tools or procedures.
In addition to the potential of computational methods for selection and
analysis of technological alternatives, there has been significant development of
new concepts toward more efficient processes from an environmental protection
standpoint. Among these is the concept of “green” engineering, which include

389
390 Vasquez

ideas such as process integration to broaden the design procedures by incorporat-

ing new engineering concepts to avoid pollution at early stages of design. In other
words, the idea is to avoid the generation of waste through process design.
Regardless of the approach used, pollution prevention and control through
process analysis and design is probably the most powerful approach to dealing
with environmental issues in the chemical process industry (CPI). Increasing
research efforts are reported in the literature involving the applications of con-
cepts of process synthesis and integration toward pollution prevention and waste
minimization. Development and application of techniques such as heat exchanger
networks (pinch analysis), mass-exchange networks (MENs), synthesis of envi-
ronmentally acceptable reaction paths coupled with mathematical optimization
techniques (e.g., linear and nonlinear programming), and knowledge-based sys-
tems (expert systems) establish the current basis used to approach pollution
prevention issues involving computational techniques. The methodologies and the
current development of new technologies are mainly in the area of process
synthesis, which is very important for new plants or expansion of existing ones.
However, there is substantial lack of methodologies for pollution prevention
analysis of existing plants based on the study of operational conditions without
making drastic modifications in the process flowsheets. Introducing changes in
process flowsheets of existing plants is a very difficult task because of the
inherent reluctance of the CPI to change processes that are already producing.
Even though computer-based tools for pollution prevention can play a very
significant on the identification and analysis of pollution prevention opportuni-
ties, current tools lack many capabilities for appropriate assessment of environ-
mental impact of new and existing plants (1). Efforts from several organizations
have been made during the last 10 years to address these needs and to prioritize
research and development. Among the organizations involved are the American
Institute of Chemical Engineering (AIChE), the U.S. Environmental Protection
Agency (EPA), and the U.S. Department of Energy (DOE). Some of the needs
identified are as follows (2–4):
1. Develop methodologies and tools to identify and evaluate alternative,
environmentally benign reaction pathways.
2. Develop algorithms for accurately tracking trace components in pro-
cess and waste streams.
3. Integrate process synthesis and process simulation.
4. Develop optimization approaches for synthesis of new processes.
5. Develop techniques to deal with the stochastic nature of variables in
process modeling and simulation.
6. Identify better techniques to assess environmental costs and impacts.
It is clear that to achieve these goals, the computer tools needed have to be
sophisticated, but with the growing computer industry this is becoming achiev-
Computer Process Design 391

able. An important issue to be considered is how to deal with uncertainties present

at different stages of process design and development. Current process simulators
and computer tools used for process design, in general, lack capabilities to deal
with this problem, which is very important in the evaluation of robustness and
probabilistic assessment of models or technological alternatives. Appropriate
risk analysis studies (economical and technical) provide a strong basis for deci-
sion making in the development of new “clean” chemical processes. Figure 1
illustrates the role of computational tools in an integrated approach to pollution
prevention and design.

2 COMPUTERS AND POLLUTION PREVENTION ANALYSIS

Computer-based pollution prevention analysis plays, and will continue to play, a
very significant role in the identification of potential opportunities for decreasing
the environmental impact of the chemical process industry and the development
of environmentally friendly technologies. The literature on pollution prevention is
vast, and the descriptions of the following sections are focused on computational
methods available to facilitate flowsheet analysis toward pollution prevention. From
a computational standpoint, four main areas can be identified: (a) process simulation
and numerical optimization, (b) process integration, (c) process synthesis and
assessment of alternatives, and (d) sensitivity and uncertainty analysis of process
design and simulation. Notice that these areas are strongly interconnected, and
the classification is mainly for discussion purposes only.
One of the main limitations of the approaches described above is how to
deal with the technical and economical uncertainty built into the processes. That
is why sensitivity and uncertainty analysis of process design and simulation is
becoming more important in assessing technological designs and alternatives.
On the other hand, most of the methods mentioned have practical limitations in
the sense that their actual application toward pollution prevention in a real plant
could be cumbersome and time consuming because of the difficulty of handling
complex processes realistically. However, for analysis and design of new plants,
they are very valuable.

3 PROCESS SIMULATION
Process simulation tools consist of a series of computer software resources based
on mathematical models used to model and analyze a given process. This section
will focus on simulators related to chemical processes. In practice, any computa-
tional procedure used to model a certain operation can be called a simulator.
Several elements characterize the process simulators available in the mar-
ket. A good chemical process simulator should be able to predict physical
properties or use databanks to retrieve them. Under many circumstances, good
 392
Vasquez

FIGURE 1 Integrated approach to pollution prevention involving computer design tools. (Adapted from Ref. 2.)
Computer Process Design 393

property prediction is essential. This is because experimental measurements are

always more expensive than predictions, and the time required to obtained them
is usually longer. For example, it is usually difficult to find experimental values
of physical properties for combustion gases at high temperatures, which are
essential for proper design of treatment technology of exhaust gases in combus-
tion processes. Thus, physical properties are an elementary component of a
chemical process simulator and their quality has to be assessed carefully before
designing or making technical decisions. The second integral part of a process
simulation tool is the unit operation models. These are subroutines or subpackages
that model various unit operations such as mixers, reactors, distillation columns,
heat exchangers, separators, and so forth.
Also, a process simulator must have a software component (usually called
the executive) that administrates the different modules and interacts with the user.
Chemical process simulators are very common in process modeling of industrial
processes design to transform raw materials into products and by-products. A
classical example of the application of chemical process simulators is in the
petroleum refining industry. Every time that the composition of the crude oil
changes due to changes in the source, process simulators are used to evaluate the
new operation conditions required to keep the quality of the products within
specifications or to estimate the new product flows caused by the change in
composition of the raw materials.
Process simulators also differ on how the models inside the package interact
and how the different systems of equations involved are solved or administrated.
There are two main architectures for process modeling tools, modular process
modeling and equation-oriented process modeling. A simplified diagram showing
the basic architecture of a modular process tool is presented in Figure 2. The
process modeling executive takes responsibility for interacting with the modules
that describe the unit operations involved in the process. Notice that in this case,
each unit operation module is responsible for calling subroutines or computa-
tional packages to obtain or estimate the physical properties required for the
calculations. Additionally, all the equations involved in a given unit operation are
solved within the unit operation module or by calling external subroutines or
packages. In the modular approach, the executive plays a key role in managing
and interacting with the different unit operations modules (5). The executive
administrates the inputs and outputs of the different modules involved in the
process. Usually, the output of a unit operation module is the input of another unit
operation module; therefore, the solution process becomes sequential. The com-
putational strategies used in the modular approach are closely related to the actual
flowsheet of the process (see Figure 2).
An example taken from Westerberg et al. (5) illustrates the sequential
modular approach. Figure 3 shows a simple flowsheet for the conversion of
propylene to 2-hexene by the reaction
394 Vasquez

FIGURE 2 Typical architecture for modular process modeling computer tools.

(Adapted from Ref. 7.)

2 propylene [CH3CH==CH2] → 2-hexene [CH2==CH(CH2)3CH3] (1)

The propylene feed is at 150˚C and 20 bar, and it contains 2 mol% propane. The
process simulator will have a flowsheeting system to build the model interactively
in the computer by placing icons representing the unit operations and then
connecting through the different streams involved in the process. The graphical
output on the screen will be something like the sketch in Figure 3. The next step
for the modular executive will be to specify the components to be used in the
simulation. Each stream will have different species, and the physical property
methods used for the different mixtures or pure components also have to be
chosen. In modern chemical process simulators, the options can be numerous and
some skills are required to choose among them. This is particularly important
when dealing with thermodynamic models. Some process simulators have expert
systems (e.g., ChemCAD) to help the user choose the thermodynamic property
package. For simple physical systems, ideal methods can be chosen to begin with
and then the calculations can be tuned using more sophisticated techniques. The
example, Figure 3 shows recycles; therefore, initial guesses for some of the
variables or conditions have to be given before starting the computations.
Commonly, the software package will present several possibilities for conditions
to be guessed. The choices made may influence significantly the convergence of
the simulation. Sometimes it is very important to have good initial guesses for
certain conditions.
Computer Process Design 395

FIGURE 3 Simplified process flowsheet for the production of 2-hexene from

2-propylene. (Adapted from Ref. 5.)

Next, the models to be used in the unit operations calculations have to be

selected. Here, the user is faced with different options depending on the unit
operation. The correct choice is very important for the degree of accuracy
required in the calculations. If only rough estimates are required, the choices
will be to use simple or ideal calculation methods. For more accurate calculations,
rigorous models have to be chosen. The user has to keep in mind that the
more complex the model, the more time is required to perform the calculations.
Sometimes convergence problems may arise with the use of complex models.
For example, for the flash unit in Figure 3, we might have three unit operation
models available. The first might be a component splitter, for which the user
specifies the percentage of components entering that has to exit in the top
vapor stream and in the bottom liquid stream. The second might be a constant
relative volatility flash unit, for which the user specifies the relative volatilities
for the species or requests the process simulator to estimate them using some
thermodynamic method such as Raoult’s law at a given temperature and pressure.
A third option might be a more sophisticated or rigorous method which uses a
physical property library to compute nonideal equilibrium K values and nonideal
mixture enthalpies.
396 Vasquez

The same process is repeated for all the units. Then the input streams have
to be specified together with the various conditions required, such as temperature,
composition, pressure, and state of aggregation (i.e., gas, liquid, solid). The user
may also have to choose specific numerical packages or methods to solve the
equations in the unit operation modules. Once all the conditions have been
specified for the streams and the unit operations, the process simulator solves the
model. Because it is a sequential modular process simulator, it solves first the
mixer, then the reactor, the flash, the heat exchanger, the splitter, and finally the
compressor. It compares the output of the compressor with the one guessed, and
if they do not agree, the recycle is re-guessed and the calculation repeated until
it converges within the accuracy specified by the user.
Figure 4 shows a typical architecture for an equation-oriented process
simulator. We can see that the structure does not differ that much with respect to
the sequential modular (see Figure 2). The fundamental difference is that the unit
operation modules no longer have the responsibility for solving the equations.
Instead, the equation-oriented process modeling executive collects all the equa-
tions to assemble a single system of equations (usually large), which is solved
with one or more numerical solvers. One of the main problems of this approach
is robustness. In other words, solving the system of equations, particularly if they
are highly nonlinear, can be difficult, but if convergence is not a problem this
approach is faster than using a sequential modular. Equation-oriented systems are
suitable for synthesis and optimization because the simulations are usually faster.

FIGURE 4 Typical architecture for equation-oriented process modeling tools.

(Adapted from Ref 7.)
Computer Process Design 397

For example, when dealing with recycles, sequential modular systems usually
have to do many iterations before they reach a solution, a problem that is avoided
in an equation-oriented simulator. However, it is substantially easier to set up a
flowsheet in a sequential modular than in an equation-oriented simulator, because
the former is more logical for the user. This is one of the reasons why most
commercial process simulators are sequential (see Table 1). In other words, they
are user-friendlier than equation-oriented ones. Another disadvantage of equation-
oriented systems is that it is not always obvious how to set up the global or total
system of equations.
Besides the two main classifications described above, process simulators
can be classified as steady state versus dynamic or unsteady-state. For many
processes (particularly batch), it is very important to determine how physical
conditions and operation characteristics change over time. Examples are simula-
tion of heat transfer processes or reaction kinetics. Usually, modeling of unsteady-

TABLE 1 Partial List of Commercial Process Simulatorsa

Simulation
package Description

ASPEN PLUS Sequential modular chemical process simulator produced

by Aspen Technologies, Inc. (see www.aspentech.com
for details).
ASCEND Equation-oriented process simulator developed at
Carnegie-Mellon University (see
www.cs.cmu.edu/~ascend for details).
ChemCAD Sequential modular chemical process simulator (see
www.c-com.net/~chemstat/ for details).
FLOWTRAN Sequential modular simulator for fluid flow in pipe
networks (see www.FlowTran.com.au for details).
HYSIM Sequential modular chemical process simulator (see
www.hyprotech.com for details).
gPROMS Equation-oriented general process modeling tool
developed by Process Systems Enterprise Ltd. (see
www.psenterprise.com/gPROMS/ for details)
PRO II Sequential modular chemical process simulator (see
www.simsci.com for details).
RIP Sequential modular stochastic process simulator oriented
to environmental applications (see www.golder.com/rip/
for details).
SPEEDUP Equation-oriented dynamic process simulator (see
www.aspentech.com).
aFor a more comprehensive list, see www.interduct.tudelft.nl/Pitools/tools.html.
398 Vasquez

state process is more difficult than modeling of steady-state calculations; there-

fore unsteady-state process simulators are more difficult to find in the market and
are usually designed for specific applications. Table 1 shows a partial list of
chemical process simulators. The list is not exhaustive, but information about
them can be easily obtained through the World Wide Web. Also, it is suggested
to review Ref. 6. Another important characteristic of modern chemical process
simulators is the capability to perform cost analysis including cost estimation of
equipment and their operation, energy, raw materials, etc. These options greatly
facilitate the economical evaluation of technical alternatives.
Another issue that modern engineering practice faces is information trans-
fer among different computational platforms or tools. For example, Figure 5

FIGURE 5 Chemical engineering information transfer practices in process

design and simulation. (Adapted from Ref 8.)
Computer Process Design 399

illustrates current practice of chemical engineering information transfer. Each of

the boxes is based on computer applications, and all of them have to interact. How
these boxes interact has been the subject of discussion in many companies and
organizations (see Refs. 7 and 8). Some initiatives such as the AIChE Process
Data eXchange Institute (pdXi) (8) and the project CAPE-OPEN (Computer
Aided Process Engineering. Open Simulation Environment) (7), involving com-
panies, academic institutions, and process engineering software vendors, were
created to address compatibility issues and data exchange among the different
computer tools available for engineering design and simulation. It is expected that
in the near future most engineering computer tools will be able to exchange data
with standard process modeling environments (PMEs). These projects will sub-
stantially enhance the role of computer simulation in designing cleaner processes
from an environmental standpoint.
There are other issues, however, that most computer process simulators do
not address properly. For instance, the development of cleaner processes involves
necessarily more factors and variables than traditional chemical process design,
so evaluation of robustness and probabilistic modeling becomes very important.
As mentioned earlier, most of the commercial process simulators do not have
capabilities for probabilistic modeling of processes. They are, however, capable
of performing some basic sensitivity analysis by performing successive incremen-
tal evaluations of certain variables such as flows or the number of stages in a
distillation column.

4 OPTIMIZATION AND MATHEMATICAL PROGRAMMING

Process simulation coupled with numerical optimization techniques is com-
monly used for process optimization, production programming, scheduling, and
alternative analysis in process synthesis. Process simulators have been useds
extensively for material and energy balances and for unit operations modeling
but, as mentioned earlier, they lack sufficient capabilities to study pollution
prevention options (2,9–11). Numerical optimization or mathematical program-
ming consists basically of identifying an n-dimensional vector of variables that
maximizes (or minimizes) an objective function subject to a series of equality and
inequality constraints (12–17). Mathematically, the problem can be stated as
follows:
Optimize Z = z(x,y)
Subject to h(x,y) = a (2)
g(x,y) = b
where Z is the objective function, x is a vector of design variables, and y can be
a vector of discrete parameters to further constrain the problem or to study
different alternatives.
400 Vasquez

Within this framework, any pollution prevention problem can be studied as

long as the setup and computations are feasible from a practical standpoint.
Mathematical programming techniques are classified mainly as linear program-
ming (LP) and nonlinear programming (NLP). Also, depending on the nature of
the optimization, further classifications are found. For instance, formulations with
real continuous variables mixed with integer variables are called mixed-integer
programming (MIP), with two flavors: mixed-integer linear programming (MILP)
and mixed-integer nonlinear programming (MINLP). The principles and theory
of these approaches can be found in several references (18–21). Development of
methods to find global solutions of MINLP problems is still a very active research
field (22–26). Notice that, in principle, these methods or techniques can be
used to set up objective functions that involve process synthesis or finding of
optimal configurations to minimize the environmental impact. This approach is
referred as structural optimization (2). The principles and theory of optimization
techniques are covered in many references (see, for example, Refs. 18–21). Also,
several commercial software packages are available. Edgar and Himmelblau (18)
present an early survey of computer packages for optimization. Optimization
tools can also be part of integrated modeling systems. Examples of these
are GAMS (27) and LINDO (28). A great deal of information about optimiza-
tion methods and packages can be found on the World Wide Web. For instance,
the NEOS Website at www-neos.mcs.anl.gov provides a guide to optimization
software, containing around 130 entries. Case studies that demonstrate the use
of the algorithms are also presented (for more details and other Websites, see
Ref. 28).

5 PROCESS INTEGRATION
Process integration involves mainly the interconnection and use of existing
equipment in more effective ways. Two main classifications are found: (a) pinch
analysis, and (b) knowledge-based approaches. In pinch analysis (29–32), the heat
flows through industrial processes are systematically analyzed based on funda-
mental thermodynamics in order to determine an optimal heat exchanger network
from an energy-savings standpoint. A trade-off between energy consumption and
capital investment has to be defined. The use of pinch technology results in
energy savings and also leads to waste minimization and emission reduction (2).
Commercial software is available for pinch technology, including ADVENT from
Aspen Technology, Inc., and SuperTarget from Linnhoff March, which evaluate
the trade-off between energy savings and the associated capital costs. Using
similar ideas from pinch analysis, El-Halwagi and Manousiouthakis (33) devel-
oped the notion of mass-exchange networks (MENs). Basically, MENs consist of
synthesizing a network of mass exchangers given a number of waste-(rich)
Computer Process Design 401

streams and a number of lean streams (MSAs). The network can transfer species
from the waste streams to the MSAs at a minimum cost. The main disadvantage
of this approach is the fact that mass-exchanging technology is substantially more
complex and unique to the application than the technology for transferring energy
(heat exchangers). However, significant efforts have been made to continue the
development and application of mass-exchange networks, with promising results.
(34–40).
On the other hand, knowledge-based systems include expert systems and
artificial intelligence applications. They are based on expert knowledge, usually
generated through experience in various processes (41–43). The use of heuristics
and numerical optimization approaches play a significant role in the development
of the process integration methods mentioned.

6 PROCESS SYNTHESIS AND ASSESSMENT

OF ALTERNATIVES
Methodologies in the area of process synthesis and assessment of chemical
processes also are very important for analysis of pollution prevention alternatives.
These methods require extensive amounts of information to generate a series of
technological options and then, through the use of assessment methodologies,
determine the most appropriate from both technical and economical standpoints.
Many times the assessment methodologies are linked directly to numerical
optimization techniques (optimization and mathematical programming). Basi-
cally, any methodology developed for process synthesis and assessment can
potentially be applied to pollution prevention analysis by setting appropriate
objective functions or goals [see, for example, Refs. 44–46. For instance, Linnin-
ger et al. (47) describe methods for synthesis and assessment of batch processes
for pollution prevention. Their work is focused toward the pharmaceutical and
specialty chemicals industry, which use primarily batch processing. Some of the
concepts presented include ideal ones such as zero avoidable pollution (ZAP) and
more pragmatic ones such as minimum avoidable pollution (MAP), which
represents a trade-off between cost and environmental impact. Environmental
acceptable reactions (EARs) is another computational-based methodology used
to identify environmentally safe overall chemical reactions (48). The task is
formulated as a mixed-integer nonlinear optimization problem that examines
overall reactions occurring to produce a given product. Other ideas include
synthesis of reactor networks (49,50), targeting waste minimization. Their success
has been limited mainly due to the mathematical complexity of the problem
(highly nonlinear differential models) and substantial uncertainty in the definition
of pollution-control cost functions.
402 Vasquez

7 SENSITIVITY AND STOCHASTIC ANALYSIS OF

PROCESS DESIGN AND SIMULATION
The values of design parameters and decision variables used in any process design
project are not 100% certain: there is always uncertainty associated with them.
The problem of uncertainty increases particularly in environmental applications, due
to the increase in the number of variables and parameters that have to be monitored
or used in the calculations. For example, every time an experimental measurement is
repeated, a different value is obtained even though, in theory, the conditions are kept
constant (experimental error). Thus, from a modeling and design standpoint, we
would like to know all the possible values that a calculation may have due to
uncertainties in the process variables. The collection of all these potential values is
called a distribution and is characterized in terms of probability. Before defining other
terms required to perform uncertainty or probabilistic analysis in process design and
simulation, let us illustrate further the role of uncertainty.
Figure 6 shows a simplified sketch of a stochastic modeling problem. In
this figure, we have a series of conditions or parameters that we use for design
and analysis, but we are not certain of their values requiring other type of
representations. In this case, probability distributions are used to represent their
values. Now we can see that the problem is not deterministic anymore. Therefore,
it is expected that the results of the modeling process will be probability
distributions. The main goal of stochastic modeling is to determine the probability
distributions of the results and use them to analyze the robustness of the design
or calculations. Probability distributions are more valuable than deterministic
calculations because they tell the designer the certainty that the results will be
within the specifications or objectives. Also, this information can be used to select
safety factors more rationally. Traditionally, safety factors in design are chosen
based on experience, by applying a percentage to the deterministic answer of the
problem and usually leading to expensive designs (significant overdesign) and
poor evaluation of robustness.
Stochastic modeling analysis can be carried out in several ways, depending
on the nature of the problem being studied. Computer-based approaches usually
follow four basic steps:
1. Specification of probability distributions for the stochastic variables
involved
2. Sampling process
3. Simulation
4. Statistical analysis of the results

7.1 Specification of Probability Distributions for

Stochastic Variables
In order to specify probability distributions for stochastic variables, reasonable
information has to be found about all the parameters and variables used in the
 Computer Process Design
403

FIGURE 6 Role of uncertainty (stochastic variables) in process design and simulation.

404 Vasquez

evaluation or design. For instance, for experimental data, probabilistic informa-

tion can be found from the uncertainties of the apparatus used to obtain the
measurements. Usually, the manufacturer will report what type of probabilistic
distribution can be used to characterize the errors (e.g., Gaussian). Various
probability distributions can be used to represent stochastic variables, depending
on the amount of information available. Figure 7 shows some classical probability

FIGURE 7 Typical probability distributions used in stochastic analysis and

modeling of chemical processes.
Computer Process Design 405

distributions. Among them, the Gaussian or normal and uniform are very common
in many processes. The parameters usually used to characterize probability
distributions are the mean, variance, skewness, and kurtosis (51).

7.2 Sampling Process

From Figure 6, we can easily conclude that it is not practical to evaluate all
possible values of the probability distributions that characterize the design variables.
Thus, a suitable sampling procedure is required to take representative values of
the probability distributions in such way that representative probability distribu-
tions are obtained for the output variables. This process involves two main steps:
(a) defining the sample size and (b) choosing an appropriate sampling technique.
7.2.1 Sample Size
For simulation purposes, the sample size is mainly an economic decision,
assuming that at least a reasonable minimum sample size is guaranteed from a
statistical viewpoint. Basically, for computer-based simulation of chemical pro-
cesses, we are interested in either the precision of the mean of the output variables
or the precision of the estimate of the median and other fractiles. The latter
involves the precision of the estimated cumulative distributions.
Let us assume that we have available a random sample of n output values
generated for a given sampling technique such as Monte Carlo simulation:
(y1,y2,y3, . . . , yn) (3)

Applying the classical definitions for the mean and standard deviation of y:
n
− y
y = ∑ ni (4)
i=1

and
n
(yi − −
y)2
s2 = ∑ n − 1
(5)
i=1

The confidence interval for y with a confidence level α can be defined as (60)
− s − s 
y − c √n ,y + c √n  (6)
 
where c is the deviation for the unit normal enclosing probability α.
In this case we want precision around the mean value, so we have to specify
the precision desired. Let us say that we want a total width w in the interval as a
maximum value. Therefore,
406 Vasquez

s
2c ≤ w (7)
√n
Rearranging,
2
 2cs 
n ≥   (8)
w
The equation above gives an estimate of the samples needed for a given w and
α. In order to use it, notice that an estimation of the standard deviation is required,
which can be obtained through simulation.
The other criterion mentioned is the precision of the estimated cumulative
distributions. It is possible to obtain a confidence interval for Yp, the pth fractile
of Y. We need to reorder the n elements of the sample y as follows:
y1 ≤ y2 ≤ . . . yn (9)
In this case, yi is an estimate of fractile Yp, where p = i/n.
The α confidence level for Yp is given by (yi, yj), where yi is the lower point
and yj is the upper one, and
i = np − c√ np(1 − p) −

(10)
i = np − c√

np(1 − p) 
where x means rounding x down and x− means rounding it up.
Equating the two expressions for i and j, the following expression for n is
obtained:
2
c
n = p(1 − p)  (11)
 ∆p 
The term ∆p corresponds to the width of the confidence interval centered on
fractile p.
For more details about this topic, see Morgan and Henrion (52). Authors
such as Vasquez and Whiting (53), Reed and Whiting (54), and Whiting and Tong
(55) have reported the use of this approach successfully in uncertainty studies
related to chemical process design and simulation. According to these authors,
100 simulations give a 90% confidence region of 0.85–0.95 about the 0.90
percentile.
7.2.2 Sampling Techniques
Several sampling methods are available in the literature for selecting samples
from probability distributions [Kalagnanam and Diwekar (56), Morgan and
Henrion (52)], such as using combinatorial scenarios, probability tree approaches,
and Monte Carlo methods. Several of them are well known in the literature
Computer Process Design 407

for the last 25 years: (a) Monte Carlo sampling (MCS), (b) Latin hypercube
sampling (LHS), (c) median Latin hypercube sampling (MLHS), and (d) shifted
Hammersley sampling (SHS), which is one of the most recent reported in the
literature (56).
In crude Monte Carlo sampling, for a sample of size n, a value is drawn at
random from the distribution n times. If the model evaluation is not expensive,
MCS has the advantage of avoiding the discretization of continuous distributions
and is very to implement for discrete distribution as well.
Most of the time, the primary objective is to obtain accurate quantile
estimations for the output of the model. In this case, stratified sampling is more
convenient. A classical stratified sampling technique is Latin hypercube sampling,
where the distribution is divided into intervals of equal probability and one
sample is taken at random from within each interval [Iman and Shortencarier
(57)]. One important advantage of this technique is that it produces a faster
approximation of the output probability distribution than would the random
Monte Carlo sampling technique. This is because, in principle, stratified sampling
always performs a more uniform coverage of the full range of the probability
distributions. Figure 8 illustrates the basic principle of stratified sampling. Notice
that the sampling is performed using the cumulative frequency distribution (cfd)
and not the probability density function (pdf). This facilitates the sampling

FIGURE 8 Stratified sampling procedure. Random numbers uniformly dis-

tributed between 0 and 1 are used on the y axis of the cumulative probability
distributions to read the samples on the x axis.
408 Vasquez

process substantially, by using uniformly distributed random numbers on the y

axis and then using the cfd curve to obtain the samples.
In LHS, the order of sampling is random, but the pairing of the selected
values is done either randomly or using a technique that enforces correlation
between the variables. For several correlated variables, the correlation can be
established using the correlation matrix. The goal is to perform the pairing process
in such way that the final sample has a correlation matrix nearly equal to the one
specified. To do this, the method introduced by Iman and Conover (58) is widely
used. This method rearranges the input variable sets so that they have nearly the
same rank correlation matrix as the specified rank correlation matrix. The values
of the input variables are not modified; only their pairing is modified. The details
are provided in Section 7.3.
A modification of the LHS technique is to choose the median of each of the
n probability intervals. This method is called median LHS or midpoint LHS.
It has the advantage that the approximation of the output probability distributions
is even faster than with the LHS. With this method the mean and variance of the
sample can often be almost exact, and if the model is linear, the mean of the
output will converge more rapidly (52). In general, median LHS is considered
better than crude LHS, but it is subject to problems when dealing with models
that present periodicity characteristics, producing misleading results in the output
(52). For example, Lee et al. (59) reported the use of median LHS in thermal
design studies.
Kalagnanam and Diwekar (56) developed a sampling technique called
shifted Hammersley sampling (SHS). This technique is similar in principle to
median LHS, with the difference that the sequence of points for sampling
from the distribution is generated by a Hammersley sequence introduced by
Wozniakowski (60). This sequence has been shown to have a minimum discrep-
ancy for the class of real continuous functions (on average) (60). When there are
many stochastic inputs and the model is highly nonlinear, all these stratified
sampling techniques including SHS are not much better than crude Monte Carlo
sampling (52). In these cases, more sophisticated sampling techniques are recom-
mended (61,62).

7.3 Correlated Samples

It is well known that the assumption of independence among input variables is
not appropriate in many cases. When performing sampling, the correlation
structures among the input variables have to be taken into account in some way.
Most of the approaches to this problem consider linear combinations of independ-
ent random variables to obtain a given correlation structure. Now, when dealing
with stratified sampling, the way of keeping the correlation patterns is not
straightforward in order to maintain the integrity of the sampling.
Computer Process Design 409

Iman and Conover (58) proposed a distribution-free approach to inducing

rank correlation among input variables. Their method has been widely used to
keep correlations structures when using stratified sampling (2,56).
Iman and Conover use the fact that a sample matrix R of n rows of k
independent random variables can be transformed into a sample matrix R* from
a correlated multivariate distribution with correlation matrix C via multiplication
of R by the upper-triangular matrix ST:

R∗ = RST (12)

where

C = ST ST (13)

where ST is the transpose of S and T is the correlation matrix of R.

When performing a Monte Carlo simulation, we first generate n sets of k
independent variables, where n is the number of simulations to be performed. It
is often useful (or necessary) to retain all of the original nk values of the variables,
but to transform the pairing so that the variables are correlated. For example, we
use Latin hypercube sampling to stratify our input variables according to their
respective distributions. We would prefer to retain this one-dimensional stratifi-
cation while inducing multidimensional correlation.
To allow the retention of the original variable values, Iman and Conover
developed a technique that matches the rank correlation matrix rather than the
correlation matrix. This is not only desirable for stratified sampling techniques
but also essential if the individual distributions for the variables are non-Gaussian.
First, Iman and Conover create a generic n × k matrix that represents n
sample sets of k independent Gaussian variables. They used “van der Waerden
scores” to sample the Gaussian distributions, but any stratified sampling tech-
nique could be used. [The van der Waerden scores are similar to LHS except that
the former uses the median of the n intervals as the samples, rather than taking a
stochastic sample within the interval as in LHS. Also, the upper and lower
(2n + 2)−1 of the Gaussian distribution is not sampled by van der Waerden scores.]
The desired rank correlation matrix is CR. For a Gaussian distribution, the
rank correlation matrix is equal to the correlation matrix. Therefore, Iman and
Conover transform the generic Gaussian sample matrix with Eqs. (12) and (13),
using CR, which is equal to C. The resultant sample matrix has a correlation
matrix and a rank correlation matrix that are equal to the desired rank correlation
matrix CR. The final step is to use the transformed sample matrix as a template
to reorder the values in our original LHS sample matrix so that the ranks (1 to n)
for each of the k variables are the same as those for the respective variable in the
transformed generic matrix. Thus, for a given sample size n × k and desired rank
410 Vasquez

correlation matrix CR, all the steps of the procedure are identical (regardless of
the distributions of the variables) until the final template matching of the ranks.
This procedure can be used with any sampling technique that starts with
independence between the variables. The individual variable distributions (which
need not be Gaussian) are not changed. Also, the procedure does not induce a
desired correlation matrix on the sample; rather, it induces a desired rank
correlation matrix. Also, it does not capture the complete interdependence of the
variables in the multivariate distribution; it only matches the rank correlation
matrix. For highly nonlinear models, the sampling technique proposed by Vas-
quez et al. (69) is recommended.

7.4 Simulation
Once the samples are generated, the input values are passed through the model.
For chemical engineering purposes, the model can represent a thermodynamic
equation plus procedures that involve the simulation of one or several unit
operations. All the most popular chemical process simulators (see Table 1), such
as FLOWTRAN, ChemCAD, ASPEN, PROVISION, and HYSIM, work in a
deterministic way, implying that the samples have to be passed through the
software manually or using batch mode.
After performing the simulations, the output information has to be pro-
cessed to construct the cumulative probability distributions. It is important to
mention that we have to take into account that there are several sources of
uncertainty in the simulators, such as modeling error and round-off error, that can
affect the output cumulative distributions. These effects are very difficult to
estimate, but it is important to be aware of them in the process of decision making.

7.5 Statistical Analysis

7.5.1 Cumulative Distributions
The cumulative probability distributions are basic tools for the understanding and
quantification of the output of stochastic models. They give a direct measure of
the probability of the occurrence of a given event. The definition of a cumulative
distribution can be found in any statistics book (63,64); summarizing, suppose
that X is a random variable that takes values on ___1. The cumulative distribution
function of X is defined as

F(x) = Pr{X ≤ x} (14)

If F(x) is differentiable, the function

ƒ(x) = F′(x) (15)

Computer Process Design 411

is called the density function of X. Then, for any real numbers a and b, we have
that the probability of X being between a and b is given by
b
Pr{a < X ≤ b} = F(b) − F(a) = ∫aƒ(x) dx (16)

Figure 9 shows performance results for the extraction of acetic acid from an
aqueous solution using diisopropyl ether as solvent. Three sets of parameters for
the NRTL equation were used to simulate the extraction [for details, see Vasquez
and Whiting (53)]. For example, analyzing Figure 9, we can see that there is
almost 30% probability of getting a percentage extracted less than the value
obtained with the optimum parameters for Set 1. In other words, there is
approximately 70% chance of getting an extraction greater than that obtained with
the optimum parameters. Thus, using the optimal parameter set (in this case, the
deterministic calculation) could lead to considerable overdesign.

FIGURE 9 Uncertainty in the percentage of acetic acid extracted in the liquid–

liquid extraction from an aqueous solution using diisopropyl ether as solvent.
The three sets corresponds to different values of the binary interaction
parameters of the NRTL equation regressed from different experimental data
sets. For details, see Vasquez and Whiting (53).
412 Vasquez

7.5.2 Sensitivity Analysis

Additional statistical measures are used to identify the importance of the input
variables in relation to the output variable probability distribution. These mea-
sures are the standardized regression coefficient (SRC), partial correlation coef-
ficient (PCC), standardized rank correlation coefficient (SRRC), and partial rank
correlation coefficient (PRCC). Rank correlation coefficients are the same as
SRCs and PCCs, but are computed on the ranks of the original observations. A
larger absolute value of any of these coefficients means a stronger effect by that
input variable on the output variable.
Standardized Regression Coefficients. The SRC provides a direct mea-
sure of the relative importance of the input variables, as long as the linear relation
used to compute the SRC provides a reasonable relation between the output and
input variables. To compute the SRC, a multilinear regression has to be performed
to get the best fit for the following model:

− −
(Y − Y) Xi − Xi
sY
= ∑ SRCi sXi
(17)
i
−
In the above equation, Y is the output variable and Xi are the input variables, Y
is the mean for input and output variables, and and are the standard deviations for
the output and input variables, respectively.
Partial Correlation Coefficients. The PCC is a measure of the linear
relationship between the output and input variables, and it explains the unique
relation between two variables that cannot be explained in terms of the relations
of these two variables with any other variables (65). In other words, the PCC
provides a way of identifying which additional variables have to be added to a
linear model with Y as a function of the input variables with the strongest linear
relationships.
In order to compute the PCC, a measure of the adequacy of the regression
model is required. The lack of fit from the analysis of variance can be used for
this purpose, so the adequacy of the model is measured as
−
Σi(Yi − Y)2
^

Ry2 = − (18)
Σi(Yi − Y)2

Ry2 varies between 0 and 1 and is called the coefficient of determination. It

provides a measure of the percent of the variation in Y explained by regression
on the X’s.
The partial correlation coefficient PCCi is computed using the following
equation:
Computer Process Design 413

(19)
where RXi2 is the coefficient of determination of the multilinear regression of Xi
as a function of Y and all the other X’s, and Ry2 is the coefficient of determination
of the multilinear regression of Y as a function of all of the X’s.
More details about these regression coefficients can be found in Iman et al.
(65), Reed et al. (55), Reed and Whiting (54), Diwekar and Rubin (11), and
Morgan and Henrion (52).
Rank Transformation. When nonlinear regressions are involved, rank cor-
relation coefficients are preferred. The only difference between the SRC and the
PCC is that the variable ranks are used in place of the actual values for the
variables. When computed in this way they are called standardized rank correla-
tion coefficients (SRRC) and partial rank correlation coefficients (PRCC), respec-
tively (65). In this work, both approaches are used for comparison and analysis.
To compute the ranks, the smallest value of each variable is assigned the
rank 1, the next smallest value is assigned rank 2, and so on up to the largest
value, which is assigned the rank n, where n is the number of observations.

7.6 Computer Tools

The computer tools available in the market for stochastic modeling of chemical
processes are very limited. There are some software packages for risk analysis in
general-purpose applications, but these are not specific to the chemical industry.
For instance, the package @RISK, from Palisade Corporation (66), is a risk
analysis and simulation add-in for Microsoft Excel or Lotus 1-2-3. A similar
package, but directed more toward environmental applications, is RIP Integrated
Probabilistic Simulation for Environmental Systems, from Golder Associates
(67). This is self-standing software, but it is very limited in terms of unit operation
modules.

7.7 Example: Modeling the Effect of Uncertainties on the

Performance of a Liquid/Liquid Extraction Operation
The purpose of this example is to show the significance of stochastic modeling
using probabilistic approaches. We chose a single unit operation (liquid–liquid
extraction) to illustrate some of the ideas discussed. In order to model a liquid–
liquid extraction operation, a suitable thermodynamic model is required. In this
case, we used the NRTL equation (68) to calculate the liquid–liquid equilibria of
the system 1,1,2-trichloroethane (1)–acetone (2)–water (3) system at 25˚C and
1 atm. The NRTL equation requires at least six binary interaction parameters,
414 Vasquez

which can be regressed using liquid–liquid equilibria experimental measure-

ments. The experimental data used to regress the binary interaction parameters
and their values are reported by Vasquez and Whiting (53).
The example selected is reported by Schweitzer (69), and for this case the
acetone concentration of 45 kg/h of feed is 50% (mass) acetone–50% water and
the solution is to be reduced in its acetone concentration with 13.5 kg/h of
1,1,2-trichloroethane as solvent in a countercurrent multiple-contact operation at
25˚C in five stages. The output variable for this case is the percentage of acetone
extracted. The extraction was simulated using the ChemCAD chemical process
simulator.
One hundred simulations were performed using this set of parameters. The
results are shown in Figure 10. The shape of the output distribution is an “S
shape,” and it is quite wide, from 75.38% to 94.43% in the percentage of acetone
extracted. Also, the value using the optimum parameters is 86.40%. Looking at
the figure, we can see that, almost 42% of the time, we get a percentage extracted
less than the extraction reached with the optimum parameters. In other words, we

FIGURE 10 Uncertainty in the percentage of acetone extracted in the liquid–

liquid extraction an aqueous solution using 1,1,2-trichloroethane as solvent.
For details, see Vasquez and Whiting (53).
Computer Process Design 415

TABLE 2 Liquid–Liquid Extractor Regression Resultsa

Input variable PCC SRC PRCC SRRC

b12 –0.795883 –1.8013 –0.562165 –1.18981

b21 –0.752577 –1.82991 –0.549007 –1.48629
b13 –0.274897 –0.17939 –0.073065 –0.05707
b31 –0.002288 –0.00143 –0.060129 –0.05175
b23 0.853524 2.566886 0.719530 2.11771
b32 0.824773 2.394583 0.670185 2.03531
aOutput variable: percentage acetone extracted. System: 1,1,2-trichloroethane (1)–
acetone (2)–Water (3).

have over 58% probability of getting an extraction greater than calculated using
the optimum parameters or deterministic value.
The SRC, PCC, SRRC, and PRCC computed are shown in Table 2.
Observing the data, we see that the parameters b13 and b31 are the least influential.
These parameters correspond to the pair 1,1,2-trichloroethane + water. If we want
to reduce the effect of uncertainty in the output variable distribution, we have to
reduce the variance associated with the parameters b12, b21, and b23. Even though
this is a very simple example, it clearly illustrates the usefulness of stochastic
modeling. With a performance cumulative frequency curve such as the one
presented in Figure 10, the design engineer can rationalize better the safety factors
selection among other applications. This example also demonstrates the hidden
underdesign and overdesign bias that can result from using the optimum param-
eter set from a regression (deterministic values). Notice that a hidden overdesign
bias is detected.

8 CONCLUSIONS
From a chemical process standpoint, pollution prevention is nothing but good
engineering practice. However, environmental control problems are very complex
in nature. Therefore, the design tasks become complex, creating a need for more
advance design tools. The future role of computer applications toward the design
of clean processes will be very significant. The capability of studying many
technical options in a reasonable period of time is going to boost and improve
current design methodologies. Tasks such as life-cycle analysis, multiobjective
optimization, and environmental impact assessment will become more and more
computerized. We believe that the role of uncertainty and therefore stochastic
modeling is very important in the determination of robustness of process model-
ing and simulation. Also, with the development of new computer architectures,
the study of many stochastic scenarios related to environmental protection will
416 Vasquez

be feasible. For instance, scenarios such as contaminants leaks, runaway reac-

tions, trace pollutants, and so forth, will create more certainty in the design of
clean processes.

REFERENCES
1. D. Petrides, V. Aelion, S. Mallick, and K. Abeliotis, Pollution Prevention through
Process Simulation. In H. Freeman (ed.), Industrial Pollution Prevention Handbook,
pp. 409–432. New York: McGraw-Hill, 1995.
2. P. Chaudhuri and U. Diwekar, Pollution Prevention Design. In R. Meyers (ed.),
Encyclopedia of Environmental Analysis and Remediation, pp. 3772–3791. New
York: Wiley, 1998.
3. J. M. Persichetti and S. M. Desai, Recent Advances in Process Simulation for
Environmental Considerations. AIChE Spring Natl. Meeting, Atlanta, GA, 1994.
4. J. Eisenhauer and S. McQueen, Environmental Considerations in Process Design and
Simulation. CWRT-AIChE, ISBN 0-8169-0614-9, 1993.
5. A. W. Westerberg, H. P. Hutchinson, R. L. Motard, and P. Winter, Process Flowsheet-
ing. Cambridge, UK: Cambridge University Press, 1979.
6. W. Marquardt, Trends in Computer-Aided Process Modeling. Comput. Chem. Eng.,
vol. 20, no. 6/7, pp. 591–609, 1996.
7. B. L. Braunschweig, C. C. Pantelides, H. I. Britt, and S. Sama, Open Software
Architectures for Process Modeling: Current Status and Future Perspectives.
FOCAPD’99, Breckenridge, CO, 1999.
8. T. L. Teague and J. T. Baldwin, The Emerging Discipline of Chemical Engineering
Info Transfer. FOCAPD’99, Breckenridge, CO, 1999.
9. U. M. Diwekar, A Process Analysis Approach to Pollution Prevention. In M. M.
El-Halwagi and D. P. Petrides (eds.), Pollution Prevention via Process and Product
Modifications, AIChE Symp. Ser., vol. 9, pp. 3772–3791. New York: American
Institute of Chemical Engineering, 1994.
10. P. D. Chaudhuri and U. M. Diwekar, Process Synthesis under Uncertainty: A Penalty
Function Approach. AIChE J, vol. 42, no. 3, pp. 742–748, 1996.
11. U. M. Diwekar and E. S. Rubin, Stochastic Modeling of Chemical Processes.
Comput. Chem. Eng., vol. 15, no. 2, pp. 105–114, 1991.
12. M. M. El-Halwagi, Introduction to Numerical Optimization Approaches to Pollution
Prevention. In A. P. Rositer (ed.), Waste Minimization through Process Design,
pp. 199–208. New York: McGraw-Hill, 1995.
13. I. Grossamnn, Mixed-Integer Programming Approach for the Synthesis of Integrated
Process Flowsheets. Comput. Chem. Eng., vol. 9, no. 5, p. 463, 1985.
14. I. Grossmann, Mixed-Integer Nonlinear Programming Techniques for the Synthesis
of Engineering Systems. Res. Eng. Des., vol. 1, p. 205, 1990.
15. G. R. Kocis and I. Grossmann, A Modeling and Decomposition Strategy for MINLP
Optimization of Process Flowsheets. Comput. Chem. Eng., vol. 13, p. 797, 1989.
16. I. P. Androulakis, C. D. Maranas, and C. A. Floudas, αBB: A Global Optimization
Method for General Constrained Nonconvex Problems. J. Global Optimization,
vol. 7, pp. 337–363, 1995.
Computer Process Design 417

17. C. S. Adjiman, I. P. Androulakis, C. D. Maranas, and C. A. Floudas, A Global

Optimization Method, αBB, for Process Design. Comput. Chem. Eng. Suppl., vol. 20,
p. S419, 1996.
18. T. F. Edgar and D. M. Himmelblau, Optimization of Chemical Processes. New York:
McGraw-Hill, 1988.
19. S. F. Hiller and G. J. Lieberman, Introduction to Operations Research. New York:
McGraw-Hill, 1986.
20. G. V. Reklaitis, A. Ravindran, and K. M. Ragsdell, Engineering Optimization. New
York: Wiley, 1983.
21. G. S. G. Beveridge and R. Schechter, Optimization: Theory and Practice. New York:
McGraw-Hill, 1970.
22. R. Vaidyanathan and M. M. El-Halwagi, Global Optimization of Nonconvex Non-
linear Programs via Interval Analysis. Comput. Chem. Eng., vol. 18, no. 10, pp. 889–
897, 1994.
23. C. D. Maranas and C. Floudas, Finding All Solutions of Nonlinearly Constrained
Systems of Equations. J. Global Optimization, vol. 7, p. 143, 1995.
24. C. S. Adjiman, S. Dallwig, C. A. Floudas, and A. Neumaier, A Global Optimization
Method, αBB, for General Twice-Differentiable NLPs-I. Theoretical Advances.
Comput. Chem. Eng., vol. 22, no. 9, pp. 1137–1158, 1998.
25. C. S. Adjiman, I. P. Androulakis, and C. A. Floudas, A Global Optimization Method,
αBB, for General Twice-Differentiable NLPs-II. Implementation and Computational
Results. Comput. Chem. Eng., vol. 22, no. 9, pp. 1159–1179, 1998.
26. C. S. Adjiman and C. A. Floudas, Rigorous Convex Underestimators for General
Twice-Differentiable Problems. J. Global Optimization, vol. 9, p. 23, 1996.
27. A. Brooke, D. Kendrick, and A. Meeraus, GAMS Release 2.25: A User’s Guide.
Danvers, MA: International Thomson, 1992.
28. S. J. Wright, Algorithms and Software for Linear and Nonlinear Programming.
FOCAPD’99, Breckenridge, CO, 1999.
29. B. Linnhoff, D. R. Mason, and L. Wardle, I. Understanding Heat Exchanger Net-
works. Comput. Chem. Eng., vol. 3, pp. 295–302, 1979.
30. B. Linnhoff and E. Hindmarsh, The Pinch Design Method of Heat Exchanger
Networks. Chem. Eng. Sci., vol. 38, no. 5, pp. 745–763, 1983.
31. B. Linnhoff and J. R. Flower, Synthesis of Heat Exchanger Networks: Part I:
Systematic Generation of Energy Optimal Networks. AIChE J., vol. 24, no. 4, pp.
633–642, 1978.
32. B. Linnhoff and J. R. Flower, Synthesis of Heat Exchanger Networks: Part II:
Evolutionary Generation of Networks with Various Criteria of Optimality. AIChE J.,
vol. 24, no. 4, pp. 642–654, 1978.
33. M. M. El-Halwagi and V. Manousiouthakis, Synthesis of Mass Exchange Networks.
AIChE J., vol. 35, no. 8, pp. 1233–1244, 1989.
34. M. M. El-Halwagi and H. D. Spriggs, Solve Design Puzzles with Mass Integration.
Chem. Eng. Prog., vol. 94, no. 8, p. 25, 1998.
35. M. M. El-Halwagi, A. A. Hamad, and G. W. Garrison, Synthesis of Waste Interception
and Allocation Networks. AIChE J., vol. 42, no. 11, pp. 3087–3101, 1996.
36. C. Stanley and M. M. El-Halwagi, Synthesis of Mass-Exchange Networks Using
418 Vasquez

Linear Programming Techniques. In A. Rositer (ed.), Waste Minimization through

Process Design, pp. 209–224. New York: McGraw-Hill, 1995.
37. M. Zhu and M. M. El-Halwagi, Synthesis of Flexible Mass-Exchange Networks.
Chem. Eng. Commun., vol. 138, pp. 193–211, 1995.
38. G. W. Garrison, W. B. Cooley, and M. M. El-Halwagi, Synthesis of Mass-Exchange
Networks with Multiple Target Mass-Separting Agents. Dev. Chem. Eng. Mineral
Proc., vol. 3, no. 1, pp. 31–49, 1995.
39. B. K. Srinivas and M. M. El-Halwagi, Synthesis of Reactive Mass-Exchange
Networks with General Nonlinear Equilibrium Functions. AIChE J., vol. 40, no. 3,
pp. 463–472, 1994.
40. B. K. Srinivas and M. M. El-Halwagi, Synthesis of Combined Heat and Reactive
Mass-Exchange Networks. Chem. Eng. Sci., vol. 49, no. 13, pp. 2059–2074, 1994.
41. K. E. Nelson, Use These Ideas to Cut Waste. Hydrocarbon Processing, vol. 69, no.
3, pp. 93–98, 1990.
42. J. M. Douglas, Process Synthesis for Waste Minimization. Ind. Eng. Chem. Res., vol.
31, pp. 238–243, 1992.
43. A. P. Rossiter, H. D. Spriggs, and H. Klee, Apply Process Integration to Waste
Minimization. Chem. Eng. Prog., vol. 89, no. 1, pp. 30–36, 1993.
44. M. M. Daichent and I. E. Grossmann, Integration of Hierarchical Decomposition and
Mathematical Programming for the Synthesis of Process Flowsheets. Comput. Chem.
Eng., vol. 22, pp. 147–175, 1998.
45. H. Yeomans and I. E. Grossmann, A Systematic Modeling Framework of Superstruc-
ture Optimization in Process Synthesis. Comput. Chem. Eng., vol. 23, no. 6, pp. 709,
1999.
46. B. T. Safrit and A. W. Westerberg, Synthesis of Azeotropic Batch Distillation
Separation Systems. Ind. Eng. Chem. Res., vol. 36, no. 5, pp. 1841–1854, 1997.
47. A. A. Linninger, S. A. Ali, E. Stephanopoulos, C. Han, and G. Stephanopoulos,
Synthesis and Assessment of Batch Processes for Pollution Prevention. In M. M.
El-Halwagi and D. P. Petrides (eds.), Pollution Prevention via Process and Product
Modifications, AIChE Symp. Ser., vol. 90, pp. 46–58. New York: American Institute
of Chemical Engineers, 1994.
48. E. W. Crabtree and M. M. El-Halwagi, Synthesis of Environmentally Acceptable
Reactions. In M. M. El-Halwagi and D. P. Petrides (eds.), Pollution Prevention via
Process and Product Modifications, AIChE Symp. Ser., vol. 90, pp. 117–127. New
York: American Institute of Chemical Engineers, 1994.
49. A. Lakshmanan and L. T. Biegler, Reactor Network Targeting for Waste Minimiza-
tion. In M. M. El-Halwagi and D. P. Petrides (eds.), Pollution Prevention via Process
and Product Modifications, AIChE Symp. Ser., vol. 90, pp. 128–138. New York:
American Institute of Chemical Engineers, 1994.
50. D. Hildebrandt and L. T. Biegler, Synthesis of Reactor Networks. In Proc. VI
FOCAPD Conf., Snowmass, CO, 1994.
51. S. Dowdy and S. Wearden, Statistics for Research, 2nd ed. New York: Wiley, 1985.
52. M. G. Morgan and M. Henrion, Uncertainty: A Guide to Dealing with Uncertainty
in Quantitative Risk and Policy Analysis. New York: Cambridge University Press,
1990.
53. V. Vasquez and W. Whiting, Uncertainty of Predicted Process Performance Due to
Computer Process Design 419

Variations in Thermodynamic Model Parameter Estimation from Different Experi-

mental Data Sets. Fluid Phase Equilibria, vol. 142, pp. 115–130, 1998.
54. M. E. Reed and W. B. Whiting, Sensitivity and Uncertainty of Process Designs to
Thermodynamic Model Parameters: A Monte Carlo Approach. Chem. Eng. Commun.,
vol. 124, pp. 39–48, 1993.
55. W. B. Whiting, T. Tong, and M. E. Reed, Effect of Uncertainties in Thermodynamic
Data and Model Parameters on Calculated Process Performance. Ind. Eng. Chem.
Res., vol. 32, pp. 1367–1371, 1993.
56. J. Kalagnanam and U. M. Diwekar, Efficient Sampling Technique for Optimization
under Uncertainty. AIChE J., vol. 43, no. 2, 1997.
57. R. L. Iman and M. J. Shortencarier, A FORTRAN 77 Program and User’s Guide for
the Generation of Latin Hypercube and Random Samples for Use with Computer
Models, Report NUREG/CR-3624. Springfield, VA: National Technical Information
Service, 1984.
58. R. L. Iman and W. J. Conover, A Distribution-Free Approach to Inducing Rank
Correlation among Input Variables. Commun. Statist.-Simul. Comput., vol. 11, no. 3,
pp. 311–334, 1982.
59. S. Lee, H. Kim, S. Park, and S. Chang, Development of Statistical Core Thermal
Design Methodology Using a Modified Latin Hypercube Sampling Method. Nuclear
Technol., vol. 94, p. 407, 1990.
60. H. Wozniakowski, Average Case Complexity of Multivariate Integration. Bull. Am.
Math. Soc., vol. 24, no. 1, p. 185, 1991.
61. V. R. Vasquez and W. B. Whiting, Uncertainty and Sensitivity Analysis of Thermo-
dynamic Models Using Equal Probability Sampling (EPS). Comput. Chem. Eng., vol.
23, pp. 1825–1838, 2000.
62. V. Vasquez, M. Meerschaert, and W. Whiting, A Sampling Technique for Correlated
Parameters in Nonlinear Models Based on Equal Probability Sampling (EPS).
Submitted, 1998.
63. M. Meerschaert, Mathematical Modeling, 2nd ed. San Diego, CA: Academic Press,
1999.
64. R. V. Hogg and E. A. Tanis, Probability and Statistical Inference, 4th ed. New York:
Macmillan, 1993.
65. R. L. Iman, M. J. Shortencarier, and J. D. Jhonson, A FORTRAN 77 Program and
User’s Guide for the Calculation of Partial Correlation and Standardized Regression
Coefficients, Report NUREG/CR-4122. Springfield, VA: National Technical Infor-
mation Service, 1985.
66. Palisade Corporation, Guide to Using @RISK: Advance Risk Analysis for Spread-
sheets. Newfield: Palisade Corporation, 1997.
67. Golder Associates, RIP Integrated Probabilistic Simulator for Environmental Sys-
tems: Theory Manual and User’s Guide, 1999 (see www.golder.com/rip/ for more
information).
68. J. P. Novak, J. Matous, and J. Pick, Liquid-Liquid Equilibria. New York: Elsevier,
1987.
69. P. A. Schweitzer, Handbook of Separation Techniques for Chemical Engineers. New
York: McGraw-Hill, 1979.
20
Minimization and Use of Coal Combustion
By-Products (CCBs): Concepts and
Applications

Harold W. Walker, Panuwat Taerakul,

Tarunjit Singh Butalia, and
William E. Wolfe
The Ohio State University, Columbus, Ohio

Warren A. Dick
The Ohio State University, Wooster, Ohio

1 INTRODUCTION AND BACKGROUND

During coal-fired electric power production, four main types of coal combustion
by-products (CCBs) are produced: fly ash, bottom ash, boiler slag, and flue gas
desulfurization (FGD) material (1,2). In 1998, 97.7 million metric tons of CCBs
were produced in the United States (see Figure 1). Fly ash was generated in the
largest quantity (57.1 million metric tons), with FGD material the second most
abundant CCB (22.7 million metric tons). Roughly 15.1 million metric tons of
bottom ash were generated and 2.7 million metric tons of boiler slag were
produced. Although the majority of CCBs produced currently enters landfills and
surface impoundments, there is great potential for the effective and environmen-
tally sound utilization of these materials.

421
422 Walker et al.

120

97.7
100
Million Metric Tons

80

57.1
60

40
22.7
20 15.1
2.7
0
Fly Ash Bottom Boiler FGD Total
Ash Slag CCPs

FIGURE 1 CCB production in million metric tons in the United States in 1998 (1).

Currently, the amount of CCBs entering landfills and surface impound-

ments is greater than half of the total municipal solid waste (MSW) disposed of
in the United States (see Table 1). Of the 97.7 million metric tons of CCBs
generated in 1998, 69.4 million metric tons of CCBs (or 70%) were disposed of
in landfills or surface impoundments (1). In 1997, the most current year for which
data are available, the total MSW disposed of in landfills was 119.6 million
tons (3). The amount of CCBs disposed each year is greater than the amount of
paper (37.4 million metric tons), plastic (15.5 million metric tons), wood (8.4
million metric tons), and glass (6.9 million metric tons) discarded.

TABLE 1 Amount of CCBs Disposed of in

Landfills in the United States in 1998 Compared
to Disposal of Municipal Solid Waste (MSW)a

Material Metric tons × 106 Reference

Total MSW 119.6 (3)

CCBs 69.4 (1)
Paper 37.4 (3)
Plastic 15.5 (3)
Wood 8.4 (3)
Glass 6.9 (3)
aData for disposal of MSW are for 1997, the most current
year for which data are available.
Minimization and Use of CCBs 423

TABLE 2 Amount of CCBs Produced in the United

States in 1998 Compared to Traditional Nonfuel
Mineral Commodities

Commodity Metric tons × 106 Reference

Crushed stone 1,500a (4)

Sand and gravel 1,020a (4)
CCBs 97.7 (1)
Cement 85.5a (4)
Iron ore 62a (4)
aEstimated.

Recently, the American Coal Ash Association (ACAA) proposed that CCBs
be considered a product, and therefore they recommend that these materials be
referred to as coal combustion products (CCPs). Considered as a commodity,
CCBs are ranked as the third largest nonfuel mineral commodity produced in the
United States (1,4). As shown in Table 2, the amount of CCBs generated every
year exceeds the amount of Portland cement generated in the United States, is
significantly greater than the production of iron ore, and falls behind the produc-
tion of crushed stone, sand, and gravel.
The purpose of this chapter is to review the current state of the art in
technology for minimizing CCB generation, maximizing CCB use, and reducing
the disposal of CCBs in landfills and surface impoundments. This chapter will
first present a review of important federal regulations influencing the generation
and utilization of CCBs in the United States. Next, the physical, chemical, and
engineering properties of CCBs will be discussed, and the operational factors
affecting CCB generation will be presented. The chapter will conclude with a
discussion regarding strategies for minimizing CCB production and maximizing
the utilization of CCBs. Potential barriers to utilization and minimization in the
future will also be discussed.

2 FEDERAL REGULATIONS INFLUENCING CCB

GENERATION AND USE
Governmental regulations of emissions from electric power plants combined with
efforts to improve air quality have had a profound effect on the amount and type
of CCBs produced in the United States over the past 25 years. The Clean Air Act
of 1967 was the first legislation to establish the authority of the federal govern-
ment to promulgate air quality criteria (5). It set the groundwork for future
“technology-forcing legislation,” i.e., legislation that sets standards unattainable
424 Walker et al.

utilizing existing technology. This regulatory approach required industry and

utilities to develop new technologies to meet promulgated standards.
The Clean Air Act Amendments of 1970 established Natural Ambient Air
Quality Standards (NAAQS) and set specific pollutant removal requirements
(New Source Performance Standards or NSPS) for both stationary and mobile
sources (5). NSPS, which are applicable to coal-fired utilities, were written in part
60, subpart D, Da, Db, and Dc of 40 CFR (Code of Federal Regulation) (6). NSPS
in 40 CFR, part 60, subpart D, set air pollutant levels for coal-fired steam
generators with heat input rates over 73 megawatts (MW), constructed or sub-
stantially modified after August 17, 1971. Amendments to the Clean Air Act in
1990 added new provisions to reduce the formation of acid rain by decreasing
sulfur and nitrogen oxide emissions. Key to these provisions was the requirement
to reduce annual SO2 emissions by 10 million tons below 1980 levels, and to
reduce NOx emissions by 2 million tons below 1980 levels. To achieve these
emission reductions, the Clean Air Act Amendments of 1990 promulgated NOx
and SO2 performance standards and set up an innovative emission trading system
for SO2 reduction. In phase I of the SO2-reduction program, the legislation
required 110 identified utilities to reduce SO2 emissions to 2.5 lb/mmBTU by
January 1995. Phase II mandated further reductions in emissions to 1.2
lb/mmBTU for all utilities generating at least 25 MW of electricity. It is estimated
that phase II requirements will affect 2128 utilities in the United States (7). The
NOx reduction program was also separated into two phases. In phase I, Group 1
boilers (dry-bottom wall and tangentially fired boilers) were required to meet NOx
performance standards by January 1996 (8). Phase II set lower NOx emission
limits for Group 1 boilers and established initial NOx emission limitations for
Group 2 boilers (cell burner technology, cyclone boilers, wet bottom boilers, and
other types of coal-fired boilers) (7).
To meet these federal regulations, coal-fired utilities have switched to
alternative fossil fuels or installed air pollution control technologies such as
electrostatic precipitators, baghouses, and wet or dry SO2 scrubbing systems.
Currently, CCBs generated as a result of air pollution control processes are
regulated under subtitle D of the Resource Conservation and Recovery Act
(RCRA), which pertains to nonhazardous solid wastes (9). In 1988, and then
again in 1999, the U.S. Environmental Protection Agency (EPA) issued a Report
to Congress examining the environmental impacts associated with CCB use and
disposal (10,11). Reports in both 1988 and 1999 concluded that CCBs were
nonhazardous and nontoxic materials. In early 2000, based on its own findings in
the Report to Congress as well as input from environmental groups, the EPA
maintained its previous ruling that CCBs will continue to be regulated under
subtitle D of the RCRA. As a result, the use and/or disposal of CCBs is regulated
at the state level. For example, regulations in the state of Ohio consider fly ash,
Minimization and Use of CCBs 425

bottom ash, boiler slag, and FGD generated from coal or other fuel combustion
sources to be exempt from regulation as hazardous waste (12).

3 PHYSICAL, CHEMICAL, AND ENGINEERING

PROPERTIES OF CCBS
Information regarding the physical, chemical, and engineering properties of
CCBs is required before these materials can be safely and effectively utilized. The
physical and engineering properties, in particular, are important parameters
affecting the behavior of CCBs in various engineering applications. Information
regarding the chemical composition is important for addressing potential environ-
mental impacts associated with CCB utilization and disposal. Chemical data are
also useful for explaining physical properties when pozzolanic or cementitious
reactions take place.
As mentioned above, the four main types of CCBs are fly ash, bottom ash,
boiler slag, and FGD material. Fly ash is a powdery material removed from
electrostatic precipitation (ESP) or baghouse operations, while bottom ash is a
granular material removed from the bottom of dry-bottom boilers. Boiler slag is
a granular material that settles to the bottom of wet-bottom and cyclone boilers.
It forms when the operating temperature in the boiler exceeds the ash fusion
temperature. Boiler slag exists in a molten state until it is drained from the boiler.
The majority of FGD material is a mixture of fly ash and dewatered scrubber
sludge. Scrubber sludge is produced when flue gases are exposed to an aqueous
solution of lime or limestone. The wet scrubber sludge is dewatered and stabilized
with fly ash and extra lime. Alternatively, the scrubber sludge can be oxidized to
calcium sulfate (CaSO4) to produce synthetic FGD gypsum. Dry FGD processes
are widely used, in which limestone is injected directly into the boiler or flue gas
stream. Dry FGD by-products are removed from the flue gas by electrostatic
precipitation or baghouse operations.

3.1 Physical and Engineering Properties of CCBs

A number of the physical and engineering properties of fly ash, bottom ash, boiler
slag, and FGD material are summarized in Table 3 (10,11,13–16,18,19). Fly ash
is usually spherical, with a diameter ranging from 1 to 100 µm. Fly ash has the
appearance of a gray cohesive silt and has low permeability when compacted.
Bottom ash and boiler slag are granular in shape, with sizes ranging from 0.1 to
10.0 mm. Boiler slag has a glassy appearance. Bottom ash has a permeability
higher than fly ash, while boiler slag has a permeability similar to that of course
sand. Fly ash, bottom ash, and boiler slag have dry densities that range between
40 and 100 lb/ft3 (10,11,15,16). Fly ash has lower shear strength than both bottom
ash and boiler slag.
426 Walker et al.

TABLE 3 Summary of Physical Characteristics and Engineering Properties

of Fly Ash, Bottom Ash, Boiler Slag, and FGD Material (10,11,13–16,18,19)

FGD material
Bottom ash/
Physical characteristics Fly ash boiler slag Wet Dry

Particle size (mm) 0.001–0.1 0.1–10.0 0.001–0.05 0.002–0.074

Compressibility (%) 1.8 1.4
Dry density (lb/ft3) 40–90 40–100 56–106 64–87
Permeability (cm/s) 10–6–10–4 10–3–10–1 10–6–10–4 10–7–10–6
Shear strength
Cohesion (psi) 0–170 0
Angle of internal
friction (deg) 24–45 24–45
Unconfined compres-
sive strength (psi) 0–1600 41–2250

The physical characteristics of FGD material depend on the type of FGD

system used: wet or dry (see Table 3). Wet FGD systems generate by-products
with diameters ranging from 0.001 to 0.05 mm. Dry FGD systems produce
by-products with diameters ranging from 0.002 to 0.074 mm. FGD material
generally has low permeability, ranging from 10–4 to 10–7 cm/s. The unconfined
compressive strength is affected by the moisture content of FGD, and the
percentages of fly ash and lime. For example, wet FGD scrubber sludge is similar
to toothpaste in consistency and has little unconfined compressive strength.
However, the strength of wet FGD is greatly improved when FGD sludge is
stabilizing by mixing with lime and fly ash.

3.2 Chemical Properties of CCBs

The chemical characteristics of fly ash, bottom ash, and boiler slag depend greatly
on the type of coal used and the operating conditions of the boiler (10,11). Over
95% of fly ash consists of oxides of silicon, aluminum, iron, and calcium, with
the remaining 5% consisting of various trace elements (10,11). The chemical
composition of fly ash is affected by the operating temperature of the boiler,
because the operating temperature influences the volatility of certain elements.
For example, sulfur may be completely volatilized at high temperature and
removed during lime scrubbing, thus reducing the amount in the fly ash, bottom
ash, and boiler slag (10,11).
Table 4 shows the trace-element content of fly ash, bottom ash, boiler slag,
and FGD material (10,11,20). The elemental composition of fly ash from two
TABLE 4 Trace Element Composition of Fly Ash, Bottom Ash, Boiler Slag (10), and FGD Material (20)
Fly ash
Bottom ash/boiler
Mechanical ESP/baghouse slag Dry FGD material
Element
(mg/kg) Range Median Range Median Range Median Range Median

Arsenic 3.3–160 25.2 2.3–279 56.7 0.50–168 4.45 44.1–186 86.5

Boron 205–714 258 10–1300 371 41.9–513 161 145–418 318
Minimization and Use of CCBs

Barium 52–1152 872 110–5400 991 300–5789 1600 100–300 235

Cadmium 0.40–14.3 4.27 0.10–18.0 1.60 0.1–4.7 0.86 1.7–4.9 2.9
Cobalt 6.22–76.9 48.3 4.90–79.0 35.9 7.1–60.4 24 8.9–45.6 26.7
Chromium 83.3–305 172 3.6–437 136 3.4–350 120 16.9–76.6 43.2
Copper 42.0–326 130 33.0–349 116 3.7–250 68.1 30.8–251 80.8
Fluorine 2.50–83.3 41.8 0.4–320 29.0 2.5–104 50.0 — —
Mercury 0.008–3.0 0.073 0.005–2.5 0.10 0.005–4.2 0.023 — —
Manganese 123–430 191 24.5–750 250 56.7–769 297 127–207 167
Lead 5.2–101 13.0 3.10–252 66.5 0.4–90.6 7.1 11.3–59.2 36.9
Selenium 0.13–11.8 5.52 0.6–19.0 9.97 0.08–14 0.601 3.6–15.2 10.0
Silver 0.08–4.0 0.70 0.04–8.0 0.501 0.1–0.51 0.20 — —
Strontium 396–2430 931 30–3855 775 170–1800 800 308–565 432
Vanadium 100–377 251 11.9–570 248 12.0–377 141 — —
Zinc 56.7–215 155 14–2300 210 4.0–798 99.6 108–208 141
427
428 Walker et al.

types of collection methods is shown. Mechanical collection methods generally

collect larger particles from the flue gas, while finer ash particles are collected by
electrostatic precipitators (ESPs) or baghouses. However, similar ranges of most
trace elements are found in both types of collection methods. Some exceptions to
this are arsenic, boron, lead, and selenium, which may be found at slightly higher
fractions in fly ash collected by ESPs or baghouses. Cadmium and fluorine may
be present at higher levels in ash collected by mechanical methods.
The chemical characteristics of FGD by-products depend on the type of
absorbent used and the sulfur content of the coal. In the United States, approxi-
mately 90% of FGD systems use lime or limestone as a sorbent (17). In
lime-based FGD processes, the absorbent reacts with sulfur in the flue gas and
forms a calcium compound, either calcium sulfite or calcium sulfate, or a calcium
sulfite–sulfate mixture (10,11). In systems that use dual-alkali scrubber technol-
ogy, sodium hydroxide, sodium sulfite, or lime is used as absorbent solution.
These types of systems generate calcium sulfite and sodium salts (10,11). In
spray-drying scrubber systems, sodium sulfate and sodium sulfite are produced
with sodium-based reagents. When fly ash is added to FGD, the quantity and
characteristics of the fly ash will also affect FGD chemical characteristics.
The most significant components in FGD include calcium and sulfur, with
lesser amounts of silica, aluminum, iron, and magnesium if fly ash is added. The
elemental composition of dry FGD materials has been determined based on data
from a variety of dry-scrubber technologies, including spray dryer systems, duct
injection, lime injection multistage burner (LIMB) processes, and a number of
fluidized bed combustion (FBC) processes (i.e., bed-ash process and cyclone ash
process) (18,19). The calcium content of dry FGD material varies in the range
from 10% to 30% depending on the particular scrubber technology. The sulfur
content of dry FGD material typically varies between 4% and 11%. The silicon
content of dry FGD may range from 2% to 11%, while the aluminum content can
vary from 1% to 7%. Table 4 shows the trace-element content of dry FGD
materials (20). Although detectable amounts of arsenic, cadmium, chromium,
copper, lead, molybdenum, nickel, selenium, and zinc are present in dry FGD
materials, levels of these constituents are typically lower than EPA land applica-
tion guidelines for sewage sludge.
For many CCB applications, it is important to understand the leaching
behavior of these materials. The EPA’s Toxicity Characteristic Leaching Proce-
dure (TCLP) is a commonly used method for characterizing the leaching potential
of organics, metals, and other inorganic constituents from CCB matrices (21).
Table 5 shows the results of TCLP analyses of dry FGD materials and ash
produced from various air pollution control technologies. Typically, very low
levels of organic materials are found in CCBs, and therefore, TCLP tests focus
on examining the leaching behavior of inorganic constituents. The TCLP values
for FGD shown in Table 5 were determined for a variety of dry scrubber
Minimization and Use of CCBs 429

TABLE 5 Range of Values Observed for TCLP Analysis of

Dry FGD Materials (19,20) and Ash (14)

Chemical constituent
(mg/liter) FGD Ash

pH 9.58–12.01 —
TDS 11,840–13,790 —
Ag <0.024 0.0–0.05
Al 0.12–0.20 —
As <0.005 0.026–0.4
B 0.543–2.17 0.5–92
Ba <0.002 0.30–2.0
Be 0.141–0.348 <0.0001–0.015
Ca 1,380–3,860 —
Cd <0.003 0.0–0.3
Co <0.014–0.026 0.0–0.22
Cr <0.005–0.028 0.023–1.4
Cu <0.013 0.0–0.43
Fe <0.029 0.0–10.0
Hg <0.0002 0.0–0.003
K 1.3–22.1 —
Li 0.04–0.18 —
Mg <0.04–1,360 —
Mn <0.001 0.0–1.9
Mo 0.025–0.088 0.19–0.23
Na 1.32–9.82 —
Ni <0.01 0.0–0.12
P <0.12 —
Pb <0.001–0.017 0.0–0.15
S 132–979 —
Sb <0.24 0.03–0.28
Se <0.001–0.005 0.011–0.869
Si 0.10–0.33 —
Sr 0.83–3.38 —
V <0.019–0.024 —
Zn <0.006 0.045–3.21
Cl– 19.6–67.8 —
SO32– <1.0–43.2 —
SO42– 236–2,800 —
430 Walker et al.

technologies. TCLP leachate typically meets most primary and secondary drink-
ing water standards. Levels of silver, arsenic, barium, cadmium, copper, iron,
mercury, manganese, nickel, phosphorus, antimony, and zinc in leachate are
typically below the limit of detection. For all FGD materials shown in Table 5,
high pH values are observed, thus making FGD an attractive product for applica-
tions requiring alkaline materials. Typically, with the exception of sulfur and
calcium, higher levels of most inorganic elements are found for TCLP tests
carried out with ash than for FGD. It should be noted that the acidic conditions
and high liquid-to-solids ratio of the TCLP test are perhaps more favorable for
leaching than conditions typically observed in field applications.

4 FACTORS AFFECTING CCB GENERATION

The physical and chemical properties of CCBs and the quantity of CCBs
produced will depend on the mechanical design and operation of the combustion
process, the type of air pollution control equipment utilized, as well as the
characteristics of the coal used in the combustion process (11). In order to
minimize CCB generation, it is important to understand how these factors affect
the type and amount of solid by-product produced. In all cases, however, efficient
energy production and low-pollutant air emissions must be maintained.

4.1 Boiler Technology

The boiler used in an electric power plant is a closed vessel that is heated from
the combustion of coal to produce hot water or stream. There are four major types
of boiler technologies in current commercial application: pulverized coal (PC)
boilers, stokers, cyclones, and fluidized bed combustion systems. Figure 2 shows
the approximate distribution of ash and slag produced by different kinds of boiler
technology.
The most widely used boiler technology is the PC boiler. The coal used in
PC boilers is finely ground prior to combustion. The large effective surface area
of finely ground coal used in PC boilers increases combustion efficiency. The
greater efficiency of combustion reduces the total volume of ash by-products
produced. There are two types of pulverized coal boilers; wet-bottom and
dry-bottom boilers. The larger-sized ash that falls to the bottom in a dry-bottom
process remains dry and becomes bottom ash. For the wet-bottom process, ash is
removed as a flowing slag. Large ash particles fall to the bottom of the furnace
and flow out of the furnace in a molten state which later solidifies as slag (10,11).
As seen in Figure 2, dry-bottom PC boilers produce 80% fly ash and 20% bottom
ash. PC boilers with a wet-bottom design produce 50% fly ash and 50% slag. The
predominance of fly ash in these two types of boilers is primarily a result of the
small particle size of ground coal used in the combustion process.
Minimization and Use of CCBs 431

Dry PC Boiler

Wet PC Boiler

Spreader
Stoker

Other Stoker

Cyclone

0 20 40 60 80 100
% Ash Proportion
Fly Ash Bottom Ash/Boiler Slag

FIGURE 2 Approximate ash distribution as a function of boiler technology

(10,11,22).

Stoker boiler technology is typically used in smaller utility plants. A stoker

boiler is classified based on the location of the stoker, the method of coal feeding,
and the method of grating coal in the furnace. Spreader stokers are the most
widely used of all stoker technologies (10,11). The bottom ash generated by
spreader stokers ranges from free-flowing ash to fused slag, while the bottom
ash created from other types of stokers is normally slag (10,11,22). Figure 2
shows that spreader stokers produce 35–60% fly ash and 40–65% bottom ash and
slag. Other types of stokers produce about 10% fly ash and 90% bottom ash
and slag.
Cyclone boilers are used for coal combustion and are designed to circulate
air to enhance the combustion of fine coal particles in suspension. This design
helps to reduce erosion and fouling problems in the boiler. Larger ash particles
stick to the molten layer of slag and flow out. Combustion occurs in a horizontal
cylindrical vessel attached to the boiler. This kind of design facilitates the flow
of molten slag and also reduces the cost of particulate collection (10,11). Most of
the by-product from the cyclone design is in the form of slag. Figure 2 shows that
cyclones produce 30% fly ash and 70% slag.
432 Walker et al.

Other technologies are also used for coal combustion. These alternative
boilers may also aid in controlling air emission. Fluidized bed combustion (FBC)
is a boiler technology that can be used with a variety of fuels (22). This
technology has a high combustion efficiency at low operating temperatures (11).
The fluidized bed combustion system consists of a blower that injects preheated
air into the fluidization vessel, and a bed material that can be sand or a reactive
solid. Injection of air into the vessel fluidizes the bed material and aids in
combustion. The amount of CCBs produced from FBC is based on the type of
FBC. Two types of FBC include bubbling fluidized bed systems and circulating
fluidized bed systems.
Bubbling fluidized bed systems have gas velocities of 5–12 ft/s. Gas flow
passes through the bed and causes the bed material to “bubble.” In bubbling FBC
systems, the particle size of bottom ash in the bed is usually larger and is packed
denser (~45 lb/ft3) than in circulating FBC systems (11,22). CCBs generated from
bubbling FBC systems include ash, sand, and other inert bed material. Lime or
limestone may be added directly to the bed to aid in sulfur emission control
(11,22). As a result, by-products from FBC boilers may also contain unreacted
lime, calcium sulfate, and/or calcium sulfite.
Circulating fluidized bed combustion systems have higher gas velocities of
about 30 ft/s. In circulating systems, some of the bed material is recovered from
the gas phase and reinjected into the fluidized bed vessel. Bottom ash in the bed
of circulating FBC systems is usually finer and more densely packed (~35 lb/ft3)
than in bubbling FBC systems (11,22). Ash generated from circulating FBC
systems consists mainly of fly ash, with lesser amounts of bottom ash (22).

4.2 Air Pollution Control Technology

The type of technology used for controlling pollutants released to the atmosphere
during coal combustion influences the generation and characteristics of CCBs.
There are two main categories of air pollution control technologies that generate
CCBs during coal combustion: particulate control and gaseous emission control
technologies.
Particulate control technologies during coal combustion capture fly ash
from the flue gases before they are released to the atmosphere. The processes
most often used for particulate control are electrostatic precipitation, fabric
filtration, scrubbers, and mechanical collectors. The electrostatic precipitator is
the most common process used for capturing fine ash particles in coal-fired
utilities (11,22). ESPs capture ash by applying an electrical charge to the ash
particles. The charged particles are subsequently attracted to oppositely charged
collector surfaces in an intense electrical field. Following collection, the particles
are sent to a hopper. This technology is appropriate for capturing fly ash from
coal with high sulfur content. In fact, sulfur oxides in the flue gas may increase
the efficiency of particle capture in the ESP (10,11,22). The capability of ESP to
Minimization and Use of CCBs 433

capture fly ash in the flue gas is more than 99% when this process is properly
operated and maintained (10,11).
A fabric filter unit, or baghouse, is an appropriate technology for particulate
control in combustion processes that use coal with low sulfur content. This
technology operates by forcing the flue gas through a fine mesh filter. Fly ash is
trapped and builds up on the filter surface. The ash on the filter forms a cake,
which is then periodically removed. The efficiency of the filter increases as ash
forms a thick layer on the filter surface. However, thick cake formation also leads
to greater head losses in the process. Fabric filters can remove over 99% of fly
ash from the flue gas in coal-fired utilities (10,11).
Scrubbers can also be used for particulate control and operate by applying
water to contact the fly ash in the flue gas in a spray tower. This technology also
can remove over 99% of large ash particles, but less than 50% for particles with
size smaller than 1 or 2 µm (11,22). Mechanical collectors are instruments used
for removing primarily large ash particles. They operate by forcing the ash
particles against a collector wall, where the dry ash by-product is collected. The
efficiency is lower than 90% for small particles.
Desulfurization technology is used for capturing gaseous sulfur oxides
from flue gas in coal-fired utilities. The use of desulfurization technology re-
sults in the generation of FGD material. There are two major types of FGD sys-
tems: nonrecovery and recovery systems. In the United States, 95% of
FGD systems are nonrecovery systems (11,22). Nonrecovery systems pro-
duce by-product material, mainly calcium sulfate or sulfite, that has to be
disposed or used. The nonrecovery FGD process can be separated into two
types, wet and dry systems. Wet systems operate by contacting the flue gas
with a slurry of water and sorbent. Examples of wet scrubber systems include
direct lime, direct limestone, alkaline fly ash, and dual-alkali. As mentioned
earlier, approximately 90% of FGD systems in the United States use lime or
limestone as a sorbent (17). Typically, the calcium sulfite/sulfate sludge produced
in wet systems is dewatered and mixed with fly ash and lime to produce
“stabilized” FGD. Examples of dry nonrecovery FGD systems include spray
drying and dry sorbent injection (11). Wet FGD systems produce more FGD per
pound of coal than that of dry systems because of the use of water in the process.
Recovery systems produce materials that can be used again in the FGD process,
because the sorbent can be recycled. Recovery FGD processes also have wet and
dry systems. Examples of recovery FGD systems include Wellman-Lord and
magnesium oxide systems and aluminum sorbent and activated-carbon sorbent
systems (11).

4.3 Types of Coal

Different types of coal have different heating values and also different ash
contents. The highest-ranked coal with respect to heating value is anthracite,
434 Walker et al.

while the lowest-ranked coal is lignite (22). The generation of ash and slag from
the combustion process is affected by the ash content, which is determined by the
rank and geographic source of the coal. Certain anthracite coals in the United
States generate about 30% ash. Bituminous coal ash content ranges from 6% to
12%, and subbituminous and lignite coals have ash contents ranging from 6%
to 19% (10,11). The rank of coal is affected by the specific region of coal mining,
mine, seam, and production method (11,22). The use of low-ash-content coal
reduces the management cost for removing particulate matter. In the United
States, the average ash content of coal used decreased from 13.5% in 1975 to
9.22% in 1996 (11,22,23). In addition, coal can be cleaned before the combustion
process to reduce the quantity of CCBs. The cleaning process as a pretreatment
for coal can reduce the ash content by 50–70% (22).
The generation of FGD depends on the sulfur content of the coal. The sulfur
content varies from region to region. Some coal produced in Iowa has a sulfur
content as high as 8% by weight, while coal from Wyoming may have an average
sulfur content of less than 1% by weight (22). Coal cleaning processes that reduce
the ash content of coal can reduce sulfur emissions by removing pyrites and other
metal sulfides from the coal.
Technologies for precombustion coal desulfurization are characterized as
physical, chemical, or biological. Currently, physical cleaning processes are the
most widely used. Physical processes use density differences to separate out
pyrites from the coal (24). Chemical processes use a chemical agent to desulfurize
coal. Chemical desulfurization processes that use chemical reagents such as ferric
salts, chlorine, and ozone are currently not cost effective, due to high chemical
recovery costs. In addition, chemical processes are also energy intensive due to
high operating pressures (600–1000 psi) and temperatures (100–500˚C) (25–28).
Recently, new chemical agents have been developed that may provide a cheaper
approach to the chemical cleaning of coal (29).
In biological processes, microorganisms are used to remove sulfur from
coal. Biological processes can be operated at room temperature and atmospheric
pressure, and therefore have lower costs than some physical and chemical coal
cleaning techniques. In addition, biological pretreatment of coal does not reduce
the BTU value of the coal, but instead, may increase coal energy content due to
the remaining biomass (25–27,30). Biological coal cleaning processes can re-
move both inorganic and organic sulfur. Thiobacillus ferrooxidans can oxidize
inorganic pyrite (FeS2) in coal, while microorganisms such as Rhodococcus
rhodochrous, Sulfologus brierleyi, and Sulfolobus acidocaldarius remove organic
sulfur compounds (31–35). Although precombustion cleaning can reduce flue gas
desulfurization requirements, additional by-products may be produced and energy
may be required.
Minimization and Use of CCBs 435

5 STRATEGIES FOR MINIMIZATION OF CCBS

The established hierarchy for minimization of waste materials in any process
consists of the following: reduction > recycle/reuse > treatment > disposal. In this
hierarchy, strategies for reduction and recycle/reuse are favored over “end-of-
pipe” treatment and disposal options for waste material.

5.1 Reduction at Source

Strategies for the reduction of CCBs at the source include process modifications,
feedstock improvements, improvements in efficiency of equipment, better man-
agement practices, and recycling of material within or between processes. Possi-
ble process modifications may include changes to boiler operating conditions,
selection of wet versus dry scrubbing technologies, or addition of precombustion
coal cleaning, to name a few. Changes in feedstock may also aid in CCB
minimization. Recent advances are providing more efficient technologies for
removal of pollutants from flue gas in coal combustion facilities. For example,
new sorbents with high efficiency for SO2 capture do a better job of removing
SO2 from flue gas while at the same time reducing the amount of solid by-prod-
ucts produced during the process. Better management of the coal combustion
process may also lead to improved CCB minimization. Internal audits provide
opportunities to optimize process operations, thus maximizing energy production
and minimizing CCB production. For example, one factor limiting use of CCBs
is the variability in CCB properties. Better management practices may reduce this
variability, lower total volume of CCBs produced, and enhance utilization.

5.2 Use of Coal Combustion By-products

A number of applications of CCBs have been developed and demonstrated in
order to reduce the amount of CCBs disposed of in landfills. The first column in
Table 6 shows demonstrated applications of CCBs. In 1998, the American Coal
Ash Association (ACAA) reported that 28 million metric tons of CCBs were used
in the United States. This represented 30% of all CCBs generated in that year.
One demonstrated application is the use of CCBs in cement, concrete, and grout.
CCBs may also be used in flowable and structural fill applications. Demonstrated
applications of FGD material include use in the production of wallboard or in
mine land reclamation.
The use of CCBs reduces the need for landfill space and also reduces the
utilization of natural resources. Table 6 presents some of the environmental
benefits associated with CCB use. Approximately 10 million metric tons of fly
ash and bottom ash were used as a replacement for cement in nonfill applications.
This was the largest application of fly ash and represented roughly half of all the
436 Walker et al.

TABLE 6 Major CCB Applications and Environmental Benefits of CCB Use

Reduction in emission
or natural resource utilization per year

Current
CCB utilization utilization rate 100% CCB utilizationa

Cement/concrete/grout 10 × 106 tons cement 32 × 106 tons cement

10 × 106 tons CO2b 32 × 106 tons CO2
Flowable fill 0.36 × 106 tons CO2 1.08 × 106 tons CO2
Structural fill 3.7 × 106 tons Soilc 11.2 × 106 tons soil
Wallboard gypsum 1.6 × 106 tons gypsum 16.5 × 106 tons gypsum
Mining applications 0.6 × 106 tons clay 2.2 × 106 tons clay
Total CCB reduction
in landfills 17 × 106 m3 59 × 106 m3
aThe environmental benefit for 100% CCB utilization is calculated assuming the percent
of material used for a particular application (e.g., cement/concrete/grout) is
independent of overall CCB utilization.
bAssumes 1 lb CO /1 lb cement not used, and 1 lb of cement replaced/1 lb fly ash used.
2
cAssumes 1 lb soil/1 lb of fly ash used.

fly ash used in 1998. It has been estimated that every ton of cement replaced with
fly ash eliminates the emission of approximately 1 ton of CO2 to the atmosphere
(36). Based on this figure, the use of fly ash in cement applications reduced the
release of CO2 by 10 million tons in 1998. Also, in 1998, 0.36 million metric tons
of fly ash were used for flowable fill. This application also replaced cement and
represented approximately 1.8% of all fly ash used. The use of fly ash in flowable
fills reduced the emission of CO2 by 0.36 million metric tons in 1998. If all of
the fly ash generated were used at current utilization percentages (50% for cement
replacement and 1.8% of flowable fill), CO2 emissions would be reduced by
roughly 32 million metric tons (31 million tons as a result of cement replacement
and 1 million ton due to the use of fly ash in flowable fill). Structural fill is an
application in which CCBs, and in particular fly ash, are used to replace natural
soil. In 1998, the use of fly ash and bottom ash for structural fill saved 3.6 million
metric tons of soil and could save up to 11.2 million metric tons of soil if 100%
of all CCBs were used at the current utilization rate for structural fill (13.2%).
In 1998, 2.2 million metric tons of FGD were used, which represented only
8% of all FGD produced. Of this amount, 1.6 million metric tons of FGD were
used in the production of wallboard. This represented the single largest use of
FGD and 73% of all FGD used in that year. As a result, approximately 1.6 million
metric tons of natural gypsum were saved by using FGD gypsum in the wallboard
industry. FGD could replace up to 18 million tons of gypsum if all FGD in the
Minimization and Use of CCBs 437

United States were reused, 73% of which for wallboard manufacturing. This
amount of FGD wallboard could supply a good fraction of the 1.6 million new
houses built in the United States every year. About 0.6 million tons of clay used
for mining applications were saved in 1998 by replacement with CCBs, and up
to 2.2 million tons could be saved if all CCBs were used (6.8% for mining
applications). The use of CCBs in 1998 reduced landfill space consumption by
about 17 million m3. If all CCBs were used, landfill space consumption would
be reduced by 59 million m3 each year.
5.2.1 Cement/Concrete/Grout Application
Using fly ash in cement and concrete increases the strength, workability, and
resistance to alkali–silica reactivity and sulfate, and reduces permeability, bleed-
ing, and heat of hydration (13). The Federal Highway Administration (FHWA)
and ACAA have reported that fly ash-enhanced concrete has lower strength than
pure Portland cement in early periods, but provides higher strength in the long
term (13). In 1998, 10.2 million metric tons of CCBs were used in concrete
applications. The amount of fly ash used with cement varies from 15% to 20%
of the total weight. Typically, 1–1.5 kg of fly ash is used for every 1 kg of cement
replaced in concrete applications (13). Lowering the cement content reduces the
emission of CO2 caused by the calcination of limestone and fuel burning during
cement production.
Future efforts to reduce nitrogen oxide emissions from coal combustion
facilities may negatively impact the utilization of fly ash in concrete. To control
emission of NOx, many coal-fired utilities may utilize new low-NOx emission
technologies. Some of these technologies operate at lower temperatures than
traditional boilers. Operation at lower temperature may result in an increase in
carbon and ammonia content of fly ash (13). Carbon content, which is typically
measured by determining the weight loss on ignition (LOI), affects concrete
strength development and so is restricted by industry standards. In the United
State, the specification of LOI of fly ash is between 3% and 5% for use in
ready-mix concrete (37,38). The use of low-NOx technology may increase the
LOI above 5% and thus reduce the utilization of fly ash in concrete applications
(39). For example, in 1998, the ACAA reported that 19 out of 20 coal-fired
utilities in Ohio may be affected by the low-NOx emission technologies, and 46%
of all coal-fired utilities in the United States may be required to control NOx (39).
5.2.2 Flowable Fill
Flowable fill is another application in which fly ash can be used in place of
cement. Flowable fill is defined as a self-leveling, self-compacting cementitious
material that is in a flowable condition at the time of placement and has a
compressive strength of 1200 psi or less at 28 days (40). Flowable fill is also
known as control density fill (CDF), controlled low-strength material (CLSM),
438 Walker et al.

unshrinkable fill, flowable mortar, plastic-soil cement slurry, K-Krete, and/or

Flash Fill (40). Flowable fill may contain a mixture of fly ash, bottom ash, water,
and Portland cement. This application is especially suitable for filling in void
spaces that are difficult to reach.
The use of fly ash and bottom ash in flowable fill applications is increasing.
Recently, a technical guidance manual for flowable fill applications has been
written: the ACI229 Committee Report on Controlled Low Strength Material
(40). However, the current use of CCBs in this application is still relatively small.
As mentioned above, only 360,000 metric tons of fly ash and bottom ash were
used in flowable fill applications in the United States, which represented only
1.8% of all fly ash and 0.3% of all bottom ash used.
5.2.3 Embankment/Structural Fill
Embankment and structural fill is currently the second largest application of
CCBs. CCBs are used to replace conventional soil in structural fill applications.
CCBs have many advantages over natural soil for use as structural fill, including
lower unit weight, high shear strength-to-unit weight ratio, and good availability
in bulk (13). These benefits lead to significant savings in material costs when
CCBs are used. For example, in 1993, FGD material from pressurized fluidized
bed combustors (PFBCs) was used as embankment to repair part of Ohio State
Route (SR) 541 near Coshocton, Ohio (41–43). The cost of this project was
$77,000, while the estimated cost of using conventional materials would have
been between $105,000 and $120,000. This represented a savings of between
26% and 36%, before counting the environmental benefits associated with saving
existing natural resources.
The use of CCBs for structural fill has increased in recent years. American
Society for Testing and Materials (ASTM) Standard E1861 defines the appropri-
ate guidelines for the use of CCBs for structural fill (36,44). In 1998, approxi-
mately 4 million tons (12.7%) of CCBs were used as structural fill. Fly ash and
bottom ash were used the most, 2.5 and 1.1 million metric tons respectively.
Although only 18,000 metric tons of FGD were used in 1998, this material has
demonstrated excellent strength over a wide range of moisture contents compared
to natural soils (41–43). FGD could be an excellent alternative for use in
embankment and structural fill applications in the future (41–43).
5.2.4 Stabilized Base/Subbase
Mixing fly ash with lime and aggregate can produce a good-quality road base and
subbase. This material is also called lime–fly ash aggregate (LFA) or pozzolanic-
stabilized mixture (PSM) base. The fly ash content in the material for road base
typically varies from 12% to 14%. Lime content also varies from 3% to 5% (13).
The proportion of lime can be replaced with Portland cement or cement kiln dust.
The advantages of using LFA or PSM for road base and subbase applications
Minimization and Use of CCBs 439

include increased strength and durability of the mixture, lower cost, autogenous
(self-generating) healing, and less energy consumption (13). Use of LFA reduces
the energy to produce cement. In addition, it does not require heat, as an asphalt
base does.
In 1998, 3.6 million tons of CCBs were used in road base and subbase
applications. The CCBs used most commonly for these applications were fly ash
and bottom ash (1.4 million tons of fly ash and 1.6 million tons of bottom ash).
For example, 10 municipal and commercial projects in and around the City of
Toledo used approximately 1 million tons of LFA from 1970 to 1985 (45). In
1998, Hunt et al. (46) developed an economic analysis comparing the use of LFA
and other pavement materials. It was found that LFA base pavement was 20%
cheaper than aggregate base pavement and 15% cheaper than bituminous base
pavement (46).
5.2.5 Mining Applications
Use of CCBs in mining applications can aid in the abatement of acid mine
drainage (AMD), reduce subsidence, and reduce off-site sedimentation control.
Acid mine drainage is an environmental problem caused by water drainage from
abandoned mines and coal refuse piles. Water in abandoned mines reacts with
pyrite and other metal sulfides in the presence of oxygen and produces acidity
(47). Fly ash and/or FGD can be used to minimize the exposure of pyrite to water
and oxygen. Also, the alkalinity in FGD can neutralize AMD already generated.
In 1998, 2 million metric tons of CCBs in the United States were used for mining
applications. This represented 6.8% of all CCBs used. Most of material used in
this application was fly ash (1.9 million tons).
At mine sites, refuse waste materials such as soil, rock, slate, and coal are
commonly found and can pose serious environmental problems. These refuse
waste piles, commonly called “gob piles,” contain pyrite and produce acidity.
FGD material may be used as a liner to construct ponds to collect runoff from
gob piles. The low permeability of FGD may also be used as a cap to minimize
the amount of water entering the gob pile. For example, the Rock Run reclamation
site at New Straitsville, Ohio, had approximately 14 acres of gob piles prior to
reclamation utilizing CCBs (39). The drainage from these gob piles to Rock Run
had a pH of 2.27. At this site, 2 ft of stabilized FGD from the Conesville coal-fired
power plant in Ohio was used to cover the gob piles. Utilization of FGD as a cap
material resulted in improvements in water quality at the site. Estimated cost
saving from using FGD for gob pile reclamation, instead of clay, ranged from
$8,350 to $12,600 (39).
Another potential application of FGD is in the reclamation of abandoned
surface mines (39,48). An example of such an application was carried out at the
Fleming site located in Franklin Township of Tuscarawas County, Ohio (49,50).
The Fleming site was an abandoned clay and coal mine. In the past, flooding
440 Walker et al.

downstream of the Fleming site resulted in offsite soil sedimentation at an

estimated rate of 450 tons/acre/year. In 1994, several AMD treatment approaches
were developed at this site utilizing limestone, a dry FGD material from an PFBC
plant, and a 2.5:1 mixture of FGD and yard waste. All treatments resulted in
neutralization of mine drainage. Trace-metal analysis showed water quality
improved and, in fact, met all drinking water standards.
5.2.6 Wallboard Manufacture
Instead of being stabilized with fly ash and sent to landfill, FGD material can be
used as a material to manufacture wallboard. The calcium sulfite in FGD can be
oxidized to calcium sulfate and dewatered to produce synthetic FGD gypsum
(39,51). In 1998, 1.6 million tons of FGD were used in wallboard manufacturing
(1). This represented 72.9% of all FGD used. There will be a $20 million
investment in oxidation and dewatering equipment to produce synthetic FGD
gypsum at the Zimmer plant (located in Moscow, Ohio) by Cinergy, American
Electric Power, and Dayton Power and Light (52–54). The synthetic FGD gypsum
produced at the Zimmer plant will supply a wallboard plant at Silver Grove,
Kentucky. Upon completion, the Silver Grove wallboard plant will have the
highest wallboard production capacity in the world, 900 million ft2 per year.
5.2.7 Agricultural Applications
There are many advantages to using CCBs in agricultural applications, and
addition of CCBs can improve the properties of soil and increase plant growth.
For example, the lime in CCBs can raise the pH of acidic soil to neutral levels,
and can increase yields of alfalfa above those observed using agricultural lime-
stone (18). Trace elements in FGD can be utilized by plants and may aid in plant
growth (24). In 1998, 920,000 metric tons of CCBs were used in agricultural
applications. It has been estimated that 25% of the agricultural lime used in Ohio
could be replaced by FGD (39). Assuming that every ton of agricultural lime is
equivalent to 1.67 tons of FGD, the potential use of FGD for replacing agricul-
tural lime in Ohio alone would be 365,000 tons per year (55).
CCBs can also be used in animal production facilities as a base for livestock
feeding and hay storage (56). The moisture from mud can deteriorate the quality
of hay bales and decrease animal yields. Utilizing CCBs as a base material can
reduce muddy conditions. In 1997, 24 livestock feeding and hay storage pads
(ranging size from 1,500 to 15,000 ft2) were built in eastern and southern Ohio.
Over 150 FGD pads were constructed in 12 counties in Ohio in 1998 (56). An
economic analysis of the construction of FGD pads performed for Gallia County,
Ohio, in 1997 showed that FGD pads were 26% cheaper than aggregate pads and
65% cheaper than concrete pads (57).
FGD can also be used for constructing low-permeability liners for water-
holding ponds and manure storage. Normally, the construction of manure facili-
Minimization and Use of CCBs 441

ties in the United States utilizes compacted clay. The successful utilization of
FGD requires that the material provide sufficiently low permeability and not
degrade groundwater quality. A study has been conducted to evaluate the perme-
ability and the water quality of the leachate from an FGD liner (58–60). It was
found that the FGD liner has a permeability as low as 10–7 cm/s. Moreover,
trace-element concentrations in the leachate were generally lower than the
drinking water standards (60). It has been estimated that replacing clay or
geomembranes with FGD material for pond liners could save construction costs
by as much as $2–$3 per square foot (58–60).

5.3 Treatment and Disposal

Options for treatment and disposal are important considerations for evaluating the
ultimate fate CCBs. For example, in wet scrubbing technologies a calcium sulfite
slurry is produced that is dewatered prior to disposal. The dewatering step reduces
the total volume of material going to final disposal, but it also generates a liquid
waste stream. The dewatered scrubber sludge is typically mixed with equal
amounts of fly ash in order to improve the handling characteristics of the material.
The exact proportion of fly ash to dewatered scrubber sludge has a significant
impact on the performance of these materials during disposal and/or use.

6 LIFE CYCLE ASSESSMENT (LCA) MODEL FOR

MINIMIZATION OF CCBS
A life cycle assessment model (61) can be used to determine optimum strategies
for minimizing the environmental impacts associated with coal combustion
processes. The benefit of this approach is that it does not trade gains made in
minimizing the amount of CCBs with other environmental impacts. In a life cycle
assessment model, the coal combustion process can be broken down into the
following elements: resource extraction, electricity generation, electricity trans-
mission, and electricity use. In developing an LCA model, each element of the
coal combustion process must be evaluated with respect to the following vari-
ables: materials choice, energy use, solid residues, liquid residues, and gaseous
residues. Once this has been carried out, an LCA matrix can be developed which
includes a numerical score (from 1 to 4) for each variable and element. The rank
of the entire process can then be determined as

R = ∑ ∑ Mij (1)
i j

where R is the rank of the process and Mij is the numerical score summed over
each variable (i) and element (j). It should be noted that in certain cases it might
442 Walker et al.

not be possible or appropriate to develop a single score defining a process. In such

cases, ISO 14040 provides criteria by which a comparison can be made between
two or more potential options using the LCA framework (62).
The LCA model provides a framework for analyzing different process
alternatives and determining the most environmentally sound option. For example, it
was mentioned above that one strategy to minimize CCB generation is to utilize a
low-sulfur, low-ash coal source. However, utilizing low-sulfur, low-ash coal may not
be the most environmentally sound alternative in all cases. If a preferable coal source
is located far from the coal-fired utility, significant environmental impacts may be
associated with the transportation of these materials.
Recently, a comparison between the use of fly ash and natural soil for
structural fill was conducted using the LCA framework (63). This study consid-
ered a number of factors, such as natural resource use, energy consumption, air
emissions, and solid waste. It was concluded that the use of fly ash as structural
fill resulted in less use of natural raw materials, significant reductions in solid
waste disposal, and was more energy efficient than the use of natural resources.
In this particular study, however, the impact of fly ash on water quality was not
assessed quantitatively, due to variations in soil and fly-ash leaching behavior.

7 BARRIERS TO CCB UTILIZATION

Currently, the cost of disposing and managing CCBs in landfills is relatively low,
and this reduces the incentive to utilize CCBs. For example, landfill costs for
coal-fired utilities in Ohio range from $2 to $40 per ton (18,39). A lack of
standards for using CCBs is also a barrier for CCB utilization. For example, the
use of CCBs in structural fill has typically fluctuated in the past. Increasing use
of CCBs for structural fill is expected in the future as a result of ASTM Standard
E1862 (44). New, large-volume, cost-effective applications of CCBs are needed.
Because CCBs are secondary products of energy production, few controls
on their generation are in place. As a result, the physical, chemical, and engineer-
ing properties of CCBs may vary. Significant variations can be observed in CCB
properties within a given plant, as well as variations between plants. Future
emission standards for coal-fired power plants will also affect the utilization of
CCBs. As mentioned above, NOx control technologies influence the carbon
content of fly ash. The increased carbon content of fly ash generated from
low-NOx boilers could reduce the potential for use in cement applications, the
biggest current CCB market.
Although extensive testing has shown that CCBs are nontoxic and nonhaz-
ardous, public concern regarding the environmental impacts of CCBs may limit
utilization. Continued research on the environmental impacts associated with
CCB use and public education are needed. As is the case for any product or
Minimization and Use of CCBs 443

process, efforts should be focused primarily on reducing CCB generation, while

maintaining efficient energy production and air pollution controls.

8 CONCLUSIONS
Fly ash, bottom ash, boiler slag and FGD material are by-products from the
combustion of coal and are considered to be solid wastes from a federal regulatory
perspective. Currently, 100 million tons of CCBs are produced in the United
States every year. Approximately 70 million tons of CCBs are disposed of in
landfills and surface impoundments.
The two primary strategies for minimizing CCBs include reduction at
source and effective utilization. A life cycle assessment model should be used to
determine the most environmentally beneficial approach for minimizing CCB
generation and disposal. A number of applications have been developed for using
CCBs, including the use of fly ash as a substitute for cement in concrete and grout
applications, the use of fly ash in flowable and structural fill, the use of calcium
sulfate-rich FGD scrubber sludge as a replacement for natural gypsum in wall-
board manufacturing, and a variety of mine reclamation applications. The utiliza-
tion of CCBs reduces the consumption of natural resources, reduces emissions of
greenhouse gases to the atmosphere, and reduces the need for new landfill
construction. Potential barriers to CCB use include the low cost of landfilling, the
lack of available large-volume or high-value applications, and variations in CCB
material properties.

ACKNOWLEDGMENTS
The authors would like to thank the Ohio Coal Development Office (OCDO) for
its support of much of the research cited herein. We would also like to thank the
reviewers for their helpful comments.

REFERENCES
1. B. R. Stewart, S. S. Tyson, and G. J. Deinhart, American Coal Ash Association
Programs to Advance the Management and Use of Coal Combustion Products
(CCPs). 1999 Int. Ash Utilization Symp., October 18–20,1999.
2. American Coal Ash Association, Coal Combustion Byproduct (CCB) Production &
Use: 1966–1994. Report of the Coal Burning Electric Utilities in the United States,
Alexandria, VA, May 1996.
3. U.S. Environmental Protection Agency, Characterization of Municipal Solid Waste in
the United States 1998 Update, July 1999.
4. U.S. Geological Survey, Mineral Commodity Summaries, January 1999.
5. C. D. Cooper and F. C. Alley, Air Pollution Control: A Design Approach, 2nd ed.,
Waveland Press, 1994.
444 Walker et al.

6. Code of Federal Regulations, Title 40, Part 60, Standards of Performance for New
Stationary Sources. Washington, DC: Office of Federal Register, July 1, 1999.
7. U.S. Environmental Protection Agency, Profile of the Fossil Fuel Electric Power
Generation Industry, EPA /310-R-97-007, September 1997.
8. Code of Federal Regulations, Title 40, Part 76, Acid Rain Nitrogen Oxides Emission
Reduction Program. Washington, D.C.: Office of Federal Register, July 1, 1999.
9. United States Code, Title 42, Section 6921 (b)(3)(A)(i), Identification and Listing of
Hazardous Waste. Washington DC: U.S. Government Printing Office, 1995.
10. U.S. Environmental Protection Agency, Wastes from the Combustion of Coal by
Electric Utility Power Plants, Report to Congress, EPA/530-SW-88-002, February
1988.
11. U.S. Environmental Protection Agency, Wastes from the Combustion of Fossil Fuels,
Volumes 1 and 2—Methods, Findings, and Recommendations, Report to Congress,
EPA 530-S-99-010, March 1999.
12. Ohio Administrative Code, Chapter 3745-51-04 (B) (4), Exclusion: Identification and
Listing of Hazardous Waste, June 25, 1998.
13. American Coal Ash Association, Fly Ash Facts for Highway Engineers, FHWA-SA-
94-081. Washington, DC: Federal Highway Administration, August 1995.
14. U.S. Environmental Protection Agency, The Class V Underground Injection Control
Study, Volume 10, Mining, Sand, or Other Backfill Wells, EPA/816-R-99-014,
September 30, 1999.
15. L. K. Moulton, Bottom Ash and Boiler Slag. Proc. Third Int. Ash Utilization Symp.,
Information Circular No.8640. Washington, DC: U.S. Bureau of Mines, 1973.
16. U.S. Department of Transportation. Coal Fly Ash—User Guideline—Embankment or
Fill, http://www.tfhre.gov///////hnr20/recycle/waste/cfa54.htm, 2000.
17. R. Kalyoncu, Coal Combustion Products. U.S Geological Survey, Mineral Informa-
tion, 1997.
18. R. Stehouwer, W. Dick, J. Bigham, R. Forster, F. Hitzhusen, E. McCoy, S. Traina,
W. Wolfe, R. Haefner, and G. Rowe, Land Application Uses for Dry FGD By-prod-
ucts. Phase 2. EPRI Rep. TR-109652. Palo Alto, CA: Electrical Power Research
Institute, 1998.
19. W. A. Dick, J. M. Bigham, R. Forster, F. Hitzhusen, R. Lal, R. Stehouwer, S. Traina,
W. Wolfe, R. Haefner, and G. Rowe, Land Application Uses of Dry FGD By-product.
Phase 3. EPRI Rep. TR-112916. Palo Alto, CA: Electrical Power Research Institute,
1999.
20. W. A. Dick, J. M. Bigham, R. Forster, F. Hitzhusen, R. Lal, R. Stehouwer, S. Traina,
W. Wolfe, R. Haefner, and G. Rowe, Land Application Uses for Dry Flue Gas
Desulfurization By-products, Executive Summary. U.S. Geological Survey, January
1999.
21. U.S. Environmental Protection Agency, Test Methods for Evaluating Solid Waste
Physical/Chemical Methods, EPA SW-846, 3rd ed. Washington, DC: U.S. Govern-
ment Printing Office, 1986.
22. S. C. Stultz and J. B. Kitto (eds.), Steam: Its Generation and Use. Barberton, OH:
Babcock & Wilcox, 1992.
23. Energy Information Administration, Cost and Quality of Fuels for Electric Utility
Plants, 1996 Tables. U.S. Department of Energy, May 1996.
Minimization and Use of CCBs 445

24. R. F. Korcak, Agricultural Uses of Coal Combustion Byproducts. In R. J. Wright,

W. D. Kemper, P. D. Milner, J. F. Power, and R. F. Korcak (eds.), Agricultural Uses
of Municipal, Animal, and Industrial Byproducts. U.S. Department of Agriculture,
January 1998.
25. F. Kargi, Microbial Coal Desulfurization. Enzyme Microb. Technol., vol. 4, pp. 13–19,
1982.
26. F. Kargi, Microbial Desulfurization of Coal. In A. Mizrahi and A. L. van Wezel (eds.),
Advances in Biotechnological Process, vol.3, pp. 241–272. New York: Alan R. Liss,
1984.
27. T. Omori, L. Monna, Y. Saiki, and T. Kodama, Desulfurization of Dibenzothiophene
by Corynebacterium sp. Stain SY1. Appl. Environ. Microbiol., vol. 58, pp. 911–915,
1992.
28. T. D. Wheelock, Coal Desulfurization (Chemical and Physical Methods), ACS Symp.
Ser. Washington, DC: American Chemical Society, 1977.
29. K. K. Ho, G. V. Smith, R. D. Gaston, R. Song, J. Cheng, F. Shi, and K. L. Gholson,
Desulfurization of Coal with Hydroperoxides of Vegetable Oils, Tech. Rep. Decem-
ber 1, 1994 through February 28, 1995. U.S. Department of Energy, March 1996.
30. P. Bos and J. G. Kuenen, Microbial Treatment of Coal. In H. L. Ehrlich and C. L.
Brierly (eds.), Microbial Mineral Recovery, pp.343–377. New York: McGraw-Hill,
1990.
31. S. A. Denome, E. S. Olson, and K. D. Young, Identification and Cloning Genes
Involved in Specific Desulfurization of Dibenzothiopene by Rhodococus sp. Strain
IGTS8. Appl. Environ. Microbiol., vol. 59, pp. 2837–2843, 1993.
32. J. R. Gallagher, E. S. Olson, and D. C. Stanley, Microbial Desulfurization of
Dibenzothiophene, A Sulfur-Specific Pathway. FEMS, vol. 107, pp. 31–36, 1993.
33. Y. Izumi, T. Ohshiri, H. Ogino, Y. Hine, and M. Shimao, Selective Desulfurization of
Dibenzothiophene by Rhodococcus erythropolis D-1. Appl. Environ. Microbiol., vol.
60, pp. 223–226, 1994.
34. B, Lei and S. C. Tu, Gene Overexpression, Purification, and Identification of a
Desulfurization Enzyme from Rhodococcus sp Strain IGTS8 as a Sulfide/Sulfoxide
Monoxygenase. J. Bacteriol., vol. 178, pp. 5699–5705, 1996.
35. C. S. Piddington, B. R. Kovacevich, and J. Rambosek, Sequence and Molecular
Characterization of a DNA Region Encoding the Dibenzothiophene Desulfurization
Operon of Rhodococcus sp. Strain IGTS8. Appl. Environ. Microbiol., vol. 61,
pp. 468–475, 1995.
36. B. R. Stewart and R. S. Kalyoncu, Materials Flow in the Production and Use of Coal
Combustion Products. 1999 Int. Ash Utilization Symp., October 18–20, 1999.
37. Pittsburgh Mineral & Environmental Technology, Coal Ash Utilization Tech-
nologies—Introduction, 30 December 1999, http://www.pmetinc.com/coal_ash/
intro.html.
38. American Society for Testing and Materials (ASTM), ASTM C618-92a, Standard
Specification for Fly Ash and Raw or Calcined Natural Pozzolan for Use as Mineral
Admicture in Portland Cement Concrete. ASTM Designation C618-92a-1994. West
Conshohocken, PA: ASTM, 1994.
39. T. S. Butalia and W. E. Wolfe, Market Opportunities for Utilization of Ohio Flue Gas
446 Walker et al.

Desulfurization (FGD) and Other Coal Combustion Products (CCPs). Columbus,

OH: The Ohio State University, May, 2000.
40. American Concrete Institute, Controlled Low Strength Materials (CLSM), American
Concrete Institute Committee 229, Rep. 229R-94. Detroit, MI: American Concrete
Institute, July 1994.
41. S. M. Nodjomian, Clean Coal Technology By-products Used in a Highway Embank-
ment Stabilization Demonstration Project, M.S. thesis, The Ohio State University,
Columbus, OH, 1994.
42. S. M. Nodjomian and W. E. Wolfe, Field Demonstration Projects Using Clean Coal
Technology By-products. Second Annual Great Lakes Geotechnical/Geoenvironmen-
tal Conf., West Lafayette, IN, May 1994.
43. S. H. Kim, S. Nodjomian, and W. E. Wolfe, Field Demonstration Project Using Clean
Coal Technology By-products (CCBs). American Coal Ash Association and Electric
Power Research Institute, EPRI TR-104657, vol. 1, p. 16(1–15), Orlando, FL,
January 15–19, 1995.
44. American Society for Testing and Materials (ASTM), ASTM E1861 Standard Guide
for Use of Coal Combustion By-products in Structural Fills, ASTM Designation
E1861-97. West Conshohocken, PA: ASTM, 1997.
45. Transportation Research Board, Synthesis of Highway Practice 199: Recycling and
Use of Waste Materials and By-products in Highway Construction. Washington, DC:
National Academy Press, 1994.
46. R. G. Hunt, L. E. Seitter, J. C. Collins, R. H. Miller, and B. S. Brindley, Final Report:
Data Collection and Analyses Pertinent to EPA’s Development of Guidelines for
Procurement of Highway Construction Products Containing Recovered Materials,
Volume K: Issues and Technical Summary, EPA Contract 68-01-6014, July 6, 1981.
47. A. G. Kim, B. S. Heisey, R. L. P. Kleinmann, and M. Deul, Acid Mine Drainage:
Control and Abatement Research, BuMines IC 8905, 1982.
48. T. S. Butalia, W. E. Wolfe, and W. A. Dick, Developments in Utilization of CCPs in
Ohio. Proc. 13th Int. Symp. on Use and Management of Coal Combustion Products
(CCPs), Orlando, FL, January 11–15, 1999.
49. W. Dick, R. Stehouwer, J. Bigham, and R. Lal, Dry Flue Gas Desulfurization
By-products as Amendments for Reclamation of Acid Minespoil. Proc. Int. Land
Reclamation and Mine Drainage Conf., Pittsburgh, PA, April 24–29, 1994, pp. 129–138.
50. R. C. Stehouwer and W. Dick, Soil and Water Quality Impacts of a Clean Coal
Combustion By-product Used for Abandoned Mined Land Reclamation. Proc. 12th
Int. Symp. on Coal Combustion By-product (CCB) Management and Use, American
Coal Ash Association and Electric Power Research Institute, vol. 1, p. 7(1–12).
51. J. H. Beeghly, K. J. Smith, and M. Babu, The Dewatering and Agglomeration of FGD
Gypsum into Micropellets for Use as a Soil Amendment. Proc. 1997 Int. Ash
Utilization Symp., Lexington, KY, October 20–22, 1997.
52. Zimmer to Provide Raw Material, The Cincinnati Enquirer, January 28, 1999.
53. WH Zimmer, Generating Station to Invest $20 Mil to Make High Quality Synthetic
Gypsum to Be Sold to Lafarge Gypsum Div for a $90 Mil Wallboard Plant, Power
Eng., vol. 103, no. 5, p. 12, 1999.
54. Lafarge Corp Plans to Build a US $90 Mil, 900 Mil Square Feet/Year Greenfield
Minimization and Use of CCBs 447

Gypsum Wallboard Plant in Silver Grove, KY. Industrial Specialties News, vol. 13,
no. 3, February 8, 1999.
55. R. Stehouwer, W. Dick, J. Bigham, L. Forster, F. Hitzhusen, E. McCoy, S. Traina,
and W. E. Wolfe, Land Application Uses for Dry FGD By-products, Rep. TR-105264,
Res. Proj. 2769-02. Palo Alto, CA: Electrical Power Research Institute, 1995.
56. T. S. Butalia, P. Dyer, R. Stowell, and W. E. Wolfe, Construction of Livestock Feeding
and Hay Bale Storage Pads Using FGD Material, The Ohio State Extension Fact
Sheet, AEX-332-99. Columbus, OH: The Ohio State University, 1999.
57. Gallia Country NRCS, Heavy Use Livestock Pads Constructed of Stabilized FGD
By-product, Report Prepared by Gallia County NRCS and Radian Corporation,
Federal Energy Technology Center, Pittsburgh, PA, 1997.
58. T. S. Butalia, S. Mafi, and W. E. Wolfe, Design of Full Scale Demonstration Lagoon
Using Clean Coal Technology By-products. 13th Int. Conf. on Solid Waste Technol-
ogy and Management, Philadelphia, PA, November 16–19, 1997.
59. W. E. Wolfe and T. S. Butalia, Use of FGD as an Impervious Liner, 23rd Int. Tech.
Conf. on Coal Utilization and Fuel Systems, Clearwater, FL, March 9–13, 1998.
60. W. E. Wolfe, T. S. Butalia, and C. Fortner, Preliminary Performance Assessment of
an FGD-Lined Pond Facility. Proc. 13th Int. Symp. on Use and Management of Coal
Combustion Products (CCPs), Orlando, FL, January 11–15, 1999.
61. T. E. Graedel and B. R. Allenby, Design for Environment. Prentice-Hall, Englewood
CLiffs, NJ, 1996.
62. Draft ISO/DIS 14040, Environmental Management, Life Cycle Assessment, Princi-
ples and Framework, 1996.
63. American Coal Ash Association, Comparison of Coal Combustion Products (CCPs)
Used for Structural Fill Material vs. Disposal in a Landfill Using the Life Cycle
Analysis Framework. Alexandria, VA: ACAA, September 1997.
21
Engineered Wetlands for Metal
Mining-Impacted Water Treatment

Herold J. Gerbrandt
Montana Tech of the University of Montana, Butte, Montana

1 INTRODUCTION
Mining is a major point source of water pollution in America and in the world.
In mining zones containing pyritic ore (sulfur-bearing ore), acid mine drainage
(AMD) contributes degraded water quality in general and potentially toxic metals
in specific to receiving streams, lakes, and oceans. This chapter explains the
production of AMD, summarizes common solutions to the problem, and presents
engineered wetlands as a possible new control of this very serious pollutant.
Waste minimization is not addressed here, although many possibilities for waste
minimization exist within the mining industry.

2 BACKGROUND
U.S. mining has supplied national and world demand for gold, iron, copper, silver,
molybdenum, and a host of other metals for two centuries. Butte, Montana’s
“Richest Hill on Earth” supplied copper to electrify America in the 1920s through
the 1940s. U.S. metals production helped turn the tide of World War II with an
overwhelming armory of guns, planes, tanks, and ships. In more recent times,

449
450 Gerbrandt

demand for jewelry and electronic parts has also been met by the national
mining industry.
The success of this extractive enterprise has come at a cost to the environ-
ment. While surface impact is often limited to isolated areas, groundwater and
surface water contamination has been more extensive and far more difficult
to address. Sulfuric ore bodies, when brought to the surface, have the potential
to produce acidic waters when exposed to water and oxygen (air). Acidic waters
(pH values between 2 and 5 Standard Units) can then mobilize metals by
dissolving them and transporting them downstream or vertically down to the
water table.
This chapter outlines the problem of mine-impacted surface and groundwa-
ter, then discusses conventional approaches to reduce environmental impact.
Lastly, results of a demonstration-scale engineered wetlands to clean metals-laden
waters are presented as a natural alternative to conventional methods. Throughout
the chapter, illustrations of various topics are drawn for the nation’s largest
Superfund site, the Butte Area/Clark Fork River Superfund Project. The site
stretches from the hills above Butte, Montana, 120 miles down the Clark Fork
River to the Milltown Reservoir just outside of Missoula, Montana. The Butte
area has a rich history of mining beginning in the 1860s, as well as a legacy of
environmental degradation resulting from the mining and smelting in the vicinity.
Major sources of acidic and metals-laden waters include the following.
Mine adits. These horizontal tunnels access ore bodies or veins within mountains.
Once the underground workings are abandoned, the adits become unwanted
drains for the tunnels, shafts, drifts, and stopes that make up the mine works.
Waste dumps. Made up of overburden soil and rock as well as ore too low
in grade to process, waste dumps can become sources of acid rock
drainage (ARD) if not isolated from the environment.
Tailings. Metals ore is milled down into fine particles so that the metals can
be removed by physical/chemical processes. The remaining fine-grade
material is discarded. Modern tailings disposal places this material in
lined empoundments. Historic as well as present-day operations such as
Freeport-McMoRan’s Grasberg mine (1) often discharged tailings into
nearby waterways to be washed downstream.
Pit lakes. Large open pit mines have, to date, not been required to backfill
after the economical ore has been removed. If the pit intercepts the
groundwater table, the groundwater will begin filling the pit when
dewatering pumps are turned off. Groundwater flowing into the pit will
carry dissolved metals into the lake. When the lake surface rises to the
natural groundwater level, acidic water carrying dissolved metals may
reverse directions and began flowing out of the lake into the aquifer (2).
In the cases of adits and pit lakes, groundwater flowing through the unmined ore
body surrounding the mine workings can generate acid and dissolve metals. Waste
Engineered Wetlands for Mining-Impacted Water 451

dumps and tailings become ARD producers when rainfall and snowmelt move
through the material. The resulting surface flow or infiltration carries metals and
acidity to down-gradient receptors (streams, lakes, aquifers).

3 CHEMISTRY OF METALS-IMPACTED WATERS

Iron sulfide ores marcasite and pyrite (both FeS2) are prevalent above and below
coal seams as well as in metal deposits. They differ by crystal structure. Pyrite is
the predominant of the two, and is the major source of acid mine drainage.
Stoichiometric reactions are

FeS2 + 7⁄2O2 + H2O = Fe2+ + 2SO42− + 2H+ (1)

Once the reaction is started and the ferrous iron is generated, this reaction looses
significance compared to the following three.

Fe2+ + 1⁄4O2 + H+ = Fe3+ + 1⁄2H2O (2)

The oxygenation of ferrous iron in the chemistry lab is very slow, approximately
1000 days.

Fe3+ + 3H2O = Fe(OH)3(s) + 3H+ (3)

Ferric iron is hydrolyzed and 3 moles of acid (H+) are produced.

FeS2 + 14Fe3+ + 8H2O = 15Fe2+ + 2SO42− + 16H+ (4)

This reaction rate is rapid (20–1000 min), and 16 moles of acid are produced.
It can be seen that the ferrous iron produced in Eq. (4) will feed the reaction
represented by Eq. (2). In the presence of excess pyrite, reactions (2), (3), and (4)
are now self-sustaining, yielding large quantities of acid in steps 3 and 4.
How is the “acid” produced? The hydrogen ions generated in Eqs. (3) and
(4) combine with sulfate or phosphate ions to form their respective acids:

2H+ + SO42− = H2SO4 sulfuric acid (5)

3H+ + PO42− = H3PO4 phosphoric acid (6)

Sulfate ions, and often phosphate ions, are present in excess in pyritic ore.
In a hypothetical chemistry lab, acid generation should not be a problem,
as Eq. (2), the oxygenation of ferrous iron, has an extremely slow rate of reaction.
In nature, microorganisms speed up the rate of reaction by mediating the
oxidation of ferrous iron. This is performed principally by Thiobacillus and
Ferrobacillus ferrooxidans, as well as other sulfide-reducing bacteria (SRB). pH
is also important, as the reactions are slow unless pH < 4. Thus, potential ARD
material may lie nonreactive for years until the Eq. (1) reaction lowers the pH to
452 Gerbrandt

around 4. Then, the generation of acid is fast. ARD from adits, waste dumps,
tailings, or pit lakes may have pH values less than 2.
The final result of the acidification of water in contact with ARD materials
is the dissolution of metals in the same materials or in the downstream environ-
ment. It has long been known that most metals will dissolve at extremely low or
extremely high pH. A typical metals concentration-versus-pH plot shows a
V-shaped zone of solids precipitate, with the bottom of the V near neutral (see
Figure 1). Outside the V-shaped zone, metals tend to be in solution as dissolved

FIGURE 1 Zone of precipitation of Fe(OH)3(s) in contact with hydroxoiron (III)

complexes at 25˚C (3).
Engineered Wetlands for Mining-Impacted Water 453

ions. Figure 1 shows equilibrium concentrations of hydroxo iron(III) complexes

in a solution in contact with freshly precipitated Fe(OH)3(solid) (3). Most metals
show a similar trend. It can be seen, then, that at pH ranges between 2 and 4,
many metals will dissolve and be carried down-gradient is surface water and
infiltration.

4 IMPACTS ON THE ENVIRONMENT

The acidic water leaving the mine site may prove toxic to aquatic flora and fauna
in the receiving streams and lakes. Most aquatic life flourishes at near-neutral pH,
and may die or fail to flourish when pH values drop significantly. A lower pH
may also cause reactions in plant nutrients, changing them to forms in which they
are unavailable to the aquatic floral community (4).
The low pH values additionally can cause large concentrations of metals to
enter the aquatic environment: iron, aluminum, manganese, copper, cadmium,
zinc, nickel, and lead, to name the typical major constituents. In higher concen-
trations, these metals can themselves prove toxic to aquatic life. Additionally,
precipitation of metals (principally iron oxides) in downstream channels and lakes
will affect the aquatic life found on the beds and banks of streams and lakes.

5 TRADITIONAL SOLUTIONS
Traditional solutions can be grouped into two categories: source control and
downstream treatment (5). Source control seeks to prevent or stop ARD produc-
tion at the source—the waste dumps, adits, and tailings.

5.1 Source Control

For dumps and tailings, source control usually consists of isolating the waste
materials from water or oxygen or both, covering them with either soil or water.
This is often successful when waste materials are contained within impoundments
or engineered dumps. When waste rock or tailings are spread over large areas, it
is difficult to locate, excavate, transport, and isolate these ARD-producing wastes.
Adit discharge has proven even more difficult to address at the source.
Conventional approaches include plugging up the adit and attempting to prevent
percolation through the mine workings drained by the adit. Plugging an adit often
results in the creation of ARD springs in the vicinity of the plugged adit, as the
hydraulic pressures inside the mountain create new passages to drain the mine
works. Schemes to prevent percolation through the mine workings include
attempting to grout fissures which carry rainfall percolation down to the mine
works. Large recharge areas and complex geology often foil such attempts (6).
Source control for large pit lakes is impossible using current technology, as
the lake bank surface area is too large to seal, and the inflowing groundwater too
454 Gerbrandt

difficult to control. Backfilling of small pits attempts to limit contact between

groundwater and oxygen, and may or may not prove successful. Approaches for
large pits include institutional controls (fencing, restriction of pumping) if the lake
will not affect the groundwater. If the pit lake water will significantly affect the
groundwater, then conventional technology calls for perpetual pumping with
surface treatment before discharge (7).

5.2 Downstream Treatment

Most downstream treatment approaches for surface flow ARD use neutralization
followed by precipitation of metal compounds. Neutralization may be achieved
by adding lime: limestone (CaCO3), slaked lime [Ca(OH)2], or baked lime (CaO).
Limestone is often used to line channels through which the ARD is directed. It is
inexpensive and dissolves slowly; however, iron precipitation in the form of FeO2
or FeOH often coats the limestone particles, decreasing their effectiveness. Slaked
lime and baked lime are most often added to ARD streams as a slurry (lime mixed
with water). Lime addition to the Clark Fork River near Warm Springs, Montana,
is a good example. The Clark Fork is a potential trout stream heavily impacted
by metal mining in its watershed. Both waste rock and tailings contribute acidity
and dissolved metals to surface and groundwater, which in turn deliver these
pollutants to the Clark Fork. Flow rates average 40 cubic feet per second (cfs) in
spring and 8 cfs during the rest of the year (8). To raise the pH and drop out
metals, a lime slurry is injected into the stream, followed by oxidation using a
small waterfall. The stream flow then passes through a series of settling ponds,
where metal hydroxides and oxides precipitate out. The water leaves the ponds
at a pH 4 range of 7.5–9.0 and with typical metals concentrations several orders
of magnitude lower than those in the upstream water.
While lime precipitation is successful at reducing metals concentrations
down to levels that meet discharge regulations, several factors argue for a better
solution. First, the cost of lime is significant, and increases with the purity and
fineness of the lime. The Warm Springs lime treatment station spent approxi-
mately $89/ton in 1999 for slaked lime, which resulted in an annual cost of
approximately $720,000 (9). Since treatment for surface runoff may continue for
hundreds of years and pit lake treatment for perpetuity, a less costly alternative
with minimal operations is desirable. Second, lime precipitation creates a large
volume of CaOH, CaCO3, and CaO sludges. These sludges have various waters
of hydration attached, and may contain large amounts of water when pumped out
of settling chambers. Additionally, these sludges are unstable and may redissolve
under changing water conditions.
An innovative and still experimental approach for treating mine adit ARD
is the U.S. Bureau of Mine’s (USBM) “In-Mine Treatment of Acidic Drainage
Using Anaerobic Bio-reactors” (10). The USBM built a water treatment bioreac-
Engineered Wetlands for Mining-Impacted Water 455

tor inside an abandoned mine adit and filled the reactor with limestone and
compost. The mine adit discharged around 4 liters per minute with a pH of 3, 500
ppm iron, and 30 ppm aluminum. Mine drainage was directed to the reactor,
where, in an anaerobic environment, sulfur-reducing bacteria (SRB) are encour-
aged to consume organic materials such as sugar and generate hydrogen sulfide
(H2S). The H2S reacts with metal ions to produce metal sulfide precipitates. The
metal sulfide precipitates are retained within the reactor, and have a much smaller
volume and are more stable than hydroxide or oxide precipitates. Alkalinity,
generated by dissolution of the limestone, raises the pH and precipitates alumi-
num as hydroxide [Al(OH)3]. The bioreactor reportedly raised the pH to 6 and
reduced the iron and aluminum concentrations by 50% and 99%, respectively.
The treatment system was located inside the mine and could thus be operated
year-around. The cost of the sugar was expected to be $0.67 per million liters
of water.

6 ENGINEERED WETLANDS TO TREAT

MINE-IMPACTED WATER
The USBM bioreactor described in the previous section, while using natural
mechanisms to precipitate metal sulfides, is still operationally demanding, requir-
ing addition of sugar on a semiweekly basis. Additionally, it is limited to warm
climates or protected areas where freezing will not inhibit the operation. Also, the
discharge treated was approximately a gallon per minute, on the small side of adit
discharges and much less than storm discharges of surface water from mining-
impacted areas. Thus, a natural technology with larger discharge capacity and
reduced operational requirements is desired to address the problem of mining-
impacted waters.
In the past two decades, natural and created wetlands have been used
successfully to clean a variety of contaminants from water, including domestic
sewage and industrial waste streams. It is logical to investigate such wetlands for
removal of metals from water as well. Dominant reactions and treatments
occurring within wetlands are represented by the following five processes (11).
Photosynthesis uses light energy and CO2 to form organic matter and O2
and reduce the partial pressure of CO2. The production of organic matter
allows other zones of the system to become anaerobic. Production of O2
increases the oxidation potential, most probably at microenvironment
sites. The change in the partial pressure of CO2 shifts the water chemis-
try, usually toward higher pH.
Aerobic respiration involves the formation of energy for cell use by the
consumption of organic matter and oxygen. The main benefit of aerobic
respiration is the consumption of oxygen. Through aerobic respiration,
456 Gerbrandt

the deeper portions of the free water surfaces have decreased oxygen
concentrations, as do the treatment walls. The decrease of oxygen allows
other organisms to exist, which use other electron acceptors for anaerobic
respiration.
Anaerobic respiration involves the formation of energy for cell use by the
consumption of organic matter and electron acceptors such as ferric iron,
sulfate, nitrate, and CO2. The main benefits are the reduction of sulfate
and nitrate and the formation of metal sulfide complexes. Metal sulfide
complexes are more stable than metal oxides/hydroxides, which are more
likely to form in aerobic environments.
Settling and filtering involves the removal of suspended solids from the
water. Plantings along the banks and berms of the cells will increase
settling and filtering. It is anticipated that solids that are settled and
filtered out will reside in and help to create anaerobic zones and
transitional zones between aerobic and anaerobic.
Oxidation reactions in zones of higher oxidation potential (near the water
surface) are expected to form metal oxides/hydroxides, which will settle
in the open water system.
These complex natural processes are expected to combine to adjust pH values and
remove excess metals from the influent water.

7 CASE STUDY: ARCO DEMONSTRATION WETLANDS

In 1996, the Atlantic Richfield Corporation (ARCO) began a series of demonstra-
tion projects to treat mining-impacted surface and groundwater in Butte, Mon-
tana. ARCO had been named in 1983 as one of the Potentially Responsible Parties
(PRPs) for the Clark Fork Basin Superfund sites listed in the very first National
Priorities List (12). Water components included surface water running through or
over widespread waste rock and tailings as well as precipitation percolating
through waste rock and tailings and reaching the groundwater. While some adit
discharge may add to the storm water, adit flow is considered insignificant. The
site includes a pit lake with severely degraded water quality. No pit lake water
was treated in the demonstration projects.
Three demonstration-scale wetlands were constructed by ARCO to test the
hypothesis that natural and chemical processes in engineered wetlands could
remove metals from mining-impacted surface and groundwater. The three wet-
lands are identified below, with salient features of each presented.
Wetlands Demonstration Project 1 (WDP1). Seven cells were constructed,
including an initial storage pond, four anaerobic subsurface cells, and
two aerobic surface water cells. The volumes of the four anaerobic cells
measured 30,000 ft3, 20,000 ft3, 12,000 ft3, and 15,000 ft3 for Cells 1–4,
Engineered Wetlands for Mining-Impacted Water 457

respectively, with depths varying from 2.5 to 6 ft. Hydraulic retention

times varied from 3.7 days to 9.4 days at 5 gallons per minute (gpm).
Three different flow paths were possible using the seven cells. The
anaerobic cells’ substrate was a 20% limestone/80% river cobble mix.
Compost was added to three of the four anaerobic cells. Cattails were
planted in the anaerobic cells as a renewable source of organic carbon
(13). All of the cells except the storage pond were lined with HDPE
geomembrane. Inflow water came from the Metro Storm Drain, a small
channel collecting storm water at the foot of the “Richest Hill on Earth.”
Inflow characteristics are presented in Table 1.
Wetlands Demonstration Project 2, Butte Reduction Works (BRW). This
wetland consisted of three open ponds separated by two porous treatment
walls (13), the total surface area measuring slightly less than an acre with
a depth of 4–5 ft. Water flowed sequentially through the five cells. The
treatment walls were constructed of river cobble. The second treatment
wall contained compost as well. None of the cells was lined. This wetland
treated Missoula Gulch surface water, which is also impacted by mine

TABLE 1 Dissolved Metals and Major Anions (14)

LAO
Parameter Units Missoula Gulch groundwater Metro Storm Drain

PH mg/liter 7.51 3.58 7.05

Aluminum mg/liter <0.03 2.52 0.04
Arsenic mg/liter <0.04 <0.04 0.05
Cadmium mg/liter 0.019 0.115 0.007
Calcium mg/liter 84.8 134 214
Copper mg/liter 0.096 14.6 0.067
Iron mg/liter <0.021 0.087 <0.021
Lead mg/liter <0.04 0.39 <0.04
Magnesium mg/liter
Manganese mg/liter 1.38 17.6 13
Phosphorus mg/liter <0.1 <0.1 <0.1
Sodium mg/liter 37.3
Zinc mg/liter 2.22 18 15.1
Chloride mg/liter 35 0.58 69
Nitrate mg/liter 3.34 0.15 0.87
Phosphate mg/liter 0.09 0.11
Sulfate mg/liter 184 594 650
aCombined POTW: POTW water, LAO groundwater, and LMG water.
bCombined w/o POTW: Metro Storm Drain (MSD) water, LAO groundwater, and LMG
water.
458 Gerbrandt

wastes distributed throughout the watershed. Missoula Gulch water

characteristics can be found in Table 1.
Wetlands Demonstration Project 2, Colorado Tailings (CT). The wetland
design was similar to BRW, consisting of three sequential open ponds
separated by two cobble treatment walls. The five cells are unlined, and
are approximately 425 ft long by 185 ft wide, with a depth of around
4.5 ft. The wetlands floor penetrated the groundwater, so the bottom
6–18 in. can be considered groundwater flow, while the top 2–4 ft of
water is surface flow. CT included lime addition at the inlet to raise the
pH, remove excess metals, and pretreat the water. Influent water was
pumped from a collection trench intercepting groundwater, known as
Lower Area One (LAO) groundwater. This water is the most degraded
of the three. Characteristics are presented in Table 1.

7.1 Summary of Performance on Each Wetland

It was hoped that the ARCO wetlands would precipitate metal sulfides via
anaerobic respiration. Theoretically, in an anaerobic environment, sulfur-reducing
bacteria (SRB) generate hydrogen sulfide (H2S). The H2S reacts with metal ions
to produce metal sulfide precipitates. As stated previously, metal sulfide precipi-
tates are more stable and of a much smaller volume than metal oxides or
hydroxides. Anaerobic zones are produced in the subsurface cells in WDP1 and
in the treatment walls of BRW and CT, as well as at the bottom of surface cells
where sediments and decaying organic matter form an anaerobic layer. However,
any of the other mechanisms—photosynthesis, oxidation, aerobic respiration,
settling, and filtration—may play the major role as well. The metals of interest
for the ARCO wetlands are copper, manganese, and zinc. Other metals are present
at concentrations low enough to meet anticipated discharge standards. A summary
of the performance of each of the wetlands is presented below.

7.2 WDP1 Performance

Cell 3, the only upflow cell of the four, failed hydraulically within the first three
months. The remaining three horizontal flow cells were monitored continuously
from the spring of 1996 until December of 1998. Contaminants of major concern
were copper, cadmium, and zinc. Manganese may also be a problematic constit-
uent, depending on site-specific cleanup levels yet to be determined. Table 2
presents typical summer and winter performance data from Cell 2. It can be
seen that copper and cadmium concentrations were reduced more than 97%
on both sampling days. Zinc concentrations were reduced by more than 99% on
the summer day, but only 94% reduction was achieved for unfiltered Zn on the
winter sampling date. The decrease in Zn removal efficiency is probably due to
both a reduction in hydraulic retention time because of freezing of the upper
Engineered Wetlands for Mining-Impacted Water 459

TABLE 2 Concentration of Selected Parameters in Cell 2 in Summer

Versus Winter

Summer (8/17/98) Winter (2/19/98

Percent Percent
Influent Effluent removed Influent Effluent removed

T (˚C) 19.0 19.1 2.0 2.7

PH 6.80 6.84 6.42 6.73
SC (µmho/cm) 872 1070 821 897
Alkalinity 66.3 178 72 149
DIC (mg ˚C/liter) 16.1 54.4 18.3 36.4
Eh (mV) 476 –153 493 76
As 8.6 11.9 (–38%) 7.3 5.2 28.8%
Cd 40.5 0.51 98.7% 32.7 0.52 98.4%
Cu 123 3.6 97.0% 216 1.0 99.5%
Fe 28.8 18.0 37.5% 171 528 (–209%)
Mn 6680 3640 45.5% 7070 8080 (–14%)
Zn (filtered) 9330 10.5 99.9% 9950 101 99.0%
Zn (unfiltered) 9850 42.9 99.6% 10,000 640 93.6%
Sulfate
(mg SO4/liter) 472 289 38.8% 336 313 6.8%
H2S
(mg H2S/liter) 0.0 34.1 0.0 2.0

Heavy metals = filtered (0.45 µm), µg/liter; alkalinity = mg CaCo3/liter.

Source: From Seasonal Influences on Heavy Metal Metal Attenutation in an Anaerobic
Treatment Wetland, Butte, Montana, 1999.

portion of the wetland and a decrease in the activity of SRBs during cold
temperatures (15).
It should be kept in mind that Cell 2 has a volume of 20,000 ft3 and a flow
rate of approximately 5 gpm. It was thought that, upon scaling up, this volume-
to-flow rate ratio would require excessive land surface for treating the required
flow rate for the Superfund site. Early results of the WDP1 performance led
ARCO to modify its wetlands design for the second and third demonstration
projects. The projected large volume and surface area requirements based on
WDP1 performance resulted in large costs of substrate for subsurface cells and
geosynthetic lining for all cells. Therefore, the next generation of demonstration
wetlands utilized a greater surface water volume to subsurface volume ratio, and
subsequent ponds were unlined. Additionally, piping and enclosed flow meters
were eliminated wherever possible, as these utilities were shown to be prone to
freezing and clogging in WDP1. Ditches and flumes were utilized in subsequent
installations.
460 Gerbrandt

7.3 BRW Performance

BRW wetland consisted of three open ponds separated by two porous treatment
walls, with flow passing sequentially through the five cells. It was hoped that the
cobble treatment walls would develop an anaerobic treatment zone where SRBs
would generate H2S and metal sulfides would precipitate. Additional treatment
mechanisms were described previously.
The following trends were observed during the first two years of operation
of the BRW wetland (13,16).
Sulfate did not show any removal trends.
Arsenic was not removed, but concentrations are below anticipated cleanup
standards.
Cadmium concentrations were reduced by about half but remained above
anticipated cleanup standards.
Filtered and unfiltered copper concentrations were reduced significantly.
Unfiltered concentration reductions varied from 39% (2/27/97) to 99%
(6/19/97). Filtered concentrations varied from 33% to 93%.
Zinc was removed significantly, but not enough to meet treatment objec-
tives. Most of the zinc was contained in the dissolved phase.

7.4 CT Performance
The CT wetland achieved around 99% removal of copper, iron, manganese, and
zinc (17). Of this removal, 96% to 99% occurred in the first surface water
treatment cell. Recall that the CT system includes injection of lime at the inlet.
It is likely that pH adjustment led to metal hydroxide, oxide, and carbonate
precipitation in the first cell. Mulholland (16) concluded: “Even though the results
obtained by the wetlands are primarily due to pH precipitation, there still may be
some advantages to this treatment technology over a traditional lime treatment
operation.” These advantages include:
Compaction of precipitates over time, resulting in a lower volume of sludge
than obtained in a settling tank.
The longer residence time allows formation of kinetically slower, more
stable minerals such as manganese oxides.
Reducing environments may eventually transform precipitates into denser,
more stable metal sulfides.

8 CONCLUSIONS
Acid mine drainage from mining of sulfur-bearing ores seriously impacts surface
and groundwater quality. While waste minimization and pollution prevention can
prevent generation of AMD in current operations, centuries of historic mining has
Engineered Wetlands for Mining-Impacted Water 461

resulted in acid mine drainage from thousands of mine sites across the United
States. The control of these pollution sources is mostly limited to treatment. This
chapter has explained the production of AMD, summarized common solutions to
the problem, and presented engineered wetlands as a possible new control of
this very serious pollutant. A demonstration-scale wetlands was presented to
demonstrate treatability of AMD.
After two years of operation, the three demonstration wetlands in Butte,
Montana, are successful to some degree in removing metals from mining-
impacted surface and groundwater. While the subsurface cells performed well in
reducing metals concentrations, they require a large surface area and are prone to
flow rate reduction with time. The BRW wetland used surface water cells with
cobble treatment walls between the surface cells. Significant reductions were
achieved, but the treatment objective could not be reached. The CT wetlands were
similar in design to the BRW setup, but lime addition at the inlet to the wetlands
allowed this facility to meet treatment goals. Lime addition, however, was at a
much smaller dosage than conventional lime treatment for metals removal.
The Butte climate is extreme, with a long and cold winter. Piping, valving,
and meters were prone to freezing, so recommendations are to use open channels
to deliver water and flumes or weirs to measure flow rate. Wetlands must be
scaled to account for freezing in winter. Freezing of the top 12 in. of the treatment
cells will obviously reduce the flow area and decrease hydraulic retention time.
Also, bacterial and plant communities become much less active during the cold
season, and storage of contaminated water may be required during the winter until
natural degradation processes are once again active in the spring.
Precipitation of contaminants such as metal sulfides is desirable, and
theoretically achievable using natural processes in engineered wetlands. At two
years of operation, the ARCO wetlands were not mature enough to document
precipitation of metal sulfides.
Certainly, the ARCO wetlands showed promise for removing metals from
water. The extent of contaminant reduction and long-term effectiveness must wait
to be proven until the wetlands mature. Cold water temperatures (in the winter
season) obviously reduce the effectiveness of the wetlands. At a minimum,
engineered wetlands can serve as a passive polishing step. With the addition of a
small amount of lime to the influent, the wetlands may be able, with time, to meet
treatment goals.

ACKNOWLEDGMENTS
The information concerning passive engineered wetlands presented in this chapter
would not have been generated without the support and guidance of the Atlantic
Richfield Company and their project manager, Dr. John Pantano.
462 Gerbrandt

REFERENCES
1. G. A. Mealey, Grasberg, Freeport-McMoRan Copper & Gold Inc., 1996.
2. Berkeley Pit Public Education Committee, Pitwatch, Vol. 2, no. 2, Butte-Silver Bow
Planning Department, Butte, MT.
3. V. L. Snoeyink and D. Jenkins, Water Chemistry. New York: Wiley, 1980.
4. F. F. Munshower, The Practical Handbook of Disturbed Land Revegetation. Lewis
Publishers, 1994.
5. Educational Communications, Inc., for U.S. Environmental Protection Agency, Sur-
face Mining and the Natural Environment, 1985.
6. L. McCloskey, D. B. Kelly, and J. Gilbert, Source Control—Surface Waste Pile
Demonstration Project, 2000 Billings Land Reclamation Symp., March, 2000.
7. ARCO, Environmental Action Plan for the Clark Fork River Basin, Spring 1998.
8. F. D. Searle, Evaluation of Factors Affecting Heavy Metals in Discharge from Warn
Springs Ponds, Master’s thesis, Montana College of Mineral Science and Technology,
Butte, MT, May 1981.
9. B. Duff, Atlantic Ridhfield Company, personal correspondence, June 2000.
10. Bureau of Mines, U.S. Department of the Interior, In-Mine Treatment of Acidic
Drainage Using Anaerobic Bioreactors. Technol. News, no. 444, December 1994.
11. ARCO, Wetlands Demonstration Project 2 Butte Reduction Works Final Design
Report, December 1996.
12. Montana Department of Health and Environmental Sciences and the U.S. Environ-
mental Protection Agency, Progress—Clark Fork Basin Superfund Sites, May 1990.
13. A. Frandsen, Review of ARCO Wetlands Demonstration Projects: Most Significant
Findings through January 1999, unpublished, April 1999.
14. ARCO, Draft Silver Bow Creek/Butte Area NPL Site Butte Priority Soils Operable
Unit, Wetlands Demonstration Project 2 Butte Reduction Works Preliminary Design
Report, Anaconda, MT, August 1996.
15. C. H. Gammons, W. J. Drury, and Y. Li, Seasonal Influences on Heavy Metal
Attenuation in an Anaerobic Treatment Wetland, Butte, Montana, unpublished, 1999.
16. Montana Tech, Wetlands Demonstration Project Progress, September 1997.
17. T. P. Mulholland, A Study of the Mass Deposited, Removal Efficiency, and Types of
Minerals Formed in Colorado Tailings Constructed Wetlands, Master’s thesis, Mon-
tana Tech of the University of Montana, May 1999.
22
Fuel Blends and Alkali Diagnostics:
European Case Study

Ingo F. W. Romey
University of Essen, Essen, Germany

1 INTRODUCTION
The main goal of the project, “Advanced Combustion and Gasification of Fuel
Blends and Diagnostics of Alkali and Heavy Metal Release,” was the investiga-
tion and development of data and technologies for commercial use of waste in
industrial applications. The R&D program has been set up in three main clusters
featuring textile waste co-processing, plastic waste co-processing, and alkali and
heavy metal diagnostics.
Ten main items have been investigated, covering the following topics and
working packages: Determination and characterization of the supplied production
wastes with regard to physical and chemical characteristics
Investigations on special preparation methods for ecologically and econom-
ically effective waste preparation (for instance, sorting, separation, shred-
ding, grinding)
Upgrading and blending with coal or biomass and variation of coal/waste
biomass/waste ratios
Supply of representative fuel blends from waste and coal to the partners
performing the combustion/gasification tests
463
464 Romey

Determination of logistic and economic principles of the fuel blend prepa-

ration (transport, storage, location for the preparation facilities), environ-
mental aspects
Combustion and gasification of fuel blends
Residue valorization concepts
Alkali sensor market study
Alkali diagnostics—measuring campaigns and evaluation
Techno-economic assessment studies
Environmental aspects of co-combustion

A very important part of the work deals with the preparation of fuel blends,
transport of the materials, and development of feeding systems. In addition to the
combustion and gasification behavior of the materials, special attention was paid
to the composition of the fuel and flue gases and the utilization of the ashes.
Especially for the later commercial use of the fuel blends in advanced combustion
and gasification technologies, it was necessary to develop on-line measurements
for alkali and heavy-metal diagnostics. Three different systems have been devel-
oped and investigated in the project to guarantee an immediate analysis of alkalis
and heavy metals in the flue gas stream and to ensure 100% protection of gas
turbines in combined-cycle technologies as well as protection of the environment
in advanced combustion and incineration technologies. The R&D program cov-
ered basic research work at the laboratory scale as well as large-scale tests in
commercial units to guarantee quick transmission of the results for later commer-
cial use.
Due to the large amount of waste plastics in Europe, thermal utilization
could be an attractive option in a number of industrial applications; however,
standardization and equalization of national laws and reduction of preparation
costs must be considered and solutions must be developed.

2 WORKING PACKAGES
In the project, 20 partners have worked together in a very cooperative way to
reach the main goals of the planned 10 working packages.

2.1 Upgrading and Preparation of Waste (Textiles and

Plastics) for Further Utilization with Coal in
Combustion and Gasification
2.1.1 Sächsisches Textilforschungsinstitut (STFI)
The aim of the research work of STFI in the overall project was the transfer of
production waste (not suitable for material recycling) by means of mechanical
Fuel Blends and Alkali Diagnostics 465

preparation methods into such a form that it can be combusted after blending with
coal in combustion plants.
The following textile waste with different material composition has been
investigated:

Cotton, viscose, viscose-acetate, polyester, polyamide 6

and regarding their textile materials structure,

Dust, fluff, fabric selvages, heavy screen fabrics

The following methods of mechanical preparation and thermal treatment for

selected textile waste have been investigated:

Grinding process, with knife mills

Cutting process, with cutting machines
Breaking process, with tearing machines
Agglomeration process, with a plast-agglomerator-system

Compaction/briquetting tests of textile waste from used drying screens from

the paper industry (PA) and a mixture of fabric selvages consisting of 50%
polyester and 50% acetate were carried out, followed by combustion tests of the
briquettes and emission measurements of combustion gases as well as chemical
analysis of ashes. The permitted emission values can be kept for further optimi-
zation of the combustion conditions. The obtained ashes cannot be used further
and must be landfilled.
Furthermore, agglomeration of synthetic textile waste has been investi-
gated. The waste samples contained used wipes for machines, fabric selvages,
and reclaimed fibers from used carpets. The materials were treated by a cutting
process and than reprocessed by a plast-agglomerator system. The wipes, consist-
ing of 100% cotton, were blended with cut film waste material of polyethylene.
The results in bulk density obtained are between 210 and 350 g/liter.
Knife mills have to be tested where very small (dustlike) particles can be
produced. Blending and dosage of small particles requires special constructive
changes of transport and dosage systems at combustion plants suitable for textile
matters.
The different combustion or gasification technologies are connected to
special feeding systems. For these feeding systems a defined degree of treatment
of textile waste is necessary

In principle, a higher degree of treatment leads to higher costs. However,

the heat value, depending on the materials, the composition of the waste
and the theoretical costs for landfilling have to be considered.
466 Romey

Blends of textile waste with coal are very difficult to obtain, because of the
great differences in the materials density and the bulk density.
Orientating briquetting trials have shown that a very large surface of the
textile waste is necessary. The reduction of particle size from 30 mm to
10 mm is connected with a decrease of the throughput of the cutting
machine of about 60%.
The obtained densities for textile briquettes without addition of coal is
about 0.7 g/cm3. A reduction of storage and transport volume of 1/7 can
be achieved if there are no high requirements regarding the strength of
briquettes. Briquetting with lignite requires a share of coal of about 60%.
The densities are > 0.9 g/cm3. Problems of blendability occurred in the
trials. Segregation during the feeding of the compression mold was
observed. This led to reduced briquette quality.
The investigated industrial waste textiles from the paper industry (H1-10
and H1-30) are applicable in combustion plants under practical condi-
tions. The combustion should be a combined process because of the
problems arising during monocombustion (melting, coke-like sediments).
The waste textiles can be used as textile cuttings blended with a second
fuel as well as like briquettes with coal. During a combustion trial with
blended briquettes (70% coal/30% waste textiles) in a slow-burning stove
(nominal capacity 5.6 kW), it was shown that blended briquettes have a
stable roasting residue, i.e., the briquettes do not break down and form a
coarse ember bed with uniform roasting residues. The problems that
occurred in monocombustion of these textiles could not be observed (1).
The measured emissions are under control with regard to legal conditions,
after-treatment of the gases, and optimization of the combustion conditions.
2.1.2 Thyssen Umwelttechnik Beratung GmbH (TUB)
TUB has compared different methods of separation of waste plastics from used
cars (shredder light fraction, or SLF). Based on analysis of the results, options
have been determined for the adjustment of a plant configuration for the given
application.
The licensing procedure with the authority, Bezirksregierung Düsseldorf,
led to a license according to Bundes-Immissionsschutzgesetz (Federal Immission
Protection Act) being granted for the erection and operation of such a trial plant.
In several discussions with the regulatory authority it was possible to lower the
requirements thus making the process viable.
As the shredder light-weight fraction contains a large amount of plastics
and as plastics producers have a joint responsibility for this share, the Verband
Kunststofferzeugende Industrie e.V. (VKE, Association of plastic producing
industry) carried out a study on thermal and material recovery. This report states
the procedure TUB presented to the VKE to be the most appropriate system. It is
Fuel Blends and Alkali Diagnostics 467

the only system offering the most suitable option for utilization of the output
material.
Under licensing aspects, there is no problem in using the organic fraction
via Thyssen Schachtbau by adding coal and by palletizing, since the Thyssen blast
furnace operation has a trial licence from the authorities; however, during the
running time of the project it was not possible to reach final agreement with
Thyssen Steel.
TUB also got in touch with the Belgian and French cement industry to
discuss an alternative use for the plastic fraction, but, the approach has not led to
a final result to date.
2.1.3 Thyssen Schachtbau Kohletechnik GmbH
TSK has mainly investigated ways for thermal utilization of waste plastics from
cars (SLF). Due to its high content of inerts, ranging from 20% to 50% (dust,
metals, glass, etc.), preparation has been carried out to recover the organic
components.
The following aspects have been investigated:
Analysis of SLF and suitable coal components. Besides the immediate
analyses of the accompanying components chlorine and metals such as
Cu, Zn, and Pb, investigation with regard to identifying the fractions of
the SLF where these components are accumulated was carried out.
Methods of blending and conditioning with hard coals, including evaluation
of possible industrial methods in terms of technical and commercial
aspects in order to blend and condition such diverse materials to get a
homogenous, pneumatic-conveyable fuel. Industrial tests and investiga-
tions showed that one possible way to get a homogenous solid fuel to
blend the SLF with dried coal in a conveyor screw mixer, followed by
pelletizing in a pellet press.
Transport/pneumatic conveyance of pellets obtained from the tests has been
done successfully under industrial conditions.
Combustion behavior of the coal-enriched shredder fraction (shredder
carbon pellets, SCP) has been tested in a small combustion test rig at
DMT (fluidized bed furnace).
In general, due to its composition, the industrial kilns favored for the
thermal utilization of SCP are
Blast furnaces
Cement rotary kilns
Fluidized bed combustors
The work programs by Thyssen Sonnenberg Umweltberatung and Thyssen
Schachtbau Kohletechnik have provided the basis for the design of a SLF
468 Romey

preparation plant. A material balance with the average composition of SLF

differentiated by material groups is shown in Table 1.
It is obvious that this inhomogeneous shredder composition “run of plant”
results in an undefined calorific value which can range from 10–22 GJ/t, accom-
panied by undesired components for industrial kilns such as metals, glass, etc.
Therefore, a serial investigation was carried out regarding the immediate analysis
of SLF at different sizes (<10 mm, <5 mm, >10 mm).
Due to the high inert content in the fraction <10 mm, a future prepara-
tion plant will operate with a prescreening to <10 mm followed by a further
preparation of the materials >10 mm. The second step to recover an acceptable
material for thermal conditioning is to separate metals and other inorganic
compounds. The program work done on this matter was carried out success-
fully in more detail by project partner Thyssen Umwelttechnik Beratung
GmbH. The materials obtained from the preparation tests showed calorific values
ranging from 16 to 22,000 kJ/kg, but with a typical consistency unusable for
pneumatic transport.
In order to stabilize the calorific value and to decrease the metal and
chlorine contents, coal has to be added to the SLF; therefore, technical operations
such as combined grinding, briquetting, etc., have been evaluated in technical and
commercial terms.
Tests to pelletize SLF showed that a significant improvement in pellet
strength can be achieved by adding 20% pulverized coal. Photographs taken at
millimeter scale proved the effect that the plastic compounds starts softening and
thus binding the pulverized coal.
A semiindustrial test run on a 300-kg/h pellet press confirmed these results,
and approximately 1 t of pellets were obtained for Thyssen Stahl AG to investi-
gate the mechanical properties as well as pneumatic conveyance.

TABLE 1 Composition of SLF by

Material Groups (Sommer, D.,
Grünenberg, H., Gotthelf, H.)

Material Weight percent

Plastics 30–35
Elastomers, tyres 20–30
Glass 10–16
Textiles 3–5
Woods/grains 3–5
Lacs 3–5
Metals 0.5–4
Inerts (sand/dust) 10–20
Fuel Blends and Alkali Diagnostics 469

The pellets showed good performance on both criteria so that plant design
could be completed to a 5-t/h pellet output capacity.
Table 2 presents the expected fuel qualities. The combustion tests showed
no particular differences when firing pure SLF pellets or shredder carbon pellets
(SCP). In both cases emissions were kept to the same level, which at least proved
a homogenous fuel suitable for fluidized bed combustion.
A suitable way to recover fuel from raw shredder light fraction for industrial
kilns such as rotary kilns or blast furnaces is to separate the inerts, e.g., metals,
and to pelletize the residual organic material in a pellet press after blending with
pulverized coal.
The SCP are pneumatic conveyable and have a calorific value of at least
22,000 kJ/kg with ash contents below 20%. The chlorine content ranks below 1%,
which allows combustion in cement kilns. The metal content, in terms of the
critical parameters for steel works, Pb, Cu, and Zn, ranks below 1%, which allows
blast furnace operations.
Firing in industrial fluidized bed furnaces seems to work without problems
regarding emissions as well as feeding and dosing behavior.
2.1.4 Fechner GmbH & Co. KG
Fechner, in cooperation with Krupp Hoesch Stahl (now Thyssen Krupp Stahl),
jointly investigated a concept for blast furnace co-injection of pulverized mixed
plastics from postconsumer packaging materials and pulverized coal as a ready-
to-inject fuel blend. The idea behind this fuel blends project was to assess the
technical and commercial viability of a low-investment and low-development-

TABLE 2 Expected Fuel Qualities

Component/blend

Prepared Pulverized Pellet Pellet

Specification SLF coal 70/30 80/20

Total moisture (wt%) 2 1 1 1

Ash (d.b. wt%) 23 7 18 20
Volatile matter (d.b. wt%) 60 25 50 53
Sulfur (d.b. wt%) 1,0 0,7 0.9 0.9
NCV (a.r. cal/g) 4,600 7,500 5,500 5,150
NCV (J/g) 19,200 31,400 23,000 21,700
Chlorine (d.b. wt%) 1 0 0.7 0.8
Pb (ppm) 3,000 50 2,100 2,400
Cu (ppm) 5,000 35 3,500 4,000
Zn (ppm) 9,000 6,100 6,000 7,000
470 Romey

risk concept for mixed plastics injection by utilizing existing PCI installations and
operational experience at Krupp Hoesch steelworks.
Final evaluation of the operational and commercial viability of this fuel
blends concept in comparison to the alternate concept of separate injection of
granular plastics or mixed plastics pellets, which is also under investigation by
Thyssen Krupp Stahl, was one of the major outcomes of the project part.
The technical program investigated in the project dealt basically with the
pulverization of thermally agglomerated mixed plastics with a sizing between 0
and 10 mm to be milled down to a particle size of less than 2 mm.
Following a critical assessment of candidate impact pulverizer systems, an
air flow rotor impact mill (turbo-rotor mill) was finally selected for pulverizing
mixed plastics agglomerates. The mill design throughput was 5 t/h of thermally
agglomerated plastics from packaging waste, although the output capacity varied
within the limits of 3–4 t/h of pulverized product.
The subsequent manufacturing of fuel blends with variable mixture ratios
of pulverized plastics and coal dust in the range between 10 and 30 wt% of
plastics was performed in the production, handling, and storage facilities at
Fechner’s Lünen works; and the ready-to-inject fuel blends were finally shipped
by silo trucks to the blast furnace site at Dortmund.
The major technical objectives achieved were as follows.
Pulverization of thermally agglomerated mixed plastics, including evalua-
tion of different impact pulveriser systems, semitechnical-scale grinding
trials with potential equipment suppliers, and selection and installation
of a specific impact pulverizing mill for large-scale production of pulver-
ized plastics. Laboratory scale investigations of the pneumatic conveying
properties of coal dust and pulverized plastics mixtures with respect to
the maximum/optimum plastics ratio in the fuel blends were carried out.
Pulverized fuel blends manufacture, including evaluation of conceivable
problems regarding the large-scale manufacturing of almost homogenous
fuel blends; assessment of the dosing, mixing, and storage equipment
requirements. Development of a low-cost approach using Fechner’s
available handling and mixing installations.
Sufficient fuel blending production trials with sieve analyzing and shearing
resistance measurements in order to optimize the mixture ratios and the highest
possible uniformity of the fuel blends were performed.
Blast furnace injection trials were carried out jointly with Krupp Hoesch;
production of fuel blends for long-term blast furnace injection runs (up to the end
of 1996 a total of 2500 tons of mixed plastics was injected) were carried out.
Based on the above technical assessments and operational experience
gained from the mixed plastics grinding, fuel blends manufacturing, and blast
furnace injection runs, the technical viability of the fuel blends concept has been
Fuel Blends and Alkali Diagnostics 471

demonstrated. However, a straightforward evaluation of the commercial viability

of this concept for a normal routine blast furnace operation in comparison to the
separate and direct injection of plastics agglomerates or pellets (mono-injection
concept) needs still more large-scale investigation with the two different mixed
plastics injection modes which have been investigated.

2.2 Co-combustion Behavior of Wastes (Textiles and

Plastics) and Coal in Entrained Flow and Fluidized Bed
Combustion
2.2.1 Instituto Superior Técnico (IST)
The work undertaken at IST dealt with the investigation of co-combustion of
natural gas and textile waste entrained in a combustion air stream. The IST
contribution can be divided into three main fields of research and action:
Design and construction of a textile waste feeding system and gas/textile
waste burner
Simple gas combustion experiments
Mixed gas and textile wastes combustion experiments
Design and construction of a feeding system to continuously feed textile residues
in a gas stream, for co-combustion with gas, was developed and tested.
The textile waste feeding system consists of a feed hopper that delivers
pulverized textile waste into a vibrating chute and then to a balanced injector
system in the primary air line, from where it is pneumatically delivered to the
burner through a high-quality, metal-shielded weighbridge, so that the solids in
the feed hopper can be weighed. The loss in weight is measured every few
microseconds, this information being averaged and stored as a weight loss rate in
a computer.
The design and construction of a burner for co-combustion of textile wastes
with gas was successful.
IST has undertaken several flue gas measurements for co-firing two kinds
of agriculture waste (pine shell and peach stone) and two types of textile waste
with propane. The influence of thermal ratio (waste/propane) in the flue gas
composition was sought. Detailed in-flame measurements for major local mean
gas species (O2, CO, CO2, unburned hydrocarbons, and NOx) and local mean gas
temperatures for combined flames of gas + textile waste and gas + biomass were
collected. Detailed flue gas measurements for different air staging configurations
were also conducted.
Overall, the complete set of results allowed drawing the following conclusions:
1. NOx emissions increase with the waste/propane thermal ratio regardless
of the type of waste. The increase is remarkable in the case of one type
472 Romey

of textile waste owing to its high nitrogen content (fuel NO) and in the
case of peach stone, probably because of its smaller particle size
distribution (thermal NO).
2. In general, as the waste/propane thermal ratio increases, the CO and
UHC emissions increase, particularly for the propane + biomass
flames.
3. Attempts to co-fire pulverized coal with textiles were impossible due
to their different physical proprieties.
4. NOx emissions from propane + textiles and propane + biomass flames
can be effectively controlled using air staging.
2.2.2 University of Bochum (LEAT)
LEAT investigated the combustion behavior of coal/textile blends in a CFBC.
The first step was the construction of a new heat exchanger for the CFBC
test facility as a replacement of the old one, which had a low efficiency. For the
collection of fly-ash samples needed for the development of valorization concepts
at DMT-SysTec, an additional filter was built. The new devices were tested
successfully.
The first combustion trials using blends from coal and shear dust failed due
to the fact that the existing metering system was not able to feed the textile fibers
continuously. The mixtures separated in the hopper, and also bridging occurred.
So the most important task became the development of a reliable feeding system
for the different kinds of textile waste, all of them having very specific physical
properties. LEAT planned to integrate a separate metering system for textile waste
into the test facility. In cooperation with the company Emde, a metering system
employing a unique stirring device was developed. This system can feed most of
the textile waste considered. The greatest advantage of two separate metering
systems is that fluctuations of the textile mass flow have only a small effect on
the operation of the facility. The reason is that the ratio of energy input of the
textile waste is low in comparison to that of the coal. The dosing of coal using
an independent metering system is very precise. In addition, fuel preparation is
not necessary. Problems with the metering of textile waste do not lead to
shutdown of the facility, since the energy input of the textiles can be easily
compensated by additional coal. In view of the later planned industrial im-
plementation of co-combustion of wastes, this is relevant.
Different kinds of textile wastes have been investigated, and the operating
parameters of the CFBC in order to optimize the co-combustion of textile wastes
have been improved.
The optimization of the burning process inside the CFBC is influenced by
primary measures. Possible parameters such as air staging through different
injection ports and variation of average combustion temperature were examined.
In these series of investigations, shear dust I, consisting of 50% polyacrylics next
Fuel Blends and Alkali Diagnostics 473

to 50% cotton, shear dust II, consisting of 100% polyacrylic agglomerated textile
waste TA, which is production waste of the car industry, and TD consisting of
100% polypropylene were applied. The major difference between those textile
wastes was the nitrogen content (shear dust I, N = 14%; shear dust II, N = 24.34%;
TA, N = 1%; and TD, N = 0%). This content establishes the formation of N2O
and NOx.
The pollutant formation of fuel blends from coal and textile waste in a
CFBC is influenced mainly by two factors. On one hand, the physical properties
of the textiles lead to difficulties running the combustion process itself. The high
CO and CxHy emissions are caused by the physical properties, especially the low
bulk density of the textiles, because they lead to fast transportation through the
combustion chamber without giving enough residence time for burning. On the
other hand, the high nitrogen content of the synthetic fibers results in increased
nitric pollution. The reduction of NOx and N2O is possible due to primary
measures such as air staging, and variation of the air ratio and the combustion
temperature. The dosability of textile blends and scraps are disadvantageous but
excellent for agglomerates. The major disadvantage of burning the agglomerates
is that the emissions lead to high CO and CxHy-emissions due to their physical
properties.
Summarizing, the co-combustion of textile waste inside a CFBC is a
possible but not favorable solution. If it is intended to employ this type of
co-combustion, the textile fraction of the fuel ought to be minimal. For the future,
more experiments applying cotton are needed. As soon as the CO problems has
been overcome, the combustion of fuel blends from coal and cotton in a CFBC
will be a good solution.
In a second field, LEAT investigated the use of textiles with a high nitrogen
content as a substitute for NH3 in the SNCR process. This textile waste consists
of a milled agglomerated shear dust with a particle size smaller than 100 µm.
These investigations were performed in a drop tube furnace with an electrical
capacity of 50 kW. For simulation of the flue gas in a coal-fired power plant, dried
flue gas of a natural gas burner with a fixed fraction of NO was used. The addition
of the textiles was intended to reduce the NOx emission.
Prepared textile dust with a high nitrogen content is an excellent substitute
for NH3 for the SNCR process. With the addition of NH3, the NOx emissions may
be reduced to about 33%. The experimental results using textile waste as a
substitute for NH3 showed a reduction of NOx up to 85%. The main conclusions
are that the molar fraction of the fuel bound nitrogen from the additive, the gas
temperature, and the residence within the furnace are the major parameters for
NOx reduction in the SNCR process. Further investigations are necessary in the
drop tube furnace for exact predictions of the SNCR process. Variation of
operating parameters such as gas temperature, molar ratio of the additive, and
474 Romey

residence time within the furnace are required, accompanied by further experi-
ments in technical-scale plants.
2.2.3 KEMA
KEMA has worked on reactivity measurements as selection criteria for combus-
tion of mixed plastic waste. The activity consisted of several research steps, such
as sampling, analysis, prediction of quality of waste products, prediction of
emissions, and drop tube furnace experiments.
The topics on sampling/analysis were evaluated successfully, and tests
dealing with quality of waste products, emission measurements, and drop tube
furnace experiments have been carried out.
2.2.4 International Flame Research Foundation (IFRF)
The primary objective of the IFRF work has been the investigation at a semi-
technical scale of firing solid fuels through the tuyeres of a blast furnace, as a
means of reducing the coke content while maintaining and/or improving the
reduction of iron oxide.
The program examined the possibilities of firing a blend of European
coal with plastics in simulated blast air. The following parameters have been
investigated:
Blending ratio
Fuel heat input
Fuel velocity
Flame length and flame boundary conditions
Flame penetration based on the kinetic energy of the blast
Burnout
The emphasis has been toward the testing of systems for the feeding of
granulated and powder plastics. The results from these tests have defined the
particle size that can be fed with the coal to make specific blends.
It is desirable for the blast furnace simulation that the particles are as small
as possible, in order to expose a large surface area which should ensure complete
burnout.
Plastics have been identified as possible fuels that can be used with coal of
various rank to form a blended fuel that is suitable for injection into the blast
furnace. Typical composition of plastics relative to other common fuels is given
in Table 3.
A detailed evaluation of the combustion behavior of plastic blended with
coal with different injection methods has been carried out. Furthermore, ash
samples have been taken for ash valorization with other project partners.
Finally, the potential for plastic fuel as a blend fuel with coal for replace-
ment of coke in the blast furnace has been ascertained.
Fuel Blends and Alkali Diagnostics 475

TABLE 3 Typical Composition of Plastics Relative to Other Common Fuels

Plastic Delayed Municipal
Element (% w/w) scrap coke HVB/MVB coal sewage sludge

Ash 1–2 <2 5–15 40–50

Water 0–15 <2 <3 <10
Volatiles >95 10–13 25–40 50
C 60–80 92 70–78 30
H 6–14 3 4–5 4
S 0 1–3 1–2 1–2
N 0 0.5–1.5 1–2 3–5
Cl 0.1–1.5 <0.1 <0.2 <0.1
F 0–0.5 <0.1 — <0.1
LCV MJ/kg (dry) 27–40 31–33 28–31 11–14

2.2.5 ICI Films

ICI, as one of the industrial project partners, took over the large-scale combustion
of fuel blends consisting of coal and plastics film wastes in a 40-MW circulating
fluidized bed boiler (CFB). Due to serious damage in the boiler, however, ICI
decided to leave the project.
However, first trials in 1996 showed that the system after further optimiza-
tion will work sufficiently. ICI has plastic waste samples available in a 100-t scale
and was able to deliver the material for further combustion tests to the other
project partners.
IST together with industrial partners in Portugal has taken over a project
part and has carried out combustion tests in large-scale boilers.
In addition, IST has conducted detailed flue gas measurements of O2, CO,
CO2, unburned hydrocarbons, and NOx at the Portucel full-scale biomass boiler.
During the tests around 40 t of plastics of four different types were burned. The
plastics were burned in combination with biomass at different thermal ratios.
From the data collected, it was possible to draw the following conclusions.

1. There is a value of plastics/biomass ratio that corresponds to a maxi-

mum value of NOx emissions; above and below this value, NOx
emissions decrease. Consistently, this is the value to which corresponds
a higher value of the mean boiler temperature.
2. SOx emissions from the combined combustion of biomass + plastics are
always insignificant. However, they increase marginally as the plas-
tics thermal input increases because of the sulfur present in one type
of plastic.
476 Romey

3. CO and UHC emissions were not detected in the flue gases for any of
the tested conditions.
4. During extended and continuous boiler operation, firing small quanti-
ties of biomass plus high amounts of plastics, there was a propensity
for the latter to melt on the grate, with the combustion efficiency
suffering accordingly.

2.3 Co-gasification Behavior of Wastes (Textiles and

Plastics) and Coal in Fluidized Bed Gasification
2.3.1 University of Lund
TU Lund has carried out work on gasification experiments of fuel blends of coal
and textile and plastic wastes. The experience showed a need for development in
mixing procedures for fuel blends and also a necessity for modifications to the
existing feeding system.
Several successful and stable gasification experiments were done with coal
and textile waste blends. The main goals in these experiments were
To study the effect of the various mixing methods on feeding process
To see whether different mixtures behave differently in the gasifier
To study the reliability of the gasifier and reproducibility of the gasification
results
The experiments conducted showed a great improvement both in the
gasifier operation and in the gasification quality; however, the intensive gasifica-
tion of the coal blends in the gasifier resulted in high-temperature-induced
material exhaustion of the reactor tube followed by melting of it and breaking
down of two reactor furnaces. Therefore, further tests (after repairs) were carried
out exclusively with biomass/textile waste blends.
Gas analyses showed a fluctuation in the product gas composition, which
is a new phenomenon in the gas analyses for the Lund gasifier. It is believed
that sometimes a separation of fuel blends results in an uneven oxygen distribu-
tion in the bed. The input oxygen reacts first with the textiles, which are more
reactive than coal or biomass and results in total combustion of the textile
fraction. In this case the amount of CO2 in the gas increases at the expense of the
combustible gases.
According to preliminary calculations, the volumetric heating value of the
product gas varies between 3.5 and 4.5 MJ/Nm3, depending on the operational
conditions.
Study of the concentrations of trace gases shows that during coal/textile
experiments more sulfur compounds are produced. The concentration of H2S
in these experiments exceeds 700 ppmv in the gas. The amounts of SO2 and
COS in the gas are also higher than those in the biomass/textile gasification.
Fuel Blends and Alkali Diagnostics 477

The content of the sulfur compounds is not considerably affected by the textile
mixing ratio.
Adding textile to biomass does not affect the composition of the product
gas. However, at high mixing ratios the amount of unburned carbon in the fly ash
increases.
There is a clear relationship between the biomass/textile mixing ratio and
the concentrations of trace gases such as NH3, HCl, and H2S. While the S, Cl,
and N contents of the textile are higher than those of the biomass, increased textile
addition increases the amounts of these traces in the gas.
Ammonia formation in the gasifier is regulated by three important factors:
the fuel nitrogen, the gasification temperature, and the ER value. The relationship
between the fuel nitrogen and the ammonia concentration in the gas is linear.
Higher temperature results in decrease of the ammonia. The effect of higher ER
in ammonia decreasing is due to both gas dilution and also the improved oxygen
availability in the gasifier.
The amount of the lighter PAHs in the tar is larger in the case of mixture
gasification. Compounds with molecular weight larger than 168 (dibenzofurane)
make up about 84–95% of the total PAHs for the pure biomass, while the
corresponding value for the mixture is below 55% at its highest value.
Textile addition contributes to increased formation of light hydrocarbons
such as benzene, toluene, ethane, and butane.
The heating value of the gas is affected mostly by the ER value. Including
the combustion heats of the light hydrocarbons improves the gas heating value
by 15–28%.

2.4 Analytics of Gaseous Emissions, Fly Ash, and

Residues; Utilization of Residues such as Fly Ash and
Ash in Subsequent Processes; Difference from
Utilization and Treatment of Pure Coal Ashes
One partner has carried out valorization concepts for residues. Residues of
combustion and gasification of fuel blends were delivered by other project
partners working on co-combustion and co-gasification.
2.4.1 DMT-SysTec
In existing literature only a few reports are found in this context. One of these
reports deals with thermal utilization of textile floor matting materials in a
stationary fluidized bed (single-fuel) firing system. The composition of the
resulting ashes with high calcium oxide concentrations should allow manifold
applications in cement industries. Utilization possibilities quoted in the project
are proportioning to concrete mixes as well as an additive for cement production
feedstock. Supposedly, there are further utilization possibilities, such as produc-
478 Romey

tion of sand-lime bricks and lightweight concrete. The ashes should contain only
small concentrations of heavy metal and the elution rates of these heavy metals
should be low as well, thus allowing environmentally benign and cost-effective
disposal on mineral-matter dumps.
Further examples are the use of textile carpet waste as fuel or feedstock in
blast furnaces or in a modified refuse incineration plant. Initial field trials have
shown that the resulting blast furnace slags are particularly suited for producing
cement qualities intended for underwater use, and the refuse incineration slags
are suited for use as sand for various purposes in civil engineering and road
construction. Dusts from dust separation systems as well as residual ashes need
to be dumped.
Bochum University has carried out combustion tests in a fluidized bed
combustion furnace with fuel blends of coal and textiles (blends of propylene and
flax, TA, and TD). The investigations into the utilization possibilities obtained
with these residues have shown that the levels of, for example, heavy metals are
low enough and that they remain under the limit values for construction materials
for restricted installation. Furthermore, it would be possible to use the residues as
underground mortar, e.g., for backfilling or roadside packs. At the same time, the
criteria to be fulfilled by the fluidized bed ash include the following: homogenous
chemical composition and constituents, constant particle size distribution, suffi-
cient strength, controlled carbon and CaO, and availability of adequate quantities
Special construction material tests to establish the suitability of the residues
in certain construction materials, e.g., in mortars or mining mortars, or preparation
tests, e.g., to separate interfering components, must be investigated with larger
quantities of ashes.
Lund University has also conducted gasification tests (with different ER
values and bed material) with biomass and textiles. The loss on ignition, some
heavy metal levels, and the PAH values exceed the limiting values for utilization
as construction material and for dumping on a mineral dump. Reducing the textile
portion down to, for example, 5% does not produce any basic advantage. In
addition, the chemical composition, with more than 75% MgO, also renders them
unsuitable for such a use.
If the high MgO level is maintained with an optimization of the gasification
process, however, it might be possible to recover the MgO at very high cost. Since
the MgO exhibits a higher density than the other ash portions, consideration can
be given to the procedures of wash table grading or grading in autogenous heavy
media (upstream classification). It has not been possible to conduct examinations
in this respect because the quantities of the samples were too small. The drawback
of this procedure is the necessary drying of the separated material. The remaining
residue with high carbon levels could then be returned to the reactor. In view of
the small ash portions both in the biomass and in the textiles, only a small residue
Fuel Blends and Alkali Diagnostics 479

would have to be dumped in total. The advantages of this procedure would be a

saving in fuel and bed material.
Further investigations were carried out with residues from the combustion
of biomass and the co-combustion of biomass and high-ash-containing plastics
(IST) and from the combustion of fuel blends from coal and plastics in a 1-MW
plant (a pulverized dry firing system, KEMA). Without additional treatment the
residues from the combustion of biomass/plastic and coal/plastic cannot be used
in the construction materials domain due to, e.g., high Cl, sulfate, and heavy metal
content, and partly due to very high electrical conductivity.
It is very laborious and costly to wash out the interfering constituents,
e.g., to reduce the chlorine content, because the material washed out would then
have to be dried and the contaminated solvent would have to undergo further
treatment. In addition, it is not possible to say anything about the requisite plant
concept and possible investment costs without more detailed knowledge of the
quantities arising and the actual residue qualities.
One possible form of recycling would be use in a cement works as a
supplier of fuel and minerals. For this purpose, however, the cement works would
have to be equipped with the requisite gas cleaning facilities and the mineral
composition of the residue would have to fit the raw materials concept.
A further possibility for recycling is integration of the ashes or polluting
constituents in cement. In this case the total material would be evaluated in terms
of its properties. On top of the installation costs, the integration would involve a
high additional expenditure. Beyond this there is only the possibility of dumping
on a household waste dump.

2.5 Market Study on the Worldwide Availability of Gas

Analyzing Sensors, Especially Alkali Sensors, and
Future Marketing Chances
All partners of the alkali measuring project group have contributed in the
preparation of a market study for on-line measurement of alkali and heavy metal
species. The study has been published under the title, “Diagnostics of Alkali and
Heavy Metal Release” (EUR 18291 EN; ISBN 3-00-002948-6).

2.5.1 BTU Cottbus

In order to gain an overview about the current state of the art of the techniques
and about the existing opportunities for the use of on-line alkali and heavy metal
sensors, information was requested from more than 50 selected companies and
institutions. The majority of written questionnaires was answered. The results of
this survey, with phone calls and literature searches, contributed to the market
study. It could be shown that on-line detectors are not yet available worldwide,
480 Romey

although both in Japan and in the United States, there is ongoing research to
develop corresponding measuring devices for commercial availability.
According to the survey, a potential market for on-line alkali measuring
devices exists in the following areas: combined-cycle plants using pressurized
fluidized bed combustion (PFBC, CPFBC); combined-cycle plants using inte-
grated gasification cycles and hot gas cleaning (IGCC); combined-cycle plants
using pressurized pulverized coal combustion (PPCC); plants for the thermal
exploitation of waste, especially plastic waste (combustion, pyrolysis, gasifica-
tion) and municipal solid waste (conversion); and plants for the thermal conver-
sion of biomass. Correspondingly, potential buyers and users of on-line systems
would be gas turbine producers, boiler producers, engineering companies, energy
providers, and municipal waste combustion companies.
Together with DMT and the University of Heidelberg, a comparative test
program with off-line measurements carried out by BTU Cottbus and the ELIF
method of the University of Heidelberg was outlined. The measurements were
carried out at DMT’s atmospheric fluidizing reactor, ALFRED, at a temperature
of about 850˚C. Sampling was made with three gas washing bottles, and for the
analysis of alkali species ion-selective measurements were chosen. The tests were
carried out with two different coals (Westerholt bituminous coal and Rheinbraun
lignite), both with and without additives (CH3COONa, CH3COOK), and mea-
surements were made between the cyclone and hot gas filter and behind the filter.
Potassium off-line measurements are 2 to 200 times higher than on-line
measurements, and sodium off-line measurements are 1.25 to 40 times higher
than corresponding on-line measurements. A reason for the bad results was not
identified.
Alkali concentrations measured behind the cyclone were much higher than
those measured behind the filter. An explanation was supposed in the high amount
of ash particles in the flue gas before filtration. There could have been alkali
deposition on the surface of ash particles or there could have remained a
water-soluble alkali fraction in the ash. In order to see if this assumption was
right, BTU Cottbus tested several samples of ash for their content of water-soluble
alkali compounds. At the cyclone there was only a very little amount of alkali
compounds in the ash. It could be shown that the ash particles do not have a
significant influence on alkali concentration before the filter. The alkali content
at the filter was much higher than that at the cyclone. One explanation could be
that alkali deposition took place at the filter in previous experiments and these
deposits have been removed now.
Thermochemical Equilibrium Calculations. These calculations were car-
ried out with the program ChemSage, produced by GTT Technologies. GTT
Technologies also delivered a data file which was originally created for the
investigation of gas species and condensed phases formed during coal combustion
Fuel Blends and Alkali Diagnostics 481

processes and which was extended for the calculation of alkali release processes.
This data file uses for the slag phase the quasi-chemical excess model for the slag
phase. For the other phases it uses the ideal mixing model.
BTU Cottbus carried out calculations comparable to the on-line measur-
ing results of the measurement campaigns at Foster Wheeler Energia Oy in
Karhula/Finland carried out within this program. Furthermore, calculations have
been carried out on the influence of temperature, limestone addition, and addition
of various alkali getter materials.
Results of the calculations are 4 to 10 times higher than on-line measure-
ments. At 800–1500 K, which is the temperature region of the Karhula tests,
mainly NaCl and KCl are present, whereas at more than 1500 K, NaOH and KOH
also show high concentrations. Na2SO4 becomes unstable at T > 800 K, and
K2SO4 reaches its maximum at 1500 K.
To summarize the results, it can be said that SiO2 is the only getter material
of importance. Other materials probably act only via their content of SiO2 (e.g.,
bauxite, CaO⋅MgO⋅2SiO2), and their use is only convenient if those materials are
less expensive. But attention should be given to the fact that for the materials
tested, more than the amount corresponding to pure SiO2 is necessary.

2.6 Alkali Measurements Applying Three Different

Analyzers in Flue Gas Streams of Combustion
Facilities; Refinement and Extension of the Alkali
Measurement Methods

2.6.1 University of Heidelberg (PCI)

The ELIF measurement method for PCI covered the following activities:

Design, construction, and testing of in-situ optical access for use in indus-
trial-scale systems under realistic PFBC conditions. The new design
allows coupling of the laser beam and collection of fluorescence at just
one window. The durability of this optical access was demonstrated for
conditions of 10 bar total pressure and ca. 800˚C during continuous plant
operation over periods of about two weeks.
Integration of a new detection setup in the ELIF measuring system using
fiber optics to connect to the optical access. Simultaneous two-channel
detection is also still feasible. The fiber optic cable allows the detection
system to be placed together with the remaining electronics in a compact
way and also much reduces adjustment effort. The components of the
detection unit are in a fixed arrangement, requiring no adjustment except
the simple rotation of a wheel to change neutral-density filters for
different fluorescence intensities.
482 Romey

Redesign of evaluation software, allowing more efficient, user-friendly

operation of the ELIF system and a fast overview of results.
In-situ acquisition of on-line data on alkali release from the 10-MW PFBC
plant of Foster Wheeler Energia/Karhula, Finland, including observation
of short-term (e.g., pulsing of hot-gas filter during cleaning, addition of
limestone) and longer-term effects (e.g., variations in load and oxygen
excess). Measurements were made during two periods of plant operation
of 10–14 days each and were interrupted only briefly for refilling the
laser (ca. 15 min) and securing the data. The ELIF system could be
operated nonstop for at least 24 h. In addition, simultaneous measure-
ments were made with the Tampere and Göteborg groups, using the
PEARLS and SI techniques respectively, so that direct comparisons
could be made. Measured concentrations are of the same order of
magnitude for all three measuring techniques.
Calibration of the ELIF signals was performed on a facility designed and
constructed by the group from Tampere.
Survey of possibilities for detection of heavy metals using the ELIF
method, including assessment of collisional effects. Some detection
schemes for Cd, Ni, and Zn compounds were selected, whereby single-
photon excitation is to be preferred in the first instance, since lower laser
energies will be required and interpretation of signals should be more
straightforward.

2.6.2 University of Tampere (TUT)

The PEARLS alkali measuring instrument of the TUT group has been tested in
the 10-MW PCFB of Foster Wheeler Energia Oy. The instrument has been used
for simultaneous alkali measurements together with the ELIF and surface ioniza-
tion instruments during different operations of the PCFB.
The PEARLS alkali measuring instrument has been developed by utilizing
a new photodetector which substantially improves performance.

Improved detection limit below 1 ppb (previously 5 ppb)

Improved data acquisition rate, 2 values/s (previously 10 values/minute)

The improved data acquisition rate makes it possible to analyze fast phenomena,
such as alkali behavior during the hot-gas filter cleaning pulse.
A pressurized and transportable test/calibration unit was designed and
manufactured. The operation of the unit is based on the known vapor pressure of
alkali salts at a determined temperature. The unit will help to improve the
accuracy and develop the components of the measuring instrument.
TUT provided background information for other partners by establishing
the format for preparative actions and sending a full copy of the technical
Fuel Blends and Alkali Diagnostics 483

drawings of the PEARLS instrument to the partners. The following features were
utilized in the design of the ELIF and surface ionization instruments:

Identical DN150/DN50 reducer flange design for all instruments

Architecture of the instrument body (surface ionization)
Division of the sampling line to heating zones (surface ionization)
Protective arrangements of the sampling line (surface ionization)
Ball valve for window protection (ELIF)
The PEARLS instrument has been improved significantly in the test runs;
comparison with the two other on-line instruments showed comparable alkali
levels.

2.6.3 University of Göteborg (GU)

The main objective of the project has been to develop an on-line alkali measure-
ment instrument for power plant applications based on surface ionization (SI).
Alkali metal atoms are well known to ionize easily in contact with hot metal
surfaces, and the instrument developed in this project is based on this principle.
This is an almost unique property of alkali compounds, which implies that a
sensitive and selective detector can be developed. The primary parts of the
detector are a filament, generally made of platinum, and an ion collector.
The platinum filament is supported between two electrodes and is resistively
heated for alkali vaporization in ionic form. The filament is biased at a positive
voltage of 500 V to repel the positive alkali ions formed. The ion collector is
situated close to the filament and is grounded through a sensitive electrometer.
The measured current is proportional to the arrival rate of alkali atoms onto
the filament.
In this project an instrument consisting of a SI alkali detector, a hot-gas
sampling line, and a control system has been constructed according to prevailing
high-pressure and high-temperature standards. The instrument has been tested
during two extended measurement campaigns in a PCFB combustion pilot plant
at Foster Wheeler’s R&D Center in Karhula, Finland. During the campaigns, the
instrument was under operation for more than 500 h during pressurized combus-
tion of coal. Simultaneous alkali measurements with the groups from PCI and
TUT were performed on several occasions during the campaigns. The measure-
ment campaigns have been evaluated in cooperation with PCI, TUT, and FW, and
this work also included the calibration of the SI instrument. It can be concluded
that the developed instrument has performed well and the experimental runs
confirmed that the instrument can operate during extended times in the hostile
environment present at a power plant.
After the campaigns the SI instrument was further developed. The SI
instrument operation has been simplified by increased computer control, and the
reliability of the measurement procedure has been further improved. The SI
484 Romey

detector has also been tested in a laboratory-scale pressurized gasification unit,

in order to analyze the requirements for instrument use in gasification. A descrip-
tion of the SI instrument, together with descriptions of the other detectors
developed in the program, has been published in a market study on “Diagnostics
of Alkali and Heavy Metal Release.”
The developed instrument monitors the total alkali concentration (Na + K)
in the hot flue gas of a pressurized system. In an extension of the technique, the
instrument is also intended to differentiate between Na and K as well as
discriminate between alkali species in molecular and particulate form. It was
found that platinum might be a suitable material for the ionizing hot filament,
considering both ionization efficiency and filament lifetime. A large number of
other filament materials were also tested with respect to thermal stability and their
surface ionization efficiencies for Na and K. A few of the tested materials can be
considered as alternatives to Pt in future SI applications, including platinum
alloys, high-temperature steels, and Kanthal. Based on the investigations, it can
be concluded that all tested materials (except Pt) are oxidized at the high
temperatures used in the instrument and are therefore not suitable for differential
measurements of Na and K. It is suggested that the combination of SI detection
and ion mobility spectroscopy or mass spectroscopy be investigated in future
work on Na and K separation.
Laboratory investigations have been performed with alkali-containing
particles in the size range 0.05–1 µm. The SI technique can be used to detect
single particles in this important size range, and a selection between alkali in
molecular form and alkali-containing particles can be based on this principle. The
results also suggest that a new type of particle analyzer can be constructed based
on SI technique. This simple and very sensitive device may prove ideal as an
“early warning” system for measurements of increased particle concentrations in
a gas turbine inlet.
2.6.4 DMT-FuelTec
DMT pursued various tasks in the project, including the adaptation and mod-
ification of the analyzers to applications in industrial plant. In particular, a
rugged but versatile hot-gas extraction probe system permitting reliable sam-
pling for accurate on-line measurements was designed and constructed. The
retractable probe system included a pressure lock and incorporated a measuring
chamber for the ELIF analyzer, providing a number of windows in order to let
the exciting laser irradiation in and to detect the fluorescence signal, and
providing an opportunity in its pressurized fluidized bed combustion facility,
FRED, to test developments in analyzer improvements and to carry out combus-
tion experiments at conditions most suited to instrument development needs.
These runs complemented the field measurement campaigns of the instrument
developing teams.
Fuel Blends and Alkali Diagnostics 485

Based on observations and measurements with the ELIF method, drawing

on experience with an extractive sampling line forming part of one of the three
analyzers and including DMT expertise from other instrument developments, a
new retractable hot-gas extraction probe system was conceived, designed, and
constructed that incorporated the following features.
The system is strictly modular in design. The various functions of the
system are allotted to individual components.
Components needing servicing or exchange/replacement are readily acces-
sible.
The extracting gas tube is heated indirectly and can be replaced with tubes
made from other materials if required.
The probe system is fitted to the plant via pressure lock. It can be withdrawn
under pressure, isolated, and independently depressurized.
The measuring cell provides four rectangularly arranged access ports for
observation. The measuring cell has its own individual heating system.
The sampling gas is passed, downstream of the measuring chamber, to a
cooler, followed by a strainer, to the pressure letdown valve that also
controls the total gas flow through the system. The gas can then be vented
or used for other purposes, e.g., conventional gas analysis.

2.6.5 Foster Wheeler Energia Oy (FW)

The Foster Wheeler Pressurized Circulating Fluidized Bed Test Facility in
Karhula, Finland, provided over 1000 h of operation on coal to be utilized for
alkali measurements under industrial conditions during the two test segments in
1997. The measuring teas, from the University of Göteborg, Heidelberg Univer-
sity, and Tampere University of Technology, were able to take advantage of a
large part of the operation. Technically, instruments of all three measuring teams
showed remarkable potential for industrial use.
During summer 1998, the final results of the measurements of all teams
were submitted to Foster Wheeler. The alkali emission results were reviewed
together with steady-state process data, and possible correlations with process
parameters such as temperature and pressure were investigated. Clear connections
between the alkali emission and the amount of particulate in the flue gas as well
as the flue gas temperature could be observed.
In general, with dust-free flue gas, the sodium and potassium emissions
were in the range of 25 ppb (6% O2) at temperatures of 800–850˚C, and the range
of 10 ppb is reachable at temperatures of 700–750˚C already without special
getters or process modifications with the coal used during the test campaign.
The main results of the project were published during the presentation of
the alkali group and in the poster session of the final project meeting in Brussels
on November 11–12, 1998.
486 Romey

A joint publication on alkali measurements was submitted together with the

measurement teams to the organizers of the 6th International CFB Conference,
held in Würzburg in 1999.
2.6.6 RWE Energie AG
The objective of RWE’s participation in the project was to get actual and detailed
information on possibilities of on-line measuring techniques of alkali in pressur-
ized combustion or gasification processes.
To get results with the alkali on-line measuring systems at the pressurized
fluidized bed combustion test facilities of Foster Wheeler, it was necessary to
work out a main test program, in which the basic requirements (types of coal and
additives to be used, constant and variable process parameters during the test
runs) are described. This main test program has been discussed with all partners
involved. The program has taken into account all important parameters and
operating conditions of the plant and the needs of the three different measuring
systems.
Concerning this main test program, RWE intended to participate in the
following:
1. Participation in planning the experimental program for runs in different
PFBC test units in cooperation with Technische Universität Cottbus and
plant manufacturers
2. Adaptation and study of the requirements of the analyzing system in
gasification plants
The requirements of the analyzing system for gasification plants have been
discussed with the University of Heidelberg. The considered aspects for feasibil-
ity of the ELIF measurements are as follows:
Fluorescence quenching
Absorption of laser beam
Background emission
Dimensions and geometry of the system
The result was that measurements under the stated conditions (HTW gasification)
will be feasible.
The potential applications relating to gasification of the measurement
system are future power plant concepts with combined cycle such as “IGCC” or
“PFBC second generation.”
The PFBC process with extremely staged combustion could be one possible
option for a high-efficiency power plant concept. The process will be investigated
at the laboratory scale at different places and is in the very beginning of the
development.
Fuel Blends and Alkali Diagnostics 487

To get an impression what alkaline values are to be expected in a future

combined-cycle process such as staged combustion, the University of Cottbus did
some calculations with their program ChemSage for the use of lignite.
After finishing the investigation of the potential applications of the mea-
surement system in gasification plants, it was discussed to use the systems in
atmospheric combustion in conventional boilers. This had the following back-
ground.
The alkali content of the Rhenish lignite of some mining areas will increase
during the next years. Important for RWE Energie is to obtain knowledge of the
influence of higher ranges of alkali content. Suitable measuring systems need to
be tested. Therefore theoretical investigations have been carried out concerning
the possible application of the Elif system.
It is especially important to obtain a better understanding of alkali deposi-
tion and the incipience of fouling and corrosion.
The test employment of one or more measurement systems in conventional
firing systems at RWE Energie is under consideration.

2.7 Database on the Occurrence of Textile Wastes in

Industry and Their Characterization (Amount,
Composition, Contamination, etc.)
Two project partners were involved in this topic.

2.7.1 University of Münster (FATM)

FATM carried out a database study on the occurrence of textile wastes from
different textile branches in Germany. The main aim of this study was the
calculation of the textile production waste that arises in the Federal Republic of
Germany, according to production phases, in order to gain a basic idea about the
amount of production waste that would be available for the combustion and
gasification as an addition to coal. Typical waste quotas are determined for the
individual production processes of the textile industry. In connection with the
amounts and types of raw materials used, these quotas will make it possible to
give a differentiated estimate of the amount of waste arising, according to its type
and the raw material structure. These are basics for a following market study on
the real and potential markets.
The calculation of the amount of textile production waste arising is based
mainly on two sources. The first is the official statistics on production in German
industry in 1995, which provides production data for the different branches of the
textile industry. The other is a questionnaire sent to companies by the FATM in
1996, which was used to get and calculate waste quotas for different production
processes.
488 Romey

In addition to waste quotas and waste volumes, a breakdown by types of

fiber for the raw materials is given.
There is a significant concentration of textile mills in three German
states. Taking the number of employees as a measure for the regional distribution
of production, about 76% of the textile production waste in 1995, i.e.,
83,000 t (83 kt), is concentrated in Nordrhein-Westfalen (NRW), 32 kt; Baden-
Württemberg (BW), 16 kt; and Bayern (BY), 14 kt. At the county level, about
60% of the occurence of all textile waste is concentrated in only six regions:
Münster (NRW), 12 kt; Düsseldorf (NRW), 12 kt; Oberfranken (BY), 8 kt;
Freiburg (BW), 7 kt; Stuttgart (BW), 5 kt; and Kassel (Hessen), 5 kt. We find a
similar result for the regional occurrence of textile dusts, which are of special
interest due to problems in recycling and current prices of disposal on one hand,
and opportunities in preparing the dusts for co-combustion with coal on the other
hand. About 12 kt, i.e., 65%, of all textile dusts from production arise in only
seven counties: Münster (NRW), Düsseldorf (NRW), and Köln (NRW), with a
combined amount of dusts of 5.2 kt; Freiburg (BW), and Stuttgart (BW), with a
combined amount of 4.4 kt; Oberfranken (BY), 1.4 kt; and Kassel (Hessen), 900 t.
Disposal Costs of and Profits from Textile Production Waste. Textile waste
is handled as a commercial good. A positive price for a type of waste indicates a
path for recycling the waste inside or outside the textile chain. If there is a
negative price for textile waste, the textile companies have to pay for disposal
and only in this case will textile waste be offered for co-combustion instead of
landfilling. Table 4 offers information on the availability of textile waste for
co-combustion with coal in regard to the market situation.
The operational result is that mainly textile dusts will be offered for
co-combustion with coal, because fiber waste from the nonwovens production can
be classified as textile dusts due to the very short length of residual fibers.
Collecting Systems and Remarks on the Logistics for Textile Waste, Espe-
cially Dusts. Concerning the regional occurrence of textile waste, a few com-

TABLE 4 Average Disposal Costs/Earnings of Textile Waste in the GE

Textile Industry, 1995

Textile Fiber Yarn Fabric

dusts waste waste waste

Spinning mills –230 DM/t +640 DM/t +270 DM/t —

Nonwoven mills –320 DM/t –320 DM/t — – 320 DM/t
Weaving mills –170 DM/t — +110 DM/t +190 DM/t
Source: Personal calculation, based on FATM Questionnaire, 1996.
Fuel Blends and Alkali Diagnostics 489

panies in these regions have specialized in buying and selling textile waste. It is
therefore possible to refer to an existing logistics system for collecting textile dust
from a great number of mills.
In spite of the very low density of textile dust we have a cost-intensive ratio
of volume and weight in the transporting system. Since the three major waste-
producing regions are located in three corners of Germany, a central point for
preparing textile dust for co-combustion in a power plant nearby may not
necessarily be the optimal solution with respect to transportation cost. Thus
co-combustion plants in the main three regions might be the optimum solution
with respect to logistics.

Conclusion. From a total of currently about 83,000 t per year of all types
of textile waste from the textile production chain, only 18,000 t per year of textile
dusts, including short fibers, are offered for co-combustion with coal. Although
this type of textile waste has a high content of energy, it has to be collected from
a great number of plants and prepared for co-combustion in a power plant with
an extra feeding system. Technical and environmental problems with co-combustion,
as well as the low rise of textile dusts over time in hundreds of mills located in
several regions of Germany, seem to call for small technology concepts for using
the energy content of textile dusts at the place of origin.

2.7.2 CITEVE
CITEVE worked on the calculation of the main textile production waste arising
in Portugal, according to production phases of four representative industrial
sectors in the country and, on other hand, the determination and characterization
of the supplied production waste with regard to its fundamental characteristics.
CITEVE also supplied representative wastes to the Portuguese partner (IST)
performing the combustion tests (package 1).
The results developed, gathered from different companies, allowed the
calculation of median percent figures of waste on production phases. It was
possible to calculate figures for secondary raw material, primary waste, and
secondary waste.
In terms of waste samples and quantities, samples were supplied to the
University of Lund and to the partner IST consisting of about 1 t of seeds, dust,
and very short fibers of cotton, 500 kg from a blowing at a spinning mill and the
other 500 kg from the gig operation at finishing. This last kind consisted of short
fibers (short woolens) with color and finishing/dyestuff.
Those two kinds of waste supplied to IST and another four, for a total of
six main kinds of waste, have been identified as the most representative secondary
waste in Portugal.
The kind and quantity of waste produced depends mainly on the production
process. CITEVE collected information about different production processes of
490 Romey

the most representative industrial sectors in the country: wool, cotton, knitting,
and clothing. This waste was divided into three categories:
A: secondary raw material
B: primary waste
C: secondary waste
Since A and B waste can be recycled, only C waste were submitted to combustion.
Also, the clothing and knitting industries do not produce A and B waste, so only
the wool and cotton industries were analyzed in terms of category C waste.
Quantity of Waste Produced. The calculation of the amount of textile
production waste arising was based mainly on the following sources:
Gathering of industrial quotes of waste per process by means of diagnostics
on site
Statistics of Interlaine and L’Atlas du Textile Mondial de Françoise Depin,
the first given for 1995 and the second for 1994, concerning the Portu-
guese textile production of the main four industrial sectors in the country.
Combining both sources, it was possible to approximate calculations of national
textile production waste, as shown in Table 5.
It was established that in Portugal, in 1995, about 65,100 t of textile waste
were produced. If we take into account just the most representative four textile
sectors, considering nevertheless all kind of waste, either those which can be

TABLE 5 Amount of Waste/Year

Total
Gross amount of Share of Share of Share of
Products production (t) waste (t) kind A (t) kind B (t) kind C (t)

Woolen yarns 19,060 2,995 1,256 1,545 194

(worsted)
Woolen yarns 9,300 628 438 190
(carded)
Woolen fabrics 7,283 303 303
(worsted)
Woolen fabrics 3,921 941 920 21
(gig)
Cotton yarns 129,400 38,856 20,704 5,375 12,777
Cotton fabrics 49,000 1,945 1,515 430
Knitting fabrics 143,750 4,446 4,446
Clothing 85,000 15,000 15,000
Total 65,114 21,960 29,542 13,612
Fuel Blends and Alkali Diagnostics 491

recycled or not, regarding the waste which cannot be raw material of another
process (not recyclable), the amount is just about 13,600 t.
The regional occurrence of textile waste for combustion or gasification has
to be determined under aspects of logistics optimization between regions. Regard-
ing this aspect, the distribution of the Portuguese companies in land was found
with the support of a study from CENESTAP, the Centro de Estudos Têxteis
Aplicados, of 06/08/97, with about 1800 companies. With this study, it was
possible to establish the distribution of the six more representative Portuguese
districts, considering the number of companies per district (Braga, Porto, Guarda,
Castelo Branco, Lisboa, and Setúbal). Lisboa and Setúbal are not considered as
important as the others for the Portuguese economy, but we must refer to them if
we look to the amount of waste produced by its companies, mainly clothing.
According to Table 4, the four main industrial sectors represent the produc-
tion amounts in Table 6.

2.8 Techno-economic Study on Industrial Utilization of

Fuel Blends Produced from Industrial Wastes and Coal
One partner has worked on this topic using data and test results from all other
participants in the project.
2.8.1 University of Ulster (UUE)
The work of the Energy Research Centre of the University of Ulster was to
perform technical, environmental, and economic assessment studies of a range of
power generation technologies using coal/textile waste/plastic waste blends.
Models have been developed for many different technologies, including combus-
tion in fixed beds, bubbling beds, circulating fluidized beds, and entrained flow
reactors, and gasification in a fixed bed gasifier with a gas engine, and a fluidized
bed gasifier using a gas turbine in simple cycle, combined cycle, STIG, and ISTIG
modes. These models have been used to integrate the results from the other
partners. The specific work program called for:
Collection of data on the wastes which are available and assessment of the
cost and energy required to convert them to a feedstock suitable for
co-processing with coal

TABLE 6 Production Amounts of Four Main Industrial Sectors

Clothing 85,000 t
Spinning/weaving (wool and blends) 39,564 t
Spinning/weaving/home textiles (cotton and blends) 178,400 t
Knitting 143,750 t
492 Romey

Validation of the pilot plant results and scale-up to a suitable size of

operation
Performance of techno-economic assessment studies for the range of pro-
posed feedstocks and the alternative co-processing technologies at a
suitable size of operation
Performance of sensitivity analyses to determine the conditions required to
make these alternatives competitive with existing disposal routes
Information was obtained from the other partners on the situation with
regard to waste textiles arising in Germany and Portugal and plastic waste arising
in Germany. Further information with regard to the geographic location was
required in order to determine the relationship between cost of transportation and
size of waste processing plant.
The technologies modeled included:
Small-scale combustion in a fixed bed furnace
Medium-scale combustion in a bubbling or circulating fluidized bed
Large-scale entrained flow combustion
Small-scale gasification in a fixed bed gasifier with a gas engine
Medium-scale fluidized bed gasification using either a gas engine or a gas
turbine
Medium- to large-scale fluidized bed gasification systems using gas turbine
topping cycle and steam turbine bottoming cycle
Medium-sized fluidized bed gasifier with STIG and ISTIG configuration
The overall results from the work using the ECLIPSE process simulator can
be summarized as follows.
The addition of 5% plastic waste to a coal-fired PF power station reduces
its efficiency by 0.3%, due to the high moisture content of the plastics. There is
a small increase in capital cost associated with feeding the waste plastic to the
boiler. For the overall economics to be the same as the coal-only case, the power
station could afford to pay up to 20 ECU/dry t for the plastic waste, compared
with 37 ECU/t for the coal. The disposal of plastic waste normally costs
approximately 94 ECU/dry t. Therefore, provided the cost of transporting, han-
dling, and preparing the plastic waste for use in the power station does not exceed
114 ECU/dry t, then it is economic for the power station to use plastic waste.
The addition of 20% plastic waste to a Texaco IGCC power station
increases the net efficiency by 0.2%, due to the improved gasification qualities
of plastic over coal. There is an increased capital cost for handling the plastic
waste, which means that for the overall economics to be the same as the coal-only
case, the power station could afford to pay up to 38 ECU/dry t for the plastic
waste, similar to the coal price. Again, the disposal of plastic waste normally costs
approximately 94 ECU/dry t and therefore, provided the cost of transporting,
Fuel Blends and Alkali Diagnostics 493

handling, and preparing the plastic waste for use in the power station does
not exceed 132 ECU/dry t, then it is economic for the power station to use
plastic waste.
While there are sufficiently large amounts of plastic waste available, the
amount of textile waste suitable for co-combustion in Portugal and Germany is
limited. If all the available textile waste in Germany is used in a 450-MWe British
coal air-blown gasifier (BCABG), less than 2% of the coal thermal input can be
replaced by textiles. This generally has a small positive benefit to the BCABG
power station, which means that the textiles have a value of 32 ECU/t to this
power station.
With textile waste suitable for co-processing of only 18,400 t/year in
Germany, this imposes limits on the maximum size of a power station. When
using a 5% textile share in a CFBC power station, a 122-MWe plant can be
sustained with the available textile dust, while with a 20% textile blend this is
reduced to 24 MWe. For textile waste arising in Germany, the ideal location for
a power plant is Giessen for larger-scale plants, while Freiburg or Münster are
best suited for smaller plants. Using textiles in a CFBC power station reduces the
efficiency and increases the capital cost. However, the textiles have a positive
value to the power station if it can be assumed that the textile disposal cost of
about 135 ECU/t is avoided. Hence, although the textiles have a higher transport
cost than coal, the overall fuel cost is improved when textiles are employed. As
a result, a reduction in BESP can be achieved. The best reduction in electricity
price for co-combustion of coal with textiles can be accomplished for a 5-MWe
power station with a 20% textile share. For a 5% textile share the most advanta-
geous size is a 20-MWe CFBC plant.
The co-combustion of a wide range of blends of waste plastic film with
biomass was successfully modeled. The value of waste plastic as a feedstock was
shown to be between 13 and 21 ECU/t greater than the equivalent ECU/GJ value
of the biomass feedstock, due to its improved combustion performance. In the
specific case studied, there are problems with the availability of suitable plastic
waste and the high transport cost. However, it has been shown that this particular
technology can successfully recover the energy from blends of waste plastic with
biomass. Under more favorable waste plastic supply conditions and waste plastic
transport distances, it has the potential to be economically viable.

3 CONCLUSION
Thermal utilization of residues and industrial waste in combination with fossil
fuels and/or biomass will become a very important factor in future energy
production from the viewpoint of environmental aspects, the careful use of
resources such as fossil fuels, as well as developing/increasing new market
segments for decentralized energy production. But also in existing markets and
494 Romey

technologies such as iron and steel or cement production, the utilization of

residues or waste might play a more important role in the future. The main goal
of the project, “Advanced Combustion and Gasification of Fuel Blends and
Diagnostics of Alkali and Heavy Metal Release,” was the investigation and
development of data and technologies for commercial use of waste in industrial
applications.
The first aim of the project part “Utilization of Residues” was development
of basic knowledge for characterization and assessment of these novel residues.
On this basis, environmentally compatible valorization concepts for the various
residues were developed for making up marketable products to be recycled to the
economy. Residue samples (fly ashes, bottom ashes, filter material, etc.) were
supplied by the partners performing the combustion/gasification tests of the fuel
blends in their facilities and were subsequently investigated according to the work
program. After analysis and characterization of the samples, valorization concepts
have been developed or modified in comparison to known concepts. For devel-
oping the above-mentioned aims and for categorization of the residues in view of
different valorization lines, exact analyses of the matters fed to the process via
secondary fuels in the coal have been carried out. Due to the different properties
and environmental impacts of the residues, specific preparation and valorization
concepts were developed for each case. Above all, the fly ashes and filter cakes,
which may be contaminated by organic noxious matters, in particular dioxins,
furanes, and halogenated hydrocarbons, as well as by inorganic noxious matters,
such as salts, desulfurization products, or heavy metals, have been investigated
and the ashes have been classified for potential market segments, mainly in
construction industries.
The conventional method of measuring alkali levels in hot gas consists of
extracting a gas sample by means of a condensing probe followed by a number
of impingers for the absorption of condensed aerosols. The probe is rinsed after
taking the sample, and the aqueous liquids are then analyzed by flame photometry
on alkalis. Sampling takes a long time because of the rather low sensitivity of the
analyzing method and is prone to very many errors of unknown magnitude. No
continuous monitoring of the alkali levels was possible at the beginning of the
project, and plants had to be run at steady-state conditions during the sampling
period, lasting several hours.
A market study has been performed concerning the availability of gas
analyzing sensors, especially alkali sensors, and the future marketing possibilities
promised after successful development of sensors. The study has shown on the
one hand the worldwide novelty and uniqueness of the alkali sensor development
to be undertaken in this project and on the other hand has clarified future
marketing chances. In a joint effort, instrument developers, plant manufacturers,
and operators have used worldwide business contacts and screen literature to meet
the above-mentioned objectives.
Fuel Blends and Alkali Diagnostics 495

It was possible to develop three different on-line alkali measuring devices

which have been demonstrated under comparable conditions in a 10-MW plant,
and the instruments might be ready for commercial use in industry.
The waste which has been investigated arose from different processing and
production processes and varied in chemical composition and physical properties.
Depending on their origin or their pretreatment, the waste was contaminated by
heavy metals and organic compounds arising from colors and special coatings,
e.g., for flammability reduction, and they occurred in the form of dust or pieces.
One way for efficient energy recovery from those waste was co-combustion with
coal, which is preferred compared with pure waste incineration, since this was
not optimized for energy production and caused some problems with low melting
points of the plastic waste. However, based on the results of our R&D program,
it should be considered in later projects.
Three techniques for the on-line measurement of alkali species have been
developed based on physical detection of alkali atoms. Simultaneously, all three
analyzers in several individual experimental campaigns of the hot-gas duct of two
pressurized fluidized bed combustion pilot plants of different design have been
demonstrated and optimized. The project aim was to obtain an overview of alkali
concentration levels in hot gas from coal combustion for a variety of fuels and
plant operating conditions.
The alkali concentration levels have been correlated to the parameters set
or found in the experimental programs and an attempt has been made to find
relations between hot-gas conditions and alkali levels and to draw conclusions
with respect to generalization of the results obtained. It has been shown that the
risk alkali vapors present in the technology of pressurized fluidized combustion
of solid fuels can be assessed on a fairly well-defined basis of data at the end of
this project. Furthermore, it has been demonstrated that the accuracy and reliabil-
ity of the three different alkali analyzers could be significantly improved and their
capability to be used in commercial applications might be possible in future.
It can be stated at the end of the project that all 20 partners have worked
together in a very cooperative way, that all groups have reached the main goals
of their planned working packages, and that a number of results have been
obtained for later technical/industrial use as described above. On the other side,
some of the work has shown that there is still demand for additional development
before using the results commercially.
Due to the large amount of waste plastics, thermal utilization will be an
attractive option in a number of industrial applications; however, as already
described, standardization equalization of national laws and reduction of prepa-
ration costs must be considered and solutions must be developed.
The textile waste market seems to be very limited, and it can be concluded
that due to the limited amount of textile wastes for thermal utilization, the
material will only be of interest for local application in grate firing in some
496 Romey

regions; however, the material can be used in combination with biomass and can
improve the combustion behavior of boilers.
The developed on-line alkali and heavy metal devices might be ready for
technical application in advanced power generation technologies but also in
conventional boilers, incineration plants, and in boilers where biomass is burnt.

REFERENCES
1. I. Romey, JOULE-THERMIE Clean Coal Technology R&D, Fuels Blends and Alkali
Diagnostics, European Commission, 1999.
2. I. Romey, J. Barnish, and J. M. Bemtgen, (eds.), Diagnostics of Alkali and Heavy Metal
Release, European Commission, 1998.
3. J. M. Bemtgen and I. Romey, JOULE-THERMIE Clean Technologies for Solid Fuels
R&D (1996–1998), European Commission, 1996.
4. I. Romey, Co-utilization of Coal with Biomass and Wastes, Proc. Final Conference,
Volumes I, II, and III: Executive Summary, Final Reports, APAS Clean Coal Technol-
ogy, 1993–1994.
23
Best Practices for the Oil and
Gas Exploration, Production, and
Pipeline Transportation Industry

Bart Sims
Railroad Commission of Texas, Austin, Texas

1 INTRODUCTION
Many oil and gas exploration, production, and pipeline companies (E&P compa-
nies), both large and small, have applied the concepts of waste minimization in
their operations. As a result, E&P companies have achieved progress in pollution
prevention by eliminating oil and gas wastes at the source. As well as pollution
prevention efforts, E&P companies have implemented effective reuse and recy-
cling alternatives for various oil and gas wastes.
Several E&P companies have provided case studies which highlight suc-
cessful and beneficial waste minimization. The following case studies illustrate
how E&P companies have applied waste minimization techniques, or “best
management practices,” to either reduce the quantity of waste generated at the
source or to recycle, reclaim, or reuse waste streams that could not be reduced at
the source. E&P companies have implemented equipment modifications, proce-
dural or process changes, product substitution, and reuse of spent materials to
achieve beneficial waste minimization. Importantly, the case studies demonstrate
that, in addition to preventing pollution, waste minimization in oil and gas
497
498 Sims

operations can provide valuable benefits such as cost savings, increased revenue,
improved operating efficiency, reduced regulatory compliance concerns, and
reduced future potential liability concerns.
The following case studies have been provided to the Railroad Commission
of Texas (RRC) Waste Minimization Program. Most are included in the RRC
publication, Waste Minimization in the Oil Field (1). The case studies are
presented for the various E&P operations in the following sequence: drilling
operations, production operations, natural gas treating and processing operations,
and pipeline operations. Each case study is presented as a “problem” which is
addressed by a waste minimization “solution.” The benefits gained close each
case study.

2 CASE STUDIES OF SUCCESSFUL WASTE

MINIMIZATION IN THE OIL AND GAS EXPLORATION
AND PRODUCTION INDUSTRY
2.1 Waste Drilling Fluid in Conventional Earthen Pits at a
Drilling Operation
A small independent operator was concerned about the volume of drilling waste
in conventional reserve pits at his drilling locations. Waste management costs
were a concern, as well as the costs associated with the impact of drilling fluid
on adjacent land, due to earthen reserve pit failures. The operator was concerned
about the potential for surface water or ground water contamination and the
associated potential liabilities. As a solution, the operator implemented a proce-
dure change, using a closed-loop drilling fluid system rather than conventional
earthen reserve pits. The wells were to be drilled in relatively shallow, normally
pressured strata and, therefore, were amenable to use of a closed-loop system.
The operator negotiated with drilling contractors to obtain a turnkey contract that
required the drilling company to use a closed-loop system and take responsibility
for recycling the waste drilling fluid generated by the operation. The turnkey
contract was incrementally more expensive. However, the operator realized a
savings of about $10,000 per well because of reduced drill site construction and
closure costs, reduced waste management costs; and reduced surface damage
payments. Importantly, the drilling mud was reclaimed and reused, rather than
disposed of by burial or land application. Also, the operator reduced the potential
for environmental impact and associated potential liability concerns.

2.2 Used Lubricating Oil and Filters from Power Units on

a Drilling Rig
This case history is taken from the Society of Petroleum Engineers paper titled
“Monitoring Engine Oil” (2). A drilling company was concerned with the quantity
Best Practices for the Oil and Gas Industry 499

of waste lubricating oil and filters generated by diesel power plants on its rigs,
and the costs of new oil, new filters, and maintenance. The drilling company
recognized that the problem stemmed from performing oil and filter changes at
500-h operating intervals as recommended by the engine manufacturer. The
solution the company selected was a procedural change which resulted in source
reduction, even though their main goal was to reduce operating costs. The
company extended the operating interval between lube oil changes for the diesel
power plants by performing sampling and analysis of the lube oil to determine
when lube oil changes were actually needed. The company established threshold
concentrations for specific analytes, such as contaminants, additives, and metals.
Whenever a threshold value was exceeded, a lube oil and filter change was made.
In any event, the maximum operating interval was set at 1250 h. This procedural
change resulted in a decrease in oil costs from $64/day to $41/day, which
translates to a 36% reduction in waste generation. Additional cost savings were
realized due to decreased maintenance requirements, improved operating effi-
ciency, and reduced waste management requirements. Importantly, no harm or
unusual wear was experienced in the diesel power plants.

2.3 Coalescor Panels from Horizontal Separators in an

Oil Field
An oil and gas company was concerned about the use of flow stabilizing coalescor
pads (14 in. deep with 1-in. by 1⁄2-in. openings) in over 300 horizontal oil/water
separators used in a large field. The coalescor pads would plug with solids and
ice, resulting in need for replacement. Occasionally, differential pressure would
cause coalescor pad segments to break free and damage the separator’s interior
components (e.g., level controls and cathodic protection anodes), resulting in
costly repairs. A typical coalescor pad replacement cost about $4000, and, if the
separator was damaged, cost about $1000 to $5000 for repair. Frequent replace-
ment of the pads resulted in high maintenance and waste management costs
(15–20 separator coalescor change-outs would generate about 25 cubic yards of
waste, costing about $1000 for disposal). Also, when a separator had to be
serviced due to coalescor pad problems, production had to be shut-in. As a
solution, the operator employed a simple equipment modification by designing a
1⁄2-in.-thick fiberglass baffle perforated with 1-in.-diameter holes to replace the
coalescor pads. The new baffle cost about $200 to fabricate and install in each
separator. The new baffles performed well in the separators, providing flow
characteristics adequate for optimum performance. The operator enjoyed im-
pressive benefits from the equipment modification. The elimination of the old
coalescor pads in turn eliminated all of the previously experienced separator
maintenance, waste generation and management, and the associated costs. A
simple $200 modification eliminated the waste disposal, repair, and maintenance
500 Sims

costs. Therefore, it is apparent that the small capital investment was quickly
recovered. Additionally, the operator gained benefits such as increased produc-
tion, reduced regulatory compliance concerns, and improved worker safety due
to reduced confined entry.

2.4 Crude Oil and Saltwater Releases from Flowline

Ruptures in an Oil Field
A very small independent oil producer operated a shallow, 40-well oil field in an
area subject to extremely cold winters. The operator’s poly flowlines would freeze
and rupture in the cold weather. Increased paraffin deposition also contributed to
this problem. The flowline ruptures resulted in crude oil- and saltwater-contami-
nated soil that had to be cleaned up under state environmental regulations. Also,
to prevent rupture of the flowlines, the rod pumps had to be shut down when
freezing weather was predicted, resulting in loss of production and revenue. The
operator employed an innovative and low-cost equipment modification to resolve
the problem. The operator designed and installed automatic pump shut-off
systems at a cost of about $75 per well, which included parts and labor. The total
investment for installing the devices on the 40 wells was about $3000. The
automatic shut-off system was made using an automotive brake light switch
(pressure switch), copper tubing, hydraulic fluid, and a simple relay switch. The
circuit was designed such that the pump had to be manually restarted after the
cause of shut-off was determined. The operator’s equipment modification resulted
in impressive source reduction. Since installation of the automatic shut-off
systems, the operator has not experienced a single flowline rupture. As a result,
the operator eliminated the generation of contaminated soil and the associated
cleanups and loss of production and revenue. The operator did not provide the
specific economics for this project; however, it was clearly cost effective. The
savings from reduced soil cleanups and increased revenue from more efficient
production easily recovered the nominal capital investment.

2.5 Pump-Jack Gear-Box Lube Oil at an Oil Field

A small independent operator felt that the cost of replacing lube oil in pump-jack
gear boxes, including maintenance and waste lube oil management, was exces-
sive. The operator investigated opportunities to reduce these costs and the
management of the waste lube oil. The operator’s simple solution, which resulted
in source reduction, was to contract a service company to service the pump-jack.
The service company filters and treats (i.e., purifies) the lube oil on-site, and then
returns the reclaimed lube oil to the gear box for reuse. The cost of this service
was about $35–$40 per pump-jack in 1995. The operator’s use of this service
eliminated his generation and management of waste lube oil and the associated
Best Practices for the Oil and Gas Industry 501

maintenance requirements. The change in procedure was cost effective. New

replacement lube oil costs about $175 per pump jack; therefore, a savings of about
$135 per pump jack was realized. Additional savings were realized because of
reduced waste management and maintenance costs.

2.6 Produced Water Filters at an Enhanced Oil

Recovery Project
A small independent operator generated a large quantity of waste filters from a
produced water injection system used in an enhanced oil recovery project. About
14,000 bbl of produced water were injected each day. The operator was replacing
two produced water filter units at each of 36 injection wells twice per month,
resulting in about 1700 waste filters per year. The operator spent $4148 per year
for new replacement filters. Additional expense was incurred from waste filter
management and maintenance. The operator made a simple equipment modifica-
tion and procedure change to achieve significant source reduction. The operator
began basing filter replacements on differential pressure rather than on a twice-
monthly schedule. The operator installed valves on each filter inlet and outlet to
accommodate a temporary pressure gauge hookup for differential pressure mea-
surement. A capital investment of $1800 was required for installation of the
valves. The operator’s simple equipment modification and procedure change
resulted in a significant reduction in the quantity of waste filters. In the year
following the change, a total of 28 waste filters were generated—a 98% reduction
in waste generation. The change was very cost effective due to reduced mainte-
nance requirements, reduced waste management, and reduced filter replacements.
The operator saved about $4000 per year due only to the reduced filter replace-
ment costs. The capital investment was recovered in less than five months.

2.7 Heat-Medium Oil Filters at a Natural Gas

Processing Plant
A natural gas processing plant operator was generating a large quantity of
conventional filters from the operation of a heat-medium oil system. The system
had three sets of filters: a 36-filter set changed weekly; a 3-filter set changed
bi-weekly; and a 54-filter set changed bi-monthly. The costs associated with the
filter change-outs, new filters, and lost oil were high. Also, the plant had changed
to a new type of heat-medium oil that more effectively cut deposits in the system,
resulting in the need for more efficient filtering. To address the waste filter
generation problem, the operator made an equipment modification. The operator
installed spinner (i.e., centrifugal) filter units in place of the conventional 36-filter
set and 3-filter set. The heat-medium oil is now circulated out of the system,
cooled, and run through the spinner filter units. The spinner filter units require
502 Sims

clean-out three times per week, rather than removal and disposal of a filter casing
and media. By eliminating the conventional filter sets, the operator eliminated the
generation of 1950 waste filters per year. The cost savings were significant.
Approximately $18,500 per year was saved due to reduced filter replacement
costs, reduced waste management costs, reduced labor and maintenance costs,
and reduced lost oil costs.

2.8 Screen-Type Lube Oil Filters on Drive Engines at a

Natural Gas Processing Plant
A natural gas processing plant operator was concerned about the quantity of waste
sock lube oil filters and lost lube oil created by filter changes on the plant’s 16
large drive engines. The seven sock filters on each drive engine were changed
four times per year or every 2000 operating hours. Each filter change required
about 40 gal of new lube oil, as well as new filter installation and spent filter
disposal. The total filter management cost per year for the 16 units was
$10,315.00. The operator made a simple equipment modification. Two reusable
screen-type filters were installed in parallel with each other and in series with the
sock filters on each of the drive engine’s lube oil systems. The screen filters could
then be alternately bypassed for cleaning while the unit continued to operate. As
a result, the sock filters required replacement only once per year. This simple
equipment modification was cost effective. The reduced sock filter changes saved
$7736.00 per year, not including the savings from reduced management and
disposal costs for the eliminated waste sock filters. The operator also gained the
benefit of reduced regulatory compliance.

2.9 Soda Ash Reagent in a Sulfur Dioxide Scrubber at a

Natural Gas Processing Plant
This case history is taken from the Society of Petroleum Engineers paper titled
“Reuse of Spent Natural Gas Liquid Sweetening Solutions” (3). A natural gas
company was using sour gas fuel in their operations and necessarily operated
exhaust gas scrubbers to control sulfur dioxide (SO2) emissions. The company
conducted a study to determine if partially spent caustic natural gas liquid
sweetening solution could be used in place of soda ash solution as the reagent in
the SO2 scrubbers. To achieve acceptable performance using partially spent
caustic solutions, they found that necessary changes to the scrubber operation
were reagent feed rate, scrubber liquid pH and specific gravity, and blowdown
rate. The company found that the SO2 scrubber could be operated using the
partially spent caustic sweetening solution as the reagent without negative effects
on performance, regulatory compliance, or operating costs. The reuse of the
sweetening agent eliminated its disposal as waste and eliminated the purchase and
Best Practices for the Oil and Gas Industry 503

handling of soda ash. As a result, the company realized the benefits of cost
savings and reduced regulatory compliance concerns.

2.10 Spent Sand-Blasting Media at Pipeline

Compressor Stations
A natural gas pipeline company was using conventional sand-blasting to pre-
pare compressors and other equipment surfaces for painting. This procedure
had generated relatively large quantities of spent sand-blast media. In this par-
ticular instance, the sand-blasting of a compressor generated 27 drums of
spent sand-blast media, which was characteristically hazardous waste due to
lead toxicity. The result was significant waste management concerns and
costs. First, the cost of disposal was $6426 for the 27 drums of hazardous
waste ($238 per drum). Also, the operator was required to register as a large-
quantity generator (LQG) of hazardous waste, resulting in compliance with
the state’s stringent hazardous waste management requirements and a waste
generation fee of $2000. To address the high waste management costs and the
regulatory compliance concerns, the operator implemented a procedural change.
Rather than use conventional sand-blasting, the operator began using pneumatic
needle scalers to remove old paint and prepare surfaces for painting. The process
was somewhat slower, but it resulted in much less waste. And labor requirements
were actually reduced, because less cleanup was required. As a result of this
procedural change, the operator realized a cost savings of $8188. Much of the
cost savings were obtained by reducing the quantity of hazardous waste from
27 drums to one drum. This source reduction also significantly reduced the
operator’s regulatory compliance concerns with respect to hazardous waste
management.

3 CONCLUSIONS
These case studies are clear evidence that waste minimization efforts in the oil
and gas exploration and production industry can provide impressive benefits, as
well as environmental protection. The case studies provide excellent examples of
source reduction through equipment modification, procedural change, preventive
maintenance, and excellent examples of waste minimization through recycling
and reuse. In each instance, the oil and gas company has saved money, improved
operating efficiency, reduced regulatory compliance concerns, and reduced po-
tential future liability. More and more oil and gas companies are recognizing that
waste minimization can obtain the environmental benefit of pollution prevention
and be a good business decision.
504 Sims

REFERENCES
1. Railroad Commission of Texas, Waste Minimization in the Oil Field. Austin, TX: RRC,
revised April 1999.
2. Fullerton et. al., Monitoring Engine Oil, Society of Petroleum Engineers Paper 18663.
Proc. SPE/IADC Drilling Conference, New Orleans, LA, February 28–March 3, 1989.
3. Hahn et al., Reuse of Spent Natural Gas Liquid Sweetening Solutions, Society of
Petroleum Engineers Paper 29733. Proc. SPE/EPA Exploration & Production Environ-
mental Conference, Houston, Texas, March 27–29, 1995.
Index

Acid mine waters/drainage, 108, 217, Clean Air Act, 13, 37, 251, 424
439, 449, 460–461 amendments, 111
Acid rock drainage, 450–454 Clean Water Act, 13, 251, 263–264
Aerobic respiration, 455 Coal combustion by-products, 421–443
American Coal Ash Association, 423, Code of federal regulations, 424
435, 437 Coefficient
Anaerobic respiration, 456 heat transfer, 161
Analytical hierarchy process, 313 mass transfer, 161
Arrhenius equation, 163 Common Sense Initiative, 265
Artificial intelligence applications, 401
Composting, 90
Co-product allocation, 292
Bernoulli equation, 167 Cumulative frequency distribution, 407
Biofilters, 188
Bioscrubbers, 188
Biosensors, 228 Data format, 73
Biosorption, 195 Decision making, 77, 79, 352, 370, 380,
Biotowers, 186 366
Biotran, 337 discrete, 352, 357, 363, 365
Boiler Decontamination and decommissioning,
cyclones, 430–431 10, 12, 21
fluidized, 430–432 Design for the environment, 81, 252, 253,
pulverized coal, 430 255, 257–260
stokers, 430–431 Desulfurization technology, 433–434
flue gas, 421–443
Chemical process industry, 390 Diffusion, 155, 159
Chemical substitution, 13 Discounted cash flow, 286

505
506 Index

Ecological risk assessment (“ecorisk”), Hazardous air pollutants, 261

334, 337–341 Heat island effect, 123
Einstein’s Theory of Relativity, 149 Hydraulic loading rate, 185
Electrochemical probes, 226
Electrostatic precipitation, 432 Industrial pollution prevention, 270, 277
Energy efficiency, 114 International flame research foundation,
Energy sources 474
coal, 101 ISO 14000, 50
natural gas, 102, 122
petroleum, 102 Lab-on-cable, 228–231
Entropy, 151, 157 Landfill, 90, 196
Environmental Life cycle
decision problems, 65, 68 assessment, 274, 275, 286–290, 295
decision support systems, 58, 81, 82 costs, 314
health and safety, 260 impacts, 311
impact, 297, 450 interpretation, 314
integrated environmental information Lime fly ash aggregate, 438
system, 59 Linear programming, 78, 400
management, 49–50, 57, 68, 75, 269 Los Alamos National Laboratory, 233,
management information systems, 63, 234, 248, 328
253, 259 Love Canal, 44
models, 71, 76
performance, 53 Mass exchange networks, 400
Protection Agency (EPA), 252, 253, Material and energy balance, 276
256, 263, 265, 269, 424 Microbe, 171–180
protection strategies, 270 Microbial growth, 179
remediation, 238 Mine pollution, 216, 450–453
restoration, 10, 21 Modeling
science, 7 equation-oriented process, 393–399
Ethanol/methanol, 126 modular process, 393–399
European standards, 88 stochastic, 69, 402, 413
Eutrophication, 309 Moore’s law, 259
Extractive membranes, 197 Municipal solid waste, 13, 26

Fanning friction factor, 168 Nanofiltrational, 216, 218

Fourier’s first law, 143 National priorities list, 11
New source performance standards, 424
Generator set-aside free-funded, 332 Newton’s second law, 152
Geographic Nonlinear programming, 400
area, 38 Nonradionuclides, 335
information systems, 64, 72 Nuclear Regulatory Commission (NRC),
Gibbs free energy, 164–166 12
Global warming, 99, 102, 110, 303–306, Nusselt number, 141
311
Grashof number, 140 Organic loading rate, 185
Green engineering, 389 Oxidation reactions, 456
Index 507

Ozone, 104–106 Sampling

Ozone depletion, 302–303 methods, 407
Monte Carlo, 405–409
Parabolic trough, 130 Science Advisory Board, 256
Particle correlation coefficient, 412 Second law of thermodynamics, 152
Passive/active solar technologies, 116, Sector notebooks, 52
123, 130, 325 Source reduction, 4, 48
Photosynthesis, 455 Standardized regression coefficient, 412
Photovoltaic cells, 130, 132 Status quo point/nadir, 370–372, 378
Pollution Stefan-Boltzmann constant, 159
air, 432 Stress tensors, 159–160
control, 3, 37, 390 Stripping analysis, 226
costs, 281, 283–285 Sulfide-reducing bacteria , 451, 455, 458
non-point-source pollution, 5 Superfund sites, 11, 44, 324, 450. 456
prevention, 2–9, 15, 17, 21–24, 30, 43, Sustainable America, 255
48, 49, 234, 248, 258, 270, 273,
276, 280, 283 Terminal election acceptor, 174, 196
prevention act, 252, 270, 390, 497 Textile waste, 465
prevention assessment, 271 Thermal conductivity, 138
prevention measurement, 276, 281 Thermochemical equilibrium calculations,
production adjusted pollution preven- 480
tion, 280 Toxicity characteristics leaching proce-
road-mapping, 233, 234, 237, 248 dure, 428
solvent substitution, 2 Toxic emissions, 17
violation notices, 13, 15 Treaty of Amsterdam, 85
Potential release sites, 331 Trickling filters, 184
President’s Council on Sustainable Devel-
opment, 255 Union Carbide, 46
Probability density function, 407 United Nations Environmental Program,
Process simulators, 391 270
Project XL, 53 U.S. Bureau of Mines, 454–455

Radionuclides, 336 Waste

Reactive/proactive company, 18 facility, 67
Recycling, 4, 43, 479 generation, 16, 144
cost, 28 hazardous, 47, 48, 332, 503
materials, 7, 87 incineration, 92–97
ratio, 185 information, 7
waste, 7 low-level radioactive, 238, 240, 332
Resource conservation, 5 management, 5, 7, 10, 17, 26–29, 35,
Resource Conservation and Recovery Act 38, 85, 498
(RCRA), 251, 252, 258, 327 minimization, 3, 8, 17, 22, 145, 174,
Resource depletion factor, 308 234, 251, 390, 497
Reverse osmosis (RO), 216, 218, 221 mixed, 332
Reynolds number, 140 natural laws, 39
Rotating biological contactors, 186–187 packaging, 87–90
508 Index

[Waste] [Waste]
plastic, 471, 492–493 solid, 47, 48, 422
primary, 11, 22, 490 streams, 243, 246
sanitary, 239 textile, 471, 487–490
secondary, 11, 22, 490 treatment, 7
segregation, 7, 21 Wetlands demonstration, 456–459