A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Assessment of Precious Metal

Flows During Preprocessing
of Waste Electrical and
Electronic Equipment
Perrine Chancerel, Christina E.M. Meskers, Christian Hagelüken,
and Vera Susanne Rotter

Keywords:
Summary
industrial ecology
printed circuit board (PCB) The manufacturing of electronic and electrical equipment
recycling (EEE) is a major demand sector for precious and special met-
resource recovery als with a strong growth potential. Both precious and special
substance flow analysis (SFA) metals are contained in complex components with only small
waste electrical and electronic
concentrations per unit. After the use-phase, waste electronic
equipment (WEEE)
and electrical equipment (WEEE) is an important source of
these “trace elements.” Their recycling requires appropriate
Supplementary material is available processes in order to cope with the hazardous substances
on the JIE Web site contained in WEEE and to recover efficiently the valuable ma-
terials. Although state-of-the-art preprocessing facilities are
optimized for recovering mass-relevant materials such as iron
and copper, trace elements are often lost. The objective of this
article is to show how a substance flow analysis (SFA) on a pro-
cess level can be used for a holistic approach, covering techni-
cal improvement at process scale, optimization of product life
cycles, and contributing to knowledge on economy-wide ma-
terial cycles. An SFA in a full-scale preprocessing facility shows
that only 11.5 wt.% of the silver and 25.6 wt.% of the gold and
of the palladium reach output fractions from which they may
potentially be recovered. For copper this percentage is 60.
Considering the environmental rucksack of precious metals,
Address correspondence to:
an improvement of the recycling chain would significantly con-
Perrine Chancerel tribute to the optimization of the product life cycle impact of
Institute of Environmental Technology EEE and to ensuring the long-term supply of precious metals.
Technische Universität Berlin
Sekr. Z2
Strasse des 17 Juni 135
10623 Berlin, Germany
[email protected]
www.aw.tu-berlin.de


c 2009 by Yale University
DOI: 10.1111/j.1530-9290.2009.00171.x

Volume 13, Number 5

www.blackwellpublishing.com/jie Journal of Industrial Ecology 791

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

• to determine the technical and economic

Introduction
implications for process optimization on a
Due to the leverage of huge unit sales glob- business and enterprise level based on the
ally, the manufacturing of electrical and elec- SFA results; and
tronic equipment (EEE) is a major demand sec- • to illustrate the necessity of applying SFA
tor for precious and special metals with a strong on a process level to create a sufficient and
further growth potential. Both precious metals reliable database for modeling regional and
(gold, silver, and platinum-group metals) and global material and substance flows for a
special metals such as selenium, tellurium, bis- holistic approach to systems optimization.
muth, antimony, and indium are contained in
complex components with only small concen-
trations per unit. After the use phase, the waste Background
electronic and electrical equipment (WEEE) is
Significance and Characteristics of WEEE
an important source of these “trace elements.”
Appropriate processes are required to cope with Waste electrical and electronic equipment
the hazardous substances contained in WEEE and (WEEE) is defined as “EEE which is waste
to recover efficiently the resources. [. . .], including all components, subassemblies
To assess the recovery options for trace ele- and consumables which are part of the product at
ments, the recycling chain for WEEE has to be the time of discarding” (WEEE-Directive 2003,
analyzed. The recycling chain consists of sev- Article 3). Alternatively, the terminologies “e-
eral steps: collection, preprocessing, and end- scrap” and “e-waste” are widely used.
processing. Each step is usually carried out by The use of EEE has increased exponentially in
specialized operators. The achieved yield for in- the last decades. The generation of obsolete prod-
dividual substances depends on the recovery effi- ucts reached 8.3 million tonnes1 in the European
ciency of each individual step and on the coop- Union (EU-15) in 2005. From these only 2.2 mil-
eration along the recycling chain. Preprocessing, lion tonnes were collected and treated (Huisman
that is, presorting, liberation through manual dis- et al. 2007). Although the per-capita production
mantling, and/or shredding and manual and/or rate in populous developing countries such as
mechanical separation plays a key role in the China and India is still relatively low and esti-
overall chain. Looking at the state-of-the-art pro- mated to be less than 1 kilogram (kg)2 WEEE per
cess configurations the following argument can be capita per year, the absolute volume of end-of-life
formulated: products generated in these countries is already
Preprocessing determines which fractions and huge (Widmer et al. 2005). The generation of
thus substances of WEEE are steered into ap- WEEE is continuously increasing (COM 2005;
propriate end-processing streams. Any sorting of Huisman et al. 2007; Rotter and Janz 2006; Wid-
a specific target substance into the “wrong” out- mer et al. 2005). This increase varies per prod-
put stream from preprocessing in most cases leads uct group. In Germany, the generation of waste
to the loss of this substance in the final recovery information technology (IT) equipment is ex-
step (end-processing). Thus the effectiveness of pected to increase by 72 wt.% between 2008 and
preprocessing has a major impact on the recovery 2013 while the generation of small waste con-
of a specific substance over the entire recycling sumer equipment may increase by 10% (Rotter
chain.
and Janz 2006).
The goal of this article is to substantiate As a consequence of continuous modifications
his statement by using a substance flow analysis of function and design of EEE, WEEE is a highly
(SFA). heterogeneous mix of materials. Essential con-
The specific objectives are as follows: stituents of many electronic products, in particu-
lar IT and communication equipment (Behrendt
• to quantify and characterize the flows of pre- et al. 2007; Hagelüken 2006), are precious met-
cious metals (silver, gold, palladium) in and als (gold, silver, palladium) and special metals
out of a preprocessing facility for WEEE; (indium, selenium, tellurium, tantalum, bismuth,

792 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Table 1 Average concentration of precious metals in printed circuit boards from different equipment types
(literature review)
Equipment type (origin of Silver Gold Palladium Platinum
Reference the printed circuit board) (g/t) (g/t) (g/t) (g/t)
Angerer et al. 1993 Audio and video 674 31
equipment
Huisman et al. 2007∗ Radio set 520 68 8
Huisman et al. 2007∗ DVD player 700 100 21
Angerer et al. 1993 Personal computer 905 81
Hagelüken 2006 Personal computer 1000 250 110
Huisman et al. 2007∗ Personal computer 1000 230 90
Keller 2006 Personal computer 775 156 99
Kramer 1994 Personal computer 600 300
Legarth et al. 1995 Personal computer 700 600 100 40
Art 2008 Computer keyboard and 700 70 30 0
mouse
Huisman et al. 2007∗ Computer CRT Monitor 150 9 3
Huisman et al. 2007∗ Computer LCD Monitor 1300 490 99
Huisman et al. 2007∗ Printer 350 47 9
Ernst et al. 2003 Telephone 2244 50 241
Ernst et al. 2003 Mobile telephone 3573 368 287
Hagelüken and Buchert 2008 Mobile telephone 5540 980 285 7
Huisman et al. 2007∗ Small IT and 5700 1300 470
telecommunication
equipment
Hagelüken 2006 TV set—CRT-Monitor 280 17 10
Huisman et al. 2007∗ TV set—CRT-Monitor 1600 110 41
Huisman et al. 2007∗ TV set—LCD-Monitor 250 60 19
∗ Combination of data from different sources.

antimony). The precious metals are mainly found of the secondary production in state-of-the-art
in printed circuit boards (PCBs). Table 1 gives operations are much lower than primary produc-
the average concentration of silver, gold, and tion (Hagelüken and Buchert 2008).
palladium in PCBs from different types of equip- Besides environmental protection and legisla-
ment. The concentration of precious metals in tive pressure, recycling is also driven by economic
PCBs is usually much higher than the concen- interests. Precious metals contribute to more than
tration of precious metals in ores. In general, 80% of the materials’ market value of obsolete
presently mined ores for the extraction of gold personal computers, despite their small quantity
and palladium contain less than 10 g/t of pre- (Streicher-Porte et al. 2005). Moreover, the high
cious metals (Hagelüken et al. 2005). Compared economic value of precious metals on the world
with the concentrations in PCBs of personal com- market as well as the limited available reserves of
puters of 250 g/t of gold and of 110 g/t of pal- precious metals (Behrendt et al. 2007) provides
ladium (table 1), the importance of recovering additional incentives to improve the recovery of
precious metals from electronics becomes obvi- precious metals from WEEE.
ous. Moreover, the extraction of precious metals
through mining is associated with negative envi-
Role of Preprocessing
ronmental impacts through significant emissions
of greenhouse gases and energy, water, and land EEE have a product life starting with prod-
usage (Ayres 1997). The environmental impacts uct manufacture—as the design of the product

Chancerel et al., Precious Metal Flows During the Preprocessing of Electronic Waste 793
 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Figure 1 Simplified recycling chain for WEEE, focusing on recovery of precious metals in the “End-of-Life”
and the “Raw materials production” phases.

determines the product characteristics and, processing stage. Preprocessing can be carried out
therefore, the demand for raw materials—and manually, automatically (shredding and sorting
ending with waste management (end of life). of materials), or with a semiautomatic process
The overall task of the end-of-life phase is to combining manual and automatic techniques.
recover the raw materials as effectively as possi- Preprocessing is divided into the following major
ble and to destroy or remove components with stages: (1) sorting, (2) selective disassembly, tar-
a high hazardous potential. Three steps can be geted on singling out hazardous or valuable com-
distinguished (figure 1): collection of WEEE, pre- ponents, and (3) upgrading, using mechanical/
processing, and disposal. The material generated physical processing and/or metallurgical process-
by preprocessing can be converted to useful ma- ing to prepare the materials for the final refining
terials by recovery facilities (for example, metal process (Cui and Forssberg 2003). Preprocessing
smelters), which belong to the phase “Raw mate- must also ensure that hazardous materials are pre-
rials production.” pared for disposal in an environmentally sound
After collection, WEEE is preprocessed to way. The chain of preprocessing, recovery, and
concentrate the substances and generate appro- disposal processes used to recover the materials
priate material streams that are sold to the end- of WEEE is called “recycling process chain.”

794 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

As first step of the recycling process chain, leagues (2000). Based on MFA data, other tools
preprocessing determines to which recovery pro- such as exergy analysis can be used to further in-
cesses the materials are fed (Chancerel and Rot- vestigate and evaluate the recycling system while
ter 2008). If a substance is sent to a recovery considering the quality of the material and energy
process that is not able to recover it, it is “lost” flows (Reuter and Van Schaik 2008; Ignatenko et
for the recycle-economy. During preprocessing of al. 2007; Meskers et al. 2008). The combination
WEEE, the so-called “grade-recovery function” of different tools enables thorough quantitative
applies (Hagelüken 2006). The recovery of a spe- evaluation of the system from different perspec-
cific material (for example metal) from an input tives and the development of strategies to im-
stream decreases with increasing purity require- prove it.
ments of that material separated into an output In the past, SFA has been used to analyze
fraction. The optimum operating conditions for flows of metals on a local, regional, or global
preprocessing are, therefore, a compromise be- scale. Table 2 presents an overview on SFA of
tween grade (quality) and recovery (quantity). precious metals with their scope of investiga-
To determine this optimum for each metal tion. The previous studies had very large system
and for the overall preprocessing of WEEE and boundaries, and the data at process level were
to minimize the losses of valuable metals, the estimated through rough assumptions. Johnson
distribution of precious metals over the outputs and colleagues (2005b), for example, quantified
has to be known. silver in WEEE based on an approximation of the
amount of silver contained in the circuit boards
of discarded electrical and electronic equipment
SFA Applied to Precious Metals
as 0.2% of the circuit boards by weight. These
and WEEE Recycling
assumptions are accompanied by high uncertain-
MFA is a systematic assessment of the flows ties as a consequence of the lack of data related
and stocks of materials within a system defined in to precious metals in WEEE and on the actual
time and space (Brunner and Rechberger 2004). recycling processes.
As such, MFA is the first step of every life cy- Some publications report on MFA and SFA
cle assessment (LCA) (Brunner and Rechberger specifically for WEEE processing. Morf and Tav-
2004; Tukker et al. 1997). The term material erna (2004) carried out a SFA in a Swiss pre-
stands both for substances and goods. An SFA, processing facility for WEEE. Based on a sample
which tracks elements or molecules within a sys- of 230 tonnes of WEEE, some base and heavy
tem, is therefore a specific kind of MFA. The metals, chlorine, bromine, potassium, polychlori-
historic background and a detailed overview on nated biphenyl, and brominated flame retardants
the concepts of MFA and SFA are available in were tracked, but precious metals were not con-
publications by Brunner and Rechberger (2004), sidered. Hischier and colleagues (2005) report
Van der Voet (1996) and Udo de Haes and col- on a MFA in Switzerland to gather data for an

Table 2 SFAs focusing on precious metals—Scope of published case studies

Metal Scope of investigation Reference
Silver Flows in Europe Lanzano et al. 2006
Flows in 64 countries, 9 world regions, planet Johnson et al. 2005b
Flows in Asia Johnson et al. 2005a
Gold Process-oriented: processes for gold recovery from Keller 2006
printed circuit boards in Bangalore, India
Platinum- group Flows in Germany Hagelüken et al. 2005
metals (PGM)
Product-oriented: platinum in fuel cell vehicles Elshkaki 2007
Platinum in Russia Babakina and Graedel 2005
PGM in the environment Ravindra et al. 2004

Chancerel et al., Precious Metal Flows During the Preprocessing of Electronic Waste 795
 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

impact assessment. Streicher-Porte and col- products or nonelectronic waste (missorted waste
leagues (2005) carried out a material flow analysis equipment).
of the recycling of personal computers in Delhi. Figure 2 presents a general flow sheet of the
The SFA of Tasaki and colleagues (2004) focused preprocess. First, during manual dismantling the
on brominated flame retardants in Japan. Bertram visible hazardous and problematic components
and colleagues (2002) investigated the flows of such as batteries, large metal sheets, and motors
copper in the European waste management sub- are manually removed from the end-of-life prod-
system, considering explicitly WEEE. Huisman ucts. Furthermore the easy-to-remove PCBs are
(2003, 2004) evaluated the environmental per- manually taken out. After a coarse preshredding,
formance of different recycling scenarios for sev- a second manual sorting takes place to remove
eral equipment types. Reuter and Van Schaik car- the remaining hazardous and disturbing compo-
ried out SFA as part of their investigations, fo- nents. Finally, the rest of the scrap is shredded
cusing on WEEE and end-of-life vehicles (Reuter and sorted automatically. Within approximately
et al. 2005; Reuter and Van Schaik 2008; Van 24 hours, the 44 output material fractions were
Schaik and Reuter 2007). None of these publi- generated. The flows of gold, palladium, silver,
cations made quantitative data on the flows of copper, iron, and aluminum were investigated, of
precious metals during automated preprocessing which the flows of the precious metals are dis-
of WEEE available. cussed in more detail.
Before the test, the input material was weighed
Methodology and qualitatively described. After processing,
all output fractions were collected separately
The investigation took place on March 31 and
and weighed. The 24 most relevant fractions
April 1, 2008, in a full-scale preprocessing facil-
regarding precious metals were sampled ac-
ity for WEEE in Germany using state-of-the-art
cording to the industry standard for WEEE.
technology. Twenty-seven tonnes of WEEE clas-
Figure 3 summarizes the applied methodology.
sified in collection group 3 (IT, telecommuni-
cations, and consumer equipment) as defined in
article 9 of the German ElektroG (2005) were
Characterization of the Input Material
processed. In practice, the treated scrap did not
only contain IT, telecommunications, and con- The qualitative description of the input by ex-
sumer equipment, but also wrongly sorted waste emplary counting of end-of-life products gave an

Figure 2 Overview of the unit processes applied in the preprocessing facility.

796 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Figure 3 Methodology for the analysis of the substance flows through the preprocessing.

approximation of the distribution of equipment Determination of the Concentration

types in the input. The amount of precious metals of Precious Metals in the Outputs
contained in the input material was estimated by
The concentration of precious metals in 15
quantifying following parameters:
output fractions, including all outputs of the auto-
mated sorting process and the manually removed
• the distribution of equipment types found PCBs, was determined at Umicore Precious Met-
in the input mix; als Refining in Hoboken, Belgium. These outputs
• the average content of PCBs in the equip- were selected because of their expected high con-
ment types taken from the WEEE charac- tent of precious metals, due either to a high con-
terization database (Chancerel and Rotter centration of precious metals or to a high mass
2007); and flow. The applied sampling and analysis methods
• the average content of precious metals in to analyze the output fractions were developed
the PCBs (table 1). The precious metals at Umicore considering the specific characteris-
concentrations in PCBs were assumed when tics of WEEE (UPMR 2005). The challenge is
results of measurements were not available to obtain representative samples from the still
in the literature. heterogeneous fractions, in which precious met-
als are often only present at parts per million
Only the precious metals present in the PCBs (ppm) level. The concentration of precious met-
(base material and components) were quantified, als was measured with inductively coupled plasma
although precious metals are known to be con- atomic emission spectrometry (ICP-AES).
tained in other parts such as connectors, contacts, Nine other output fractions such as motors,
cables, and solders (Gmünder 2007; Hagelüken computer drives, and transformers were manu-
2006), but very few quantitative data are avail- ally dismantled. The obtained material fractions
able regarding the precious metals contained in (plastics, ferrous metals, nonferrous metals, ca-
these components. bles, PCBs, and so on) were manually separated

Chancerel et al., Precious Metal Flows During the Preprocessing of Electronic Waste 797
 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

and weighed. Information about the concentra- dard deviation squared) of the result is the sum
tion of precious metals in the material fractions of the variances of the input variables. In our
was either taken from the literature or estimated. case, the variance of the calculated mass flow
By multiplying the material composition with the of precious metals Var(MFPM in Output ) is calcu-
concentrations of precious metals in the different lated by adding the variance of the weight of
materials, the content of precious metals in the the outputs Var(Weight), the variance of sam-
output fractions was calculated. pling Var(Sampling), and the variance of the
For the remaining 20 output fractions, the analysis of the precious metal concentrations
content of precious metals was estimated based Var(Analysis):
on visual observations. These output fractions
Var(MFPM in Output ) = Var(Weight)
such as wood and plastic foils had presumably
+ Var(Sampling) (1)
low concentrations of precious metals. For most
+ Var(Analysis)
of the output fractions, it was assumed that the
concentration of precious metals is zero. If the With:
outputs contained small quantities of PCB pieces, Var: variance
low concentrations up to 20 g/t of silver and MFPM in Output : mass flow of precious metals
5 g/t of gold were estimated on the basis of a The uncertainty is quantified by calculating
visual quantification of the PCBs present in the the standard deviation, which is the square root
output. of the variance.

Determination of Uncertainties Data Processing

The quantification of the uncertainties in the The substance flows were determined by mul-
measurements is based on the following consid- tiplying the substance concentrations measured
erations: in the output materials with the mass flows:

• One of the scales used to weigh the input Mmetal i in flow j = Mflow j × cmetal i in flow j (2)
and output material fractions had an uncer-
With:
tainty of ±1 kg; the other scale had an un-
Mmetal i in flow j : mass flow of substance (metal) i
certainty of ±10 kg. Because two weighings
contained in flow j
were necessary per container (one weigh-
Mflow j : mass flow of flow j
ing of the full container, one weighing of
cmetal i in flow j : concentration of substance
the empty container to determine the tare),
(metal) i in flow j
each weight had an uncertainty of ±2 or
±20 kg. For some larger fractions, several The calculations were done using the STAN
containers were used, which increased the freeware (TU Vienna 2008). STAN (short for
uncertainty. subSTance flow ANalysis) helps to perform
• To facilitate the SFA, the uncertainty for MFA (Cencic and Rechberger 2008). To get
sampling is assumed to be 10% because of clear and meaningful results, the material flows
the heterogeneity of the material. were aggregated according to the recovery pro-
• The uncertainty for determining the con- cesses they will be fed to. Sankey diagrams were
centration of precious metals is assumed to drawn for visualization of the outcomes of the
be 2% for the samples analyzed with ICP- SFA using the software E!Sankey (E-Sankey
AES, 20% for the samples that were man- 2008).
ually dismantled, and 50% for the visual
estimation.
Results
The uncertainties were calculated by apply- Figure 4 presents the results of the mass bal-
ing Gauss’s Law of error propagation. Because ance for processing one tonne of WEEE (see also
the variables are independent, the variance (stan- appendix 1 available as supplementary material

798 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Figure 4 Mass balance of the preprocessing of 1,000 kg of input WEEE.

on the Journal’s Web site). Twenty-seven wt.% Flows of Precious Metals

of the input is sorted out manually before any
Based on the overall mass flows (figure 4), the
shredding, 4 wt.% is sorted out manually after
mass flows of precious metals are calculated using
preshredding, and the majority of the mass (69
formula 2. The results are presented in figure 5
wt.%) is fed to the shredding and automated sort-
for silver, gold, and palladium.
ing process. The largest output fractions of the
The fractions “Printed circuit boards,”
process are ferrous metals (almost 33 wt.%) and
“Precious-metals rich material” and “Copper-rich
plastics (26 wt.%). Copper-rich materials such
material” are sent to facilities for recovery of the
as motors, cables, and other composite materials
precious metals. Precious metals in the “Copper-
containing a high fraction of copper represent 12
rich material” are also assumed recovered because
wt.% of the outputs. Precious metal-rich materi-
state-of-the-art metallurgical processes for cop-
als consist of the nonliberated PCBs in, for ex-
per recovery are designed to also recover precious
ample, CD drives and power supplies of personal
metals. The recovery R of a metal in the investi-
computers. Hazardous materials, wood, and other
gated preprocessing facility is therefore calculated
(precious) metal-poor materials are classified as
with the following formula:
“Other material.”
Min PCB + Min PM−rich mat. + Min Cu−rich mat.
R=
Minput
(3)
Characterization of the Input Material
With:
Table 3 presents the results of the estimated R: Recovery
quantification of precious metals in one tonne of Min PCB : mass flow of metal contained in the
input material. The input mainly contains per- output “PCB”
sonal computers, printers, keyboards, and radio Min PM−rich mat. : mass flow of metal contained
sets. It was estimated that around 67.6 grams (g)3 in the output “Precious-metals rich material”
of silver, 11.2 grams of gold, and 4.4 grams of Min Cu−rich mat. : mass flow of metal contained
palladium were in one tonne of input WEEE (ta- in the output “Copper-rich material”
ble 3). These values, however, are only used as a Minput : mass flow of metal contained in the
cross-reference. For the calculation of yields, the input
input composition was obtained by summing the By adding the flows of silver of the outputs, it
contents in the output streams. was calculated that 313.3 g of silver is present in

Chancerel et al., Precious Metal Flows During the Preprocessing of Electronic Waste 799
 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Table 3 Estimation of the content of silver (Ag), gold (Au), and palladium (Pd) in 1,000 kg of input (source:
see table 1)
Metal
concentration in the Mass of metals in the
∗
PCBs (g/t of PCB) input (g/t of input)
Amount in the PCBs in equipment
Equipment type input (kg/tonne) type (wt.%) Ag Au Pd Ag Au Pd
Computer keyboard 100 2% 700 70 30 1.40 0.14 0.06
LCD monitor 10 4% 1,300 490 99 0.52 0.20 0.04
Computer mouse 5 8% 700 70 30 0.28 0.03 0.01
DVD player 40 10% 700 100 21 2.80 0.40 0.08
Hi-fi unit 50 8% 674 31 10 2.70 0.12 0.04
Laptop 5 15% 1,000 250 110 0.75 0.19 0.08
Loudspeaker 100 2% 674 31 10 1.35 0.06 0.02
Mobile telephone 1 22% 5,540 980 285 1.22 0.22 0.06
Personal computer 210 13% 1,000 250 110 27.30 6.83 3.00
Printer, fax 230 8% 350 47 9 6.44 0.86 0.17
Radio set 100 20% 520 68 8 10.40 1.36 0.16
Telephone 10 22% 2,244 50 241 4.94 0.11 0.53
Video recorder 50 10% 674 31 10 3.37 0.16 0.05
Others 89 9% 520 68 8 4.17 0.54 0.06
Sum: 1,000 9% 67.6 11.2 4.4
∗ PCB = printed circuit boards.

one tonne of input. With 35.9 g of silver analyzed tive metals can be recovered in subsequent pro-
in the output fractions for precious metals recov- cesses. This is significantly higher than the yield
ery, the recovery for silver is only 11.5%. Almost for the precious metals and underlines that the
35% of the silver ends up in the output “ferrous recovery of small and trace amounts of (precious)
metals”; another 29% is found in plastics. metals from complex materials requires special
For gold, 40% of the 22.4 g/t of gold put in attention.
goes to the ferrous metals output. This is simi-
lar to the value found for silver. The recovery,
Comparison of Input Estimation
conversely, is much higher compared with RAg :
and Output Analysis
25.6%, which means that 74.4 wt.% is still dis-
tributed to fractions from which gold is not likely The estimated precious metal content of the
to be recovered. input is 67.6 g/t of silver, 11.2 g/t of gold and
As with gold, only 25.6% of the 7.16 grams 4.4 g/t of palladium (table 3). However, the sum
palladium is sent to a fraction where precious of precious metals in the outputs showed that
metals are recovered. The plastic output con- the input contains 313.3 g/t of silver, 22.2 g/t
tains the biggest fraction of palladium (one of gold and 7.16 g/t of palladium. The estimated
third). Filter dust and rubbish (what was swept content of precious metals of the input material
from the floor after the test) contain almost is, therefore, lower than the sum of the flows of
5% of the palladium, which shows a tendency precious metals in the outputs, especially regard-
of palladium to be released into the air during ing silver. The discrepancy can be explained by
shredding. uncertainties due to the application of literature
The SFA for copper and iron showed that data but above all by the fact that precious met-
60% of the 43.5 kg of copper per tonne of in- als are not only found in PCBs, but also in other
put and 95.6% of the 402 kg of iron per tonne parts of WEEE, such as silver in contacts, plugs, or
of input end up in fractions where the respec- solders.

800 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Figure 5 Flows of precious metals during preprocessing of one tonne of input WEEE (wt.%).

The estimation of the input, presented less, the results also show that the simpli-
in preceding sections, leads to higher un- fied approach of estimating the input quality
certainties than the calculation of the input and, specifically, the precious metal content
flow by adding the output flows. Neverthe- based on counting the equipment types is as a

Chancerel et al., Precious Metal Flows During the Preprocessing of Electronic Waste 801
 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Table 4 Concentration of precious metals measured in the outputs ‘unshredded PCB’, “preshredded PCB”
and “shredded PCB”
Concentration of metal Silver (g/t) Gold (g/t) Palladium (g/t)
Unshredded printed circuit boards 669 135 50
Preshredded printed circuit boards (<8 mm) 562 126 48
Shredded printed circuit boards (<2.5 mm) 481 48 18

first approximation at least correct to an order of due to the shredding process and the extent to
magnitude. which it is due to the intrinsic characteristics of
the PCBs. However, it is most likely that shred-
ding indeed has a negative impact on the precious
Losses of Precious Metals due to metal recovery.
Shredding of Printed Circuit Boards
The test provided results on the concentration
of precious metals in PCBs at the following points Discussion
in the preprocessing chain (table 4): (1) unshred-
Process SFA as a Tool for Process
ded PCBs from manually sorting, (2) pieces of
Operators
PCB (8 millimeter [mm]4 ) from preshredding,
and (3) pieces of PCB (2.5 mm) from shredding. Quantifying the Improvement Potentials
The SFA indicates that more size reduction From the point of view of the process oper-
(shredding) of the material seems to lower the ators, the results of the test can be qualified as
concentration of precious metals in the result- “disappointing” because only about a quarter of
ing PCB fraction (table 4). The difference in the gold and palladium and a tenth of the sil-
concentration between unshredded PCBs and ver are sent to output fractions from which pre-
preshredded PCBs (7% less precious metals in cious metals will be directly recovered. Compared
the preshredded PCBs compared with the un- with the recovery rates of major elements such as
shredded PCBs) is not as large as the difference iron, aluminum, and copper, the recovery rates
between preshredded PCBs and shredded PCBs for precious metals are very low. Most of the
(62% less precious metals in the shredded PCBs precious metals go to the most mass-relevant
compared with the preshredded PCBs). At first fractions (plastics and ferrous metals). These
glance it appears that the precious metals are fractions have relatively low concentrations of
spread over all the output fractions because of precious metals (24 g/t of gold and 8 g/t of pal-
shredding, so that the concentration of precious ladium in the plastics, 24 g/t of gold and 5 g/t of
metals in PCBs decreases. palladium in the ferrous metals; see appendix 1
The results may be influenced by the process available as supplementary material on the Jour-
characteristics, however, because the manually nal’s Web site), but the considerable mass of the
sorted PCBs (before or after preshredding) are outputs makes the flows of precious metals very
mainly large and easily accessible with a high con- relevant.
centration of precious metals (e.g., main boards These results imply that the operators of the
of personal computers). The PCBs that end up process do not get any revenue for almost three
in the shredder are frequently encased in waste quarters of the gold and the palladium (figure 5).
products such as printers, radio sets, or other small Per tonne of input WEEE, the company operating
products and are not easily accessible for man- the facility does not get any revenues for around
ual sorting. In these products, PCBs often have 16.5 g of gold and 5.3 g of palladium. At a price
low concentrations of precious metals. During the of $900 per ounce of gold and $370 per ounce of
test, no experiment could be carried out to verify palladium (average price for 2008 [USGS 2009]),
the extent to which the lower concentration of this means that a metal value of $524 for gold and
precious metals in the shredded circuit boards is almost $70 for palladium per tonne of treated

802 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

WEEE is lost. By comparison, the gross intrinsic subsequent sorting, the first and most straight-
value5 of the 43.5 kg of copper contained in one forward approach is to reduce the quantity of
tonne of input scrap is $307 (at a price of $3.20 precious metals entering the shredder. In this
per pound of copper in 2008 [USGS 2009]). manner the distribution of precious metals over
a large number of fractions during the automatic
Analysis of the Causes of Resource Losses
sorting will be minimized. This implies adjusting
The outcomes of the SFA provide data for a
the manual sorting step at the beginning of the
better understanding of the process, as indicated
process to remove most precious metal-rich mate-
by Morf and Taverna (2004). Two phenomena
rials. This requires knowledge about the location
became obvious:
of precious metals in WEEE, which is currently
1. The low concentrations of precious met- partially missing. More experimental research, es-
als in mass-relevant fractions (plastic and pecially to quantify the amounts of precious met-
ferrous metals) generate high mass flows of als contained in other parts of WEEE than PCBs,
precious metals in fractions that are not and a better communication with the manufac-
subsequently sold to processes for precious turers of EEE would provide the missing data.
metals recovery. A second approach would be adapting the
2. More shredding results in a decrease of the mechanical liberation process and the used sort-
concentration of precious metals in PCBs. ing technologies to recover more precious met-
als. This can be done through development and
These two observations are supporting the implementation of advanced preprocessing tech-
hypothesis that an unselective fine shredding is nologies addressing more specifically the recovery
causing unwanted losses of precious metals. The of precious and other trace metals.
dilution of precious metals into other outputs However, the limits of technology will not al-
happens mainly after shredding. This agrees with low achieving recoveries of 100% (Reuter et al.
the results of Hagelüken (2006), Kreibe and col- 2006). The grade-recovery curve implies that a
leagues (1996), and Veit and colleagues (2002). compromise has to be found between concentra-
Precious metals occur in PCBs as a component tion of precious metals in the outputs for precious
of complex material mixtures as they are mixed metal recovery and losses of precious metal in
with or connected to other metals in contacts, other fractions. In case of high-grade and highly
connectors, solders, and hard disk drives; with complex products such as mobile phones and dig-
ceramics in multilayer capacitors (MLCC), inte- ital cameras, this means that shredding should
grated circuits (ICs), and hybrid ceramics; or with be completely avoided (Huisman 2004), and—
plastics in PCBs tracks, interboard layers, and ICs unless manual dismantling is affordable—these
(Hagelüken 2006). The liberation of layered ma- devices are (after removal of the battery) directly
terials and of (metallic) coatings is not possible fed into appropriate metallurgical recovery pro-
in a shredder (Castro et al. 2005). The mechan- cesses for precious and special metals.
ical preprocessing disperses the precious metals
depending on the materials with which they are Strengths and Weaknesses
mixed or connected. For example, palladium is Usually, the operational decisions are mainly
often found in ceramics that are broken down driven by maximizing the recovery of mass-
to dust during shredding, which explains why al- relevant material outputs such as copper, ferrous
most 5% of the palladium ends up in the filter metals, and aluminum. For trace elements, deci-
dust. Small pieces of boards may, after shredding, sions are mainly based on empirical knowledge.
still contain a magnetic part and are pulled out SFA is the basis for a more systematic approach
because of the strong magnets in the magnetic to improve the recovery of trace metals because
separators. it is able to quantify resource losses on a mass
and monetary basis at the process level and thus
Suggestions for Technical Improvement to determine the potential for optimization. The
To reduce the losses of precious metals in pre- data have to be processed, aggregated, and pre-
processing, in particular during shredding and sented in a user-friendly manner. As emphasized

Chancerel et al., Precious Metal Flows During the Preprocessing of Electronic Waste 803
 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

by Schmidt (2008) and illustrated in this article, input for optimizing a product life cycle and its
Sankey diagrams are a potent means of visualiz- subsystems. Figure 6 shows actors in the life cycle
ing inefficiencies in the system and also are useful and how exchange of SFA-related data allows for
for a nonspecialist audience. improvement in the product life cycle.
The SFA presented in this article generated
data on the characteristics of the input waste flow Interface Preprocessing/Recovery
as an aggregation of the data on the output flows. The importance of optimization of inter-
SFA on a process level is, therefore, a possible faces between “preprocessing” and “recovery”
method for generating data on waste character- within the end-of-life phase was pointed out by
istics because these data are difficult to gather Hagelüken (2006). Through suboptimal inter-
directly due to difficulties of sampling and ana- faces between preprocessing and recovery pro-
lyzing complex materials such as WEEE. cesses, precious metals are collected in streams
This investigation is an initial SFA giving a from which they are not recovered, such as pre-
snapshot to describe the flows of precious met- cious metals in plastic, steel, or aluminum. The
als within the system defined in time and space. operators of preprocessing act under the bound-
Sampling issues in the broader sense such as the ary conditions of the existing market situation
representativeness of the WEEE input mixture and legislative framework. They can influence
and of the applied process at the day of the test as the substance flows in their process through
well as the sampling of the outputs for analysis are
key factors in properly evaluating the reliability of • technical decisions to determine the quan-
the results. The uncertainties of the sampling of tity and quality of the outputs and
the output materials could have been reduced by • management decisions on the subsequent
sampling and analyzing the input and the inter- destination of the outputs.
mediary material flows in the facility in addition
to sampling and analysis of the output fractions. In practice, the communication of data on
Additional measurements allow for description of substance flows out of preprocessing and in
the system in a redundant way and then for de- the recovery processes (see arrows 3 and 4 of
termination of the best-fitting values by using a figure 6) should develop iteratively driven by the
method for data reconciliation, for example, the expertise of preprocessing and recovery facility
method of least squares of Gauss (Brunner and operators. This should go beyond the exchange
Rechberger 2004). This could not be done in this of data to include technical discussions and tests
research for cost reasons. to develop an optimal fit between output fractions
Though the effects of modifications of the ap- in preprocessing (quantity, quality) and capabil-
plied preprocessing technology or in the charac- ities of the specific recovery plant. The commu-
teristics of the input waste were not investigated, nication between operators of preprocessing and
a periodic monitoring of metal flows (instead of recovery processes is suggested in article 7 of the
a trial-and-error approach) will allow identify- WEEE-Directive (2003), which requires that the
ing further potential improvements and observ- operators of recycling processes keep records on
ing the effects of changes in the overall process. the material flows in and out of their facility to
Both the company and society in general will calculate recovery rates.
benefit as economic viability is ensured and ma- An approach to assessing and improving the
terial resources are conserved. interface between preprocessing and recovery is
“process modeling.” The data gathered by the pro-
cess SFA can be used for designing and calibrat-
Process SFA for Improving the Product
ing models of the preprocessing process for any
Life Cycle
type of waste consumer product (Chancerel and
In addition to the application of the results to Rotter 2008; Reuter et al. 2005). By expressing
analyze and improve the preprocessing of WEEE in mathematical terms the SFA data on the be-
described in the previous section, the results of havior of precious metals in preprocessing facil-
process SFA can be aggregated and used as an ities and combining it with scientific knowledge

804 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Figure 6 Communication between the system actors to optimize the substance flows.

in the fields of mechanics, chemistry, and pro- the product does not require a very high purity
cess engineering, a process model can be built. of metals, the recovery facility can sell suitable
The characteristics of the waste and of the recy- materials. This implies that manufacturers share
cling process are parameters. By formulating as- information on their needs for raw materials with
sumptions or by carrying experimentally further the operators of the recovery facility.
SFA out, the results of the SFA can be extrapo-
lated to other process configurations (other input Interface Manufacturer/Preprocessing
characteristics and/or other process parameters). Here the idea of “producer’s responsibility”
The model can be expanded and used to simulate comes in. The Organization for Economic Co-
and improve the process. These considerations operation and Development (OECD) (2001) de-
have been included in the models for recycling of fines “extended producer responsibility” as “an
end-of-life vehicles developed by Van Schaik and environmental policy approach in which a pro-
Reuter (2007). The improvement of the system ducer’s responsibility for a product is extended to
must be multicriteria and take into account the the post-consumer stage of a product’s life cycle”
different materials (precious metals, base metals, (p. 18). Article 4 of the WEEE-Directive (2003)
plastics, hazardous substances, and so on) and the introduces the responsibility of the manufacturer
process conflicts (e.g., iron is not recovered in a in the European Union by stipulating that “Mem-
recovery process for copper). ber States shall encourage the design and produc-
tion of electrical and electronic equipment which
Interface Recovery/Manufacturer takes into account and facilitates dismantling and
Arrows 5 and 6 of figure 6 show how infor- recovery, in particular the reuse and recycling of
mation on substance flows can be exchanged to WEEE, their components and materials” (Article
optimize the interface between material recovery 4).
and manufacturing of new products. According The design of the product essentially deter-
to the demand of the manufacturer for raw ma- mines the materials used in the product, the
terials, the recovery process can be adapted. If material combinations, and the type of

Chancerel et al., Precious Metal Flows During the Preprocessing of Electronic Waste 805
 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

connections (screws, glue, and so on). For this be aware of the limitations in the data reliability
reason, the design of electrical and electronic and avoid uncontrolled error propagation.
equipment determines the maximum possible ef- For a systematic evaluation and optimization
ficiency of the end-of-life process (preprocess- of the life cycle, tools such as LCA or exergy
ing and recovery process). Design for recycling analysis can be used. To do this, the SFA data
aims at increasing both the quality of the recy- need to consider all relevant elements (not only
clates and the recovery rates (Reuter and Van precious metals) and have to be completed with
Schaik 2008). A concrete example of this is the data on process emissions, energy consumption,
accessibility of printed circuit boards. The man- transport, and so on. These aspects are usually not
ual removal of printed circuit boards would par- considered in MFA and SFA and have been left
tially prevent the losses of precious metals in out in this investigation. The combination of dif-
sidestreams. The ease of the manual removal ferent tools enables thorough quantitative eval-
depends on product design. Design-for-recycling uation of the system and developing strategies to
models, linking product design to the amount improve it, as shown by Unger and colleagues
and composition of the output fractions of me- (2008), whose research aimed at comparing eco-
chanical pretreatment and to the products of the design and environmental assessment tools for
subsequent material recovery processes, were de- EEE taking the life cycle perspective.
veloped by Van Schaik and Reuter (2004, 2007).
The communication between a manufacturer
Process SFA for Economy-wide
and an operator of preprocessing aims at exchang-
Applications in Industrial Ecology
ing data
The life cycle of EEE includes many mate-
• on the composition of the EEE, the location rial and metal cycles such as copper, aluminum,
of the substances, and their connections nickel, cadmium, lead, gold, palladium, silver, in-
(arrow 1 of figure 6) for the preprocessing dium, and plastics. Although the amount of a par-
to be adapted to the product characteristics, ticular metal in a single unit of EEE can be small,
and the number of EEE manufactured and entering
• on the requirements on product design for the end-of-life phase is very large. Therefore, the
a more efficient preprocessing (arrow 2). amount of metal used in EEE and which can
be potentially recovered becomes considerable.
In case of considering new or uncommon com- Knowledge about flows of WEEE and WEEE-
binations of substances in a product design, it is related substances within a country or region can
advisable that the manufacturer consults the op- be valuable but is often not available, as em-
erators of preprocessing and recovery facilities. phasized by various authors (Bertram et al. 2002;
The literature review done for this investiga- Reck et al. 2008).
tion showed a lack of available information, for The results of SFA from industrial trials at the
instance, regarding the exact location of precious process level provide data required for SFA and
metals in (W)EEE, which led in this investigation MFA at the macroscopic level (regions, coun-
to an underestimation of the amount of precious tries, world). In the European Union, the WEEE-
metals in the input material, especially for silver. directive (2003) requires the calculation of re-
An adequate knowledge of the characteristics of covery rates on a country level based on the data
input materials would allow anticipating their be- provided by the recycling industry (arrow 7 of
havior during processing (ex-ante approach). figure 6). These data could be completed and
compared with data from EEE manufacturers (ar-
rows 1 and 8).
Strengths and Weaknesses The next research project aims at incorporat-
SFA and MFA provide very useful data for op- ing the results of the SFA presented in this article
timization of interfaces between the subsystems in a larger investigation on the cycles of precious
of the life cycle. The uncertainties and the rep- metals by quantifying the flows of precious metals
resentativeness of SFA data have to be known to related to recycling of WEEE at local, regional,

806 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

or global level. Clearly, the questions relating to • The results of the process SFA also provide
representativeness and uncertainties of the SFA a database for larger investigations, for ex-
results have to be considered for the extrapola- ample, aiming at assessing the performance
tion of the results at the process level to the macro of end-of-life process chains regarding re-
level. Besides precious metals, quantitative re- covery of precious metals. A scheme for ex-
search on the flows of other trace metals present change of information on substance flows
in WEEE should be encouraged. Recently, more between actors was proposed.
discussions and reports on the limits of metal re-
serves, especially of the “critical metals,” have Limitations of using SFA are related to its rep-
appeared, for example through the Raw Mate- resentativeness. Static SFAs deal with a defined
rials Initiative of the European Union. In these input material within a defined time frame. The
discussions, the contribution of recycling plays an extrapolation of the data for further investiga-
important role. To properly determine “scarcity” tions must take into account uncertainties linked
and the contribution of recycling, good data are with the system boundaries. This emphasizes the
key. In this light, strategic considerations are a fact that the correct application of the method is
driving force for improving the recovery of these crucial to get reliable results.
metals. Especially regarding trace elements such as
precious metals, SFA allows a better understand-
ing of resource flows, identifying weaknesses, and
Conclusions and
developing strategies to improve the system. As
Recommendations
far as possible, the method applied to carry out the
As a starting point for the conclusion, it has SFA should be standardized to make the results
to be noted that the investigation was carried out comparable.
at a state-of-the-art facility fulfilling all European
legal requirements of the WEEE-directive (2003)
Acknowledgements
and the German ElektroG (2005) and in partic-
ular achieving the required mass-based rates of The authors thank the Deutsche Bundess-
recovery. The recovery of precious metals is not tiftung Umwelt (DBU) for providing financial
driven by those legal requirements but by eco- support through the Scholarship Programme, the
nomic interest of stakeholders. Therefore, SFA is operators of the investigated preprocessing facil-
a useful tool for optimizing material and product ity for their valuable support, and Umicore Pre-
life cycles beyond the regulatory framework in cious Metals Refining for executing sampling and
the interest of single companies and the society: analysis.

• At the process level, the SFA in general al-

Notes
lows monitoring and assessing the process
to identify improvement potentials. This 1. All tons in this article are metric tons (as denoted
investigation showed that after preprocess- by “tonnes”). One tonne (t) = 103 kilograms (kg,
ing, despite the high recovery rates for mass- SI) ≈ 1.102 short tons.
2. One kilogram (kg, SI) ≈ 2.204 pounds (lb).
relevant elements such as ferrous and cop-
3. One gram (g) = 10−3 kilograms (kg, SI) ≈ 0.035
per, only a quarter of gold and palladium
ounces (oz).
ends up in outputs from which precious 4. One millimeter (mm) = 10−3 meters (m, SI) ≈
metals may be recovered. This and a de- 0.039 inches.
tailed insight in the process are relevant 5. Gross intrinsic value is the economic value of the
information for technical adaptations. A metal without considering the costs of recovering it.
possible solution to reduce the losses of pre-
cious metals is to manually remove the rele-
vant materials, for instance PCBs, to avoid
References
shredding them and dispersing part of the Angerer, G., K. Bätcher, and P. Bars. 1993. Verw-
metal content. ertung von Elektronikschrott – Stand der Technik,

Chancerel et al., Precious Metal Flows During the Preprocessing of Electronic Waste 807
 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Forschungs- und Technologiebedarf [Recycling of A Thematic strategy on the prevention and recycling
electronic scrap – State-of-the-art of the need for of waste. COM(2005) 666. Brussels: Commission
technique, research and technology]. Bundesmin- of the European Communities.
isterium für Forschung und Technologie. Berlin: Cui J. and E. Forssberg. 2003. Mechanical recycling
Erich Schmidt Verlag. of waste electric and electronic equipment: A re-
Art, S. 2008. Personal communication with S. Art, view. Journal of Hazardous Materials 99(3): 243–
Sales and Customer Service. Umicore Precious 263.
Metals Refining, Hoboken, Belgium, 21 August ElektroG. 2005. Act Governing the Sale, Return and En-
2008. vironmentally Sound Disposal of Electrical and Elec-
Ayres, R.U. 1997. Metals recycling: Economic and tronic Equipment of March 16 2005 (Electrical and
environmental implications. Resources, Conserva- Electronic Equipment Act, in German: Gesetz
tion and Recycling 21(3): 145–173. über das Inverkehrbringen, die Rücknahme und
Babakina, O. and T.E. Graedel. 2005. The industrial die umweltverträgliche Entsorgung von Elektro-
platinum cycle for Russia: A case study of materi- und Elektronikgeräten, ElektroG).
als accounting. Working Paper Number 8, New Elshkaki, A. 2007. Systems analysis of stock buffering:
Haven, CT, USA: Yale School of Forestry and Development of a dynamic substance flow-stock
Environmental Studies. model for the identification and estimation of fu-
Behrendt, S., M. Scharp, L. Erdmann, W. Kahlenborn, ture resources, waste streams and emissions. Ph.D.
M. Feil, C. Dereje, R. Bleischwitz and R. Delzeit. Thesis, Leiden University, Leiden, The Nether-
2007. Rare metals – Measures and concepts for the so- lands.
lution of the problem of conflict-aggravating raw ma- Ernst, T., R. Popp, M. Wolf and R. van Eldik. 2003.
terial extraction – The example of coltan. Research Analysis of eco-relevant elements and noble met-
Report 363 01 124, Text 23/07. Dessau, Germany: als in printed wiring boards using AAS, ICP-AES
Federal Environmental Agency (Umweltbunde- and EDXRF. Analytical and Bioanalytical Chemistry
samt). 375(6): 805–814.
Bertram, M., T.E. Graedel, H. Rechberger and S. E-Sankey. 2008. www.e-sankey.com. Accessed August
Spatari. 2002. The contemporary European cop- 2008.
per cycle: Waste management subsystem. Ecolog- Gmünder, S. 2007. Recycling – From Waste to Re-
ical Economics 42(1/2): 43–57. source – Assessment of optimal manual disman-
Brunner, P.H. and H. Rechberger. 2004. Practical hand- tling depth of a desktop PC in China based
book of material flow analysis. New York: Lewis on eco-efficiency calculations. Diploma thesis,
Publishers. Eidgenössische Technische Hochschule Zürich,
Castro, M.B., J.A.M. Remmerswaal, J.C. Brezet, A. Zürich.
van Schaik and M.A. Reuter. 2005. A simula- Hagelüken, C. and M. Buchert. 2008. The mine
tion model of the comminution-liberation of re- above ground – opportunities & challenges to
cycling systems – relationships between product recover scarce and valuable metals from EOL
design and the liberation of materials during re- electronic devices. In Proceedings of 7th Inter-
cycling. International Journal of Mineral Processing national Electronics Recycling Congress, 16–18
75(3–4): 255–281. January, Salzburg, Austria.
Cencic, O. and H. Rechberger. 2008. Material flow Hagelüken, C. 2006. Improving metal returns and
analysis with software STAN. Journal of Environ- eco-efficiency in electronics recycling – a holis-
mental Engineering Management 18(1): 3–7. tic approach for interface optimisation between
Chancerel, P. and S. Rotter. 2007. Recycling oriented pre-processing and integrated metals smelting
characterisation of WEEE. In Proceedings of Eco- and refining. In Proceedings of the IEEE In-
X conference, 9–11 May, Vienna: 205–212. ternational Symposium on Electronics & the
Chancerel, P. and S. Rotter. 2008. Stoffstromanalyse Environment, 8–11 May, San Francisco: 218–
und Modellierung von mechanischen Aufbere- 223.
itungsprozessen für Elektro- und Elektronikalt- Hagelüken, C., M. Buchert and H. Stahl. 2005. Ma-
geräte [Substance flow analysis and modeling of terials flow of platinum group metals. Umicore AG
mechanical pre-processing for waste electrical and & Co. KG and Öko-Institut e.V. London: GFMS
electronic equipment]. In Proceedings of Abfall- Limited.
forschungstage. Göttingen, Germany: Cuvillier. Hischier, R., P. Wäger and J. Gauglhofer. 2005. Does
Commission of the European Communities (COM). WEEE recycling make sense from an environmen-
2005. Taking sustainable use of resources forward: tal perspective? The environmental impacts of the

808 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Swiss take-back and recycling systems for waste European silver cycle. Resources, Conservation and
electrical and electronic equipment (WEEE). En- Recycling 46(1): 27–43.
vironmental Impact Assessment Review 25(5): 525– Legarth, J.B., L. Altmg, B. Danzer, D. Tartler, K.
539. Brodersen, H. Scheller, and K. Feldmann. 1995.
Huisman, J. 2003. The QWERTY/EE concept, quan- New strategy in the recycling of printed circuit.
tifying recyclability and eco-efficiency for end-of- Circuit World 21(3): 10–15.
life treatment of consumer electronic products. Meskers, C.E.M., A. Kvithyld, M.A. Reuter, and T.A.
Ph.D. thesis, Delft University of Technology, Engh. 2008. A fundamental metric for metal recy-
Delft, Netherlands. cling applied to coated magnesium. Metallurgical
Huisman, J. 2004. QWERTY and eco-efficiency analy- and materials transactions 39(3): 500–517.
sis on cellular phone treatment in Sweden. The eco- Morf, L. and R. Taverna. 2004. Metallische und nicht-
efficiency of the direct smelter route versus mandatory metallische Stoffe im Elektronikschrott – Stoffflus-
disassembly of printed circuit boards. Stockholm, sanalyse [Metallic and non-metallic substances
Sweden: El-Kretsen. in electronic scrap – substance flow analysis].
Huisman, J., F. Magalini, R. Kuehr, C. Maurer, C. Del- Schriftenreihe Umwelt Nr. 374. Bern: Bundesamt
gado, E. Artim and A.L.N. Stevels. 2007. 2008 für Umwelt, Wald und Landschaft BUWAL.
review of directive 2002/96 on waste electrical and Organisation for Economic Co-operation and Develop-
electronic equipment (WEEE). Bonn, Germany: ment (OECD). 2001. Extended Producer Respon-
United Nations University. sibility: A Guidance Manual for Governments.
Ignatenko, O., A. van Schaik and M.A. Reuter. 2007. Environment & Sustainable Development 2001(5):
Exergy as tool for evaluation of the resource effi- 1–159.
ciency of recycling systems. Minerals Engineering Ravindra, K., L. Bencs, and R. Van Grieken. 2004.
20(9): 862–874. Review Platinum group elements in the environ-
Johnson, J., M. Bertram, K. Henderson, J. Jirikowic, and ment and their health risk. Science of the Total
T.E. Graedel. 2005a. The contemporary Asian Environment 318(1–3): 1–43.
silver cycle: One-year stocks and flows. Journal of Reck, B., D.B. Müller, K. Rostkowski, and T.E.
Material Cycles and Waste Management 7(2): 93– Graedel. 2008. The anthropogenic nickel cycle:
103. Insights into use, trade, and recycling. Article and
Johnson, J., J. Jirikowic, M. Bertram, D. van Beers, R.B. Supporting Information. Environmental Science &
Gordon, K. Henderson, R. J. Klee, T. Lanzano, Technology 42(9): 3394–3400.
R. J. Lifset, L. Oetjen, and T.E. Graedel. 2005b. Reuter, M.A., U.M.J. Boin, A. van Schaik, E. Verhoef,
Contemporary anthropogenic silver cycle: A mul- K. Heiskanen, Y. Yang, and G. Georgalli. 2005.
tilevel analysis. Environmental Science & Technol- The metrics of material and metal ecology: harmoniz-
ogy 39(12): 4655–4665. ing the resource, technology and environmental cycles.
Keller, M. 2006. Assessment of gold recovery pro- Amsterdam: Elsevier Science.
cesses in Bangalore, India and evaluation of Reuter, M.A., A. van Schaik, O. Ignatenko, and G.J. de
an alternative recycling path for printed wiring Haan. 2006. Fundamental limits for the recycling
boards. Diploma thesis, Eidgenössische Technis- of end-of-life vehicles. Minerals Engineering 19(5):
che Hochschule Zürich, Zürich. 433–449.
Kramer, T. 1994. Mechanische Aufbereitung von Reuter, M.A. and A. van Schaik. 2008. Thermody-
Leiterplatten [Mechanical processing of printed namic metrics for measuring the “sustainability”
circuit boards]. In Verwertung von Elektro- und of design for recycling. JOM Journal of the Miner-
Elektronikgeräten [Recycling of electrical and elec- als, Metals and Materials Society 60(8): 39–46.
tronic appliances]. Materialien Nr. 3., proceed- Rotter, S. and A. Janz. 2006. Charakterisierung elek-
ings of 6. Aachener Kolloquium Abfallwirtschaft. trischer und elektronischer Altgeräte (EAG)–
Essen, Germany: Landesumweltamt Nordrhein- 1. Teil: Mengenprognosen und und Zusam-
Westfalen. mensetzung von Kleingeräten [Characterisation
Kreibe, S., J. Wagner and K. Rommel. 1996. Verwertung of small Waste Electric and Electronic Equipment
und Beseitigung von Leiterplattenschrott [Recovery (WEEE) – Part 1: Definition of strategies for sepa-
and disposal of waste printed circuit boards]. BIfA- rate collection and recycling]. Müll und Abfall (7):
Text Nr. 7. Böblingen, Germany: Bayerisches In- 365–373.
stitut für Abfallforschung. Schmidt, M. 2008. The Sankey diagram in energy and
Lanzano, T., M. Bertram, M. De Paloa, C. Wagner, K. material flow management. Journal of Industrial
Zyla, and T.E. Graedel. 2006. The contemporary Ecology 12(2): 173–185.

Chancerel et al., Precious Metal Flows During the Preprocessing of Electronic Waste 809
 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Streicher-Porte, M., R. Widmer, A. Jain, H.P. Bader, of substance flows through the economy and the
R. Scheidegger, and S. Kytzia. 2005. Key drivers of environment of a region. Ph.D. Thesis, Leiden
the e-scrap recycling system: Assessing and mod- University, Leiden, The Netherlands.
elling e-scrap processing in the informal sector Van Schaik, A. and M.A. Reuter. 2004. The optimiza-
in Delhi. Environmental Impact Assessment Review tion of end-of-life vehicle recycling in the Euro-
25(5): 472–491. pean Union. JOM Journal of the Minerals, Metals
Tasaki, T., T. Takasuga, M. Osako and S. Sakai. 2004. and Materials Society 56(8): 39–47.
Substance flow analysis of brominated flame re- Van Schaik, A. and M.A. Reuter. 2007. The use of fuzzy
tardants and related compounds in waste TV rule models to link automotive design to recycling
sets in Japan. Waste Management 24(6): 571– rate calculation. Minerals Engineering 20(9): 875–
580. 890.
Tukker, A., R. Kleijn, L. Oers and E. Smeets. 1997. Veit, H., C. de Pereira and A. Bernardes. 2002. Using
Combining SFA and LCA. Journal of Industrial mechanical processing in recycling printed wiring
Ecology 1(4): 93–116. boards. JOM Journal of the Minerals, Metals and
TU Vienna. 2008. Software for Substance Flow Materials Society 54(6): 45–47.
Analysis STAN. www.iwa.tuwien.ac.at/iwa226_ WEEE-Directive. 2003. Directive 2002/96/EC of the Eu-
english/stan.html. Accessed August 2008. ropean parliament and of the council of 27 January
Udo de Haes, H., R. Heijungs, G. Huppes, E. van der 2003 on waste electrical and electronic equipment.
Voet and J. Hettelingh. 2000. Full mode and at- Widmer, R., H. Oswald-Krapf, D. Sinha-Khetriwal, M.
tribution mode in environmental analysis. Journal Schnellmann, and H. Böni. 2005. Global perspec-
of Industrial Ecology 4(1): 45–56. tives on e-scrap. Environmental Impact Assessment
Umicore Precious Metals Refining (UPMR). 2005. Ex- Review 25(5): 436–458.
ploring Sampling and Assaying. Hoboken, Belgium.
Unger, N., F. Schneider and S. Salhofer. 2008. A re-
view of ecodesign and environmental assessment About the Authors
tools and their appropriateness for electrical and
Perrine Chancerel is a Ph.D. student at Berlin
electronic equipment. Progress in Industrial Ecol-
ogy 5(1–2): 13–29. University of Technology, Germany. Christina
United States Geological Survey (USGS). 2009. Min- Meskers and Christian Hagelüken are responsi-
eral Commodity Summary 2009. http://minerals. ble for business development and market research
usgs.gov/minerals/pubs/mcs/. Accessed May 2009. at Umicore Precious Metals Refining in Hobo-
Van der Voet, E. 1996. Substance from cradle to grave: ken, Belgium. Vera Susanne Rotter is a professor
development of a methodology for the analysis at Berlin University of Technology, Germany.

Supplementary Material
Additional Supplementary Material may be found in the online version of this article:
Supplement S1. This supplement contains an appendix with a table displaying the mass of the
output fractions and concentration of silver, gold, palladium, copper, aluminum, and iron in
WEEE.
Please note: Wiley-Blackwell is not responsible for the content or functionality of any supple-
mentary materials supplied by the authors. Any queries (other than missing material) should be
directed to the corresponding author for the article.

810 Journal of Industrial Ecology