Vmgsim

Modelling and optimization of a
Biomass-to-Liquid process using

VMGSim
Riccardo Costanzi
Chemical Engineering
Submission date: Januar 2016
Supervisor: Magne Hillestad, IKP
Norwegian University of Science and Technology

Department of Chemical Engineering
Contents
1 Introduction 3
1.1 Biomass and its products . . . . . . . . . . . . . . . . . . . . . . . 4
1.1.1 Products of Biomass . . . . . . . . . . . . . . . . . . . . . 5
1.1.2 Biomass conversion . . . . . . . . . . . . . . . . . . . . . . 6
2 Gasication 8
2.1 Modelling of pretreatment of the biomass . . . . . . . . . . . . . . 8
2.2 Gasication Process . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2.1 Reactor . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.3 Modeling of Gasier . . . . . . . . . . . . . . . . . . . . . . . . . 15
3 Concentration and purication of Syngas 19

3.1 Concentration: Water-gas-shift Reactor . . . . . . . . . . . . . . . 19
3.1.1 Modeling Water-Gas-Shift Reactor . . . . . . . . . . . . . 21
3.2 Modeling purication of Syngas . . . . . . . . . . . . . . . . . . . 28
4 Fischer-Tropsch Reactor 31
4.1 Fischer-Tropsch Process . . . . . . . . . . . . . . . . . . . . . . . 31
4.1.1 Chemical concepts . . . . . . . . . . . . . . . . . . . . . . 31
4.2 Modeling the Fischer-Tropsch Product Distribution and Model Im-
plementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4.3 Modeling FT Reactor . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.3.1 Optimization . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.3.2 Rigorous Approach . . . . . . . . . . . . . . . . . . . . . . 56
1
CONTENTS 2
5 Conclusion 60
Chapter 1
Introduction
In the last 10 years the global energy request has increased by approximately 26%
and is predicted to grow by 47% by 2035. nowadays this request is primarily sup-
plied by fossil resources i.e. natural gas, oil and coal. While considerable reserves
of coal are available, it is partly responsible for CO2 gas emissions and it has to
be transfered from where it is abundant to where required for power production.
It is supposed that utilization of coal will rise by an average of 1.5% per year
from 2015 to 2035. Natural gas is plentiful and its providing has recently grown
owing to technology allowing access to gas in shale beds. Coal bed methane is
an another fount of natural gas. All these elements mantain the global market
well provided and the use is supposed to rise by 1.6% from 2015 to 2035. Since
the utilization is quickly catching up with the production and accessibility, the
share of liquid fuel in world energy demand is projected to decrease by 2035.
Against these valutations the use of liquid fuels in transportation sector will rise.
The transportation sector depends almost exclusively on the hydrocarbon fuels as
energy fount due to its high volumetric energy density. In order to protract peak
oil, increased request in transportation sector must be supplied by alternative
sources. Except fossil crude oil, hydrocarbon fuels can be generated from photo,
bio-chemical and thermochemical way. These alternative ways use methane, coal,
sun and biomass (alternative resource) as feedstock. Liquid fuels can be generated
from natural gas by rst producing syngas and then transforming the syngas to liq-
3
CHAPTER 1. INTRODUCTION 4
uid fuels. Natural gas is a plentiful fossil fuel and thus provides to greenhouse gas
(GHG) emissions. There is an incentive to produce renewable fuels from biomass.
Liquid fuels can be produce from biomass by rst transforming it to syngas via
gasication. Biomass is a valuable renewable feedstock, its global request is pre-
dicted to rise over the next 20 years due to bring down the dependence on fossil
liquid fuel. during the industrial revolution after the introduction of coal, The
search for renewable sustainable energy resources has given biomass a importance
it had lost. The share of biomass in meeting current world's primary energy mix
is at a small level of 10%, but given the increasing preoccupation about global
warming and sustainability, this share is inclined to grow. The most prevalent
use of biomass for energy is direct combustion, followed by pyrolysis,gasication
and carbonization. This thesis deals with gasication process. [1] Gasication
is a chemical process that transform carbonaceous materials like biomass into
functional convenient gaseous fuels or chemical feedstock. Pyrolysis, partial oxi-
dation,and hydrogenation are connected processes. Combustion also transforms
carbonaceous materials into product gases but with some important distinctions.
For example, the outlet gas from combustion does not have any convenient heat-
ing value, but the outlet gas from gasication does. Gasication wraps energy
into chemical bonds in the product while combustion liberates it. The process
occurs in reducing (oxygen-decient) environment releasing heat, while combus-
tion occurs in a oxidizing environment releasing heat. The object of gasication
or pyrolysis is not just energy conversion but another important application is the
production of chemical feedstock.
1.1 Biomass and its products

Biomass is made of alive species like animals and plants; anything that is now
alive or was alive a short time ago produces biomass. It is formed as soon as
a seed germinates or an organism is born. Unlike fossil fuels, biomass doesn't
request millions of years to evolve and it can be reproduced, and for that reason,
it is considered renewable. Every year, a huge quantity of biomass is produced

through photosynthesis by absorbing CO2 from the atmosphere. When it burns,
it liberates the carbon dioxide that the plants had assimilated from atmosphere
only recently (a few years to a few hours). Thus, the combustion of biomass
doesn't give a contribution to the world's carbon dioxide levels. This liberation
also occurs for fossil fuels. So,by comparison,there is no addition to the CO2
level by the burning of biomass and one can consider biomass a carbon-neutral'.
Only 5% of the huge quantity of biomass in the world can be potentially use to
produce energy. Even this amount is big enough to furnish about 26% of world's
energy waste which is equivalent to 6 billion tons of oil. The principal types of
accumulated biomass are sugar (cereal),cellulosic (non cereal), and starch.
1.1.1 Products of Biomass
There are three types of mainly fuels that could be generated from biomass and
are as follows:
Liquid fuels (ethanol, biodiesel, methanol, vegetable oil, and pyrolysis oil)
Gaseous fuels (biogas (CH4 , CO2 ), producer gas ( CO, H2 , CH4 , CO2 ), syn-
gas, substitute natural gas
Solid fuels (charcoal, torreed biomass, biocoke, biochar)
These biomass products nd application in following four main types of industries:
1. Chemical industries for production of methanol, fertilizer, synthetic ber
and other chemicals.
2. Energy industries for generation of heat and electricity.
3. Transportation industries for production of gasoline and diesel.
4. Environmental industries for capture of CO2 and other pollutants.
The application of ethanol and biodiesel as trasport fuels lessens the emission of
CO2 per unit of energy production. It also reduces our reliance on fossil fuels.
Thus, the energy produced from biomass is not only renewable but also purify
from the point of view of greenhouse gas (GHG) emission, and so it can become
a crucial point in the global energy scene.
1.1.2 Biomass conversion
Bulkiness, low energy density, and awkward form of biomass are the main ob-
stacles to a rapid conversion from fossil to biomass fuels. Unlike gas or liquid,
biomass cannot be manipulated, accumulated, or transported easily. This gives an
important reason for the transformation of solid biomass into liquid and gaseous
fuels, which have a greater energy density and can be manipulated and stored
with relative ease. This transition can be reached through one of two important
ways (Figure 1.1): biochemical conversion (fermentation) and thermochemical
conversion (pyrolysis, gasication). Thedisadvantage of bulkiness and other aws
of solid biomass are exceeded some extent through the production of more useful
puree solid fuel through carbonization and torrefaction. Biochemical conversion
is maybe the oldest process of biomass gasication. For local energy demand,
India and China produced methane gas by anaerobic microbial digestion of ani-
mal wastes. nowadays the large amount of ethanol used for automotive fuels is
produced from corn using fermentation. Thermochemical conversion of biomass
into gases has developed much later. Large-scale consumption of small biomass
gasiers started during the Second World War, when more than a million units
were operating. [2] .
.
Figure 1.1: Dierent options for conversion of biomass
In this thesis the process from biomass to liquid (BtL) is examined. In the
BtL process, biomass, such as woodchips and straw stalk, is rstly converted into
biomass-derived syngas (bio-syngas) by gasication. Then, a cleaning process
is applied to remove impurities from the bio-syngas to produce clean bio-syngas
which meets the FischerTropsch synthesis requirements. Cleaned bio-syngas is
then conducted into a FischerTropsch catalytic reactor to produce green gasoline,
diesel and other clean biofuels.
This review will analyze the four main steps of BtL process:
1. Gasication of Biomass
2. Water Gas Shift Reactor
3. Fisher Tropsch Reactor
The main object is to explain the four main steps of the process, that can be
processed with a new simulator called VMGSim.
The aim is to nd the optimize variables to maximize the production.

Chapter 2
Gasication
2.1 Modelling of pretreatment of the biomass

First of all, the kind of biomass used is described.
Like in most other simulators, the thermodynamic model must be determined
rst of all in the VMGSim; without this, it is impossible to calculate any equilib-
rium. After adding the needed components for the reaction, the property package
has to opened and one of the models divided into chemical system categories is
selected. The termodynamic model that represents in better way the rst part of
the project is Gasication_2010, in fact the rst main operation for BtL is the
gasication.
The second step is to characterize the biomass: the simulator oers the possi-
bility to create an hypothetical compound from existing components.
Dry raw hardwood is created using three main inputs:
Molecular weight: in this case it is considered, like default, 10000.
Lower Heating Value: for wood the value is 18,5 MJ/Kg
Chemical formula and composition
8
CHAPTER 2. GASIFICATION 9
The elementary wood's composition has been added and
it is 48.2% carbon, 45.5% oxygen and 6.3% hydrogen. The
composition can change a little bit but approximately this
is correct for all kinds of wood. The hypotetical compound Figure 2.1: charac-
has been created and now the rst stream can be inserted teristic biomass
in the owsheet: Biomass.
This stream is characterized by 50% mass fraction of DryRawHardwood and
50% mass fraction for water. In this stream there aresome suldes that will be
subsequently absorbed by a chemical agent.
In the real process there are a lots of steps before the rst main unit operation
of Gasication. In fact the biomass chips have to be subjected to a process of
torrefaction, then the chips are crushed inside the grinder and nally adding water
is indispensable to create a slurry mixture, that can be putted into the gasier.
Torrefaction is a thermochemical process in an inert or limited oxygen envi-
ronment where biomass is slowly heated into a specied temperature range and
retained there for a stipulated time such that it results in near complete degra-
dation of its hemicellulose content while maximizing mass and energy yield of
°
solid product. A typical temperature range for this process is between 200 C and
°
300 C. Though other ranges have been suggested, none exceeds the maximum
°
temperature of 300 C. Torrefaction above this temperature would result in ex-
tensive devolatilization and carbonization of the polymers. Both processes are
undesirable for torrefaction. Also, the loss of lignin in biomass is very high above
°
300 C. This loss could make it dicult to form pellets from torreed products.
Furthermore, fast thermal cracking of cellulose causing tar formation starts at
°
temperature 300-320 C. These reasons x the upper limit of torrefaction temper-
°
ature as 300 C. [3]
A simple illustration of the torrefaction process is shown in Figure 2.2. It
shows how the mass and energy content of biomass changes as it is converted into
a torreed product.
Figure 2.2: Mass and energy changes of a feed undergoing torrefaction
The heating medium here is represented by a ame but it also could be a hot
substance, dry or wet. In wet torrefaction, the biomass is subjected to heating in
hot compressed water. Dry torrefaction involves heating either by a hot inert gas
or by indirect heating. The dry process is the accepted method for commercial
torrefaction. The mass loss of the wood at this stage is about 20%.
Heating stages during the torrefaction are:
1. Predrying
2. Drying
3. Postdrying Heating
4. Torrefaction
5. Cooling
In the simulator it is not possible to study the torrefaction in detail. Never-
theless certain requirements need to be fullled, like charge at the temperature of
°
100 C that has evaporated losing water and a little bit of wood and it is pumpable
until 20 bar, necessary for the gasier.
Therefore a series of operations can be approximated using just one unit oper-
ation called Dryer. It has a physical function of separating the most of water from
the rest of biomass. So the change from a solid biomass to a liquid and pumpable
biomass is included and approximated in a single step.

After the pump, the stream has the following
composition: 91% of Dry raw Hardwood and 9% of
water.
2.2 Gasication Process

Figure 2.3: pretreatment
Gasication is convenient because it can transform
biomass
solid or liquid feedstock into gaseous fuel or chemi-
cal feedstock that can be burned to produce energy
or used for create output of value-added chemicals.
Gasication and combustion are two closely con-
nected thermochemical processes, but there is an important dierence between
them. Gasicationwraps energy into chemical bonds in the outlet gas; combustion
breaks those bonds to liberate the energy. The gasication process adds hydrogen
and strips carbon away from the hydrocarbon feedstock to produce gases with a
higher hydrogen-to-carbon (H/C) ratio. In contrast combustion oxidizes the hy-
drogen and carbon into water and carbon dioxide. A typical biomass gasication
process may contain the following steps: Drying Thermal decomposition or py-
rolysis, partial combustion of some gases, vapors, and char Gasication of decom-
posed products. Gasication needs a gasifying medium like steam, air, or oxygen
to reorganize the molecular structure of the feedstock in order to transform the
solid feedstock into gases or liquids; it can also add hydrogen to the product. The
use of a medium is essential for the gasication process, which is not the same for
pyrolysis or torrefaction. Gasifying medium (also called agent) reacts with solid
carbon and heavier hydrocarbons to transform them into low-molecular-weight
gases like CO and H2 . The main gasifying agents used for gasication are:
Oxygen
Steam
Air
For this process, oxygen has been used. Oxygen is a popular gasifying agent
although it mostly serves for the combustion or the partial gasication in a gasier.
It may be provided to a gasier both in pure form or via air. The heating value
and the composition of the gas created in a gasier are strong characteristics of
the nature and amount of the gasifying agent used. The conversion path proceed
to the oxygen node. Its products include CO for low amount of oxygen and
CO2 for high amount of oxygen. When the amount of oxygen overcomes a certain
(stoichiometric) level, the processchanges from gasication to combustion, and the
product is ue gas instead of fuel gas. The ue gas or the combustion product
doesn't include remaing residual heating value.[4] The following is a delineation
of some of those reactions with carbon, carbon dioxide, hydrogen, steam, and
methane.
C + H2 O ←→ CO + H2 4H = 131KJ/mol (1)
CH4 + H2 O ←→ CO + 3H2 4H = 206KJ/mol (2)
C + O2 ←→ CO2 4H = −394KJ/mol (3)
CO + H2 O ←→ CO2 + H2 4H = −41KJ/mol (4)
C + CO2 ←→ 2CO 4H = 172KJ/mol (5)
The total reaction are generally endothermic but there are some exothermic
reactions because of the presence of partial combustion and reactions like water-
gas-shift reaction. The rate of the consumption of char (comprising mainly of
carbon) is inuenced primarily by its reactivity and the reaction potential of the
gasifying medium, in this case oxygen. In fact, amongst gasication medium,

oxygen is most active, followed by steam and carbon dioxide.
The reaction (1) is known as Water-gas-shift reaction and it is perhaps the
most important gasication reaction. The big amount of hydrogen has a
strong inhibiting eect on the gasicartion; in fact the best solution is to
continuous removal of hydrogen from the reaction site. For example, 30%
hydrogen in the gasication atmosphere can reduce the gasication rate by
a factor as high as 15.
The reaction (4) is called Shift reaction. The rst feature is that it is a gas-
phase reaction and it is less exothermic than the combustion. Unlike the
above reactions, shift reaction takes place between an intermediate product
of the gasication reaction and steam. At the expense of CO, this reaction
increases the H2 amount of the gasication product. The heat of enthalpy is

KJ
−41 mol , so, even if the yield decreases with the temperature, the equilibrium
is not much inuenced by the temperature. Anyway, the changing of the
°
pressure has no importance on the process. Above 1000 C, the shift reaction
rapidly reaches equilibrium also without the usage of catalyst, but at a
lower temperature, it needs heterogeneous catalysts. As every exothermic
reaction, the diculty is to nd the compromise between the thermodynamic
and the kinetic. With increasing temperature, the yield decreases, but the
reaction rate increases. The optimum yield is obtained at about 225 °C.
The reaction (5) is the gasication of char in carbon dioxide and it is pop-
ularly known as the Boudouard reaction. The rate of the char gasication
reaction inCO2 is insignicant below 1000 K.
2.2.1 Reactor
The succession of gasication reactions is based on some extent on the type of
gas solid contacting reactors used. There are dierent kinds of reactor: Moving
bed, Fluidized bed, Entrained ow reactor. In this thesis a Fluidized bed reactor
has been considered. The uidized-bed gasier includes nonfuel granular solids
(bed solids) that work as a heat carrier and mixer. In a bubbling uidized bed,
the fuel fed from either the top or the sides blends quite fast over the entire body
of the uid bed. The gasifying agent (air, oxygen, steam, or their mixture) also
is used as the uidizing gas and so it is fed through the bottom of the reactor.
Although the bed solids are well amalgameted, the uidizing gas keeps usually
the plug-ow mode, getting in from the bottom and leaving from the top. In the
moment it enters the bottom of the bed, the oxygen reacts quickly in exothermic
reactions with char mixed with bed materials. The bed materials directly dissolve
the heat produced by these reactions to the whole uidized bed. The amount
of heat produced close to the bottom grid is based on the oxygen amount of the
uidizing gas and the amount of char that comes in contact with it. The tem-
perature of this region is based on how strongly the bed solids dissolve the heat
from the combustion zone. The following gasication reactions occur further up
as the gas increases. The bubbles of the uidized bed can be used as the main
channel to the top. They contain relatively no solids. While they support in
mixing, the bubbles can also consent gas to bypass the solids without taking part
in the gasication reactions. The pyrolysis products coming in contact with the
hot solids collapsed into noncondensable gases. If they escape the bed and go up
to the cooler freeboard, tar and char are formed. A bubbling uidized bed cannot
obtain the entire char conversion because of the back-mixing of solids. The high
level of solid mixing supports a bubbling uidized-bed gasier to obtain the tem-
perature uniformity, but due to the intimate mixing of completely gasied and
partially gasied fuel particles, any solids leaving the bed carry some partially
gasied char. Char particles dragged from a bubbling bed can also contribute to
the waste in a gasier. The other important question with uidized-bed gasiers
is the slow diusion of oxygen from the bubbles to the emulsion phase. This
favors the combustion reaction in the bubble phase, which reduces gasication
eciency. In a circulating uidized bed (CFB), solids ow around a loop that
is characterized by severe mixing and longer solid residence time within its solid
circulation loop. The absence of any bubbles avoids the gas-bypassing problem
of bubbling uidized beds. Fluidized-bed gasiers typically operate in the tem-

°
perature range of 800-1000 C to avoid ash agglomeration. This is satisfactory for
reactive fuels such as biomass, municipal solid waste (MSW), and lignite. Since
uidized-bed gasiers operate at relatively low temperatures, most high-ash fuels,
depending on ash chemistry, can be gasied without the problem of ash sintering
and agglomeration.[5]
This gure describes the model gasier and its temperature trend.
Figure 2.4: Trend of temperature inside the gasier
2.3 Modeling of Gasier

As indicated in Figure 2.5, there are two streams entering into the gasier. The
rst stream is pretreated biomass, this has the following composition: 91% Dry
Raw Hardwood and 9% water, so the most of water is evaporated into the dryer.
The second stream is composed by pure oxygen that represents the gasifying
medium.
The product of gassication is contained in the stream called Raw gas out,
but it can include solid particles coming out from the bottom of gasier.
The simulator allows to specify the function properties of the gasier; for this
reason a controller has been used to follow our needs.
The controller is composed by three main vari-
ables:
PV: The connected PV (process variable).
This is the variable that the controller mon-
itors while manipulating the output variable
(OP) until it matches the set point (SP).
SP: The connected SP variable. This connec-
tion is optional, and would be used if it was

Figure 2.5: Gasier simu-
desired to connect to another port in the sim-
lated
ulation to obtain a set point.
OP: The connected OP variable. This will be
the variable that the controller manipulates
in order to have the process variable (PV) match the set point (SP).
Temperature of the process and the H2 /CO ratio depends on the owrate of
gasifying. So we connect the controller to the oxygen stream, xing the top tem-
°
perature. The temperature of 1000 C has been xed because that's the optimal
temperature for gasication and for the correct ratios of products that the simu-
lator calculates.
This optimal specication is reached after the
following discussions: the gasier works at high
° °
temperature and it is run at 800 C, 900 C, 1000 C, °
°
1100 C and 1200 C. °
It is possible to realize the dierent composition
that the outlet stream has, changing the operation's
temperature.
The concentration of the syngas and the carbon
dioxide change in these cases, due to the inuence Figure 2.6: Controller
of the temperature in the reactions of gasication.
components °
800 C °
900 C °
1000 C °
1100 C
H2 O 200.23 kmol/h 150.7 kmol/h 183 kmol/h 211 kmol/h

H2 310.86 kmol/h 437 kmol/h 481 kmol/h 467 kmol/h
CO 252.47 kmol/h 553 kmol/h 604 kmol/h 615 kmol/h
CO2 178.95 kmol/h 152 kmol/h 141 kmol/h 136 kmol/h
CH4 85 kmol/h 46 kmol/h 8 kmol/h 1.33 kmol/h
Table 2.1: Stream outlet of gasication
The owrate of oxygen is the variable that has most inuence on temperature
prole. An increase in the owrate promotes the combustion, which leads to
an increasing temperature. However it is necessary not to have excess oxygen
because the combustion needs to be stopped in the partial oxydation; having
a total oxydation prevents CO2 from forming and for subsequent treatment to
absorb the gas, moreover CO is needed for the next step of water gas shift reaction.
The purpose is to increase the concentration of syngas, avoiding the total
combustion and the huge presence of CO2 . For this reason, the top temperature
°
of 1000 C is chosen. In fact, as seen in the Table 2.1, this temperature represents
a compromise between high composition of hydrogen and carbon monoxide, and
as little as possibile carbon dioxide concentration.
The oxygen molar owrate to have the determined temperature is 222.31
kmol/h.
As can be seen in the Figure 2.7, the most important feature is the reaction
model that we consider as the Gibbs reaction, in which the kinetics is not described
because every reaction is at equilibrium. In the simulator the process can work at
high temperature using Plasma Duty or just due to the presence of combustion,
it is chosen just the heat released by oxydation. The other two settings, that are
set for the gasier, is the number of reactor and the Predicted Particulate C.
First the simplest case of only one reactor is considered; for the predicted
particulated C (fraction) only a very small part it can be found like a particulate.
With the input given by user, the most important output are:
Bottom and Top Temperature

O/C ratio
H2 O/C ratio
H2 /CO mole ratio
In the gure below, it can be seen every output found, according our assumptions.
Figure 2.7: Specications of gasier

Chapter 3
Concentration and purication of
Syngas
3.1 Concentration: Water-gas-shift Reactor

The water-gas shift is a reversible, exothermic chemical reaction, usually assisted
by a catalyst. It is the reaction of steam with carbon monoxide to produce carbon
dioxide and hydrogen gas:
CO + H2 O ←→ CO2 + H2 4H = −40.6 KJ/mol
Many theoretical and mechanistic investigations has been done about the catal-
ysis of this reaction, and the reaction represents one of the most used industrial
processes. In fact it can be used in ammonia and hydrogen production. Hydrogen
is mainly produced today by steam reforming of hydrocarbons and partial oxi-
dation of hydrocarbons. However, the result of these processes give mixtures of
hydrogen and carbon monoxide not suitable for many industrial applications. For
instance, in ammonia production the amount of carbon monoxide must be very
low, since CO can deactivate the ammonia synthesis catalyst. Hence the water
gas shift reaction is used as a step in the overall process to minimize the CO levels
and produce additional hydrogen.
19
CHAPTER 3. CONCENTRATION AND PURIFICATION OF SYNGAS 20
Figure 3.1: Equilibrium constant as a function of temperature
Some thermodynamic considerations can be made: the equilibrium constant
Kp increases with decreasing temperature as the Figure 3.1 shows.
The following expression provides a right way to describe this trend in the
range of operating temperature.
4577.8

Kp = exp − 4.88
T
From the point of view of increased hydrogen production and reduced carbon
monoxide content, it is desirable to conduct the reaction at low temperatures;
from the point of view of steam economy, a lower reaction temperature is desir-
able. However, in order to achieve the necessary reaction rates, higher tempera-
tures are often required. It can be seen that an ideal condition for the water-gas
shift reaction would be the use of a catalyst active enough to operate at low tem-
peratures where equilibrium is very favorable. Reaction pressure does not have
any eect on the equilibrium of the shift reaction, since the number of moles of
material in the shift reaction does not change during the course of the reaction.
Generally in the industrial processes two kinds of catalysts are used: iron based
catalysts and copper based catalysts.

The iron based are the most popular catalysts used in industrial operation,
°
because of operating in the range of temperature 320-450 C, they are called high-
temperature-shift catalysts. The advantage is that iron oxide can tolerate small
presence of solphur and are fairly rugged overall, the disadvantage is the low rate
of reaction at high temperature.
The copper based are called low temperature shift catalysts and are therefore
attractive since equilibrium is more favorable at lower temperatures. In addition
to higher activity, another advantage claimed for the low-temperature shift cata-
lysts is more selectivity and fewer side reactions at elevated pressures than high
temperature shift catalysts. Copper-based low-temperature shift catalysts are
completely sulfur intolerant, however, being irreversibly poisoned by even small
quantities of sulfur compounds.
Another catalyst of remarkable interest (during last years) is sulded cobalt
oxide-molybdenum oxide on alumina, that, with its other variations, has been
characterized in various papers and patents. This kind of catalyst is absolutely
inert towards sulfur (doesn't suer poisoning by sulfur) and some its variations
are proved to be very active at both high and low temperatures. Industrial shift
reactors are tipically adiabatic so the temperature will increase along catalytic
bed, as the reaction is a little bit exothermic; if high purity is required, two
stage reactor is generally used, while if purity isn't so important, single stage is
considered better. Two-stage systems usually work with a high-temperature iron-
based catalyst in the rst stage, then the gases are cooled and steam is added.
Furthermore, the shift of the residual CO over a copper-based low temperature
shift catalyst is carried out. Thanks to acid gas and sulfur removal, the second-
stage catalyst is free from poisons, so its life is much longer under these controlled
conditions.[6]
3.1.1 Modeling Water-Gas-Shift Reactor
The stream leaving the gasier is composed mostly by hydrogen and carbon
monoxide, but there is also the presence of carbon dioxide, as you can see in
the Table 3.1.

Compounds Molar fraction Molar owrate
Water 0.12 150.7

Hydrogen 0.32 437.03
Carbon monoxide 0.41 553.23
Carbon dioxide 0.11 152.02
Methane 0.04 46.65
Table 3.1: Stream inlet WGS reactor
The task of Water-Gas-Shift reactor is to have a stream with a composition
as close as possible to the optimal inlet stream the Fisher-Tropsch reactor.
It's useful to explain the composition of stream entering the FT reactor: the
H2
most of important property is
CO
≈2; in this case it is xed the value 1.9. The
H2
reason is to avoid that increases during the reactor and to avoid that C1-C3
CO
product privileges on product with high content of carbon. In the nal product it's
better to have as less as possible the light carbon because of low energy content.
The reactor in VMG Simulator is an equilibrium reactor in which it is chosen
an adiabatic type without pressure drop inside the equipment. In the section
reaction the stoichiometich coecients of water-gas-shift reaction are added.
About the reaction constant, there is the possibility to choose between Gibb Free
Energy, K-T Table or K specied; in this case the Gibb Free Energy and so the

4G
following regression equation K = exp RT
are considered.
The focus is on the operating conditions of Water Gas Shift reactor; concerning
the pressure, there is no change of number of moles so it makes no sense to change
the pressure from gasier that was 20 bar. In terms of temperature, the situation
is more complicated beacuse it depends on which type of stream is added before
the reactor.
Figure 3.2: WGS reactor
As seen from the Figure 3.2, before the equilibrium reactor, there is a mixer
in which can be added hydrogen or water. In the following the dierence between
both possibilities is explained. The WGSR is an exothermic reaction in nature,
from thermodynamics it is known that an increase in reaction temperature will
impede the forward reaction for H2 production in the WGSR.
using steam
It is considered the reaction from CO, H2 O, and added steam to H2 and CO2 , the
H2
reaction moves to the product increasing the . Anyway there is one important
CO
consequence: the rise of carbon dioxide. It is a disadvantage for both gas emissions
and for the reduction of carbon eciency of the whole plant. The carbon eciency
describes how much carbon we can nd among the products. Increasing the carbon
dioxide owrate, a bigger portion of carbon leaves the plant like gas and not like
hydrocarbons, so the intention is to minimize the production of CO2 .
For this reason the analysis of the inuence of the temperature and of steam
owrate entering in the mixer, before the reactor, will do.

H2
About the owrate, increasing the owrate of the steam, the increases.
CO
There is the focus on this rise because the stream has to have the ratio around
2 before the FT reactor. However it is disdvantageous to use an excessive steam
owrate, in fact this implies a bigger production of CO2 .
Figure 3.3: Trend steam owrate-CO2 production
As shown in the Figure 3.3, the increasing of the steam owrate corresponds
to a decrease of the eciency because it happens a loss of the carbon that it will
go away as CO2 .
About the temperature, the reaction temperature on the design of the WGSR.
Varies according with the type of catalyst used. For HTC, increasing reaction
temperature, the concentration of CO declines with respect to the temperature,
° °
thus the conversion of WGSR increases, but between 400 C and 500 C the change
in this propriety is small, so temperatures between 350°C and 400°C the major
reaction occur.
°
So it is xed a range 315-380 C, it is analyzed the trend of product with the
increasing temperature. The water-gas-shift is an exothermic reaction and so we
expect the decrease in products with temperature, anyway it works to minimize

H2
the gas emission but, at same time, trying to not reduce the below 1.9.
CO
The Figure 3.4 shows, in the best way, this concept.
Temperature WGSR ( C) ° CO2 euent ow (kmol/h)

H2
CO
inlet FTR
315 373 2,01

325 369 2
350 360 1.95
375 351 1.68
(a) Trend Temperature-CO2
Figure 3.4: Inuence of temperature
After these considerations, a compromise is chosen to have high productivity
and low CO2 outgoing:
1. Flowrate steam = 240 kmol/h
2. Temperature of reactor = 350 C °

using Hydrogen
The approach to the problem, in this case, is totally opposite to the reactor with
the introduction of steam. Now the main task of Water-Gas-Shift reactor is not
H2
to reach a specic
CO
but to have as less possible CO2 leaving the plant. To meet
these needs, it has to work with the highest temperature possible, proting by the
high temperature from the gasication.
The reaction is not working with the conditions thermodynamically favored
because our intent is to consume carbon dioxide, without lowering excessively the
hydrogen to carbon monoxide ratio.

The composition of the outlet of the gasier is summarised in the following
table:
Compound Flowrate (kmol/h)
Hydrogen 437
Carbon monoxide 553.23
Carbon dioxide 152.02
Water 150
Methane 46.05
Table 3.2: Outlet the gasier
Our purpose will be to reduce as much as possible the content of CO2 , working
with the highest temperature to move to left the reaction.
So there is the possibility to omit a cooler, described for the case without
hydrogen, and take an advantage of this energetic content.
A controller is used for changing the hydrogen molar inlet owrate to have
H2
CO
= 1.9 outlet of the reactor. It ensures that the inlet WGSR stream has a
°
temperature of 1000 C. In fact this temperature is used to take advantage of the
high temperature outlet of the gasier.
The water gas shift reactor takes place with dierent types of catalyst. It
depends on low temperature or high temperature operating. For each case it
needs a catalyst to work. It doesn't happen for this plant, using a temperature of
°
1000 C. The reaction takes place anyway, using or not the catalyst, because it is
a thermodynamic equilibrium.
Actually, there are two dierent plants: the rst is to add a part of the total
hydrogen before the WGSR and the rest in the section before the FT reactor.
The nal goal is to have a determined ratio in the stream inlet of the reactor.
The second one is to add enough owrate of hydrogen only before the water gas
H2
shift reactor to have
CO
= 1.9 outlet.
It is preferable to choose the last for a simple reason: as said before, the
reaction takes place anyway, and so adding all hydrogen in the rst section means
to move to left the water gas shift reactor. The main consequence is the reduction
of the amount of carbon dioxide and the increasing of carbon eciency.

Compounds Flowrate
Hydrogen 1261.06
Carbon monoxide 662
Carbon dioxide 82.62
Water 242.06
Methane 8
Table 3.3: Outlet the WGSR
The calculated hydrogen owrate is 838 kmol/h.
The product stream from the Water-Gas-Shift reactor is the following:
As you can see from the Table 3.3, there is, even if small, the conversion of
CO2 in CO and the ratio is perfectly as it is xed.
About the hydrogen owrate, it is put two dierent make up streams in the
owsheet: one of them before the WGS reator and the other one before the second
stage of FT reactor (it depends on how many reactor are used). The second stream
is controlled and manipulated by the controller ( mentioned above ), in which the

H2
specic is
CO
= 1.9. Therefore it is not possible to manipulate the second stream,
but just the stream useful for the water-gas-shift reactor.
In conclusion:
1. Hydrogen owrate ( before WGSR ) =838 kmol/h
2. Hydrogen owrate ( before second FT reactor ) = 100 kmol/h
Two possible designs of a plant have just described ; the rst one oers economic
advantages but lower carbon eciency, the second one is more expensive but it
increases exponentially the production.
The project includes not only material streams but also energy streams. The
biomass is the energy source of the system. The energy balance shows how we nd
in the products the most of energy conferred by biomass and inlet streams. The
dierence, between the plant runed with steam or with hydrogen, is also indicated
by the energy input: using hydrogen, there is a greater energy input than using
steam and so the nal products have a bigger energetic charge.

The work is nalized to maximize the production and so it is chosen to run
the plant with hydrogen, for the reasons given above.
3.2 Modeling purication of Syngas

The stream, that comes from the separator, is composed by Syngas but also carbon
dioxide and a very small part of sulphur.
There is the necessity to remove these gases out the plant, cleaning the main
stream. We know that the typical used process is the absorption with amines.
Monoethanolamine (MEA), diethanolamine (DEA), and methyldiethanolamine
(MDEA) are the chemical solvents that have proved to be of principal commer-
cial interest for gas purication. Triethanolamine has been asided largely from
the commercial usage because of its low capacity (resulting from higher equiv-
alent weight), its low reactivity (as a tertiary amine), and its relatively poor
stability. Amines with two hydrogen atoms directly attached to a nitrogen atom
exist and they are called primary amines and are the most alkaline, such as mo-
noethanolamine (MEA) and 2-(2-aminoethoxy) ethanol (DGA). Then it is pos-
sible called secondary: amines diethanolamine (DEA) and Diisopropanolamine
(DIPA). They have one hydrogen atom directly attached to the nitrogen atom.
Triethanolamine (TEA) and Methyldiethanolamine (MDEA) are called tertiary
amines and represent completely substituted ammonia molecules with no hydro-
gen atoms attached to the nitrogen.
It is used an absorber with seven total stages in which the main gas stream
enters from the bottom and the lean solution ( liquid ) from the top of the column.
In this case the solution is composed by water and Diethanoamine (DEA).
For many years the streams that contain amount of CO2 and H2 S are treated
by aqueous solutions of diethanolamine (DEA). This amine shows the low va-
por pressure and for this reason it is suitable for low-pressure operations where
the vaporization losses are quite negligible. One disadvantage of diethanolamine
solutions is that the cleaning of contaminated solutions may require vacuum dis-
tillation and then, absorbing carbon dioxide some problems of corrosion could
Figure 3.5: Amine absorber
exist.[7]
Generally, monoethanolamine and diethanolamine should be used in concen-
tration from 20 wt% to 30 wt%. Expecially for treatment of renery gases typically
range in concentration is 20 to 25 wt%. Increasing the concentration there is a
reduction of the required solution circulation rate and thereforethe plant cost.
However, the eects are not as positive, as might expected. The principal reason
for this is that the acid-gas vapor pressure is higher over more concentrated solu-
tions at equivalent acid-gas/amine mole ratios. In addition, when an attempt is
made to absorb the same quantity of acid gas in a smaller volume of solution, the
heat of reaction results in a greater increase in temperature and a consequently
increased acid-gas vapor pressure over the solution.
As you can see from the Figure 3.6, gas to be puried is passed upward through
the absorber, counter-current to a stream of the solution. The rich solution from
the bottom of the absorber is sent to a ash drum in which the most of gases is
separated from the rest of lean solution and then it ows at some point near the
top.
In the following the processes in this loop are explained;
The stream containing the carbon monoxide is fed on the bottom of the col-
umn, the lean solution on the top. In the absorber the pressure drop is 0.1 bar
and the raise of the temperature is 4T = 40°C. In fact on the bottom, the solu-
°
tion, rich of the gases, is obtained at the temperature of 70 C. Our intention is to
recover the solution and so it is possible to recirculate, introducing it on the top
of the column.
The rich solution then ows to the low pressure ash unit, which operates
close to atmospheric pressure, using a valve, as can be seen in the Figure 3.5.
The low pressure ash tank needs to improve the eciency of CO2 removal in
this vessel. A signicant portion of the CO2 contained in the rich solution is
stripped in the low pressure ash unit. Completely regenerated solution from this
physical stripping is fed to the pump that it increases the pressure until 20
bar and, after that, we need unit operation to cool down the temperature until
°
30 C. In fact the absorption is a exothermic process, favored by high pressure and
low temperature. The last unit is the make up, because there is the possibility
to lose a small amount of solution and, using a controller, we make sure to have
a xed stream inlet the column. The solution enters to the top absorption stage,
completing the cycle.

Chapter 4
Fischer-Tropsch Reactor
4.1 Fischer-Tropsch Process

4.1.1 Chemical concepts
The main conversion processes of lignocellulosic feedstock into transportation fu-
els, a topic that is being more and more investigated, are fast pyrolysis and Fischer-
Tropsch synthesis. The two processes have been proven to be faisable procedures
in transforming biomass into liquid fuels.
The Fischer-Tropsch synthesis is a process of surface-catalyzed polymerization
that produces hydrocarbons with a wide range of chain length and functionality
by means of CHx monomers composed by hydrogenation of adsorbed CO. The
addition of surface methylene species to adsorbed alkyl groups results in chain
growth. The alkyl groups can undergo β-hydrogen abstraction to compose linear α-
oleins or hydrogen addition to compose n-parans on Co catalysts. In presence of
typical Fischer-Tropsch synthesis conditions, β-hydrogen abstraction is a reversible

chain termination step while the readsorption of α=olens causes the reinitiation
of alkyl surface chains and the generation/development of larger hydrocarbons.
Studies are currently being focused on producing syngas from gasication of
lignocellulosic feedstock. FT synthesis can take place either at low temperatures
° °
(200 C-240 C, LTFT) using an iron or cobalt catalyst, or at high temperatures
31
CHAPTER 4. FISCHER-TROPSCH REACTOR 32
°
(300-350 C, HTFT) using iron catalyst. Most commonly, slurry and xed bed
reactors are employed in LTFT, and uized bed reactors in HTFT. The complex
thermochemistry occurring in an FT reactor can be generalized as follows:
Methanation Reaction
CO + 3H2 −→ CH4 + H2 O
Fischer-Tropsch (FT) Reaction
nCO + 2nH2 −→ (−CH2 −)n + nH2 O
Water gas shift (WGS)
CO + H2 O ←→ CO2 + H2
Studies are currently being focused on producing syngas from gasication of lig-
nocellulosic feedstock. FT synthesis can take place either at low temperatures
° °
(200 C-240 C, LTFT) using an iron or cobalt catalyst, or at high temperatures
°
(300-350 C, HTFT) using iron catalyst. Most commonly, slurry and xed bed
reactors are employed in LTFT, and uized bed reactors in HTFT. The complex
thermochemistry occurring in an FT reactor can be generalized as follows:
The usage ratio may be profoundly aected by the water gas shift reaction.
Being the extent of the WGS reaction negligible for the best cobalt catalyst, it can
then be treated as a one way reaction producing a small quantity of carbon dioxide.
In this case, carbon dioxide is commonly treated as a carbon containing product.
When using iron catalysts at the upper end of the operating temperature range,
the water gas shift reaction approaching equilibrium occurs. In such a situation,
the mostused reactant is carbondioxide. The direction of the WGS reaction is
determined by the composition of the dominating gas.
While when using cobalt catalysts, the usage ratio depends primarily on FT
reaction and is signicantly inuenced by methanation reaction. The usage ratio

depends on the extent of methane formation, the olen content in the longer chain
hydrocarbons and the slight water gas shift activity. Its range is typically 2.06 -
2.16.
At the other extreme, when the WGS reaction is in equilibrium, the combined
usage ratio of hydrogen and carbon monoxide for the Fischer-Tropsch and water
gas shift reactions together, is a moving target that depends on the feed gas
composition.
Reaction conditions and catalyst composition strongly inuence C2+ hydro-

H2
carbon selectivity. For example, when
CO
ratio increases, C2+ hydrocarbon pro-
duction decreases while methane production increases, increasing reaction tem-
perature, and decreasing pressure
From the thermodynamic point of view, formation of methane and hydrocar-
bons is energetically favorable. To produce ethane and propane 4Go values at
°
227 C are -122 and -156 KJ/mol, respectively, while at 200 C ° 4H o values to
create of methane, ethane, and propane are -209, -176, and -121 kJ/mol, respec-
tively. Furthermore, equilibrium calculations show that the production of methan
is highly favored over formation of hydrocarbons of heavier molecular weight.
Therefore, FTS catalysts must be conceived with high selectivity for hydrocar-
bons other than methane; while Co, Fe, and Ru catalysts can be easily designed
to meet this requirement, Ni catalyst is not useful since it produces too much
methane.
One of the major cost factors is the production of puried synthesis gas (
syngas ), suitable as a feed gas to the FT reactor. Depending on the complexity
of the overall downstream plant, syngas can represent up to about 70% of both
the capital and operating cost, whether the raw material is coal or methane.
Hence, two things are of extreme importance: that the conversion of the syngas
to hydrocarbon products from syngas in the FT reactors is as ecient and as
complete as possible, and that the highest quantity possible of the reactans (
CO, H2 , and CO2 ) is consumed to produce useful products. The FT reaction
selectivity should be supervised in order to minimize the production of unwanted
products (e.g. methane).

To achieve the above-mentioned objectives, the syngas composition and the
usage ratio of the FT reactions should match. The usage ratio will be determined
by the overall product selectivity, which in turn is determined by several other
factors. The eect of selectivity on the H2 to CO usage ratio is illustrated in the
following table.
H2
FT product Reactions
CO
CH4 CO + 3H2 −→ CH4 + H2 O 3
C2 H6 2CO + 5H2 −→ C2 H6 + 2H2 O 2.5
Alkanes nCO + (2n + 1)H2 −→ Cn H(2n+2) + nH2 O (2n+1)/n
Alkenes nCO + 2nH2 −→ Cn H2n + nH2 O 2
Alcohols nCO + 2nH2 −→ Cn H(2n+1) OH + (n − 1)H2 O 2
Table 4.1: Usage ratio of FT reactions
For all alkenes and alcohols the usage ratio is 2, independently of chain length.
While in the case of alkanes, the more the chain length increases, the more the
usage ratio decreases. The extent of the water gas shift reaction that takes place
over the FT catalyst used is also important. For a cobalt catalyst (presenting
little or no WGS activity) the overall H2 to CO usage ratio under typical FT syn-
thesis conditions is 2.05-2.15. The usage ratio of precipitated iron based catalyst
°
operating in a xed bed reactor at about 225 C is approximately 1.65.
For cobalt catalysts (presenting little or no WGS activity), increasing FT con-
version levels do not inuence the hydrogen to carbon monoxide usage ratio. On
the contrary, iron catalysts are active for the WGS reaction, and when using
them at the lower temperatures, the overall usage ratio decreases as the conver-
sion increases. When using iron catalysts at the higher temperatures, however,
the presence of excess hydrogen in the FT reactor would be extremely useful to
minimize carbon deposition in the catalyst. If, as it is normally practiced, there
is an excess of hydrogen in the FT reactor, it is theoretically impossible to attain
a high % (CO + H2 ) conversion.
A wide range of hydrocarbon and oxygenated products results from The FT
synthesis. Mechanistic and kinetic factors appear to control the formation of the
various products and the product spectra dier signicantly from what would
be expected from thermodynamic point of view. It is calculated that if a gas

H2
with =1 were to go to complete equilibrium at 0.1 MPa and at a typical
CO
FT temperature then the main products would be methane, carbon dioxide and
graphite and the amounts of higher hydrocarbons wuold be negligible. Similar

H2
calculations were then presented for 600 K, 1.6 MPa and a gas with =2, and
CO
it is possible to recognize the dierent kind of products. It is clear that, under
normal operating conditions, the FT reactions are nowhere near thermodynamic
equilibrium, the observed C2 and higher products are produced in huge quantities
relative in comparison with thermodynamic expectation.
About the alkanes, they should theoretically be completely hydrogenated to
alkanes, they are the most produced hydrocarbons. This shows that alkenes must
be primary FT products and that, if it occurs, their subsequent hydrogenation over
iron catalyst in the gas atmosphere prevailing in the FT reactors must be slow.
Even with cobalt catalysts (much more active for hydrogenation reactions than
iron), the alkenes exceed the alkanes for the lower molecular mass hydrocarbons.
Thermodynamically, ethene hydrogenation should be more complete than propene
hydrogenation.
In the following gure, you can see the typical products, changing type of
catalyst and temperature.

Figure 4.1: FT hydrocarbon isomers
FT reaction always produces a wide range of hydrocarbon and oxygenated
hydrocarbon products, irrespective of the operating conditions. Methane, is an
unwanted product that is always present and whose selectivity can change from
as low as about 1% up to 100%. At the other end of product spectrum, the se-
lectivity of long chain linear waxes can vary from zero to 70%. The alteration of
the operating temperature, the type of catalyst, the amount or type of promoter
present, the feed gas composition, the operating pressure, or type of reactor used
can change the spread in carbon number products . Whatever the process condi-
tions, clearly all the products formed are always interrelated. [8]
Figure 4.2 illustrates the link between the lower molecular mass hydrocarbons
produced with a fused iron catalyst operating at about 330 C.°

Figure 4.3 illustrates the relationship between the higher molecular mass hy-
drocarbons produced over a precipitated iron catalyst operating at about 225 C. °

( Remember that Hard wax is the hydrocarbon fraction boiling above 500 C). °
Figure 4.4 shows how the selectivity of some dierent C2 compounds varies
with the methane selectivity.

Figure 4.2: Hydrocarbons product selectivities relative to the methane selectivity
Figure 4.3: Selectivity for product distillation cuts relative to the hard wax selec-
tivity
Figure 4.4: C2 product selectivities relative to the methane selectivity
4.2 Modeling the Fischer-Tropsch Product Distri-

bution and Model Implementation
The selectivities of hydrocarbon product are determined by chain propagation
rate relative to chain termination rate. In FTS the distribution of the products is
given by in part by a chain polymerization kinetics model ascribed to Anderson,
Schulz, and Flory, henceforth referred to as Anderson-Schulz-Flory (ASF) model.
The Anderson-Schulz-Flory distribution, with propagation probability α (growth

factor) as the only parameter, is the ideal ASF distribution. There are some as-
sumtions: no chain limitation with the ASF distribution and having no chain lim-
itation; but it may seems to be conceptually problematic. The probability of pro-
ducing chain lenghts larger than N carbons from a distribution is: p(i > N ) = αN .
By the combination of two or more ideal ASF distributions and addition of single
reactions, it is possible to represent most observed product distributions.
The ASF product distribution is mathematically dened by the equations:
wn mn+1 rp
= αn−1 (1 − α)2 α= =
n mn rp + rt
where n is the number of carbon atoms in the product, wn and mn are the
weight and mole fractions of product containing n carbon atoms, rp is the rate of
chain propagation, and rt is the rate of chain termination
It is observed that the growth factors are dierent between parans and
olens, instead oxygenates are often neglected because only very small amounts
are formed.
In conclusion the followig expressions describe the production of parans and
olens with two dierent reactions, rst one for n-parans and the second for
distribution of 1-olens. It is easy to understand that we use two dierent growth
factors, α1 and α2 .
∞
X
CO + U1 H2 −→ νi,1 Ci H2i+2 + H2 O (1)
i=1
∞
X
CO + U2 H2 −→ ν1,2 CH4 + ν1,2 Ci H2i + H2 O (2)
i=2
Notice that the second reaction also included CH4 . Even if it is not an olen,
there is a good reason for including also in the second reaction. So the total
production of methane from these two reactions is therefore RCH4 = ν1,1 r1 + ν1,2 r2
while the parans are RCi H2i+2 = νi,1 r1 and olens are RCi H2i = νi,2 r2 .[9]
Stoichiometry
The stoichiometric coecients will follow the ASF distribution and must be given
as:
νi,1 = (1 − α1 )2 α1i−1
νi,2 = (1 − α2 )2 α2i−1
Using the stoichiometric coecient, the sum for reaction 1 is 1 − α1 and, of
course, this is the number of molecules of hydrocarbon for each consumed molecule
of CO. The same consideration is for the second reaction but using α2 .
In the equation (1), the symbol U1 is used but the exactly stoichiometric
coecient can be calculated:
∞
X
U1 = (i + 1)νi,1 + 1 = 3 − α1
i=1
∞
iνi,2 + 1 = (1 − α2 )2 + 2
X
U2 = 2ν1,2 +
i=2
lumping of components
It can be useful, especially for the simulations, to introduce the concepts of lump-
ing of components. The advantage is the ability to describe an innite number
of component in a model. The interest is in calculating the molar fractions, the
stoichiometric coecient and the average carbon number of a lump. It is indis-
pensable to provide a detailed description of the lump.
The nite sum of molar fraction up to N carbon atoms is:
N
αi−1 = 1 − αN
X
S[1,N ] = (1 − α)
i=1
If, instead, the sum is from N to M carbon atoms:
M
αi−1 = S[1,M ] − S[1,N −1] = αN −1 − αM
X
S[N,M ] = (1 − α)
i=N
About the number average carbon number of a lump of components, it is very
important to create an hypothetical compounds with a specic number of carbons.
If the lump is composed by closed interval from N to M, the equation is:
M
iαi−1
P
i=N N αN −1 − (N − 1)αN − (M + 1) αM + M αM +1
n̄n,[N,M ] = =
M (1 − α) (αN −1 − αM )
αi−1
P
i=N
The number average of the lump [N, ∞] is reduced to:

α
n̄n,[N,∞] = N +
1−α
kinetic model
The model should be able to predict reaction rates as function of temperature and
composition and preferably also the product selectivity.
Most of the rate expressions originate from Langmuir-Hinshelwood (LH) or
Eley-Rideal (ER).
Instead, in this case a more complex kinetic model, based on the study done
by Todic et al., is used. The model is based on the CO-insertion mechanism.
Both parans and olens are included; the ratio between the paran and olen
growth factors is a constant.
The model describes the rate for each component and introduces dierent α
for each polymerization step. In general the growth factors depends on the vacant
site fraction, [S], which again is a function of changing growth factors, making
the model implicit. Because of the changing growth factor, the model does not
follow the ASF distribution and the formalism shown here can't be applied. On
the other hand, the model can be simplied without much loss of accuracy. The
main simplication, done in the FT reactor, is for all polymerization steps growth
factors are assumed to be constant; α1 for parans and α2 for olens, equal to
the upper limit predicted by the original model. This makes the growth factors
independent on the fraction of vacant sites [S] and the model becomes explicit.
1
α1 = √
k7 K2 pH2
1+ k3 K1 pCO
α2 = α1 e−0.27
1
αM = √
k7,M K2 pH2
1+ k3 K1 pCO
After describing of the alpha value, this is the following semplication about
the fraction of vacant sites, overall rates of reactions (1) and (2), respectively.
q αM
r1 = k7 K2 pH2 [S]2
(1 − α1 )2
αM
r2 = k8 [S]
(1 − α2 )2
!
−1
q pH 2 O q αM
[S] = 1 + K1 pCO + K 2 pH 2 + + K 2 pH 2
K22 K4 K5 K6 p2H2 (1 − α1 )
The rate expressions are given in [kmol/(kg h)] and the partial pressures are
given in [MPa].
If the problem is analyzed more accurately, it could be said that the selectivity
of methane is higher than predicted by ASF distribution, while ethylene is lower.
To have the real distribution, another reaction can be introduced, such as the
hydrogenolytic cleavage of ethylene; r4 is the rate of the following reaction:
C2 H4 + 2H2 −→ 2CH4
The rate r4 will be the dierence between the prediction according to the ideal
ASF distribution of olen production and the observed ethylene production.
r4 = (1 − α2 )2 α2 r2 − rC2 H4
About methane, the production is higher than the one predicted from r1 and
from the ethylene cleavage. So a reaction of methanation can be introduced. r3

is the rate of the following reaction:
CO + 3H2 −→ CH4 + H2 O
The reaction rate will be the dierence between the observed rate rCH4 and
the prediction according to the ideal ASF.
r3 = rCH4 − (1 − α1 )2 r1 − (1 − α2 )2 r2 − 2r4
So, according with this reasoning, we can insert the Todic model and the
reaction rate will be the following:
q
r3 = k7,M K2 pH2 [S]2 αM − (1 − α1 )2 r1 − (1 − α2 )2 r2 − 2r4
r4 = (1 − α2 )2 α2 r2 − k8,E [S] αM α2
4.3 Modeling FT Reactor

The simulation can be divided into two parts: base case using vapour (before the
Water-Gas-Shift Reactor) and advanced case using hydrogen (before the WGSR).
In this part, the eects, that some specications lead on the products, are
analyzed. The purpose of these considerations is to optimize the process and
maximize the production of nC5+ .

At the beginning, the optimization on the base case will be done and their
results will be compared with the advanced case to nd the best possible ow
sheet. The choice won't be the most economical choice but the most productive
one.
First of all, the stream, entering in the FT reactor, should be xed in temper-
ature and composition.
About temperature, it works with the typical temperature used for LTFT
process. Above it has been explained the types of reactor and for LTFT the range
° °
of temperature is 200 C-230 C. So a heater is used to bring the temperature up
°
to 210 C.
As shown in the previous section, two types of catalystscan be used: iron type
and cobalt type. In the simulation the Todic's expression is followed about the
kinetics and Todic used a particular catalyst: 25%Co/0.48%Re/Al2 O3 .

Instead, about composition, we have to divided our simulation into two sec-
tions: a plant with hydrogen and a plant without hydrogen.
with hydrogen
The right composition is given by
make-up stream of hydrogen, before
the Water-Gas-Shift reactor.

H2
The specication is
CO
= 1.9. This
value has been xed because the main
FT reaction works with the ratio equal
to 2 or little bit more. So working
with lower ratio, the reaction moves to-
wards hydrocarbons with higher molec-
ular weight, getting closer to higher
material and energy eciency. In con-

H2
trast , working with
CO
= 2.1, it pro-
duces a lot of C1 , C2 , C3 , C4 , which is
useless for the main production.
Now all the unit operations are an- Figure 4.5: Controller of composition
alyzed. On the left side of this Figure,
the unit M1 can be seen. This unit mixes the stream from the main plant with
the one from the recycler. On the lower left of the gure the unit CSP1 is shown,
which is a component splitter. This unit is useful for dividing hydrogen from the
rest of the stream.
In this way, the hydrogen stream S26 is created. It is directed into the con-
troller.
We remember that this type of plant is more energy-intensive. As seen from
the Figure 4.6, a unit calculator called Set3 is necessary. It manipulates the make
H2
up hydrogen stream in order x the ratio of to 1.9. Actually, the only stream,
CO
that has to be manipulated, is that of make up; in fact in this Set, the molar
owrate of CO is xed and multiplied for the hydrogen to carbon monoxide ratio.
Figure 4.6: Set controller
In the following, a detailed explanation of the processes in the FT reactor is
given.
In the VMGSim, the ideal Fischer-Tropsch reactor models the hydrogenation
of carbon monoxide over a catalyst bed. The products of this unit operation
are vapor, liquid, and wax outlet material streams. Enthalpy of formation is
accounted for all products generated and enthalpy of fusion is calculated for any
wax drop out. Carbon monoxide conversion, carbon dioxide selectivity, and paraf-
n/olen/alcohol product distribution kinetics can be entered as signal values by
the user to customize this operation. Alternatively, these values can be automat-
ically estimated through methodologies from literature for both iron and cobalt
based catalysts.
Figure 4.7: Model FTR
If the kinetic of the reactor is checked , it could be realized that it is too
restrictive using this pre-set model. The kinetic and CO selectivity, used in the
simulator, is the following:
−0.2 0.7
−r = aPCO PH2
E
conversion CO = Ae− RT PCO
−0.2
PH2 (catalystactivity)
For this reasons, it would be better to build it yourself, using PF reactor where
we have to set everything.
In the Plug Flow (PFR) or tubular
ow reactor the feed enters at one end
of a cylindrical tube and the product
exits at the other end. The PFR model
used in VMGSim assumes that there is
no mixing in the axial direction and a
complete mixing in the radial direction.

Figure 4.8: Single tube
This unit operation works like a single
tube. So the total mass owrate is di-
vided in so many tubes as it is necessary to have 0.5 m/s the space velocity inside
the single tube. As you can see from the Figure 4.8, two splitters are inserted,
rispectively before and after the FT reactor.
The stoichiometric coecients has to be set, introducing dierent reaction.
In the following part, an approximation will be done, about alpha value, already
explained in the modeling FT section. This value depends on the temperature

H2
and the ratio . This ratio can change inside the reactor, especially for a high
CO
raise of temperature. At the moment of the rst valutation, α1 and α2 are xed
and the reaction and his stoichiometric coecients, inserted, is shown in the next
table.
component ν n̄n
C1 (1 − α1 )2 1
2
C2 (1 − α1 ) α1 2
2 2
nC3 (1 − α1 ) α1 3
α1
nC21+ (1 − α1 ) α120 20 + 1−α 1
component ν n̄n
2
C2 (1 − α2 ) α2 2
2
1 − C3 (1 − α2 ) α2 3
9 20 10α92 −9α10 20
2 −21α2 +20α2
21
1 − C10−20 (1 − α2 ) (α2 − α2 ) (1−α2 )(α2 −α2 )
9 20
α2
1 − C21+ (1 − α2 ) α220 20 + 1−α 2
Table 4.2: component lumping with the stoichiometric coecients and number
average carbon number (rst table for n-parans, second table for 1-olens)
As you can see, there are also the lumps; they are compounds that describe a
range of hydrocarbons . They are created as hypotetical compounds from already
existing components, with the same molecular weight and number average carbon
that is calculated in the section before. The stoichiometric coecients should be
balanced and we could modify the error tolerated from the settings of reactor.
The second step is to integrate the kinetics. Using the advanced kinetic, it is
edited the dierent expressions, for every reaction that exists in the reactor. Every
constant is specied and its dependencies and, due to the gas phase reaction, every
partial pressure.
Instead, the traditional kinetic could be used adding Ea and A, respectively
energy of activation and pre exponantial factor, but it is not allowed to use more
complex kinetics, even if it is easier to manage. In the advanced section, every
type of kinetics can be written. The VMGSim works with python language, so
every variable introduced has to be write in the consistency way.
The Figure 4.9 is a example of kinetic for the reaction that produces n-
parans.
Every constant as k, the growth probability factor as ap, aq and am, nally
the partial pressure as P are dened.

Figure 4.9: kinetic expression
A microscopic analysis are done on the reactor, but the FT section is not
only consists of the reactor; it is composed by some dierent unit operations are
explained in the following:
This part includes the separator immediately after the PF reactor. In fact
there are two outlet streams from the real reactor, not just one as you can see
from the Figure 4.10; so a separator has to add to model perfectly the real case.
The liquid stream, from the separator, is formed mainly by high hydrocarbon,
instead the gas stream will be cooled down and split through three phases sepa-
rator. This unit includes water, hydrocarbons and vapour composed by methane,
that represents inert for the process, hydrogen, CO unconverted and other light
hydrocarbons. This vapour stream will be recycled to increase the productivity.
The following picture shows that the product is in the stream S10.
Figure 4.10: FT section
First of all, it is necessary to describe the reactor and its properties.
There is a need to x some specications, without these, the reactor would
not converge.
The reactor is working with catalyzed reaction so the most suitable correlation
to calculate the pressure drop has to be chosen: Ergun expression; inside the
reactor, the pressure drop, are of around 1.33 bar. This value depends also on
reaction phase, that in our case is vapour.
The second specication is about
the overall heat transfer coecient, it
depends on what type of utility we are
using. The Fischer-Tropsch reaction is
an exothermic reaction and the tem-
perature rises to not have a stable con-
trol of it and the following embrittle-
ment of tubes walls that are made of
steel carbon. So, by using the boiler
feed water, the temperature decreases

Figure 4.11: Tren temperature and pres-
and temperature control becomes sta-
sure inside the reactor
ble. As you can see from th Figure
4.11, the trend shows a maximum where the rate reaction is very high. The
°
following decrease is due to the presence of water at 220 C. The simulator shows
how much heat is removed by the cold utility, and the main inuence is given by
W
the overall heat tranfer coecient (U); which for water, it is 1000 .
m2 K
Then, it's absolutely needful to insert some properties about the catalyst:
catalyst shape, packing, diameter of catalyst and aspect ratio that, if a catalyst
shapedierent forma sphere is chosen the aspect ratio of the catalyst must be
provided here. This value is taken as the Length/Diameter ratio.
The last values are the length of the tubes and the inner diameter. The inner
diameter is strongly connected to the number of tubes, because the space velocity,
the velocity inside the tube section, is set to 0.5 m/s.
The next Figure shows the input elds for these variables.
Figure 4.12: Properties FT reactor
without hydrogen
In this case, the right composition is given by the steam introduced in the plant.
The inlet composition is given in Table 4.4. As you can see it is composed by
less CO and H2 .
Compounds Flowrate (kmol/h)
Hydrogen 664.22
Carbon monoxide 320.14
Methane 46.24
Water 0.66
Table 4.3: Inlet FT composition
However this case is not taken in consideration because the liquid production
is too low and we are forced to work with one step because of the drop of hydrogen
to carbon monoxide ratio. That's the main reason why it has been decided to use
the previous plant. Using the same perfectly conditions of the optimized case
H2
(described above), the is too low, equal to 1.16. So without introducing of
CO
hydrogen, it is impossibile to consider two FT reactor. The total nC5+ production
is 2132.75 kg/h. The dierence is considerable and so we prefer also do the
optimization with the hydrogen case.
4.3.1 Optimization
In this section, our aim is to optimize the FT reactor and the connected unit
operations . The goal will be to maxime the liquid production, without an interest
of minimizing the costs. You could realize this purpose because we are using
hydrogen despite the cost due by air separation.
In Figure 4.13 there are all specications which it allows to manipulate to
increase the productivity (kg/h) and these are: the cooling water temperature
in the FT reactor, the length of the tube, the temperature used to separate the
vapour, the recycle ratio (RR) and the possibility of useing two stages.
Cooling water temperature
The productivity is stronghly linked to cooling water temperature. In fact the
total rate reaction depends on how much heat the cold utility removes. The
maximum of the curve, that describes the trend of the tempurature inside the
Figure 4.13: Variables to manipulate
reactor, moves to the right if less cold water is used. Additionally all rate reactions
seem to be higher, and also the reacting species increases.
Of course, it is not possible to work with too high cooling water temperatures,
because a unstable hot spots, inside the reactor, have to avoid. This water is
boiler feed water that it is opposed to the exothermic reactions but, in the same
time, it uses this heat to pass from liquid to medium pressure vapour at 20 bar.
So the range of this cooling water is 210-220 C. °

H2
Actually, another real limit exists: the outlet to the FT reactor. The
CO
stream, inlet of the unit operation, is a substoichiometric feed, so there is a de-
crease of this ratio. An excessive reduction represents a limit because it requires

H2
a substantial usage of hydrogen from make up, to have
CO
= 1.9.
In the next Figure, you can see the trend of these variables.
Figure 4.14: Trend changing cooling water temperature
° °
By lowering the cooling water temperature, from 220 C to 210 C, the CO
conversion drops from 55.4% to 33.1%. So it's convenient to work with higher
temperature. Even if there is a drop of hydrogen to carbon monoxide ratio, it
doesn't decrease until minimum ratio and so it is used the highest conversion that
it is given by cooling water temperature of 220 C.°

Length of tube
As explained in the previous section, we have to set a CO conversion limit, for
the reasons already given. The limit is 70% and so the length of the tube is
changed until the CO conversion reaches a level of 65% , to work safe and because
these simulations are without recycle so open-loop, so we expect a raise of the
conversion when we will close the loop.
H2
Length (m) CO conversion (%)
CO
nC5+ (kg/h)
15 55.4 1.64 3628.34

16 57.9 1.6 3788
17 60.1 1.58 4931.7
18 62.2 1.55 4060
19 64 1.53 4175
20 65 1.5 4276
Table 4.4: Eects of length of tube
As you can see from the Figure 4.15, most of the reaction happens in the rst
segments of the reactor. Nevertheless the reaction continues until the temperature
is equal of the one of the cold utility. So the volume of the reactor can be increased
by changing the length of tube as far as possible. If the prole temperature
°
remains constant at 220 C, there is no or just a little reaction. So the changing
of the length wouldn't have a benet.
Temperature of separator
Now there is a discussion about the manipulation of the outlet temperature before
the three-phases separator .
This is the least important specication. A variation of temperature doesn't
°
have a big impact if it takes place below 70 C. The plant is simulated within a
°
temperature range from 35 C (minimum temperature using industrial water as
cold utility) to 60 C. °
Temperature cooler ( C) ° nC5+ (kg/h)
35 4276
40 4270.31
45 4260.15
50 4220.92
60 4170.24
Table 4.5: Eect of the separator's temperature
To have a bigger production, we choose T=35 C. °

Stages
In this section, the discussion is about whether on or two stagesare favourable.
The conguration described, is an examplefor one stage, with the length of 20 m.
The CO conversion is xed to 65%. The total volume is around 38 m3 . The second
possibility is to work with two reactors, that together have the same volume as the
previous reactor. Our intention is to understand which conguration represents
the highest production. From the simulation it is known that two reactors with
the same volume are the optimum conguration. This is shown in the following
table.
Stages volume 1 ° reactor (m3) volume 2 ° reactor (m3) 3

total volume (m )
1 37.8 / 37.8
2 26.44 11.36 37.8
2 37.8 13.17 51
Stages 1 ° CO conversion (%) 2° CO conversion (%) total conversion (%)
1 65 / 65
2 53.6 51.6 77.6
2 65 65 88
Table 4.6: Eects of stages
After understanding that, because our purpose is to increase more and more
the conversion, we force the reactor to work with the limit conversion (65%) and
so each reactor works very close to the limit condition. It is obvious that the total
conversion increases and reachs 88%.
Recycle ratio (RR)
The recycle ratio is dened as the ratio between the recycled stream and the main
stream (entering to the mixer). The recycled stream is composed by methane (a
huge part of ), unreacted hydrogen and carbon monoxide and small part of C1 ,C2 ,
C3 , C4 . If we consider also this stream, the conversion increases as well as the
production. So there is an economical benet because of the re-use of unreacted
stream.
The plant is simulated using dierent RR which shows two dierent eects:
using a higher RR, there are more reagents and the possibility to convert becomes
bigger. At the same time, there is a raise of methane that it is considered aa
a disadvantage. In fact the presence of methane, that represents a product for
the FT reaction, moves to left the reaction. The combination of these two eects
means the presence of the maximum in the graphic RR-nC5+ .
To maximize nC5+ , we will choose RR=0.13 in which the liquid production is
6100 kg/h. The CO conversion in the rst reactor is 66%, in the second reactor
it is 68%, instead the total CO conversion is 88.8%.
In the last step, remembering that the limit conversion is 70%,each reactor is
Figure 4.15: Eects of RR on the production
forced to work in this condition. The lenght of the tubes of the rst reactor is
23m, for the second one it is 21m.
The optimized plant will be with total CO conversion equal to 90% and its
nal productivity is 6200 kg/h.
4.3.2 Rigorous Approach
In the previous section, the values for alpha could be calculated because they
are considered as constants in the reactors. Actually these values change, if the
temperature and the hydrogen to carbon monoxide ratio is changed.
For this reason, the reaction is splitted up into many small reactions. So it is
possible to analyze the reaction in detail.
First of all, it seems to be better to divided into a lot of number of reactions,
so we can analyze every reaction.
We used to work with a xed stoichiometric coecients, but in this case this
is no longer possible, because the coecients depend on the value of alpha that
H2
depends on the temperature and . This ratio decreases along the reactor and
CO
the temperature shows a maximum, so there is a change of temperature.
It could be possible to bypass the problem of stoichiometric coecients in the

following expressions for the parans:
iνi CO + ((i + 1) + iνi )H2 −→ νi Ci H2i+2 + iνi H2 O
in which r = rF T · iνi ; we divide for iνi :
1 1

CO + 2 + H2 −→ Ci H2i+2 + H2 O
i i
q αM
iνi · rF T = k7 k2 PH2 S 2 · iνi
(1 − α1 )2
νi = (1 − α1 )2 αi−1
q
iνi · rF T = k7 k2 PH2 S 2 αM · iα1i−1
For olens, the reaction will be:
iνi CO + 2iνi H2 −→ νi Ci H2i + iνi H2 O
1
CO + 2H2 −→ Ci H2i + H2 O
i
αM
iνi · rF T = k8 S · iνi
(1 − α2 )2
νi = (1 − α2 )2 α2i−1
The diculty is the characterisation of the lump. It depends on the growth
probability factor and we know that this value is not xed, but it depends on tem-
H2
perature, and molecular weight. So it is possible to calculate the stoichiometric
CO
coecients or to describe in the simulator for a specic lump.
A way of overcoming this problem is by representing the lump of varying α

with two lumps of constant α; one with low, αL , and one with hight, αH . The
original and the modied reactions will then become as shown below.
CO + U1 H2 −→ ... + νl (α)Cl + ...
CO + U1 H2 −→ ... + φνl (αL )Cl,L + (1 − φ)νl (αH )Cl,H + ...
A dierent range is chosen for α1 and α2 . Of course, for parans the bounds
could be [0.7,0.97], and for olens, the bounds could be [0.7,0.8]. As the growth
factor α is changing, this will be reected in a changing distribution between the
two constant lumps, φ. The conservation of mass requires that the stoichiometric
coecient multiplied with the molecular mass of the lump remains the same.[10]
νl (α)M̄n,L (α) = φνl (αL )M̄n,L (αL ) + (1 − φ)νl (αH )M̄n,H (αH )
So the reaction will be:
CO + H2 −→ νl (αL ) Cl,L
1 1

CO + 2 + H2 −→ Cl,L + H2 O
n̄L n̄L
q αM
rF T · n̄L (1 − αL ) αLN −1 = k7 k2 PH2 S 2 n̄L (1 − αL ) αLN −1 φ
(1 − α1 )2
where n̄L and φ are:
αL
n̄L = N +
1 − αL
νl (α) M̄n − νl (αH ) M̄n (αH )

φ=
νl (αL ) M̄n (αL ) − νL (αH ) M̄n (αH )
The liquid production is calculated in this case, with the optimized conditions.
The productivity is lower and the reason comes from the variable alpha value.
H2
Even if the is decreasing, the alpha value doesn't increase; probably it depends
CO
on PH−0.5
2
that it increases more than the hydrogen to carbon monoxide decreases.
So nC5+ will be 5000 kg/h.

Chapter 5
Conclusion
This project wants to give a green way to produce fuel. The aim is to optimize
the plant, increasing the CO conversion and the carbon eciency.
Certainly, gas to liquid process will have higher eciency, but at same time
the burning of traditional fossil fuels can also emit greenhouse gas, such as carbon
dioxide, into the atmosphere and cause some other severe environmental issue.
This plant is produced in Norway, of course, the most of common type of
biomass, in this country, has been chosen: Raw Wood.
Literature considers wood as one of the best kind of biomass, because of the
high carbon eciency and it is one of the richest in carbon, 48.2 wt%.
Pretreatment of the biomass should be properly performed to create suitable
gasication feed with low cost and ecient logic chain. In the gasication pro-
cess, some parameters have to be optimized, such as gasication temperature and
gasifying agent; they are connected between them and so it should be optimized
to convert biomass into hydrogen sucient raw bio-syngas. With the cleaning
process, the organic and inorganic impurities, such as sulfur and carbon dioxide,
will need to be removed to meet requirements in the following catalytic conver-
sion. The most of the attention is paid to optimize the FT reactor: cooling water
temperature, lenght of reactor, number of stages, amount of the recycle. The cat-
alyst is the heart of Fischer-Tropsch synthesis and in this case, using a determined
kinetic ( Todic model ), the respective type of catalyst has been used ( Cobalt
60
CHAPTER 5. CONCLUSION 61
based ).
Two cases are done: alpha value xed and alpha variable. It's possible to
understand the dierence from the following mass balance.
Figure 5.1: Mass balance ( approximate case )
Figure 5.2: Mass balance ( rigorous approach )

Bibliography
[1] Syed Ali Gardezi, BabuJoseph, Faustino Prado, Alejandro Barbosa. October
2013. article Thermochemical biomass to liquid (BTL) processes: Bench-
scale experimental results and projected processes economics of a commercial
scale processes.
[2] Prabir Basu, Dalhouse university and greeneld research incorporated. Sec-
ond edition. Biomasss gasication, pyrolysis, and torrefaction. Chapter 1,
Chapter 3
ond edition. Biomasss gasication, pyrolysis, and torrefaction. Chapter 1
[6] David S. Newsome, reviews: Science and Engineering. 2006 The water gas
shift reactor
[7] Arthur L. Kohl, Richard B. Nielsen. 1997. Alkanolamines for Hydrogen Sul-
de and Carbon Dioxide Removal. Chapter 2
[8] Andre Steynberg. 2004. Fischer-Tropsch Technology. Chapter 3
[9] Magne Hillestad. 2015. Modeling the Fischer-Tropsch product distribution
and model implementation
62
BIBLIOGRAPHY 63
[10] Magne Hillestad. 2015. Modeling the Fischer-Tropsch product distribution
and model implementation

Vmgsim

Uploaded by

Copyright:

Available Formats

Vmgsim

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Vmgsim

Uploaded by

Copyright:

Available Formats

Modelling and optimization of a

Biomass-to-Liquid process using

Norwegian University of Science and Technology

1.1.1 Products of Biomass . . . . . . . . . . . . . . . . . . . . . 5

1.1.2 Biomass conversion . . . . . . . . . . . . . . . . . . . . . . 6

2.2 Gasication Process . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2.3 Modeling of Gasier . . . . . . . . . . . . . . . . . . . . . . . . . 15

3 Concentration and purication of Syngas 19

3.1.1 Modeling Water-Gas-Shift Reactor . . . . . . . . . . . . . 21

3.2 Modeling purication of Syngas . . . . . . . . . . . . . . . . . . . 28

4.1.1 Chemical concepts . . . . . . . . . . . . . . . . . . . . . . 31

4.2 Modeling the Fischer-Tropsch Product Distribution and Model Im-

4.3 Modeling FT Reactor . . . . . . . . . . . . . . . . . . . . . . . . . 43

4.3.2 Rigorous Approach . . . . . . . . . . . . . . . . . . . . . . 56

be transfered from where it is abundant to where required for power production.

share of liquid fuel in world energy demand is projected to decrease by 2035.

The transportation sector depends almost exclusively on the hydrocarbon fuels as

oil, increased request in transportation sector must be supplied by alternative

(GHG) emissions. There is an incentive to produce renewable fuels from biomass.

gasication. Biomass is a valuable renewable feedstock, its global request is pre-

use of biomass for energy is direct combustion, followed by pyrolysis,gasication

is a chemical process that transform carbonaceous materials like biomass into

functional convenient gaseous fuels or chemical feedstock. Pyrolysis, partial oxi-

dation,and hydrogenation are connected processes. Combustion also transforms

occurs in reducing (oxygen-decient) environment releasing heat, while combus-

tion occurs in a oxidizing environment releasing heat. The object of gasication

production of chemical feedstock.

1.1 Biomass and its products

a seed germinates or an organism is born. Unlike fossil fuels, biomass doesn't

it is considered renewable. Every year, a huge quantity of biomass is produced

through photosynthesis by absorbing CO2 from the atmosphere. When it burns,

accumulated biomass are sugar (cereal),cellulosic (non cereal), and starch.

1.1.1 Products of Biomass

gas, substitute natural gas

 Solid fuels (charcoal, torreed biomass, biocoke, biochar)

1. Chemical industries for production of methanol, fertilizer, synthetic ber

and other chemicals.

2. Energy industries for generation of heat and electricity.

3. Transportation industries for production of gasoline and diesel.

4. Environmental industries for capture of CO2 and other pollutants.

a crucial point in the global energy scene.

1.1.2 Biomass conversion

biomass cannot be manipulated, accumulated, or transported easily. This gives an

ways (Figure 1.1): biochemical conversion (fermentation) and thermochemical

conversion (pyrolysis, gasication). Thedisadvantage of bulkiness and other aws

puree solid fuel through carbonization and torrefaction. Biochemical conversion

produced from corn using fermentation. Thermochemical conversion of biomass

were operating. [2] .

Figure 1.1: Dierent options for conversion of biomass

biomass-derived syngas (bio-syngas) by gasication. Then, a cleaning process

is applied to remove impurities from the bio-syngas to produce clean bio-syngas

which meets the FischerTropsch synthesis requirements. Cleaned bio-syngas is

then conducted into a FischerTropsch catalytic reactor to produce green gasoline,

diesel and other clean biofuels.

2. Water Gas Shift Reactor

3. Fisher Tropsch Reactor

processed with a new simulator called VMGSim.

The aim is to nd the optimize variables to maximize the production.

2.2 Gasication Process . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2.3 Modeling of Gasier . . . . . . . . . . . . . . . . . . . . . . . . . 15

3 Concentration and purication of Syngas 19

3.2 Modeling purication of Syngas . . . . . . . . . . . . . . . . . . . 28

share of liquid fuel in world energy demand is projected to decrease by 2035.

gasication. Biomass is a valuable renewable feedstock, its global request is pre-

use of biomass for energy is direct combustion, followed by pyrolysis,gasication

occurs in reducing (oxygen-decient) environment releasing heat, while combus-

tion occurs in a oxidizing environment releasing heat. The object of gasication

Solid fuels (charcoal, torreed biomass, biocoke, biochar)

1. Chemical industries for production of methanol, fertilizer, synthetic ber

conversion (pyrolysis, gasication). Thedisadvantage of bulkiness and other aws

Figure 1.1: Dierent options for conversion of biomass

biomass-derived syngas (bio-syngas) by gasication. Then, a cleaning process

which meets the FischerTropsch synthesis requirements. Cleaned bio-syngas is

then conducted into a FischerTropsch catalytic reactor to produce green gasoline,

The aim is to nd the optimize variables to maximize the production.

Molecular weight: in this case it is considered, like default, 10000.

Lower Heating Value: for wood the value is 18,5 MJ/Kg

Chemical formula and composition

in the owsheet: Biomass.

of Gasication. In fact the biomass chips have to be subjected to a process of

theless certain requirements need to be fullled, like charge at the temperature of

until 20 bar, necessary for the gasier.

2.2 Gasication Process

Gasication and combustion are two closely con-

nected thermochemical processes, but there is an important dierence between

(stoichiometric) level, the processchanges from gasication to combustion, and the

gasifying medium, in this case oxygen. In fact, amongst gasication medium,

The reaction (1) is known as Water-gas-shift reaction and it is perhaps the

most important gasication reaction. The big amount of hydrogen has a

strong inhibiting eect on the gasicartion; in fact the best solution is to

hydrogen in the gasication atmosphere can reduce the gasication rate by

increases the H2 amount of the gasication product. The heat of enthalpy is

reaction, the diculty is to nd the compromise between the thermodynamic

reaction inCO2 is insignicant below 1000 K.

The succession of gasication reactions is based on some extent on the type of

of bubbling uidized beds. Fluidized-bed gasiers typically operate in the tem-

uidized-bed gasiers operate at relatively low temperatures, most high-ash fuels,

Figure 2.4: Trend of temperature inside the gasier

2.3 Modeling of Gasier

PV: The connected PV (process variable).

SP: The connected SP variable. This connec-

OP: The connected OP variable. This will be