COWIFinal
COWIFinal
COWIFinal
on
Presented to
COWI and Borealis
Principal Investigators
Josefina Jernberg, Øyvind Nørregård, Marianne Olofsson,
Oliver Persson, Maria Thulin
Tutors
Christian Hulteberg, Hans Karlsson
This study is divided into two parts, a literature review and a feasibility study. The aim of the
literature review was to give Borealis a background comprising catalysts, available processes
and contaminants that possibly can have an impact on the process. In the second part, the
feasibility study, a design of a green ethylene plant is presented together with an economical
evaluation.
Contaminants found in 2nd generation bioethanol can possibly have an impact on catalysts and
further investigations are needed in this area. The types of catalysts that seemed to be the most
promising ones are alumina oxides, zeolites and heteropolyacids. The main problem using any
of these catalysts is coke formation. Modifications can be made in order to decrease coke
formation. There are companies claiming that they can sell processes which dehydrate ethanol
to ethylene. Some of these companies are British Petroleum, Chematur and Axens together
with Total and IFPEN.
The plant designed in the feasibility study consisted of one reactor with associated
downstream purification and recycle units. The reactor was designed to be isothermal and
contained a tube bundle filled with heteropolyacid catalyst. The process was simulated
successfully and 102,000 tonnes ethylene with a purity of 99.9 % was produced annually. The
yield of ethylene over the reactor, with recirculation of ethanol, was 96.8 %. The investment
cost of the plant was calculated to 28,105,200 USD and a total production cost of 1,136
USD/tonne ethylene was obtained. The ethanol price and the ethylene price was assumed to
be 537 USD/tonne and 1,280 USD/tonne respectively. This resulted in a pay-back time of
1.75 years in the economical evaluation.
I
Contents
1 Introduction ........................................................................................................................ 1
1.1 Purpose ....................................................................................................................... 1
1.2 Disposition ................................................................................................................. 1
2 Literature review ................................................................................................................ 2
2.1 Ethylene production from bioethanol ......................................................................... 2
2.2 Bioethanol based on lignocellulosic biomass............................................................. 2
2.3 Potential contaminants formed in ethylene process ................................................... 5
2.4 Catalysts ..................................................................................................................... 5
2.5 Commercial processes available .............................................................................. 11
2.6 Ethylene purification ................................................................................................ 12
3 Discussion regarding literature study ............................................................................... 13
4 Feasibility study ............................................................................................................... 14
4.1 Process steps ............................................................................................................. 16
4.2 Design....................................................................................................................... 17
4.3 Cost estimates ........................................................................................................... 20
5 Discussion ........................................................................................................................ 28
6 Conclusions ...................................................................................................................... 30
7 Appendix .......................................................................................................................... I-i
Appendix I – Ethanol calculations ....................................................................................... I-i
Appendix II – Heat and mass balances ............................................................................... II-i
Appendix III – Reaction specifications .............................................................................. III-i
Appendix IV – Design equations and assumptions ............................................................IV-i
Appendix V – Calculation of unit operation design ............................................................ V-i
Appendix VI – Economy equations and method ...............................................................VI-i
Appendix VII – Grass Root Capital ................................................................................. VII-i
Appendix VIII – Operating costs .................................................................................... VIII-i
Appendix IX – Investment calculations .............................................................................IX-i
II
1 Introduction
One of the most important polymer precursors in today’s society is ethylene. The traditional
production method, steam cracking, is a thermal cracking process with no catalysts involved.
A common feedstock is naphtha but lighter hydrocarbons such as ethane can be used as well.
The produced ethylene is an important monomer for a variety of plastics [1]. An ethylene
producing company is Borealis, which currently utilizes the steam cracking technique. Since
fossil fuels eventually will run low and with a desire to contribute to a more sustainable
society, an increasing interest in converting 2nd generation bioethanol, derived from
lignocellulosic material, to ethylene has occurred.
1.1 Purpose
The aim of this study was to investigate the possibility of using second generation ethanol as a
feedstock for production of ethylene and to design a plant where this is possible. The report
identifies catalysts that can be used as well as potential contaminants and their effect on the
catalysts. A review of how a plant could be designed and economic aspects of the plant is also
presented.
1.2 Disposition
The study is divided into two parts, a literature review and a feasibility study. The literature
review shows an overall process design and which reactor types that can be used. A more
detailed description of bioethanol from lignocellulosic biomass and its contaminants are
presented as well. Possible catalysts are considered and some information about commercial
available processes is given. The first part is followed by a discussion regarding the literature
review. In the feasibility study a design of a process is presented together with an economic
evaluation. The report is finalized with a discussion and a conclusion. At the end, the
references are listed.
1
2 Literature review
2.1 Ethylene production from bioethanol
To produce ethylene from ethanol, acidic catalysts are often used. Homogeneous catalysts that
traditionally have been used are sulfuric acid or phosphoric acid. More modern catalysts are
acidic heterogeneous catalysts such as zeolites treated with e.g. alumina or manganese. These
catalysts provide a high selectivity to ethylene, which results in a relatively simple process. A
general process flow diagram is depicted in Figure 1 below, where the main parts needed for
ethylene production are presented. Ethanol with a certain composition is vaporized and
delivered to the reactor. The specification of the ethanol depends on the robustness of the
catalyst. After dehydration reaction in the reactor the gaseous product is separated from the
non-reacted ethanol. By-products are removed and transported to wastewater treatment. The
crude ethylene is separated from heavier products in a distillation tower and is then further
purified in a separate section [1].
2.2.1.1 Pretreatment
The pretreatment methods can be divided into physical, chemical, physicochemical and
biological methods [7]. Physical methods, such as milling, are generally too energy intense
which makes them unfeasible [8]. One method suitable for commercial use is dilute acid
pretreatment, which can be used on woody biomass like spruce and willow. This chemical
method aims to separate hemicellulose and lignin as well as open up the lignocellulosic
structure [9]. The major disadvantage with the dilute acid pretreatment, and other acid
pretreatment methods, is that a part of the sugars that could have been used for producing
ethanol is lost in formation of lignocellulose degradation by-products [10]. These by-products
can be divided into three main groups: furans, weak acids and phenolic compounds [9]. The
furans that exist after pretreatment are 5-hydroxymethyl-2-furaldehyde (HMF) and 2-
furaldehyde (furfural). These are formed by dehydration of hexoses respectively pentoses.
The concentration of furans is strongly dependent on how the dilute acid pretreatment is
performed, for example weather it is operated as a one-step or a two-step process. Three weak
acids that are generated during the lignocellulose treatment are acetic acid, levulinic acid and
formic acid. Acetic acid is formed by deacetylation of hemicellulose. Both formic and
levulinic acid are products from breakdown of HMF, but formic acid may also be created
from reaction of furfural in the acidic environment. The phenolic compounds that are formed
consist of a wide range of compounds. Lignin interacts in different ways in different
lignocellulosic materials, consequently the biomass source have big influence on the amount
3
and type of phenolic compounds which will form. Two of the more common phenolic
compounds that are formed from dilute acid pretreatment of spruce are vanillin and
syringaldehyde. The same groups of by-products as mentioned above have also been reported
for wet oxidation of wheat straw and steam pretreatment of sugar cane bagasse and corn
stover [9]. However, the amounts of by-products vary and not all of the above mentioned
compounds exist in these types of pretreatment.
A promising alkaline pretreatment method is AFEX (ammonia fiber explosion) in which the
biomass is treated with ammonia under pressure [6]. AFEX works better for biomass with low
lignin content and is therefore particularly suitable for agricultural residues rather than
softwoods. Unlike during acidic pretreatment no fermentation inhibiting by-products are
created during AFEX [7].
2.2.1.2 Hydrolysis
After the lignocellulosic structure has been opened up in the pretreatment, hydrolysis can be
performed. The hydrolysis aims at converting the polysaccharides into fermentable sugars and
can be performed both with enzymes or acid [5]. The polysaccharides are mainly cellulose,
but depending on pretreatment method also parts of the hemicellulose remains. The present
commercialization of enzymatic hydrolysis has preceded by a possibility to decrease enzyme
loadings and lowered enzyme production costs [6]. Novel enzymes have also shown to be
very efficient in the conversion of cellulose [15]. The enzymatic hydrolysis uses a mixture of
enzymes, mainly cellulases, to release the remainder of the sugars [6].
4
2.2.1.3 Fermentation
The sugars that are to be fermented are both hexoses (glucose, galactose, mannose, rhamnose)
and pentoses (xylose, arabinose) [9]. There are many microorganisms with potential of being
used in the fermentation process, and normal baker’s yeast (Saccharomyces cerevisiae) is one
of them [6]. It is well established in large scale production and has a high ethanol tolerance.
Another advantage is that S. cerevisiae have shown relatively good resistance against the by-
products that are formed in the pretreatment, which inhibit fermentation. A large disadvantage
with S. cerevisiae is that it does not naturally ferment pentoses. Since xylose (a pentose) is the
second most occurring sugar in many plants it is highly desirable to ferment it. Different
methods have been proposed for solving this problem. One is to separate the hemicellulose
from the cellulose and fermenting it with other microorganisms while another is to modify S.
cerevisiae [5, 14]. During the metabolism of S. cerevisiae a variety of by-products is formed.
These are acetaldehyde, acetic acid/acetate, glycerol and carbon dioxide [14]. Also methanol
[3], fusel oils (n-propanol, amyl alcohol, isoamyl alcohol, isobutanol, phenethyl alcohol etc.)
and ethyl acetate are created during ethanol fermentation [16]. Volatile sulphur compounds
that can be created are diethyl sulfide and dimethyl sulfide [16]. It is further known that
recombinant S. cerevisiae slowly may reduce the furaldehydes created in the dilute acid
pretreatment to their corresponding alcohols (2-furan methanol, furan 2,5-dimethanol) [14].
2.2.1.4 Purification
The finalization of the ethanol production is to separate and purify the ethanol from the
fermentation broth. The main technique used in this process step is distillation. At 95.6 wt%
ethanol the distillation reaches the azeotropic mixture of ethanol and water. When the
azeotrope is boiled the vapor has the same composition as the unboiled mixture, resulting in
no separation in a distillation column. To purify the ethanol beyond the azeotrope the ethanol
must be dehydrated. However, for economic reasons dehydration above azeotrope is not
always implemented [3]. Most of the by-products formed and minerals solved during
bioethanol production are removed in the distillation, leaving only traces behind.
Two ethanol molecules may react to form diethyl ether [17]. However, the diethyl ether is to
be considered more as an intermediate rather than a by-product [18]. Its formation is favored
mainly between 150 °C and 300 °C. The most important by-products are acetaldehyde and
hydrogen which are formed by decomposition of ethanol [17]. Other by-products that may be
formed in small quantities are alkanes and alkenes with low number of carbons (methane,
ethane, propane, butane, propylene, butylenes) and carbon monoxide [17, 18]. Also
hydrocarbons with five carbons or more may exist. Of all the hydrocarbons butylenes and
ethane are the most occurring in the ethylene product [19]. At too high temperatures the
amount of by-products like coke are increasing [17].
2.4 Catalysts
A catalyst is used to change the reaction kinetics of a reaction. It cannot change the
equilibrium of a reaction but it can change the rate of reaction toward the equilibrium. It
lowers the activation energy, which is needed for the reaction to occur. The catalyst is not
5
consumed during the reaction but it can be deactivated and loose its ability to catalyze the
wanted reaction. There are some concepts that are important when talking about catalysts. The
main three concepts in focus within this report are activity, selectivity and deactivation. The
activity is a measurement of how fast the reaction reaches the equilibrium. Selectivity
describes the capability to produce a desired product. Deactivation is when a catalyst loses its
ability to catalyze a reaction and becomes less active [20, 21]. Deactivation of the catalysts
will be discussed later in the report.
In the catalytic dehydration of ethanol to ethylene an acid catalyst, with weak, relatively
strong or strong acid sites, can promote the reaction [22]. Research has been done on different
types of catalysts and mainly three different groups of acid catalysts have been the focus in
this report. These are oxide catalysts (γ-Al2O3 based), molecular sieve catalysts (HZSM-5
zeolites) and heteropolyacids [22]. All of these can be modified and doped which results in
varying lifetime, conversion of ethanol and selectivity towards ethylene. See Table 2 for a
comparison of the different catalysts that are studied in this review.
To increase the ethanol conversion γ-Al2O3 can be modified in different ways, for example by
adding oxides like MgO/SiO2, Cr2O3, FeOx and TiO2 [22]. There are numerous of other ways
to improve the oxide catalysts and thus increase the ethanol conversion and selectivity. One of
the problems with this type of catalyst is the high reaction temperature required [24]. Another
problem is that water can deactivate active sites on γ-Al2O3 and inhibit the formation rate of
ethylene and diethyl ether [26].
2.4.2 Zeolites
Zeolites are crystalline aluminosilicates with a fine pore structure. The surface area of a
zeolite is approximately 900 m2/g. The pores can contain exchangeable cations, which can be
exchanged to alter the performance of the zeolite. By changing the size of the cations, the
cross-section of the pores is changed and thereby allowing different types of molecules to be
adsorbed and pass through. The catalytic behavior of the zeolite can also be changed by
substituting the cations [28].
ZSM-5 zeolites have a high Si/Al ratio and have been used to catalyze the dehydration of
ethanol to ethylene [22, 28]. As well as the oxide catalysts, zeolites can be doped to enhance
the activity or to make it more stable. The stability is mainly affected by the acidity. High
acidity increases coke formation, which deactivates the catalyst. With this said one of the
difficulties is to find a way to reduce the acidity but still be able to keep a good conversion
and selectivity of ethylene [24]. Studies regarding these difficulties have been made on
different types of catalysts, which will be discussed below.
2.4.2.1 HZSM-5
HZSM-5 is a commercial zeolite that is in focus. It is able to catalyse the dehydration of
ethanol at lower temperatures, around 300 °C [24]. According to Zhang et al. [23] a
conversion of 98 % ethanol and ethylene selectivity of 95 % was achieved with HZSM-5 at
300 °C. The disadvantage with HZSM-5 is the high acidity, which promotes coking and
lowers the stability and lifetime of the catalyst. Desorption of NH3 is one method that can be
used to estimate the acidity of a catalyst. NH3 is adsorbed and corresponds to the amounts of
acid present in the catalyst; it is then desorbed by changing the temperature. This can be
made with a temperature-program which gradually increases the temperature. A higher
desorption temperature relates to stronger acid strength. HZSM-5 shows both strong and weak
sites and has a high desorption temperature. One conclusion from this is that the acid property
of HZSM-5 is high and also the activity, since the activity of the catalyst is dependent on the
acid strength and the amount of acid sites [23]. The stability on the other hand is dependent on
the deactivation of the catalyst. A strong acid site can polymerize the formed ethylene to
higher olefins and aromatics. Because of the fine microporous structure and the fact that these
substances can form structures that are not gaseous they will not be able to pass through.
Consequently coke can be formed and cover the active sites in the catalyst, which leads to
deactivation of the catalyst [29].
7
had the best catalytic ability at relatively low temperatures and also better stability than the
original HZSM-5 [31].
2.4.3 Heteropolyacids
Heteropolyacids (HPAs) has become of great interest in the category of future catalyst. The
most known form is the Keggin type, with the general formula 𝑋𝑀12 𝑂40 , where X is the
heteroatom and M the addendum atom. 𝐻3 𝑃𝑊12 𝑂40 and 𝐻4 𝑆𝑖𝑊12 𝑂40 are two examples that
are commercially available, where 𝐻3 𝑃𝑊12 𝑂40 is the most acidic, regarded as a very strong
acid [33].
The main topic of interest in the research of new HPA catalysts is the economic benefit [34].
The economic benefits provided for the conversion of ethanol to ethylene is primarily the low
operating temperature, approximately 190 °C [24, 35]. Though the low operating
temperatures achieve large benefits, high temperatures are a problem because of the low
thermal stability of HPA catalysts.
8
Another disadvantage with HPAs of the Keggin structure is the low surface area provided by
the crystal structure. This has been overcome by adding HPAs on supports with larger surface
area [37]. Though any better thermal stability of the catalyst is not always achieved, some
research has proven an increased thermal stability of the Keggin structure when added to
support materials [38]. Promising heteropolyacids added on support materials are mainly the
TPA-MCM-41 and STA-MCM-41, seen in Table 2, where MCM-41, which stands for
“Mobile Composition of Matter”, is the support. MCM-41 is a microporous silicate support
with high surface area, approximately 1,000 m2/g [39].
Some of the most promising heteropolyacids can be seen in Table 2; where especially
𝐴𝑔3 𝑃𝑊12 𝑂40 has proven high selectivity to ethylene production in the dehydration of ethanol.
One of the reasons for the exchange of hydrogen to silver as cation is to construct a water-
insoluble salt, which then is not affected by the water being produced in the process [40].
2.4.4 Coking
Coking is a severe deactivation issue for most heterogeneous acid catalysts. Coking occur
mainly due to the oligomerisation and polymerization, linked to the dehydrogenation of
hydrocarbons, as seen in reaction 1.
The higher ratio of carbon to hydrogen n/m in the hydrocarbon, and thus the more double
bonds, the more it tends to oligomerise. Thus the main issue of coking, in the conversion of
ethanol to ethylene, is the ethylene itself, which is very reactive [41].
To hinder the conversion of ethylene into coke the residence time, for the ethylene product,
has to be kept as low as possible [42]. This is one of the main problems considering zeolites,
as the micro-pores of the zeolite adds diffusion resistance and such the residence time
increases and coking evolves. As for most solid acid catalysts the strong acidity is a
contributing factor on coke formation as the reactivity increases with acidity and thus also the
chance of ethylene to react further. This can be suppressed by either decreasing the residence
time or by pretreating the catalyst to lower the reactivity [29]. Coke formation can also be
suppressed by adding water, or other substances, in the feed to compete with the precursors
[17].
9
Table 2. Comparison of the studied catalysts. Ethanol is either being inserted with nitrogen gas or air. The
reaction temperature is the lowest required to achieve the given selectivity and conversion.
Catalyst Max Ethanol Reaction GHSVa/ Comments Ref
ethylene conversion temperature WHSVb
selectivity
HZSM-5 96.4 % 100 % 260 °C 2 h-1 b 50 wt% ethanol [31]
Lab scale
0.5 g catalyst
0.5% La 2% 99.9 % 100 % 240 °C 2 h-1 b 50 wt% ethanol [31]
P HZSM-5 Lab scale
0.5 g catalyst
Nano HZSM- 98.4 % 98.4 % 240 °C 1 h-1 b 95 wt% ethanol [32]
5 Lab scale
1 g catalyst
Run time: 620 h
0.4 mol/l 99.6 % 99.7 % 265 °C 2.37 h-1 b 20 w% ethanol [29]
NaOH - Lab scale
HZSM-5
1 g catalyst
Run time: 350 h
Ag3PW12O40 99.2 % 100 % 220 °C 6000 h-1 a 15 g/m3 ethanol [43]
Lab scale
0.5 cm3 catalyst
Run time: 1 h
TPA-MCM- 99.9 % 98 % 300 °C 2.9 h-1 b 99.98 % ethanol [39]
41 Lab scale
0.2 g catalyst
Run time: 2-3 h
STA-MCM- 99.9 % 99 % 250 °C 2.9 h-1 b 48 % ethanol [44]
41 Lab scale
0.2 g catalyst
Run time: 8 h
P90 (Sasol) 99.6 % 100.0 % 400 °C 1.43 h-1 b 7.9 wt % ethanol [27]
θ-Al2O3 Lab scale
0.5 g catalyst
P200 (Sasol) 100.0 % 99.7 % 350 °C 1.43 h-1 b 7.9 wt % ethanol [27]
γ, δ-Al2O3 Lab scale
0.5 g catalyst
a
Gas Hourly Space Velocity
b
Weight Hourly Space Velocity
10
V200 (UOP) 99.4 % 99.8 % 400 °C 1.43 h-1 b 7.9 wt % ethanol [27]
γ-Al2O3 Lab scale
0.5 g catalyst
D100 99.9 % 99.3 % 350 °C 1.43 h-1 b 7.9 wt % ethanol [27]
(Degussa/ Lab scale
Evonik)
0.5 g catalyst
γ-Al2O3
SA330 100.0 % 99.8 % 350 °C 1.43 h-1 b 7.9 wt % ethanol [27]
(Strem) Lab scale
amorphous 0.5 g catalyst
11
Figure 3. Flow sheet of ethylene production – Chematur [19].
In the ethylene production process from Chematur the purification of ethylene mainly consists
of a quench column, a caustic wash column, dryers, an ethylene column and a stripper [19].
The only information that was found about the purification section in the Atol process from
Axens was that it is simplified and contains no caustic tower or C2 splitter, which separates
compounds of two carbons [49]. The purification in the Hummingbird process from BP is
simplified [46].
12
3 Discussion regarding literature study
Most of the by-products formed and minerals solved during bioethanol production are
removed in the distillation, leaving only traces behind. Small amounts of a contaminant can
however still be a problem. It is desirable to have a continuous process operating during long
time intervals without downtimes. Having traces that deactivate the catalyst may hinder the
continuous operation, leading to a lower productivity. For most of the potential contaminants
mentioned in the literature study no information was found about how or if they affect the
catalysts. Thus it would be advantageous to further investigate how these trace compounds
influences the catalyst. This can be done in lab scale experiments.
In the zeolites there are exchangeable cations. When the ethanol contains inorganic ion
impurities from minerals, these ions can replace the zeolite ions. Due to this there are less
active sites, which results in lower activity of the catalyst. If the newly attached cation is too
large it is possible that it will cover larger parts of the zeolite. Consequently the cross
sectional area of the pores will be lowered and reduce the transportation in the particle. Trees
contain less inorganic material than agricultural crop residues. Therefore it could be an
advantage to use bioethanol from wood derived ethanol over bioethanol from agricultural
crops.
Coke is formed in the dehydration process, mainly due to polymerization of ethylene. In the
studied literature, coke is often mentioned as an important factor connected to the stability of
the catalysts. The coking is a central issue for all catalyst groups investigated. When building
a process it is relevant to take the coking into account and the regeneration step should be
considered carefully.
All catalysts that were studied seemed to have a high conversion of ethanol and high ethylene
selectivity. Alumina oxide catalysts are stable but require a high reaction temperature. The
zeolite catalysts can run at lower temperatures but without modifications they seem less
stable. The modifications that have been studied were only tested in lab scale; what happens
with phenomena like pressure drop and changes in mass transfer when the process is scaled
up is not known. Heteropolyacids has the possibility to provide good process conditions in the
conversion of ethanol to ethylene. Their advantage is the same as for the zeolites; high
activity because of the high acidity results in the possibility of low operating temperatures.
One disadvantage is the low thermal stability which will require gentle operating when
decoking. The main focus on the listed HPAs has been on ethanol conversion and the ethylene
selectivity. Further investigations should focus more on the long-term stability of the catalyst
to know for certain if these types of catalysts are appropriate for the process or not.
13
4 Feasibility study
The process mainly consists of one reactor and subsequent purification, see the block diagram
in Figure 4. Unreacted ethanol that is recycled in order to obtain a higher conversion of
ethanol is also presented in Figure 4.
Figure 4. Block diagram illustrating the two main parts of the process, reactor and subsequent purification.
Besides ethylene, the raw ethylene stream shown in Figure 4 contains water, water soluble
compounds, carbon dioxide, hydrocarbons heavier than ethylene, carbon monoxide and
hydrogen. These impurities are separated in five units in order to obtain polymer grade
ethylene. The units needed for purification can be seen in Figure 5. A flowsheet over the
process is presented in Figure 6.
Figure 5. Block diagram showing the purification units used in the process.
14
Ethanol Heat
Reaction Flue
storage tank 1 exchanger 2 27
5 column gas
4
Oil heater
6
26
3 NaOH (aq)
Heat storage tank 13 Oxygen
exchanger 1
14
12 15 Pure water
10 7 17
Distillation 1
18 25
21
20
24
11
22
19
Heat Heat
Waste water exchanger 4 exchanger 5
Ethylene C2 Ethylene
treatment 23
column stripper storage
tank
4.1.2 Pumps
Pumps were placed in the system to create a flow through the process. The pumps assure a
pressure of 1 bar throughout the system. The pumps are also designed to be able to raise the
pressure in the system to 10 bar, if this would be desirable.
4.1.4 Reactor
The reactor used in this process design is a fixed bed reactor with isotherm heat supplement.
This requires a tube bundle but enables to control the reaction temperature closely. The tube
bundle of the reactor is filled with heteropolyacid catalyst that favors the reaction from
ethanol to ethylene. It is important to keep the reaction temperature constant and not too high.
The reactor is kept at constant temperature by having a jacket of circulated heating media, see
section 4.1.5.
16
4.1.8 Caustic soda wash
A caustic soda wash, an absorption tower using sodium hydroxide, is used. Carbon dioxide is
separated from the ethylene stream by letting the carbon dioxide react with sodium hydroxide.
A 50 wt % NaOH-water solution is used. [50]
4.2 Design
To calculate the designs of each operating unit in the process seen in Figure 6, heat and mass
balances were obtained by simulating the system in Aspen Plus V8.2 – AspenONE.
To produce 100,000 tonnes of ethylene annually a certain amount of ethanol is needed. The
amount of ethanol depends on the estimated conversion and selectivity.
17
In calculations the conversion of ethanol was set to 95 mole-% and the selectivity to ethylene
was set to 98 mole-%. The calculations resulted in that 192,000 tonnes of 95 wt-% ethanol
was needed. For calculations see Appendix I. To be sure that the right amount of ethylene is
produced, a higher amount of ethanol was used when modelling the process. The yield of
ethylene over the reactor, with recirculation of ethanol, was 96.8 %. For amounts given from
Aspen see Appendix II, Table 10.
18
Condenser area: 100 m2 Bottom temperature: 100 °C
2
Reboiler area: 161 m
Dryer, 2 pieces Inside diameter: 2.5 m Residence time: 12 h
Height: 5.94 m
Adsorbent: Zeolite
Adsorbent mass: 16,970 kg
Packing height: 4.94 m
Heat exchanger 4 Area: 62 m2 Hot stream: 32 – -104 °C
Cold stream: Cooling media,
-150 °C
Ethylene column Inside diameter: 1.49 m Nr of ideal stages: 15
Height: 11.3 m Top temperature: -104.2 °C
Condenser area: 216 m2 Bottom temperature: -98.3 °C
2
Reboiler area: 50 m
Heat exchanger 5 Area: 511 m2 Hot stream: -104 – -130 °C
Cold stream: Cooling media,
-150 °C
C2 stripper column Inside diameter: 2.30 m Nr of ideal stages: 10
Height: 7.5 m Top temperature: -109.7 °C
2
Condenser area: 13 m Bottom temperature: -103.9 °C
2
Reboiler area: 9.0 m
Ethylene storage tank Total vessel volume: 4,920 m3
Heat pump Heat absorption rate: 3,710 kW
Wastewater treatment Capacity: 0.0033 m3/s
plant
Thermal fluid heater Heating duty: 10,160 kW
19
4.3 Cost estimates
In order to determine the feasibility of the project, an economical evaluation was made. The
evaluation covered the capital cost needed to build the plant, annual operating costs and a
sensitivity analysis of how the pay-back time varies on feedstock cost, investment cost and the
price of ethylene. The initial case description was based on the following: an economic life
time of 15 years, a weighted average cost of capital of 10 % and an ethanol price of 537
USD/tonne [51]. The ethylene price was assumed to be 1,280 USD/tonne. The initial case
description can be seen in Table 6 and a summary of the total production cost in Table 7.
20
The grass root capital for the plant was calculated to 28,105,200 USD, which includes the
fees, contingencies and auxiliary facilities. How the costs are distributed can be seen more
clearly in Figure 7 and 8. The largest cost in the grass root capital was the heat pump, which
accounted for 46 % of the total Bare-Module Cost. The heat pump is necessary to reduce to
electrical costs and creates a more profitable process.
1%
99%
Figure 8. Detailed description of how the unit operation investment costs are distributed.
21
4.3.2.1 Fixed capital
The fixed costs include the cost of storage for both feedstock and products and also the cost
for spare parts. Feedstock and products is assumed to be stored for 10 days and the cost for
spare parts have been set to 15 % of the cost for maintenance and repair. The total amount of
fixed costs sums up to 884,360 USD. The calculations can be seen in Appendix VIII, Table
43.
Operating costs
Other operating
89% costs
Figure 9. Distribution between operating costs for the process, the raw material is ethanol.
22
Other operating costs
Natural gas
5%7% Electricity
30% 7% Storage & spare parts
14% Maintenance & repair
Personnel
20% 17%
Other materials
Annual capital cost
Figure 10. Detailed description of how the other operating costs than raw material is distributed.
23
4.3.4 Investment calculations
An investment calculation was made to determine the pay-back time for the plant. Pay-back
time is a relative method which compounds the costs and revenues. The price of ethylene has
been set according to market prices and is 1,280 USD/tonne. This price is slightly higher than
the calculated production cost. With the given price of ethylene, the pay-back time is
calculated to 1.75 years. The interest rate has been chosen to 10 %. Table 8 displays a
summary of the calculations and the equations that were used can be seen in Appendix IX.
Table 8. Summary of investment calculations and pay-back time.
Cost
Grass root capital 28,150,200 USD
Annual operating costs 112,204,680 USD/year
Revenue 130,542,800 USD/year
Pay-back time 1.75 years
14
12
Pay-back time (years)
10
0
-50% -40% -30% -20% -10% 0% 10% 20%
Ethanol price
Figure 11. Sensitivity analysis of how the pay-back time changes with different ethanol prices.
24
There is a large difference between the two scenarios and compared to the regular price of
ethanol, an increase of ethanol price has a very large impact on the pay-back time and the
annual profit. As can be seen in Figure 11, an increase in ethanol price makes a larger impact
than a decrease. If the ethanol price increases with more than 18 % there will be no annual
profit. The operating costs will be higher than the revenues and the plant will not be
profitable. How the annual profit varies with ethanol price can be seen in Figure 12.
60.000.000
50.000.000
40.000.000
Profit (USD/year)
30.000.000
20.000.000
10.000.000
0
-50% -40% -30% -20% -10% 0 5% 10% 15% 20% 30%
-10.000.000
-20.000.000
Ethanol price
Figure 12. Sensitivity analysis of how the profit changes with different ethanol prices.
When adding 25 % to the price of ethylene the pay-back time decreases to 0.6 years compared
to 1.75 years for the original price. With the premium, the ethanol price could increase up to
50 % and the process would still generate an annual profit. The results of how pay-back time
changes with ethylene price can be seen in Figure 13. The price has been increased with 20-30
% and also decreased with 10 %, the results are compared to the original price of ethylene.
25
Sensitivity analysis of changes in ethylene price
130% 0,5
125% 0,6
Ethylene price
120% 0,7
100% 1,8
90% 8,0
0,0 1,0 2,0 3,0 4,0 5,0 6,0 7,0 8,0 9,0
Pay-back time (years)
Figure 13. Sensitivity analysis of how the pay-back time changes with different changes of ethylene price.
Finally, a sensitivity analysis of the grass root capital was made. Ulrich's method does not
offer an exact accuracy, for example when calculating the installation cost. The installation
cost for the process is most likely higher than the one Ulrich's method has provided. It is
therefore interesting to see how the grass root capital will change in size and how this can
affect the process economy and pay-back time. See results in Figure 14 and 15.
25.000.000 Cost of
20.000.000 capital
15.000.000
10.000.000
5.000.000 Annual
- net
-30% 0 30% 50% 75% income
Figure 14. Sensitivity analysis of how the grass root capital, capital cost and annual net income varies when the
grass root capital increases and decreases.
As can be seen in Figure 14 when the grass root capital increases it will result in a higher
capital cost each year. The capital cost is calculated with the grass root capital and annuity
factor, see Appendix IX. Its size will reflect on the annual net income, which is the annual
profit minus the capital cost, and lower it. Figure 15 displays the pay-back time for the
26
different grass root capital costs. A higher cost consequently means a longer pay-back time
for the plant.
2,50 Pay-
2,00 back
time
1,50
1,00
0,50
-
-30% 0 30% 50% 75%
Grass root capital change
Figure 15. Sensitivity analysis of how the pay-back time changes with changes of the grass root capital costs.
27
5 Discussion
The reaction is preferably run at as low a pressure as possible to push the equilibrium as far as
possible towards ethylene production. This has been overcome by running the simulations in
Aspen at the pressure of one bar. Other parts of the process such as the quench, the absorber
and the cryogenic distillation towers will on the other hand run optimal at higher pressures, to
minimize the cooling duty required in these systems. To improve the system a higher pressure
would preferably be set directly after the ethanol storage tank with the advantage of pumping
liquid ethanol compared to compressing gases in the downstream processing. This would
cause the reaction to not reach the optimal equilibrium, but also leads to a lesser amount of
catalyst needed because of the increase in concentration and thus an increase in reaction rate.
Since the lowered conversion and selectivity is included when calculating the mass balances
the higher pressure is considered to have a positive effect on the energy consumption, in the
downstream processing and the storage of ethylene. The size of each process equipment could
also become smaller. The usage of a higher pressure has been rejected in this study because of
the problem to make the simulations converge. This results in that the operating costs and
catalyst cost probably are estimated higher than the actual cost, which is considered better
than the opposite case.
In the process butylenes are produced as by-products. These are, due to their favorable
heating value, burnt in a thermal fluid heater to supply the process with heat. However,
butylene is a valuable chemical and it would therefore be more desirable to separate and
purify it rather than burning it.
The quench heat exchanger and the gas/liquid separator used in the process operates with
large amounts of water. To make the separation more energy efficient the separation could
have been performed in many steps connected in series. Removing more water increases the
carbon dioxide concentration which should make the absorber more efficient.
The sodium hydroxide scrubber, operating to remove the CO2 in the process, produces waste
water that is sent to wastewater treatment. In reality the sodium hydroxide would need to be
regenerated to make the process more sustainable. As Zeman and Lackner [50] explains the
produced sodium carbonate (Na2CO3) can be mixed with calcium hydroxide (Ca(OH)2) to
recover sodium hydroxide, see reaction 2. This procedure also creates a precipitate of calcium
carbonate (CaCO3). This precipitate can be filtered away, dried, washed and thermally
decomposed, see reaction 3. The calcium carbonate is decomposed to lime (CaO), which then
can be hydrated to produce calcium hydroxide to close the circle, see reaction 4. The
recycling of sodium hydroxide is well known in the pulp and paper industry (Kraft process).
𝑁𝑎2 𝐶𝑂3 (𝑎𝑞) + 𝐶𝑎(𝑂𝐻)2 (𝑠) → 2 𝑁𝑎𝑂𝐻(𝑎𝑞) + 𝐶𝑎𝐶𝑂3 (𝑠) (reac. 2)
𝐶𝑎𝐶𝑂3 (𝑠) + 𝑒𝑛𝑒𝑟𝑔𝑦 → 𝐶𝑎𝑂(𝑠) + 𝐶𝑂2 (𝑔) (reac. 3)
𝐶𝑎𝑂(𝑠) + 𝐻2 𝑂(𝑙) → 𝐶𝑎(𝑂𝐻)2 (𝑠) (reac. 4)
Two dryers are used to remove the final water in the process. While one dryer is operating the
other one is regenerated. Regeneration can be performed thermally, removing the water from
the zeolite by vaporization. This requires heat, but the cost of this heat is not included in the
economical evaluation.
The simulation implemented in Aspen is not optimized due to time constraints. By tweaking
the reflux ratios, distillation rates etc. of the distillation towers the energy requirements of
these could be lowered. The changes could have been done by using Aspen design
specifications. Implementing these improvements would probably lead to lower operational
28
costs due to lower energy requirements. Another thing that may be exaggerated is the number
of ideal trays in the absorption column. Lowering the number of trays would end up in a
smaller absorption tower and therefore lower investment cost.
The last two distillation columns and the last two heat exchangers in the process operate at
very low temperatures. To achieve this low temperatures a heat pump with cooling media
have to be used, for example pure ethylene. In the economical evaluation the cost of the
cooling media is not included. As mentioned above, there would be many advantages having
the process pressurized. Another advantage is that the last two distillation columns and the
last two heat exchangers could be operated at higher temperatures, making it easier to
construct the heat pump. This would also make the cost of the heat pump implemented in the
process more reliable. Ulrich method can’t handle heat pumps that works at temperatures
under -55 °C, but is still used even though the Aspen simulation displays cooling
requirements at under -100 °C.
The ethylene is stored in a tank at low temperature. The storage tank is not completely
adiabatic and will therefore receive heat from the surrounding. To keep the ethylene at low
temperature the tank needs to be cooled. The cost of this cooling is not included in the
economical evaluation.
The installation cost and maintenance and repair cost is dependent on how difficult it is to fill
and remove the catalyst from the reactor. Handling the catalyst will be more difficult if the
reactor has a tube bundle filled with catalyst instead of just using a packed reactor without
tube bundles. This is not something that Ulrich method takes into account. The reactor used in
the constructed ethylene process uses tube bundles and the cost for the reactor may therefore
be slightly underestimated.
The economic evaluation is rather optimistic and favorable given the current assumptions.
The database method that was used for calculating the capital costs is not completely reliable.
For example, as mentioned before, the installation cost will most likely be higher. Some
simplifications were made to run the simulation in Aspen. Therefore a couple of unit
operations have not been accounted for. Some understanding on how much the capital cost
could increase and affect the economy was given in the sensitivity analysis but the cost still
seems optimistic. An additional sensitivity analysis is recommended and possibly another
method to perform the cost estimations.
The sensitivity analysis was mainly made on pay-back time and it could be seen that the
process is sensitive for changes in market prices. If ethanol prices increase with more than 18
%, there will be no annual profit. It is not unlikely that the ethanol price increases and also, if
the ethanol is bought from outside the European Union there will be additional custom duty,
which consequently would increase the price. If it would be possible for the process, and
reactor, to run on ethanol with lower quality the ethanol price could be lower. This would
have a positive effect on the economy for the process.
As mentioned earlier there is a premium on green ethylene. The premium price is based on
what the customer is willing to pay for a green product and what the market looks like. There
is a demand for green ethylene and the price can therefore be kept high. If, on the other hand,
the production increases significantly the prices and possibly the premium will decrease.
29
6 Conclusions
Second generation bioethanol can contain contaminants that may have an impact on the
process. How these contaminants affect the catalysts’ performance in the long term has not
been found in the literature and therefore further investigations is needed in this area. Coke
formation is a problem in the dehydration process. Since none of the studied catalysts have
been used in industrial scale for manufacturing of green ethylene from 2nd generation
bioethanol the choice of catalyst is complex.
The properties of heteropolyacids enable the process to operate at relatively low temperatures.
This together with good performance was the reasons why this catalyst was chosen in the
process design. In broad terms the process contained one reactor and several purification
units. The suggested process was simulated successfully at one bar. An ethanol flow of
192,000 tonnes/year resulted in 102,000 tonnes ethylene/year with a polymer grade purity of
99.9 %.
The economical evaluation displayed a total production cost of 1,136 USD/tonne ethylene and
a pay-back time of 1.75 years. Another method is needed for the economic calculations in
order to determine if the numbers are reliable. According to the sensitivity analysis market
prices had a large impact on the profitability of the plant.
30
References
[5] A. M. Shupe och S. Liu, ”Ethanol fermentation from hydrolysed hot-water wood,”
Biomass and Bioenergy, vol. 39, pp. 31-38, 2012.
[8] M. Linde, ”Process Development of Bioethanol Production from Wheat and Barley
Residues,” Ph.D. dissertation, Dept. of Chemical Engineering, Lund University, Lund,
2007.
[11] S. Clarke, P.Eng. och F. Preto, ”Biomass Burn Characteristics,” 17 01 2014. [Online].
Available: http://www.omafra.gov.on.ca/english/engineer/facts/11-033.pdf. [Använd 27
02 2015].
31
&_schema=PORTAL. [Använd 27 02 2015].
[15] S. Macrelli, ”Ethanol from sugarcane lignocellulosic residues,” Ph.D. dissertation, Dept.
of Chemical Engineering, Lund University, Lund, 2014.
[16] K. Jacques, D. R. Kelsall och T. P. Lyons, The Alcohol Textbook: A reference for the
beverage, fuel and industrial alcohol industries, Third edition, Bath: Nottingham
University Press, 1999.
[22] M. Zhang och Y. Yu, ”Dehydration of Ethanol to Ethylene,” Ind. Eng. Chemistry Res.,
vol. 52, nr 28, p. 9505–9514, 2013.
[23] X. Zhang, R. Wang, X. Yang och F. Zhang, ”Comparison of four catalysts in the
catalytic dehydration of ethanol to ethylene,” Microporous and Mesoporous Materials,
vol. 116, pp. 210-215, 2008.
[24] D. Fan, D.-J. Dai och H.-S. Wu, ”Ethylene Formation by Catalytic Dehydration of
Ethanol with Industrial Considerations,” Materials, pp. 101-115, 2013.
[25] G. Chen, S. Li, F. Jiao och Q. Yuan, ”Catalytic dehydration of bioethanol to ethylene
over TiO2/γ-Al2O3 catalysts in microchannel reactors,” Catalysis Today, vol. 125, nr 1-
2, pp. 111-119, 2007.
32
[27] T. P. Phung, A. Lagaxxo, M. Á. Rivero Crespo och V. Sánchez Escribano, ”A study of
commercial transition aluminas and of their catalytic activity in the dehydration of
ethanol,” J. of Catalysis, vol. 311, pp. 102-113, 2014.
[28] L. Smart och E. Moore, Solid State Chemistry: An introduction, 2nd red., Cheltenham:
nelson thornes, 1996, pp. 239-262.
[29] Q. Sheng, S. Guo, K. Ling och L. Zhao, ”Catalytic Dehydration of Ethanol to Ethylene
over Alkali-Treated HZSM-5 Zeolites,” J. Brazilian Chemistry Soc., vol. 25, nr 8, pp.
1365-1371, 2014.
[30] K. Ramesh, L. M. Hui, Y.-F. Han och A. Borgna, ”Structure and reactivity of
phosporous modified H-ZSM-5 catalysts for ethanol dehydration,” Catalysis Commun.,
vol. 10, pp. 567-571, 2009.
[32] J. Bi, X. Guo, M. Liu och X. Wang, ”High effective dehydration of bio-ethanol into
ethylene over nanoscale HZSM-5 zeolite catalysts,” Catalysis Today, vol. 149, pp. 143-
147, 2010.
33
”Mesoporous silica directly modified by incorporation or impregnation of some
heteropolyacids: Synthesis and structural characterization,” Materials Res. Bulletin, vol.
46, pp. 19-25, 2011.
[39] A. Ciftci, D. Varisli, K. C. Tokay, A. N. Sezgi och T. Dogu, ”Dimethyl ether, diethyl
ether & ethylene from alcohols over tungstophosphoric acid based mesoporous
catalysts,” Chemical Eng. J., nr 207-208, pp. 85-93, 2012.
[41] F. Wang, M. Luo, W. Xiao, X. Cheng och Y. Long, ”Coking behavior of a submicron
MFI catalyst during ethanol dehydration to ethylene in a pilot-scale fixed-bed reactor,”
Appl. Catalysis A: General, vol. 393, pp. 161-170, 2011.
[42] A. G. Gayubo et al., ”Kinetic modelling for the transformation of bioethanol into olefins
on a hydrothermally stable Ni–HZSM-5 catalyst considering the deactivation by coke,”
Chemical Eng. J., vol. 167, pp. 262-277, 2011.
34
http://www.gep-
aftp.com/_upload/ressources/jah_2014/presentations/jah2014_atelier2_minoux_mallet.p
df. [Använd 17 3 2015].
[50] F. S. Zeman och K. S. Lackner, ”Capturing Carbon Dioxide Direcly From The
Atmosphere,” World Resource Review, vol. 16, nr 2, pp. 154-172, 2004.
[53] ”Sodium hydroxide - caustic soda flake pearl,” Alibaba, [Online]. Available:
http://www.alibaba.com/product-detail/sodium-hydroxide-caustic-soda-flake-
pearl_1816747740.html?s=p. [Använd 12 05 2015].
[56] C. R. Branan, ”Heat exchangers,” i Rules of Thumb for Chemical Engineers, Elsevier,
2005, pp. 29-58.
[60] Interra Global Corp, ”Molecular Sieve,” Interra Global Corp, 2015. [Online]. Available:
https://www.interraglobal.com/products/molecular-sieve/3a-molecular-sieve. [Använd
28 4 2015].
35
[61] M. Alveteg, Handbook, Lund: Dept. of Chemical Eng., 2013.
[62] H. T. Karlsson, Cost Estimation, Lund: Institutionen för kemiteknik, LTH, 2015.
[66] K. S. Knaebel, ”A "How To" Guide for Adsorber Design,” Adsorption Research, Inc.,
Dublin, Ohio.
36
7 Appendix
Appendixes to the report are presented below.
The molar ration between ethylene and ethanol is 1:1. Taking conversion of ethanol and
selectivity to ethylene into account the amount ethanol needed, expressed in mole, is
calculated in Equation 2.
𝑛𝑒𝑡ℎ𝑦𝑙𝑒𝑛𝑒
𝑛𝑒𝑡ℎ𝑎𝑛𝑜𝑙 = ↔ 𝑛𝑒𝑡ℎ𝑎𝑛𝑜𝑙 = 3.829 ∙ 109 𝑚𝑜𝑙𝑒 Equation 2
𝑋𝑒𝑡ℎ𝑎𝑛𝑜𝑙 ∙𝑆𝑒𝑡ℎ𝑦𝑙𝑒𝑛𝑒
I-i
Appendix II – Heat and mass balances
Summary of the process streams generated by Aspen Plus V8.2 – AspenONE.
Table 10. Specifications from simulation in Aspen Plus V8.2 – AspenONE.
Stream ID 1 3 4 5 6 7 8 9 10 11 12
From Storage Heat Ex 1 3 & 10 Heat ex 2 Reaction Heat Ex 1 Heat Ex 3 G-L separator Distillation 1 Distillation 1 G-L separator
column
To 4 Heat ex 2 Reaction Heat Ex 1 Heat Ex 3 G-L separator Distillation 1 4 Absorber
column
Temperature C 10 81,00146 80,81845 240 240 70 25 25,00015 75,8087 99,35311 25,00015
Pressure BAR 1 1 1 1 1 1 1 1 1 1 1
Vapor Fraction 0 0,786987 0,794862 1 1 0,677533 0,467026 0 1 0 1
Liquid Fraction 1 0,213013 0,205138 0 0 0,322467 0,532974 1 0 1 0
Mass Flow KG/HR 22000 22000 22931,65 22931,65 22931,65 22931,65 22931,65 9978,459 931,3098 9047,149 12953,19
Component Mass
Flow
ETHYLENE KG/HR 0 0 1,96998 1,969984 12323,1 12323,1 12323,1 1,969911 1,969911 5,79E-23 12321,13
ETHANOL KG/HR 20900 20900 21732,96 21732,96 1085,719 1085,719 1085,719 1018,694 832,624 186,0701 67,02491
WATER KG/HR 1100 1100 1100,335 1100,335 9108,672 9108,672 9108,672 8861,41 0,3310225 8861,079 247,2627
HYDROGEN KG/HR 0 0 0,000174 0,000174 8,695498 8,695498 8,695498 0,000174241 0,000174241 0,00E+00 8,695324
PROPENE KG/HR 0 0 0,348436 0,348436 3,884474 3,884474 3,884474 0,3484269 0,3484269 1,22E-16 3,536047
ISOBUTYLENE KG/HR 0 0 70,63443 70,63443 259,2231 259,2231 259,2231 70,63458 70,63458 3,29E-09 188,5886
ETHANE KG/HR 0 0 0,573134 0,573134 21,62941 21,62941 21,62941 0,5731143 0,5731143 1,58E-21 21,0563
DIETHYL-ETHER KG/HR 0 0 1,329672 1,329672 2,160144 2,160144 2,160144 1,329756 1,32975 5,44E-06 0,830388
ACETALDEHYDE KG/HR 0 0 23,42929 23,42929 48,10788 48,10788 48,10788 23,43021 23,43021 4,79E-07 24,67768
CO KG/HR 0 0 3,81E-06 3,81E-06 0,00362 0,00362 0,00362 3,81E-06 3,81E-06 5,35E-25 0,003616
CO2 KG/HR 0 0 0,060798 0,060798 67,80514 67,80514 67,80514 0,0607944 0,0607944 1,39E-19 67,74435
METHANE KG/HR 0 0 0,007819 0,007819 2,650044 2,650044 2,650044 0,00781874 0,00781874 1,40E-23 2,642225
Component Mass
Fraction
ETHYLENE 0 0 8,59E-05 8,59E-05 0,537384 0,537384 0,537384 0,000197416 0,00211521 6,40E-27 0,951204
ETHANOL 0,95 0,95 9,48E-01 9,48E-01 0,047346 0,047346 0,047346 0,1020893 0,8940355 2,06E-02 0,005174
WATER 0,05 0,05 0,047983 0,047983 0,39721 0,39721 0,39721 0,888054 0,000355438 0,979433 0,019089
HYDROGEN 0 0 7,60E-09 7,60E-09 0,000379 0,000379 0,000379 1,75E-08 1,87E-07 0 0,000671
PROPENE 0 0 1,52E-05 1,52E-05 0,000169 0,000169 0,000169 3,49E-05 3,74E-04 1,35E-20 0,000273
ISOBUTYLENE 0 0 3,08E-03 3,08E-03 0,011304 0,011304 0,011304 7,08E-03 0,0758443 3,64E-13 0,014559
ETHANE 0 0 2,50E-05 2,50E-05 0,000943 0,000943 0,000943 5,74E-05 0,000615385 1,75E-25 0,001626
DIETHYL-ETHER 0 0 5,80E-05 5,80E-05 9,42E-05 9,42E-05 9,42E-05 1,33E-04 0,00142783 6,01E-10 6,41E-05
ACETALDEHYDE 0 0 1,02E-03 1,02E-03 2,10E-03 2,10E-03 2,10E-03 0,00234808 0,0251583 5,30E-11 1,91E-03
CO 0 0 1,66E-10 1,66E-10 1,58E-07 1,58E-07 1,58E-07 3,82E-10 4,09E-09 5,91E-29 2,79E-07
CO2 0 0 2,65E-06 2,65E-06 2,96E-03 2,96E-03 2,96E-03 6,09E-06 6,53E-05 1,54E-23 5,23E-03
METHANE 0 0 3,41E-07 3,41E-07 0,000116 0,000116 0,000116 7,84E-07 8,40E-06 1,55E-27 0,000204
Mole Flow KMOL/HR 514,7262 514,7262 534,7337 534,7337 981,0151 981,0151 981,0151 515,9036 20 495,9036 465,1115
Component Mole
Fraction
ETHYLENE 0 0 0,000131 0,000131 0,447768 0,447768 0,447768 0,000136109 0,00351096 4,16E-27 0,944283
ETHANOL 0,881375 0,881375 0,88221 0,88221 0,024023 0,024023 0,024023 0,0428613 0,9036697 0,008145 0,003128
WATER 0,118625 0,118625 0,114221 0,114221 0,515393 0,515393 0,515393 0,9534398 0,000918727 0,991855 0,029509
II-i
HYDROGEN 0 0 1,62E-07 1,62E-07 0,004397 0,004397 0,004397 1,68E-07 4,32E-06 0,00E+00 0,009274
PROPENE 0 0 1,55E-05 1,55E-05 9,41E-05 9,41E-05 9,41E-05 1,60E-05 0,000413999 5,87E-21 0,000181
ISOBUTYLENE 0 0 0,002354 0,002354 0,00471 0,00471 0,00471 0,00244021 0,0629457 1,18E-13 0,007227
ETHANE 0 0 3,56E-05 3,56E-05 0,000733 0,000733 0,000733 3,69E-05 9,53E-04 1,06E-25 0,001506
DIETHYL-ETHER 0 0 3,35E-05 3,35E-05 2,97E-05 2,97E-05 2,97E-05 3,48E-05 0,000896991 1,48E-10 2,41E-05
ACETALDEHYDE 0 0 0,000995 0,000995 0,001113 0,001113 0,001113 0,00103093 0,0265931 2,19E-11 0,001204
CO 0 0 2,54E-10 2,54E-10 1,32E-07 1,32E-07 1,32E-07 2,63E-10 6,80E-09 3,85E-29 2,78E-07
CO2 0 0 2,58E-06 2,58E-06 1,57E-03 1,57E-03 1,57E-03 2,68E-06 6,91E-05 6,38E-24 3,31E-03
METHANE 0 0 9,11E-07 9,11E-07 0,000168 0,000168 0,000168 9,45E-07 2,44E-05 1,76E-27 0,000354
Volume Flow m^3/HR 27,09045 11933,82 12515 22814,41 41854,99 18969,93 11368,06 10,30996 580,2716 9,906551 11529,74
Stream ID 14 15 16 17 18 19 21 20 22 23 24 25 26 27
From Absorber Absorber Dryer Dryer Heat ex 4 Ethylene Ethylene Heat ex 5 C2 Stripper C2 Stripper 21 & 24 Oil heater
column column
To Absorber Dryer Heat ex 4 Ethylene 25 Heat ex 5 C2 Stripper Ethylene 25 Oil heater Oil heater
column storage tank
Temperature C 25 33,24234 29,67447 32,31291 32,31291 -104 -98,3353 -104,149 -130 -103,9743 -109,7252 -104,3031 20 1252,258
Pressure BAR 1 1 1 1 1 1 1 1 1 1 1 1 1 1
Vapor Fraction 0 1 0 0 1 0,457764 0 1 0,0118454 0 1 0,463977 1 1
Liquid Fraction 1 0 1 1 0 0,542236 1 0 0,9881545 1 0 0,536023 0 0
Mass Flow KG/HR 2000 12823,71 2129,477 247,1647 12576,55 12576,55 627,7206 11948,83 11948,83 11642,29 306,5347 934,2553 5759,784 6694,039
Component Mass
Flow
ETHYLENE KG/HR 0 12299,55 21,57973 0 12299,55 12299,55 362,0619 11937,49 11937,49 11642,27 295,2183 657,2802 0 2,60E-35
ETHANOL KG/HR 0 31,08229 35,94262 0 31,08229 31,08229 3,11E+01 2,49E-97 2,49E-97 0 0 31,08229 0 2,71E-34
WATER KG/HR 1500 247,1647 1527,776 247,1647 0 0 0 0 0 0 0 0 0 1264,989
HYDROGEN KG/HR 0 8,695317 6,77E-06 0 8,695317 8,70E+00 5,94E-14 8,695317 8,70E+00 7,21E-07 8,695316 8,695316 0,00E+00 5,95E-04
PROPENE KG/HR 0 3,517747 0,018299 0 3,517747 3,517747 3,52E+00 4,96E-21 4,96E-21 0 0 3,517747 0 4,37E-36
ISOBUTYLENE KG/HR 0 185,4287 3,159874 0 185,4287 185,4287 1,85E+02 6,73E-37 6,73E-37 0 0 185,4287 0 4,23E-35
ETHANE KG/HR 0 21,012 0,044293 0 21,012 21,012 21,00915 0,002858 0,0028576 2,86E-03 8,58E-09 21,00915 0 1,04E-33
DIETHYL-ETHER KG/HR 0 0,776636 0,053753 0 0,776636 0,776636 7,77E-01 7,18E-59 7,18E-59 0 0 0,776636 0 5,54E-42
ACETALDEHYDE KG/HR 0 23,71615 0,961529 0 23,71615 23,71615 2,37E+01 2,44E-51 2,44E-51 0 0 23,71615 0 1,40E-33
CO KG/HR 0 0,003616 3,47E-07 0 0,003616 3,62E-03 1,13E-11 0,003616 3,62E-03 3,35E-06 0,00361278 0,003613 0,00E+00 2,14E-02
CO2 KG/HR 0 0,127987 0,000899 0 0,127987 0,127987 1,28E-01 5,79E-06 5,79E-06 5,79E-06 7,89E-12 0,127982 0 2832,458
METHANE KG/HR 0 2,639804 0,002421 0 2,639804 2,64E+00 1,56E-06 2,639802 2,639802 0,0223378 2,617464 2,617466 0 3,14E-25
H3O+ KG/HR 7,24E-15 0 2,57E-14 0 0 0 0 0 0 0 0 0 0 0
OH- KG/HR 212,614 0 160,3531 0 0 0 0 0 0 0 0 0 0 0
HCO3- KG/HR 0 0 0,00028 0 0 0 0 0 0 0 0 0 0 0
CO3-2 KG/HR 0 0 92,19792 0 0 0 0 0 0 0 0 0 0 0
NA+ KG/HR 287,386 0 287,386 0 0 0 0 0 0 0 0 0 0 0
OXYGE-01 KG/HR 0 0 0 0 0 0 0 0 0 0 0 0 5759,784 2596,57
Component Mass
Fraction
ETHYLENE 0 0,959125 0,010134 0 0,977975 0,977975 0,576788 0,999051 0,9990508 0,9999978 0,9630828 0,703534 0 3,88E-39
ETHANOL 0 0,002424 0,016879 0 0,002471 0,002471 0,049516 2,08E-101 2,08E-101 0 0 0,03327 0 4,04E-38
WATER 0,75 0,019274 0,717442 1 0 0 0,00E+00 0,00E+00 0 0 0 0 0 0,1889725
HYDROGEN 0 0,000678 3,18E-09 0 0,000691 0,000691 9,46E-17 0,000728 0,0007277 6,20E-11 0,0283665 0,009307 0 8,89E-08
PROPENE 0 0,000274 8,59E-06 0 0,00028 2,80E-04 0,005604 4,16E-25 4,16E-25 0 0 0,003765 0,00E+00 6,54E-40
ISOBUTYLENE 0 0,01446 1,48E-03 0 0,014744 0,014744 2,95E-01 5,63E-41 5,63E-41 0 0 0,198478 0 6,32E-39
ETHANE 0 0,001639 2,08E-05 0 0,001671 0,001671 3,35E-02 2,39E-07 2,39E-07 2,45E-07 2,80E-11 0,022488 0 1,55E-37
II-ii
DIETHYL-ETHER 0 6,06E-05 2,52E-05 0 6,18E-05 6,18E-05 1,24E-03 6,01E-63 6,01E-63 0,00E+00 0 0,000831 0 8,27E-46
ACETALDEHYDE 0 1,85E-03 4,52E-04 0 1,89E-03 0,001886 3,78E-02 2,04E-55 2,04E-55 0 0 0,025385 0 2,09E-37
CO 0 2,82E-07 1,63E-10 0 2,88E-07 2,88E-07 1,79E-14 3,03E-07 3,03E-07 2,88E-10 1,18E-05 3,87E-06 0 3,20E-06
CO2 0 9,98E-06 4,22E-07 0 1,02E-05 1,02E-05 2,04E-04 4,84E-10 4,84E-10 4,97E-10 2,57E-14 0,000137 0,00E+00 4,23E-01
METHANE 0 2,06E-04 1,14E-06 0 2,10E-04 0,00021 2,49E-09 2,21E-04 2,21E-04 1,92E-06 0,00853888 0,002802 0 4,69E-29
H3O+ 3,62E-18 0 1,21E-17 0 0 0 0 0 0 0 0 0 0 0
NA2CO3 0,00E+00 0 0,00E+00 0 0 0 0 0 0 0 0 0 0 0
OH- 0,106307 0 0,075302 0 0 0 0 0 0 0 0 0 0 0
HCO3- 0 0 1,31E-07 0 0 0 0 0 0 0 0 0 0 0
CO3-2 0 0 4,33E-02 0 0 0 0 0 0 0 0 0 0 0
NA+ 0,143693 0 0,134956 0 0 0 0 0 0 0 0 0 0 0
OXYGE-01 0 0 0 0 0 0 0 0 0 0 0 0 1 0,3878928
Mole Flow KMOL/HR 108,2645 461,9393 109,9003 13,71972 448,2196 448,2196 18,21959 430 430 415 15 33,21959 180 215,7242
Component Mole
Fraction
ETHYLENE 0 0,949103 0,006999 0 0,978154 0,978154 0,708359 0,989586 0,9895856 0,9999964 0,7015537 0,705286 0 4,29E-39
ETHANOL 0 0,001461 0,007099 0 0,001505 0,001505 0,037031 1,26E-101 1,26E-101 0 0 0,02031 0 2,72E-38
WATER 0,769067 0,0297 0,771649 1 0 0 0 0 0 0 0 0 0 0,325497
HYDROGEN 0 0,009338 3,06E-08 0 0,009623 0,009623 1,62E-15 0,010031 0,0100311 8,62E-10 0,2875606 0,129845 0 1,37E-06
PROPENE 0 0,000181 3,96E-06 0 0,000187 0,000187 4,59E-03 2,74E-25 2,74E-25 0 0 0,002516 0 4,82E-40
ISOBU-01 0 0,007154 0,000512 0 0,007373 0,007373 0,181392 2,79E-41 2,79E-41 0 0 0,099486 0 3,50E-39
ETHAN-01 0 0,001513 1,34E-05 0 0,001559 1,56E-03 0,038348 2,21E-07 2,21E-07 2,29E-07 1,90E-11 0,021032 0,00E+00 1,60E-37
DIETH-01 0 2,27E-05 6,60E-06 0 2,34E-05 2,34E-05 5,75E-04 2,25E-63 2,25E-63 0 0 0,000315 0 3,46E-46
ACETA-01 0 0,001165 0,000199 0 0,001201 0,001201 2,95E-02 1,29E-55 1,29E-55 0 0 0,016206 0 1,47E-37
CO 0 2,79E-07 1,13E-10 0 2,88E-07 2,88E-07 2,21E-14 3,00E-07 3,00E-07 2,88E-10 8,60E-06 3,88E-06 0 3,55E-06
CO2 0 6,30E-06 1,86E-07 0 6,49E-06 6,49E-06 1,60E-04 3,06E-10 3,06E-10 3,17E-10 1,19E-14 8,75E-05 0 0,2983426
METHA-01 0 0,000356 1,37E-06 0 0,000367 0,000367 5,35E-09 3,83E-04 0,0003827 3,36E-06 0,010877 0,004911 0 9,06E-29
DOWTH-01 0 0,00E+00 0,00E+00 0 0,00E+00 0,00E+00 0,00E+00 0,00E+00 0,00E+00 0,00E+00 0,00E+00 0 0,00E+00 0,00E+00
H3O+ 3,52E-18 0,00E+00 1,23E-17 0 0,00E+00 0 0,00E+00 0,00E+00 0,00E+00 0,00E+00 0,00E+00 0 0 0
NA2CO3 0 0 0,00E+00 0 0 0,00E+00 0 0 0,00E+00 0 0 0 0 0
OH- 0,115466 0 0,085788 0 0 0 0 0 0 0 0 0 0 0
HCO3- 0,00E+00 0 4,17E-08 0 0 0 0 0 0 0 0 0 0 0
CO3-2 0 0 0,01398 0 0 0 0 0 0 0 0 0 0 0
NA+ 0,115466 0 0,113748 0 0 0 0 0 0 0 0 0 0 0
NAOH 0 0 0,00E+00 0 0 0 0 0 0 0 0 0 0 0
OXYGE-01 0 0 0 0 0 0 0 0 0 0 0 0 1 0,3761555
Volume Flow m^3/HR 1,573762 11702,71 1,709154 0,250685 11383,52 2897,645 0,999688 6042,049 80,38169 20,48199 203,8151 217,3445 4387,223 27359,74
II-iii
Appendix III – Reaction specifications
Reactor
The composition of the stream after the reactor is based on data from Chematur [19]. To
obtain this composition reactions have been set up. Two Aspen reactors, RStoic and REquil,
are used to simulate the real process reactor. Reactions specified in the RStoic reactor and the
selectivities for these are displayed as reaction 1-7 in Table 11. A REquil reactor is used to
balance the carbon dioxide and the carbon monoxide, in order to make it coincide with
Chematur data. The reaction for this reactor is shown as reaction 8 in Table 11.
Table 11. Reactions specified in Aspen reactors.
# Reaction Selectivity (%)
1 𝐶2 𝐻5 𝑂𝐻 → 𝐶2 𝐻4 + 𝐻2 𝑂 98.0
2 2 𝐶2 𝐻5 𝑂𝐻 → (𝐶2 𝐻5 )2 𝑂 + 𝐻2 𝑂 0.005
3 2 𝐶2 𝐻5 𝑂𝐻 → 𝐶4 𝐻8 + 2 𝐻2 𝑂 1.5
4 𝐶2 𝐻5 𝑂𝐻 → 𝐶2 𝐻4 𝑂 + 𝐻2 0.125
5 2 𝐶2 𝐻5 𝑂𝐻 → 𝐶3 𝐻6 + 𝐶𝑂2 + 3 𝐻2 0.0375
6 2 𝐶2 𝐻5 𝑂𝐻 → 𝐶2 𝐻6 + 2 𝐶𝑂 + 3 𝐻2 0.3125
7 2 𝐶2 𝐻5 𝑂𝐻 → 3 𝐶𝐻4 + 𝐶𝑂2 0.02
8 𝐶𝑂 + 𝐻2 𝑂 ↔ 𝐶𝑂2 + 𝐻2 equilibrium
III-i
Appendix IV – Design equations and assumptions
The equations and assumptions used in the dimensioning calculations for the different unit
operations are presented below.
Reactor
The tubes in the reactor were set to be 3 m long and have a diameter of 3 cm. By assuming an
ethanol space velocity of 6000 GHSV [40], the minimum amount of catalyst required could
be calculated:
𝐹𝐸𝑡𝑂𝐻
𝑉𝑐𝑎𝑡 = Equation 4
𝐺𝐻𝑆𝑉
With the total number of tubes the cross sectional area of all the tubes was calculated with
Equation 7:
2
𝐴𝑡𝑢𝑏𝑒𝑠 = 𝑛𝑡𝑢𝑏𝑒𝑠 ∙ 𝜋𝑟𝑡𝑢𝑏𝑒 Equation 7
Where rtube is the radius of one tube. To be able to design the reactor, the packing of the tubes
inside the reactor had to be assumed. This packing factor, 𝑝𝑓𝑡𝑢𝑏𝑒 , of the tubes was set to 0.25.
The total reactor cross sectional area was calculated by using Equation 8 and the diameter of
the reactor could then be calculated with Equation 9.
𝑛𝑡𝑢𝑏𝑒𝑠 ∙𝐴𝑡𝑢𝑏𝑒𝑠
𝐴𝑟𝑒𝑎𝑐𝑡𝑜𝑟 = Equation 8
𝑝𝑓𝑡𝑢𝑏𝑒
𝐴𝑟𝑒𝑎𝑐𝑡𝑜𝑟
𝑑𝑟𝑒𝑎𝑐𝑡𝑜𝑟 = 2 ∙ √ Equation 9
𝜋
When running at industrial scale, a guard bed is usually added to adsorb impurities and thus
the prolonging the reactor catalyst lifetime. Consequently, a guard bed was added and
assumed to be 20 % of the total reactor catalyst volume:
𝑉𝑔𝑢𝑎𝑟𝑑 𝑏𝑒𝑑 = 0.2 ∙ 𝑉𝑐𝑎𝑡 Equation 10
The height of the guard bed was calculated with Equation 11. A total height of the reactor was
then calculated with Equation 12. To the total height 0.5 m was added at the top and the
bottom of the reactor.
𝑉𝑔𝑢𝑎𝑟𝑑 𝑏𝑒𝑑
ℎ𝑔𝑢𝑎𝑟𝑑 𝑏𝑒𝑑 = Equation 11
𝐴𝑡𝑢𝑏𝑒𝑠
Distillation columns
The design of the distillation columns were obtained by calculating the diameter and the
height of the columns. A maximum allowable vapor velocity, uv, was estimated from the plate
spacing, lt, with Equation 14:
𝜌𝑙 −𝜌𝑣 0.5
𝑢𝑣 = (−0.17𝑙𝑡2 + 0.27𝑙𝑡 − 0.047) ∙ ( ) Equation 14
𝜌𝑣
Where ρl is the liquid density and ρv is the vapor density. The plate spacing was assumed to
be 0.6 m. An average vapor density was estimated both at the top and the bottom of the
column and hence two maximal velocities were obtained. These two velocities were used in
Equation 15 with its vapor density respectively estimating the two column diameters:
4𝑉𝑤
𝐷𝑐 = √ Equation 15
𝜋𝜌𝑣 𝑢𝑣
Where Dc is the column diameter and Vw is the vapor mass flow in the column. The largest
diameter calculated with Equation 15 was then used for the column design. By using Equation
16 and 17 the column height could be calculated.
ℎ𝑐 = 𝑁𝑅 ∙ 𝑙𝑡 Equation 16
𝑁𝐼
𝐸0 = Equation 17
𝑁𝑅
Where hc is the column height, NR is the number of real stages, NI is the number of ideal
stages and E0 is the column efficiency. The column efficiency was set to 0.8 for all the
distillation towers. Both the condenser and the reboiler were seen as heat exchangers and
designed thereafter [57].
Heat exchangers
The required heat exchanger area, A, was calculated with Equation 18, where Q is the effect,
k is the overall heat transfer coefficient and TL is the logarithmic average temperature,
calculated with Equation 19. The numbers 1 and 2 in Equation 19 denotes different sides of
the heat exchanger. Equation 20 was used to calculate the amount of needed cooling media of
the condenser [56].
𝑄
𝐴= Equation 18
𝑘∙𝑇𝐿
𝑄
𝑚̇𝑐𝑜𝑜𝑙𝑖𝑛𝑔 = Equation 20
𝐶𝑝 ∙∆𝑇
Gas/liquid separator
The height and diameter of the vessel used for separating gas from liquid were calculated. The
gas velocity (uG) was determined by using the vapor and liquid densities (ρG and ρL):
IV-ii
𝜌𝐿
𝑢𝐺 = 0.064√ −1 Equation 21
𝜌𝐺
From the gas velocity and the gas flow rate (QG) the diameter (D) of the vessel could be
calculated:
𝑄𝐺
𝐷 = 2√ Equation 22
𝑢𝐺 𝜋
The liquid height (hL) was calculated based on an assumed residence time of the liquid
fraction (θ) of 600 seconds. QL denotes the liquid flow from the vessel.
𝑢𝐺 𝑄𝐿 𝜃
ℎ𝐿 = Equation 23
𝑄𝐺
The total height of the vessel (h) was calculated as the sum of the gas height (hg) and the
liquid height. The vessel was designed so that hg was equal to the column diameter D or, if the
column diameter was less than 1 m, was set to 1 m [58].
ℎ = ℎ𝐺 + ℎ𝐿 , 𝑤ℎ𝑒𝑟𝑒 ℎ𝐺 = max{1 𝑚, 𝐷} Equation 24
Absorption column
The absorption column was designed in the same manner as the distillations columns, but
without condenser and reboiler at top and bottom. [57]
Storage tanks
The storage tanks were designed to store ethanol, ethylene and sodium hydroxide in 10 days
respectively. The volume of the tanks (V) was calculated from the flows (F) and the amount
of storage days (θ):
𝑉 = 𝐹𝜃 Equation 25
Dryer
The breakthrough time could be calculated with Equation 26 [59], in which V is total mass
flow, c0 is concentration of water, S is adsorbent mass and X is an adsorption factor.
𝑆∙𝑋
𝑡𝑏𝑟𝑒𝑎𝑘𝑡ℎ𝑟𝑜𝑢𝑔ℎ = Equation 26
𝑉∙𝑐0
Assumptions that were made was that 21 g water can be adsorbed by 100 g zeolite, 3A
Molecular Sieve. A zeolite from Interra Global was used. This zeolite has a bulk density of ca
0.70 g/ml [60].
IV-iii
Appendix V – Calculation of unit operation design
The calculations of the unit operations were made in excel, see specifications below.
Reactor
Distillation 1
Distillation tower 1. Water/Ethanol
Table 17. Calculations for distillation tower 1.
Parameters Unit Source/Comment
Pressure column 100,000 Pa Aspen Plus V8.2
Molar mass ethanol 0.046 kg/mol [61]
Molar mass water 0.018 kg/mol [61]
Gas law constant 8.315 J/mol∙K [61]
Liquid density 967.8 kg/m3 Aspen Plus V8.2
Tray distance 0.6 m Assumption by authors
Mass flow rate vapor 0.259 kg/s Aspen Plus V8.2
Ideal number of stages 25 Aspen Plus V8.2
Column efficiency. E0 0.8 Assumption by authors
V-ii
Column height 18.75 m
Column surface area 29.82 m2
Ethylene column
Distillation tower ethylene/higher hydrocarbons
Table 20. Calculations for the ethylene column.
Parameters Unit Source/Comment
V-iii
Pressure column 100,000 Pa
Molar mass ethylene 0.028 kg/mol Aspen Plus V8.2
Molar mass butene 0.056 kg/mol Aspen Plus V8.2
Gas law constant 8.315 J/mol∙K [61]
Liquid density 627.9 kg/m3 Aspen Plus V8.2
Tray distance 0.6 m Assumption by authors
Mass flow rate 3.319 kg/s Aspen Plus V8.2
Ideal number of stages 15 Aspen Plus V8.2
Column efficiency. E0 0.8 Assumption by authors
V-iv
Table 22. Calculations for the reboiler in the ethylene column.
Reboiler
Parameters Unit Source/Comment
Temp. heat. Media in 245 °C Assumption by authors
Mass flow oil 38.9 kg/s Aspen Plus V8.2
Temp. heat. Media out 218.8 °C Aspen Plus V8.2
Temp. Liquid -98.33 °C Aspen Plus V8.2
Heat transfer coeff. 114 W/(m2·°C) Assumption by authors, made on ethylene
(k) liquid-ethylene vapor [56]
Heat transferred. Q 1,884 kW Aspen Plus V8.2
log. mean temp 330.0 °C
Area 50.07 m2
C2 Stripper
Distillation tower ethylene/light gases.
Table 23. Calculations for the C2 stripper column.
Parameters Unit Source/Comment
Pressure column 100,000 Pa
Molar mass ethylene 0.028 kg/mol Aspen Plus V8.2
Molar mass hydrogen 0.0020 kg/mol Aspen Plus V8.2
Gas law constant 8.315 J/mol∙K [61]
Liquid density 148.7 kg/m3 Aspen Plus V8.2
Tray distance 0.6 m Assumption by authors
Mass flow rate 3.319 kg/s Aspen Plus V8.2
Ideal number of stages 10 Aspen Plus V8.2
Column efficiency. E0 0.8 Assumption by authors
V-vi
Heat Exchanger 1
Ethanol feed and oil
Table 26. Calculations for heat exchanger 1.
Parameter Unit Source/Comment
Ethanol feed. in 10 °C Aspen Plus V8.2
Ethanol feed. out 80 °C Aspen Plus V8.2
Stream from reactor 240 °C Aspen Plus V8.2
(ethylene) in
Stream from reactor 70 °C Aspen Plus V8.2
(cooled ethylene) out
Temp. transfer efficiency 0.739
Calculations
Heat transfer. Q 5,808 kW Aspen Plus V8.2
Overall heat transfer 341 Assumption by authors, based on [56]
coefficient (k)
Log. mean temp. 102.0
Area 167.0 m2
Heat Exchanger 2
Ethanol preheating.
Table 27. Calculations for heat exchanger 2.
Parameter Unit Source/Comment
Cold stream in (ethanol) 81 °C Aspen Plus V8.2
Cold stream out (ethanol) 240 °C Aspen Plus V8.2
Hot stream in (dowtherm A) 288 °C Aspen Plus V8.2
Hot stream out (dowtherm A) 288 °C Aspen Plus V8.2
Calculations
Heat transfer, Q 3,088 kW Aspen Plus V8.2
Overall heat transfer 100 Assumption by authors, based on [56]
coefficinet (k)
Log. mean temp. 108.8
Area 283.8 m2
V-vii
Heat Exchanger 3
Heat exchanger connected to quench column.
Table 28. Calculations for heat exchanger 3.
Parameters Unit Source/Comment
Temp. Ethylene in 70 °C Aspen Plus V8.2
Temp. Ethylene out 25 °C Aspen Plus V8.2
Temp. Cooling media 15 °C Water, assumption by authors
in
Temp. Cooling media 30 °C Water, assumption by authors
out
Calculations
Heat transfer coeff (k) 350 W/(m2·°C) Assumption by authors, based on
ethylene vapor-chilled water [56]
Heat transferred, Q 3,097 kW Aspen Plus V8.2
log.Temp. 21.6 °C
Area 408.9 m2
Mass flow cooling 49.40 kg/s
water
Heat Exchanger 4
Cooling of stream before ethylene column.
Table 29. Calculations for heat exchanger 4.
Parameters Unit Source/Comment
Temp. Ethylene in 32 °C Aspen Plus V8.2
Temp. Ethylene out -104 °C Aspen Plus V8.2
Temp. Cooling media -150 °C Assumption by authors
in
Temp. Cooling media -150 °C Assumption by authors
out
V-viii
Heat Exchanger 5
Cooling of stream before C2 Stripper column.
Table 30. Calculations for heat exchanger 5.
Parameters Unit Source/Comment
Temp. Ethylene in -104 °C Aspen Plus V8.2
Temp. Ethylene out -130 °C Aspen Plus V8.2
Temp. Coooling -150 °C Assumption by authors
media in
Temp. Cooling -150 °C Assumption by authors
media out
Calculations
Heat transfer coeff 114 W/(m2·°C) Assumption by authors, based on ethylene
(k) vapor-ethylene vapor [56]
Heat transferred, Q 1,819 kW Aspen Plus V8.2
log.Temp. 31.22 °C
Area 511.2 m2
Gas/Liquid separator
Separates gas from liquid after heat exchanger 3.
Table 31. Calculations for the gas/liquid separator.
Parameters Unit Source/Comment
Liquid density (bottom) 954.6 kg/ Aspen Plus V8.2
m3
Vapor density (top) 1.123 kg/ Aspen Plus V8.2
m3
Volumetric flow rate gas 3.155 m3/s Aspen Plus V8.2
3
Volumetric flow rate 0.0029 m /s Aspen Plus V8.2
liquid
Residence time 600 s Assumption by authors.
Calculations
Gas velocity 1.865 m/s
Diameter flash column 1.468 m
Height gas 1.468 m
Choose either height=1
or if D>1, choose height of gas [58].
V-ix
Height liquid 1.050 m
Height column 2.51749 m
Absorption column
Storage tanks
Ethanol tank, liquid.
Table 33. Calculations for the ethanol storage tank.
Parameter Unit
V-x
Flow ethanol 27.09 m3/h
Days of storage 10 days
Volume 6,502 m3
Dryer
Two pieces with regeneration.
Table 36. Calculations for the two dryers.
Parameters Unit Source/Comment
Adsorption factor 0.21 kg water/kg [60]
adsorbent
Concentration water 0.019 kg water/kg mass Aspen Plus V8.2
flow
Total mass flow 12,824 kg/h Aspen Plus V8.2
Drying time 12 h Assumption by authors
3
Density adsorbent 700 kg/m [60]
Calculations Unit
Adsorbent mass 14,143 kg adsorbent
Volume adsorbent 20.20 m3 adsorbent
Extra adsorbent 4.040 m3 adsorbent
Total volume 24.24 m3 adsorbent Assuming cylindrical geometry
adsorbent
Total mass adsorbent 16,971 kg adsorbent
V-xi
Diameter dryer 2.5 m
Height dryer 4.94 m
Centrifugal pumps
Pump 1.
Table 37. Calculations for centrifugal pump 1.
Parameter Unit Source/Comment
Inlet pressure 1 bar
Outlet pressure 10 bar
Mass flow 22,000 kg/h Ethanol/water
Pump 2.
Table 38. Calculations for centrifugal pump 2.
Parameter Unit Source/Comment
Inlet pressure 1 bar
Outlet pressure 10 bar
Mass flow 2,000 kg/h NaOH/water
V-xii
Appendix VI – Economy equations and method
Ulrich Method
A capital cost estimation method called Ulrich method was used for economical calculations.
Ulrich method is based on a calculation where the Bare-Module Cost, CBM, for each utility is
calculated by multiplying the Purchased Equipment Cost, Cp, with a Module Factor, FBMα, see
Equation 27 [52].
𝛼
𝐶𝐵𝑀 = 𝐶𝑝 ∙ 𝐹𝐵𝑀 Equation 27
The sum of all Bare-Module Costs are then calculated and multiplied with two factors, one
taking fees and contingencies, ffees/contingency, into account and one taking auxiliary facilities,
fauxiliary facilities, into account, Equation 28 [52]. The cost K is direct and indirect plant costs as
well as auxiliary facilities costs.
𝐾 = (∑𝑛𝑖=1 𝐶𝐵𝑀𝑖 ) ∙ 𝑓𝑓𝑒𝑒𝑠/𝑐𝑜𝑛𝑡𝑖𝑛𝑔𝑒𝑛𝑐𝑦 ∙ 𝑓𝑎𝑢𝑥𝑖𝑙𝑖𝑎𝑟𝑦 𝑓𝑎𝑐𝑖𝑙𝑖𝑡𝑖𝑒𝑠 Equation 28
Purchased Equipment Costs and Module Factors were taken from diagrams from January
2004 in which the costs were given in USD. A factor, see Equation 29, was used for updating
of the utilities cost, see Equation 30. In Equation 29 IAK stands for Equipment Construction
Cost Index and applies to US. X and Y in Equation 29 and 30 denotes which year the values
are related to [52].
(𝐼 )
𝑓 = (𝐼𝐴𝐾)𝑥 Equation 29
𝐴𝐾 𝑌
Fixed costs
The annual cost for storage of feedstock and product was calculated with Equation 31, in
which Q is annual consumption or production, P is cost or income per tonne and D is days of
storage. fA is the annuity factor, which can be calculated with Equation 32, in which X is life
time and X is the interest rate. The factor can also be used to depreciate the grass root capital
cost [62].
𝐴𝑛𝑛𝑢𝑎𝑙 𝑐𝑜𝑠𝑡 𝑓𝑜𝑟 𝑠𝑡𝑜𝑟𝑎𝑔𝑒 = 𝑄 ∙ 𝑓𝐴 ∙ 𝑃 ∙ 𝐷/(365 𝑑𝑎𝑦𝑠/𝑦𝑒𝑎𝑟) Equation 31
𝑋
𝑓𝐴 = Equation 32
1−(1+𝑋)−𝑁
Annual maintenance and repair cost were estimated to amount to 6 % of Grass Roots Capital.
The cost of spare parts was calculated to be 15 % of the cost for maintenance and repair times
the factor fA calculated with Equation 32.
Direct costs
The annual cost for operators was calculated with Equation 33. S stands for direct monthly
salary and n is number of workers per shift [62].
𝐴𝑛𝑛𝑢𝑎𝑙 𝑐𝑜𝑠𝑡 𝑓𝑜𝑟 𝑜𝑝𝑒𝑟𝑎𝑡𝑜𝑟𝑠 = 12 𝑚𝑜𝑛𝑡ℎ𝑠/𝑦𝑒𝑎𝑟 ∙ 𝑆 ∙ 𝑛 ∙ 5 Equation 33
The cost for supervisors for operators and laboratory work was 15 % on operators each.
Consumables are also included as direct costs. The cost for land and license fees was
neglected.
VI-i
Indirect costs
To calculate the overhead for staff 70 % was added on the cost for shift personnel and 50 %
was added on the cost for day personnel. Administration costs were estimated to be 25 % of
the overhead costs for staff. Costs for research, development, distribution and sale were
neglected.
VI-ii
Appendix VII – Grass Root Capital
Purchased cost for equipment and Bare-Module Cost have been retrieved from the online
database EconExpert, using Ulrich’s calculation method [63].
Table 40. All units and their pruchased cost for equipment and Bare-Module Cost.
Unit Dimensions/Operating Material Cp CBM
conditions (Purchased (bare module
cost, USD) cost, USD)
Ethanol storage 6,502 m3 Atmosperic Stainless steel 154,899 542,146
tank pressure- Cone roof
Sodium hydroxide 378 m3 Atmosperic Stainless steel 24,339 85,188
storage tank pressure- Cone roof
Ethylene storage 4,916 m3 Atmospheric Stainless steel 224,877 787,069
tank press-Gas holder
Centrifugal pump Pressure rise 10 bar Stainless steel 9,348 45,111
1 12 kW, Centrifugal pump
Heat exchanger 1 168 m2 Shell and Tube, Stainless steel 16,731 96,749
Fixed tube sheet and U-
tube
Heat exchanger 2 284 m2 Shell and Tube, Stainless steel 23,814 137,710
Fixed tube sheet and U-
tube
Reactor vessel D=1.94 m, H=4.15 m, Stainless steel 20,209 190,545
Process vessel, vertically,
no packing
A=297.3 m2, Shell and Stainless steel 24,585 142,168
tube, fixed tube
Thermal fluid 10,158 kW - 534,216 1,175,274
heater Furnaces, Thermal fluid
heater, Mineral oil heaters
Heat exchanger 3 409 m2 Stainless steel 30,845 178,368
Shell and Tube, Fixed
tube sheet and U-tube
Gas/Liquid D=1.47 m, H=2.52 m, Stainless steel 12,161 114,666
separator Process vessel, Vertically
oriented, No packing or
trays
Centrifugal pump Pressure rise 10 bar, 3kW, Stainless steel 5,477 26,433
2 Centrifugal
Absorber D=1.52 m, H=26.3 m Stainless steel and 78,444 878,861
43 Trays, Process vessel, stainless steel trays
Vertically oriented,
Sieve-trays
Dryer - 2 st D=2.5 m, H=5.94 m, Stainless steel 66,658 868,824
Packed height 4.94 m Zeolite adsorbent
Process vessel, vertically
Heat exchanger 4 62 m2 Stainless steel 9,113 52,700
Shell and Tube, Fixed tub
VII-i
Ethylene column D=1.49 m, H=11.25 m Stainless steel and 34,666 372,576
12 Trays, Process equip, stainless steel trays
vertically, sieve trays
Condensor: Heat Stainless steel 19,755 114,235
exchanger, A= 216m2
Shell and Tube, Fixed
tube sheet and U-tube
Reboiler: Heat exchanger, Stainless steel 8,080 46,727
A= 50m2
Shell and Tube, Fixed
tube sheet and U-tube
Heat exchanger 5 511 m2 Shell and Tube, Stainless steel 36,318 210,016
Fixed tub
C2 stripper D=2.3 m, H=7.5 m Stainless steel and 35,937 453,142
13 Trays, Process vessel, stainless steel trays
vertically, sieve trays
Condensor: Heat Stainless steel 4,080 23,593
exchanger, A= 12.5m2
Shell and Tube, Fixed
tube sheet and U-tube
Reboiler: Heat exchanger, Stainless steel 3,503 20,258
A= 8.7m2
Shell and Tube, Fixed
tube sheet and U-tube
Ethanol D=0.506 m, H=18.75 m Stainless steel 36,470 350,919
distillation tower 19 Trays, Process vessel,
vertically, sieve trays
Condensor: Heat Stainless steel 12,071 69,803
exchanger, A= 100 m2
Shell and Tube, Fixed
tube sheet and U-tube
Reboiler: Heat exchanger, Stainless steel 16,275 94,113
A= 161 m2
Shell and Tube, Fixed
tube sheet and U-tube
Heatpump Heat absorption Coolant temp: - 6,896,134
rate=3,708 kW - 55°C
Auxiliary, Mechanical ref
unit
Wastewater Capacity: 0.0043 m3/sek - - 991,409
treatment plant (minimum)
Total CBM 14,964,737
VII-ii
Table 41. Calculation for the investment cost for the material needed upon start-up.
Material supply
Catalyst Catalyst volume 2.68 m3
Density 700 kg/m3
Catalyst mass 1,873 kg
Table 42. Summary of the total capital cost for the plant.
Total cost of plant
Contingency and fees 1.15 15%
Auxiliary and facilities 1.25 25%
VII-iii
Appendix VIII – Operating costs
Operating costs for the plant are calculated with excel and divided into three groups; fixed capital, direct costs and indirect costs.
Fixed capital
Table 43. Fixed capital cost including storage and spare parts.
Storage of feedstock Unit Source/Comment Storage of product Unit Source/Comment
Ethanol Ethylene
Days of storage days 10 Assumption by authors Days of storage days 10 Assumption by authors
Consumption kg/hr 22,000 Aspen Plus V8.2 Production kg/hr 11,642 Aspen Plus V8.2
3
Density kg/ m 812 [61] Operating time hours/day 24
Consumption volume m3/hr 27.09 Year Days/year 365
Operating time hours/day 24 Annual production tonnes/year 101,987
Year Days/year 365
Annual consumption m3/year 237,340 Income per tonne dollar/tonne 1,280 Assumption by authors
3 3
Cost per m USD/ m 436 [51] VKK 2012 7.5 [62]
Annual cost USD/year 372,679 Annual cost USD/year 470,218
NaOH Total cost of storage USD/year 851,101
Days of storage days 10 Assumption by authors
Consumption kg/hr 500 Aspen Plus V8.2
Operating time hours/day 24 Spare parts for plant
Year Days/year 365 Percent % 0.15
Annual consumption tonne/year 4,380
Cost per tonne dollar/tonne 520 [53] Total cost spare parts USD/year 33,256
VKK 2012 7.5 [62]
Annual cost USD/year 8,204 Total fixed capital USD/year 884,357
VIII-i
Direct cost
Table 44. Direct costs for the plant, part one.
Feedstock Unit Source Dowtherm oil Unit Source
Ethanol 95 wt % Oil exchange % 5 Assumption by authors
Consumption kg/hr 22,000 Aspen Plus V8.2 Exchange of oil kg/year 3,500
3
Density kg/ m 812 [61] Oil price USD/tonne 4,400 [64]
3
Consumption volume m /hr 27.1 Total cost of oil USD/year 115,500
Operating time hours/day 24 Cooling water
Year Days/year 365 Required energy removal kW 6,992 Aspen Plus V8.2
3
Annual consumption m /year 237,340 Required water flow kg/s 96
Cost per m3 USD/ m3 436 [51] Water density kg/ m3 1,000 [61]
3
Total cost of USD/year 103,463,894 Water flow m /s 0.096
feedstock
Solvents m3/year 3,027,456
NaOH Cost per m3 USD/m3 0.013 Assumption by authors
Consumption kg/hr 500 Aspen Plus V8.2 Total cost of cooling USD/year 40,366
water
Operating time hours/day 24 Supervisors
Year Days/year 365 Percent 10% 0.1 [62]
Cost per tonne USD/tonne 520 [53] Total cost supervisors USD/year 104,000
Total cost NaOH USD/year 2,277,600 Laboratory work
Electricity Percent 10% 0.1 [62]
Required energy kW 757 Total cost laboratory USD/year 104,000
hours/year h/year 8,760 Maintenance & repair
Energy hours/year kWh/year 6,627,816 Percent 6% 0.06 Assumption by authors
Cost per kWh USD/kWh 0.13 Assumption by authors Total cost of maintenance USD/year 1,686,309
Total cost electricity USD/year 883,709
VIII-ii
Table 45. Direct costs for the plant, part two.
Catalyst Unit Source
Catalyst changes 2 Assumption by authors
catalyst cost per set dollar 112,390 Assumption by authors
total cost for changes dollar 224,780
VIII-iii
Indirect cost
Table 46. Indirect costs for the plant.
Overhead for staff Unit Source
Shift personnel 70 % 0.7 [62]
Day personnel 50 % 0.5 [62]
Total cost for overhead USD/year 720,000
Administration
Percent 25 % 0.25 [62]
Total cost of distribution USD/year 180,000
VIII-iv
Appendix IX – Investment calculations
Costs and revenues are calculated to determine the annual net income, ai.
ai =Ii – Ui
Grass root capital, G
Annual operating costs, Ui
Annual income, Ii
Annual net income, NI
Annuity factor, fa
Interest rate, X = 10 %
The pay-back time, n, can be estimated by using Equation 34 [52].
𝑋
ln(1−𝐺∙(𝑎 ))
𝑖
𝑛 = −( ) Equation 35
ln(1+𝑋)
Annual net income is calculated with Equation 36 below and the capital cost with Equation 37
[52].
𝑁𝐼 = 𝑎𝑖 − 𝑓𝑎 ∙ 𝐺 Equation 36
𝐶𝑎𝑝𝑖𝑡𝑎𝑙 𝑐𝑜𝑠𝑡 = 𝑓𝑎 ∙ 𝐺 Equation 37
IX-i