i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 0

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier.com/locate/he

Modelling and optimization of FischereTropsch

products through iron catalyst in fixed-bed reactor

Hossein Atashi*, Fatemeh Rezaeian

Department of Chemical Engineering, University of Sistan and Baluchestan, P.O. Box 98164-161, Zahedan, Iran

article info abstract

Article history: The most important issue for alternative renewable energy source instead of fossil fuels is
Received 15 February 2017 environmental concern. Human energy supply from carbon waste has an important role in
Received in revised form environmental health. FischereTropsch synthesis is a reaction whereby carbon waste such
20 April 2017 as woods, foodstuffs, sewage, and any material that contains carbon in their structure,
Accepted 21 April 2017 produces all oil derivatives including gasoline, diesel, etc. In this paper, the products model
Available online xxx and the effective parameters of the model and also, the interaction between the parame-
ters were investigated. PH2 and T*PH2 are effective parameters and interaction in the model,
Keywords: respectively. Finally, the optimization of products was discussed. The best operating
FischereTropsch synthesis conditions for minimizing CO2 and CH4 and maximizing C2H4 and C2H6 were investigated.
Iron catalyst This method can be suitable for environmental purposes.
Waste materials © 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
Selectivity
Environment
Products model

gas (H2 þ CO). As can be seen in Fig. 1, synthesis gas is a

Introduction feedstock for FischereTropsch synthesis [3,4].
Products like olefins, diesel, gasoline, kerosene, and
Increase in greenhouse gasses resulting from fossil fuel pro- methanol are the result of FischereTropsch reaction.
cesses including transportation fuels induces a lot of climate FischereTropsch reaction is a catalytic reaction which can
changes, which, is inevitable from environmental point of produce a variety of products by changing the nature of the
view. Also, declining oil reserves, rising price of crude oil, and catalyst. The importance of these products is that they do not
increasing energy demand especially for transport, have contain any harmful substances like sulfur and aromatics.
caused scientists to think about alternative sources [1,2]. On Therefore, these products have been remarkable in terms of
the contrary, the wastes including urban, agricultural, and the environment and are in competition with fossil fuels
industrial waste which are harmful to the environment, can [5e7]. Transition metals which are active in FischereTropsch
be utilized into gasification process and to produce synthesis synthesis are listed according to the average molecular weight
of hydrocarbons produce as follows:

Ruthenium > Iron > Cobalt > Rhodium > Nickel > Iridium > Platinum > Palladium

* Corresponding author.
E-mail addresses: [email protected] (H. Atashi), [email protected] (F. Rezaeian).
http://dx.doi.org/10.1016/j.ijhydene.2017.04.224
0360-3199/© 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Atashi H, Rezaeian F, Modelling and optimization of FischereTropsch products through iron catalyst
in fixed-bed reactor, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.04.224
2 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 0

Fig. 1 e Strategies for production of oil derivatives from biomass [4].

The catalyst of Iridium (Ir), Platinum (Pt), and Palladium (Pd) Fischer Tropsch products in membrane reactor and conven-
produce more methane. Cobalt (Co) and iron (Fe) are the most tional Fischer Tropsch reactor [7]. Atashi et al. studied ther-
frequently used for commercial scale production while Nickel modynamic analysis which is combine with RSM [24]. One of
(Ni), Ruthenium (Ru) are preferred for academic research [8,9]. the advantage of selectivity models in industry is, determining
The advantage of using iron catalyst can be noted as follows: the optimal conditions for reaction products. This can maxi-
low cost, excellent performance in Water Gas Shift reaction mize the desirable products and decrease the production of
(WGS), high selectivity of olefins products, low selectivity of undesirable products as much as possible.
methane, and have high activity in FTS [10,11]. As mentioned earlier, many studies have been carried out
Several research have been carried out for improving the to study the effect of these parameters quantitatively, but few
activity, stability, and selectivity of catalyst by adding pro- efforts have been made to present a model which can express
moters. For instance, potassium is a very common metal that the importance of these factors. The aim of this study is to
can be used as a promoter because it enhances the present a model for the products selectivity of iron catalyst
FischereTropsch synthesis and Water Gas Shift reaction with promoters with the component amount of 100Fe/5Cu/4K/
catalyst. In addition, it shifts the products to the higher mo- 16SiO2 and examine the importance of factors such as tem-
lecular weight of hydrocarbons [12,13]. perature, time on stream, and partial pressure of H2 and CO. In
Copper has the same result and increases isomerization of 1- addition, the importance of parameters using the constant
alkenes to 2-alkenes and hydrogenation of olefins while potas- coefficients was determined. It also determines the interac-
sium prevents any of these reactions [14]. Potassium increases tion between these parameters via a statistical model and
the selectivity of olefins and suppression of methane and the finally discusses the optimal conditions which are very useful
formation of light hydrocarbons [15]. Fe/Cu/K/SiO2 catalyst with for increasing the desirable products.
different molecular weight is mostly utilized than the other
catalysts. Bukur used three types of this catalyst to determine
their performance [16] and Hai-Jun also used different supports Experimental
to enhance the performance of the catalyst, he also prepared Fe/
Cu/K/SiO2 and Fe/Cu/K/Fe2O3 catalysts and it was concluded In this paper, the experimental data were taken from Ref. [25].
that the catalyst with SiO2 has low deactivation and high activity Catalyst construction was performed using values presented
of FTS and also, increase the Cþ 5 selectivity [17]. in Table 1. Brunner which was used for preparing the catalysts
Recently, many studies have been conducted on products with various conditions was also used to name them. For
distributions and many factors such as operating conditions instance, for the intended catalyst (100Fe/5Cu/4K/16SiO2) was
have effect on products distributions. Manipulation of oper- named as 2UH. The number 2 indicates that it quickly adds
ating conditions like temperature, pressure, space velocity, promoter after sedimentation or drying. The letter U indicates
H2/CO ratio, time on stream, etc. can change product selec- that the catalyst does not require washing while the letter H
tivity. Increase in temperature causes the products to shift to indicates drying at high temperature (100  C).
higher hydrocarbons and increase the conversion of CO but Cþ 5 The drying was done for 16e24 h and also the calcination
is reduced. Increase in pressure causes the products to shift to was done at 300  C by air flow for 6e16 h. An amount of 0.251 g
heavier hydrocarbons. Increase in H2/CO ratio causes the of catalyst was tested in the fixed bed reactor and the data is
products to shift to heavier hydrocarbons. Decrease in space presented in Table 2 [25].
velocity causes increase in the conversion of CO and de-
creases the selectivity of methane [18e20].
Kinetic and selectivity of Fischer Tropsch products Data analysis
constitute two pillars in the reaction. Many studies have been
proposed on selectivity and variety of methods. Dragomir In FischereTropsch synthesis, there are factors which can
investigated olefin selectivity and the reaction conditions over affect the quality and selectivity of products. For example,
Fe catalyst [21]. Haghtalab studied the ability of his catalyst to temperature, the partial pressure of CO and H2, and time on
increase selectivity of long-chain hydrocarbons and reduce stream are variables which have effect on the products
the lighter hydrocarbons [22]. Shawane by employing Taguchi selectivity like methane (CH4), ethylene (C2H4), ethane (C2H6),
method provided optimal conditions and investigated cata- and carbon dioxide (CO2). In this work, the impact of four
lytic activity [23]. Forghani et al. investigated the selectivity of parameters (temperature, the partial pressure of CO and H2,

Please cite this article in press as: Atashi H, Rezaeian F, Modelling and optimization of FischereTropsch products through iron catalyst
in fixed-bed reactor, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.04.224
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 0 3

Table 1 e Materials masses for catalyst preparations. All

P_value
0.007
0.026
0.046
0.002
0.027
0.034

0.057
0.082

0.024
0.019

0.022
numbers are grams of specified moiety [25].

e
e

e
Product model for S(CO2)
Catalyst Fe(NO3)3a Cu(NO3)2b KHCO3 SiO2 NH4HCO3

R2adj ¼ 93:64%
2UH 72.492 1.836 1.077 1.611 43.799

Tvalue

3.13

5.87
3.11

2.46
2.17
a
Fe(NO3)3$9H2O.

4.36

2.68

2.89

3.21
3.40

3.27
b
Cu(NO3)2$2.5H2O.

e
e

R2 ¼ 97:88%
SE coefficient
Table 2 e Experimental data and operating conditions

0.903

0.523
26.2
59.2
2.68

29.8
10.1

3.46

62.2
2.40

3.22
[25].

e
e

e
T ( C) TOS (h) PCO (atm) PH2 (atm) Selectivity (% mol)
CO2 CH4 C2H4 C2H6
249 206 6 6 32.4 5.6 5.6 7

P_value
230 232 6.2 6.4 30.9 4.6 6 4.9

0.171
0.062
0.103
0.246
0.085
0.049

0.181
0.058
0.140
220 267 6.4 6.6 26.2 3.7 5.2 4.2

e
e
e
e

e
e
Product model for S(C2H6)
239 280 6.3 6.3 31.3 4.7 5.7 5.4

R2 ¼ 98:38% R2adj ¼ 96:54%

249 292 6 6 32.7 5.4 5.6 6.8
249 314 6 6 33.1 5.3 5.5 6.9

Tvalue
1.52

1.87
1.27

2.38

2.27
1.66
250 528 6.2 6.3 33.2 4.7 6.4 5.3

2.22

2.01

1.49
e
e
e
e

e
e
259 547 6.1 6.1 33.6 5.3 5.9 5.9
249 363 6.4 5.8 33 4 4.1 6.7
249 391 3.8 4.9 35 4.1 4.6 6.7

SE coefficient
249 404 11.4 4.9 30.4 7.5 4.6 11.9
249 427 4.3 5 33.5 4.5 4.7 7.3

14.8
32.1
1.40
11.6
13.2
3.03

24.7
26.9
3.18
249 448 6 2.5 32 7.4 4.4 11.9

e
e
e
e

e
e
249 468 5.4 5.9 31.1 4.7 3.9 7.3
250 495 11.6 5 31 7.6 4.9 11.9
249 512 6.1 2.6 32.4 7.6 4.8 12.2
P_value

time on stream) were evaluated by using response surface

0.003
0.004
0.004
0.004
0.004
0.005
0.012
0.003
0.003
0.008
0.003
0.004
0.003
0.005
methodology (RSM).

e
Product model for S(C2H4)

Response surface methodology (RSM) is a set of mathe-

R2adj ¼ 98:42%
matical and statistical techniques for constructing an exper-
16.60

16.16

17.70
17.03
10.92
16.93
Tvalue

imental model. The aim of RSM is to achieve a mathematical

17.33

15.46
16.85

14.73

16.04
17.62
14.74
9.18

model that describes the process in which the main factors or

e
Table 3 e Regression coefficient of the RSM modeling for all responses.

their interactions are involved. This model can be achieved to

R2 ¼ 99:79%
the most favorable response by adjusting the factors.
SE coefficient

The polynomial regression model which has acceptable

0.121
0.923
0.956
0.773

appropriateness between experimental data and obtained 0.317

11.2

1.04
4.29
10.5
2.95

9.40
21.1
1.80
23

selectivity model is considered as follows: e

X
n X X
n
f ¼ b0 þ bi zi þ bij zi zj þ bii z2i þε (1)
i¼1 ij i¼1
P_value

where f is the response, b represents the coefficient of

0.003
0.000
0.000
0.000
0.001
0.003

0.000

0.000
0.001
0.000
0.000
0.002

regression, z is variables (e.g. T, PCO , PH2 and TOS) and also, n

e
e

e
Product model for S(CH4)

indicates the number of experiments [26,27]. The accuracy of

R2 ¼ 99:98% R2adj ¼ 99:93%

regression is obtained by coefficients below:

11.12
12.13

14.12
12.68
Tvalue
6.42

6.68

9.27
11.67

19.32

13.90

SSR
8.98

6.93

R2 ¼ (2)
SSR þ SSE
e
e

The modification of R2 is adjusted R2 which is defined as

SE coefficient

follows:
0.0446

0.0717

ð1  R2 Þðn  1Þ
0.112
0.838
0.991
0.229

0.226

0.211

R2adj ¼ 1  (3)
1.11

1.79
2.01

np
2.4

e
e

Pvalue is used for important regression coefficient. The

amount of Pvalue must be lower than 0.05. The Pvalue is a
necessary condition but not sufficient, for this reason, the
PCO (atm)
PH2 (atm)

TOS  PCO
TOS  PH2
TOS*TOS
Constant

PCO *PCO
Terms

PCO *PH2
TOS (h)

PH2 *PH2

other items such as Tvalue and SE coefficient were used [28,29].

T*TOS
T*PCO
T ( C)

T*PH2

The regression coefficient and their analysis are presented in

T*T

Table 3.

Please cite this article in press as: Atashi H, Rezaeian F, Modelling and optimization of FischereTropsch products through iron catalyst
in fixed-bed reactor, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.04.224
4 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 0

FischereTropsch products selectivity such as methane,

Results and discussion ethylene, ethane, and carbon dioxide via statistical method.
The selectivity models were improved by Response Surface
As stated above, the aim of this study is to present a new Methodology (RSM). The results have acceptable appropri-
procedure for examining the effect of various parameters ateness between experimental data and obtained model. The
such as temperature (T), partial pressure of CO (PCO ) and par- result is presented in Table 4. Finally, the optimization results
tial pressure of H2 (PH2 ), time on stream (TOS), on the are shown in Table 5. The minimum amount of products

Table 4 e Selectivity model of products as well as predicted value by RSM.

No. Responses (experimental) Responses (predicted by RSM)
S(CO2) S(CH4) S(C2H4) S(C2H6) S(CO2) S(CH4) S(C2H4) S(C2H6)
1 32.4 5.6 5.6 7 32.8077 5.6238 5.3274 7.1533
2 30.9 4.6 6 4.9 31.1432 4.6048 5.7577 4.8488
3 26.2 3.7 5.2 4.2 26.1629 3.7073 4.9769 4.2320
4 31.3 4.7 5.7 5.4 30.9237 4.7197 5.4376 5.6532
5 32.7 5.4 5.6 6.8 32.7331 5.3748 5.2329 6.9926
6 33.1 5.3 5.5 6.9 32.7140 5.3111 5.2998 6.9515
7 33.2 4.7 6.4 5.3 33.1982 4.7141 6.1064 5.8804
8 33.6 5.3 5.9 5.9 33.5308 5.3122 5.5939 5.9593
9 33 4 4.1 6.7 32.6822 4.0184 3.7995 5.8779
10 35 4.1 4.6 6.7 34.9419 4.0742 4.3479 6.9949
11 30.4 7.5 4.6 11.9 30.3157 7.5088 4.3113 11.7700
12 33.5 4.5 4.7 7.3 33.5424 4.5467 4.3880 7.6081
13 32 7.4 4.4 11.9 32.2881 7.4108 4.1158 12.2311
14 31.1 4.7 3.9 7.3 31.7502 4.6858 3.6357 6.7656
15 31 7.6 4.9 11.9 31.0922 7.6139 4.6088 11.9887
16 32.4 7.6 4.8 12.2 32.1181 7.6003 4.5212 1.9280

Table 5 e Optimization results.

T ( C) TOS (h) PCO (atm) PH2 (atm) Selectivity (% mol)
CO2 CH4 C2H4 C2H6
Maximization of CH4 259 206 11.6 2.5 46.6771
T
Minimization of CH4 244.5264 227.8380 10.0933 4.3842 0
Maximization of C2H6 259 206 11.6 2.5 105.9774
T
Minimization of C2H6 249.9761 206 3.8 6.5002 0
Maximization of C2H4 220 547 11.6 2.5 1379.0617
T
Minimization of C2H4 220.1515 546.9814 4.2785 6.1449 0
Maximization of CO2 220 547 3.8 2.5 642.2513
T
Minimization of CO2 249.0650 252.3054 11.60 2.5 0
T
is Target (In order to achieve characteristic of selectivity value).

Fig. 2 e The optimization of CO2 for zero target of selectivity. Minitab software was used for all calculation, regression, and
analysis.

Please cite this article in press as: Atashi H, Rezaeian F, Modelling and optimization of FischereTropsch products through iron catalyst
in fixed-bed reactor, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.04.224
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 0 5

selectivity is negative, while the aim of this article is to decreasing time on stream. However, at constant partial
minimize these amount to zero. Fig. 2 shows the minimization pressure of CO and temperature, at high partial pressure of H2,
of carbon dioxide and methane selectivity into the target of there are little changes. With constant temperature and par-
zero. tial pressure of H2, the selectivity is increased at high partial
pressure of CO. With decreasing time on stream, the mini-
Selectivity model of methane mum amount of selectivity becomes clearly visible, the min-
imum selectivity can occur at PCO ¼ 11.6 and TOS ¼ 547.
Fig. 3 shows the effect of temperature, partial pressure of CO At constant temperature and time on stream, the
and H2, and time on stream, on the selectivity of methane. As maximum selectivity can be obtained by increasing PH2 and
can be seen at constant partial pressure of CO and H2, the decreasing PCO . It can be seen that the maximum amount of
selectivity of methane is raised by increasing temperature but selectivity occurs at PCO ¼11.6, PH2 ¼2.5. The selectivity model
no changes could be seen in selectivity by increasing or of CH4 is as follows:

Fig. 3 e Effect of temperature, partial pressure of H2 and CO and also, time on stream on selectivity of methane: T ¼ 239.5  C,
TOS ¼ 376.5 h, PCO ¼7.7 atm, PH2 ¼4.55 atm.

Please cite this article in press as: Atashi H, Rezaeian F, Modelling and optimization of FischereTropsch products through iron catalyst
in fixed-bed reactor, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.04.224
6 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 0

SðCH4 Þ ¼ 491:57875 þ 2:96657 T þ 0:041496 TOS Selectivity model of ethane

 79:94073 PCO þ 113:85282 PH2  0:0040222 T*T
The fluctuation of selectivity of ethane with temperature, the
þ 0:20487 PH2 *PH2 þ 0:32697 T*PCO  0:46682 T*PH2
partial pressure of H2 and CO and time on stream is shown in
 0:00479721 TOS*PCO  0:00260139 TOS*PH2
Fig. 4. As can be seen, at constant partial pressure of CO and
þ 0:18310 PCO *PH2 H2, the selectivity of ethane is increased with rise in temper-
(4) ature, while there are no changes of selectivity in time on
stream. With constant time on stream and partial pressure of
TOS*TOS, PCO *PCO and T*TOS have minimum effect on the
CO and also at constant time on stream and partial pressure of
model and due to Pvalue, regardless of their coefficient. As
H2, the selectivity of ethane increases with rise in tempera-
demonstrated by the model, the maximum effect of the
ture. The maximum amount of selectivity occurs at T ¼ 259,
parameter belongs to partial pressure of H2. Besides that, the
PH2 ¼2.5, PCO ¼11.6. At constant temperature and PH2 , by
important interaction is T*PH2 , because it has the most coef-
increasing the TOS, at low partial pressure of CO, the selec-
ficient. In optimization part, the minimization of ethane was
tivity is raised and at high partial pressure, the selectivity is
set as a target and could be found in T ¼ 244.5264  C,
decreased.
TOS ¼ 227.8380 h, PCO ¼10.0933 atm, PH2 ¼4.3842 atm.

Fig. 4 e Effect of temperature, partial pressure of H2 and CO and also, time on stream on selectivity of ethane: T ¼ 239.5  C,
TOS ¼ 376.5 h, PCO ¼7.7 atm, PH2 ¼4.55 atm.

Please cite this article in press as: Atashi H, Rezaeian F, Modelling and optimization of FischereTropsch products through iron catalyst
in fixed-bed reactor, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.04.224
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 0 7

The maximum amount of ethane occurs in constant T and has the most effect on the selectivity of ethane and the
TOS in PCO ¼11.6, PH2 ¼2.5. important interaction belongs to T*PH2 , due to its negative
The selectivity model of ethane can be proposed as follows: coefficient. In optimization part, the minimization of ethane
was set as a target and could be achieved in T ¼ 249.9761  C,
SðC2 H6 Þ ¼ 2803:36109 þ 15:95781 T þ 0:045775 TOS TOS ¼ 206 h, PCO ¼3.8 atm and PH2 ¼6.5002 atm.
 116:55699 PCO þ 378:28193 PH2  0:018967 T*T
þ 0:48340 T*PCO  1:52555T*PH2 Selectivity model of ethylene
 0:00794059 TOS*PCO (5)
Ethylene selectivity changes with temperature, partial pres-
Parameters such as TOS*TOS, PCO *PCO , PH2 *PH2 , T*TOS, sure of CO and H2, and time on stream as shown in Fig. 5. As
TOS*PH2 and PH2 *PCO have been ignored due to their effects on can be seen in the figure, at constant partial pressure of CO
the model and the value of Pvalue. As can be seen, the and H2, ethylene selectivity decreases with increasing tem-
maximum coefficient belongs to PH2 , so partial pressure of H2 perature. At constant TOS and PH2 , the selectivity increases

Fig. 5 e Effect of temperature, partial pressure of H2 and CO and also, time on stream on selectivity of ethylene: T ¼ 239.5  C,
TOS ¼ 376.5 h, PCO ¼7.7 atm, PH2 ¼4.55 atm.

Please cite this article in press as: Atashi H, Rezaeian F, Modelling and optimization of FischereTropsch products through iron catalyst
in fixed-bed reactor, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.04.224
8 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 0

with decreasing temperature while, at constant TOS, PCO and omitted. As can be seen, the maximum effect of parameter
also, at constant TOS and PH2 , ethylene selectivity is increased belongs to PH2 because it has the most coefficient and also, the
by reducing the temperature. The minimum selectivity oc- important interaction is T*PH2 . In optimization part, maximi-
curs at T ¼ 259, PCO ¼11.6, and PH2 ¼2.5. At constant temper- zation of ethylene could be achieve in T ¼ 220  C, TOS ¼ 547 h,
ature and partial pressure of H2, by increasing TOS at low PCO , PH2 ¼11.6 atm and PCO ¼2.5 atm.
the selectivity is decreased poorly and at high PCO , the
selectivity is raised. The ethylene selectivity model is given as Selectivity model of CO2
follows:
Fig. 6 shows the changes in CO2 selectivity with temperature,
SðC2 H4 Þ ¼ 16779:86180  95:70111 T þ 0:24492 TOS the partial pressure of CO and H2, and time on stream. As can
þ 518:55481 PCO  2076:37291 PH2 þ 0:11430 T*T be seen, at constant partial pressure of CO and H2 and also at
þ 0:0000383254 TOS*TOS  1:07431 PCO *PCO constant TOS and PH2 , the CO2 selectivity decreases with
 3:87612 PH2 *PH2  0:00253839 T*TOS increasing temperature. At constant temperature and TOS and
also, at constant temperature and PCO , the CO2 selectivity
 2:09376 T*PCO þ 8:44984 T*PH2 þ 0:047781 TOS*PCO
decreases with rise in PH2 . At constant TOS and PCO , by
þ 0:013378 TOS*PH2 increasing temperature at low PH2 , the selectivity is decreased
(6) but at high PH2 , the selectivity is raised. In this case, the
PH2 *PCO has the minimum effect on the model, so it has been maximum selectivity of CO2 occurs at T ¼ 220, PH2 ¼2.5. At

Fig. 6 e Effect of temperature, partial pressure of H2 and CO and also, time on stream on selectivity of carbon dioxide:
T ¼ 239.5  C, TOS ¼ 376.5 h, PCO ¼7.7 atm, PH2 ¼4.55 atm.

Please cite this article in press as: Atashi H, Rezaeian F, Modelling and optimization of FischereTropsch products through iron catalyst
in fixed-bed reactor, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.04.224
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 0 9

constant T and PH2 by increasing TOS, at high PCO , the CO2 references
selectivity is raised. The maximum selectivity of CO2 can be
achieved at PCO ¼3.8, TOS ¼ 547. The selectivity model of CO2 is
proposed as follows: [1] Cormos Calin-Cristian. Assessment of flexible energy vectors
polygeneration based on coal and biomass/solid wastes co-
gasification with carbon capture. Int J Hydrogen Energy
SðCO2 Þ ¼ 12324:46682  68:30447 T þ 0:48846 TOS
2013;38:7855e66.
 11:97525 PCO  1247:59629 PH2 þ 0:077145 T*T [2] van Vliet Oscar PR, Faaij Andre  PC, Turkenburg Wim C.
 0:30574 PH2 *PH2  0:0022604 T*TOS þ 4:98943 T*PH2 FischereTropsch diesel production in a well-to-wheel
perspective: a carbon, energy flow and cost analysis. Energy
þ 0:012252 TOS*PCO þ 1:31925 PCO *PH2 Convers Manag 2009;50:855e76.
(7) [3] Ratchahat Sakhon, Kodama Satoshi,
Tanthapanichakoon Wiwut, Sekiguchi Hidetoshi. CO2
In the model, the parameters such as PCO *PCO , TOS*TOS, gasification of biomass wastes enhanced by Ni/Al2O3 catalyst
T*PCO and TOS*PH2 have been omitted because their Pvalue were in molten eutectic carbonate salt. Int J Hydrogen Energy
higher than the other parameter and have less effect on the 2015;40:11809e22.
model. As can be seen in the model, the main parameter is PH2 [4] Samiran Nor Afzanizam, Jaafar Mohammad Nazri Mohd,
and has the most effect on the model and also, the important Ng Jo-Han, Lam Su Shiung, Chong Cheng Tung. Progress in
interactions belong to T*PH2 . In optimization part, the mini- biomass gasification technique e with focus on Malaysian
palm biomass for syngas production. Renew Sustain Energy
mization of CO2 was set as a target and it was found in
Rev 2016;62:1047e62.
T ¼ 249.0650  C, TOS ¼ 252.3054 h, PCO ¼11.60 atm and [5] Feyzi Mostafa, Khodaei Mohammad Mehdi,
PH2 ¼2.5 atm. Shahmoradi Jahangir. Effect of sulfur on the catalytic
performance of FeeNi/Al2O3 catalysts for light olefins
production. J Taiwan Inst Chem Eng 2014;45:452e60.
Conclusions [6] Pendyala Venkat Ramana Rao, Jacobs Gary, Bertaux Clement,
Khalid Syed, Davis Burtron H. FischereTropsch synthesis:
effect of ammonia on supported cobalt catalysts. J Catal
As mentioned earlier, a lot of works have been carried out on
2016;337:80e90.
the kinetics and product selectivity of FischereTropsch syn- [7] Forghani AA, Elekaei H, Rahimpour MR. Enhancement of
thesis. However, there are very few scholars who offer a gasoline production in a novel hydrogen permselective
model which can predict the selectivity of the products. This membrane reactor in FischereTropsch synthesis of GTL
method can be used within any temperature ranges based on technology. Int J Hydrogen Energy 2009;34:3965e76.
FischereTropsch reaction. [8] Mousavi Shabbir, Zamaniyan Akbar, Irani Mohammad,
Rashidzadeh Mehdi. Generalized kinetic model for iron and
Considering the importance of optimizing the reactor with
cobalt based FischereTropsch synthesis catalysts: review
respect to the valuable materials, this article determined the and model evaluation. Appl Catal A Gen 2015;506:57e66.
products model which can predict the selectivity of products [9] Yu Shiyong, Zhang Tao, Xie Yanhong, Wang Qinghua,
based on temperature, the partial pressure of CO and H2, and Gao Xuechuan, Zhang Renfei, et al. Synthesis and
time on stream. Four models were obtained for the selectivity characterization of iron-based catalyst on mesoporous
of products such as methane, ethane, ethylene, carbon diox- titania for photo-thermal FeT synthesis. Int J Hydrogen
ide using RSM. As shown by the model, the most important Energy 2015;40:870e7.
[10] Tao Zhichao, Yang Yong, Zhang Chenghua, Li Tingzhen,
parameter which has effect on the selectivity model, is PH2 . In
Ding Mingyue, Xiang Hongwei, et al. Study of manganese
addition, the main interaction belongs to T*PH2 , because its promoter on a precipitated iron-based catalyst for
coefficient has the highest amount. Finally by these models, FischereTropsch synthesis. J Nat Gas Chem 2007;16:278e85.
the optimal conditions were calculated and the maximum [11] Al-Dossary M, Fierro JLG. Effect of high-temperature pre-
conditions for ethylene were found as T ¼ 220  C, TOS ¼ 547 h, reduction in FischereTropsch synthesis on Fe/ZrO2 catalysts.
PCO ¼11.6 atm and PH2 ¼2.5 atm. Also, the conditions that can Appl Catal A Gen 2015;499:109e17.
[12] Galvis Hirsa M Torres, Bitter Johannes H, Khare Chaitanya B,
minimize the amount of CO2 were determined as
Ruitenbeek Matthijs, Dugulan A Iulian, de Jong Krijn P.
T ¼ 249.065  C, TOS ¼ 252.3054 h, PCO ¼11.60 atm and
Supported iron nanoparticles as catalysts for sustainable
PH2 ¼2.5 atm. production of lower olefins. Science 2012;335:835e8.
In industries, the application of these models has partic- [13] Hallac Basseem B, Brown Jared C, Baxter Larry L,
ular importance because it is used to achieve maximum pro- Argyle Morris D. A kinetic study on the structural and
duction by manipulating the operating parameters. So they functional roles of lanthana in iron-based high temperature
can produce the products which are very useful for environ- wateregas shift catalysts. Int J Hydrogen Energy
2014;39:7306e17.
mental purpose.
[14] Bukur Dragomir B, Mukesh Doble, Patel Snehal A. Promoter
effects on precipitated iron catalysts for FischereTropsch
synthesis. Ind Eng Chem Res 1990;29:194e204.
Acknowledgement [15] Yang Yong, Xiang Hong-Wei, Xu Yuan-Yuan, Bai Liang,
Li Yong-Wang. Effect of potassium promoter on precipitated
iron-manganese catalyst for FischereTropsch synthesis.
The authors express gratitude for the financial assistance Appl Catal A Gen 2004;266:181e94.
given by Iran's Ministry of Science and Research (NT19451827). [16] Bukur Dragomir B, Nowlcki Lech, Lang Xiaosu.
They are also grateful for further funding given by the Uni- FischereTropsch synthesis in a stirred Tank slurry reactor.
versity of Sistan and Baluchestan (NP1323102). Chem Eng Sci 1994;49:4615e25.

Please cite this article in press as: Atashi H, Rezaeian F, Modelling and optimization of FischereTropsch products through iron catalyst
in fixed-bed reactor, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.04.224
10 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 7 ) 1 e1 0

[17] Wan Hai-jun, Wu Bao-shan, Li Ting-zhen, Tao Zhi-chao, [23] Dhawane Sumit H, Kumar Tarkeshwar, Halder Gopinath.
Xia AN, Xiang Hong-wei, et al. Effects of SiO2 and Al2O3 on Parametric effects and optimization on synthesis of iron (II)
performances of iron-based catalysts for slurry doped carbonaceous catalyst for the production of biodiesel.
FischereTropsch synthesis. J Fuel Chem Technol Energy Convers Manag 2016;122:310e20.
2007;35:589e94. [24] Atashi Hossein, Gholizadeh Jaber, Tabrizi Farshad Farshchi,
[18] Yang Jia, Ma Wenping, Chen De, Holmen Anders, Tayebi Jaber, Mousavi Seyed Amir Hossein Seyed.
Davis Burtron H. Fischer-Tropsch synthesis: a review of the Thermodynamic analysis of carbon dioxide reforming of
effect of CO conversion on methane selectivity. Appl Catal A methane to syngas with statistical methods. Int J Hydrogen
Gen 2014;470:250e60. Energy 2017;23:5464e71.
[19] Arsalanfar M, Mirzaei AA, Bozorgzadeh HR, Atashi H. Effect [25] Brunner Kyle Martin. Novel iron catalyst and fixed-bed
of process conditions on the surface reaction rates and reactor model for the Fischer-Tropsch synthesis. 1st ed.
catalytic performance of MgO supported FeeCoeMn catalyst Brigham Young University; 2012.
for CO hydrogenation. J Ind Eng Chem 2012;18:2092e102. [26] Atashi Hossein, Shiva Mehdi, Farshchi Tabrizi Farshad,
[20] Golestan Somayeh, Mirzaei Ali Akbar, Atashi Hossein. Mirzaei Ali Akbar. Study of syngas conversion to light olefins
FischereTropsch synthesis over an ironecobaltemanganese by response surface methodology. J Chem 2013;2013:1e12.
(ternary) nanocatalyst prepared by hydrothermal procedure: [27] Shiva Mehdi, Atashi Hossein, Mirzaei Ali Akbar,
effects of nanocatalyst composition and operational Arsalanfar Maryam, Zare Akbar. Study of syngas conversion
conditions. Int J Hydrogen Energy 2017;42:9816e30. to light olefins by statistical models. Fuel 2014;123:205e10.
[21] Bukur Dragomir B, Lang Xiaosu, Akgerman Aydin, [28] Samimi Abdolreza, Zakeri Mozhgan, Maleki Basir, Mohebbi-
Feng Zhentao. Effect of process conditions on olefin Kalhori Davod. Experimental and statistical assessments of
selectivity during conventional and supercritical the mechanical strength reliability of gamma alumina
FischereTropsch synthesis. Ind Eng Chem Res catalyst supports. Particuology 2015;21:74e81.
1997;36:2580e7. [29] Atashi Hossein, Razmjooei Sam, Khorashadizadeh Mahdi,
[22] Haghtalab Ali, Mosayebi Amir. Co@Ru nanoparticle with Shivab Mehdi, Farshchi Tabrizi Farshad, Mousavi Seyed Amir
coreeshell structure supported over g-Al2O3 for Hossein Seyed. Effects of operating conditions on selectivity
FischereTropsch synthesis. Int J Hydrogen Energy of FeeCoeMn/MgO at high temperature CO hydrogenation. J
2014;39:18882e93. Taiwan Inst Chem Eng 2015;54:83e90.

Please cite this article in press as: Atashi H, Rezaeian F, Modelling and optimization of FischereTropsch products through iron catalyst
in fixed-bed reactor, International Journal of Hydrogen Energy (2017), http://dx.doi.org/10.1016/j.ijhydene.2017.04.224