You are on page 1of 7

CJCHE-00839; No of Pages 7

Chinese Journal of Chemical Engineering xxx (2017) xxxxxx

Contents lists available at ScienceDirect

Chinese Journal of Chemical Engineering

journal homepage:

Fluid Dynamics and Transport Phenomena

Effect of the operation parameters on the FischerTropsch synthesis in uidized

bed reactors
Xiaolai Zhang 1, Weixin Qian 1, Haitao Zhang 1, Qiwen Sun 2, Weiyong Ying 1,
Engineering Research Center of Large Scale Reactor Engineering and Technology, Ministry of Education, State Key Laboratory of Chemical Engineering,
East China University of Science and Technology, Shanghai 200237, China
State Key Laboratory of Coal Liquefaction and Coal Chemical Technology, Shanghai 201203, China

a r t i c l e i n f o a b s t r a c t

Article history: For the FischerTropsch synthesis (FTS), this paper presents a numerical investigation in a 3D uidized bed reac-
Received 12 April 2017 tor. The effect of the operation parameters such as bed temperature, supercial gas velocities, particle size and
Received in revised form 17 May 2017 bed heights is discussed. A 3D-CFD model coupled with FTS chemical kinetics was set up. The computational re-
Accepted 19 May 2017
sults are compared with experimental data in terms of the components production rates, etc. The analysis shows
Available online xxxx
that the bed heights, the bed temperature, the supercial gas velocities and particle sizes affect the C5+ selectivity
and the reaction rates. Product yields are dependent on the operating conditions especially the temperature.
Fluidized bed 2017 The Chemical Industry and Engineering Society of China, and Chemical Industry Press. All rights reserved.
FischerTropsch synthesis

1. Introduction technology is heat-removal. Krishna and Sie [13,14].mentioned that

MTFBR cannot prevent axial and radial temperature proles near the
FischerTropsch synthesis (FTS) is demonstrated as a feasible and inlet of the reactor. Fixed-bed technology also imposes limits on the
environmentally benign way for production of liquid to fuels from minimum catalyst particle size that can be applied. When the particles
coal, natural gas and biomass via synthesis gas. Also, the synthetic are very small, it will lead to a high pressure drop, While the particles
fuels have high quality due to the absence of aromatics, sulfur, and ni- are very large, the selectivity and activity will down. A clear advantage
trogen compounds [1]. It is a strategic alternative technology for liquid of a slurry-phase reactor is the well-mixed liquid phase resulting in
fuels and chemical materials in case of oil shortage and more stringent nearly isothermal operation [3]. Maretto and Krishna [15] showed that
environmental legislation on liquid fuels [2]. back mixing in slurry bubble-column reactors leads to signicant
Much attention has been paid for the development of FTS technology decreases in conversion per pass and reactor productivity. CFBR has a
in the past few years. The FTS process is rather exothermal with a heat of much higher capacity than MTFBR and MR, but the circulation of a
reaction of about 170 kJmol1 [3], So, the sufcient capability of the large amount of catalyst results in considerable compression of the recy-
heat removal is a critical point for developing the commercial FT reac- cle gas associated with capital and operation costs [16]. As uidized
tors. Increased temperature results in higher methane production and beds have excellent heat transfer, easy heat removal and temperature
faster catalyst deactivation due to sintering and coking. In general, uniformity, the main advantages of FBR are lower operation and con-
there are ve main types of commercial FT reactors are commonly structing costs and higher capacities [9].
used, which are known as multi tubular xed-bed reactor (MTFBR), Because of the complexity of the FTS process, mathematical model-
slurry phase reactor(SPR), circulating uidized-bed reactor(CFBR), ing and numerical simulation have been a focus of its researches
uidized-bed reactor (FBR), and micro channel or micro structured re- for many years, among which a number of models have been
actors (MR) [411]. proposed on xed bed reactors, SPRs and MRs. These model can be
It is generally considered that the main advantages of the MTFBR are roughly divided into three categories: (a) plug ow model [17,18];
easy scaling-up, less need for expensive demonstration units, and for (b) pseudo-homogeneous models (both 1D [8,1924]and 2D [5,25]);
separation of catalyst and liquid product, lower catalyst attrition and (c) computational uid dynamics (CFD) models, with or without the
higher catalyst load [12]. The big challenge in applying the xed-bed FTS chemistry [11,2628]. However, studies on the simulation of FBRs
are seldom reported. Fernandes [29,30] proposed a mathematical
model of FBR to describe the effect of several operation parameters on
Supported by the National High-Tech R&D Program of China [2011AA05A204] and the
Fundamental Research Funds for the Central Universities [222201717013].
syngas conversion and the distribution of hydrocarbons. In order to sim-
Corresponding author. ulate the combination of uidized and xed bed membrane reactors, a
E-mail address: (W. Ying). one-dimensional heterogeneous model is made up based on the
1004-9541/ 2017 The Chemical Industry and Engineering Society of China, and Chemical Industry Press. All rights reserved.

Please cite this article as: X. Zhang, et al., Effect of the operation parameters on the FischerTropsch synthesis in uidized bed reactors, Chin. J.
Chem. Eng. (2017),
2 X. Zhang et al. / Chinese Journal of Chemical Engineering xxx (2017) xxxxxx

kinetics of the main reactions, the mass transfer equations and the heat includes a kinetic and a collisional term. The shear viscosity of particles
transfer equations. The Sasol R&D group developed a pilot plant for the is consisted of three parts, including the collision viscosity, the frictional
FTS [31], reporting similar selectivity and higher conversion than those viscosity and the kinetic viscosity [22].
in commercial CFBR. A transport equation for static enthalpy is used to describe the heat
CFD models are advantageous for the optimization of many complex ux in gas phase, and in solid phase the heat transfer is considered sim-
processes, especially for the gassolid ow in FRBs with close and com- ilar to a continuous phase [53]. The interphase heat transfer resistance is
plex interactions between the two phases and the heat and mass trans- regarded to be associated with the heat transfer coefcient and the tem-
fer, etc. The underlying physicochemical processes can be clearly perature difference [54] (Table 2).
understood through CFD simulation [3236]. The EulerianEulerian
models combined with the kinetic theory are the most popular method 2.3. Chemical model
for engineering simulation of gassolid ow. The main features of the
ows in uidized beds has been obtained by either laminar or turbu- There are many different authors proposed the kinetic mechanism
lence models [3740]. The turbulent interactions between the solid during the F-T synthesis, which can be roughly divided into two
and the gas phase have been elucidated by several authors [4143]. Re- categories: one is depended on the Langmuir Hinshelwood Hogan
cently, the turbulence model enforced in the CFD code MFIX (Multi- Watson(LHHW) model [55,56], while the other one is based on the
phase Flow with Interphase Exchanges) developed by Simonin's group power law of chemical rate expression [57,58]. The mechanism of the
has predicted dilute gassolid ows under an appropriate boundary Fe-HZSM5 Bifunctional Catalyst proposed by Marvast et al. [58].was
conditions reasonably [4446]. adopted in this study. It was supposed that the F-T reaction include
In this work, the FTS process in a three-dimension (3D) uidized bed CO, H2, CO2, H2O, CH4, C2H4, C2H6, C3H8, n-C4H10, i-C4H10, and C5+ spe-
reactor was numerically studied aiming to get a better description. Thus, cies. The following reactions (1)(8) were considered as dominating
a 3D model coupled with gassolid momentum, mass and heat trans- F-T reactions.
port, and FT reactions was employed. The operation parameters on
the production rates, the temperature distribution and the selectivity CO 3H2 CH4 H2 O 1
were analyzed based on the simulations in this model.
2CO 4H2 C2 H4 2H2 O 2
2. Models and Methods
A 3D computational uid dynamics (CFD) model coupled with FTS 2CO 5H2 C2 H6 2H2 O 3
reaction which considered the uid ow, the heat transfer and the
mass transfer mechanism was employed in this paper. 3CO 7H2 C3 H8 3H2 O 4

2.1. Conservation equations R5

4CO 9H2 nC4 H10 4H2 O 5

The gassolid ow conservation equations is described briey in R6

4CO 9H2 iC4 H10 4H2 O 6
Table 1, and more details can be found in [45,47].
2.2. Constitutive correlations 6:05CO 12:23H2 C6:05 H12:36 C5 6:05H2 O 7

Constitutive correlations are required for closing the governing CO H2 OCO2 H2 WGS reaction 8
conservation equations. The gas-phase stress tensor always uses the
Newtonian closure. The gassolid momentum exchange coefcient The reactions (1)(7) are a general form for the rate equations. It is a
(i.e., the interphase drag coefcient), which is a crucial parameter for function of the yield pressures of hydrogen and carbon monoxide, are
the successful simulation, is taken the same as Syamlal et al. [47,48]. expressed as:
The solid stress tensor is calculated from the kinetic theory of the h i
granular [49], using two completely different models, for viscous shear- Ri ki exp Ei =RT P m ni
mol  kg  s1 9
ing and plastic regimes [47,48], where an algebraic expression of the
granular temperature m is used. The solid pressure, which considers where P is partial pressure and k is the apparent rate constant. Table 3
the normal forces because of particle-particle interactions [47,4952], presented the kinetic parameters.

Table 1
Conservation equations ([45,47])

Conserved quantities Phase Equation

Mass Gas and solid
l l x l l U li Rl

Momentum Gas P g gij

g g U gi x g g U g j U gi g xi
g x j
Igmi g g g i f gi Rm gm U mi 1gm gm 

Solid P
x m m U m j U mi g xg m xmij I gmi m m g i Rm gm m 1gm gm 
m m U mi
j i j

Turbulent energy Gas kg k k U

g g t U g j xg x g gk xg g gij xgi kg g g g
j i i j

g g tg U g j xg x g g xg g kgg gij xgi g g g

Solid m m tm U m j m
x m m 2 m
m m mij U mi
km m m m
x j x
i i x j
m 121e2 2m m g 0 dmp
Energy Gas T
g g C pg tg g T g  qg Hgm H g Hgw
Solid T
g g C pg tg g T g  qg Hgm H g Hgw
Species Gas and solid
ln l X ln  l l X ln l Rln

Please cite this article as: X. Zhang, et al., Effect of the operation parameters on the FischerTropsch synthesis in uidized bed reactors, Chin. J.
Chem. Eng. (2017),
X. Zhang et al. / Chinese Journal of Chemical Engineering xxx (2017) xxxxxx 3

Table 2
Constitutive relations ([45,47])

Type Constitutive relation

Stress tensor for the gas phase U
gij P g g xgi ij 2 g Sgij

Gas phase deformation rate U U U

Sgij 12 xgi xg j 13 ij xgi
 j i
Solid phase tress tensor mij P m b U mi
ij 2 m Smij
Solid phase deformation rate U m j
Smij 2 x x 13 ij U
1 U mi
j i i

Solid pressure Pm = mmm[1 + 2(1 + emm)mg0,mm]

Bulk viscosity 1
b 83 m m dm g om s 2
Solid phase frictional tress 1025 g g 10 sin
p b g
I 2D
Shear viscosity 1=2 10 dm m 2
m m;col m;kin 45 m m dm g 0;mm 1 emm m 96mm1e mm g 0;ss
1 45 1 emm m g 0;mm 
  1 q 
Solids granular conductivity
2 m m 9=10kgm t 32 m 1 wc m g0 9=5 x cc 1 18=5 m g0 1e 2 m g 0 1 edp

gm m

Table 3
Kinetic parameters for FTS reactions in Eqs. (1)(7)

Reaction no. ki Ei mi ni

1 142583.8 83423.9 1.0889 1.5662

2 51.556 65018 0.7622 0.0728
3 24.717 49782 0.5645 1.3155
4 0.4632 34885.5 0.4051 0.6635
5 0.00474 27728.9 0.4728 1.1389
6 0.00832 25730.1 0.8204 0.5026
7 0.02316 23564.3 0.5850 0.5982

The water-gas shift (WGS) reaction is reversible as shown in Eq. (8).

The corresponding reaction rate is presented as follows [59]:

r WGS 15:7 exp 45080=RT P CO P H2 O =P 0:5 0:5
H2 P CO2 P H2 =K WGS
h  i 10
1 1:13  103 P CO P H2 O =P 0:5

For a watergas shift reaction, KWGS is the equilibrium constant,

which is:

K WGS 5:8972089 13:958689 Fig. 1. Schematic diagram of the simulated uidized bed reactor.
 104 T27:592844  108 T 2 11

To solve the transport and reaction for the model a suitable time
3. Simulation Setup step should be used, to ensure the accuracy and stability during
the calculation, was 0.00001 s in this study. The segregated solution
3.1. Reactor geometry strategy has been adopted to prevent the iterative divergent. In order
to eliminate the pressurevelocity coupling, The traditional SIMPLE
The simulated uidized bed is described in Fig. 1, the inner diameter method was used. Convective uxes of the transport equations were
of which is 30 mm and the height is 500 mm. all approximated by the Super bee spatial scheme with a second-
The synthesis gas enters from the bottom of the uidized bed which order accuracy. Linear discretization for the pressure was adopted. A
consists of CO and H2. The FischerTropsch synthesis takes place over convergence criterion of 1 105 was taken here for all the transport
the iron based catalyst powder, the effect is reected on the reaction ac- equations.
tivation energy. After the reactions, the products and reactant leave
from the outlet in the upper part of the reactor. All experiments were 3.3. Initial and boundary conditions
carried out at constant pressure conditions.
The gas distribution is assumed to be uniform at the reactor en-
3.2. Numerical method and parameters trance. In order to reach the desired bed expansion, the ow rate was
regulated based on the particle size. The inlet gas temperature is xed
In this study, the MFIX, which is described in Section 2, is used. The at a given value, and the reactor wall was assigned a series of given tem-
discretization of all the governing non-linear partial differential equa- peratures in different simulations. The gas phase used a no slip bound-
tions is accomplished using the nite volume method on the staggered ary and the solid phase used a free slip boundary. The outlet used a
grid. It is found that the grid size of 1.0 mm 1.0 mm 2.0 mm is suf- condition of the pressure outlet. Table 4 species the parameters in
cient for neglecting the numerical errors due to the grid resolution. these initial and boundary conditions and the particle properties.

Please cite this article as: X. Zhang, et al., Effect of the operation parameters on the FischerTropsch synthesis in uidized bed reactors, Chin. J.
Chem. Eng. (2017),
4 X. Zhang et al. / Chinese Journal of Chemical Engineering xxx (2017) xxxxxx

Table 4
Operating conditions and material properties of the simulated uidized bed reactor

Parameters Value

Supercial gas velocity 512umf

Mole ratio of H2 to CO in feed 1.13
Particle Sauter mean diameter/m 100300
Particle real density/kgm3 1290
Minimum void age of the particles 0.488
Static bed height/cm 4080
Pressure/kPa 17
Minimum uidization velocity/cms1 3.5412.2
Inlet gas temperature/K 560

4. Results and Discussion

4.1. Model validation

The data of the simulations and the experiments of Rahmati et al.

[57] about the methane production rates are plotted in Fig. 2. Compared
Fig. 3. Effect of the cooling temperature on the production selectivity and the production
with the experimental data, the average relative error of simulation re-
rates (inlet gas temperature = 560 K, inlet H2:CO mole ratio = 1.13, P = 1700 kPa, Dp =
sults is 27.0%, which is in qualitatively agreement to the experiment. 200 m).
Considering the difference of the reaction conditions between the
xed bed reactor in the literature and in the uidized bed reactor in
this study, the simulation results are still reasonable for a qualitative Fig. 4 presents the axial temperature distribution at different cooling
study on the effective of major operation parameters, though quantita- temperatures for the gas phase. In general, with the growing of the
tive relationships can hardly be found further. cooling temperature the average bed temperature in the reactor rises,
which causes all production rates to increase with the rising of the
cooling temperature (see Fig. 3(b)). Based on the Arrhenius equation,
the value of the activation energy of Eq. (1) is greater than the activation
energy of Eq. (7), the increase of methane production with the temper-
ature could be much faster than the methane production of C5+. This
might be the reason that the C5+ selectivity sharply decreased when re-
action temperature was raised. Other production selectivity has little
change with the temperature within 10.0%. It should be noted that the
cooling temperature is different from the average bed temperature in
corresponding condition. For the cooling temperatures of 450.0 K and
556.0 K, the temperature differences are 45.0 K and 69.3 K, which are
1.22 times of those for the cooling temperatures of 500.0 K and 520 K.
Larger temperature differences must lead to more energy loss.

Fig. 2. Methane production rates obtained from the simulation and from literature
(Rahmati [57]).

4.2. Effect of reaction temperature

During the FT reactions, the surface temperature of the contact mix-

ture must be kept at the required level, and the process is rather exo-
thermal, therefore the cooling mode is vital to the FTS process. As seen
in Fig. 3 (a), the effects of wall cooling temperature (300556 K) on
the production selectivity, which is dened as Xi/Xprodcuction (Xi and
Xproduction are the mass ow rate of the ith product and all products leav- Fig. 4. The axial temperature distribution of gas phase at different cooling
ing the reactor, respectively). The selectivity of C5+ decreases quickly temperature. (Inlet gas temperature = 560 K, inlet H 2 :CO mole ratio = 1.13, P =
1700 kPa, Dp = 200 m).
with the rise of the cooling temperature, especially when temperature
exceeds 450 K. The C5+ selectivity is 56.4% at 300 K of the cooling tem-
perature. And to 550 K, it is less than 20% while the light production, es- 4.3. Effect of particle size
pecially methane, which is not the product of choice in modern FTS [31]
increases when the cooling temperature was raised, especially after it The FTS product distributions of the catalyst with the different parti-
exceeded 450 K. The phenomenon was also observed by Zhu et al. [60]. cle diameters (Dp) are shown in Fig. 5. For C5+ production, with the rise

Please cite this article as: X. Zhang, et al., Effect of the operation parameters on the FischerTropsch synthesis in uidized bed reactors, Chin. J.
Chem. Eng. (2017),
X. Zhang et al. / Chinese Journal of Chemical Engineering xxx (2017) xxxxxx 5

Fig. 5. Effect of catalyst particle diameter on the production selectivity and the production
Fig. 7. The production selectivity versus catalyst particle diameter (inlet gas temperature =
rates(inlet gas temperature = 560 K, inlet H2:CO mole ratio = 1.13, P = 1700 kPa, bed
560 K, inlet H2:CO mole ratio = 1.63, P = 1700 kPa, bed height = 50 cm, cooling
height = 50 cm, cooling temperature = 520 K).
temperature = 520 K).

of the particle diameter the selectivity rstly grows, the maximum value velocity is 8 times lower than Umf, the product rates are nearly linear
of 33.4% is achieved at 200 m diameter, which then slightly decreases. growth when the supercial gas velocity is growing, then decreases.
While the change trend of the methane production rate is opposite to it. For a uidized bed reactor, the rate-limiting step is usually decided by
The phenomena can be clearly explained by the axial temperature dis- mass transport or surface reaction kinetics [62,63]. It can be concluded
tribution of the gas phase as plotted in Fig. 6. It illustrates that the that when the syngas velocity is 8 times lower than Umf, the mass trans-
axial temperature difference is larger for smaller particle, which reaches port limits the process, but when velocity is 8 times over Umf, the limi-
the largest value of 65.0 K for 100 m particles, and the lowest value of tation is the reaction kinetics. When the syngas velocity is 8 times less
7.1 K for 200 m particles. The peak value of the temperature for 100 m than Umf, the C5+ selectivity and its production rate show the opposite
particles is also the maximum at 588.3 K. Because the FTS is a strong trend shown in Fig. 7(a).
exothermic reaction, the small particles tend to aggregate, which leads Fig. 8 shows the axial gas temperature distribution for different su-
to poor heat transfer. Meanwhile, smaller particles have higher reaction percial gas velocities. For smaller gas velocity, bed temperature is
activity [61]. These cause higher solid and gas temperature and uneven lower, which may explain the above phenomena. As the supercial
temperature distribution. As observed above, when the reaction tem- gas velocity grows, the gas residence time is shorter, which will reduce
perature is high, the C5+ production rate is low. This may explain that the reaction rate and the generation of the heat. Moreover, the heat
with the growth of the catalyst particle diameter, the production rates transfer rate among the gas and solid phase is higher.
sharply increase and then decrease, as shown in Fig. 5(b).

4.4. Effect of supercial gas velocity

The effect of supercial gas velocity is given in Fig. 7. The supercial

gas velocity greatly effects on the production rates. As the supercial gas

Fig. 8. Axial temperature distribution for different supercial gas velocity (inlet gas
temperature = 560 K, inlet H2:CO mole ratio = 1.63, P = 1700 kPa, bed height = 50 cm,
cooling temperature = 520 K).

4.5. Effect of bed height

Fig. 6. Axial gas temperature distribution with different diameter catalyst particle
(inlet gas temperature = 560 K, inlet H2:CO mole ratio = 1.13, P = 1700 kPa, bed Fig. 9 illustrates the effects of bed height on FTS products. Methane,
height = 50 cm, cooling temperature = 520 K). propane, ethane and C5 + production rates slightly increase at the

Please cite this article as: X. Zhang, et al., Effect of the operation parameters on the FischerTropsch synthesis in uidized bed reactors, Chin. J.
Chem. Eng. (2017),
6 X. Zhang et al. / Chinese Journal of Chemical Engineering xxx (2017) xxxxxx

q conductive heat ux, kJmol1

ij Kronecker Delta
m the granular temperature, K
2 solid granular conductivity, Wm1K1
,b bulk viscosity, Pas
m shear viscosity, Pas
density, kgm3
ij second-order stress tensor
xgm particle relaxation time, s
cm collisional time-scale, s

g gas phase
i,j indices to identify vector and tensor components
l liguid phase
m solid phase
WGS watergas shift
Fig. 9. Effect of bed height on the selectivity and rate of the products (inlet gas
temperature = 560 K, inlet H2:CO mole ratio = 1.13, P = 1700 kPa, Dp = 200 m,
supercial gas velocity = 8Umf, cooling temperature = 520 K). References
[1] M.R. Rahimpour, M.H. Khademi, A.M. Bahmanpour, A comparison of conventional
and optimized thermally coupled reactors for FischerTropsch synthesis in GTL
average rates of about 5.0 105 molcm3s1 per cm bed height. technology, Chem. Eng. Sci. 65 (23) (2010) 62066214.
Higher bed height can provide more particle-surface area, which leads [2] K.J. Woo, S.H. Kang, S.M. Kim, et al., Performance of a slurry bubble column reactor
for FischerTropsch synthesis: Determination of optimum condition, Fuel Process.
to more heat generation and bed temperature rise. Thus, the selectivity Technol. 91 (4) (2010) 434439.
of C5 + slightly declines when the bed height is increased, while the [3] R.M. de Deugd, F. Kapteijn, J.A. Moulijn, Trends in FischerTropsch reactor
methane selectivity has increased accordingly as shown in Fig. 8(a). technologyOpportunities for structured reactors, Top. Catal. 26 (14) (2013) 2939.
[4] P. Narataruksa, S. Tungkamani, K. Pana-Suppamassadu, et al., Conversion enhance-
Other production selectivity has less change. ment of tubular xed-bed reactor for FischerTropsch synthesis using static mixer,
J. Nat. Gas Chem. 21 (4) (2012) 435444.
5. Conclusions [5] X.P. Dai, P.Z. Liu, Y. Shi, et al., FischerTropsch synthesis in a bench-scale two-stage
multitubular xed-bed reactor: Simulation and enhancement in conversion and
diesel selectivity, Chem. Eng. Sci. 105 (2014) 111.
The FTS process in a uidized bed was simulated using the CFD code [6] G.P. Laan, A. Beenackers, R. Krishna, Multicomponent reaction engineering model
MFIX for gassolid ow and heat and mass transfer, together with a for Fe-catalysed FischerTropsch synthesis in commercial scale bubble column slur-
ry reactors, Chem. Eng. Sci. (1999) 54.
reaction kinetic model of Marvast et al. [58]. The effect of the operation
[7] S.T. Sie, R. Krishna, Fundamentals and selection of advanced Fischer-Tropsch reac-
parameters was investigated. It was found that the effective reaction tors, Appl. Catal. A Gen. 186 (1) (1999) 5570.
rate and selectivity depended on the cooling temperature (reaction [8] N. Rados, M.H. Al-Dahhan, M.P. Dudukovic, Modeling of the FischerTropsch syn-
thesis in slurry bubble column reactors, Catal. Today 79 (2003) 211218.
temperature), the particle diameter, the supercial gas velocity and
[9] M.E. Dry, High quality diesel via the FischerTropsch processA review, J. Chem.
the bed height. The order of signicance of the effects is: the reaction Technol. Biotechnol. 77 (1) (2002) 4350.
temperature N the particle diameter N the supercial gas velocity N the [10] L.C. Almeida, O. Sanz, J. D'olhaberriague, et al., Microchannel reactor for Fischer
bed height. The selectivity of C5+ decreased sharply with the rising of Tropsch synthesis: Adaptation of a commercial unit for testing microchannel blocks,
Fuel 110 (2013) 171177.
the reaction temperature, but less affected by other reaction parame- [11] M.S. Shin, N. Park, M.J. Park, et al., Computational uid dynamics model of a modular
ters. The results of the simulations emphasize that temperature control multichannel reactor for FischerTropsch synthesis: Maximum utilization of catalytic
is important. Product yields are dependent on operating conditions es- bed by microchannel heat exchangers, Chem. Eng. J. 234 (2013) 2332.
[12] J.J.C. Geerlings, J.H. Wilson, G.J. Kramer, et al., FischerTropsch technologyFrom
pecially the temperature. active site to commercial process, Appl. Catal. A Gen. 186 (1) (1999) 2740.
Generally, the continuum model used in this study agrees reason- [13] R. Krishna, S.T. Sie, Design and scale-up of the FischerTropsch bubble column slurry
ably with the experiments in terms of reaction rate and C5+ selectivity. reactor, Fuel Process. Technol. 64 (1) (2000) 73105.
[14] S.T. Sie, R. Krishna, Fundamentals and selection of advanced FischerTropsch
However, improvements are required for the model to further investi- reactors, Appl. Catal. A Gen. 186 (1) (1999) 5570.
gate the uidized bed reactors for FTS process. [15] C. Maretto, R. Krishna, Modelling of a bubble column slurry reactor for Fischer
Tropsch synthesis, Catal. Today 52 (2) (1999) 279289.
[16] B. Jager, M.E. Dry, T. Shingles, et al., Experience with a new type of reactor for
Nomenclature FischerTropsch synthesis, Catal. Lett. 7 (1) (1990) 293301.
Cp specic heat capacity, kJmol1K1 [17] M.R. Rahimpour, S.M. Jokar, Z. Jamshidnejad, A novel slurry bubble column mem-
e restitution coefcient brane reactor concept for FischerTropsch synthesis in GTL technology, Chem. Eng.
g gravity, ms2 Res. Des. 90 (3) (2012) 383396.
[18] L.C. Almeida, O. Sanz, D. Merino, et al., Kinetic analysis and microstructured
gom radial distribution function at contact, reactors modeling for the FischerTropsch synthesis over a CoRe/Al2O3 catalyst,
H heat transfer, kJmol1 Catal. Today 215 (2013) 103111.
I interaction force representing the momentum transfer, [19] Y.N. Wang, Y.Y. Xu, Y.W. Li, et al., Heterogeneous modeling for xed-bed Fischer
Tropsch synthesis: Reactor model and its applications, Chem. Eng. Sci. 58 (3)
ms2 (2003) 867875.
P phase pressure, Pa [20] M.R. Rahimpour, H. Elekaei, A comparative study of combination of FischerTropsch
Rl the inter phase mass transfer terms between the gas and solid synthesis reactors with hydrogen-permselective membrane in GTL technology,
Fuel Process. Technol. 90 (6) (2009) 747761.
phase [21] M. Bayat, M.R. Rahimpour, B. Moghtaderi, Genetic algorithm strategy (GA) for
T the temperature, K optimization of a novel dual-stage slurry bubble column membrane conguration
U velocity component, ms1 for FischerTropsch synthesis in gas to liquid (GTL) technology, J. Nat. Gas Sci. Eng.
3 (4) (2011) 555570.
Xln mass fraction of n species in the lth phase, [22] N.A. Mamonov, L.M. Kustov, S.A. Alkhimov, et al., One-dimensional heterogeneous
volume fraction model of a FischerTropsch synthesis reactor with a xed catalyst bed in the
H the heat of reaction, kJmol1 isothermal granules approximation, Catal. Ind. 5 (3) (2013) 223231.

Please cite this article as: X. Zhang, et al., Effect of the operation parameters on the FischerTropsch synthesis in uidized bed reactors, Chin. J.
Chem. Eng. (2017),
X. Zhang et al. / Chinese Journal of Chemical Engineering xxx (2017) xxxxxx 7

[23] N. Moazami, H. Mahmoudi, K. Rahbar, et al., Catalytic performance of cobaltsilica [43] C.M. Hrenya, J.L. Sinclair, Effects of particle-phase turbulence in gassolid ows,
catalyst for FischerTropsch synthesis: Effects of reaction rates on efciency of AIChE J. 43 (4) (1997) 853869.
liquid synthesis, Chem. Eng. Sci. 134 (2015) 374384. [44] J. He, O. Simonin, Non-equilibrium prediction of the particle-phase stress tensor in
[24] N. Moazami, M.L. Wyszynski, H. Mahmoudi, et al., Modelling of a xed bed reactor vertical pneumatic conveying, ASME 166 (1993) 253263.
for FischerTropsch synthesis of simulated N2-rich syngas over Co/SiO2: Hydrocar- [45] R. Garg, J. Galvin, T. Li, Open-source MFIX-DEM software for gas-solids ows: Part I
bon production, Fuel 154 (2015) 140151. Verication studies, Powder Technol. 220 (2012) 122137.
[25] N. Park, J.R. Kim, Y. Yoo, et al., Modeling of a pilot-scale xed-bed reactor for iron- [46] S. Benyahia, M. Syamlal, T.J. O'Brien, Evaluation of boundary conditions used to
based FischerTropsch synthesis: Two-dimensional approach for optimal tube model dilute, turbulent gas/solids ows in a pipe, Powder Technol. 156 (2) (2005)
diameter, Fuel 122 (2014) 229235. 6272.
[26] J. Na, K.S. Kshetrimayum, U. Lee, et al., Multi-objective optimization of microchannel [47] S. Benyahia, M. Syamlal, T.J. O'Brien, Study of the ability of multiphase continuum
reactor for Fischer-Tropsch synthesis using computational uid dynamics and models to predict core-annulus ow, AIChE J. 53.10 (2007) 25492568.
genetic algorithm, Chem. Eng. J. 313 (2017) 15211534. [48] P.C. Johnson, R. Jackson, Frictionalcollisional constitutive relations for granular
[27] Y. Li, G. Xie, T. Lei, et al., A CFD model for gas uniform distribution in turbulent ow materials, with application to plane shearing, J. Fluid Mech. 176 (1987) 6793.
for the production of titanium pigment in chloride process, Chin. J. Chem. Eng. 24 (6) [49] C. Lun, S.B. Savage, D.J. Jeffrey, et al., Kinetic theories of granular ow: Simple shear
(2016) 749756. of inelastic particles and general deformations of nearly elastic particles, J. Fluid
[28] S. Wang, B. Shao, R. Liu, et al., Comparison of numerical simulations and experiments Mech. 140 (1984) 223256.
in conical gassolid spouted bed, Chin. J. Chem. Eng. 23 (10) (2015) 15791586. [50] D. Gidaspow, L. Huilin, Collisional viscosity of FCC particles in a CFB, AIChE J. 42 (9)
[29] F.A.N. Fernandes, E.M.M. Sousa, FischerTropsch synthesis product grade optimiza- (1996) 25032510.
tion in a uidized bed reactor, AIChE J. 52 (8) (2006) 28442850. [51] L. Huilin, H. Yurong, D. Gidaspow, Hydrodynamic modelling of binary mixture in a
[30] M.R. Rahimpour, H. Elekaei, Optimization of a novel combination of xed and gas bubbling uidized bed using the kinetic theory of granular ow, Chem. Eng.
uidized-bed hydrogen-permselective membrane reactors for FischerTropsch Sci. 58 (7) (2003) 11971205.
synthesis in GTL technology, Chem. Eng. J. 152 (2) (2009) 543555. [52] D.G. Schaeffer, Instability in the evolution equations describing incompressible
[31] M.E. Dry, Practical and theoretical aspects of the catalytic Fischer-Tropsch process, granular ow, J. Differ. Equ. 66 (1) (1987) 1950.
Appl. Catal. A Gen. 138 (2) (1996) 319344. [53] M. Syamlal, D. Gidaspow, Hydrodynamics of uidization: Prediction of wall to bed
[32] X. Lan, C. Xu, G. Wang, et al., CFD modeling of gassolid ow and cracking reaction in heat transfer coefcients, AIChE J. 31 (1) (1985) 127135.
two-stage riser FCC reactors, Chem. Eng. Sci. 64 (17) (2009) 38473858. [54] D.J. Gunn, Transfer of heat or mass to particles in xed and uidised beds, Int. J. Heat
[33] C. Wu, Y. Cheng, Y. Ding, et al., CFDDEM simulation of gassolid reacting ows in Mass Transf. 21 (4) (1978) 467476.
uid catalytic cracking (FCC) process, Chem. Eng. Sci. 65 (1) (2010) 542549. [55] H. Atashi, F. Siami, A.A. Mirzaei, et al., Kinetic study of FischerTropsch process
[34] Q. Xue, T.J. Heindel, R.O. Fox, A CFD model for biomass fast pyrolysis in uidized-bed on titania-supported cobaltmanganese catalyst, J. Ind. Eng. Chem. 16 (6) (2010)
reactors, Chem. Eng. Sci. 66 (11) (2011) 24402452. 952961.
[35] K. Papadikis, S. Gu, A.V. Bridgwater, CFD modelling of the fast pyrolysis of biomass in [56] J. Yang, Y. Liu, J. Chang, et al., Detailed kinetics of FischerTropsch synthesis on an
uidised bed reactors. Part B: Heat, momentum and mass transport in bubbling industrial FeMn catalyst, Ind. Eng. Chem. Res. 42 (21) (2003) 50665090.
uidised beds, Chem. Eng. Sci. 64 (5) (2009) 10361045. [57] M. Rahmati, M. Mehdi, M. Bargah-Soleimani, Rate equations for the FischerTropsch
[36] K. Papadikis, S. Gu, A.V. Bridgwater, CFD modelling of the fast pyrolysis of biomass in reaction on a promoted iron catalyst, Can. J. Chem. Eng. 79 (5) (2001) 800804.
uidised bed reactors: Modelling the impact of biomass shrinkage, Chem. Eng. J. 149 [58] M.A. Marvast, M. Sohrabi, S. Zarrinpashne, et al., FischerTropsch synthesis:
(1) (2009) 417427. Modeling and performance study for Fe-HZSM5 bifunctional catalyst, Chem. Eng.
[37] M.J.V. Goldschmidt, J.A.M. Kuipers, W.P.M. Van Swaaij, Hydrodynamic modelling of Technol. 28 (1) (2005) 7886.
dense gas-uidised beds using the kinetic theory of granular ow: Effect of coef- [59] Y.N. Wang, Y.Y. Xu, H.W. Xiang, et al., Modeling of catalyst pellets for Fischer
cient of restitution on bed dynamics, Chem. Eng. Sci. 56 (2) (2001) 571578. Tropsch synthesis, Ind. Eng. Chem. Res. 40 (20) (2001) 43244335.
[38] S. Benyahia, H. Arastoopour, T.M. Knowlton, et al., Simulation of particles and gas [60] X. Zhu, X. Lu, X. Liu, et al., Heat transfer study with and without FischerTropsch
ow behavior in the riser section of a circulating uidized bed using the kinetic reaction in a xed bed reactor with TiO2, SiO2, and SiC supported cobalt catalysts,
theory approach for the particulate phase, Powder Technol. 112 (1) (2000) 2433. Chem. Eng. J. 247 (2014) 7584.
[39] A. Almuttahar, F. Taghipour, Computational uid dynamics of high density circu- [61] G. Chabot, R. Guilet, P. Cognet, et al., A mathematical modeling of catalytic milli-xed
lating uidized bed riser: Study of modeling parameters, Powder Technol. 185 (1) bed reactor for FischerTropsch synthesis: Inuence of tube diameter on Fischer
(2008) 1123. Tropsch selectivity and thermal behavior, Chem. Eng. Sci. 127 (2015) 7283.
[40] A. Neri, D. Gidaspow, Riser hydrodynamics: Simulation using kinetic theory, AIChE J. [62] F.M. Dautzenberg, M. Mukherjee, Process intensication using multifunctional
46 (1) (2000) 5267. reactors, Chem. Eng. Sci. 56 (2) (2001) 251267.
[41] S. Dasgupta, R. Jackson, S. Sundaresan, Turbulent gas-particle ow in vertical risers, [63] P. Zhang, J.H. Duan, G.H. Chen, et al., Effect of bed characters on the direct synthesis
AIChE J. 40 (2) (1994) 215228. of dimethyldichlorosilane in uidized bed reactor, Sci. Rep. UK 5 (2015) 18.
[42] E.J. Bolio, J.L. Sinclair, Gas turbulence modulation in the pneumatic conveying of
massive particles in vertical tubes, Int. J. Multiphase Flow 21 (6) (1995) 9851001.

Please cite this article as: X. Zhang, et al., Effect of the operation parameters on the FischerTropsch synthesis in uidized bed reactors, Chin. J.
Chem. Eng. (2017),