Sie sind auf Seite 1von 8

Journal of Energy Chemistry 24(2015)23–30

Numerical simulation of packed-bed reactor


for oxidative coupling of methane
Zhao Zhang, Ziqi Guo, Shengfu Ji∗
State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, Beijing 100029, China
[ Manuscript received June 27, 2014; revised August 13, 2014 ]

Abstract
A three-dimensional geometric model of the oxidative coupling of methane (OCM) packed-bed reactor loaded with Na2 WO4 -Mn/SiO2 partic-
ulate catalyst was set up, and an improved Stansch kinetic model was established to calculate the OCM reactions using the computational fluid
dynamics method and Fluent software. The simulation conditions were completely the same with the experimental conditions that the volume
velocity of the reactant was 80 mL/min under standard state, the ratio of CH4 /O2 was 3, the temperature and pressure were 800 ◦ C and 1 atm,
respectively. The contour of the characteristics parameters in the catalyst bed was analyzed, such as the species mass fractions, temperature,
the heat flux on side wall surface, pressure, fluid density and velocity. The results showed that the calculated values matched well with the
experimental values on the conversion of CH4 and the selectivity to products (C2 H6 , C2 H4 , CO2 , CO) in the reactor outlet with an error range
of ±2%. The mass fractions of CH4 and O2 decreased from 0.6 and 0.4 in the catalyst bed inlet to 0.436 and 0.142 in the outlet, where the
mass fractions of C2 H6 , C2 H4 , CO and CO2 were 0.035, 0.061, 0.032 and 0.106, respectively. Due to the existence of laminar boundary layer,
the contours of each component bent upwards in the vicinity of the boundary layer. This OCM reaction was volume increase reaction and the
total moles of products were greater than those of reactants. The flow field in the catalyst bed maintained constant temperature and pressure.
The fluid density decreased gradually from 2.28 kg/m3 in the inlet of the catalyst bed to 2.22 kg/m3 in the outlet of the catalyst bed, while the
velocity increased from 0.108 m/s to 0.115 m/s.
Key words
oxidative coupling of methane; packed-bed reactor; computational fluid dynamics; numerical simulation

1. Introduction Mn/SiO2 catalyst. They confirm that WO4 , which is apt to


form transition state compounds with CH4 , works as the ac-
tive center of Na2 WO4 -Mn/SiO2 catalyst. Some other studies
The conversions of methane to chemicals and liquid fuels [8,9] also find that the presence of La and Li has positive effect
are achieved mainly by indirect methods and direct methods. on activity and selectivity of Na2 WO4 -Mn/SiO2 catalyst.
The direct methods are more potential since they can avoid the The OCM reaction is known to be highly exothermic,
syngas step [1]. Oxidative coupling of methane (OCM) is one so it is easy to form hot spots in a reactor. No less than
of the direct methods and regarded as a promising way for 100 ◦ C of temperature gap is measured by researchers dur-
methane conversion [2]. Various catalysts have been inves- ing OCM reaction over La2 O3 /CaO [10], Li/MgO [11], Mn-
tigated by researchers for OCM reaction [3]. Among these, Na2 WO4 /SiO2 and Mn-Na2WO4 /MgO [12] catalysts. The hot
Na2 WO4 -Mn/SiO2 particle, invented by Fang et al. in 1992 spot is a decisive factor in the performance of reactor, which
[4], is believed to be one of the most effective catalyst over may lead to temperature runaway, catalyst deactivation and
the past thirty years. Jiang et al. [5] reports that the existence even thermal cracking of the main products [13]. To inhibit
of Na2 WO4 can decrease the phase transition temperature of the formation of hot spots and improve the selectivity of re-
SiO2 from amorphous to cristobalite, which plays an impor- actant, the selection of reactor and the alteration of contact
tant role in the activity of Na2 WO4 -Mn/SiO2 catalyst. Ji et al. mode between reactant and catalysts function a lot apart from
[6,7] studies the effects of Na, W, Mn and other alkali met- the exploration of highly selective catalysts. Taniewski et al.
als on the structure and reaction performance of Na2 WO4 - [11] show that successive active catalyst layers can be involved

∗ Corresponding author. Tel: +86-10-64419619; Fax: +86-10-64419619; E-mail: jisf@mail.buct.edu.cn

Copyright©2015, Science Press and Dalian Institute of Chemical Physics. All rights reserved.
doi: 10.1016/S2095-4956(15)60280-7
24 Zhao Zhang et al./ Journal of Energy Chemistry Vol. 24 No. 1 2015

and a nearly unchanged overall selectivity is obtained during 2. Model and numerical method
the process of CH4 transformation over the bed by changing
the feed inlet locations. Talebizadeh et al. [13] study the 2.1. Geometric model and meshes
OCM in a two-zone fluidized-bed reactor (TZFBR). In their
work, diluted oxygen in argon is introduced into the bottom
A packed-bed reactor with GHSV of 9600 h−1 was se-
of the TZFBR, while methane is entered at higher positions
lected as the subject of study in this work. To diminish the er-
along the fluidized bed. They find that the TZFBR gives
ror, the established geometric model was completely the same
a higher C2 (both C2 H6 and C2 H4 ) selectivity than that ob-
with the experimental device [15] that is a circular tube with a
tained in the fluidized-bed reactor, since the introduced CH4
diameter of 8 mm and a length of 600 mm. The height of cat-
in medium height decrease the concentration of O2 , inhibit-
alyst bed was 10 mm and two sections with a length of 95 mm
ing the oxidation reaction occurred in gas phase. Pan et al.
were filled with quartz at both ends (see Figure 1a).
[14] study the OCM reaction in a dual-bed reactor compris-
ing of Na2 WO4 -Mn/SiO2 particle catalyst and Ce-Na2 WO4 -
Mn/SiO2 /cordierite monolithic catalyst. A maximum C2 yield
of 23.6% is obtained when the bed heights of particle catalyst
and monolithic catalyst are 10 mm and 50 mm, respectively. It
is noted that the raw gas must go through the particle catalyst
first. Wang et al. [15] make extensive researches about OCM
reaction in the above reactor. They confirm that similar re-
sults can be obtained if the monolithic catalyst is replaced by
Na3 PO4 -Mn/SiO2 /cordierite monolithic catalyst. They also
claim that the OCM reaction is restrained by low concentra-
tion of O2 in monolithic catalyst bed since the raw gases firstly
go through the particle catalyst layer. Therefore, an extra ad-
dition of O2 between the two beds is necessary [16]. After a
series of researches, they improve the C2 yield to 24.3% when
the flow of extra O2 is 15% of the inlet flow.
However, the experimental researches involve many dis-
advantages, such as high cost and long research cycle. By con-
trast, the simulation method can rapidly and accurately predict
the reaction characteristics of OCM reactor, and provide basic
data for the development of OCM technology [17−21]. The
classic models of OCM fixed bed include 1D models [17,18] Figure 1. Sketch (a), meshes (b) and geometric model (c) of packed-bed re-
actor. (1) Preheating section, (2) quartz section, (3) catalyst section, (4) quartz
and 2D models [21], which discard the influence of three- section, (5) adiabatic section
dimensional flow field on heat transfer and chemical reaction.
When the reactor is scaled up, the three-dimensional effect is
The meshes of cross-section, which was perpendicular to
enhanced, so the classical model of OCM reactor will bring
the symmetry axis of the reactor, were quadrilateral (see Fig-
large error, which is unable to instruct the amplification of
ure 1b), and all three dimensional meshes of the reactor were
OCM process. Computational fluid dynamics (CFD) method
hexahedral for reducing numerical viscosity. To ensure the
is a simulation method based on flow field analysis. It can
convergence of the numerical iteration and the independence
accurately predict the impact of three-dimensional flow field
of mesh, the aspect ratio of meshes in the bulk of the cata-
inside the reactor for heat transfer and chemical reaction. This
lyst bed was about 1 : 1 and other sections were 1 : 3, respec-
method has been successfully applied to predict the reaction
tively (see Figure 1c). In order to accurately predict the lam-
performance of the packed-bed reactor loaded with SnBaTiO3
inar boundary layer effects on flow and chemical reactions,
catalyst [19,20]. However, the application of CFD on the
the vicinity of reactor wall was portioned by inflation meshes
packed-bed reactor filled with Na2 WO4 -Mn/SiO2 catalyst has
of hexahedron (see Figure 1b). The number of meshes in the
not been seen in literature.
reactor model was about 410000.
In this work, we improved the Stansch reaction kinetics
model [22] using the existing experimental data [15], mak-
ing it suitable to simulate the packed-bed reactor filled with 2.2. Governing equations
Na2 WO4 -Mn/SiO2 particle catalysts, and established a three-
dimensional numerical model of OCM tubular packed-bed re- According to the operating conditions in literature [15],
actor. The Fluent solver was used to solve the Navier-Stokes the flow in OCM packed-bed reactor is laminar (Re = 5.7),
equations and species transport equations, and the reaction ki- which is suitable to be described by the Navier-Stocks equa-
netics model added by the user-defined function (UDF) of tions [23]. The set of governing equations of flow is given
Fluent software was adopted to estimate the actual reaction as:
performance of reactor. (1) Continuity equation
Journal of Energy Chemistry Vol. 24 No. 1 2015 25

∂ρ ~)=0 3.5 (1 − ε)
+ · (ρV (1) c= (6)
∂t Dp ε3
(2) Momentum equation
where, Dp is the mean particle diameter of catalyst, 50 mesh,
~
∂V and ε is bed porosity, 0.6 [25].
ρ + ρV ~ = · T~
~· V ~ +S
~m (2)
∂t In OCM packed-bed reactor, the presence of molecular
(3) Energy equation diffusion and convection diffusion between components make
it necessary to use species transport equation. The species
∂H ~· H= ~ ~ transport equation [26] for a component k in laminar flow
ρ + ρV · (T~ · V )+ · (k T ) + Sh (3)
∂t field can be expressed as Equation (7):
where, t is time, ρ is density, V~ is velocity vector, S
~m is mo- ∂Yk ~ · Yk = · (ρDk Yk ) + Sk
ρ + ρV (7)
mentum equation source term, such as the drag caused by cat- ∂t
alyst, T is temperature, Sh is energy equation source term, where, Yk is the mass fraction of component k, Dk is diffusion
such as the energy released by chemical reaction, H is total coefficients of component k in other components, and Sk is
energy, i.e., the sum of kinetic energy and internal energy, and source term, i.e., increase and decrease of the components
~
T~ is surface stress tensor. mass caused by chemical reaction.
The catalyst particles and quartz particles in the reactor
can produce resistance to the flow. The drag can be calcu- 2.3. Kinetic model
lated by porous medium model which is added to the momen-
tum equations in the form of source terms. For the isotropic The OCM reaction system consists of two parts, the cat-
porous medium, source terms of the momentum equations can alytic heterogeneous reaction and the homogenous reaction
be simplified as: without catalysis [27]. Then, two factors must be considered

µ 1 ~
 if we plan to select a suitable reaction kinetics model for the
Sm,i = − u i + c ρ V s u i (4) OCM simulation using CFD. On the one hand, the blocking
α 2

effect of CO2 on the catalyst cannot be ignored. Su et al. [28]
where, α is the permeability and c is the inertial resistance claim that the conversion of methane and the yields of prod-
factor. ucts in OCM reaction are limited, which is contrary to the
For the packed-bed of catalytic particles, α and c can be conclusion that CH4 will be soon totally converted with the
obtained from empirical equations [24], which are expressed reaction time without considering the blocking effect of CO2
as: on the catalyst. On the other hand, no reaction equation based
Dp2 ε3 on the active center of catalyst surface should be included in
α= (5)
150 (1 − ε)2 the model to reduce the calculation time of CFD.

Table 1. Reaction equations and reaction rate equations of Stansch kinetic model

Entry Reaction equations Reaction rate equations


m1 n1
k0,1 e−Ea,1 /RT PCH P
4 O2
1 CH4 + 2O2 = CO2 + 2H2 O r1 =  n1
−D Had,CO2 /RT
1 + K1,CO2 e PCO2
 n2
m2
k0,2 e−Ea,2 /RT KO2 e−D Had,O2 /RT PO2 PCH 4
2 2CH4 + 0.5O2 = C2 H6 + H2 O r2 = h  n2 i2
−D Had,O2 /RT −D Had,CO2 /RT
1 + K O2 e P O2 + K2,CO2 e PCO2
m n
k0,3 e−Ea,3 /RT PCH34 PO23
3 CH4 + O2 = CO + H2 O + H2 r3 =  n3
1 + K3,CO2 e−D Had,CO2 /RT PCO2
m4 n4
k0,4 e−Ea,4 /RT PCO P O2
4 CO + 0.5O2 = CO2 r4 =  n4
−D Had,CO2 /RT
1 + K4,CO2 e PCO2
m n
k0,5 e−Ea,5 /RT PC2 H5 6 PO25
5 C2 H6 + 0.5O2 = C2 H4 + H2 O r5 =  n5
1 + K5,CO2 e−D Had,CO2 /RT PCO2
m n
k0,6 e−Ea,6 /RT PCO6 PO26
6 C2 H4 + 2O2 = 2CO + 2H2 O r6 =  n6
1 + K6,CO2 e−D Had,CO2 /RT PCO2
m
7 C2 H6 = C2 H4 + H2 r7 = k0,7 e−Ea,7 /RT PC2 H7 6
m n
8 C2 H4 + 2H2 O = 2CO + 4H2 r8 = k0,8 e−Ea,8 /RT PC2 H8 4 PH28O
m n
9 CO + H2 O = CO2 + H2 r9 = k0,9 e−Ea,9 /RT PCO9 PH29O
m n
10 CO2 + H2 = CO + H2 O r10 = k0,10 e−Ea,10 /RT PCO10
2
PH210
26 Zhao Zhang et al./ Journal of Energy Chemistry Vol. 24 No. 1 2015

The application of kinetic model on Na2 WO4 -Mn/SiO2 sch et al. [22] build a kinetic model containing 10 chemical
particle catalyst has been reported by many studies reactions and 8 components and apply it to the OCM reac-
[25,29−31]. Daneshpayeh et al. [25] and Shahri et al. tion catalyzed by La2 O3 /CaO particles. The blocking effect
[29] establish the kinetics models including ten-step reaction of CO2 is taken into account in 6 of these reactions (See Ta-
and seven-step reaction, respectively. In those reactions, the ble 1), and the simulation results meet the requirements very
blocking effects of O2 and C2 are partly considered, while the well. Nakisa et al. [20] modify the parameters of Stansch
effect of CO2 is ignored. Lee et al. [30] and Ahari et al. [31] model according to their experimental data, and successfully
establish a kinetic model based on the important elementary apply them to the simulation of a packed-bed reactor packed
reaction and multiple surface reactions are included. How- with Sn/BaTiO3 .
ever, all of them do not meet with our selected subjects. Stan-

Table 2. Kinetic parameters


k0,j Ea,j Kj,CO2 Had,CO2 K O2 Had,O2
Entry mj nj
(mol·g−1 ·s−1 ·Pa−(m+n) ) (kJ·mol−1 ) (Pa−1 ) (kJ·mol−1 ) (Pa−1 ) (kJ·mol−1 )
1 1.25e-7 48 2.50e-13 −175 0.24 0.76
2 4.64 182 4.15e-13 −186 1.15e-11 −124 1.00 0.40
3 1.73e-8 68 3.60e-14 −187 0.57 0.85
4 9.17e-6 104 2.00e-11 −168 1.00 0.55
5 1.06e-2 157 1.80e-10 −166 0.95 0.74
6 7.50e-5 166 1.60e-13 −211 1.00 0.96
7 6.00e+5 226 0.00 0
8 1.86e+5 300 0.97 0
9 9.50e-5 173 1.00 1.00
10 2.60e-2 220 1.00 1.00

In this work, we adopted the strategy of Nakisa, modified transport equations in the reactor. The variable gradients were
the parameters of Stansch model based on our experimental discretized by least squares method based on the cell cen-
data, and established a kinetics model suitable for the packed- ter. The discrete methods of momentum and species trans-
bed reactor over Na2 WO4 -Mn/SiO2 particle catalyst (See Ta- port equations were second-order upwind scheme. In order
ble 2). The modified kinetics model was added to Fluent soft- to avoid the divergence of the iteration, the under-relaxation
ware in the form of User Defined Function (UDF). The ther- factors of density and species mass fractions were set to 0.6.
mal chemical parameters of each component were mainly pro-
vided by the Fluent database, and the other parameters were
3. Results and discussion
obtained from NIST database. Since the reactor is regarded as
constant temperature, the viscosity coefficient, thermal con-
ductivity and diffusion coefficient between each component 3.1. Comparison of the simulated and experimental reactor
were assumed to be constant. performance

2.4. Operating conditions, boundary conditions and numeri- Based on the experimental data, we simulated the OCM
cal methods reaction phenomenon of a packed-bed reactor with GHSV of
9600 h−1 using the model and calculation conditions intro-
To compare with the experimental results, we calculated duced above. The conversion of CH4 at reactor outlet and the
the above model using the same operating conditions as the lit- selectivities to the main products and byproducts obtained by
erature. The volume flow rate of the reactant was 80 mL/min simulation and experiment were all listed in Table 3. As seen
under standard state with a CH4 /O2 ratio of 3 and the operat- from Table 3, the calculated and experimental values of CH4
ing temperature and pressure were 800 ◦ C and 1 atm, respec- conversion were 27.3% and 27.8%, respectively. The rela-
tively. tive error of the calculated value to experimental value was
In order to calculate the pressure drop in the flow field −1.9%. For the selectivities to products, C2 H4 , C2 H6 , CO
accurately, the type of reactor outlet was defined as pressure and CO2 , the calculated values were 42.5%, 22.9%, 11.1%
outlet, and then the type of inlet was designated as the mass and 23.5%, while the experimental values were 42.1%, 22.9%,
flow inlet corresponding to the type of outlet. The thermal 11.2% and 23.8%. The calculated values only made a relative
boundary conditions of the reactor walls should be constant error of 1.0%, 0, −0.9% and −1.2%. Based on these val-
temperature so as to maintain the constant temperature within ues, we could obtain the sum selectivity to C2 and the ratio
the reactor. of C2 H4 /C2 H6 , which were 65.4%, 1.9 and 65.0%, 1.8 by cal-
The SIMPLE Algorithm, which is one of the pressure- culation and experiment, respectively. Then, the calculated
based segregated algorithms of Fluent software, was used to values took a relative error of 0.6% and 3.4%. In the same
solve the equations of incompressible flow and the species way, we could get the yield of C2 according to CH4 conver-
Journal of Energy Chemistry Vol. 24 No. 1 2015 27

sion and C2 selectivity, which was 17.8% and 18.1% by cal- C2 H4 /C2 H6 reached 3.4%. This indicated that the calculated
culation and experiment, leading to a relative error of −1.5%. parameters values were agree well with those obtained by ex-
In conclusion, the relative error range of calculated values to periment, proving that the OCM reactor model established in
experimental values is ±2% except that the relative error of this paper was reliable.

Table 3. Performance parameters of the simulated and experimental reactors


CH4 conversion Selectivity (%) Selectivity to C2 Yield of C2
Reactors C2 H4 /C2 H6
(%) C2 H4 C2 H6 CO CO2 (%) (%)
Simulated 27.3 42.5 22.9 11.1 23.5 65.4 1.9 17.8
Experimental 27.8 42.1 22.9 11.2 23.8 65.0 1.8 18.1
Error (%) −1.9 1.0 0 −0.9 −1.2 0.6 3.4 −1.5

3.2. Species mass fraction to 0.035, 0.061, 0.032 and 0.107, respectively at the outlet of
catalyst bed. It can also be seen that the mass fraction con-
It is assumed that the reactor was laid vertically, and the tours of CH4 , O2 and all the products were denser in the en-
gas flow entered from the top of the reactor. In view of trance of catalyst bed than other regions. This phenomenon
the symmetry of catalytic bed, we intercepted a plane pass- can be obviously observed by the color transitions in Figures
ing through the symmetry axis of the catalytic bed, analyzed 2−4. Moreover, the decrease or increase rate of reactants and
the contours of parameters on this plane, such as species products concentration was also faster in the vicinity of inlet
mass fraction, and investigated the reaction performance of than other regions. For example, CH4 concentration decreased
the OCM packed-bed reactor. This plane was marked as P1 in sharply from 0.6 to 0.562 in the vicinity of catalyst inlet (the
this article. red region), while it only reduced to 0.4360 from 0.4734 in
Figure 2 shows that the mass fraction of CH4 and O2 de- the whole bottom half (the blue region). This indicated that
creased from 0.6 and 0.4 at the inlet of catalyst bed to 0.436 the OCM reaction carried out faster in the vicinity of catalyst
and 0.142 at the outlet of catalyst bed. Correspondingly, Fig- inlet than other regions, caused maybe by the higher concen-
ures 3 and 4 revealed that the mass fractions of C2 H6 , C2 H4 , tration of CH4 and O2 and the corresponding higher partial
CO and CO2 increased from zero at the inlet of catalyst bed pressure of the reactants in the inlet.

Figure 2. The contours of CH4 (a) and O2 (b) mass fractions on plane P1

Figure 3. The contours of C2 H6 (a) and C2 H4 (b) mass fractions on plane P1


28 Zhao Zhang et al./ Journal of Energy Chemistry Vol. 24 No. 1 2015

We can also see that the mass fraction contours of each reaction equation of OCM packed-bed reactor as follows:
component had a trend of bending upwards in the vicinity of
catalyst bed wall (Figures 2−4). This phenomenon told us CH4 + 0.782O2 = 0.114C2 H6 + 0.213C2H4 + 0.235CO2+
that the mass fraction of reactant in this region was lower than 0.111CO + 0.982H2O + 0.250H2
other regions in the parallel height, while the mass fractions (8)
of products (C2 H6 , C2 H4 , CO and CO2 ) had a converse trend. According to the stoichiometric coefficient in Equation
This is easy to be understood since the velocity is slower in (8), we get the expansion factor of the reaction based on CH4
the boundary layer, which provides longer residence time. to be 0.123, so this reaction belonged to volumetric increase
According to the concentrations of reactant and products reaction [32]. Liu et al. [33] find that the conversion of CH4
at the inlet and outlet of the catalyst bed, we can get the overall decreases with the increase of pressure and Chou et al. [34]

Figure 4. The contours of CO (a) and CO2 (b) mass fractions on plane P1

also have the same conclusion on this reaction. Their conclu- upper portion of the catalyst bed, and they were symmetric
sions imply that the reaction in OCM packed-bed reactor is a along the axis of reactor. In this work, the wall boundary con-
volumetric increase reaction. ditions of the packed-bed reactor model were constant tem-
perature, and it was appointed as 1073.0 K, so the extra heat
3.3. Temperature and heat f lux through the wall of catalytic from OCM reaction would be released through the wall.
bed Figure 5(b) shows that the heat flux of the wall sur-
rounding the catalyst bed varied from −17500 W/m2 to
Figure 5(a) shows that the catalyst bed can be considered −3500 W/m2 , and the minimum value appeared in the upper
as constant temperature since they ranged within 1073.0 K portion of the catalyst bed. This illustrated that the reaction
and 1073.5 K. The higher temperature zone distributed in the rate in upper portion of the catalyst bed was faster than other

Figure 5. The contours of temperature on plane P1 (a) and heat flux through the wall of catalytic bed (b)

parts, and it is agreed well with the results above (see Fig- 3.4. Pressure, density and velocity
ure 2, Figure 3, and Figure 4). In this region, more reaction
heat was released, so there must be more heat dispersed from The pressure in the reactor gradually decreased from the
the wall surface to environment so as to maintain a constant upper section to the lower section due to the friction drag
temperature of catalyst. produced by the catalyst particles and the wall surface (see
Journal of Energy Chemistry Vol. 24 No. 1 2015 29

Figure 6). In this work, the gauge pressure at the outlet of the the catalyst bed to 2.22 kg/m3 in the outlet of the catalyst bed.
reactor was set to zero, and then the gauge pressure of the cat- Since the flow field in the catalyst bed maintained constant
alyst bed would be 10.6−12.6 Pa, so the pressure drop of total temperature and pressure (see Figure 5a and Figure 6), the
catalyst bed was about 2 Pa. Therefore, the catalyst bed can density was determined only by chemical reaction. Moreover,
be considered as constant state. the volume of OCM reaction was variable, and the total moles
of products were greater than those of reactants. As the chem-
ical reaction decreased the density of fluid flow, the fluid ve-
locity would be increased gradually (see Figure 7b). The ve-
locity increased from 0.108 m/s in the inlet of the catalyst bed
to 0.115 m/s in the outlet of the catalyst bed.
From Figure 7, we can also see that a laminar bound-
ary layer existed near the catalyst bed. Compared with the
fluid velocity near the symmetry axis of the catalyst bed, it
was slower in the boundary layer, making the convective mass
transfer much weaker in this region, so the reactants and prod-
ucts cannot be inputted or outputted in time. Therefore, the
mass fraction of reactant in boundary layer was lower than
other regions in the parallel height, while the products (C2 H6 ,
C2 H4 , CO and CO2 ) had a converse trend. At the same time,
the mass fraction contours of each component bent upwards
in boundary layer (see Figures 2–4).
Figure 6. The contours of gauge pressure on plane P1

4. Conclusions

In this paper, an improved Stansch kinetic model was


established to simulate a packed-bed OCM reactor loaded
with Na2 WO4 -Mn/SiO2 particulate catalyst using the compu-
tational fluid dynamics (CFD) methods and Fluent software.
The volume velocity of reactant was 80 mL/min under stan-
dard state with a CH4 /O2 ratio of 3 and the operating tempera-
ture and pressure were 800 ◦ C and 1 atm, respectively. The es-
tablished OCM reactor model was proved reliable by compar-
ing the calculated and experimental values of CH4 conversion,
product selectivity and other parameters in the outlet of reac-
tor. The results showed that the reaction rate near the inlet of
reactor was faster than other parts. Therefore, the temperature
in this region was a little higher and more heat was released
through the reactor wall (the maximum heat flux in this work
was 17500 W/m2 ). In the laminar boundary layer, slower ve-
locity provided longer residence time, so the contours of each
component bent upwards. The volume of OCM reaction was
changing, and the total moles of products were greater than
those of reactants. The density of the fluid was gradually re-
duced, while the velocity was increased as the proceeding of
the OCM reaction.

Nomenclatures
c the inertial resistance factor (m−1 )
Dp the mean particle diameter of catalyst (m)
Dk diffusion coefficients of species k in other components
(m2 ·s−1 )
H total energy (J·kg−1 )
Figure 7. The contours of pressure (a) and velocity magnitude and vectors (b) Sh energy equation source term (kg·m−2 ·s−2 )
~m momentum equation source term (J·m−3 ·s−1 )
S
Figure 7(a) reveals that with the reaction of CH4 , the fluid Sk source term of species transport equation of species k
density decreased gradually from 2.28 kg/m3 in the inlet of (kg·m−3 ·s−1 )
30 Zhao Zhang et al./ Journal of Energy Chemistry Vol. 24 No. 1 2015

t time (s) [10] Schweer D, Mleczko L, Baerns M. Catal Today, 1994, 21(2-3):
T temperature (K) 357
~ [11] Taniewski M, Lachowicz A, Skutil K. Chem Eng Sci, 1997,
T~ surface stress tensor (Pa) 52(6): 935
u a component of velocity vector (m·s−1 ) [12] Pak S, Lunsford J H. Appl Catal A, 1998, 168(1): 131
~
V velocity vector (m·s−1 ) [13] Talebizadeh A, Mortazavi Y, Khodadadi A A. Fuel Process
Technol, 2009, 90(10): 1319
~s
V superficial velocity (m·s−1 ) [14] Pan D, Ji S F, Wang W H, Li C Y. J Nat Gas Chem, 2010, 19(6):
Yk mass fraction of species k 600
[15] Wang W H, Ji S F, Pan D, Li C Y. Fuel Process Technol, 2011,
92(3): 541
Greek letters
[16] Ji S F, Wang W H. Int J Chem Reactor Eng, 2012, 10(1): 11
α permeability (m−2 ) [17] Hoebink J H B J, Couwenberg P M, Marin G B. Chem Eng Sci,
ε bed porosity 1994, 49(24B): 5453
µ molecular viscosity (Pa·s) [18] Tye C T, Mohamed A R, Bhatia S. Chem Eng J, 2002, 87(1): 49
[19] Yaghobi N, Ghoreishy M H R. J Nat Gas Chem, 2008, 17(1): 8
ρ Density (kg·m−3·s−1 )
[20] Nakisa Y, Reza G M H. J Nat Gas Chem, 2009, 18(1): 39
[21] Holst N, Jašo S, Godini H R, Glöser S, Arellano-Garcia H,
Subscript Wozny G, Steinbach J. Chem Eng Technol, 2012, 35(2): 294
i identifier for the ith component of coordinate [22] Stansch Z, Mleczko L, Baerns M. Ind Eng Chem Res, 1997,
36(7): 2568
k species k [23] Batchelor G K. An Introduction to Fluid Dynamics. 2nd Ed.
r reaction Cambridge: Cambridge University Press, 2000. 131
[24] Ergun S. Chem Eng Prog, 1952, 48(2): 89
[25] Daneshpayeh M, Khodadadi A, Mostoufi N, Mortazavi Y,
References Sotudeh-Gharebagh R, Talebizadeh A. Fuel Process Technol,
2009, 90(3): 403
[1] Lunsford J H. Catal Today, 2000, 63(2-4): 165 [26] Poinsot T, Veynante D. Theoretical and Numerical Combustion.
[2] Gesser H D, Hunter N R. Catal Today, 1998, 42(3): 183 2nd Ed. Philadelphia: Edwards, 2005. 13
[3] Khammona K, Assabumrungrat S, Wiyaratn W. J Eng Appl Sci, [27] Pak S, Qiu P, Lunsford J H. J Catal, 1998, 179(1): 222
2012, 7(7): 447 [28] Su Y S, Ying J Y, Green W H. J Catal, 2003, 218(2): 321
[4] Fang X P, Li S B, Lin J Z, Chu Y L. J Mol Catal (Fenzi Cuihua), [29] Shahri S M K, Alavi S M. J Nat Gas Chem, 2009, 18(1): 25
1992, 6(6): 427 [30] Lee M R, Park M J, Jeon W, Choi J W, Suh Y W, Suh D J. Fuel
[5] Jiang Z C, He J H. J Mol Catal (Fenzi Cuihua), 1992, 6(6): 477 Process Technol, 2012, 96: 175
[6] Ji S F, Xiao T C, Li S B, Xu C Z, Hou R L, Coleman K S, Green [31] Ahari J S, Zarrinpashne S, Sadeghi M T. Fuel Process Technol,
M L H. Appl Catal A, 2002, 225(1-2): 271 2013, 115: 79
[7] Ji S F, Xiao T C, Li S B, Chou L J, Zhang B, Xu C Z, Hou R L, [32] Fogler H S. Elements of Chemical Reaction Engineering. 4th
York A P E, Green M L H. J Catal, 2003, 220(1): 47 Ed. London: Prentice Hall, 2006. 108
[8] Malekzadeh A, Khodadadi A, Dalai A K, Abedini M. J Nat Gas [33] Liu Y, Liu X X, Xue J Z, Hou R L, Zhang B, Yu C C, Shen S K.
Chem, 2007, 16(2): 121 Appl Catal A, 1998, 168(1): 139
[9] Wu J J, Zhang H L, Qin S, Hu C W. Appl Catal A, 2007, 323: [34] Chou L J, Cai Y C, Zhang B, Niu J Z, Ji S F, Li S B. J Nat Gas
126 Chem, 2002, 11(3): 131

Das könnte Ihnen auch gefallen