J Ijhydene 2014 04 010

A simulation study on methanol steam reforming
in the silica membrane reactor for hydrogen

production
K. Ghasemzadeh
a,b
, P. Morrone
c
, A.A. Babalou
a
, A. Basile
c,
*
a
Department of Chemical Engineering, Nanostructure Material Research Center, Sahand University of Technology,
Tabriz 51335-1996, Iran
b
Department of Chemical Engineering, Urmia University of Technology, Urmia 57166-93187, Iran
c
ITM-CNR, c/o University of Calabria, Via Pietro Bucci, Cubo 17/C, 87036 Rende (CS), Italy
a r t i c l e i n f o
Article history:
Received 31 July 2013
Received in revised form
31 March 2014
Accepted 2 April 2014
Available online xxx
Keywords:
Silica membrane reactor
Simulation study
Hydrogen production
Methanol steam reforming
a b s t r a c t
The main aim of this study is the presentation of a quantitative operating analysis of silica
membrane reactor for carrying out the methanol steam reforming reaction to produce
hydrogen. For this, a 1-dimensional isothermal model is developed and the simulation
study is carried out for evaluating the catalyst loading, sweep gas ow rate and silica
membrane hydrogen selectivity effects on silica MR performance in terms of methanol
conversion and hydrogen recovery. These parameters effects are investigated at different
reaction temperatures and pressures. The model validation results show good agreement
with literature experimental data. As a main consequence, simulation results indicate that
the catalyst loading and sweep gas ow rate have positive effects on silica MR perfor-
mance. However, this result is devoted to lower catalyst contents (<3 g) and sweep factor
values (SF < 6). Moreover, the increase of silica membrane hydrogen selectivity increases
the hydrogen recovery and methanol conversion. However, the improvement of mem-
brane hydrogen selectivity in lower hydrogen permeance ranges cannot be more useful on
silica MR performance in MSR reaction. According to the reaction pressure effect on silica
MR performance, methanol conversion trends are various at different orders of hydrogen
permeance.
Copyright 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
reserved.
Introduction
With increasing concerns over gradual depletion of non-
renewable fossil fuels and global warming, considerable
attention has been paid to hydrogen for fuel cells in various
applications owing to their zero emission and high efciency
[1]. One of the major challenges associated with the potential
implementation of H
2
economy is the issue of hydrogen pro-
duction. It has been produced mainly via the steam reforming
of methane, methanol and ethanol where the conversion of
feed is limited due to the thermodynamic equilibrium. Unlike
hydrocarbon fuels, methanol is a sulphur free fuel and this
avoids additional sulphur removal steps in the fuel
* Corresponding author. Tel.: 39 (0)984492013; fax: 39 (0)984401071.
E-mail address: a.basile@itm.cnr.it (A. Basile).
Available online at www.sciencedirect.com
ScienceDirect
j ournal homepage: www. el sevi er. com/ l ocat e/ he
i nt e r na t i o na l j o ur na l of hyd r o g e n e ne r g y x x x ( 2 0 1 4 ) 1 e1 0
Please cite this article in press as: Ghasemzadeh K, et al., A simulation study on methanol steam reforming in the silica
membrane reactor for hydrogen production, International Journal of Hydrogen Energy (2014), http://dx.doi.org/10.1016/
j.ijhydene.2014.04.010
http://dx.doi.org/10.1016/j.ijhydene.2014.04.010
0360-3199/Copyright 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
processing. In addition, methanol can be reformed at rela-
tively lower temperatures around 573 K, and this makes the
fuel processing relatively simple. The traditional methanol
steam reforming (MSR) process consists of different process
steps such as feed gas preheating, pretreatment (for example
hydrodesulphurisation), primary and secondary reformers
(often multi-tubular xed-bed reactors), high and low tem-
perature shift converters, CO
2
removal and methanation units
[2,3].
In recent years, membrane reactors (MRs) as a new
approach combine reaction and separation and have attracted
attention because of their compactness and possible cost-
savings over traditional reactors (TRs). To avoid higher pro-
cess costs, Pd-based MRs for pure hydrogen production have
been proposed in literature for different reaction systems
such as MSR reaction [4e16]. Nevertheless, there are not many
studies about other kinds of membrane performances in MRs
for MSR reaction. Regarding the membrane kinds to be housed
in an MR, both MR cost and performance should be consid-
ered. Owing to the high cost of Pd-based MRs, a cheaper so-
lution is strongly needed. Although these membranes are only
permeable to hydrogen, their applications have limited due to
lower permeability compared to porous inorganic membranes
[17,18]. As a possible alternative, silica membranes can be
considered because of their lower cost and higher perme-
abilities with respect to Pd-based membranes. On the other
hand, as a drawback, they present lower hydrogen selectiv-
ities with respect to all the other gases.
Indeed, according to our best knowledge, Lee et al. [19,20]
have presented several experimental studies comparing a TR
with silica MR. They concluded that higher methanol con-
versioncan be realized in MR with respect to TR. Nevertheless,
a low hydrogen recovery was achieved. Therefore, in order to
enhance the hydrogen recovery, the authors have carried out
the water gas shift (WGS) reaction in the permeate side
simultaneously with MSR reaction in the retentate side,
improving the silica MR performance in terms of both meth-
anol conversion and hydrogen recovery. These promising re-
sults have justied the need for a detailed study of potential
advantages achievable using the silica MR. To evaluate the
silica MR performance as an alternative in MSR reaction, a
comprehensive analysis seems necessary.
In order to avoid high experimental costs and to develop a
better understanding of the effects of several parameters for
the study of MR, the numerical simulation can be useful.
Nevertheless, few works have been presented from modelling
point of viewfor silica MR performance or other inorganic MR.
Sa et al. [21,22] have compared the performances of palladium
MR with a carbon molecular sieve (CMS) MR. They developed a
1-dimensional model and validated it with literature experi-
mental data. Ghasemzadeh et al. [23] evaluated silica and
palladium MR performances for hydrogen production via MSR
reaction by 1-dimensional model. In this case, a 1-
dimensional isothermal numerical model was developed
and its validation was realized by comparing the theoretical
results with experimental data. Also, the effect of some of the
important operating parameters (reaction temperature, pres-
sure and GHSV) on the performance of both MRs was studied
in terms of methanol conversion and hydrogen recovery.
Furthermore, Ghasemzadeh et al. [24] was investigated
hydrogen production in silica MR via MSR reaction using
HAZOP study and modelling analysis. In fact, the main aim of
the mentioned work was the presentation of both qualitative
safety and quantitative operating analyses of silica MR for
carrying out the MSR reaction to produce hydrogen.
In the present study, regarding the leakage of numerical
studies about silica MR performance in MSR reaction, a 1-
dimensional isothermal model was developed for evaluating
sweep gas ow rate and catalyst loading effects. Moreover, as a
rst approach, asimulationstudyis carriedout for investigating
silica membranes hydrogen selectivity effect on silica MR per-
formance during the MSR reaction. A set of simulation results
shows some key points about the silica MR performance in
terms of methanol conversion and hydrogen recovery.
Model development
The following assumptions were made to develop a model for
silica MR.
1. Steady state, isothermal operation.
2. Uniform ow in shell and tube sides, so no radial distri-
bution of components exists.
3. Permeation through the membrane is proportional to the
partial pressure difference in shell and tube sides.
4. The catalysts and packing materials that are used to dilute
the catalyst bed have the same size and shape.
5. Ideal gas behaviour.
6. Steam effect on silica membrane is negligible.
7. Flux equations of permeating components for silica
membrane were derived fromsingle gas permeation test at
different temperatures.
8. Pseudo-homogeneous model is assumed in the catalyst
bed.
The model takes into account all the reactions (Eqs. (1)e(3))
for silica MR, along with the methanol steam reforming re-
action [25,26],
CH
3
OHH
2
O43H
2
CO
2
DH 49:4 kJ mol
1
(1)
the methanol decomposition reaction
CH
3
OH42H
2
CO DH 90:5 kJ mol
1
(2)
and the water gas shift reaction
COH
2
O4H
2
CO
2
DH 41:1 kJ mol
1
(3)
take place in the MR.
Fig. 1 e Schematic of the silica MR model.
i nt e r na t i o na l j our na l o f hy d r og e n e ne r gy x x x ( 2 0 1 4 ) 1 e1 0 2
Fig. 1 shows the schematic of the silica MR model, which
has a simple shell and tube geometry. The co-current ow
mode for silica MR was considered. The commercial Cu/ZnO/
Al
2
O
3
catalyst is lled in retentate side. The model has been
developed taking into account the mass balance equation in
retentate and permeate sides for each species, the pressure
drops in retentate side through the Ergun equation, the rate of
each reaction for retentate side and transport equation for the
permeating species through the membrane.
In the following, the governing equations of silica MR are
presented in the separate subsections namely, retentate side,
permeate side and membrane area.
2.1. Retentate side (reaction side)
Mass balance equation
For each species the following mass balance equation was
used:
dF
i
dz

W
V
X
n
j1
n
ij
r
i
J
i
pD
A
(4)
where F
i
is the molar ow rate, z the axial coordinate, W the
weight of catalyst, V the volume of catalyst, n
ij
the stoichi-
ometry coefcients, r
i
the rate of reaction, J
i
the ux of species,
D the membrane diameter and A the surface area of retentate
side.
Pressure drop equation
Presence of the catalyst in retentate side can increase
pressure drop. Therefore, the Ergun equation was used [27]:
dp
dz
150
mU
d
2
p
1
2
7
4
r
gas
V
2
d
p
1
(5)
where p is the pressure, the catalyst bed porosity, d
p
the
catalyst diameter, U the gas velocity, m the gas viscosity and
r
gas
the gas density.
Reaction rate expressions
The reaction rate expressions and parameters were indi-
cated in our previous studies [23,24] in details. In general,
corresponding to three main reactions that take place during
MSR reaction; three rate reaction correlations were used.
Methanol steam reforming (Eq. (1))
Methanol decomposition reaction (Eq. (2))
Water gas shift reaction (Eq. (3))
where k
j
and K
eq
j
are the reaction rate and equilibrium con-
stants for reaction j, respectively; K
i
the adsorption coefcient
for surface species i, C
T
S1
and C
T
S2
the total catalyst surface
concentration of sites 1 and 2, respectively, and C
T
S1a
and C
T
S2a
the total catalyst surface concentration of sites 1a and 2a,
respectively. All of the kinetic parameters were used as pre-
sented in our previous work [24].
Boundary conditions
In Table 1, the boundary conditions for retentate side are
reported.
r
SR

k
SR
K
CH
3
O
1
p
CH
3
OH
.
p
1=2
H
2
p
3
H
2
p
CO
2
.
K
eq
SR
p
CH
3
OH
p
H
2
O
C
T
S1
C
T
S1a
1 K
CH
3
O
1
p
CH
3
OH
.
p
1=2
H
2
HCOO
1
p
CO
2
p
1=2
H
2
K
OH
1
p
H
2
O
.
p
1=2
H
2
1 K
1=2
H
1a
p
1=2
H
2
(6)
r
MD

k
MD
K
CH
3
O
2
p
CH
3
OH
.
p
1=2
H
2
p
2
H
2
p
CO
.
K
eq
MD
p
CH
3
OH
C
T
S2
C
T
S2a
1 K
CH
3
O
2
p
CH
3
OH
.
p
1=2
H
2
OH
2
p
H
2
O
.
p
1=2
H
2
1 K
1=2
H
2a
p
1=2
H
2
(7)
r
WGS

k
WGS
K
OH
1
p
CO
p
H
2
O
.
p
1=2
H
2
p
H
2
p
CO
2
.
K
eq
WGS
p
CO
p
H
2
O
C
T
2
S1
1 K
CH
3
O
1
p
CH
3
OH
.
p
1=2
H
2
HCOO
1
p
CO
2
p
1=2
H
2
K
OH
1
p
H
2
O
.
p
1=2
H
2
2
(8)
i nt e r na t i o na l j o ur na l of hyd r o g e n e ne r g y x x x ( 2 0 1 4 ) 1 e1 0 3
2.2. Permeate side
For this domain, only mass balance equation was used.
Mass balance equation
For each species the mass balance was considered as
follow:
dF
i
dz
J
i
pD
A
(9)
Boundary conditions
In Table 2, the boundary conditions for permeate side are
presented.
2.3. Membrane
Regarding the silica membrane, all components permeate
through the membrane. Therefore, in general, permeating
ux through the silica membrane for each component can be
dened by the following equation:
J
i
Pe
p
i;retentate
p
i;permeate
(10)
where Pe
i
Pe
i
=d is the permeance for species i, p
i,retentate
and p
i,permeate
the partial pressures for species i in the reten-
tate and permeate sides, respectively. Pe
i
can be calculated by
using the following equation:
Pe
i
Pe
i;0
exp
E
a;i
RT
(11)
where Pe
0;i
is a permeance coefcient and E
a,i
the transport
activation energy for each species. Values of Pe
0;i
and E
a,i
for
each species i, shown in Table 3, were used from Costa et al.
[28] and Lee et al. [19].
In Table 4, the parameter values used in the modelling
analysis are presented.
2.4. Post-processing correlations
The following denitions were used for describing the silica
MR performance:
Methanol conversion%
CH
3
OH
in
CH
3
OH
out
CH
3
OH
in
100 (12)
where CH
3
OH
in
and CH
3
OH
out
are the inlet and outlet meth-
anol ow rates in the reactor, respectively.
Hydrogen recovery%
H
2permeate
H
2permeate
H
2retentate
100 (13)
where H
2-permeate
is the hydrogen molar ow rate that per-
meates through the membrane and H
2-retentate
the hydrogen
molar ow rate in retentate side.
Also, the sweep factor correlation is dened by following
equation;
Sweep factorSF
N
2
molarflow rate in the permeate side
CH
3
OH
in
molarflow rate
(14)
Procedure of numerical solution
All of the model equations were imported to C builder 2006
environment and a numerical package was developed. A
predictorecorrector method [29] was used to solve the gov-
erning equations. Both prediction and correction steps are
explicit methods. It should be noted that, a variable dis-
cretization of the domain has been used. The number of in-
tervals was considered 10,000.
Table 2 e Boundary conditions used in permeate side for
the silica MR model.
Positions Parameters Values
Inlet F
7
(sweep gas molar ow rate) It was varied
Inlet F
8
(hydrogen molar ow rate) 110
30
F
7
Inlet F
9
(carbon monoxide molar ow
rate)
110
30
F
7
Inlet F
10
(carbon dioxide molar ow rate) 110
30
F
7
Inlet F
11
(methanol molar ow rate) 110
30
F
7
Inlet F
12
(water molar ow rate) 110
30
F
7
Table 3 e List of parameters used in this model for silica
membrane [19,28].
Gas species, i Pe
0,i
* (mol m
2
s
1
Pa
1
) E
a,i
(kJ mol
1
)
H
2
2.72 10
6
8.78
CO
2
3.29 10
8
1.72
CO 7.06 10
8
5.33
N
2
4.62 10
8
3.55
H
2
O 4.11 10
8
3.49
CH
3
OH 9.52 10
9
2.52
Table 4 eList of parameter values used in the simulation.
Parameters Values
Silica membrane layer (m) 5 10
7
Silica membrane outer diameter (m) 1 10
2
Silica membrane length (m) 5 10
2
Catalyst particle size (m) 5 10
4
Catalyst particle surface area (m
2
/g) 50
Catalyst bed density (kg/m
3
) 1300
Catalyst bed porosity 0.6
Mass of catalyst (g) 3
R (m
3
atm/mol K) 8.314
Table 1 e Boundary conditions used in retentate side for
Positions Parameters values
Inlet F
1
(methanol molar ow rate) 5.36 10
6
(mol/s)
Inlet F
2
(water molar ow rate) 1.61 10
5
(mol/s)
Inlet F
3
(hydrogen molar ow rate) 1 10
30
F
1
Inlet F
4
(carbon monoxide molar
ow rate)
0
Inlet F
5
(carbon dioxide molar
ow rate)
0
Inlet F
6
(standard nitrogen molar
ow rate)
1.875 10
5
(mol/s)
One of the most important advantages of explicit methods
with respect to implicit methods is that in the rst method the
variable at the actual position z is completely calculated by
considering the previous position z-1 (already computed),
while in the implicit methods the variable at the actual posi-
tion is solved by using the position z and z-1. Therefore, in the
explicit method the equations are solved sequentially starting
from the boundary condition to the end of the MR, while for
the implicit method the equations need to be solved simul-
taneously for all the domain steps, resulting in a system of
non-linear equations. The disadvantage of the explicit
method is the intrinsic instability that requires very low step
for the discretization of the domain. Nevertheless, this
drawback can be avoided by using good initial parameters for
modelling analysis. The predictorecorrector method is a good
choice for modelling the systems such as silica MR.
Results and discussion
Model validation
In order to validate the mathematical model, simulation re-
sults for silica MR model have been compared with experi-
mental data from Lee et al. [19]. It should be noted that the
model validation was discussed in our previous work [24], in
detail. Fig. 2 shows methanol conversion and hydrogen re-
covery versus the reaction temperature.
The validation of model results at reaction pressure of
1.42 bar, feed ow rate of 0.03 mL/min and feed molar ratio
(H
2
O/CH
3
OH) of 3 is realized. It is quite evident that there is
good agreement between experimental data and the results of
Operating conditions and analysis
According to our knowledge, some critical factors namely,
catalyst loading, sweep factor and membrane hydrogen
selectivity effects on silica MR performance during MSR re-
action have not been analysed in literature. Therefore, a
simulationstudy has been carried out for comparing the effect
of sweep gas ow rate and catalyst loading at different
reaction temperatures on silica MR performance, in terms of
methanol conversion and hydrogen recovery. Table 5 reports
the operating conditions used for carrying out MSR reaction in
silica MR. In particular, this section of simulation analysis can
be divided into two separate parts, in that one part sweep
factor was changed for different temperatures, while in other
part the catalyst loading was studied at different reaction
temperatures.
Furthermore, in this theoretical work, H
2
selectivity effect
at different reaction pressures and hydrogen permeance on
silica MR performance was investigated, in terms of methanol
conversion and hydrogen recovery. Table 6 exhibits the
operating conditions used for this case study. In particular,
the simulation analysis of this parameter can be divided into
three parts, in that each part H
2
selectivity of silica membrane,
for a constant order of hydrogen permeance, was changed at
different reaction pressures.
Evaluation of the sweep factor effects
Having in mind to maximize the hydrogen streamcollected in
the permeate side, the sweep factor (SF) inuence on the silica
MR performance was analysed. The increase of SF reduces the
hydrogen partial pressure in the permeate side, favouring the
hydrogen permeation driving force and, consequently, the
methanol conversion and hydrogen recovery enhancement.
As illustrated in Fig. 3, methanol conversion is rapidly
increased by increasing the sweep factor at lower sweep factor
ranges (SF < 6) while, at higher ranges (SF > 6), enhancement
of methanol conversion is slight.
Fig. 2 e Comparison of modeling results and experimental
data [24].
Table 5 e The investigated conditions for analysis the
catalyst loading and sweep factor effects on silica MR
performance.
Operating
parameters
Conditions for catalyst
loading analysis
Conditions for
sweep factor
analysis
Temperature (K) 493e553 493e553
Pressure (bar) 1.5 1.5
Feed ow rate
(mL/min)
0.03 0.03
H
2
O/CH
3
OH () 3 3
Catalyst loading
(g)
0.5e10 3
Sweep factor () 6 0e60
Bold value indicates the range of parameter variations.
Table 6 e The investigated conditions for analysis the
silica membrane selectivity and hydrogen permeance
effect on silica MR performance.
Operating
parameters
Conditions
for case 1
Conditions
for case 2
Conditions
for case 3
Temperature (K) 523 523 523
Pressure (bar) 1.5e10 1.5e10 1.5e10
Order of H
2
/x*
selectivity
10e10000 10e10000 10e10000
Order of hydrogen
permeance
(mol m
2
s
1
Pa
1
)
10
8
10
7
10
6
Bold value indicates the range of parameter variations.
Fig. 4 shows that hydrogen recovery is increased by
enhancement of sweep factor. The same with methanol
conversion results, increasing trend of hydrogen recovery at
lower sweep factor (SF < 6) is more noticeable with respect to
higher values (SF > 6).
In particular, at 553 K, SF 6 and 1.5 bar, 94.68% methanol
conversion and 32.6% hydrogen recovery were achieved with
respect to SF 0, where 90.39% methanol conversion and
4.03% hydrogen recovery were obtained. Therefore, the
hydrogen recovery is more sensible to SF values over meth-
anol conversion.
Evaluation of the catalyst loading effects
A further parameter that can strongly affect the methanol
conversion and also the collected hydrogen stream in the
permeate side is the catalyst loading. It is obvious that an
enhancement of catalyst loading can shift the MSR reaction
towards the products by decreasing the reaction activation
energy.
As depicted in Fig. 5, methanol conversion is increased by
increasing the catalyst loading. However, this result is devoted
to lower catalyst contents.
As shown in Fig. 6, hydrogen recovery is improved by
increasing the catalyst contents in the retentate side at three
different temperatures. As a consequence, it is found that
catalyst loading more than 3 g does not show noticeable ef-
fects on the hydrogen recovery and methanol conversion.
In particular, for a value of catalyst loading about 3 g, a
94.68% methanol conversion and a 32.6% hydrogen recovery
were achieved, at 553 K temperature, SF of 6 and at pressure of
1.5 bar; on the other side, with a higher value of loading, 10 g, a
94.91% methanol conversion and 33.12% hydrogen recovery
were reached in the same conditions.
Evaluation of membrane H
2
selectivity effects at different
hydrogen permeance and reaction pressures
Effective investigation of membrane hydrogen selectivity on
the silica MR performance was the main reasonto analyse this
factor effects on silica MR performance in terms of methanol
Fig. 3 e Methanol conversion versus sweep factor at three
different temperatures, 1.5 bar, H
2
O/CH
3
OH [ 3 and
0.03 ml/min.
Fig. 4 e Hydrogen recovery versus sweep factor at three
2
O/CH
3
OH [ 3 and
0.03 ml/min.
Fig. 5 e Methanol conversion versus catalyst loading at
three different temperatures, 1.5 bar, H
2
O/CH
3
OH [ 3 and
0.03 ml/min.
Fig. 6 e Hydrogen recovery versus catalyst loading at three
2
O/CH
3
OH [ 3 and
0.03 ml/min.
conversion and hydrogen recovery. As reported in Table 6, by
considering the hydrogen permeance in order of 10
8
, 10
7
and 10
6
(mol m
2
s
1
Pa
1
), membrane hydrogen selectivity
was changed from 10 to 10,000 times at different reaction
pressures. Therefore, in the following, the results are pre-
sented in three separate parts.
Low hydrogen permeance (case 1). As illustrated in Figs. 7 and
8, two kinds of results can be discussed. Fig. 7 shows that
methanol conversion is increased by increasing of hydrogen
selectivity. It is probably related to the higher hydrogen per-
meance with respect to the lower methanol and steam per-
meance. Moreover, as shown in Fig. 8, the enhancement of
silica membrane selectivity increases the hydrogen separa-
tion factor, favouring the hydrogen permeation driving force
and, consequently, the hydrogen recovery increment. How-
ever, the results obtained for silica membranes with various
hydrogen selectivities were not much different.
As a further consequence, methanol conversion versus
reaction pressure shows an absolute decrease (see Fig. 7).
As a general consideration, an increase of reaction pres-
sure in the silica MR can result in competitive effects on
methanol conversion:
A negative effect owing to the thermodynamic character-
istics of the MSR reaction, which proceeds with an increase
of the mole number, so the reaction is not favoured by
increasing the reaction pressure.
A further negative effect caused by reactants loss due to
the permeation through silica membrane.
A positive effect due to the selective removal of the reac-
tion products by means of the permeation through the
silica membrane. As a result, the MSR reaction is shifted
towards further products formation and methanol con-
version is improved (shift effect).
As a consequence, in this case, shift effect cannot over-
come thermodynamic limitation.
According to simulation results, it should be noted that the
hydrogen selectivity effect on methanol conversion and
hydrogen recovery at higher reaction pressure is more
noticeable with respect to lower reaction pressure (see Fig. 8).
Moderate hydrogen permeance (case 2). In this case, as pre-
sented in Fig. 9, it is realized that reaction pressure has a dual
effect on methanol conversion. It means that in the lower
pressures (less than 3 bar) shift effect cannot overcome ther-
modynamic limitation, while it appears in the higher
pressures.
The hydrogen selectivity effect on methanol conversion in
this case is more noticeable over the case 1. However,
increasing the hydrogen selectivity from 10 to 100 is much
effective compared to the higher selectivity variations.
Fig. 10 shows that the hydrogen recovery is increased by
increasing the reaction pressure and hydrogen selectivity of
silica membrane. Moreover, enhancement of hydrogen re-
covery versus hydrogen selectivity at higher reaction pressure
ranges is more noticeable with respect to lower values, which
can be related to higher hydrogen permeance driving force at
higher reaction pressures.
Fig. 7 e Methanol conversion versus reaction pressure at
different membrane H
2
selectivity, 523 K, H
2
O/CH
3
OH [ 3,
0.03 ml/min and 10
L8
order of hydrogen permeance.
Fig. 8 e Hydrogen recovery versus reaction pressure at
2
2
O/CH
3
OH [ 3,
0.03 ml/min and 10
L8
2
2
O/CH
3
OH [ 3,
0.03 ml/min and 10
L7
High hydrogen permeance (case 3). According to the model
results, as illustrated in Figs. 11 and 12, it is found that
methanol conversion and hydrogen recovery are increased by
increasing the reaction pressure.
In this case, the improvement of silica MR performance by
increasing the hydrogen selectivity is more noticeable
compared to other cases. However, increasing the hydrogen
selectivity more than 100 seems effectless.
As a consequence, in this hydrogen permeance value, shift
effect plays a key role inenhancement of methanol conversion.
By comparing the all cases, the optimum order of hydrogen
permeance and selectivity for the silica MR application in MSR
reaction can be reported as 10
6
and 100, respectively.
Conclusion
This simulation study was focused on the investigation of the
methanol conversion and hydrogen recovery reachable in the
silica MR by varying sweep factor and catalyst loading at
different reaction temperatures. Also, H
2
selectivity effect of
silica membrane on the MR performance was analysed at
three different orders of hydrogen permeance and reaction
pressures. It can be concluded that 3 g of catalyst and SF 6
are optimum values in the silica MR application for MSR re-
action. Moreover, as another consequence, higher selectivity
of silica membrane shows more effect on silica MR perfor-
mance in higher ranges of hydrogen permeance. However,
effect of hydrogen selectivity more than 100 in terms of
hydrogen recovery and methanol conversion in all cases was
negligible. Finally, as a general result, the improvement of
silica selectivity with lower hydrogen permeance is not useful
on the silica MR performance for MSR reaction in terms of
methanol conversion and hydrogen recovery, however the
silica membrane selectivity has more effect on permeate
stream purity. Therefore, the hydrogen selectivity and
hydrogen permeance should be balanced for the silica MR
application in MSR reaction.
Acronyms list
GHSV gas hourly space velocity
MD methanol decomposition
MR membrane reactor
MSR methanol steam reforming
MR membrane reactor
SF sweep gas factor
SR steam reforming
TR traditional reactor
WGS water gas shift
Nomenclature
A surface area, m
2
C
T
S
i
total surface concentration of site i, mol m
2
CH
3
OH
in
methanol molar ow rate fed to the reactor, mol s
1
CH
3
OH
out
methanol molar ow rate going out from the
reactor, mol s
1
D diameter, m
d
p
catalyst diameter, m
E
a
apparent activation energy for membrane, J mol
1
2
2
O/CH
3
OH [ 3,
0.03 ml/min and 10
L7
2
2
O/CH
3
OH [ 3,
0.03 ml/min and 10
L6
2
2
O/CH
3
OH [ 3,
0.03 ml/min and 10
L6
E
a,i
transport activation energy for species in the silica
membrane, J mol
1
E
i
activation energy for rate constant of reaction i,
kJ mol
1
F
i
molar ow rate, mol s
1
J
i
ux of component i, mol s
1
k
j
rate constant for reaction j (units are specic to the
form of the rate expression)
k
N
i
pre-exponential term in Arrhenius expression; this
is the intercept of the Arrhenius equation as T /N
K
eq
j
equilibrium constant of reaction j
K
i
equilibrium constant or adsorption coefcient for
surface species i
m feed molar ratio coefcient
P total pressure, bar
p
i
partial pressure of component i, bar
Pe
i
permeance coefcient for silica membranes of
species i, mol s
1
cm
2
atm
0.5
Pe permeability coefcient, mol s
1
cm
1
atm
0.5
r
i
rate of reaction i (mol s
1
m
2
) or rate of formation of
species i, mol s
1
R gas constant, kPa m
3
mol
1
K
1
T temperature, K
V volume of catalyst or reactor, m
3
U gas velocity, ms
1
W mass of catalyst, g
z axial coordinate
Greek letter
porosity of the catalyst,
m gas viscosity, kg m
1
s
1
n
ij
stoichiometric coefcient
r
b
density of the catalyst bed, kg m
3
r
g
density of gas, kg m
3
d membrane thickness, m
r e f e r e n c e s
[1] Mueller-Langer F, Tzimas E, Kaltschmitt M, Peteves S.
Techno economic assessment of hydrogen production
processes for the hydrogen economy for the short and
medium term. Int J Hydrogen Energy 2007;32:3797e810.
[2] Donitz W. Fuel cells for mobile applications, status,
requirements and future application potential. Int J
Hydrogen Energy 1998;23:611e5.
[3] Ogden JM, Steinbugler MM, Kreutz TG. A comparison of
hydrogen, methanol and gasoline as fuels for fuel cell
vehicle: implications for vehicle design and infrastructure
development. J Power Sources 1999;79:143e68.
[4] Iulianelli A, Longo T, Basile A. Methanol steam reforming in a
dense PdeAg membrane reactor: the pressure and WHSV
effects on CO-free H
2
production. J Membr Sci
2008;323:235e40.
[5] Basile A, Parmaliana A, Tosti S, Iulianelli A, Gallucci F,
Espro C, Spooren J. Hydrogen production by methanol steam
reforming carried out in membrane reactor on Cu/Zn/Mg-
based catalyst. Catal Today 2008;137:17e22.
[6] Basile A, Gallucci F, Paturzo L. A dense Pd/Ag membrane
reactor for methanol steam reforming: experimental study.
Catal Today 2005;104:244e50.
[7] Basile A, Tereschenko GF, Orekhova NV, Ermilova MM,
Gallucci F, Iulianelli A. An experimental investigation on
methanol steam reforming with oxygen addition in a at
PdeAg membrane reactor. Int J Hydrogen Energy
2006;31:1615e22.
[8] Harold MP, Nair B, Kolios G. Hydrogen generation in a Pd
membrane fuel processor: assessment of methanol-based
reaction systems. Chem Eng Sci 2003;58:2551e71.
[9] Nair BKR, Harold MP. Hydrogen generation in a Pd membrane
fuel processor: productivity effects during methanol steam
reforming. Chem Eng Sci 2006;61:6616e36.
[10] Wieland IS, Melin IT, Lamm IA. Membrane reactors for
hydrogen production. Chem Eng Sci 2002;57:1571e6.
[11] Iulianelli A, Longo T, Basile A. Methanol steam reforming
reaction in a PdeAg membrane reactor for CO-free hydrogen
production. Int J Hydrogen Energy 2008;33:5583e8.
[12] Gallucci F, Basile A. Palladium membrane reactor for steam
reforming reaction: a comparison between different fuels.
Int J Hydrogen Energy 2008;33:1671e87.
[13] Gallucci F, Basile A. Co-current and counter-current modes
for methanol steam reforming membrane reactor. Int J
Hydrogen Energy 2006;31:2243e9.
[14] Gallucci F, Basile A, Tosti S, Iulianelli A, Drioli E. Methanol
and ethanol steam reforming in membrane reactors: an
experimental study. Int J Hydrogen Energy 2007;32:1201e10.
[15] Amandusson H, Ekedahl LG, Dannetun H. Hydrogen
permeation through surface modied Pd and PdeAg
membranes. J Membr Sci 2001;193:35e47.
[16] Ghasemzadeh K, Liguori S, Morrone P, Iulianelli A,
Piemonte V, Babaluo AA, Basile A. H
2
production by low
pressure methanol steam reforming in a dense PdeAg
membrane reactor in co-current ow conguration:
experimental and modeling analysis. Int J Hydrogen Energy
13 December 2013;38(36):16685e97.
[17] Ismail AF, David LIB. A review on the latest development of
carbon membranes for gas separation. J Membr Sci
2001;193:1e18.
[18] Sznejer GA, Efremenko I, Sheintuch M. Carbon membranes
for high temperature gas separations: experiment and
theory. AIChE J 2004;50:224e31.
[19] Lee DW, Park S-J, Yu C-Y, Ihm S-K, Lee KH. Study on
methanol reformingeinorganic membrane reactors
combined with wateregas shift reaction and relationship
between membrane performance and methanol conversion.
J Membr Sci 2008;316:63e72.
[20] Lee DW, Nam SE, Sea B, Ihm SK, Lee KH. Preparation of Pt-
loaded hydrogen selective membranes for methanol
reforming. Catal Today 2006;118:198e204.
[21] Sa S, Silva H, Sousa Jose M, Mendes A. hydrogen production
by methanol reforming in a membrane reactor: palladium vs
carbon molecular sieve membranes. J Membr Sci
2009;339:160e70.
[22] Sa S, Sousa Jose M, Mendes A. Methanol steam reforming in a
double bed membrane reactor for producing PEMFC grade
hydrogen. Catal Today 2010;156:254e60.
[23] Ghasemzadeh K, Morrone P, Liguori S, Babaluo AA, Basile A.
Evaluation of dense PdeAg and silica membrane reactors
performance for hydrogen production via methanol steam
reforming: modeling study. Int J Hydrogen Energy
13 December 2013;38(36):16698e709.
[24] Ghasemzadeh K, Morrone P, Iulianelli A, Liguori S,
Babaluo AA, Basile A. H
2
production in silica membrane
reactor via methanol steam reforming: modeling and HAZOP
analysis. Int J Hydrogen Energy 21 August
2013;38(25):10315e26.
[25] Peppley BA, Amphlett JC, Kearns LM, Mann RF. Methanol
steam reforming on Cu/ZnO/Al
2
O
3
. Part 1: the reaction
network. Appl Catal A 1999;179:21e9.
[26] Peppley BA, Amphlett JC, Kearns LM, Mann RF. Methanol
steam reforming on Cu/ZnO/Al
2
O
3
catalysts. Part 2. A
comprehensive kinetic model. Appl Catal A
1999;179:31e49.
[27] Ergun S. Fluid ow through packed columns. Chem Eng Prog
1952;48:89e94.
[28] Abdel-Jawad MM, Gopalakrishnan S, Duke MC,
Macrossan MN, Schneider PS, Diniz da Costa JC. Flow elds
on feed and permeate sides of tubular molecular sieving
silica (MSS) membranes. J Membr Sci 2007;299:229e35.
[29] Anderson JD. Computational Fluids Dynamics. New York:
McGraw Hill; 1995.

J Ijhydene 2014 04 010

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

J Ijhydene 2014 04 010

Uploaded by

Copyright:

Available Formats

A simulation study on methanol steam reforming

in the silica membrane reactor for hydrogen

You might also like