i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 3 ( 2 0 1 8 ) 1 5 3 2 1 e1 5 3 2 9

Available online at www.sciencedirect.com

ScienceDirect

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

The evaluation of methane mixed reforming

reaction in an industrial membrane reformer for
hydrogen production

Seyyed Mohammad Jokar a,*, Payam Parvasi a, Angelo Basile b

a
Department of Chemical, Petroleum and Gas Engineering, Shiraz University of Technology, Shiraz, Iran
b
CNR-ITM, C/o University of Calabria, Via Pietro Bucci, Cubo 17/C, 87030 Rende CS, Italy

article info abstract

Article history: In this work, the performance of an industrial dense PdeAg membrane reformer for
Received 18 April 2018 hydrogen production with methane mixed reforming reaction was evaluated. The rate
Received in revised form parameters of mixed reforming reaction on a Ni based catalyst optimized by using the
21 June 2018 experimental results. One-dimensional models have been considered to model the steam
Accepted 25 June 2018 reforming industrial membrane reformer (SRIMR) and mixed reforming industrial mem-
Available online 18 July 2018 brane reformer (MRIMR). The models are validated by experimental data.
The proficiency of MRIMR and SRIMR at similar conditions used as a basis of compar-
Keywords: ison in terms of temperature, methane conversion, hydrogen yield, syngas production rate
Mixed reforming and CO2 flow rate. Results revealed that the methane conversion, hydrogen yield and
Industrial membrane reformer syngas production rate in MRIMR is considerably higher than SRIMR. Furthermore, the
Hydrogen production operation temperature of MRIMR could be 195
C lower than that for SRIMR. This would
CO2 consumption contribute to a major decrease in process costs as well as a reduction in catalyst sintering.
On the other hand, although MRIMR consumes CO2, the exited CO2 flow rate at the SRIMR is
three times more than that of at the MRIMR, which is a main advantage of MRIMR from the
environmental issues point of view.
© 2018 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

The conventional reformers including a number of

Introduction reforming tubes loaded by the catalyst and placed in a
furnace. The following two endothermic reactions, namely
Methane Steam reforming (MSR) is a common industrial steam reforming, are taken place in the reformer [3]:
process in which methane is converted into useful products,
such as hydrogen and synthesis gas (syngas). In this process, CH4 þ H2 O ⇔ CO þ 3H2 DH ¼ þ206 kJ=mol (1)
Ni-based catalysts are mostly employed to produce syngas at
high temperatures (800e1,000
C) [1,2]. The AL2O3 is mainly CH4 þ 2H2 O ⇔ CO2 þ 4H2 DH ¼ þ165 kJ=mol (2)
used as a support in industrial methane reformers. There have
Furthermore, the following reaction known as the water
been many researches dedicated to MSR reactions on Ni/
gas shift reaction should be considered in the MSR reaction
AL2O3 catalysts [3e9].
system.

* Corresponding author.
E-mail address: jokar@sutech.ac.ir (S.M. Jokar).
https://doi.org/10.1016/j.ijhydene.2018.06.142
0360-3199/© 2018 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
15322 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 3 ( 2 0 1 8 ) 1 5 3 2 1 e1 5 3 2 9
CO þ H2 O ⇔ CO2 þ H2 DH ¼ 41:7 kJ=mol
Dry reforming of methane (equation (4)) is an alternative
process for syngas production. The carbon dioxide could
provide from waste products [10].
CH4 þ CO2 ⇔ 2CO þ 2H2 DH ¼ þ247:9 kJ=mol
The dry reforming reaction is an appropriate method for
consuming CO2 and methane simultaneously. Therefore, it
has been taken into consideration from environmental as-
pects [2]. However, there are two disadvantages for dry
reforming reaction:
1- The carbon deposits causing deactivation of the catalyst
during the reaction [11e13].
2- At the same operating conditions, steam reforming pro-
vides higher methane conversion than dry reforming [14].
One alternative to overcome the mentioned problems
would be to combine the dry and steam reforming (mixed
reforming). The methane mixed reforming (MMR) process
could decrease catalyst deactivation via oxidation of the
carbonaceous products [15,16]. The MMR reactions are as
follows [17]:
CH4 þ H2 O4CO þ 3H2 ðsteam reformingÞ
CH4 þ 2H2 O4CO2 þ 4H2 ðsteam reformingÞ
CO þ H2 O4CO2 þ H2 ðwater  gas shiftÞ
CH4 þ CO2 42CO þ 2H2 ðdry reformingÞ
Choudhary and Rajput [18] indicated that methane could
be totally converted to syngas with high selectivity for carbon
monoxide and hydrogen by MMR. Furthermore, the
enhancing effect of combined steam and carbon dioxide
reforming process on methane activity and hydrogen to car-
bon monoxide ratio was proved by Hegarty et al. [19].
Metal support catalysts are commonly used for MMR due to
their high thermal conductivity [20]. The application of Ni
based catalysts for MMR reaction was studied by some re-
searchers [18,19,21e24].
The MMR reaction was performed on Ni/YeAl2O3 catalyst
at high temperatures by the work by Sona et al. [25]. By
applying this catalyst the methane conversion of 98.3% ach-
ieved and also a low amount of carbon coke deposit formation
observed in this process (3.6% after 200 h).
The application of membrane reactors in methane
reforming has recently drawn particular attention. By using
membrane reactor the full conversion could occur at low
temperatures for methane reforming with Ni based catalysts
[26].
Different techniques recently used to make supported Pd
based membranes with elevated selectivity and permeability
[27e31]. Unfortunately, such membranes exhibit high
permeability and low selectivity or vice versa. Therefore,
supported Pd based membranes are unsuitable to produce
pure hydrogen because of their low hydrogen selectivity [32].
Dense Pd-based membranes have high hydrogen selectivity
(3) and good permeability. Furthermore, in comparison with
supported membranes, dense membranes have the following
two advantages:
1- They can operate at high pressure and temperature.
(4) 2- They have a considerable mechanical strength and could
be easily acquired [33].
There are few studies investigating the industrial produc-
tion of hydrogen by methane reforming in membrane re-
formers. Basile et al. [34] experimentally studied the methane
steam reforming reaction in a dense PdeAg membrane
reformer.
The methane conversion of 65% was achieved at the
moderate temperature of 450
C and the pressure of 5 bar. In
the other work, their team modeled an industrial membrane
reformer based on the experimental tests [35]. They indicated
that at the optimum conditions (550
C, P ¼ 25 bar and S/
C ¼ 2.5), 98.9% of the methane conversion could be obtained in
an industrial membrane reformer with the membrane thick-
ness of 20 mm and 6 m length [35].
Jokar et al. [36] investigated the MMR in an experimental
membrane reactor. They used Ni/CeO2 as a catalyst and a
membrane reactor with 150 mm thickness and 10 cm length.
The best experimental results were observed at T ¼ 450
C and
(5)
p ¼ 350 kPa. At these conditions, 64% of conversion and 52% of
hydrogen yield was achieved in the reactor.
(6)
As mentioned, all the previous works focused on modeling
of industrial MSR reformers. Unfortunately, there is no
(7)
investigation on the application of MMR for an industrial
membrane reformer. Consequently, there is no exact rate for
(8)
MMR reactions in a membrane reformer which normally
operates at low temperatures (<550
C). The objective of this
work is to predict the performance of this kind of reformer,
based on the experimental results obtained from Jokar et al.
[36]. In this regard, the reaction rate constants are estimated
and the industrial reformer is modeled. The results are then
compared with common industrial membrane reformer with
steam reforming reaction at optimal conditions given in De
Falco et al.'s work [35].
Membrane reforming reactor model
A detailed kinetic model has been developed to evaluate the
performance of an industrial membrane reactor.
Description of membrane reforming reactor
Fig. 1 shows the schematic diagram of an industrial mem-
brane reformer. Syngas is produced in the shell side by pass-
ing gas mixture comprising methane, steam, inert gas (such as
nitrogen) and steam and/or carbon oxide over Ni-based cata-
lysts. The pure hydrogen is permeated to the tube side, from
where it is removed by sweep gas, such as nitrogen or steam.
The reactions take place in the shell side to give a higher heat-
transfer rate due to the direct contact of reactants with the hot
wall.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 3 ( 2 0 1 8 ) 1 5 3 2 1 e1 5 3 2 9 15323

dFi X
¼ hrj rB i ¼ CH4 ; H2 O; CO; CO2 (9)
dz j

Reaction feed
 
dFH2 X pffiffiffiffiffiffi qffiffiffiffiffiffi

Sweep gas
¼ hrj rB  aH Amem prH  ptH (10)
dz j

P
3   P
6  
hrB rj  DHf ;j AC  Cpi dF
dz
i
T  Tref
dTreac j¼1 i¼1
¼
dz P
6
Cpi Fi
i¼1
 
pDto Utube Tperm  Treac þ pDso Ushell ðTshell  Treac Þ
þ (11)
P6
Cpi Fi
i¼1

where Fi is the molar flow rate of component i, aH is hydrogen

permeability, rB is bed density, prH and ptH are hydrogen partial
Heating zone

Heating zone

pressures in reaction and tube sides, Ac is cross-sectional area,

and Dto and Dso are respectively outside diameter of tube and
reaction sides.
The hydrogen permeability (aH ) could be calculated from
equation (12).

 
p0H2 Ea
aH ¼ exp  (12)
d RT

In this equation p0H2 , d, Ea and R are the pre-exponential

factor, membrane thickness, activation energy and universal
gas constant respectively. A dense PdeAg membrane is
considered for this work.
Fig. 1 e Schematic diagram of an industrial membrane
As mentioned, the proposed model has a steady-state
reformer.
distribution and there is no initial condition. The boundary
conditions for reaction side are as follows:
Mathematical model of membrane reactor
z¼0; Fi ¼ Fi feed ; Treac ¼ Tfeed (13)
An element of length Dz, is considered as shown in Fig. 2.
Tube side
The system was represented mathematically by a material
For the tube side, the mass and energy balance equations are
balance for each species and energy balance for the gas
as follows:
mixture in tube and reaction sides over an element of the
membrane reformer. A one-dimensional homogeneous  
steady-state model has been used to model the reformer. For dFH2 pffiffiffiffiffiffi qffiffiffiffiffiffi
¼ aH Amem prH  ptH (14)
the homogeneous model, the temperature and concentration dz
gradients between gas and catalyst phases can be ignored,
and the equations for these phases can be combined [37].  
dTperm CpH2 dFi
Treac  Tperm þ pDto Utube ðT  Ttube Þ
¼ dz
(15)
dz CpH2 FH2 þ CpN2 FN2
Reaction side
According to the mentioned explanation above, the mass and
Like the reaction side, the boundary conditions for the tube
energy balances of the reaction side can be represented as:
side could be written as follows:

z¼0; FH2 ¼ FH2 sweep ; Tperm ¼ Tsweep (16)

(Freaction)z+∆z (Fsweep)z+∆z
Reaction kinetics
The reaction rate equations for steam, dry and mixed
reforming reactions are presented in this section.
∆z
Steam reforming. There are a large number of kinetic models
and rate expressions for MSR reactions in the literature. Xu
(Freaction)z (Fsweep)z and Froment introduced the most realistic rate equations over
a commercial NieAl2O3 catalyst in an integral flow reactor [3].
The rate expressions are as follows [3]:
Fig. 2 e An elemental volume of membrane reformer.
15324 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 3 ( 2 0 1 8 ) 1 5 3 2 1 e1 5 3 2 9

   
P3 PCO 4300:0
k1
2:5
PH
PCH4 PH2 O  H2K1 k4 ¼ k40 exp (29)
T
R1 ¼  2
 2 (17)
1 þ KCH4 PCH4 þ KH2 PH2 þ KCO PCO þ KH2 O PH2 O PH2  
7500:0
k5 ¼ k50 exp (30)
  T
P4 PCO
k2
P3:5
PCH4 P2H2 O  H2K2 2
H The k40 and k50 parameters are listed in Table 2.
R2 ¼  2
 2 (18)
1 þ KCH4 PCH4 þ KH2 PH2 þ KCO PCO þ KH2 O PH2 O PH2
Mixed reforming. For MMR reaction over a Ni-based catalyst,
 P P
equations 17e30 could be applied by modifying the constants'
k3
PH2
PCO PH2 O  H2K3CO2
R3 ¼   2 (19) values. Park et al. [39] optimized k10 ; k20 ; k30 ; k40 and k50 pa-
1 þ KCH4 PCH4 þ KH2 PH2 þ KCO PCO þ KH2 O PH2 O PH2 rameters for mixed reforming reaction at the temperatures
The rate, equilibrium and adsorption constants of equa- higher than 750
C. It should be mentioned that the membrane
tions 17e19 should be calculated by equations 20e27 [3]: reformers usually operate at relatively low temperatures
(<550
C) and consequently the mentioned parameters should
  
240; 100 1 1 be optimized based on experimental data. The optimization
k1 ¼ k10 exp  (20)
R 648 T process is discussed later in section Mixed reforming model
validation.
  
243; 900 1 1
k2 ¼ k20 exp  (21)
R 648 T
Numerical solution
  
67; 100 1 1
k3 ¼ k30 exp  (22)
R 648 T The modeling codes were written by MATLAB 2015a software.
The modified Rosenbrock method (ode23s) was adopted to
 
DG+rxni solve the set of stiff ordinary differential equations (ODEs). 100
Ki ¼ exp  i ¼ 1; 2 and 3 (23)
RT nodes were considered for the numerical solution of ODEs.
The reaction and tube side equations were solved simulta-
  
38; 300 1 1 neously. The methane conversion, carbon dioxide conversion
KCH4 ¼ 1:8  101 exp  (24)
R 823 T and hydrogen production yields were obtained based on the
following equations:
  
82; 900 1 1
KH2 ¼ 2:9  102 exp  (25) feed
FCH4  FCH4
product
R 648 T
XCH4 ¼ feed
 100 (31)
FCH4
  
70; 700 1 1
KCO ¼ 40:9 exp  (26)
R 648 T feed
FCO2  FCO2
product
XCO2 ¼ feed
 100 (32)
   FCO2
88; 700 1 1
KH2 O ¼ 0:4 exp  (27)
R 823 T
product feed
FH2  FH2
The k10 ; k20 and k30 parameters are listed in Table 1. YH2 ¼ feed
*100 (33)
FCH4

Dry reforming. The following rate expression for methane dry

reforming (equation (4)) over Ni/La2O3 catalyst was developed Optimization and results
by Verykios [38]:
At first, the model is validated with experimental data. The
k4 k5 K4 PCH4 PCO2 validated model is then used to predict and optimize the
R4 ¼ (28)
k4 K4 PCH4 PCO2 þ k4 PCH4 þ k5 K4 PCO2
performance of an industrial reactor.
where K4 , k4 and k5 constants should be derived from equa-
tions (28)e(30) respectively. Steam reforming model validation

In order to verify the model, results of the membrane reformer

model for MSR reaction have been compared with Basile et al.
Table 1 e The k10 ; k20 ; k30 parameters of equations 20e22
for steam reforming reaction [3].
Parameters Value Table 2 e The k40 and k50 parameters of equations 29 and
  6 30 for dry reforming reaction [38].
kmol:bar0:5 1.8  10
k10 Parameters Value
kgcat :h
  
kmol:bar0:5 2.2  107 1  2.61  103
k20 mol:Pa
kgcat :h k40
kgcat :s
   
kmol:bar1 2.28 mol 5.35  102
k30 k50
kgcat :h kgcat :s
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 3 ( 2 0 1 8 ) 1 5 3 2 1 e1 5 3 2 9 15325

Table 5 e The MRIMR characterizations [35].

Reactor length (L) 6.82 m
Outer tube external radius 10.16 cm
Outer tube internal radius 9.16 cm
Inner tube external radius 7 cm
Inner tube internal radius 6 cm
PdeAg membrane thickness (d) 20 mm

Fig. 3 e Comparison of modeling and experimental data for

MSR reaction over a Ni-based catalyst (S/C ¼ 2, Sweep
factor ¼ 1.2 at three different temperatures 400, 450 and
500
C).

Fig. 5 e Methane conversion profile in MRIMR (S/C ¼ 2.5,

Sweep factor ¼ 1.2, T ¼ 550
C, P ¼ 25 bar).

Fig. 4 e Comparison of modeling and experimental data for

MMR reaction over a Ni-based catalyst (S/C ¼ 2.64, Sweep
factor ¼ 1.2, T ¼ 450
C).

Table 3 e Optimal values of the mixed reforming reaction

rate constants. Fig. 6 e Carbon dioxide Conversion profile in MRIMR (S/
k10 7.24166  107 C ¼ 2.5, Sweep factor ¼ 1.2, T ¼ 550
C, P ¼ 25 bar).
k20 1.6882  109
k30 2.5440
k40 2.633550  103
k50 8.474329  102 at different temperatures [34]. The results are shown in Fig. 3.
As it could be observed, there is a good complementation
between the model and experimental data.

Table 4 e Feed composition of MRIMR [35,36]. Mixed reforming model validation

Methane (kmol/h) 1.75
Steam (kmol/h) 6.61 As mentioned, the k10, k20, k30, k40 and k50 parameters of Tables
Carbon Dioxide (kmol/h) 0.89 1 and 2 should be optimized for MMR reaction. The optimal
15326 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 3 ( 2 0 1 8 ) 1 5 3 2 1 e1 5 3 2 9
values of these constants were obtained by the differential
evolution (DE) optimization algorithm. The DE algorithm has a
remarkable ability to optimize nonlinear functions. It is a
powerful and fast method for solving the complex optimiza-
tion problems in continuous spaces. The algorithm is pre-
sented to overcome the genetic algorithms (GA) difficulties
due to the lack of local search ability. More details about DE
method are available in our previous works [37,40]. In the
optimization calculations, 80% of the Jokar et al. [36] data used
for optimization while the rest of data used for validation
(Fig. 4). The optimal values of the k10, k20, k30, k40 and k50 pa-
rameters, which are obtained in this study, are given in Table
3.
According to Fig. 4, it is clear that the developed model is
capable for modeling of MMR reaction process. Furthermore,
another available data from Iulianelli et al.’s work [41] at
Fig. 7 e Hydrogen yield profile in MRIMR (S/C ¼ 2.5, Sweep
factor ¼ 1.2, T ¼ 550
C, P ¼ 25 bar).
Fig. 8 e Effect of reaction temperature on final methane
conversion in MRIMR (L ¼ 6.82 m, S/C ¼ 2.5, Sweep
factor ¼ 1.2, P ¼ 25 bar).
380
C, 2.0 bar, S/C ¼ 1.8 and GHSV ¼ 9000 h1 used to test the
model prediction performance. A deviation of 4.2% was seen
between the experiment conversion (15%) and predicted
conversion (14.37%).
Mixed reforming industrial membrane reformer (MRIMR)
The performances of an MRIMR were investigated in this
section. The feed composition of the industrial membrane
reformer is presented in Table 4.
As mentioned in the introduction section, the optimal
conditions of a steam reforming industrial membrane
reformer (SRIMR) have been considered before by De Falco
et al. [35]. In order to make sense of the operation of a MRIMR,
the model was run at SRIMR optimal conditions (S/C ¼ 2.5,
(a)
(b)
Fig. 9 e (a) Comparison of methane conversion profile in
MRIMR and SRIMR (Sweep factor ¼ 1.2, T ¼ 550
C,
P ¼ 25 bar). (b) Comparison of hydrogen yield profile in
MRIMR and SRIMR (Sweep factor ¼ 1.2, T ¼ 550
C,
P ¼ 25 bar).
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 3 ( 2 0 1 8 ) 1 5 3 2 1 e1 5 3 2 9 15327

Fig. 10 e Comparison of carbon dioxide flowrate profile in (a)

MRIMR and SRIMR (Sweep factor ¼ 1.2, T ¼ 550
C,
P ¼ 25 bar).

Sweep factor ¼ 1.2, T ¼ 550
C, P ¼ 25 bar) and with its

charcterizations (Table 5). The results are demonstrated in
Figs. 5e7.
As it is shown in Fig. 5, at z ¼ 3.8874 m, 99% conversion
could be achieved by MMR reaction. Therefore, the MRIMR is
able to reach high conversion at approximately the mid-
length of the reformer.
The carbon dioxide conversion profile is shown in Fig. 6.
There is a descending-ascending strength curve for CO2 con-
version. Due to the high rate of MSR reaction, carbon dioxide is
initially produced, and the curve is descending to the mini-
mum. Subsequently, carbon dioxide is consumed extensively
due to the dry reforming reaction, and the curve is ascending
away from the minimum.
Fig. 7 shows the hydrogen yield profile along the reactor
(b)
length. High hydrogen production rate could be seen during Fig. 11 e (a) Comparison of H2/CO ratio profile in reaction
the mixed reforming process. side of MRIMR and SRIMR (Sweep factor ¼ 1.2, T ¼ 550
C,
Temperature is an important factor that plays a significant P ¼ 25 bar). (b) Comparison of syngas production rate
role in the cost of reforming process. Furthermore, there is profile in reaction side of MRIMR and SRIMR (Sweep
less risk of catalyst sintering at lower temperatures. The effect factor ¼ 1.2, T ¼ 550
C, P ¼ 25 bar).
of the reaction temperature on the final methane conversion
is depicted in Fig. 8. The model shows that, at L ¼ 6.82 m, the
mixed reforming process is completed at temperatures higher In regard to the environmental dimension, the mixed
than 355
C. As the reforming process was experimentally reforming process offers notable features such as high CO2
tested for T  400
C, the MRIMR could certainly operate at consumption rates and low global warming potential. The CO2
400
C which is considerably lower than SRIMR temperature molar flow rate in MRIMR and SRIMR are calculated and
(550
C). This can lead to reducing energy costs and thus lower compared in Fig. 10. As it is observed, MRIMR produces less
operating and capital costs. CO2 than SRIMR, and unlike SRIMR, it has an ascending-
descending curve. The ascending slope of the curve at the
Comparison of MRIMR and SRIMR beginning of the reactor is related to the low concentration of
In this section, the performances of MRIMR and SRIMR are CO2 and high rate of CO2 production during steam reforming
compared. Fig. 9 (a) and (b) demonstrates the methane con- reaction. Subsequently, the descending slope of the curve is
version and hydrogen yield profiles along the reactor length due to high concentration of CO2 and high rate of CO2 con-
for both reformers. As it is shown in this figure, the higher sumption during the dry reforming reaction, meaning the
methane conversion and hydrogen yield is achieved in amount of its consumption is more than that of generation.
simultaneous dry and steam reforming process. The hydrogen Therefore, although the MRIMR consumes CO2, the amount of
production rate of MRIMR is 50% higher than SRIMR. carbon dioxide at the reformer exit is approximately 1/3 of
15328 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 3 ( 2 0 1 8 ) 1 5 3 2 1 e1 5 3 2 9
that for steam reforming process (Fig. 10). Consequently, from
an environmental point of view, the MRIMR has the following
advantages:
1- It consumes CH4 and CO2, which are the two major carbon
containing greenhouse gases causing global warming.
2- The produced CO2 is much lower than conventional
membrane reformer.
Syngas is a mixture of hydrogen and carbon monoxide
with variable H2/CO ratios [42].
Fig. 11 shows the comparison between H2/CO ratio and
syngas production rate profiles along the reactor length in the
reaction side of MRIMR and SRIMR. As it is depicted in Fig. 11 (a),
the exited H2/CO ratio of syngas is about 2 for both reformers.
This is the desired syngas ratio for methanol synthesis [43].
Furthermore, at the same S/C, the syngas production rate of
MRIMR is two times more than SRIMR (Fig. 11 (b)). Hence, the
MRIMR produces the maximum amount of pure hydrogen and
syngas with a proper H2/CO ratio in the tube and reaction sides
respectively (Figs. 7 and 11).
Conclusion
The steam reforming of methane in a PdeAg membrane
reformer is considered as an industrial process for producing
pure hydrogen and syngas. The membrane reformers have a
capability to combine reaction and hydrogen separation in a
single unit. Unfortunately, the SRIMR operates at tempera-
tures higher than 550
C and the process produces a large
amount of carbon dioxide. This work was focused on the
application of mixed reforming reaction, which could be an
alternative process for industrial membrane reformers.
A one-dimensional model was proposed and validated.
The rate parameters of mixed reforming reactions (k10 ; k20 ; k30 ;
k40 ) were optimized using the experimental data at the reac-
tion temperature. The SRIMR and MRIMR were modeled and
compared.
The MRIMR showed higher methane conversion, as well as
higher hydrogen yield with respect to SRIMR operated at
optimal conditions (S/C ¼ 2.5, Sweep factor ¼ 1.2, T ¼ 550
C,
P ¼ 25 bar). These advantages were due to the combination of
dry and steam reforming in MRIMR.
The results showed that the MRIMR could operate at the
temperatures 195
C lower than SRIMR operating temperature.
This could be able to decrease the operating cost and improve
the process reliability. On the other hand, despite the con-
sumption of CO2 in MRIMR, the amount of carbon dioxide at
the reformer exit was three times less than SRIMR. Therefore,
utilization of mixed reforming reaction for industrial mem-
brane reformers would be environmentally friendly. Further-
more, MRIMR produces a high amount of syngas with the
appropriate ratio (2/1) at optimal conditions.
The study demonstrates that the MRIMR is a process with
huge advantages, since it allows achieving high reactants
conversion at lower operating temperature, producing large
amount of pure hydrogen and syngas, consuming carbon di-
oxide and decreasing the operation cost as well as CO2
emission.
Acknowledgments
The authors are grateful to the Shiraz University of Technol-
ogy for supporting this work.
references
[1] Kyriakou V, Garagounis I, Vourros A, Vasileiou E,
Manerbino A, Coors WG, et al. Methane steam reforming at
low temperatures in a BaZr0.7Ce0.2Y0.1O2.9 proton conducting
membrane reactor. Appl Catal B Environ 2016;186:1e9.
[2] Huang TJ, Jhao SY. Ni-Cu/samaria-doped ceria catalysts for
steam reforming of methane in the presence of carbon
dioxide. Appl Catal Gen 2006;302:325e32.
[3] Xu J, Froment GF. Methane steam reforming, methanation
and water-gas shift: I. Intrinsic kinetics. Am Inst Chem Eng J
1989;35:88e103.
[4] German ED, Sheintuch M. Methane steam reforming rates
over Pt, Rh and Ni (111) accounting for H tunneling and for
metal lattice vibrations. Surf Sci 2017;656:126e39.
[5] Chen J, Yan L, Song W, Xu D. Methane steam reforming
thermally coupled with catalytic combustion in catalytic
microreactors for hydrogen production. Int J Hydrogen
Energy 2017;42(1):664e80.
[6] Mundhwa M, Thurgood CP. Numerical study of methane
steam reforming and methane combustion over the
segmented and continuously coated layers of catalysts in a
plate reactor. Fuel Process Technol 2017;158:57e72.
[7] Laprune D, Theodoridi C, Tuel A, Farrusseng D, Meunier FC.
Effect of polyaromatic tars on the activity for methane steam
reforming of nickel particles embedded in silicalite-1. Appl
Catal B Environ 2017;204:515e24.
[8] Lee SY, Lim H, Woo HC. Catalytic activity and
characterizations of Ni/K2TixOyeAl2O3 catalyst for steam
methane reforming. Int J Hydrogen Energy 2014;39:
17645e55.
[9] Kim TW, Park JC, Lim TH, Jung H, Hyun Chun D, Lee HT, et al.
The kinetics of steam methane reforming over a Ni/g-Al2O3
catalyst for the development of small stationary reformers.
Int J Hydrogen Energy 2015;40:4512e8.
[10] Richardson JT, Paripatyadar SA. Carbon dioxide reforming of
methane with supported rhodium. Appl Catal
1990;61:293e309.
[11] Liao C, Horng R. Experimental study of syngas production
from methane dry reforming with heat recovery strategy. Int
J Hydrogen Energy 2017;42:25213e24.
[12] Shamskar FR, Rezaei M, Meshkani F. The influence of Ni
loading on the activity and coke formation of ultrasound-
assisted co-precipitated NieAl2O3 nanocatalyst in dry
reforming of methane. Int J Hydrogen Energy
2017;42(7):4155e64.
[13] Abdullah B, Abd Ghani NA, Vo DN. Recent advances in dry
reforming of methane over Ni-based catalysts. J Clean Prod
2017;162:170e85.
[14] Abashar MEE. Coupling of steam and dry reforming of
methane in catalytic & fluidized bed membrane reactors. Int
J Hydrogen Energy 2004;29:799e808.
[15] Ozkara-Aydinoglu S. Thermodynamic equilibrium analysis
of combined carbon dioxide reforming with steam reforming
of methane to synthesis gas. Int J Hydrogen Energy
2010;35:12821e8.
[16] Demidov DV, Mishin IV, Mikhailov MN. Gibbs free energy
minimization as a way to optimize the combined steam and
carbon dioxide reforming of methane. Int J Hydrogen Energy
2011;36:5941e50.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 3 ( 2 0 1 8 ) 1 5 3 2 1 e1 5 3 2 9
[17] Wang Q, Shi H, Yan B, Jin Y, Cheng Y. Steam enhanced
carbon dioxide reforming of methane in DBD plasma reactor.
Int J Hydrogen Energy 2011;36:8301e6.
[18] Choudhary VR, Rajput AM. Simultaneous carbon dioxide and
steam reforming of methane to syngas over NiOeCaO
catalyst. Ind Eng Chem Res 1996;35:3934e9.
[19] Hegarty MES, O'Connor AM, Ross JRH. Syngas production
from natural gas using ZrO2-supported metals. Catal Today
1998;42:225e32.
[20] Sheng M, Yang H, Cahela D, Tatarchuk BJ. Novel catalyst
structures with enhanced heat transfer characteristics. J
Catal 2011;281:254e62.
[21] Ozkara-Aydınoglu S. Thermodynamic equilibrium analysis
of combined carbon dioxide reforming with steam reforming
of methane to synthesis gas. Int J Hydrogen Energy
2010;35:12821e8.
[22] Choudhary VR, Mondal KC. CO2 reforming of methane
combined with steam reforming or partial oxidation of
methane to syngas over NdCoO3 perovskite-type mixed
metal-oxide catalyst. Appl Energy 2006;83:1024e32.
[23] Noronha FB, Shamsi A, Taylor C, Fendley EC, Stagg-
William S, Resasco DE. Catalytic performance of Pt/ZrO2 and
Pt/Ce-ZrO2 catalysts on CO2 reforming of CH4 coupled with
steam reforming or under high pressure. Catal Lett
2003;90:13e21.
[24] Jabbour K, Massiani P, Davidson A, Casale S, El Hassan N.
Ordered mesoporous “one-pot” synthesized Ni-Mg(Ca)-Al2O3
as effective and remarkably stable catalysts for combined
steam and dry reforming of methane (CSDRM). Appl Catal B
Environ 2017;201:527e42.
[25] Sona IH, Lee SJ, Soon A, Roh H, Leed H. Steam treatment on
Ni/Y-Al2O3 for enhanced carbon resistance in combined
steam and carbon dioxide reforming of methane. Appl Catal
B Environ 2013;134e135:103e9.
[26] Roux JF. Membrane reactor modeling for hydrogen
production through methane steam reforming. Polytechnic
Institute; 2011. p. 21e4. Master thesis Submitted to the
Faculty of the Worcester.
[27] Gade SK, Thoen PM, Way JD. Unsupported palladium alloy
foil membranes fabricated by electroless plating. J Membr Sci
2008;316(1e2):112e8.
[28] Guo Y, Zhang X, Deng H, Wang X, Wang Y, Qiu J, et al. A
novel approach for the preparation of highly stable Pd
membrane on macroporous a-Al2O3 tube. J Membr Sci
2010;362(1e2):241e8.
[29] Basile A, Palma V, Ruocco C, Bagnato G, Jokar SM,
Rahimpour MR, et al. Pure hydrogen production via ethanol
steam reforming reaction over a novel Pt-Co based catalyst
in a dense Pd-Ag membrane reactor (an experimental study).
Int J Membr Sci Technol 2015;2:5e14.
[30] Ghasemzadeh K, Zeynali R, Basile A. Theoretical study of
hydrogen production using inorganic membrane reactors
during WGS reaction. Int J Hydrogen Energy 2016;41:
8696e705.
15329
[31] Sharma R, Kumar A, Upadhyay RK. Characteristic of a multi-
pass membrane separator for hydrogen separation through
self-supported PdAg membranes. Int J Hydrogen Energy
2018;43:5019e32.
[32] Santucci A, Borgognoni F, Vadrucci M, Tosti S. Testing of
dense PdeAg tubes: effect of pressure and membrane
thickness on the hydrogen permeability. J Membr Sci
2013;444:378e83.
[33] Morreale B, Ciferno J, Howard B, Ciocco M, Marano J, Iyoha O,
et al. Gasification and associated degradation mechanisms
applicable to dense metal hydrogen membranes. In: Bose AC,
editor. Inorganic membranes for energy and environmental
applications. New York: Springer Science and Business
Media, LLC; 2009. p. 173e201.
[34] Basile A, Campanari S, Manzolini G, Iulianelli A, Longo T,
Liguori S, et al. Methane steam reforming in a Pd-Ag
membrane reformer: an experimental study on reaction
pressure influence at middle temperature. Int J Hydrogen
Energy 2011;36:1531e9.
[35] De Falco M, Piemonte V, Di Paola L, Basile A. Methane
membrane steam reforming: heat duty assessment. Int J
Hydrogen Energy 2014;39:4761e70.
[36] Jokar SM, Rahimpour MR, Shariati A, Iulianelli A, Bagnato G,
Vita A, et al. Pure hydrogen production in membrane reactor
with mixed reforming reaction by utilizing waste gas: a case
study. Processes 2016;4:33.
[37] Rahimpour MR, Parvasi P, Setoodeh P. Dynamic optimization
of a novel radial-flow, spherical-bed methanol synthesis
reactor in the presence of catalyst deactivation using
Differential Evolution (DE) algorithm. Int J Hydrogen Energy
2009;33:221e30.
[38] Verykios XE. Catalytic dry reforming of natural gas for the
production of chemicals and hydrogen. Int J Hydrogen
Energy 2003;28:1045.
[39] Park N, Park M, Baek S, Ha K, Lee Y, Kwak G, et al. Modeling
and optimization of the mixed reforming of methane:
maximizing CO2 utilization for non-equilibrated reaction.
Fuel 2014;115:357e65.
[40] Parvasi P, Khosravanipour Mostafazadeh A, Rahimpour MR.
Dynamic modeling and optimization of a novel methanol
synthesis loop with hydrogen-permselective membrane
reactor. Int J Hydrogen Energy 2009;34:3717e33.
[41] Iulianellia A, Liguoria S, Huang Y, Basile A. Model biogas
steam reforming in a thin Pd-supported membrane reactor
to generate clean hydrogen for fuel cells. J Power Sources
2015;273:25e32.
[42] Shamkhali A, Omidkhah MR, Towfighi J, Jafari Nasr MR. The
production of synthesis gas by a combination of steam and
dry reforming using GHR. Petrol Sci Technol
2012;30(6):594e604.
[43] Alarifi Abdulaziz, Alsobhi Saad, Elkamel Ali, Croiset Eric.
Multiobjective optimization of methanol synthesis loop from
synthesis gas via a multibed adiabatic reactor with additional
interstage CO2 quenching. Energy Fuels 2015;29(2):530e7.