Accepted Manuscript

Optimal Conditions in Converting Methanol to Dimethyl Ether, Methyl Formate, and

Hydrogen Utilizing a Double Membrane Heat Exchanger Reactor

Ali Bakhtyari, Mahboubeh Parhoudeh, Mohammad Reza Rahimpour

PII: S1875-5100(15)30269-9
DOI: 10.1016/j.jngse.2015.11.028
Reference: JNGSE 1124

To appear in: Journal of Natural Gas Science and Engineering

Received Date: 9 September 2015

Revised Date: 11 November 2015
Accepted Date: 12 November 2015

Please cite this article as: Bakhtyari, A., Parhoudeh, M., Rahimpour, M.R., Optimal Conditions in
Converting Methanol to Dimethyl Ether, Methyl Formate, and Hydrogen Utilizing a Double Membrane
Heat Exchanger Reactor, Journal of Natural Gas Science & Engineering (2015), doi: 10.1016/
j.jngse.2015.11.028.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

Optimal Conditions in Converting Methanol to Dimethyl Ether, Methyl Formate, and

Hydrogen Utilizing a Double Membrane Heat Exchanger Reactor

Ali Bakhtyari a, Mahboubeh Parhoudeh a, Mohammad Reza Rahimpour a,b,1

a
Department of Chemical Engineering, School of Chemical and Petroleum Engineering, Shiraz

PT
University, Shiraz 71345, Iran

RI
b
Department of Chemical Engineering and Materials Science, University of California, Davis, 1

SC
Shields Avenue, Davis, CA 95616, United States

U
AN
Abstract
M

Following importance of investigating new sources of energy with highly efficient processes, a
new configuration for simultaneous production of high purity dimethyl ether (DME), hydrogen
D

and methyl formate (MF) is numerically studied in this work. In this regard, a catalytic heat-
exchanger reactor assisted with two different membranes for methanol conversion and in-situ
TE

separation of products is simulated. The interesting feature of this system is utilizing only one
feedstock (i.e. methanol) to produce different valuable products. Methanol is dehydrated through
an exothermic reaction and supplies required energy for the methanol dehydrogenation reaction.
EP

Produced water in the exothermic side and produced hydrogen in the endothermic side are
separated by permeation to particular membranes. A steady state one-dimensional plug flow
model is developed to evaluate molar and thermal behavior of the system. Optimum operating
C

conditions are determined applying genetic algorithm as a powerful optimization method. The
proposed configuration working under optimum conditions promotes methanol conversion to
AC

DME to % 95.1 and methanol conversion to MF to % 99.6 providing high purity products in the
output streams.

1
Corresponding author. Department of Chemical Engineering, School of Chemical and Petroleum Engineering,
Shiraz University, Shiraz 71345, Iran. Tel.: +98 711 2303071; fax: +98 711 6287294.
E-mail address: [email protected], [email protected] (M. R. Rahimpour).
 ACCEPTED MANUSCRIPT

Key words: Methanol Dehydration, Methanol Dehydrogenation, Heat exchanger Reactors,

Membrane Reactors, Genetic Algorithm.

PT
1. Introduction

Diverse fuels and chemical intermediates such as dimethyl ether (DME), formaldehyde,

RI
formalin, methyl formate (MF), and acetic acid are produced directly or indirectly from
methanol. Besides, methanol is a potential medium for hydrogen reserve [1, 2]. Accordingly, any
enhancement in methanol associated processes leading to higher efficiency and lower energy

SC
consumption is promising for petrochemical and chemical industries. Heat integrated processes
with energy efficient separation techniques for converting methanol to downstream products are
currently of a great interest.

U
1.1. Process Intensification: An Advanced Methodology in Reducing Energy Consumption
AN
Given increasing worldwide energy demand, environmental problems and limited sources of
fossil fuels, seeking for clean and non-hydrocarbon energy resources and more energy efficient
processes is crucial. Hydrocarbons e.g. natural gas, liquefied petroleum gas (LPG), and natural
M

gas liquid (NGL) are main energy sources of chemical industries as a major energy consumer
[3]. Reconfiguration and reformation of equipment may result in heat integration and then
reducing energy consumption [4-6]. Hence, proposing processes with lower energy consumption
D

without decline in efficiency and productivity is still challenging.

TE

An elegant approach in chemical process technology called process intensification (PI) has been
running the art of increasing production rate and products quality since 1970s. These are
achieved by compacting the equipment and reducing energy and raw material consumption [7-9].
EP

Based on this approach, the generated heat in a part of system could be recovered and reused for
heating or driving the endothermic processes in the other part. Such a point of view led to design
of heat exchanger reactors [10].
C
AC

1.2. Heat Exchanger Reactors: Thermally Coupling and Reusing Heat of Reactions

Chemical reactors with exothermic and endothermic reactions, have a key role in PI. The
generated heat in the exothermic reactions may be recovered and reused as a heat source in
endothermic reactions. Such a capability is obtained using a heat exchanger reactor. Two
reactors are integrated in a shell and tube and heat of exothermic reaction is transferred through
the reactor wall. The transferred heat makes the endothermic reaction proceed. As a result,
equipment size is reduced, thermal efficiency of the reactor is enhanced, and energy
 ACCEPTED MANUSCRIPT

consumption is noticeably lowered. Hence, heat exchanger reactors are vast field of study and
they have attracted attentions of researchers and industries [10].

Heat exchanger reactors with coupling and double coupling of different exothermic and
endothermic reactions were studied by Rahimpour and coworkers. Both fluidized and fixed beds
were examined for different reactions in diverse processes in chemical, petrochemical, and

PT
refining industries. GTL technology and Fischer–Tropsch synthesis [11, 12], naphtha reforming
[13, 14], various methane reforming and syngas production [15], methanol synthesis, DME
synthesis, and dehydrogenation reactions are various processes tried to enhance with heat

RI
exchanger reactors [16-22]. Benefits and drawbacks of heat exchanger reactors are summarized
in a review published by Rahimpour et al. [10].

SC
1.3. Membrane Reactors: Low Energy Consuming Separations

U
Membrane characteristics, e.g. high surface area per unit volume of membrane, high
selectivity/permeability, and controllable permeation rate, make it promising for integrating
AN
reaction and separation in a single system [23, 24]. Unlike the traditional separation methods e.g.
distillation (consuming about 3% of whole energy in the world [25]), there is no phase change in
membrane based separations. Hence, thermal equipment such as reboliler is eliminated. On the
M

one hand, highly efficient phase separations are achieved applying perm-selective membranes.
Decline in high temperature and pressure systems, limited pH range, and problems with
membrane clean up are some disadvantages of membrane reactors [26, 27]. Yield/selectivity
D

enhancement of products by shifting the equilibrium reactions toward products using membranes
are widely studied in systems containing hydrogen and water vapor [28-30]. Hydrogen recovery
TE

from tail gas emissions, in-situ hydrogen production and separations in dehydrogenation
reactions, water recovery from dehydration reactions, and simultaneous hydrogen and oxygen
production from water are currently state of the art [13, 19, 28, 31]. As a whole, recuperative
EP

reactors with simultaneous heat and mass exchange are a vast field of study and research.
C

1.4. Dimethyl Ether (DME): A Clean Fuel with Emissions Standards

AC

Hydrocarbons supply more than 80% of the energy required in household and industrial sections.
Global warming, climate change, and defects in biodiversity caused by hydrocarbons emissions
make the researchers and industries seek for clean sources of energy.

DME (IUPAC name: methoxymethane) is an isomer of ethanol, the simplest ether, volatile, non-
toxic and colorless gas. It is liquefied when subjected to the pressures higher than 0.5 Mpa.
Hence, its storage and transport is possible without high-pressure containers [32, 33]. Low
maintenance costs due to simple engines and no need to spark plugs are advantages of DME as
 ACCEPTED MANUSCRIPT

an engine fuel [32]. Followings are some benefits of DME that make it a promising energy
resource with strict standards of emissions:

• Ease and safety of transport and storage due to lack of the explosive peroxides
• Low CO emission in comparison with natural gas (due to lack of carbon-carbon bonds)
• High cetane number with no NOx emission

PT
It’s potential to become a hydrogen carrier in the fuel cells and intermediate in producing
dimethyl sulfate, methyl acetate, and lower olefins are some applications of DME that have

RI
attracted researchers and industries attention [34-40].

Conventionally, DME is produced from syngas in single-step or two-step methods. In the single-

SC
step method, both methanol synthesis and dehydration occur in a single reactor filled with hybrid
catalyst [41-43]. In the two-step method, DME is produced by dehydration of methanol primarily
obtained from syngas [1, 16, 18]. Presence of hybrid catalyst in the single-step method and in-
situ methanol consumption can overcome thermodynamic restrictions of CO and CO2 conversion

U
to methanol. Besides, methanol separation step is eliminated in this way [44-46]. Although, the
AN
single-step production of DME seems to be more efficient, presence of unreacted syngas and
CO2 in the product streams make the next purification steps difficult [44, 45]. Accordingly
enhancing the two-step DME synthesis method resulting in higher purity of the product and
higher methanol conversion makes some profits for the associated industries.
M
D

1.5. Methyl Formate (MF): Next Candidate of C1 Chemistry

TE

C1 chemistry concerns with producing valuable multi-carbon molecules from single-carbon

molecules (e.g. syngas, methanol, carbon dioxide, and carbon monoxide). A certain building
block in C1 chemistry must be produced in different routes and handled, stored and transported
EP

easily. Additionally, downstream products must be efficiently produced from a certain building
block [47, 48]. MF (IUPAC name: methyl methanoate) is coming up as a building block in C1
chemistry in the near future due to the following potentials [49-53]:
C

• various methods to synthesis e.g. direct synthesis from syngas, dehydrogenation of

methanol, oxidative dehydrogenation of methanol, dimerization of formaldehyde,
AC

hydrocondensation of carbon dioxide and with methanol and carbonylation of methanol

• Storing, handling and transporting similar to LPG
• Source of CO/H2 mixture due to one more CO in the molecule comparing methanol
• Raw material for formic acid, formamide and dimethylformamide

MF is a colorless and toxic compound with ethereal smell. At normal pressures and
temperatures, it is in liquid state with high vapor pressure (normal boiling point is 31.5 °c). MF
is commercially produced by BASF with 96% selectivity [52]. Investigating new processes to
 ACCEPTED MANUSCRIPT

produce MF with considerations of energy enhancement is still challenging. Dehydrogenation of

methanol and in-situ hydrogen production seems an efficient way to produce MF [47, 48, 53].

1.6. Hydrogen: Non-pollutant Energy Carrier

PT
As previously mentioned, environmental problems of burning hydrocarbons, e.g. global
warming, NOx and SOx emissions, climate change, defects in biodiversity and finite sources of
hydrocarbons are incentives of researchers and industries to seek for new clean, renewable and

RI
sustainable energy alternatives [16, 54].

Hydrogen is considered as a candidate for energy delivery. Hydrogen features e.g.

SC
environmentally friendly nature and efficient ignition characteristics are reasons of ascending
demand for hydrogen. Besides, hydrogen is a promising energy carrier due to the following
advantages [55, 56]:

•
U
High heating value comparing hydrocarbons
AN
• Non-pollutant and non-toxic emissions
• Fuel for vehicle engines and power plants
• Raw material in diverse chemical, petrochemical and refining processes
M

• Renewability and sustainability

Hydrogen is commercially produced by steam or dry reforming of hydrocarbons, coal and

D

biomass gasification, partial oxidation of methane and heavy oils, natural gas catalytic
decomposition, thermal decomposition of H2S, decomposition of methanol, thermolysis, and
TE

electrolysis. However, thermolysis and electrolysis are not cost effective in large-scale
applications [57-59]. To avoid difficulties of storage and transportation of hydrogen, on-board
hydrogen production and use is commercially of a great interest. Dehydrogenation of
EP

hydrocarbons and in-situ hydrogen separation using hydrogen perm-selective membranes is

currently state of the art [17, 18].
C

1.7. Literature Review

AC

Over the years, different researchers have studied single-step and two-step synthesis of DME in
heat exchanger reactors. Vakili et al. proposed a heat exchanger reactor for single-step synthesis
of DME and production of benzene hydrogen via dehydrogenation of cyclohexane. Smaller
dimensions of reactor with enhanced conversion, production of valuable by products and low
temperature of the products are mentioned as improvements of their work. In further work, they
suggested some strategies e.g. changing flow directions, utilizing a membrane, and determining
operating conditions by an optimization procedure to boost hydrogen and DME production [60-
62]. Farniaei et al. and Rahimpour et al. introduced thermally double-coupling concept for
 ACCEPTED MANUSCRIPT

simultaneous production of methanol, DME, and hydrogen. Increasing methanol, DME and
hydrogen production rates are advancements of their system [16, 18]. Farsi et al. and Samimi et
al. proposed different thermally coupling reactors for production of DME from methanol inside
different endothermic reactions such as cyclohexane hydrogenation [20, 21] and methyl
cyclohexane dehydrogenation [63]. Goosheneshin et al. studied production and utilization of
methanol for MF production in a looped heat exchanger reactor. Produced methanol in the

PT
exothermic side was recycled and reused for MF production in the endothermic side [22].
Recently, Bakhtyari et al. presented a heat exchanger reactor for simultaneous production of
DME, MF and hydrogen from methanol. Increasing methanol conversion in both sides and

RI
hydrogen production rate in the endothermic side were the advancements of their work [64].

SC
1.8. Objectives

The present study aims to propose an integrated thermally coupled double membrane reactor for

U
simultaneous production of DME, MF, hydrogen, and water vapor from methanol. Methanol
dehydration to DME is an exothermic reaction being coupled with methanol dehydrogenation to
AN
MF, which is an endothermic reaction. Heat of methanol dehydration is transferred to the
methanol dehydrogenation side through the reactor wall and it makes the endothermic reaction
proceed. A hydroxy sodalite (H-SOD) membrane with high water perm-selectivity in a gaseous
M

mixture is utilized to separate water vapor in the exothermic side. Besides, produced hydrogen in
the endothermic side is separated utilizing a hydrogen perm-selective Pd/Ag membrane. Because
D

of water separation from the exothermic side, methanol conversion, and DME production rate is
boosted and water vapor can be recovered and reused. Separation of hydrogen from the
TE

endothermic side results in enhanced methanol conversion to MF hydrogen recovery as a

beneficial by product. Simultaneous reaction and separation in this reactor can decrease size and
cost of separation steps and consequently energy consumption. To show the performance of the
proposed heat exchanger reactor, a mathematical model based on energy and mass balance is
EP

developed. Genetic algorithm (GA) method as a powerful optimization technique is applied to

find the optimum values of the operating parameters and then a system with better performance
and efficiency.
C
AC

2. Process Description

2.1. Conventional DME Reactor (CDMR)

DME production by methanol dehydration is performed in an adiabatic fixed bed reactor filled
with γ-Al2O3 catalyst. CDMR is schematically depicted in Figure 1. As can be seen, hot product
streams are used to preheat the feed stream via a shell and tube heat exchanger. The operating
 ACCEPTED MANUSCRIPT

conditions, reactor specifications, and catalyst characteristics for the typical methanol
dehydration reactor in a commercial plant are summarized in Table 1.

2.2. Thermally Coupled Heat Exchanger Reactor (TCHR)

Figure 2 shows a conceptual schematic of a two concentric TCHR configuration working in co-
current mode. The tube side is filled with γ-Al2O3 catalyst for methanol dehydration to DME

PT
while the shell side is filled with copper-chromite catalyst for methanol dehydrogenation to MF.
Thereupon, the dehydrogenation reaction performs as a coolant medium to removing part of the
generated heat in the tube side. As a result, dehydration reaction proceeds and drives the

RI
dehydrogenation reaction. Typical operating conditions for the endothermic side and catalyst
characteristics are summarized in Table 2. It should be mentioned that TCHR is considered as a

SC
“base case” for evaluation of the optimized thermally coupled membrane reactor performance
(OTCMHR).

2.3. Thermally Coupled Double Membrane Heat Exchanger Reactor (TCMHR)

U
Figure 3 shows a conceptual schematic of a three concentric TCMHR configuration with co-
AN
current sweep gases. Outer permeation surface is equipped with a Pd/Ag membrane for in-situ
separation of hydrogen in the endothermic side while in the inner permeation surface, an H-SOD
with α-Al2O3 support membrane is devised for separation of water from the exothermic side. The
M

Results of removing water from the exothermic side are shifting the dehydration reaction toward
products, more DME production, more heat generation, and less water in the output stream.
Removing hydrogen from the endothermic side results in higher methanol conversion, more MF
D

production, higher heat removal from the exothermic side and less hydrogen in the output
stream. Inlet sweep gases contain an inert carrier such as argon. Optimum temperatures,
TE

pressures, and flow rates of inlet sweep gases are determined by optimization procedure.
Dimensions of TCHR and TCMHR are given in Table 3.
EP

3. Reaction Schemes and Kinetics

3.1. Methanol to DME Reaction

C

Methanol is dehydrated to DME over γ-Al2O3 catalyst by an exothermic and equilibrium

reaction:
AC

kJ
2CH OH ⇄ CH OCH + H O ∆H = −23.4 (1)
mol
Different authors focused on the kinetic modeling of this reaction and extracting reaction rate
[65-69]. Bai et al. proposed the following rate equation and validated the model against data of a
DME plant [68]:
 ACCEPTED MANUSCRIPT

"# $ 
 =  
 !1 − ' (2)
%& 


Where fi is the fugacity of each component. Followings are the rate constant k and the
equilibrium constant Keq [68, 69]:

78072.55
 = 1457.024 +,- .− 2 (3)

PT
01
4019
ln %& = −26.64 + 3.707 ln 1 + − 2.783 × 107 1 + 3.8 × 1078 1  + 6.561
1

RI
× 109 1  (4)

SC
It should be mentioned that the fugacities of components are calculated using Peng-Robinson
equation of state (PR EOS). PR EOS and critical properties of the components are presented in
Appendix A.

U
3.2. Methanol to MF
AN
The equilibrium endothermic reaction of methanol dehydrogenation to MF is as following:

kJ
2CH OH ⇄ CH OCHO + 2H ∆H = 98.9 (5)
mol
M

Huang et al. investigated kinetics of methanol dehydrogenation to MF over a commercial

copper-chromite catalyst and extracted the following rate expression [22, 64, 70]:
D


TE

− %:
;.< 7;.<
=>  $
= (6)
7;.< 
?1 + %: %9
;.< 7;.<
=>  $ + %
7:
=> = + % %
7;.< 7;.< ;.<
=> = + %9
7;.<
$ @
EP

Where fi is the fugacity of each component calculated using PR EOS and k2=0.43 mol.gr-1.hr-1,
K21=0.48×10-5 Pa-1, K22=3.66×103 Pa, K23=0.6×103 Pa and K24=30×103 Pa.
C
AC

4. Mathematical Modeling and Optimization

4.1. Governing Equations

A mathematical model based on heat and mass balance equations and following assumptions is
developed to describe the performance of the proposed configurations:

• Reaction on the surface of catalyst particles

• Negligible intra-particle heat and mass diffusion in catalyst pellet (i.e. homogeneous
model)
 ACCEPTED MANUSCRIPT

• Negligible radial gradient in all of sides and plug flow pattern (i.e. one-dimensional
model)
• Steady state operation
• Negligible mass and heat dispersion due to high gas velocity (i.e. plug flow model)
• Constant bed porosity in both sides (i.e. bed symmetry)
• Adiabatic operation (no heat loss from inside and outside of the reactor)

PT
• Non-ideal gaseous mixtures
• Ergun equation for pressure profiles in the exothermic and endothermic sides
•

RI
Negligible pressure drop in the permeation sides

The obtained sets of ordinary differential equations (ODEs) with associated boundary conditions

SC
are summarized in Table 4. Auxiliary correlations to determine thermophysical properties of the
reacting mixtures and heat transfer coefficients are presented in Tables B1-B4 in Appendix B.

4.2. Water Removal via an H-SOD membrane

U
AN
H-SOD membranes with sodalite cages are useful in water removal from streams because of high
water permeation rate and thermal and mechanical stability at high temperatures and pressures.
H-SOD membranes are promising in separation of small molecules such as water (kinetic
M

diameter of water molecule is 2.68 A˚). Rohde et al. declared that it is highly selective to water
(almost 100%) even in the mixtures of hydrogen and carbon monoxide [28]. Hence, in-situ water
D

removal in the DME synthesis side using H-SOD membranes is of a great interest. Samimi et al.
[63] and Rahimpour et al. [71] studied application of H-SOD membranes for water separation in
TE

different dehydration reactions and streams containing water. Water permeation rate as a
function of partial pressure of water in both sides of the membrane is obtained using the
following equation [28-30, 63]:

B$ CD
EP

A$  = ?G$ ,: − G$ ,9 @ (15)

EF

Where, As is the surface area, B$  is permeance of water (1-10×10-7 mol.s-1.m-2.Pa-1) and Vr is
C

reactor volume. G$ ,: and G$ ,9 are partial pressures of water in both sides of the membrane
AC

(i.e. exothermic and permeation sides).

4.3. Hydrogen Recovery via a Pd/Ag Membrane

Palladium membranes are widely used for separation of hydrogen from hydrogen mixtures.
Hydrogen perm-selectivity of the membrane surface is enhanced using promoters such as silver.
[26, 27, 72]. Hereupon, a continuous thin layer of Pd-Ag alloy is equipped on the surface of a
thermo-stable support and subjected to the hydrogen mixtures. The hydrogen permeation rate
 ACCEPTED MANUSCRIPT

through the Pd-Ag membrane is assumed to follow the Sievert’s law in which Q0 =1.65×10-5
mol.m-1.s-1.kPa-0.5 and EJ$ = 15.7 kJ. mol7: [15, 73]:

B; L
A$ = +,- .− $ 2 !MG$,  − MG$,  ' (16)
K$ 01

G$,  and G$, are partial pressures of hydrogen in both sides of the membrane (i.e. endothermic

PT
and permeation sides) and K$ is the membrane thickness.

RI
5. Numerical Solution and Optimization

5.1 Numerical Solution

SC
The developed mathematical model contains ODEs, boundary conditions, auxiliary equations for
physical properties and fugacity equations of PR EOS. Backward finite difference method was
applied to convert the ODEs to algebraic equations. Then, a set of nonlinear algebraic equations

U
is obtained. Length of the reactor was divided to 100 grids. Hence, a stable solution with low
numerical error is guaranteed. The algebraic equations in each grid were solved in MATLAB
AN
media. The obtained set of nonlinear algebraic equations in each grid was solved using Gauss-
Newton method. The initial guess of each grid was the output of the previous grid. Simulation
results obtained in the present study are compared with output data of a commercial DME plant
M

in Table 5. Based on the obtained relative errors, the developed model is acceptable.

5.2 Optimization
D

Optimum operating conditions such as flow rate, temperature, and pressure of feed streams are
TE

determined to achieve higher product quality and process efficiency [74]. In fact, optimization
procedures are primarily aimed to find the optimum values of decision variables affecting the
process efficiency and then lead to maxima or minima in the appropriate objective functions.
EP

Modern optimization techniques e.g. differential evolution (DE), genetic algorithms (GAs),
simulated annealing (SA) and evolution strategies (ESs) [75-79] are running the art to finding the
global optima in nonlinear problems.
C

The present study seeks to find simultaneous maximization of methanol conversion to DME in
the exothermic side, methanol conversion to MF in the endothermic side and hydrogen recovery
AC

yield from the endothermic side. The aforementioned parameters are defined as following:
PQRST,UV,W 7PQRST,XYZ,W
Methanol conversion: N,O = PQRST,UV,W
(17)

PT$ ,>
Hydrogen recovery yield: [$ = P (18)
QRST,UV,$

In this regard, sum of methanol conversions in both sides and hydrogen recovery yield is
considered as the objective function:
 ACCEPTED MANUSCRIPT

A = N,: + N, + [$ (19)

Thirteen decision variables are manipulated during optimization procedure. Selected decision
variables and their feasible bounds are summarized in Table 6. A genetic algorithm is applied for
maximization of the objective function. GA that is simulated form of natural evolution is
considered as a powerful technique between stochastic methods. Depending on the degree of

PT
non-linearity of the equations and initial values, most of the gradient-based optimization methods
have the potential to trap in the local optima. Hence, they do not warrant that the global optima
are found. GA that is simulated form of natural evolution is considered as a powerful technique

RI
between stochastic methods. This method is useful for global search in both constrained and
unconstrained systems. GA is made up of components that are inspiration from evolutionary
biologic systems i.e. inherence, mutation, selection and crossover. An initial population with

SC
constant size during the procedure and consisted of random series of genes is created. Fitness
function is a criterion for evaluating the individuals (genomes) in the population at each
generation. In fact, selection of individuals in the formation of new population is according to

U
their fitness. Selection, crossover, and mutation are mechanisms of producing new individuals.
Crossover is the mechanism of producing a new individual (child) by exchanging parts of other
AN
two picked individuals (parents). Mutation is the mechanism of changing genes of parents in a
random manner. Implication of the mentioned mechanisms in a successive generation procedure
is to drive the population to the global optima. As a whole, GA consists of the following steps:
M

• Random initial population creation

D

• Testing the fitness function

• Employing genetic operators (i.e. selection, crossover, and mutation)

TE

• Attaining the criteria to stop the search

EP

This method is independent of initial guess and function derivatives. Thus, it can use the
information of other regions without trapping in the local optima. On the one hand, by finding
better fitness function values in the other regions of the solution domain, the population is forced
to move away from the local optima. More details on basic components of GA such as gene,
C

chromosome, population, selection, and crossover could be found elsewhere [77, 79, 80].
AC

Obtained optimum values of the decision variables are tabulated in Table 7.

6. Results and Discussion

The observed thermal and molar behaviors of TCHR and TCMHR are discussed in the following
section. TCHR with the characteristics given in Table 1-3 is considered as a “base case”. Hence,
performance of OTCMHR is compared with the base case.

6.1. OTCMHR
 ACCEPTED MANUSCRIPT

In this section, molar and thermal behavior of OTCMHR is discussed. Figure 4 and 5 shows
material profile in both reaction sides. As shown in Figure 4, a sharp change in the upper section
of the exothermic side is observed. Flow of water reaches a maximum and then decreases. The
maximum point of water flow rate coincides with a change in trends of MeOH and DME flow
rates. In this point, rate of water permeation to the inner permeation side overcomes that of water
production. Then a sharp decrease in water flow rate beside slight increase in DME flow rate and

PT
slight decrease in MeOH flow rate is observed. As a result, elimination of water from the
reaction side leads to more methanol conversion and DME production.

RI
MeOH, MF and hydrogen flow rates in the endothermic side of the OTCMHR are shown in
Figure 5. A gradual decrease in MeOH flow rate and gradual increase in MF flow rate is
observed while trend of hydrogen is different. As can be seen, hydrogen flow rate immediately

SC
increases in the upper section due to reaction. Separation of hydrogen using outer membrane
causes hydrogen reduction in the endothermic side. Hence, produced hydrogen is totally
separated in the whole length of reactor. Separation of hydrogen shifts the reaction toward more

U
MeOH consumption, more MF production, and then higher purity of output stream. MeOH to
MF reaction reaches chemical equilibrium in the lower section of reactor because MeOH is
AN
totally converted. MeOH conversion profiles in both reaction sides are compared in Figure 6.
Obviously, total MeOH conversion in both sides is obtained utilizing inner and outer
membranes.
M

Axial temperature profile in both reaction sides and permeation sides is presented in Figure 7. A
hot point in the upper section of both reaction sides is obtained. Thermal pattern of inner
D

permeation side (which is inside the exothermic side) is close to the exothermic side while
thermal pattern of outer permeation side (which is inside the endothermic side) is a little different
TE

with that of endothermic side. All the streams tend to become isotherm in the output of the
reactor. Flow direction in the heat exchanger reactors is a determinant factor in the temperature
profile. Accordingly, co-current flow in this configuration is major contributor to the distribution
EP

of temperature. On the other hand, isothermal flows at the reactor outlet is due to probable
established chemical equilibrium in both sides and terminating heat generation and consumption
in the reactor with co-current streams.
C

Axial pressure profiles in exothermic and endothermic sides are compared in Figure 8. A nearly
AC

linear pattern is obtained for both sides. Total pressure of the exothermic sides shows 0.32 bar
(% 1.68) decrease while the endothermic side shows 0.72 (% 14.25) bar pressure drop. Different
pressure drops in two sides is a result of different dimensions, different physical properties of
components (e.g. viscosity) and different permeation rates from reaction sides. Besides,
differences in the stoichiometry of the reactions and may affect the pressure
distributions.
Figure 9 and 10 show performance of two permeation sides. Water mole fraction in the inner
permeation side and hydrogen mole fraction in the outer permeation side are shown in Figure 9.
 ACCEPTED MANUSCRIPT

Besides, hydrogen and water recovery in both sides are given in Figure 10. Hydrogen begins to
permeate right reactor inlet while water permeation does not begins from the reactor inlet. On the
other hand, profiles of hydrogen and water in the permeation sides as well as reaction sides are
affected by contest of permeation rates and component production.

6.2. OTCMHR vs. TCHR

PT
Performance of OTCMHR and TCHR are compared in this section. Axial temperature profile of
OTCMHR and TCHR are shown in Figure 11. Product streams of TCHR leave the reactor in
higher temperature. However, temperatures of feed streams are lower in TCHR than OTCMHR.

RI
Besides, the hot point TCHR has higher temperature and occurs in a far distance from reactor
inlet. This may be due to presence of permeation sides. In fact, a part of heat is transferred to the

SC
permeation side leading to lower output temperature. Lower temperature of streams eliminates
the probability of catalyst sintering. MeOH conversion in both endothermic and exothermic sides
of OTCMHR and TCHR are compared in Figure 12. Obviously, higher conversion is obtained in

U
OTCMHR. Such an achievement is due to lateral expelling of products and making the reactions
proceed in forward direction. Conversion enhancement in the endothermic side is more than
AN
exothermic side. Figure 13 shows products flow rates in OTCMHR and TCHR. Production of
both DME and MF is boosted utilizing the selective-membranes. Output methanol conversion of
exothermic and endothermic sides in OTCMHR and TCHR are given in Figure 14. MeOH to
M

DME conversion shows % 10 increase while MeOH to MF conversion shows % 38 increase.

Generally, higher flow rate of desired products and lower output temperature are observed by
utilizing the double membrane reactor.
D

7. Conclusion
TE

DME production by methanol dehydration and MF/hydrogen production by methanol

dehydrogenation in a catalytic heat-exchanger double membrane reactor was studied
numerically. The proposed configuration contains a shell and tube filled with catalysts of
EP

methanol dehydration and dehydrogenation. Besides, two membranes were equipped on the
inner and outer surfaces of the reactor. Methanol dehydration plays the role of heat source for
driving the endothermic reaction i.e. methanol dehydrogenation. In addition, a water perm-
C

selective membrane and a hydrogen perm-selective membrane are running the art of in-situ
separation of water and hydrogen from exothermic and endothermic sides, respectively.
AC

Optimum parameters of the configuration were determined in an optimization procedure using

genetic algorithm. The proposed configuration working under optimum conditions provides
significant improvement comparing the thermally coupled reactor proposed in the previous
studies. Following achievements are pointed out:

• Reducing energy consumption by direct autothermality in the reactor and in-situ energy
efficient separations
• Producing various value-added products from only one feedstock
 ACCEPTED MANUSCRIPT

• Obtaining higher DME production rate and purity in the output stream
• Obtaining higher MF production rate and purity in the output stream
• Recovery of pure hydrogen and water vapor
• Products with lower temperature and reducing risk of catalyst sintering

Utilizing OTCMHR configuration promotes methanol conversion to DME to % 95.1 and

PT
methanol conversion to MF to % 99.6. Besides, comparing with TCHR, % 10 increase in
methanol conversion to DME and % 38 increase in methanol conversion to MF was achieved
using the optimized thermally coupled double membrane heat exchanger reactor.

RI
SC
Acknowledgements

The authors are grateful to the Shiraz University for supporting this research.

U
AN
Appendix A: Peng-Robinson Equation of State

RT ab α
P= −
v − b v(v + b) + b(v − b)
(A1)
M

R Tb
ab = 0.457235
Pb
(A2)
RTb
b = 0.077796
D

Pb
(A3)
m = 0.37464 + 1.5422ω − 0.26992ω
TE

(A4)

α = e1 + m?1 − fTg @h (A5)
no no

(ab α)i = j j yl ym fal am αl αm (A6)

EP

lp: mp:
no

bi = j yl bl (A7)
C

lp:
(ab α)i P
A=
(RT)
AC

(A8)
(ab α)i P
A=
(RT)
(A9)
bi P
B=
RT
(A10)
Z + (B − 1)Z + (A − 2B − 3B )Z − (AB − B − B  ) = 0 (A11)
fl bl A 2 ∑ xm fal am αl αm bl Z + ?1 + √2@B
ln = (Z − 1) − ln(Z − B) − v − y ln v y
yl P bi 2√2B (ab α)i bi Z + ?1 − √2@B
(A12)
 ACCEPTED MANUSCRIPT

Appendix B: Correlations of Physical Properties and Heat Transfer Coefficients

Table B1: correlation of specific heat capacity of the components and mixture [64]
} }
z~ ⁄ z† ⁄
(J.kmol-1.K-1) z{ = z| + z} v‚ƒ„(z~ ⁄)y + z v‡ˆ„(z† ⁄)y
Heat Capacity
(B1)

C1×10-5 C2×10-5 C3×10-3 C4×10-5 C5

PT
Component

MeOH 0.3925 0.879 1.9165 0.5365 896.7

DME 0.5148 1.442 1.6034 0.7747 725.4

RI
H2O 0.3336 0.2679 2.6105 0.089 1169
MF 0.506 1.219 1.637 0.894 743
H2 0.2762 0.0956 2.466 0.0376 567.6

SC
Ar 0.2079 - - - -
=

Mixture ‰Š,‹Œ = j ŽŒ ‰Š,Œ (B2)

Œp:

U
AN
Table B2: correlation of viscosity of the components and mixture [64]

”
 = ‘’{ .“ + + z •ˆ–  + —˜ 2

Viscosity (Pa.s) (B3)
M

Component A B C D E MW (gr.mol-1)
D

MeOH -7.288 1065.3 -0.6657 0 0 32

DME -10.62 448.99 8.396×10-5 0 0 46
-5.349×10-29 10 18
TE

H2O -51.964 3670.6 5.7331

“”
=
| + z7| + —˜
Viscosity (Pa.s) (B4)
EP

A×106 B C D E MW (gr.mol-1)

MF 1.6672 0.47948 516.01 0 0 60.05

C

H2 1.0009 0.47949 290.01 0 0 2

Ar 4.71912 0.40949 390.01 0 0 39.95
∑=
Œp: ŽŒ ™Œ š›Œ
;.<
AC

™‹Œ =
∑=
Œp: ŽŒ š›Œ
;.<
Mixture (B5)
 ACCEPTED MANUSCRIPT

Table B3: correlation of thermal conductivity of the components and mixture [64]

Thermal Conductivity (J/m.K.sec) œž = “ + ” + z} + —~ (B6)

Component A×102 B×105 C×107 D×1011

MeOH -1.8683 8.7959 0.82324 -2.8949

DME -0.83897 5.8126 1.1178 -4.0608

PT
H2O 0.56199 1.5699 1.0106 -2.4282
MF 0.085 0.60228 1.2439 0
H2 1.0979 66.411 -3.4378 9.7283

RI
Ar -0.30142 6.8128 -0.3182 0.6854
=

Mixture Ÿ ,‹Œ = j ŽŒ Œ (B7)

SC
Œp:

Table B4: Heat transfer coefficient of reaction side, membrane side and overall heat

U
transfer coefficient

ℎ ‰Š ™ ⁄ 0.4548 ¢£¥Š 7;.9;8

AN
. 2 = . 2
¢£‰Š Ÿ ¤ ™
Reaction sides [81] (B8)

Ÿ
ℎ = 0.0214 G ;.9 (0+";. − 100)
M

¦
(B9)
Membrane Sides [82]
1 1 CŒ ln(¦¨ ⁄¦Œ ) CŒ 1
= + +
§ ℎŒ 2©ª« C ¨ ℎ¨
Overall coefficient (B10)
D
TE

It should be mentioned that in equation B8, Cp is used in mass units (J.kg-1.K-1) therefore the Cp
calculated from equation B2 must be multiplied by average molecular weight.
C EP
AC
 ACCEPTED MANUSCRIPT

Nomenclature
Ac cross-section area, m2
Ai inside area of inner tube, m2
Ao outside area of inner tube, m2
c1, c2, c3, c4, c5 constants of the specific heat correlation
Cp specific heat of the gas at constant pressure, J.mol−1.K
dp particle diameter, m
Di tube inside diameter, m

PT
Do tube outside diameter, m
Dj diameter of each side, m
EH2 activation energy in hydrogen permeation rate model, kJ.mol-1
fi fugacity of component i, Pa

RI
Fi molar flow rate of the components, mol.s−1
Ft total molar flow rate, mol.s−1
hi heat transfer coefficient between fluid phase and reactor wall in the exothermic side, W.m−2.K−1
ho heat transfer coefficient between fluid phase and reactor wall in the endothermic side, W.m−2.K−1

SC
∆Hf,i enthalpy of formation of component i, J.mol−1
JH2 hydrogen permeation flux, mol.m-2.s-1
JH2O water permeation flux, mol.m-3.s-1
k rate constant of methanol dehydration reaction, mol.g-1.hr-1.Pa-2

U
k2 rate constant of methanol dehydrogenation reaction, mol.gr-1.hr-1
kth thermal conductivity of the components, W.m−1.K−1
K21 rate constant of methanol dehydrogenation reaction, Pa-1
AN
K22 rate constant of methanol dehydrogenation reaction, Pa
K23 rate constant of methanol dehydrogenation reaction, Pa
K24 rate constant of methanol dehydrogenation reaction, Pa
Keq equilibrium constant for the methanol dehydration reaction, dimensionless
M

Kw thermal conductivity of the reactor wall, W.m−1.K−1

L reactor length, m
MW molecular weight of component i, g.mol−1
NC number of components (NC=3 for the exothermic side and NC=4 for the endothermic side)
D

P total pressure, Pa
Pi partial pressure of component i, Pa
TE

Q total volumetric flow rate, m3.s−1

Q0 constant in hydrogen permeation rate model, mol.m-1.s-1.kPa-0.5
rMeOH,1 rate of methanol dehydration reaction, mol.g-1.hr-1
rMeOH,2 rate of methanol dehydrogenation reaction, mol.gr-1.hr-1
R universal gas constant, J.mol−1.K−1
EP

Rp particle radius, m
T temperature, K
u superficial velocity of the fluid phase, m.s−1
U1-2 overall heat transfer coefficient between exothermic and endothermic sides, W.m−2.K−1
C

U2-3 overall heat transfer coefficient between endothermic and permeation sides, W.m−2.K−1
XMeOH methanol conversion, dimensionless
AC

yi mole fraction of component i

YH2 hydrogen recovery yield
z axial reactor coordinate, m
Greek Letters
ε porosity (void fraction) of the catalytic bed
µ viscosity of the fluid phase, kg.m−1.s−1(Pa.s)
ϕs sphericity factor of the catalyst particles
ρ density of the fluid phase, kg.m−3
ρB density of catalytic bed (ρB= ρ(1- ε)), kg.m−3
η catalyst effectiveness factor
β β=0 for exothermic side and β=1 for endothermic side
φ φ=1 for hydrogen and φ=0 for argon
 ACCEPTED MANUSCRIPT

δH2 membrane thickness, m

Superscripts
i chemical species
j reactor side (j=1 for exothermic side, j=2 for endothermic side, j=3 for permeation side)
Subscripts
g in bulk gas phase

PT
References

RI
[1] T. A. Semelsberger, R. L. Borup, H. L. Greene, Dimethyl ether (DME) as an alternative fuel. Journal of Power
Sources 156 (2006) 497-511.

[2] M. Ball, M. Wietschel, The future of hydrogen - opportunities and challenges. Int. J. Hydrogen Energy 34 (2009)

SC
615-627.

[3] M. Ross, Energy Consumption by Industry. Ann. Rev. Energy 6 (1981) 379-416.

U
[4] H. Axelsson, A. Asblad, T. Berntsson, A new methodology for greenhouse gas reduction in industry through
improved heat exchanging and/or integration of combined heat and power, Applied Thermal Engineering 19 (1999)
707-731.
AN
[5] M. Ebrahim, A. Kawari, Pinch technology: an efficient tool for chemical-plant energy and capital-cost saving,
Applied Energy 65 (2000) 45-49.
M

[6] Z. Jegla, P. Stehlı´k, J. Kohoutek, Plant energy saving through efficient retrofit of furnaces, Applied Thermal
Engineering (2000) 1545-1560.

[7] F. Friedler, Process integration, modeling and optimization for energy saving and pollution reduction. Appl.
D

Therm. Eng. 30 (2010) 2270-2280.

[8] R. Smith, State of the art in process integration. Appl. Therm. Eng. 20 (2000) 1337-1345.
TE

[9] A. P. Rossiter, Succeeding in process integration. In: Process industries expo user conference. 2003.

[10] M. R. Rahimpour, M. R. Dehnavi, F. Allahgholipour, D. Iranshahi, S. M. Jokar, Assessment and comparison of

EP

different catalytic coupling exothermic and endothermic reactions: A review. Applied Energy 99 (2012) 496-512.

[11] M.R. Rahimpour, A. M. Bahmanpour, Optimization of hydrogen production via coupling of the Fischer-
Tropsch synthesis reaction and dehydrogenation of cyclohexane in GTL technology. Appl. Energy 88 (2011) 2027-
2036.
C

[12] M. R. Rahimpour, M. H. Khademi, A.M. Bahmanpour, A comparison of conventional and optimized thermally
AC

coupled reactors for Fischer–Tropsch synthesis in GTL technology. Chem. Eng. Sci. 65 (2010) 6206-6214.

[13] M .R. Rahimpour, Enhancement of hydrogen production in a novel fluidized bed membrane reactor for naphtha
reforming. Int. J. Hydrogen Energy 34 (2009) 2235–2251.

[14] D. Iranshahi, A. M. Bahmanpour, E. Pourazadi, M. R. Rahimpour, Mathematical modeling of a multi-stage

naphtha reforming process using novel thermally coupled recuperative reactors to enhance aromatic production. Int.
J. Hydrogen Energy 35 (2010) 10984-10993.

[15] H. Rahnama, M. Farniaei, M. Abbasi, M. R. Rahimpour, Modeling of synthesis gas and hydrogen production in
a thermally coupling of steam and tri-reforming of methane with membranes. Journal of Industrial and Engineering
Chemistry 20 (2014) 1779-1792.
 ACCEPTED MANUSCRIPT

[16] M. R. Rahimpour, M. Farniaei, M. Abbasi, J. Javanmardi, S. Kabiri, Comparative study on simultaneous

production of methanol, hydrogen, and DME using a novel integrated thermally double-coupled reactor. Energy
Fuels 27 (2013) 1982-1993.

[17] M. R. Rahimpour, E. Pourazadi, A comparison of hydrogen and methanol production in a thermally coupled
membrane reactor for co-current and counter-current flows. Int. J. Energy Res. 35 (2011) 863-882.

[18] M. Farniaei, M. Abbasi, A. Rasoolzadeh, M. R. Rahimpour, Enhancement of methanol, DME and hydrogen

PT
production via employing hydrogen perm-selective membranes in a novel integrated thermally double-coupled two-
membrane reactor. J. Nat. Gas Sci. Eng. 14 (2013) 158-173.

[19] M. Bayat, M. R. Rahimpour, Production of ultrapure hydrogen via utilizing fluidization concept from coupling
of methanol and benzene synthesis in a hydrogen perm-selective membrane reactor. Int. J. Hydrogen Energy 36

RI
(2011) 6616-6627.

[20] M. Farsi, M. H. Khademi, A. Jahanmiri, M. R. Rahimpour, Novel recuperative configuration for coupling of

SC
methanol dehydration to dimethyl ether with cyclohexane dehydrogenation to benzene. Ind. Eng. Chem. Res. 49
(2010) 4633-4643.

[21] M. H. Khademi, M. Farsi, M. R. Rahimpour, A. Jahanmiri, DME synthesis and cyclohexane dehydrogenation
reaction in an optimized thermally coupled reactor. Chem. Eng. and Proc. 50 (2011) 113-123.

U
[22] A. Goosheneshin, R. Maleki, D. Iranshahi, M. R. Rahimpour, A. Jahanmiri, Simultaneous production and
AN
utilization of methanol for methyl formate synthesis in a looped heat exchanger reactor configuration. Journal of
Natural Gas Chemistry 21 (2012) 661-672.

[23] M. R. Rahimpour, Membrane reactors for biodiesel production and processing. Membrane Reactors for Energy
Applications and Basic Chemical Production. http://dx.doi.org/10.1016/B978-1-78242-223-5.00010-8.
M

[24] M. R. Rahimpour, Butene oligomerization, phenol synthesis from benzene, butane partial oxidation, and other
reactions carried out in membrane reactors. Membrane Reactors for Energy Applications and Basic Chemical
Production. http://dx.doi.org/10.1016/B978-1-78242-223-5.00021-2.
D

[25] A. K. Jana, Heat integrated distillation operation. Appl. Energy 87(2010) 1477-1494.
TE

[26] J. Shu, B. P. A. Grandjean, A. V. Neste, S. Kaliaguine, Catalytic palladium-based membrane reactors: a review.
Canadian Journal of Chemical Engineering 69 (1991) 1036-1060.

[27] S. Uemiya, T. Matsuda, E. Kikuchi, Hydrogen permeable palladium-alloy membrane supported on porous
EP

ceramics. Journal of Membrane Science 56 (1991) 315-325.

[28] M. P. Rohde , G. Schaub , S. Khajavi, J. C. Jansen, F. Kapteijn, Fischere Tropsch synthesis with in situ H2O
removal - directions of membrane development. Microp Mesop Mater 115 (2008) 123-136.
C

[29] F. A. N. Fernandes, A. B. Soares Jr., Methane steam reforming modeling in a palladium membrane reactor.
Fuel 85 (2006) 569-573.
AC

[30] S. Khajavi, J. C. Jansen, F. Kapteijn, Application of a sodalite membrane reactor in esterification coupling
reaction and separation. Catal. Today 156 (2010) 132-139.

[31] R. P. Omorjan, R. N. Paunovic, M. N. Tekic, Applicability of a double-membrane reactor for thermal

decomposition of water: a computer analysis. Journal of Membrane Science 154 (1999) 273-280.

[32] Z. Azizi, M. Rezaeimanesh, T. Tohidiana, M.R. Rahimpour, Dimethyl ether: A review of technologies and
production challenges. Chem. Eng. and Proc.: Process Int. 82 (2014) 150-172.

[33] L. Zhang, J. Wang, P. Wu, Z. Hou, J. Fei, X. Zheng, Synthesis of dimethyl ether via methanol dehydration over
combined Al2O3-HZSM-5 solid-acids. Chin. J. Catal. 31 (2010) 987-992.
 ACCEPTED MANUSCRIPT

[34] F. Raoof, M. Taghizadeh, A. Eliassi, F. Yaripour, Effects of temperature and feed composition on catalytic
dehydration of methanol to dimethyl ether over-alumina. Fuel 87 (2008) 2967-2971.

[35] W. H. Chen, B. J. Lin, H. M. Lee, M. H. Huang, One-step synthesis of dimethylether from the gas mixture
containing CO2 with high space velocity. Appl. Energy 98 (2012) 92-101.

[36] R. Ladera, E. Finocchio, S. Rojas, J. L. G. Fierro, M. Ojeda, Supported niobium catalysts for methanol
dehydration to dimethyl ether: FTIR studies of acid properties. Catal. Today 192 (2012) 136-143.

PT
[37] K. C. Tokay, T. Dogu, G. Dogu, Dimethyl ether synthesis over alumina based catalysts. Chem. Eng. J. 184
(2012) 278-285.

[38] F. Hayer, H. Bakhtiary Davijany, R. Myrstad, A. Holmen, P. Pfeifer, H. J. Venvik, Characteristics of integrated

RI
micro packed bed reactor-heat exchanger configurations in the direct synthesis of dimethyl ether. Chem. Eng.
Process. 70 (2013) 77-85.

SC
[39] K. Faungnawakij, Y. Tanaka, N. Shimoda, T. Fukunaga, R. Kikuchi, K. Eguchi, Hydrogen production from
dimethyl ether steam reforming over composite catalysts of copper ferrite spinel and alumina. Appl. Catal. B:
Environ. 74 (2007) 144-151.

[40] E. Ren, J. F. Wang, H. S. Li, Direct mass production technique of dimethyl ether from synthesis gas in a

U
circulating slurry bed reactor. Stud. Surf. Sci. Catal. 159 (2006) 489-492.
AN
[41] M. Gadek, R. Kubica, E. Jedrysik, Production of methanol and dimethyl ether from biomass-derived syngas: a
comparison of the different synthesis pathways by means of flowsheet simulation. in: The 23rd European
Symposium on Computer Aided Process Engineering (ESCAPE 23) 2013.

[42] Y. Hu, Z, Nie, D. Fang, Simulation and model design of pipe-shell reactor for the direct synthesis of dimethyl
M

ether from syngas. Journal of Natural Gas Chemistry 17 (2008)195-200.

[43] W.Z. Lu, L. H. Teng, W. D. Xiao, Simulation and experiment study of Dimethyl ether synthesis from syngas in
a fluidized-bed reactor. Chem. Eng. Sci. 59 (2004) 5455-5464.
D

[44] Zh. Nie, H. Liu, D. Liu, W. Ying, D. Fang. Intrinsic kinetics of dimethyl ether synthesis from syngas, Journal of
Natural Gas Chemistry 14 (2005) 22-28.
TE

[45] Q. You, Z. Liu, W. Li, X. Zhou, Synthesis of dimethyl ether from methane mediated by HBr, J. Nat. Gas Chem.
18 (2009) 306-311.
EP

[46] Y. Wang, W. L. Wang, Y. X. Chen, J. J. Zheng, R. F. Li, Synthesis of dimethyl ether from syngas using a
hierarchically porous composite zeolite as the methanol dehydration catalyst, J. Fuel Chem. Technol. 41 (2013) 873-
882.

[47] J. S. Lee, J. C. Kim, Y. G. Kim, Methyl Formate as a new building block in C1 chemistry. Applied Catalysis, 57
C

(1990) l-30.
AC

[48] G. Jenner, Homogeneous catalytic reactions involving methyl formate. Applied Catalysis A: General 12 (1995)
25-44.

[49] W. Keim, Catalysis in C1 Chemistry. Springer Netherlands 1983.

[50] I. Wender, Chemicals from methanol. Catal. Rev. 26 (1984) 303-321.

[51] W.H. Calkins, Chemicals from methanol. Catal. Rev. 26 (1984) 347-358.

[52] W. Couteau, J. Ramioulle, US Patent US4216339.

 ACCEPTED MANUSCRIPT

[53] R. Jiang, W. Guo, M. Li, H. Zhu, L. Zhao, X. Lu, H. Shan, Methanol dehydrogenation on Rh(1 1 1): A density
functional and microkinetic modeling study. Journal of Molecular Catalysis A: Chemical 344 (2011) 99-110.

[54] W. Wang, Y. Wang, Thermodynamic analysis of hydrogen production via partial oxidation of ethanol. Int. J.
Hydrogen Energy 33 (2008) 5035-5044.

[55] L. F. Brown, A comparative study of fuels for on-board hydrogen production for fuel-cell-powered
automobiles. Int. J. Hydrogen Energy 26 (2001) 381-397.

PT
[56] N. Itoh, S. Watanabe, K. Kawasoe, T. Sato, T. A. Tsuji, A membrane reactor for hydrogen storage and transport
system using cyclohexane–methylcyclohexane mixtures. Desalination 234 (2008) 261-269.

[57] M. Momirlan, T. Veziroglu, Current status of hydrogen energy. Renewable and Sustainable Energy Reviews 6

RI
(2002)141-179.

[58] V. Palma, V. Vaiano, D. Barba , M. Colozzi , E. Palo , L. Barbato , S. Cortese, H2 production by thermal

SC
decomposition of H2S in the presence of oxygen. Int. J. Hydrogen Energy 40 (2015 ) 106-113.

[59] J. Christopher Brown, Erdogan Gulari, Hydrogen production from methanol decomposition over Pt/Al2O3 and
ceria promoted Pt/Al2O3 catalysts. Catalysis Communications 5 (2004) 431–436.

U
[60] R. Vakili, E. Pourazadi, P. Setoodeh, R. Eslamloueyan, M. R. Rahimpour, Direct dimethyl ether (DME)
synthesis through a thermally coupled heat exchanger reactor. Appl. Energy 88 (2011) 1211–1223.
AN
[61] R. Vakili, M.R. Rahimpour, R. Eslamloueyan, Incorporating differential evolution (DE) optimization
strategy to boost hydrogen and DME production rate through a membrane assisted single-step DME heat
exchanger reactor, J. of Nat. Gas Sci. and Eng. 9 (2012) 28-38.
M

[62] R. Vakili, H. Rahmanifard, P. Maroufi, R. Eslamloueyan, M.R. Rahimpour, The effect of flow type
patterns in a novel thermally coupled reactor for simultaneous direct dimethyl ether (DME) and hydrogen
production, Int. J. Hydrogen Energy 36 (2011) 4354-4365.
D

[63] F. Samimi, S. Kabiri , M. R. Rahimpour, The optimal operating conditions of a thermally double coupled,
dual membrane reactor for simultaneous methanol synthesis, methanol dehydration and methyl cyclohexane
TE

dehydrogenation, J. of Nat. Gas Sci. and Eng.19 (2014) 175-189.

[64] A. Bakhtyari, M. Mohammadi, M. R. Rahimpour, Simultaneous production of dimethyl ether (DME),

methyl formate (MF) and hydrogen from methanol in an integrated thermally coupled membrane reactor, J. of
Nat. Gas Sci. and Eng. 26 (2015) 595-607.
EP

[65] G. Berc´ic´, J. Levec, Intrinsic and global reaction rate of methanol dehydration over Al2O3 pellets. Ind. Eng.
Chem. Res. 31 (1992) 1035-1040.
C

[66] L. Zhang, H. T. Zhang, W. Y. Ying, D. Y. Fang, Intrinsic kinetics of methanol dehydration over Al2O3 Catalyst.
World Academy of Science, Engineering and Technology 5 (2011) 11-20.
AC

[67] Gh. Moradi, F. Yaripour, H. Abbasian, M. Rahmanzadeh, Intrinsic reaction rate and the effects of operating
conditions in dimethyl ether synthesis from methanol dehydration. Korean J. Chem. Eng. 27 (2010) 1435-1440.

[68] Z. Bai, H. Ma, H. Zhang, W. Ying, D. Fang, Process simulation of dimethyl ether synthesis via methanol vapor
phase dehydration. Polish Journal of Chemical Technology 15 (2013) 122-127.

[69] M. Alavi, H. Jazayeri rad, R. M. Behbahani, Optimizing the Feed Conditions in a DME Production Process to
Maximize the Methanol Conversion. Science and Technology 3 (2013) 61-66.

[70] X. Huang, N. W. Cant, M. S. Wainwright, L. Ma, The dehydrogenation of methanol to methyl formate Part I:
Kinetic studies using copper-based catalysts Chemical Engineering and Processing 44 (2005) 393-402.
 ACCEPTED MANUSCRIPT

[71] M.R. Rahimpour, A. Mirvakili, K. Paymooni, B. Moghtaderi, A comparative study between a fluidized-bed and
a fixed-bed water perm-selective membrane reactor with in situ H2O removal for Fischere Tropsch synthesis of GTL
technology, J. of Nat. Gas Sci. and Eng. 3 (2011) 484-495.

[72] L. S. McLeod, Hydrogen permeation through micro fabricated palladium-silver alloy membranes. PhD Thesis.
Georgia Institute of Technology, December 2008.

[73] G. Barbieri, F.P. Di Maio, Simulation of the methane steam re-forming process in a catalytic Pd-membrane

PT
reactor. Ind. Eng. Chem. Res. 36 (1997) 2121-2127.

[74] G. Pandu Rangaiah, Multi objective optimization: techniques and applications in chemical engineering.
Advances in process systems engineering Vol. 1. National University of Singapore, World Scientific Publishing Co.

RI
Pte. Ltd. 2008.

[75] S. Kirkpatrik, C. D. Gelatt, M. P. Vechhi, Optimization by simulated annealing, Science 220 (1983) 671-680.

SC
[76] H.P. Schwefel, Numerical Optimization of Computer Models, John Wiley & Sons, New York, 1981.

[77] D.E. Goldberg, Genetic algorithms in search, optimization, and machine learning, Addison-Wesley, Reading,
MA, 1989.

U
[78] K. Price, R. Storn, Differential evolution - a simple evolution strategy for fast optimization, Dr. Dobb’s J. 22
(1997) 18-24.
AN
[79] M. Farsi, M. Asemani, M. R. Rahimpour, Mathematical modeling and optimization of multi-stage spherical
reactor configurations for large scale dimethyl ether production. Fuel Processing Technology 126 (2014) 207-214.

[80] M. R. Rahimpour, H. Elekaei Behjati, Dynamic optimization of membrane dual-type methanol reactor in the
M

presence of catalyst deactivation using genetic algorithm, Fuel Processing Technology 90 (2009) 279-291.

[81] P. N. Dwivedi and S. N. Upadhyay, Particle-Fluid Mass Transfer in Fixed and Fluidized Beds. Ind. Eng.
Chem., Process Des. Dev. 16 (1977) 157-165.
D

[82] Holman JP. Heat transfer. New York: McGraw-Hill; 1989.

TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 1: Operating conditions for the typical methanol dehydration process and catalyst characteristics [73]

Parameter Value Unit

Feed Composition (Mole Fraction)
MeOH 0.936 -
DME 0.055 -
H2O 0.009 -
mol.s-1

PT
Total Molar Flow Rate of CDMR 1558.28
Total Molar Flow Rate of TCHR 0.1 mol.s-1
Inlet Temperature 533 K
Inlet Pressure 18.18×105 Pa

RI
Diameter of CDMR 4 m
Reactor Length 8.08 m
wall thermal conductivity 48 J.m-1.K-1.s-1

SC
Catalyst Particle
Density 2010 kg.m-3
Particle Diameter 3.175×10-3 m
Specific Surface Area 673 m2.m-3

U
Ratio of Void Fraction to Tortuosity 0.066 -
Bed Void Fraction 0.5 -
Density of Catalyst Bed 1005 kg.m-3
AN
Table 2: Typical operating conditions for the methanol dehydrogenation reactor in TCHR [73]

Parameter Value Unit

M

Feed Composition (Mole Fraction)

MeOH 0.1 -
MF 0 -
D

H2 0 -
Ar 0.9 -
mol.s-1
TE

Total Molar Flow Rate 0.139

Inlet Temperature 510 K
Inlet Pressure 5×105 Pa
Catalyst Particle
kg.m-3
EP

Density 4500
-3
Particle Diameter 3.55×10 m
Bed Void Fraction 0.39 -
Density of Catalyst Bed 2745 kg.m-3
C

Table 3: Dimensions of TCHR and TCMHR

AC

Parameter Value Unit

TCHR
Inner tube (MeOH to DME) diameter 3.8×10-2 m
Shell (MeOH to MF) diameter 8.8×10-2 m
TCMHR
Inner permeation side diameter 5×10-2 m
Inner tube (MeOH to DME) diameter 7×10-2 m
Middle tube (MeOH to MF) diameter 10.5×10-2 m
Outer permeation side diameter 15×10-2 m
Length of the reactor 8.08 m
 ACCEPTED MANUSCRIPT

Table 4: Governing equations, boundary conditions and auxiliary equations

Definition Equation
Mass Balance

−1  , 
+ ,  , −   = 0 ;  = 1    !"  = 0
,  ,  
Exothermic side (7)

−1  , 

PT
+ ,  , −   = 0 ;  = 1   !"  = 0
,  , 
Endothermic Side (8)

 ,#
− + $ #   = 0 ; $ = 1    !" $ = 0
Inner Permeation
(9)
Sides 

RI
 ,#
− + $ #  = 0 ; $ = 1   !" $ = 0
Outer Permeation
Sides 

SC
Energy Balance
' 01
−%&, (), * + 
Exothermic side + , -  , (−∆/, + − 34* − * 5 −   (* − *&, 6 + = 0 (10)
,  ,
2

U
' 01
−%&, (), * + 
Endothermic Side +  -  , (−∆/, + − 34* − * 5 −  (* − *&,78) + = 0 (11)
,  ,
AN
2
;<

' (),&,9 *&,9 + '

−%&,9 + 9   : %&, * + # 3# (*# − *&,9 + = 0
Inner Permeation
(12)
Sides  
;=,>
M

;<

' (),&,9 *&,9 + '

−%&,9 + 9  : %&, * + # 3# (*# − *&,9 + = 0
Outer Permeation
Side 
;=,>
D

Pressure drop

? 150A 41 − D5 F 1.75 41 − D5 F

TE

=   +
 BC & D E  BC & D E 
Pressure drop (13)

Boundary conditions
EP

Reaction Sides = 0 ⟹  ,# =  ,#,J ; *# = *#,J ; K# = K#,J ; L = 1,2

= 0 ⟹  ,9 =  ,9,J ; *&,9 = *&,9,J ; ! = 3,4
(14)
Permeation Sides
C

Table 5: Comparison between simulation results and plant data of CDMR [27, 73]
AC

Conventional DME Reactor

Output Simulation Plant Data Relative Error %

DME molar flow rate (kmol.h-1) 2469.1 2506 -1.47

MeOH molar flow rate (kmol.h-1) 934.3 937.7 -0.36
Temperature (K) 655 644 +1.71
 ACCEPTED MANUSCRIPT

Table 6: Decision variables and feasible bounds

Parameter Feasible Bounds Unit

Inlet Flow Rate of Exothermic Side 0.1 ≤ )J, ≤ 0.5 mol.s-1
Inlet Temperature of Exothermic Side 495 ≤ *J, ≤ 673 K
Inlet Pressure of Exothermic Side 15 × 10T ≤ K)J, ≤ 20 × 10T Pa
Inlet Flow Rate of Endothermic Side 0.1 ≤ )J, ≤ 0.5 mol.s-1
0.1 ≤ UVW J, ≤ 0.3

PT
Inlet MeOH Mole Fraction of Endothermic Side -
Inlet Temperature of Endothermic Side 453 ≤ *J, ≤ 523 K
Inlet Pressure of Endothermic Side 4 × 10T ≤ K)J, ≤ 10 × 10T Pa
Inlet Flow Rate of Outer Permeation Side 0.5 ≤ XJ,78)WY ≤ 2 mol.s-1

RI
Inlet Temperature of Outer Permeation Side 500 ≤ *XJ,78)WY ≤ 543 K
Inlet Pressure of Outer Permeation Side 0.05 × 10T ≤ K)J,78)WY ≤ 0.1 × 10T Pa
Inlet Flow Rate of Inner Permeation Side 0.5 ≤ XJ, 66WY ≤ 2 mol.s-1

SC
Inlet Temperature of Inner Permeation Side 453 ≤ *XJ, 66WY ≤ 500 K
Inlet Pressure of Inner Permeation Side 0.05 × 10T ≤ K)J, 66WY ≤ 0.1 × 10T Pa

U
Table 7: Optimum parameters of OTCMHR
AN
Parameter Value Unit
)J, 0.1244 mol.s-1
*J, 562.84 K
KJ, 1891803.18 Pa
M

)J, 0.1009 mol.s-1

UVW J, 0.1002 -
*J, 502.96 K
KJ,
D

501040.53 Pa
XJ,78)WY 1.4805 mol.s-1
*J,78)WY 493.12 K
TE

KJ,78)WY 5178.16 Pa
XJ, 66WY 1.3942 mol.s-1
*J, 66WY 502.86 K
KJ, 66WY 5745.61 Pa
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
Figure 1: Schematic diagram of conventional DME reactor (CDMR).
M
D
TE
C EP
AC

Figure 2: Schematic diagram of thermally coupled heat exchanger reactor (TCHR).

 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
Figure 3: Schematic diagram of thermally coupled double membrane heat exchanger reactor (TCMHR).
M
D
TE
C EP
AC

Figure 4: Molar flow rates of MeOH, DME and water in the exothermic side of OTCMHR.
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
Figure 5: Molar flow rates of MeOH, MF and hydrogen in the endothermic side of OTCMHR.
M
D
TE
C EP
AC

Figure 6: Comparing MeOH conversion in the exothermic and endothermic sides of OTCMHR.
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
Figure 7: Axial temperature profiles in reaction sides and permeation sides.
M
D
TE
C EP
AC

Figure 8: Axial pressure profiles in the exothermic and endothermic sides.

 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
Figure 9: Water mole fraction in the inner permeation side and hydrogen mole fraction in the outer
M

permeation side.
D
TE
C EP
AC

Figure 10: Hydrogen and water recovery in the permeation sides.

 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
Figure 11: Comparing axial temperature profiles of OTCMHR and TCHR.
M
D
TE
C EP
AC

Figure 12: Comparing MeOH conversion of OTCMHR and TCHR.

 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
Figure 13: Comparing DME and MF flow rates in OTCMHR and TCHR.
M
D
TE
C EP
AC

Figure 14: Comparing output methanol conversion of exothermic and endothermic sides in OTCMHR and
TCHR.
 ACCEPTED MANUSCRIPT

Highlights:
• Modeling a recuperative reactor for producing DME, MF and H2 from methanol
• Improving MF production rate and purity utilizing a hydrogen selective membrane
• Improving DME production rate and purity utilizing a water selective membrane
• Optimization of the reactor by using genetic algorithm to obtain better performance

PT
RI
U SC
AN
M
D
TE
C EP
AC