Numerical study of heat and mass transfer in a plate
methanol steam micro reformer with methanol
catalytic combustor
Ching-Yi Hsueh
a, Hsin-Sen Chu
a,b, Wei-Mon Yan
c,*
, Chiun-Hsun Chen
aaDepartment of Mechanical Engineering, National Chiao Tung University, Hsin-Chu 300, Taiwan, ROC bIndustrial Technology Research Institute, Chu-Tung, Hsin-Chu 310, Taiwan, ROC
cDepartment of Greenergy, National University of Tainan, Tainan 700, Taiwan, ROC
a r t i c l e i n f o
Article history:Received 16 December 2009 Received in revised form 1 March 2010
Accepted 7 March 2010 Available online 24 April 2010 Keywords: Micro reformer Catalytic combustor Methanol Numerical analysis
a b s t r a c t
A numerical study is performed to examine the characteristics of heat and mass transfer and the performance of a plate methanol steam micro reformer with a methanol catalytic combustor. The effects of the flow configurations for co- and counter-current flows are explored in the present study. The influences of the Reynolds number (Re) and various geometric parameters on heat and mass transfer phenomena in the channels are also investigated numerically. It is expected that the Reynolds number (Re) and various geometric parameters can be improved by thermal management to enhance the chemical reaction and thus augment the micro reformer performance. Comparing the co- and counter-current flows via numerical simulation, the results show that the methanol conversion for counter-current flow could be improved by 10%. This is due to the fact that counter-current flow leads to a better thermal management, which in turn improves fuel conversion efficiency. With a higher Reynolds number on the combustor side, the wall temperature is increased and the methanol conversion can thus be enhanced. Meanwhile, a reduced Reynolds number on the micro reformer side would increase the methanol conversion. The results also reveal that appropriate geometric parameters exist for a micro reformer with a combustor to obtain better thermal management and methanol conversion.
ª 2010 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
The proton exchange membrane fuel cell (PEMFC) is a prom-ising alternative energy source due to the characteristics of high energy density, low noise and low pollution. Applications of PEMFC in portable power sources need to carry enough hydrogen fuel. However, the hydrogen storage problem is still difficult to overcome. To solve this technical difficulty, one possible solution is to employ a reformer. Methanol clearly has distinct advantages as a fuel for fuel cell applications due
to its higher hydrogen-to-carbon ratio, low reforming temperature and greater environmental friendliness[1]. The PEMFC generates electronic energy using hydrogen produced by the reformer from methanol. Therefore, the methanol steam micro reformer with small PEMFC has become a potential candidate for portable electronic products in the near future.
There have appeared a number of experiments on the plate methanol steam micro reformer in the published literature. The micro reformer for methanol steam reforming was * Corresponding author. Tel.:þ886 6 260 2251; fax: þ886 6 260 2205.
E-mail address:[email protected](W.-M. Yan).
A v a i l a b l e a t w w w . s c i e n c e d i r e c t . c o m
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / h e
0360-3199/$e see front matter ª 2010 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijhydene.2010.03.036
designed to produce hydrogen for small PEMFCs. As the steam reforming reaction is an endothermic reaction, electric heaters to provide heat for the steam reforming were used by several investigators [2e7]. The plate integrated fuel processor, consisting of various micro structured modules, was developed to produce hydrogen for fuel cell systems. The fuel processor includes a fuel vaporizer, a catalytic combustor and a steam reformer. The catalytic combustor supplied heat to the steam reforming reaction, and the hydrogen was produced by the micro methanol steam reformer[8e15]. For the PEM fuel cell, the CO concentration must be less than 10 ppm, so a cleanup step is required after methanol steam reforming. The integration of the PrOx or water-gas shift reaction equipment to reduce the CO concentration in the gas from the methanol steam reformer has been used by several researchers[16,17].
To reduce the research cost and shorten the design cycle, modeling and simulation are being used extensively in research institutions and industries across the world to gain a better understanding of the fundamental processes in a methanol steam reformer. Therefore, a vast amount of literature has been focused on theoretical modeling of the methanol steam reformer. Several articles have employed
a cylindrical mathematical model of a packed bed reformer to investigate heat and mass transport phenomena in a meth-anol reformer [18e23]. A numerical plate micro reformer model to analyze the micro reformer performance and reac-tant gas transport characteristics was presented by several investigators[24e29].
The systems are fed by hydrocarbons which convert hydrogen and generate heat, and studies have developed numerical models of a micro reformer with a combustor to explore the heat and mass transport phenomena and fuel conversion efficiency. Deshmukh and Vlachos[30,31]presented a two-dimensional model of propane (C3H8) combustion with ammonia (NH3) decomposition to investigate micro reactors for hydrogen production. The results show that the H2production rate and the temperatures generated via the C3H8combustion both increase as the flow rate of ammonia increases. The results also show that the co-current flow configuration has a lower reactor temperature and allows a wider spectrum of materials to be used than the counter-current flow configuration. A micro-channel model of the thermal integration of a steam reformer and a catalytic combustor was established by Arzamendi and collaborators [32,33]. Using the hydrogen produced by the reforming reaction from methanol and methane, the results Nomenclature
Ci concentration of species i (mol m3)
cp specific heat at constant pressure (J kg1K1)
D hydraulic diameter (m)
Deff effective mass diffusivity (m2s1) Dk mass diffusion coefficient (m2s1) Dp catalyst particle diameter (m) Ea activation energy (J mol1)
H channel height (m)
HC combustion flow channel (m)
HR reforming flow channel (m)
Hw solid wall thickness (m) DH enthalpy of reaction (J mol1)
I, J, K grid points in the x-, y- and z-directions, respectively
keff effective thermal conductivity (W m1K1) kf fluid phase thermal conductivity (W m1K1) kp permeability (m2)
ks solid medium thermal conductivity (W m1K1) k1 pre-exponential factor for steam reforming k2 pre-exponential factor for the reverse water-gas
shift
k3 pre-exponential factor for decomposition reaction k4 pre-exponential factor for combustion reaction
L flow channel length (m)
Mi mole fraction of species i
Mw,i molecular weight of species i (kg mol1)
p pressure (Pa)
R universal gas constant
RSR Arrhenius reaction rate coefficient for steam reforming (mol m3s1)
RrWSG Arrhenius reaction rate coefficient for the reverse water-gas shift (mol m3s1)
RMD Arrhenius reaction rate coefficient for decomposition reaction (mol m3s1) RCombustion Arrhenius reaction rate coefficient for
combustion reaction (mol m3s1)
Re Reynolds number, Re¼ ruD/m
T temperature (C)
T0 inlet temperature (C) Tw wall temperature (C)
u, v, w velocity components in the x-, y- and z-directions, respectively, (m s1)
u0,C inlet flow velocity on the combustion channel side (m s1) g dimensionless coordinate, g¼ y/HCþ dcþ HWþ HRþ dR x, y, z coordinates (m) WR channel width (m) WL steel width (m) Greek symbols
dC combustion catalyst layer thickness (m) dR reforming catalyst layer thickness (m)
3 porosity
h methanol conversion
s tortuosity of the porous medium
m viscosity (kg m1s1) r density (kg m3) rs catalyst density (kg m3) Subscripts eff effective u x-direction v y-direction w z-direction 0 inlet
showed the short diffusion distance and higher area to volume ratio required for using the micro reactor. The results also indicated that complete combustion of methane takes place over a very short distance. The reforming fuel is rapidly heated and then the methane reactor has a more uniform temperature distribution. Pan and Wang[34]designed a plate-fin reformer which integrated endothermic and exothermic reactions into one unit. The combustor supplied the heat for the methanol steam reformer. Their numerical model accurately predicted the methanol conversion rate and the gas distributions. A two-dimensional model of a plate methane reformer with methane combustor to investigate thermo-fluid parameters and geometric parameters was developed by Zanfir and Gavriilidis [35]. Their results showed that the micro reformers have better performance than traditional reformers due to their better heat and mass transfer. The results also showed that a higher channel height produces a lower conversion and much more uniform temperature distribution. Varesano et al. [36] used a one-dimensional transient mathematical model to study the transport behavior in a steam reforming reactor with a burner that supplies heat.
From the literature survey presented above, it was found that some literature is available on mathematical models of the methanol steam micro reformer, but little information is available on mathematical models of a micro reformer with a catalytic combustor. Therefore, the objective of the present study is to investigate the transport phenomena and the fuel conversion efficiency in a methanol steam micro reformer with methanol catalytic combustor. A three-dimensional numerical model of a micro reformer with combustor is developed to examine the effects of various flow configurations and geometric parameters on micro reformer performance.
2.
Analysis
The reactor consists of a methanol steam micro reformer and a methanol catalytic combustion chamber. A schematic diagram of the physical system under consideration is shown inFig. 1. The system consists of a solid wall, two catalyst layers and two flow channels each at the catalytic combustion/steam reforming side. It is seen that the methanol catalytic combus-tion chamber and the methanol steam reforming chamber are separated by a solid wall. Both sides of each channel are coated with a combustion catalyst layer and a steam reforming cata-lyst layer. The heat from the combustion reaction is used to drive the steam reforming reaction. To simplify the analysis for the present study, the flowing assumptions are made: (1) the flow is steady state;
(2) the inlet fuel is an ideal gas;
(3) the flow is laminar and incompressible; (4) the catalyst layer is isotropic;
(5) the chemical reaction occurs only in the catalyst layer; (6) thermal radiation and conduction in the gas phase are
negligible compared to convection.
With the above assumptions, the gas transport equations for the three-dimensional reactor can be described as follows. Continuity equation: r vu vxþvvvyþvwvz ¼ 0 (1) X-momentum equation: 3r uvu vxþ vvuvyþ wvuvz ¼ 3vpvxþ 3m v2u vx2þ v2u vy2þ v2u vz2 þ Su (2) Y-momentum equation: 3r uvv vxþ vvvvyþ wvvvz ¼ 3vpvyþ 3m v2v vx2þ v2v vy2þ v2v vz2 þ Sv (3) Z-momentum equation: 3r uvw vxþ v vw vyþ w vw vz ¼ 3vp vzþ 3m v2w vx2þ v2w vy2 þ v2w vz2 þ Sw (4) In the above equations, e stands for the porosity of the medium, and the mixture viscosity mmix based on kinetic theory is[37] mmix¼ X5 i¼1 Mimi P5 j¼1Mjfij (5) where fij¼ X i h 1þmi mj 1 2Mw;j Mw;i 1 42 h 81þMw;i Mw;j i1 2 (6)
Su, Svand Sware corrected terms of the reactant gas flow in a porous material in the catalyst layer. The source terms, Su, Sv and Swin the momentum equations are listed in Eqs. (7e9), respectively. Among them, the source terms, Su, Sv and Sw account for the Ergun equations[38]in the x-, y- and z-directions, respectively. The parameter kpstands for the permeability and bis the inertial loss coefficient in each component direction. Su¼ mu kp bur 2 ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u2þ v2þ w2 p (7) Sv¼ mv kp bvr 2 ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u2þ v2þ w2 p (8) Sw¼ mw kp bwr 2 ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u2þ v2þ w2 p (9) where kp¼ Dp33 150ð1 3Þ2 (10) b¼3:5ð1 3Þ Dp33 (11) and where Dpis the catalyst particle diameter.
Species equation: uvmi vx þ vvm i vy þ wvm i vz ¼ Deff v2m i vx2 þ v2m i vy2 þ v2m i vz2 þ ð1 3ÞrsSc (12) In the species equation, midenotes the mass fraction of the ith species, where the various species are CH3OH, H2O, H2, CO2, CO and O2. In these expressions, the concentrations of
CH3OH, H2O, H2, CO2, CO are calculated on the steam reforming side and CH3OH, H2O, CO2, O2are calculated on the combustion side. The effective diffusion coefficient, Deff is determined by the StefaneMaxwell equations[37]. Eq.(13)is employed to describe the influence of the porosity on the diffusion coefficient:
Deff¼ Dk3s (13)
The diffusion coefficient Dkfor the methanol steam micro reformer was derived from the StefaneMaxwell equations which were used to calculate the mean effective binary diffusivity[21]. Scis the source term of chemical reaction in the species equation, and differs according to the reactant gases in the catalyst layer. In the present study, there is no chemical reaction in the flow channel. Therefore, Scis zero in the flow channel. In the catalyst layer, the source term of the Fig. 1e Schematic diagram of the present study.
species equation, Sc, can be described by the following modi-fied concentration term:
Sc¼ 8 < : Mw;iðRSRþ RrWGSþ RMDÞ
l00i l0i for steam reforming Mw;iðRCombustionÞ
l00i l0i for combustion reaction (14) where Mw,iis the molecular weight of species i, and Ri,ris the Arrhenius molar rate of creation and destruction of species i in the reaction. l00i and l0i are the stoichiometric coefficient for reaction i and product i, respectively, in the reaction.
According to the chemical kinetics of Pepply et al.[39], the methanol steam reforming reaction consists of three overall reactions: one is a primary process in methanol steam reforming and the others are the decomposition reaction and water-gas shift reaction. Therefore, the steam reforming reaction, Eq.(15), the reverse water-gas shift reaction, Eq.(16), and the decomposition reaction, Eq.(17), are considered in this study. CH3OHþ H2O4 k1 k1 CO2þ 3H2 (15) CO2þ H24 k2 k2 COþ H2O (16) CH3OH/ k3 COþ 2H2 (17)
In this work, to simplify the analysis, the model of Mastalir et al. [40]for methanol steam reforming is used, and the Arrhenius equation is employed to calculate the reactant gases generated by the chemical reaction.
RSR¼ k1C0:6CH3OHC0:4H2Oexp Ea RT k1CCO2CH2exp Ea RT (18) RrWGS¼ k2CCO2CH2exp Ea RT k2CCOCH2Oexp Ea RT (19) RMD¼ k3C1:3CH3OHexp Ea RT (20) where the steam reforming reaction and reverse water-gas shift reaction are reversible reactions and the decomposition reaction is a non-reversible reaction. The constants k1, k2and k3are forward rate constants, and the constant k1and k2are the backward rate constants.
The reaction of the combustion catalyst layer can be represented by the following reaction, Eq.(21). The reaction rate of methanol over the Pt/Al2O3 catalyst was calculated with Eq.(22), as proposed by Pasel et al.[41]
CH3OHþ 1:5O2/ k4 CO2þ 2H2O (21) RCombustion¼ k4C1:3CH3OHexp Ea RT (22) In order to evaluate the distributions of the local tempera-ture, the energy equations must be solved.Energy equation: rcp uvT vxþ vvTvyþ wvTvz ¼ keff v2T vx2þ v2T vy2þ v2T vz2 þ ð1 3ÞrsSt (23)
The effective thermal conductivity is modified to account for the porous medium effect:
keff¼ 3kfþ ð1 3Þks (24)
where kfis the fluid phase thermal conductivity, ksis the solid medium thermal conductivity and e is the porosity of the medium.
The source term St in the energy equation due to the chemical reactions is determined by
St¼ 8 < :
ðDHSRRSRþDHrWGSRrWGSþDHMDRMDÞ for steam reforming ðDHCombustionRCombustionÞ for combustion reaction
(25) As for the energy equation of the solid wall, one has v2T vx2þ v2T vy2þ v2T vz2¼ 0 (26)
The boundary conditions of the present computation include those at the inlet, the outlet, the wall, and the inter-face between the flow channel and the catalyst layer. (1) The boundary conditions for inlets at the flow channel and
the catalyst layer: the inlet flow velocity is constant, the inlet gas composition is constant, and the inlet tempera-ture is constant.
(2) The boundary conditions for outlets at the flow channel and the catalyst layer: there is fully developed flow. (3) The boundary conditions for the interface between the
solid wall and the insulated walls: the temperature gradi-ents are zero.
(4) The boundary conditions for the interface between the flow channel and solid wall: no slip and zero fluxes hold the velocities and the concentration gradients are zero. (5) The boundary conditions for the interface between the
flow channel and the catalyst layer: the velocities, temperatures, species concentrations and species fluxes are continuous.
The geometrical dimensions and parameters used are lis-ted inTable 1.
3.
Numerical method
The solution to the governing equations is found by employ-ing a finite volume scheme with the model domain divided into a number of cells as control volumes. The governing equations are numerically integrated over each of these computational cells or control volumes. The method exploits a collocated cell-centered variable arrangement with the local or cell-averaged values of the physical quantities evaluated and stored at each cell center.
The governing equations can be expressed in the form of a generalized convectionediffusion type of transport equation:
V$ðrfVÞ ¼ V$ðGVfÞ þ Sf (27)
vector, Sfis the source per unit volume and r is the density. With the discretization of the governing equations, the coupled finite-difference equations become
aPfP¼ aEfEþ aWfWþ aNfNþ aSfSþ Sf (28) where fPis the value of f at the current point P, fE.fSstand for the values of the grid points adjacent to the point P, and aP.aSare known as the link coefficients. All equations were numerically solved using the commercial fluid dynamics program Fluent. The SIMPLE algorithm was employed to solve the convectionediffusion equations. The convergence criteria for the normalized residuals for each variable were restricted to less than 106.
In this work, a grid system of 121 81 21 points was used. To examine the grid independence of the predictions, three grid systems were considered and their influences on the prediction of local temperature distributions for a typical case are presented inTable 2. It is found that the maximum devi-ation among the computdevi-ations using grids of 101 62 11, 121 81 21 and 141 100 31 is less than 1.3% and the results on the 121 81 21 and the 141 100 31 grids are quite close. Therefore, the grid system of 121 81 21 points seems sufficient to resolve the behaviors of local temperature distributions in the present micro reformer model. To further check the adequacy of the numerical scheme, it is clearly seen fromFig. 2that the present predictions agree reasonably well with the experimental data of Won[10]. The above prelimi-nary runs confirm that the present model and the numerical method used are generally appropriate for the analysis of the problem.
4.
Results and discussion
In the present paper, a three-dimensional model is analyzed to understand heat and mass transfer in the channels of a methanol steam micro reformer with methanol catalytic combustor. The influences of wall conduction effects on the transport phenomena of heat and mass transfer in a micro reformer with combustor are important. Therefore, the local temperature distributions along the centerline of the top reforming channel (Y¼ 0.333) and the CH3OH mole fraction distributions along the centerline of the reforming channel (Y¼ 0.167) are presented inFig. 3. In this work, X denotes the dimensionless distance from the flow channel inlet to the outlet. It is clearly seen in Fig. 3(a) that the temperature distributions with the wall thermal conduction effect show a lower and more uniform distribution than that without a wall conduction effect. This implies that the effects of wall
Table 1e Parameters used in this study.
Flow channel length L (m) 4 103 Combustion catalyst layer thickness dC(m) 5.0 105 Reforming catalyst layer thickness dR(m) 5.0 105 Combustion flow channel HC(m) 4.5 104 Reforming flow channel HR(m) 4.5 104 Average inlet temperature (C) 120
Operating pressure (atm) 1
Catalyst density (kg m3)[21] 1480 Catalyst thermal conductivity (W m1K1)[21] 0.3 Catalyst layer porosity[24] 0.38 Catalyst permeability (m2)[24] 2.379 1012 Mass diffusion coefficient (m2s1)[21] 6.8 105 Activation energy for steam reforming
(J mol1)[40]
1.09 105 Activation energy for the reverse water-gas
shift (J mol1)[40]
1.15 105 Activation energy for decomposition reaction
(J mol1)[40]
1.42 105 Activation energy for combustion reaction
(J mol1)[41]
1.3 104
Table 2e Temperature distributions (C) for the various grid tests at different axial locations.
X (I J K ) 0.125 0.250 0.375 0.500 0.625 0.750 0.875 1.000 101 62 11 239.0 239.8 240.6 241.4 241.9 242.1 241.7 210.0 121 81 21 235.9 236.7 237.6 238.3 238.9 239.1 238.7 207.6 141 100 31 233.4 234.2 235.1 235.8 236.3 236.6 236.1 205.7 180 200 220 240 260 280 300 320 0.3 0.35 0.4 0.45 Present Results
Experimental Study of Won et al. [10]
T w ( o C) u 0,C(m s -1 ) 0 20 40 60 80 100 200 220 240 260 280 300 Present Results
Experimental Study of Won et al. [10]
(% ) T w( o C)
a
b
Fig. 2e Comparison of theoretical simulation of the present results with previous experimental data by Won et al.[10].
conduction on the thermal development in a micro reformer with combustor are important. It is also found inFig. 3that the effects of wall conduction lead to a higher methanol distri-bution than without a wall conduction effect due to a smaller value of the temperature distribution. However, the wall
conduction effects on the heat and mass transport phenomena are remarkable and cannot be neglected in the modeling. Therefore, their effect should be considered in this work.
The influences of the flow configurations on the transport phenomena and the performance of micro reformers are important. To this end, the effects of the flow configurations for co- and counter-current flow on the temperature distri-butions along different axial location lines and on the local distributions of the mole fractions of various species along the centerline of the reforming channel are presented inFig. 4. Fig. 4(a) discloses that the temperature distributions are much more uniform due to the shorter thermal entrance length. It is also obvious that a higher temperature distribution is noted for the counter-current flow. This is due to the fact that counter-current flow leads to better thermal management. Fig. 4(b) shows the local distributions of the different species for co- and counter-current flow along the centerline of the reforming channel (Y¼ 0.167). An overall inspection ofFig. 4(b) reveals that both the mole fractions of CH3OH and H2O decrease as the fluid moves downstream, while the H2, CO2 and CO mole fractions increase with axial location. In addi-tion, a lower CH3OH mole fraction along the centerline of the channel represents a higher methanol conversion rate. The methanol conversion rate is greater than 91% for the
counter-120 140 160 180 200 220 240 260 280 0 0.1 0.2 0.3 0.4 0.5 0 0.2 0.4 0.6 0.8 Without wall conduction With wall conduction
MCH OH X Tw ( oC) Counter-current flow ReC=13.8, ReR=0.7 H C=0.45mm, HR=0.45mm, WR=0.5mm, WL=0.5mm 1
Fig. 3e Comparisons of the simulation results with and without wall conduction effects for the temperature distributions and CH3OH mole fraction distributions along the centerline of the channel.
120 140 160 180 200 220 240
a
b
260 0 0.2 0.4 0.6 0.8 1 Y=0.167 Y=0.333 Y=0.667 Y=0.833 T( oC) X Counter-current flow Co-current flow Re C=13.8, ReR=0.7, HC=0.45mm, HR=0.45mm, WR=0.5mm, WL=0.5mm 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 CH 3OH H 2O H 2 CO 2 CO Co-current flow Counter-current flow Mi X Re C=13.8, ReR=0.7, HC=0.45mm, HR=0.45mm, WR=0.5mm, WL=0.5mmFig. 4e Effects of co- and counter-current flow
configurations on (a) the temperature distributions along different axial location lines (Y [ 0.333) and (b) the local distributions of the mole fractions of the various species along the centerline of the reforming channel (Y [ 0.167).
180 200 220 240 260 280 0 0.2 0.4 0.6 0.8 1 Re C=6.9 Re C=13.8 Re C=27.6 X T w ( oC) Re R=0.7, HC=0.45mm, HR=0.45mm, WR=0.5mm, WL=0.5mm Counter-current flow 0 0.1 0.2 0.3 0.4 0.5 0 0.2 0.4 0.6 0.8 1 Re C=6.9 Re C=13.8 Re C=27.6 X MCH OH Re R=0.7, HC=0.45mm, HR=0.45mm, WR=0.5mm, WL=0.5mm Counter-current flow
a
b
Fig. 5e Effects of inlet fuel velocities for the combustor on (a) the temperature distributions along the top centerline of the reforming channel (Y [ 0.333) and (b) the CH3OH mole fraction distributions along the centerline of the reforming channel (Y [ 0.167).
current flow, with a product gas composition of 73.2% H2, 25.1% CO2and 1.7% CO at the outlet of the reforming channel. Comparing co- and counter-current flow via numerical simulation, the results show that the methanol conversion efficiency for counter-current flow could be improved by 10% due to a higher temperature distribution.
In the study of the effects of the Reynolds number (Re) and various geometric parameters on the transport phenomena and micro reformer performance, understanding the detailed distributions of heat and mass transfer is important to the design of a micro reformer with a combustor. The Reynolds number (Re) is one of the key thermo-fluid parameters in the micro reformer and combustor channels which would affect the micro reformer performance. Therefore, effects of the Reynolds number (ReC) on the combustion channel side on the temperature distributions along the centerline of the top reforming channel (Y¼ 0.333) and CH3OH mole fraction distributions along the centerline of the reforming channel (Y¼ 0.333) with counter-current flow are examined inFig. 5. A careful examination ofFig. 5(a) shows that the temperature increases with an increase in the Reynolds number (ReC) on the combustor side due to a higher inlet flow rate. This is because higher combustion energy is released for a higher inlet flow rate (Reynolds number). Afterwards, an overall
inspection ofFig. 5(b) reveals that the CH3OH mole fractions decrease as the fluid moves downstream due to the reforming chemical reaction. It is also shown that the CH3OH mole fraction distributions decrease with increasing Reynolds number in the combustor channels. This can be made plau-sible by noting the fact that the chemical reaction increases as the temperature distributions increase.
Fig. 6presents the effects of the Reynolds number (ReR) on the micro reformer side on the temperature distributions along the centerline of the top reforming channel (Y¼ 0.333) and CH3OH mole fraction distributions along the centerline of the reforming channel (Y¼ 0.167) with counter-current flow. The results inFig. 6(a) reveal that the temperature distribu-tions are enhanced by the decreased Reynolds number (ReR) on the micro reformer channels. This is due to a higher Reynolds number (ReR) significantly increasing the heat leaving the flow channel, which decreases the temperature rise. In addition, it is seen inFig. 6(b) that the CH3OH mole fraction distributions decrease with a lower Reynolds number due to a longer time of gas residence and a higher tempera-ture, which results in better methanol conversion.
In order to explain the effectiveness of the geometric parameters for a micro reformer with combustor in thermal management, the temperature and CH3OH mole fraction 180 200 220 240 260 280
a
b
0 0.2 0.4 0.6 0.8 1 Re R=0.4 Re R=0.7 Re R=1.4 X Tw ( o C) Re C=13.8, HC=0.45mm, HR=0.45mm, WR=0.5mm, WL=0.5mm Counter-current flow 0 0.1 0.2 0.3 0.4 0.5 0 0.2 0.4 0.6 0.8 1 Re R=0.4 Re R=0.7 Re R=1.4 X MCH OH Counter-current flow Re C=13.8, HC=0.45mm, HR=0.45mm, WR=0.5mm, WL=0.5mmFig. 6e Effects of inlet fuel velocities for the reformer on (a) the temperature distributions along the top centerline of the reforming channel (Y [ 0.333) and (b) the CH3OH mole fraction distributions along the centerline of the reforming channel (Y [ 0.167). 180 190 200 210 220 230 240 250 260
a
b
0 0.2 0.4 0.6 0.8 1 H C=0.20mm H C=0.45mm H C=0.95mm X T w ( oC) Re C=13.8, ReR=0.7, HR=0.45mm, WR=0.5mm, WL=0.5mm Counter-current flow 0 0.1 0.2 0.3 0.4 0.5 0 0.2 0.4 0.6 0.8 1 H C=0.20mm H C=0.45mm H C=0.95mm X M CH OH Re C=13.8, ReR=0.7, HR=0.45mm, WR=0.5mm, WL=0.5mm Counter-current flowFig. 7e Effects of the channel height of the combustor on (a) the temperature distributions along the top centerline of the reforming channel and (b) the CH3OH mole fraction distributions along the centerline of the reforming channel.
distributions for various geometric parameters were investi-gated. For a fixed Reynolds number,Fig. 7demonstrates the effects of the combustion flow channel heights on the temperature distributions along the centerline of the top reforming channel and on the CH3OH mole fraction distribu-tions along the centerline of the reforming channel with counter-current flow. It is shown inFig. 7(a) that along the reforming channel, the temperature distributions are very uniform. Overall inspection ofFig. 7(b) disclosed that a lower CH3OH mole fraction is noted for a system with a greater combustion flow channel height due to a stronger chemical reaction for a higher temperature distribution. This means that a higher efficiency methanol conversion is enhanced via a greater combustion flow channel height.
An exploration of the temperature distributions for various reforming flow channel heights along the centerline of the top reforming channel is presented inFig. 8(a). For fixed Re, the results show that a higher temperature distribution is found for a micro reformer channel with a lower reforming flow channel height. This is because a higher channel height has a greater hydraulic diameter. As for the effects of reforming flow channel heights on the CH3OH mole fraction distribu-tions along the centerline of the reforming channel, an overall inspection of Fig. 8(b) reveals that better micro reformer performance is noted for a lower reforming channel height.
This implies that the chemical reaction rate is slower for a system with a greater reforming channel height. This seems plausible as a stronger chemical reaction is experienced for a micro reformer channel with a higher wall temperature.
The effects of the channel widths on the temperature distributions and CH3OH mole fraction distributions for a fixed Reynolds number were also investigated. Fig. 9(a) presents the effects of channel widths on the temperature distributions along the centerline of the top reforming channel (Y¼ 0.333). It is shown in Fig. 9(a) that the local temperature distribution increases with a decrease in the channel width. This may be because a rather narrow channel decreases the heat leaving the flow channel. InFig. 9(b), the methanol conversion of the micro reformer is slightly enhanced with a wider channel width. It is important to note that a higher temperature distribution will not necessarily provide better methanol conversion, because the channel width increases with increasing catalyst reaction area, which in turn increases the chemical reaction rate.
Fig. 10 shows the effects of the steel widths on the temperature distributions and CH3OH mole fraction distribu-tions.Fig. 10(a) presents the effects of the steel widths on the temperature distributions along the centerline of the top
reforming channel (Y¼ 0.333). Comparison of the
180 200 220 240 260 280 0 0.2 0.4 0.6 0.8 1 H R=0.20mm H R=0.45mm H R=0.95mm X T w ( oC)
a
b
Re C=13.8, ReR=0.7, HC=0.45mm, WR=0.5mm, WL=0.5mm Counter-current flow 0 0.1 0.2 0.3 0.4 0.5 0 0.2 0.4 0.6 0.8 1 H R=0.20mm H R=0.45mm H R=0.95mm X MCH OH Re C=13.8, ReR=0.7, HC=0.45mm, WR=0.5mm, WL=0.5mm Counter-current flowFig. 8e Effects of the channel height of the reformer on (a) the temperature distributions along the top centerline of the reforming channel and (b) the CH3OH mole fraction distributions along the centerline of the reforming channel.
200 210 220 230 240 250 260 0 0.2 0.4 0.6 0.8 1 W R=0.25mm W R=0.50mm W R=1.00mm X T( oC)
a
b
Re C=13.8, ReR=0.7, HC=0.45mm, HR=0.45mm, WL=0.5mm Counter-current flow 0 0.1 0.2 0.3 0.4 0.5 0 0.2 0.4 0.6 0.8 1 W R=0.25mm W R=0.50mm W R=1.00mm X M CH OH (b) Re C=13.8, ReR=0.7, HC=0.45mm, HR=0.45mm, WL=0.5mm Counter-current flowFig. 9e Effects of the channel width on (a) the temperature distributions along the top centerline of the reforming channel (Y [ 0.333) and (b) the CH3OH mole fraction distributions along the centerline of the reforming channel (Y [ 0.167).
corresponding curves of the steel widths WL¼ 0.25 mm, 0.5 mm, and 1.0 mm indicates that a higher temperature distribution is found with wider steel. As for the effects of the steel width on the CH3OH mole fraction distributions along the centerline of the reforming channel (Y¼ 0.167), the results reveal that they have similar CH3OH mole fraction distribu-tions. This is due to having similar temperature distribudistribu-tions. Therefore, the steel width does not have a significant impact on the methanol conversion.
Fig. 11demonstrates the effects of the Reynolds number (ReC) for the combustor on methanol conversion and wall temperature of the reforming channel in the plate methanol steam micro reformer. For comparison, the results without the wall conduction effect in the model are also presented. An overall inspection ofFig. 11shows that the wall tempera-ture increases with increase of the Reynolds number (ReC) on the combustion channel side. This is plausible because the inlet fuel velocity increases in the channel as the Reynolds number increases. As for the methanol conversion, the results show that the methanol conversion increases as the Reynolds number of the combustion channel increases, which in turn increases the wall temperature. In addition, the wall conduction effect is also shown inFig. 11. The deviations in
methanol conversion between the results with and without consideration of wall conduction effects are larger. This means that the wall conduction effect on methanol conver-sion and wall temperature become significant and cannot be neglected in the modeling.
5.
Conclusions
With an appropriate design of the flow configurations in the reactor, the thermal management of a micro reformer with combustor can be achieved efficiently. To reach this end, various flow configurations for co- and counter-current flow have been proposed. In addition, the effects of the Reynolds number and various geometric parameters have also been investigated by using numerical simulations of the detailed gas transport phenomena and micro reformer performance. What follows are the major findings.
(1) The deviations between the predictions with and without consideration of the wall conduction effects reveal the influences of the wall conduction on the transport phenomena and the performance of the micro reformer. The present results provide evidence for the significance of a wall conduction effect and imply that it is necessary to include this effect in modeling and analysis.
(2) The application of the flow configuration design in a plate methanol steam micro reformer with methanol catalytic combustor leads to improved thermal management and micro reformer performance. The effects can be enhanced with a counter-current flow configuration in a plate methanol steam micro reformer with methanol catalytic combustor.
(3) With a higher Reynolds number on the combustor side, the wall temperature is increased and methanol conversion can thus be enhanced. Meanwhile, a reduced Reynolds number for the reactant gas on the micro reformer side will raise the reactant gas residence time, which in turn increases the methanol conversion and improves the temperature distributions.
(4) The effects of channel geometry have a significant impact on the methanol conversion and heat and mass transfer in
200 210 220 230 240 250 260 0 0.2 0.4 0.6 0.8 1 W L=0.25mm W L=0.50mm W L=1.00mm X T( o C)
a
Re C=13.8, ReR=0.7, HC=0.45mm, HR=0.45mm, WR=0.5mm Counter-current flow 0 0.1 0.2 0.3 0.4 0.5 0 0.2 0.4 0.6 0.8 1 W L=0.25mm W L=0.50mm W L=1.00mm X MCH OHb
Re C=13.8, ReR=0.7, HC=0.45mm, HR=0.45mm, WR=0.5mm Counter-current flowFig. 10e Effects of the steel widths on (a) the temperature distributions along the top centerline of the reforming channel (Y [ 0.333) and (b) the CH3OH mole fraction distributions along the centerline of the reforming channel (Y [ 0.167). 0 50 100 150 200 250 300 0 20 40 60 80 100 5 10 15 20 25
Without wall conduction With wall conduction
Re C Tw ( oC) (%) Re R=0.7, HC=0.45mm, HR=0.45mm, WR=0.5mm, WL=0.5mm Counter-current flow
Fig. 11e Effects of the Reynolds number (Re) of the combustor on wall temperature and methanol conversion.
the plate methanol micro reformer with methanol cata-lytic combustor.
Acknowledgement
The study was supported by the National Science Council, the Republic of China, through the grants NSC 97-2221-E-211-015-MY2.
r e f e r e n c e s
[1] Holladay JD, Wang Y, Jones E. Review of developments in portable hydrogen production using microreactor technology. Chem. Rev. 2004;104:4767e90.
[2] Pfeifer P, Schubert K, Liauw MA, Emig G. Electrically heated microreactors for methanol steam reforming. Chem. Eng. Res. Des. 2003;81(7):711e20.
[3] Park GG, Seo DJ, Park SH, Yoon YG, Kim CS, Yoon WL. Development of microchannel methanol steam reformer. Chem. Eng. J. 2004;101(1):87e92.
[4] Lim MS, Kim MR, Noh J, Woo SI. A plate-type reactor coated with zirconia-sol and catalyst mixture for methanol steam-reforming. J. Power Sources 2005;140(1):66e71.
[5] Kundu A, Jang JH, Lee HR, Kim SH, Gil JH, Jung CR, et al. MEMS-based micro-fuel processor for application in a cell phone. J. Power Sources 2006;162(1):572e8.
[6] Kwon OJ, Hwang SM, Ahn JG, Kim JJ. Silicon-based miniaturized-reformer for portable fuel cell applications. J. Power Sources 2006;156(2):253e9.
[7] Ha JW, Jang JH, Gil JH, Kim SH. The fabrication and performance of a poly(dimethylsiloxane) (PDMS)-based microreformer for application to electronics. Int. J. Hydrogen Energy 2008;33(8):2059e63.
[8] Kwon OJ, Yoon DH, Kim JJ. Silicon-based miniaturized reformer with methanol catalytic burner. Chem. Eng. J. 2008; 140(1):466e72.
[9] Sohn JM, Byun YC, Cho JY, Choe J, Song KH. Development of the integrated methanol fuel processor using micro-channel patterned devices and its performance for steam reforming of methanol. Int. J. Hydrogen Energy 2007;32(18): 5103e8.
[10] Won JY, Jun HK, Jeon MK, Woo SI. Performance of microchannel reactor combined with combustor for methanol steam reforming. Catal. Today 2006;111(3): 158e63.
[11] Kim T. Micro methanol reformer combined with a catalytic combustor for a PEM fuel cell. Int. J. Hydrogen Energy 2009;34 (16):6790e8.
[12] Park DE, Kim T, Kwon S, Kim CK, Yoon E. Micromachined methanol steam reforming system as a hydrogen supplier for portable proton exchange membrane fuel cells. Sens. Actuators A 2007;135(1):58e66.
[13] Yoshida K, Tanaka S, Hiraki H, Esashi M. A micro fuel reformer integrated with a combustor and a microchannel evaporator. J. Micromech. Microeng. 2006;16:191e7. [14] Park GG, Yim SD, Yoon YG, Kim CS, Seo DJ, Eguchi K.
Hydrogen production with integrated microchannel fuel processor using methanol for portable fuel cell systems. Catal. Today 2005;110(1e2):108e13.
[15] Park GG, Yim SD, Yoon YG, Lee WY, Kim WS, Seo DJ, et al. Hydrogen production with integrated microchannel fuel processor for portable fuel cell systems. J. Power Sources 2005;145(2):702e6.
[16] Kwon OJ, Hwang SM, Chae JH, Kang MS, Kim JJ. Performance of a miniaturized silicon reformer-PrOx-fuel cell system. J. Power Sources 2007;165(1):342e6.
[17] Kim T, Kwon S. MEMS fuel cell system integrated with a methanol reformer for a portable power source. Sens. Actuators A 2009;154(2):204e11.
[18] Suh JS, Lee MT, Greif R, Grigoropoulos CP. A study of steam methanol reforming in a microreactor. J. Power Sources 2007; 173(1):458e66.
[19] Suh JS, Lee MT, Greif R, Grigoropoulos CP. Transport
phenomena in a steam-methanol reforming microreactor with internal heating. Int. J. Hydrogen Energy 2009;34(1):314e22. [20] Karim A, Bravo J, Datye A. Nonisothermality in packed bed
reactors for steam reforming of methanol. Appl. Catal. A 2005;282(1):101e9.
[21] Karim A, Bravo J, Gorm D, Conant T, Datye A. Comparison of wall-coated and packed-bed reactors for steam reforming of methanol. Catal. Today 2005;110(1):86e91.
[22] Cao C, Wang Y, Holladay JD, Jones EO, Palo DR. Design of micro-scale fuel processors assisted by numerical modeling. AlChE J. 2005;51(3):982e8.
[23] Pattekar AV, Kothare MV. A radial microfluidic fuel processor. J. Power Sources 2005;147(1e2):116e27. [24] Pattekar AV, Kothare MV. A microreactor for hydrogen
production in micro fuel cell applications. J. Microelectromech. Syst. 2004;13(1):7e18.
[25] Park HG, Malen JA, Piggott III WT, Morse JD, Greif R, Grigoropoulos CP. Methanol steam reformer on a silicon wafer. J. Microelectromech. Syst. 2006;15(4):976e85. [26] Kawamura Y, Ogura N, Yamamoto T, Igarashi A. A
miniaturized methanol reformer with Si-based microreactor for a small PEMFC. Chem. Eng. Sci. 2006;61(4):1092e101. [27] Hsueh CY, Chu HS, Yan WM. Numerical study on
micro-reformer performance and local transport phenomena of the plate methanol steam micro-reformer. J. Power Sources 2009; 187(2):535e43.
[28] Chen F, Chang MH, Kuo CY, Hsueh CY, Yan WM. Analysis of a plate-type microreformer for methanol steam reforming reaction. Energy Fuels 2009;23(10):5092e8.
[29] Kim T, Kwon S. Design, fabrication and testing of a catalytic microreactor for hydrogen production. J. Micromech. Microeng. 2006;16(1):1760e8.
[30] Deshmukh SR, Vlachos DG. Effect of flow configuration on the operation of coupled combustor/reformer microdevices for hydrogen production. Chem. Eng. Sci. 2005;60(21):5718e28. [31] Deshmukh SR, Vlachos DG. CFD simulations of coupled,
countercurrent combustor/reformer microdevices for hydrogen production. Ind. Eng. Chem. Res. 2005;44(14): 4982e92.
[32] Arzamendi G, Dieguez PM, Montes M, Centeno MA, Odriozola JA, Gandia LM. Integration of methanol steam reforming and combustion in a microchannel reactor for H2 production: a CFD simulation study. Catal. Today 2009;143 (1e2):25e31.
[33] Arzamendi G, Dieguez PM, Montes M, Odriozola JA, Sousa-Aguiar EF, Gandia LM. Methane steam reforming in a microchannel reactor for GTL intensification:
a computational fluid dynamics simulation study. Chem. Eng. Sci. 2009;154(1e3):168e73.
[34] Pan L, Wang S. Modeling of a compact plate-fin reformer for methanol steam reforming in fuel cell systems. Chem. Eng. J. 2005;108(1):51e8.
[35] Zanfir M, Gavriilidis A. Catalytic combustion assisted methane steam reforming in a catalytic plate reactor. Chem. Eng. Sci. 2003;58(17):3947e60.
[36] Varesano A, Guaglio I, Saracco G, Maffettone PL. Dynamics of a methanol reformer for automotive applications. Ind. Eng. Chem. Res. 2005;44(4):759e68.
[37] White FM. Viscous fluid flow. 2nd ed. McGraw-Hill; 1991. [38] Ergun S. Fluid flow through packed columns. Chem. Eng.
Prog. 1952;48(2):89e94.
[39] Pepply A, Amphlett JC, Kearns LM, Mann RF. Methanol steam reforming on Cu/ZnO/Al2O3catalysts. Part 2. A
comprehensive kinetic model. Appl. Catal. A 1999;179(1e2): 31e49.
[40] Mastalir A, Frank B, Szizybalski A, Soerijanto H, Deshpande A, Niederberger M, et al. Steam reforming of methanol over Cu/ZrO2/CeO2catalysts: a kinetic study. J. Catal. 2005;230(2):464e75.
[41] Pasel J, Emonts B, Peters R, Stolten D. A structured test reactor for the evaporation of methanol on the basis of a catalytic combustion. Catal. Today 2001;69(1e4):193e200.
