Performance assessment of a spiral methanol to hydrogen fuel processor for fuel cell applications

Available online at www.sciencedirect.comJOURNALOFScienceDirectNATURAL GASCHEMISTRYEL SEVIERJournal of Natural Gas Chemistry 21(2012)526- -533www.elsevier.com/locate/jagcPerformance assessment of a spiral methanol to hydrogenfuel processor for fuel cell applicationsFoad Mehri,，Majid Taghizadeh*Chemical Engineering Deparment, Babol University of Technology, P. O. Box 484, 4714871167 Babol, Iran[ Manusripc reived March 27, 2012; revised May 21, 2012 ]AbstractA novel design of plate-type microchannel reactor has been devcloped for fuel cell-grade hydrogen production. Commercial Cw/ZnVAl2O3 wasused as catalyst for the reforming reaction, and its effectiveness was evaluated on the mole fraction of products, methanol conversion, hydrogenyield and the amount of carbon monoxide under various operating conditions. Subsequently, 0.5 wt% Ru/Al2O3 as methanation catalyst wasprepared by impregnation method and coupled with MSR step to evaluate the capability of methanol processor for CO reduction. Based onthe experimental results, the optimum conditions were obtained as feed flow rate of 5 mLh and temperature of 250 C; leading to a low COselectivity and high H2 yield. The designed reformer with catalyst coated layer was compared with the conventional packed bed reformer at thesame operating conditions. The constructed fuel processor had a good performance and excellent capability for on board hydrogen production.Key wordsspiral fuel processor; hydrogen; fuel cell; methanol steam reforming1. Introductioning [4]), steam reforming of methanol is a well. developed pro-cess with a high efficiency, as hydrogen is formed from bothHydrogen gas is seen as a potential future energy sourcemethanol and water. In this way hydrogen yield is typicallyby virtue of the fact that it is renewable, clean, liberateshigh, with hydrogen and carbon monoxide concentrations arelarge amounts of energy per unit weight in combustion, and70% -75% and 1%- 3%, respectively. However, the reactionis easily converted to electricity by fuel cells. Hydrogenis highly endothermic and requires external heat source. Fur-can be produced from both hydrocarbons and renewable re-thermore, if hydrogen is utilized as a fuel for polymer elec-sources. Hydrocarbon resources are more suitable option fortrolyte fuel cells (PEFC), the concentration of Co should beportable applications. Among different hydrocarbon fuels,decreased to Jess than 20 ppm, although some of those thatthose which have lower working temperature, higher hydro-operate at higher temperatures, are more tolerant to carbongen content, low byproducts generation, and more desirably inmonoxide poisoning, occupy a much smaller volume, andliquid form, are most suitable for hydrogen production. Withmay be much less expensive.this approach, liquid oxygenated hydrocarbons like alcoholsThree common processes can be used to further reduceare appropriate options. Methanol offers several advantagesCO in the feed: preferential or selective oxidation, metha-for hydrogen production compared with other hydrocarbonsnation and membrane separation [5]. Methanation has at-[{1,2]. Methanol, as a primary alcohol, is an excellent hydro-tracted much more attention in recent ycars, becausc there isgen carrier with a high hydrogen to-carbon ratio and one ofno need to oxidizing and/or mixing inert gases with the re-the largest bulk chemicals in the world. Fuel methanol is sul-formate stream, and the produced methane which is inert tofur free and can be produced from renewable resources, suchthe most of fuel cells, can be utilized in the afterburner. Theas biomass. Furthermore, the molecule contains no carbon-mechanism and kinetics of methanation reaction, its thermo-carbon bonds and is rather easily converted to a hydrogen-richdynamics, and catalyst deactivation have been investigated in-gas at moderate temperatures.tensively [6,7]. Ruthenium, in particular, is outstandingly ac-Despite diverse methods for hydrogen production fromtive even at low temperature (e.g. 160- 200 °C) and undermethanol, (i.e. partial oxidation [3] and autothermal reform-mild hydrogenotiannditin. t in orhaps, the most active中国煤化工●Corresponding author. Tel: +98-111-3234204; Fax: +98-111-3234201; E-mail: m. taghizThis work was financially supported by the lran National Science Foundation.iYHCNMH GCopyright@2012, Dalian Insiute of Chemical Physics, Chinese Academy of Sciences. All rights reseved.doi: 10.1016/S1003- 9953(1 1)60401-5Journal of Natural Gas Chermistry Vol. 21 No. 5 2012529der to prepare clear and transparent sols without any precipi-The coating steps and subsequent drying was repeated fortation. The molar ratio of D. D. W. to alkoxide was 5. Then,4 -5 times to obtain a catalytic layer with appropriate thick-a mixture of acetic acid and nitric acid was added such that ness. Finally, the substrate was calcined in air at 350 °C forthe molar ratio of acid solution to alkoxide was 0.2 and pH3h at a heating rate rate of 2 °C/min in order to decomposevalue of each solution was adjusted to 4.5. After all, 1 g HPCundesired species.was added to each sol and remained on ultrasonic bath for 3 h.Because no cerified method exists to quantify the adhe-Alumina sol makes a porous layer with a large surface area,sion of coatings on microstructures, in order to ensure a reli-while Titania sol makes a dense layer with a high mechanicalable coating adhesion under real working conditions, two testsstability.were conducted: (i) thermal shocks (20-800 °C) were appliedThus, a sol mixture with molar ratio of 1:1 was consid-in an oven in order to simulate the start-up and shut-down op-ered. The structured surface was covered with the sol mix-erating condition of an on-board reformer; (i) the substrateture using a syringe followed by removal of excess slurry inwas tested by mechanical stress that an on-board reactor maythe circulating air. Then, it was dried at room temperatureendure during utilization. For each test, the resulting weightfor one day and calcined in air at 500°C for 3h at a heat-loss was measured, in order to simulate the impact of relevanting rate of 5 °C/min in a tube furnace. In the second stage,stress. With the notice to the entire catalyst weight of 32 mgfor the preparation of catalytic slurries, a commercially avail-for methanol steam reforming and 148 mg for CO methana-able Cu/ZnO/Al2O3 catalyst (C79- 5GL, Sid-Chemie) in the tion, test results showed a weight loss of about 3%. Referringform of extruded pellets of 3 mm was used. For a stable andto literature data [30,31], a weight loss between 0 and 10% isuniform coating, the thickness of catalyst layer should be 20-considered as indication of a stable coating.40 pum. For this purpose, the catalyst powder was preparedby a bid milling apparatus (25000 rpm, IKA, Germany). Then2.6. Experimental setup and analysis of product gasesit was sieved with a stainless steel screen to provide a powderwith particle size of 5 um. Subsequently, the sol mixture (Alu-Schematic diagram of experimental setup is ilustrated inmina+Titania), nitric acid, catalyst powder and double dis-Figure 3. It essentially consisted of syringe pump (Korea,tilled water with the ratio of 5 mL:0.5 mL: 10g: 100 mLTOP5300, 0.1-1500 mL/h) for injection of feed containingwere mixed in a 300 mL ball mill jar with glass balls fordouble ditilledl water and methanol (Merck 99.99%), a com-3-4h to provide a homogeneous slurry for coating. Gaspact methanol processor system including vaporizer tube fol-assisted fluid-displacement method was done for the coat-lowed by a set of catalyticplates and units for the catalyticing of both methanol steam reforming and methanation mi-reduction process and gas chromatograph (GC) for analysiscrochannel plate. Each surface pre-coated with a catalyticof gases. The feed rate was varied from 3 mLh to 8 mLh.alumina/titania adhesive layer was covered with pertinent cat-The water : methanol ratio in the feed was 1.5:1 as it is gen-alyst slurry using a syringe, and then the majority of the fluiderally recommended moderate value for producing a low COwas forced out of the chamber by the measured air flow ratesconcentration. To provide heat for reaction, the microreformerranging from 50 to 100 sccm. Next, the catalyst layers werewas heated with a hot plate equipped by two cartridge heatersdried in the oven at 80 °C for 20 min.with maximum power of each unit 200 W. The temperature ofTemperaturecontrollerReduction facilitiesr⑦-83 WayvalveMFC N2 MFC H2Check !Heattrace号得GC apparatus7BallCheck ,Coldtrap品iSyringepump中国煤化工YHCNMHGFigure 3. Schematic flow diagram of experimental setup530Foad Mehri et al./ Joumal of Natural Gas Chemistry Vol. 21 No. 52012the catalytic wall and evaporator conduit was regulated by twoThe steam to carbon ratio was kept constant at 1.5: 1PID controllers (Autonics, TC4S) connected to two different(moles of water/moles of methanol) to achieve the highesttypes of thermocouples (K and P), which were set on the un-production rate of H2. Figure 4 shows the effect of weightderside of catalytic plates and the inlet stream, respectively,hourly space velocity (WHSV) on the conversion of methanolto monitor and control the reactor temperature. The fuel pro-and yield of hydrogen at two reaction temperatures. Methanolcessor was insulated in order to eliminate excessive heat lossconversions at both 220 °C and 250 °C were greatly decreasedfrom the system.with increasing WHSV due to the decrease of contact time.Prior to each run, the catalytic plates were reduced in aHowever, the conversion was increased with increasing re-gas mixture of 10% H2/N2 flowing at 300 mL/min measuredaction temperature at the same WHSV. This is because thatby a mass flow controller (MFC, Brooks 5850) and heated tohigher reaction temperature favors the improvement of the300°C for 3 h. This step was necessary for activate the cat-steam reforming reaction rate as well as the catalytic activ-alytic layer. The temperature of inlet feed evaporation sectionity. A maximum conversion was obtained at 250 °C and flowwas fixed at 120 °Crate of 3 ml ./h (WHSV = 82.2 h-l) of methanol/water liquidFirst, all the tests were performed with methanol steammixture. The conversion of methanol increased from 76.3% toreforming catalytic plate at three feed flow rates (3, 5 and91% with increasing reaction temperature from 220 to 250 °C.8 mL/h) and two reaction temperatures (220 and 250°C) toAlso, hydrogen yield decreased with the increase of WHSVevaluate the optimum conditions of commercial catalyst. Thewhich is due to the decrease in methanol conversion as it cansecond experiment was carried out by coupling two catalyticbe seen from Reaction(1). At a fixed WHSV, H2 yield wasplates which are aligned together (methanol steam reform-increased with increasing temperature; H2 yield was raised uping prior to methanation catalytic reaction). All conditionsto 72.1% at 250°C.were the same as the first experiment (feed flow rates andreaction temperatures for steam reforming process), except00 r80methanation reaction two other reaction temperatures (160and 190 °C) were used. In all experiments, the activity testswere conducted at atmospheric pressure; the feed gas stream)0 t75was passed through the catalytic plates and the product gasstream was directed through a cold trap to remove the liq-uid components. The exit dry gases were analyzed by anon-line gas chromatograph (GC), Varian CP- 3800, equippedwith MOL SIV 5A column and thermal conductivity detectors(TCD).0F55 :3. Results and discussionMethanol conversion, T=250 C50 tMethanol conversion, 7-220 CH2 yield, 7-250 C3.1. Performance of methanol microreformerH, yield, 1=220 C8010012014016018020022055Performance of individual sections of methanol microre-WHSV (h")former was examined at preliminary analysis of fuel processorto determine optimum operating conditions. Following threeFigure 4. Effects of WHSV and reaction termperature on methanol conver-reactions were found to be more effective during steam re-sion and H2 yield (S/C= 1.5, estimated catalyst weight = 32 mg)forming of methanol:Selectivities to CO2 and CO are shown in Figure 5 forMethanol steam reforming (SRM):various WHSV at two reaction temperatures. Generally, aCHgOH(g) + H20(g)→CO2(g) + 3H2(g)(1)microreformer with low CO selectivity is preferred. It is ob-SH298 = +49.4kJ/molserved that there was a tradeoff between CO2 and Co selectiv-ities: with WHSV increasing, the selectivity to co decreasedMethanol decomposition:while the selectivity to CO2 increased. Noting that increasein WHSV decreased methanol conversion, hence water use inCHsOH(g)←+ CO(g) + 2H2(g)(2)Reaction (1), the excess amount of unreacted water pushed0H298 = +90.13kJ/molReaction (3) in forward reaction, slightly increasing selectiv-ity to CO. Furthermore, at a constant WHSV, CO selectivityWater gas shift (WGS):was increased with an increase in reaction temperature whileCO(g) + H20←+ CO2(g)+ H2(g)that of CO2中国煤化工e to the thermokinetic(3)△H298 = -41.2kJ/molbehavior oflYHCNMH Gis exothermic, so anyincrease ines the reverse of wGSReaction (2) and reverse of Reaction (3) are responsiblereaction with more CO production. CO selectivity increasedfor CO production.from 1.4% at 220 °C to 2.8% at 250 °C which corresponds toJourmal of Natural Gas Chemistry Vol. 21 No. 52012531a WHSV from 219 to 82 h-1. CO2 sclectivity approximatelyTable 2. Optimum operating conditions of the micro fuel reformerchanged from 22.9% to 23 .9% when temperature decreasedcompared with that of fixed-bed reactorfrom 250 to 220°C with WHSV of 82h-1. From these re-ParametersMicroreformerFixed-bed reactorsults, operating the system at moderate WHSV (or feed flowReactor temperature (C)25250Feed flow rate (mL/h)5rate) and a higher temperature where conversion is relativelyPressure (atm)1.(1.0high, is more desirable. While CO concentration in reformateS/C (steam to carbon ratio)1.increases, a further processing afterward reduces the amountWHSV (h-')1370.37of CO produced. The optimal methanol feed rate was 5 mLhMethanol conversion (%)91.385(correspondingto WHSV = 137 h-l) which is the highest feedH2 yield6864rate with a methanol conversion above 90% at 250°C.CO2 slecivity (%)23.221.4CO selectivity (%)2.4Inner volume of reactor (cm')3.1425.0Catalyst amount (mg)323003.2. Performance of fuel processor wih both microreformnerand methanation catalyst coated microreactor。824.0Presence of 1.5- -2.8 mole percent of Co in off gas streamneeded additional treatment process to reduce its negative im-23.5 tpact on polymeric membrane fuel cell. Precious metal (e.g.Ru) methanation catalyst are often able to diminish Co con-centration from 0.9- -0.5 to below 50 ppm selctively, but in23.0十CO, setivity,T=250"C 1 1.0relatively higher levels of CO in gas stream (e.g. 1%-3%),十CO2 slectivit, 7-220 Tmulti-stage Co methanation has been used to reduce it down-0- CO seletivity, T=250 C一CO selectivity, T-220 Cto a ppm range [32]. In our experiments, a single stage2.5 Lmethanaton was used to estimate the ability of microreac-80 100 120 140 160 180 200 220tor and its influence on reformate stream purification. TheWHSV(h")following two reactions can be carried out over methanationFigure 5. Selectivities to CO2 and CO as a function of WHSV at two dfferentcatalyst in the presence of CO, CO2 and H2 which are maintemperatures (S/C= 1.5, estimated catalyst weight = 32 mg)gaseous components in the exit of methanol microreformer.A comparison of the catalyst-coated performance miCO(g) + 3H2(g)→CH4(g) + H2O(g)croreformer and fixed-bed reactor is presented in Table 2.4)SH298 = - 205.8kJ/molThis comparison was made to demonstrate the superior capa-bility of constructed microreformer when operated under itsCO2(g) + 4H2(g)→CH4(g) + 2H2O(g)(5)optimal condition to a fixed-bed reactor with the same feedSH298 = - 164.6kJ/molflow rate and reaction temperature. For this purpose, 0.3 gcatalyst was crushed, sieved (16- -25 mesh) and diluted withReaction (4) is most promising in selective methanationsilica prior to loading a tubular reforming reactor. Tubular re- reaction, but Reaction (5) as a side reaction occurs via metha-actor consisted of a stainless- steel pipe (i.d. 13 mm, lengthnation of CO. Since operating at low temperatures enhances900 mm, grade 321) surrounded by a 4 kW electrical heaterCO conversion, the performance test for methanation mi-tightly insulated so that the reactor temperature can be con-croreactor carried out at two reaction temperatures (160 andtrolled. Prepared catalyst was packed inside the reactor and190°C) while WHSV varied from 17.8 to 48.4h- 1 (corre-kept in place by a glass wool filter. Preset amount of methanolsponding feed flow rate of 3 to 8 mLh). Figure 6 ilustratesand water with steam to carbon ratio (S/C) of 1.5 was suppliedthe conversions of CO and CO2 as a function of WHSV andto the reactor by a precision syringe pump. Methanol and wa-temperature. It can be seen that both conversions of CO andter were evaporated by a vaporizer and entered the reactor atCO2 decreased as WHSV increased. Selectivities to CO and120 °C. Methanol conversion and other parameters were mea-CO2 varied from 74.2% to 93.6% and 2.6% to 6.8% respec-sured with the feed rate of 5 mL/h and reactor temperaturetively, with WHSV variation. Selectivity to Co was so supe-of 250 °C. Based on the obtained results, the microreformerrior to that of CO2 which indicates that methanation catalystwith a lower catalyst loading and also with the smaller sizehas a good selectivity toward CO. Before reaching a completeshowed a greater performance rclated to packed-bed rcactor. Itconversion of CO methanation, any increase in tcmperatureis revealed that the microreformer had a higher methanol con-leads to an i中国煤化工\lso CO2 conversionversion, hydrogen yield and CO selectivity than the packed-increases sligiY台O2 methanation rate.bed reactor. The microchannel, coated with the catalysts, en-Methanation.CNMH.S 89.5% of Co con-hanced the heat and mass transfer in a microreformer and thisversion at desirable WHSV of 29.6 h- 1 and 250 °C which hascan be seen in its higher measured performance.a low CO2 conversion of 3.7%. .532Foad Mehri et al./ Journal of Natural Gas Chemistry Vol. 21 No.5 2012Equation (6) is defined as the reactor efficiency in this100study:Pout(Chemical power)out90TreactorPim (Electrical power + chemical power)inThe power output is defined as the lower heating value30 t(LHV) of hydrogen (121 kJ/g) multiplied by the molar flowrate of hydrogen produced (nH2) in Equation (7). The input8power is defined as the electrical power to heat up the reactor(catalyst wall) plus the chemical energy of methanol supplied- co conversion, 7=190 C60 F. CO conversion, T=160 Cto the system, as shown in Equation (8). The methanol chem--0- CO2 conversion, T=190 Cical energy was calculated as the product of mtolal (premixO2 conversion, T-160 Cfeed mass flow rate), the mass fraction of methanol in the pre-mix (YCHzOH = 0.54), and the mass based LHV of methanol1520 25(EcHzOH = 19.93 kJ/g).WHSV (h*)Figure 6. Conversions of CO and CO2 versus WHSV at two dfferent tem-Pout = nH2 X EH,LHV(7)peratures for 148 mg of 0.5 wt% Ru/Al2O3 coated on microchannelPm = (R7R)etrical + (motal x YCHzOH x EchH,OH)chemialThe efctst of feed flow rate and temperature on the con-(8)centration of dry effluent gases of fuel processor which utilizeAs shown in Figure 8, with increasing feed flow rateboth methanol microreformer and methanation microreactorthe efficiency of fuel processor reached a maximum value ofare shown in Figure 7. At a constant temperature with in-79.6% and 79.1% respectively at 190 °C and 160°C of metha-creasing feed flow rate, the mole percents of CO and CH4 de-nation reaction termperature and then fell. This up and downcreased in contrary to those of CO2 and H2. With increasingbehavior was attributed to tradeoff between increment of H2feed flow rate, methanol conversion in first stage and CO con-generation rate and required electrical energy which paid aversion in second stage reduced. First case leads to low Copenalty for evaporation of additional unreacted water with in-and H2 concentration due to lower conversion of methanol viacreasing feed flow rate and temperature.Reaction (2) and second one lessens CO to be converted whichconsumes less H2. Therefore, Co and H2 concentrations in-85 -crease with low slop in exit from fuel processor. With decre-口Tsuw=250 c and Twe=190Cment of Co conversion, CH4 concentration was reduced and0F3 Tsu=220 C and Tsme=160 cCO2 concentration remained approximately the same. Hightemperature improves the methanation reaction, hence moreCH4 was produced and H2 level was decreased.Total changes in CO concentration were 0.21% to 0.44%and range of CH4 mole% was 1.7%- 4.8% for two reaction0temperatures. 72% H2 and 0.27% CO were obtained in theexit of fuel processor under best conditions at 250°C ands5 F5 mLh.60[80Liquid feed flow rate (mLh)70Figure 8. Effects of feed flow rate and temperature on eficicncy of microre-60+ H,7=250CformerH, 7-220 c-CO, T=250 C十3The electrical power with a typical fuel cell can be antic-，s(CO, 7-220Cipated by 60% efficiency of fuel cell and 80% utilization of40H2. With this assumption, power generated by fucl processorCo, 7-250 ccalculated for various feed flow rate and reaction temperature30-- co, T-220 Care ilustrated in Figure 9. In all cases, the produced power didnot exceed 7.4 W. It is observed that the power generated by20fuel processor increased with increasing feed flow rate, whichwould incre中国煤化工duction. It is observedthat at the iCN M H Geticiencor fuelpefficiency of fuel pro-Liquid feed flow rate (mL/h)Figure 7. Miroreactor exit gas composition (dry basis) with respect to feedis believed that CO2 methanation was retarded at higher tem-flow rate at two reaction temperatures using SRM and methanation catalystsperature in feed flow of 8 mL/h. Based on the performanceJourmal of Natural Gas Chemistry Vol. 21 No.5 2012533assessment of constructed fuel processor, a high efficiencyReferenceswith low CO concentration is desirable. So optimal operatingconditions were considered in feed rate of 5 mLh at 250°C[1]ChenYz,WangYz,XuHY,XiongGX.JMembrSci,2008,which resulted in 79.1% efficiency with hydrogen generation322: 453of 0.17 mole/h and CO mole percent of 2.7%. Regarding the[2] Tosti s, Basile A, Borgognoni F, Capaldo V, Cordiner s, Di Caverate of hydrogen production, the anticipated power outlet ofS, Galucci F, Rizzello C, Santucci A, Traversa E. J Membr Sci,constructed fuel processor is expected to be around 5.2 W.2008, 308: 250[3] LiCL, Lin Y C. Catal Lett, 2010, 140: 69[4] Choi K S, Choi IJ, Hwang S J, Kim H M, Dorr JL, Erickson PA. Int J Hydrogen Energy, 2010, 35: 6210口Tsev=250 C and Tswm=190 C5] Song C s. Caral Today, 2002,77: 178X Tsu-220 C and Tmen=160 c[6] Bartholomew C H. Catal Rev-Sci Eng, 1982, 24: 67[7] Van Herwijnen T, Van Docsburg H, De Jong W A.J Catal, 1973,28: 391[8] Galetti C, Specchia s, Saracco G, Specchia V. Chem Eng Sci,2010, 65: 590[9] Urasaki K, Endo K, Takahiro T, Kikuchi R, Kojima T, SatokawaS. Top Catal, 2010, 53: 70710] Panagiotopoulou P, Kondarides D I, Verykios X E. J Phys ChemC, 2011, 115: 1220[11] VanderWiel D P, Pruski M, King TS. J Catal, 1999, 188: 186[12] Lim M s, Kim M R, Noh J, Woo S I. J Power Sources, 2005,140: 66Liquid feed flow rate (mLh)[13] Kwon 0 J, Hwang S M, Ahn J G, Kim J J. J Power Sources,Figure 9. Calculated power output from fuel processor at different feed flow2006, 156: 253rates and temperatures[14] KimT, Kwon S. J Micromech Microeng, 2006, 16: 1760[15] Wang X, Zhu J, Bau H, Gorte RJ. Catal Lett, 2001, 77: 173[16] Cominos V, Hardt S, Hessel V, Kolb G, Liwe H, Wichert M,4. ConclusionsZapf R. A Methanol Steam Microreformer for Low Power FuelCell Applications. In Proceedings of the 6th Intemational Con-ference on Microreaction Technology, IMRET 6, AIChE Pub.In this work, design and fabrication of two spiral mi-No. 164, New Orleans, 2002. 113crochannel is developed to investigate the performance of fuel[17] Besser R s, Ouyang X, Chen H, Shin W C, Bednarova L, Lee W,processor on H2 production in fuel cell application. Prelim-PauS, Pai C S, Taylor J A, Mansfield W M, Ho P. Preferentialinary analysis is performed by microreformer individually.Oxidation in Microchannel Reactors for Scalable Fuel Process-Then a comparison has been made between a fixed-bed re-ing. In Proceedings of the AIChE Spring Meeting, 2004, 171aactor and microreformer under optimal operating conditions[18] Denkena B, Hoffmeister H W, Reichstein M, llenseer s, Hlavacto prove the superior performance of microreformer. It hasM. Microsyst Technol, 2006, 12: 659been shown that microreformer which has lower catalyst load-[19] Samant A N, Dahotre N B. J Eur Ceram Soc, 2009, 29: 969ing and smaller volume could have lower CO selectivity and[20] Cao C S, Wang Y, Holladay J D, Jones E O, Palo D R. AIChEJ,higher conversion and H2 yield. So it can be concluded that2005, 51: 982microreformer enhances the heat and mass transfer in com-[21] Pan L w, Wang S D. Chem Eng J, 2005, 108: 51[2] Kundu A, JangJH, Lee HR, KimSH, GilJH, JungCR, Ohparison with fixed-bed reactor.Y s. J Power Sources, 2006, 162: 572Finally subsequent assessments were made by coupling[23] JangJ Y, Huang Y X, Cheng C H. Chem Eng Sci, 2010, 65:microreformer with methanation microreactor.First mi-5495crochannel was coated with a commercial Cu/ZnO/Al2O3 cat-[24] Mies M J M, van den Bosch JL P, Rebrov E V, JansenJC, dealyst (C79 5GL, Suld-Chemie) for methanol steam reform-Croon M H J M, Schouten J C. Catal Today, 2005, 110: 38ing reaction and the second one was coated with a 0.5 wt%[25] Shingu H, Bhuiyan M M H, Ikegami T, Ebihara K Thin SolidRu/Al2O3 as synthesis catalyst for CO methanation. The Per-Films, 2006, 506-507: 111formance evaluation of the constructed fuel processor wa[26] Surangalikar H, Ouyang X, Besser R S. Chem Eng J, 2003, 93:measured under various operating conditions.217Optimal methanol conversion was 91.3% at 250 °C with[27] Men Y, Gnaser H, Zapf R, Hessel V, Ziegler C, Kolb C. ApplCatal A, 2004, 277: 835 mL/h methanol feed rate. The efficiency was 79.1% and[28] Chen H, Bednarova L, Besser R S, Lee W. Appl Catal A.2005,production rate of hydrogen was approximately 0.17 mole/h286: 186that is sufficient amount to generate 5.2 w electric power for[29] Tada S,中国煤化工aS. App Catal A.211.1a typical fuel cell.404: 14[30] Boix AYHCNMHGA,MiroEE.ApplCatalAcknowledgementsB, 2003, 46: 121This work was financially supported by the Iran National Sci-31] Valentini M, Groppi G, Cristiani C, Levi M, Tronconi E, Foratience Foundation (INSF).P. Catal Today, 2001, 69: 30732] Echigo M, Tabata T. J Chem Eng Jap, 2004, 37: 75