Hydrogen Production from Methanol Steam Reforming over Cu/ZnO/Al O /CeO /ZrO Nanocatalyst in an Adiabatic Fixed-Bed Reactor

Transcription

1 Iranica Journal of Energy & Environment 3 (4): , 0 ISSN IJEE an Official Peer Reviewed Journal of Babol Noshirvani University of Technology DOI: 0.589/idosi.ijee BUT Hydrogen Production from Methanol Steam Reforming over Cu/ZnO/Al O /CeO /ZrO Nanocatalyst in an Adiabatic Fixed-Bed Reactor 3 Hassan Sharifi Pajaie, Majid Taghizadeh and Ali Eliassi Chemical Engineering Department, Babol University of Technology, P.O. Box 484, Babol, Iran Chemical Technologies Research Department, Iranian Research Organization for Science and Technology (IROST), Tehran, Iran (Received: September 8, 0; Accepted in Revised Form: December 3, 0) Abstract: Hydrogen production from steam reforming of methanol was performed in an adiabatic fixed bed heterogeneous reactor by using Cu/ZnO/AlO 3/CeO /ZrO nanocatalyst. The catalyst was prepared by co-precipitation method and characterized by TGA, XRD, SEM and BET. By changing the mean average temperature of the catalyst bed (or reactor operation temperature) from 30 to 70 C, variations in methanol conversion were monitored. The results showed that the conversion of methanol was strongly dependent on the reaction temperature. In addition, the effects of weight hourly space velocity (WHSV) on methanol conversion were investigated. According to the results, the maximum conversion was obtained at 70 C with WHSV of 5 h. Key words: Hydrogen production Methanol steam reforming Co-precipitation Nanocatalyst Fixed-bed reactor INTRODUCTION decentralized refueling units for hydrogen-based automobiles or as on-board generation systems for Catalytic steam reforming of methanol is a hydrogen-based internal combustion engines or PEM fuel well-established process used for the production of cells [5]. hydrogen [ 4]. This process is especially important for There are three process alternatives for the polymer electrolyte fuel-cells, which generate electrical conversion of methanol to hydrogen [8]: power by electrochemical oxidation of hydrogen with atmospheric oxygen [5]. The use of methanol as an Decomposition, on-board hydrogen source is attractive for fuel-cell Steam reforming, engines in transportation applications because of its Partial oxidation. safe handling, low cost and ease of synthesis from a variety of feed stocks (biomass, coal and natural gas) [6]. The decomposition reaction is a strong endothermic Moreover, methanol has been recommended as the best reaction: source for hydrogen fuel among the high energy density liquid fuels, due to the high H/C ratio having a lower CH 3 OH CO + H H 0 =8 kj/mol () propensity for soot formation than other hydrocarbons, relatively low boiling point and easy storing [7]. The process yields high CO contents, which Hydrogen production systems via steam reforming of makes the process unsuitable for fuel cell applications. methanol (SRM) constitute a promising technology for The steam reforming process, Eq. (), is also endothermic; energy feeding in portable electronic devices, for however it yields CO as the major by-product and Corresponding Author: Majid Taghizadeh, Chemical Engineering Department, Babol University of Technology, P.O. Box 484, Babol, Iran. Tel: ; Fax: , m_taghizadehfr@yahoo.com. 307

2 can produce a high H content, up to 75% on a dry to increase the Cu dispersion and the thermal stability of basis, which makes it very favorable for fuel cell the catalysts and even to reduce the CO concentration applications: [6-8]. CH 3 OH + HO CO + 3H H 0 =3 kj/mol () In this work, Cu/ZnO/AlO 3/CeO /ZrO nanocatalyst was prepared by co-precipitation method and used for the steam reforming of methanol in an adiabatic fixed bed Partial oxidation, Eq. (3), is a strong exothermic reactor. Methanol conversion and the catalyst bed reaction, which can yield an automotive system with a fast temperature changes in a laboratory-scale system at response compared to the steam reforming reaction. various operating temperatures and various WHSVs were However, the system can only iver 66% hydrogen when investigated. pure oxygen is used and as low as 4% when air is supplied instead of oxygen, which is the realistic Experimental alternative in an automotive system [8-5]: Catalyst Preparation: The oxide precursors Cu/ZnO/AlO 3/CeO /ZrO with approximate composition of (weight ratio) Cu/Zn/Al/Ce/Zr: 40/50/5/.5/.5 was CH3OH + O CO + H H 0 = -55 kj/mol prepared by the conventional co-precipitation method. (3) A mixed aqueous solution of copper nitrate 3-hydrate (Cu(NO 3).3HO), zinc nitrate 3-hydrate (ZnO(NO 3).3H O), The steam reforming of methanol for production of H zirconium nitrate 8-hydrate (Zr(NO 3).8HO), aluminum from CH3OH involves three reactions: nitrate 9-hydrate (Al(NO 3) 3.9HO), cerium nitrate 3-hydrate (Ce(NO 3).3HO) (each M) and a solution of sodium CH3OH + HO CO + 3H (4) carbonate ( M) were added slowly and simultaneously CH3OH CO + H (5) into 300 ml of deionized water at 65 C with vigorous stirring. The ph was kept constant at 7. The precipitates CO + HO CO + H (6) were aged in mother liquid for 3.5 h at the same temperature under stirring, washed with deionized water + -3 to remove Na and NO in the filtrate and then dried at These are the steam reforming reaction (4), the 383 K for 3 h and calcined by air at 63 K for 4 h (heating methanol decomposition reaction (5) and the water gas rate 5 C min ). Each calcined catalyst was pelletized in a shift reaction (6). Reaction (4) is the preferred pathway of hydraulic press, crushed and sieved into a particle size of H production and reaction (6) plus the reverse of reaction mm and used as catalyst for steam reforming of (5) should be suppressed to prevent CO formation. methanol reaction. The platinum anode in the PEM fuel cell gathering is sensitive to CO poisoning at concentrations as low as Catalyst Characterization: The proper temperature of the 0 ppm, so the aim is to produce CO-free H to the calcination process of the dried samples was determined greatest extent possible [8, 6]. The major advantage of by means of a thermo gravimetric analyzer using Diamond SRM over MD (methanol decomposition) and POM TG/DTA instrument (PerkinElmer, USA). In order to avoid (partial oxidation of methanol) is the high H selectivity, any diffusion effect and formation of thermal gradients, which makes it favorable for fuel cell applications [7, 8]. the amount of the used samples were ranged from 0 to 5 In the literature, copper-based catalysts have received mg. The catalyst samples were heated from 5 up to 550 C considerable attention for production of H by SRM at C/min with air flow rate of 00 ml/min, till no weight [9-]. To improve the efficiency of the catalysts for the SRM loss occurred. X-ray diffraction (XRD) patterns for the reaction, several researchers have introduced a third oxide particles were obtained on a Bruker D8 instrument metallic oxide, ZrO, into CuO/ ZnO/AlO 3 catalysts. with monochromatic Cu Ka radiation ( =.5406 Å) The promoting effect of ZrO has been attributed to the operated at 40 kv and 30 ma, scanning from 0 to 80. improvement in reducibility [-4]. It is reported that The specific surface area, pore size distribution and addition of ZrO improves the specific surface area, Cu pore volume of the sample was measured using the surface area and Cu dispersion of catalyst and prevents Brunauer-Emmett-Teller (BET) method by nitrogen the sintering of Cu, improving the activity and stability of adsorption using a BEL SORP-max instrument the catalyst [5]. Likewise, CeO has also been applied (BEL, JAPAN). Prior to adsorption measurements, 308

3 Fig. : A schematic diagram of the experimental apparatus for catalytic production of H from methanol adiabatic fixed-bed reactor. The axial reactor temperature at any point of the catalyst bed was measured via a thermo-well using a thermocouple. The reactor outlet products were passed through an air cooler and a double pipe heat exchanger to cool down to the ambient temperature. The cooled products were sent to a gasliquid separator. A back pressure regulator (BP- LF690, pressure Tech 000, England) was placed on the separator to regulate the system pressure. Gaseous reaction product was sent to gas chromatograph analyzer (GC) (Varian CP-3800) for online measurement using TCD and FID detectors by packed column (HYSEP Q, Custom made, Iran) and capillary column (CP-Sil 8 CB, Fig. : Temperature profile inside the reactor in a blank Varian, USA). Also, the remaining methanol in the exit experiment (without any reaction) product was measured and the methanol conversion was calculated with respect to entrance methanol. the samples were degassed for at least 5 h at 300 C. To confirm the adiabatic operation of the applied Scanning electron microscopy (SEM) was performed in a reactor used, an experiment was performed as a blank run XL30 (Philips, Poland) microscope, operated at 5 kv. (without any reaction) and the temperature across the reactor bed was measured using thermocouple and Experimental Setup: The schematic diagram of the thermowell. The result of this experiment is shown in employed setup is shown in Fig.. Catalytic tests were Fig.. According to the illustrated data, it is reasonable performed in a fixed-bed continuous flow reactor with to assume that the reactor behaves adiabatic reactor. 5 cm height operated under atmospheric pressure. Before This experiment was performed by passing air through the each run, the catalysts were pre-reduced in a stream of catalyst bed. 0% H /N (500 ml/min) at the normal pressure according the following heating program: heated from room RESULT AND DISCUSSION temperature to 300 C with heating rate 0 C/min and was kept for at the stated temperature for hours. The molar Thermal Analysis: Fig. 3 shows the TG/DTA plot of a ratio of HO:CH3OH in the feed was.3:. Water and synthesized sample (Cu/ZnO/AlO 3/CeO /ZrO ) prepared methanol mixture was pumped from a storage tank at by co-precipitation method. This figure shows that different rates from 00 mlh up to 400 mlh to an pre-catalytic exothermic reaction causes catalyst evaporator and then to a supper heater before entering weight reduction to about 76% of its initial dried weight. the reactor. The superheated methanol was sent to an This catalyst has a weight lost between 5-00 C due to 309

4 Fig. 3: XRD patterns of Cu/ZnO/AlO 3/CeO /ZrO Fig. 5: SEM image of the Cu/ZnO/Al O /CeO /ZrO catalyst after calcination catalyst 3 heterogeneous catalyst [9, 30]. TGA/DTA analysis showed that when the temperature is set at 350 C, almost 87% of the catalyst is calcined and sintering of copper particles is avoided. Phase analysis: The XRD pattern of the synthesized Cu/ZnO/AlO 3/CeO /ZrO sample with co-precipitation method is shown in Fig. 4. This figure indicates that the sample calcined at 350 C, has a crystallite CuO phase. The crystalline structure is a characteristic that is strongly dependent on the condition of calcination catalyst construction method. The average crystallite size was calculated using Scherrer equation: Fig. 4: XRD patterns of catalyst prepared with coprecipitation method after calcination k D = (7) cos desorption of humidity and crystallized water on the where k is a constant ~0.9, is the wavelength of surface and structure of the catalyst. The major mass the X-ray, is the full width of diffraction peak at reduction appeared at temperatures between C half maximum intensity (FWHM) and is the diffracting that is due to release and evaporation of volatile angle. compounds as well as conversion of metal compounds formed during co-precipitation to stable metallic oxides. Morphology: Fig. 5 shows the SEM images of the In addition, as shown in the TGA graph, there is a small prepared catalyst by co-precipitation method in an mass reduction at temperatures above 300 C verifying the aqueous medium after calcinations. According to this residues of additional volatile compounds being figure, the crystals of the sample have good dispersion completely removed and the catalyst reached to stable and the morphology of these crystals is almost regular. structure. This result confirms the crystalline particle size obtained The DTA curve presents one major peak at 300 C by XRD analysis. that is exothermic resulting from the elimination of humidity and volatile compounds. Higher temperatures Bet and Pore Size Distribution Analysis: Physical (above 400 C) resulted in sintering of Cu-base particles properties of the selected catalysts (specific surface area, which contributed to reduction of fine pores, hence, total pore volume and average pore diameter) are decrease in active surface area. The dried deposition presented in Table. The average crystallite size was of Cu/ZnO/AlO 3catalysts is often calcined for 3 to 0 calculated using Scherrer equation and was estimated to hours at a temperature between C to achieve be about 5.63 nm. 30

5 Table : Physical properties of Cu/ZnO/Al O /CeO /ZrO catalyst 3 3 a Specific surface area (m /g) Total pore volume (cm /g) Average pore diameter (nm) Crystallite size (nm) a determinate by XRD results Fig. 6: (a) N adsorption/desorption isotherm and (b) pore size distribution of the sample Fig. 7: Methanol conversion versus inlet feed Fig. 9: Temperature profile along the catalyst bed at temperature various inlet feed temperatures and WHSV = 30 h. Solid lines show the trend of these variations Fig. 8: Temperature profile along the catalyst bed at various inlet feed temperatures and WHSV = 5 h. Solid lines show the trend of these variations Fig. 0: Catalyst bed average temperature (operating temperature) versus inlet feed temperature. Solid symbols are experimental data and solid line shows the trend of changes The pore size distributions and N adsorption/desorption isotherm of the sample are of mesoporous materials. According to IUPAC shown in Fig. 6. The nitrogen adsorption/desorption classification, the hysterics loop is type H occurs at isotherms can be classified as a type VI isotherm, typical a relative pressure range (p/p = ), indicating 3 0

6 a broad pore size distribution with uniform size and ACKNOWLEDGMENT shape. This work was financially supported by Iranian Catalytic Reaction: The obtained experimental data are Nanotechnology Initiative Council. shown in Figs In Figs. 8 and 9 variations of temperature versus reactor length for different WHSVs are REFERENCES presented.. Damle, A.S., 008. Hydrogen production by reforming The Temperature Effect: The first series of experiments of liquid hydrocarbons in a membrane reactor for were conducted to investigate the effect of operating portable power generation model simulations. J. temperature of the reactor on methanol conversion. Power Sources, 86: According to Fig. 6, it can be observed that conversion of. Löffler, D.G., K. Taylor and D. Mason, 003. A light methanol was increased by increasing the inlet feed hydrocarbon fuel processor producing high purity temperature. Since the reaction is endothermic and the hydrogen. J. Power Sources, 7: reactor was adiabatic, the temperature of catalyst bed 3. Ahmed, S. and M. Krumpelt, 00. Hydrogen from decreased from the feed inlet temperature, as illustrated in hydrocarbon fuels for fuel cells. Int. J. Hydrogen Figs. 8 and 9. Energy, 6: The change in catalyst bed average temperature 4. Telotte, J.C., J. Kern and S. Palanki, 008. (reactor operating temperature) versus inlet feed Miniaturized methanol reformer for fuel cell powered temperature is shown in Fig. 0. According to this Figure, mobile applications. Int. J. Chem. Reactor Eng., 6: there seems to be a relatively linear correlation between A64. the change in the reactor operating temperature and the 5. Lindstrom, B. and L.J. Pettersson, 00. Hydrogen feed inlet temperature. generation by steam reforming of methanol over copper-based catalysts for fuel cell applications. Int. Effect of Whsv on Methanol Conversion: The influence of J. Hydrogen Energy, 6: WHSV on the conversion of methanol at four different 6. Agarwal, V.V., S. Patel and K.K. Pant, 005. H inlet feed temperatures (30-70 C) is also shown in production by steam reforming of methanol over Figs. 7 and 8. At a constant inlet temperature, methanol Cu/ZnO/AlO3 catalysts: transient deactivation conversion increases with decreasing WHSV. Increasing kinetic modeling. Applied Catalysis A: Gen., the WHSV will cause an increase in the gas velocity, 79: which promotes mass transfer but leads to a decrease in 7. Thomas, S. and M. Zalbowitz, 999. Los Alamos the contact time of reactant species. As a result, the National Laboratory, New Mexico, Report No. conversion of methanol increases with decreasing WHSV. LA-UR Lindström, B. and L.J. Pettersson, 00. Development CONCLUSIONS of a methanol fuelled reformer for fuel cell applications. Proc. Fuel Cells-Science and A Cu/ZnO/AlO 3 nanocatalyst was prepared by Technology, Scientific Advances in Fuel Cell co-precipitation method, modified by adding small Systems. amounts of Zr and Ce oxides. This catalyst was used for 9. Agrell, J., K. Hasselbo and M. Boutonnet, 000. the production of hydrogen by methanol steam reforming Preparation and characterization of Cu/ZnO and in an adiabatic packed bed reactor at different inlet feed Pd/ZnO catalysts for partial oxidation of methanol: temperatures ranging from 30 to 70 C. The influence of Control of catalyst surface area and particle size WHSV on the conversion of methanol and bed using microemulsion technique. Studies in Surface temperature was investigated. The results showed that Science and Catalysis, 30: the methanol conversion increased with decreasing 0. Alejo, L., R. Lago and J.L.G. Fierro, 997. Partial WHSV and the temperature of catalyst bed oxidation of methanol to produce hydrogen over decreased from the feed inlet temperature along the Cu-Zn-based catalysts. Applied Catalysis A: Gen., adiabatic reactor. 6:

7 . Cubeiro, M.L. and J.L.G. Fierro, 998. Partial oxidation. Matter, P.H., D.J. Braden and U.S. Ozkan, 004. Steam of methanol over supported palladium catalysts. reforming of methanol to H over nonreduced Applied Catalysis A: Gen., 68: Zr-containing CuO/ZnO catalysts. J. Catalysis,. Cubeiro, M.L. and J.L.G. Fierro, 998. Selective 3: production of hydrogen by partial oxidation of 3. Matter, P.H. and U.S. Ozkan, 005. Effect of methanol over ZnO-supported palladium catalysts. J. pretreatment conditions on Cu/Zn/Zr-based catalysts Catalysis, 79: for the steam reforming of methanol to H. J. 3. Huang, T.J. and S.W. Wang, 988. Kinetics of partial Catalysis, 34: oxidation of methanol over a copper_zinc catalyst. 4. Velu, S., K. Suzuki, M. Okazaki, M.P. Kapoor, T. Osaki Appl. Catal. A: Gen, 40: and F. Ohashi, 000. Oxidative steam reforming of 4. Newson, E., P. Mizsey, T. Truong and P. Hottinger, methanol over CuZnAl(Zr)-oxide catalysts for the 000. The autothermal partial oxidation kinetics of selective production of hydrogen for fuel cells: methanol to produce hydrogen, Studies in Surface Catalyst characterization and performance evaluation. Science and Catalysis, 30: J. Catalysis, 94: Velu, S., K. Suzuki and T. Osaki, 999. Selective 5. Samei, E., M. Taghizadeh and M. Bahmani, 0. production of hydrogen for fuel cells via oxidative Enhancement of stability and activity of steam reforming of methanol over CuZnAl(Zr)-oxide Cu/ZnO/AlO3 catalysts by colloidal silica and metal catalysts. Applied Catalysis A: Gen, 3: oxides additives for methanol synthesis from a 6. Agrell, J., H. Birgersson and M. Boutonnet, 00, CO-rich feed. Fuel Process. Technol., 96: Steam reforming of methanol over a Cu/ZnO/AlO3 6. Liu, Y., T. Hayakawa, K. Suzuki, S. Hamakawa, catalyst: a kinetic analysis and strategies for T. Tsunoda, T. Ishii and M. Kumagai, 00. Highly suppression of CO formation. J. Power Source, active copper/ceria catalysts for steam reforming of 06: methanol. Applied Catalysis A: Gen., 3: Peppley, B.A., J.C. Amphlett, L.M. Kearns and 7. Patel, S. and K.K. Pant, 007. Selective production of R.F. Mann, 999. Methanol steam reforming on hydrogen via oxidative steam reforming of methanol Cu/ZnO/AlO 3. Part : the reaction network. Applied using Cu-Zn-Ce-Al oxide catalysts. Chem. Eng. Sci., Catalysis A: Gen., 79: -9. 6: Agrell, J., H. Birgersson, M. Boutonnet, I. Melian- 8. Meshkini, F., M. Taghizadeh and M. Bahmani, 00. Cabrera, R.M. Navarro and J.L.G. Fierro, 003. Investigating the effect of metal oxide additives on Production of hydrogen from methanol over Cu/ZnO the properties of Cu/ZnO/AlO 3 catalysts in methanol catalysts promoted by ZrO and AlO 3. J. Catalysis, synthesis from syngas using factorial experimental 9: design. Fuel, 89: Liu, Y., T. Hayakawa, K. Suzuki, S. Hamakawa and 9. Sloczynski, J., R. Grabowski, P. Olszewski, T. Tsunoda, 00. Highly active copper/ceria A. Kozlowska, J. Sloczynsk, M. Lachowska and catalysts for steam reforming of methanol. Applied J. Skrzypek, 006. Effect of metal oxide additives on Catalysis A: Gen., 3: the activity and stability of Cu/ZnO/ZrO catalysts in 0. Oguchi, H., T. Nishiguchi, T. Matsumoto, H. Kanai, the synthesis of methanol from CO and H. Applied K. Utani, Y. Matsumura and S. Imamura, 005. Steam Catalysis A: Gen., 30: reforming of methanol over Cu/CeO /ZrO catalysts. 30. Wu, J., M. Saito, M. Takeuchi and T. Watanabe, Applied Catalysis A: Gen., 8: The stability of Cu/ZnO-based catalysts in. Fukuhara, C., H. Ohkura and Y. Kamat, 008. methanol synthesis from a CO-rich feed and from a Physicochemical properties of a plate-type CO-rich feed. Applied Catalysis A: Gen., 8: copper-based catalyst, prepared on an aluminum plate by electroless plating, for steam reforming of methanol and CO shift reaction. Applied Catalysis A: Gen., 344: