Hydrogen can be used as a fuel for efficient production of electricity via fuel cells. Proton exchange membrane fuel cell (PEMFC) fed by CO-free hydrogen produced via a fuel processor (FP) is a promising option for short- and mid-term decentralized combined heat and power (CHP) stations. The combined FP-PEMFC system is a very convenient solution for producing energy to be used in residential and small scale business applications [1]. On the other hand, the FP prototypes - available commercially or studied in scientific papers -have very complicated control and structure, and they are not easy-to-operate systems. A conventional fuel processor prototype includes three reaction units in series - namely reforming, water-gas shift (WGS), and preferential oxidation (PROX) - in order to supply continuous CO-free hydrogen to the fuel cell. Imposed by the reversible nature of WGS reaction, the WGS unit has two reactors at different temperatures. Typical designs of shift reactors are in stages where high temperature shift (HTS) at 620-720 K is followed by a low-temperature shift (LTS) at 420-520 K, which together can decrease the CO level to less than 1% [2]. The only way to design an easy-to-operate FP is the use of catalysts having superior activity and selectivity specs for all three reactions. Additionally, the optimization of a FP should be performed in a combined fashion, which considers specs of the catalysts for each one of those three reactions as well as the operation and design parameters of the whole system. One alternative is to design catalysts which combine the reactions and consequently the units of the FP. A catalyst which enables reforming at lower temperatures – i.e. in the temperature range of HTS - and has both autothermal reforming (ATR) and WGS activity would combine reforming and HTS reactions in a single unit. Since all streams present for HTS unit in conventional design are eliminated, the control and operation of the overall system would be easier. Additionally, a design of a smaller FP unit with enhanced energy efficiency would be possible. A bimetallic Pt-Ni/Al2O3 catalyst was tested on ATR of propane and LPG (75% propane and 25% n-butane) in our laboratory, and the catalyst showed a superior performance in comparison to monometallic catalysts [3, 4]. The tests were performed between 620 and 740 K. The catalyst has not only permitted a decrease in the reaction temperature down to 670 K but also allowed high activity and hydrogen to carbon monoxide selectivity level (between 35 and 50). In those studies, the superior performance of the bimetallic Pt-Ni system at relatively low temperatures was explained by the presence of close - but distinct - Pt and Ni sites. The heat produced on Pt sites, which mostly catalyzed exothermic total oxidation (TOX), can be reaily transferred to Ni sites, where endothermic SR occurs [3, 4].

This work is a part of the studies conducted in our laboratory aiming to combine the reforming and HTS reactors as a single unit and, therefore, to decrease the total number of reactors to two prior to PROX unit (i.e. one reforming-HTS unit and one LTS unit). The bimetallic catalyst that we have employed provides an important advantage as it drops the reforming temperatures from ca. 1020 to ca. 670 K. The tested 620-740 K temperature range coincides with that of HTS. In the experimental studies, autothermal reforming of propane was tested over four Pt-Ni/-Al2O3 bimetallic catalysts aiming to investigate the effect of Ni:Pt loading ratio on catalyst performance. Reaction temperature, carbon to oxygen ratio, and residence time were taken as the reaction parameters. The effect of C/O2 ratio on the hydrogen production and H2/CO selectivity was found dependent on the Pt and Ni loadings. The highest hydrogen production and H2/CO ratio levels were obtained for the highest C/O2 ratio tested. Lower residence times promote hydrogen production rate mainly because of the higher flow rate of fresh hydrocarbons. On the other hand, if one seeks high H2/CO ratio, medium level W/F ratios –which enhance WGS reaction- should be preferred. Lower Pt loadings have made the catalysts more sensitive to the changes in C/O2 ratio. An optimum Ni:Pt ratio was found around 50 due to suppressed methanation and enhanced hydrogen production activities of these catalysts.

KEYWORDS: Hydrogen Production, Indirect Partial Oxidation; Catalyst Optimization; Propane; Pt-Ni Bimetallic Catalyst.

REFERENCES [1] A.K. Avcı, D.L. Trimm, A.E. Aksoylu, Z.İ. Önsan, Appl. Catal. A. 258 (2004) 235 [2] G. Jacobs, U.M. Graham, E. Chenu, P.M. Patterson, A. Dozier, B.H. Davis, J. Catal. 229 (2004) 499 [3] B.S. Çağlayan, A.K. Avcı, Z.İ. Önsan, A.E. Aksoylu, Appl. Catal. A. 280 (2005) 181 [4] B.S. Çağlayan, Z.İ. Önsan, A.E. Aksoylu, Catal. Lett. 102 (2005) 63

Extended Abstract Status: Not Uploaded