International Journal of Hydrogen Energy 32 (2007) 4040 – 4051 www.elsevier.com/locate/ijhydene Model-based investigation of a CO preferential oxidation reactor for polymer electrolyte fuel cell systems F. Cipitì ∗ , L. Pino, A. Vita, M. Laganà, V. Recupero Institute CNR-ITAE, Via S. Lucia sopra Contesse n. 5, 98126 S. Lucia, Messina, Italy Received 28 February 2007; received in revised form 26 April 2007; accepted 26 April 2007 Available online 6 June 2007 Abstract This paper deals with a two-dimensional model of a preferential oxidation (PROX) reactor to be used in a beta 5kWe hydrogen generator, named HYGen II, to integrate with polymer electrolyte fuel cells (PEFCs) for residential applications. The reactor geometrical configuration developed is a single-stage multi-tube configuration, in which a cocurrent air flow in the interspace is aimed at improving heat transfer and consequently controlling the temperature of the reactor. The aim of the model is to investigate the process performance of the reactor in order to enhance optimization and control of the PROX unit. The model concerns chemical kinetics and heat and mass transfer phenomena in the reactor. The model plays a key role in overcoming the issues of system design, by evaluating the temperature and the gas concentration profiles in the reactor. The CO removal from simulated reformate was examined with varying inlet temperature. Simulation results showed the strong dependence of the overall performance upon the reactor geometrical configuration.  2007 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved. Keywords: Fuel processor; CO preferential oxidation; Hydrogen production; PEFC; Two-dimensional catalytic modeling 1. Introduction During the last few years fuel cells have been considered to be promising tools for energy conversion from both industrial R&D departments and academia. The polymer electrolyte fuel cell (PEFC) fuelled by hydrogen appears to be the key option for both transport and small scale combined heat and power applications, due to its compactness, modularity, higher con- version efficiencies and low emissions of noise and pollutants [1,2]. A growing interest for small stationary applications (in the 0.5–10 kW electrical output range) is developing, with a large increase in the number of installed units in the world, as a decen- tralized power supply, grid support, peak shaving, power back- up or uninterruptible power supply (UPS), can be derived [3]. The absence of a hydrogen refuelling infrastructure and prob- lems concerning hydrogen storage, has led to the development of fuel processors able to convert available fuels (hydrocar- bons and/or alcohols) into hydrogen rich reformate gas [4]. ∗ Corresponding author. Tel.: +39 090 624 297; fax: +39 090 624 247. E-mail address: francesco.cipiti@itae.cnr.it (F. Cipitì). 0360-3199/$ - see front matter  2007 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijhydene.2007.04.046 The choice of a suitable fuel processor and fuel, during the transition phase to a hydrogen economy, are key aspects to the successful implementation of direct-hydrogen fuel cell systems. The key requirements for a fuel processor include rapid start- up, good dynamic-response to change in hydrogen demand, high fuel-conversion, small size and weight, simple design (construction and operation), stable performance for repeated start-up and shut-down cycles, maximum thermal integration, low cost and maintenance, high reliability and safety [5]. In small scale applications, natural gas remains the fuel most commonly employed for its wide availability and related infras- tructure. For some niche markets, such as electricity production in remote sites, LPG could be an interesting optional fuel [6,7]. However, to utilize the reformate gas as a reactant for PEFC systems, clean-up steps must be considered to reduce the CO concentration to an acceptable level (10 ppm), since the fuel cell performance is progressively degraded by CO poisoning of the anode catalyst [8,9]. The reformate stream is purified using a two-stage process. The first stage is the water gas shift reaction, that reduces the carbon monoxide, increasing hydrogen con- tent. The CO conversion is limited by equilibrium at the outlet temperature of the reactor. In the second stage the amount of