A carbon monoxide PROX reactor for PEM fuel cell automotive application

Abstract

Loughborough University have designed, constructed and evaluated a compact CO preferential oxidation (PROX) reactor for PEM fuel cell applications. The reactor design is based upon the catalyst coating of high surface area heat transfer technology. Catalyst screening studies have revealed a mixed transition metal oxide promoted platinum–ruthenium formulation to be suitable for the particular reactor application i.e. acceptable CO oxidation activity and selectivity within a temperature range of 130–200°C. The CO PROX reactor design has been sized for $\text{20}\text{kW}{}_{\mathrm{<mtext>e</mtext>}}$ PEM fuel cell applications, and is based upon 2×2 l catalyst washcoated compact fin heat exchangers. The PROX reactor has being successfully integrated and commissioned with a methanol steam reformer with reductions in fuel CO concentrations of 2.7% to <20 ppm being subsequently demonstrated.

Introduction

During the last decade, there has been a significant increase in the level of worldwide concern regarding both the release of greenhouse gases into the atmosphere, and the poor air quality in many of the world's metropolitan areas [1], [2]. The transport sector, and in particular the internal combustion engine (ICE), is known to be a major source of both CO₂ and other noxious gaseous emissions. For example, the road transport sector currently accounts for nearly a quarter of all the CO₂ emissions in the UK and is also responsible for approximately 49% NO_x, 75% CO, 41% VOC and 27% of PM₁₀ emissions released into the atmosphere [3]. Against such a backdrop of environmental and health concerns, the polymer electrolyte membrane (PEM) fuel cell, is now being recognised as an increasingly realistic clean alternative for motive propulsion. The majority of the world's leading automotive manufacturers now have PEM fuel cell vehicle R&D programs underway (e.g. DaimlerChrysler's NECAR, GM's “HydroGen 1”, Ford's FC5 & P2000, Toyota's RAV4 & Honda's FCX fuel cell vehicles) and a number of companies have publicly stated their aim of launching pre-production fuel cell vehicles by 2003–2004 [4], [5], [6], [7], [8], [9], [10]. However, the choice of fuel for the fuel cell stacks continues to be a key issue regarding the commercialisation of fuel cell vehicles. In addition to infra-structure issues, the fuel choice has serious implications upon both the design of the vehicle and the re-fuelling mechanism required [11], [12], [13], [14], [15], [16]. Each of the fuels primarily under consideration i.e. hydrogen, methanol and gasoline, has its own inherent advantages and disadvantages and as of yet no particular fuel is universally favoured. Governmental and automotive representatives worldwide, have acknowledged the complexity of the fuel choice implications and many technical issues are yet to be resolved [4], [5].

The reforming of hydrocarbons to liberate the hydrogen is an attractive proposition since the hydrogen can be stored chemically at significantly increased energy densities. Reformer technology for PEM fuel cell vehicles is currently based upon steam reforming, partial oxidation or a combination of both (autothermal reforming) techniques [17], [18], [19], [20], [21], [22], [23]. Whilst the exothermicity of the partial oxidation reaction facilitates fast start-up, increased hydrogen yields and improved system efficiency are possible via steam reforming. Due to both kinetic and thermodynamic constraints, significant concentrations of carbon monoxide are produced from all methods of hydrocarbon reformation. At typical PEM stack operating temperatures of ⩽85°C, the carbon monoxide rapidly and strongly adsorbs onto the platinum electro-catalyst surface, which prevents hydrogen adsorption and electro-oxidation and results in a large and rapid decrease in cell performance [24]. Although advances in improving the tolerance of the PEM fuel cell to carbon monoxide are being made, a stage of carbon monoxide removal between the reformer and the fuel cells is still currently required [25], [26], [27], [28], [29], [30]. Such methods of CO removal include palladium diffusion membranes and catalytic oxidation reactors [15], [19], [31], [32], [33]. Whilst the use of diffusion membranes facilitates the supply of ultra-pure hydrogen to the fuel cells, the technology is inherently expensive and operates under high-pressure differentials, e.g. 10–20 bar. Alternatively, catalytic oxidation may be regarded as a lower cost and more practical alternative. Although conventional packed-bed-type catalytic reactors are used extensively in the chemical industry, the technology may be too bulky for automotive applications. In addition to the large reactor volumes often required, both unacceptable warm-up times and poor thermal management (resulting in both reduced catalyst activity and selectivity for carbon monoxide oxidation in the presence of a large hydrogen excess) are often apparent during reactor operation. Therefore, for increased compactness, minimised pressure drops and improved reaction thermal management, research at LU has focused upon the determination and integration of a suitable CO oxidation catalyst with lightweight, compact and high surface area heat transfer technology.