(543e) Steam Reforming of Ethanol (SRE) In a Palladium Based Membrane Reactor Over Nickel Catalyst
AIChE Annual Meeting
2008
2008 Annual Meeting
Separations Division
Membrane Reactors
Wednesday, November 19, 2008 - 4:31pm to 4:50pm
In recent years, there have been renewed interests in bio-ethanol as one of the renewable energy resources for the future energy. It can be easily extracted or blended through biomass fermentation. Direct use of biomass may somewhat reduce our dependence on fossil fuel; its detrimental environmental impact due to CO2 emission is inevitable. Reforming of bio-ethanol is considered as a promising method for hydrogen production from renewable resources. Recent advances in palladium-based thin-film membrane offer an opportunity to develop hydrogen fuel reformer (e.g., steam ethanol reformer) for on-board, one-step generation of hydrogen in a fuel cell vehicle (FCV).
In this work, we used a Pd-thin film membrane in tubular configuration as fuel reformer to study the steam reforming of ethanol (SRE) over nickel based catalyst. The membrane was fabricated in our laboratory using a newly developed surfactant induced electroless plating (SIEP) technique. The Pd-film was deposited on the outside surface of the microporous stainless steel tubular substrate. The effect of steam to ethanol ratio, temperature, and space velocity on ethanol conversion, hydrogen yield, and carbon monoxide suppression was studied for the SRE in the membrane reactor. A two-dimensional, pseudo-homogeneous membrane-reactor model for the SRE reactions was developed to study the membrane reactor performance. Under certain simplifying assumptions, a set of partial differential equations (PDEs) was derived using the continuity equations for the SRE reactions. To account the permeation of hydrogen through the membrane wall in radial direction, radial diffusion was included in the model development. Finite difference method was used to solve the highly coupled PDEs. Experimental ethanol conversions were found to be in good agreement with the model predictions. In this presentation we will discuss some of our experimental and modeling results.