Hydrogen Generation from Methanol Oxidation on Supported Pt and Cu Catalysts
Yu-Chuan Lin1, Keith L. Hohn1, and Susan M. Stagg-Williams2. (1) Department of Chemical Engineering, Kansas State University, 1005 Durland Hall, Manhattan, KS 66506-5102, (2) University of Kansas, Chemical and Petroleum Engineering, 4132 Learned Hall, 1530 West 15th Street, Lawrence, KS 66045
Hydrogen generation from methanol is an attractive means for generating hydrogen for proton exchange membrane (PEM) fuel cells because methanol is an abundant liquid fuel that can easily be transported and stored. In addition, methanol has a high atomic ratio of H/C compared to other fuels. Catalytic partial oxidation (CPO) is one way to convert methanol to hydrogen. It offers the advantages of an exothermic reaction and high reaction rates. Typically, copper containing catalysts have been used for CPO of methanol. These catalysts produce primarily H2 and CO2. Recently, CPO of alcohols (methanol and ethanol) has been studied on platinum catalysts. Platinum catalysts have the advantage of being stable, unlike many copper-containing catalysts, but produce much more CO, which is a disadvantage since CO is a known poison of the PEM fuel cell anode. The mechanism of methanol partial oxidation over copper and platinum is not well understood. Multiple reactions may play a role, including catalytic partial oxidation (CPO), methanol decomposition (MD), methanol steam reforming (MSR), and the water-gas shift reaction (WGS). An understanding of which reaction pathways are important is critical for designing better methanol CPO catalysts. In this work, the mechanism of methanol CPO was studied over copper and platinum catalysts with the objectives of determining which reaction pathways are important and why copper and platinum give much different product distributions. Methanol CPO was run at 250oC on 0.5% Pt/ZrO2 and 40% Cu/ZnO with different space velocities. As found by previous researchers, CO and H2 were the main products on Pt/ZrO2, while CO2 and H2 were formed on Cu/ZnO. Methanol conversions decreased for both catalysts with increasing GHSV. On both catalysts, increasing GHSV increased the selectivities of CO2 and H2O. To understand these results, kinetics experiments were performed to determine the reaction kinetics of individual reaction steps (oxidation, decomposition, steam reforming and water-gas shift) on Cu/ZnO and Pt/ZrO2. Using these reaction kinetics in a packed bed reactor model, methanol CPO was simulated at different space velocities and the results were compared with the experimental results. For both catalysts, oxidation reactions were by far faster than other reactions, therefore oxidation dominates the process until all of the oxygen is consumed. Once oxygen was consumed, the selectivities to CO and H2 continued to rise dramatically on Pt/ZrO2 because of methanol decomposition. On Cu/ZnO, however, methanol decomposition was much slower, so product selectivities did not change significantly after oxygen had been depleted. For both catalysts, steam reforming and water-gas shift were not fast enough to play a large role in the final product composition. These results suggest that methanol oxidation and methanol decomposition are the primary reaction pathways on both Cu/ZnO and Pt/ZrO2. Pt/ZrO2 produces a higher selectivity to CO than Cu/ZnO because the partial oxidation reaction produces more CO than Cu/ZnO and because methanol decomposition to CO and H2 is a much faster reaction on Pt/ZrO2. Water-gas shift was not found to play a significant role in methanol CPO under the experimental conditions of this study. Its measured rate was small, and the change in product selectivities (increasing CO and H2 with longer residence times) did not indicate that water-gas shift was occurring to a great extent. Water-gas shift is highly desirable in methanol CPO to produce additional H2 and to lower the concentration of CO. For this reason, the support was modified with the intent of producing a catalyst with better WGS activity. The support was doped with ceria to prepare Pt/Ce/ZrO2 catalysts, since Pt/Ce/ZrO2 has previously been reported to be capable of promoting WGS. Catalysts were prepared with 0%, 17.5%, 30%, 58% and 100% ceria. For all Ce-promoted Pt/ZrO2 catalysts, decreasing the space velocity (increasing the residence time) increased the selectivities of CO2 and H2. This suggests that WGS was occurring over these catalysts, since the contribution of WGS should increase CO2 and H2 selectivities if there is sufficient time for reaction to occur. However, increasing the amount of ceria in the samples did not improve CO and H2 yields. With increasing Ce dosage, the selectivity of CO and H2 decreased while CO2 and H2O increased except at 100% Ce loading, which had results similar to the 17.5 and 30% Pt/Ce/ZrO2. 58% Pt/Ce/ZrO2 produced the lowest CO and H2 exhaust. These results suggest that ceria enhanced complete combustion of methanol to CO2 and H2O, or decreased methanol decomposition rates. It appears that the effects of Ce promotion are not simple: Ce can affect the rates of multiple reaction steps.