Low Temperature Water Gas Shift: Evaluation of Pt/HfO2 and Correlation between Reaction Mechanism and Periodic Trends in Tetravalent (Ti, Zr, Hf, Ce, Th) Metal Oxides

It is well-known that catalysts containing small quantities of noble metals (e.g., Pt, Au) that are highly dispersed on Group Wand related tetravalent metal oxides (e.g., TiO2, ZrO2, CeO2, ThO2) display, once activated, high conversion for water-gas shift (WGS). The mechanism suggested for this by many authors is bimolecular in nature. In one view, adsorption of CO on the partially reducible oxide produces formates by reaction with active bridging OH groups; followed by water-assisted formate decomposition by dehydrogenation, with this second step being facilitated by metal particles at the metal support interface. Considering this group of oxides, HfO2, as far as we are aware, had not yet been studied as a potential candidate for low temperature water-gas shift (LTWGS). Thus, in this contribution, catalysts composed of Pt supported on HfO2 were prepared and tested. Catalyst testing results suggest that this system is, like Pt/ZrO2 and Pt/CeO2, active for the WGS reaction, although additional studies are required to obtain an ordering of reactivity. Insight into the possible mechanism was gained through the use of diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS). Formate bands were observed to form when defect-associated bridging OH groups, present on hafnia after partial reduction, were reacted with CO. Pt facilitated partial reduction of hafnia and thus the formation of the bridging OH groups on its surface. The formate bands readily disappeared in the presence of steam at 130 degrees C, converting to the second intermediate, carbonate species, which liberated CO2 and presumably H-2 products. Moreover, in switching from CO adsorption alone to steady state LTWGS, the formate bands became reaction rate limited-and more so at higher temperature. The implication is that formate is turning over more rapidly at the latter condition, with its decomposition (probably via C-H bond breaking) being the most likely rate limiting step. These experiments suggest that the reaction probably follows a bimolecular associative pathway, with adsorbed formate being an intermediate, in the same manner as observed with the other Group IV and related tetravalent oxide based LTWGS catalysts. Moreover, an interesting relationship may exist between the C-H bond strength of adsorbed formate (i.e., observed during CO adsorption) and the metal ion radius in the oxide.