Hydrodeoxygenation of γ-valerolactone on bimetallic NiMo phosphide catalysts

A series of supported Ni-Mo-P alloy catalysts was studied for the catalytic hydrodeoxygenation (HDO) of the cyclic five-membered ester γ-valerolactone (GVL-C5H8O2) as a model compound for pyrolysis oil. Alloy formation in Ni-Mo-P was indicated by X-ray diffraction analysis and X-ray absorption near-edge spectroscopy, which showed systematic shifts with composition. The number of active sites of each metal species was estimated by factor analysis combining CO-uptake measurements and infrared (IR) spectra of adsorbed CO. It was found that the catalytic activity followed the order: Ni2P/MCM-41 > NiMo(3:1)P/MCM-41 > NiMo(1:1)P/MCM-41 ≅ (Ni2P + MoP)/MCM-41 > NiMo(1:3)P/MCM-41 > MoP/MCM-41, whereas the normalized turnover frequency based on Ni sites were similar for all the catalysts, while retaining the same order. It is concluded that adjacent surface Ni atoms are the main active sites involved in the rate-determining step (rds). The Mo-containing catalysts produced more 1-pentanol and C5 hydrocarbons than Ni2P/MCM-41, indicating that while the exposed Ni sites governed catalytic activity, Mo sites controlled the selectivity to C5 hydrocarbons. Thus, steps following the rds were influenced by Mo sites, leading to preferences for different reaction pathway during the HDO of γ-valerolactone. The study reveals that the catalytic behavior of NiMoP catalysts can be tuned by the relative proportion of Ni and Mo sites.