(706d) Towards Common Design Principles Between Hydrotreating Catalysts: A DFT Investigation of Furan Hydrodeoxygenation on Pure and Cobalt-Promoted MoO3(010) and RuO2(110)

The high oxygen content (35-40 %) of bio-oil produced from fast pyrolysis of biomass reduces the heating value and stability, and limits its subsequent use as a transportation fuel. MoO₃ is a promising candidate with high reactivity and selectivity for bio-oil upgrade via catalytic hydrodeoxygenation (HDO) [1]. Its sulfur counterpart, MoS₂, is used in the well-studied hydrodesulfurization process (HDS). Here, we attempt an atomic-level comparison between the two related chemistries based on a prominent density functional theory (DFT) study of thiophene HDS on MoS₂ [2].

Similar to HDS, HDO requires the initial creation of an oxygen-vacancy site where the feed molecule can adsorb. We assessed the thermodynamic stability of three distinct oxygen vacancy sites on the MoO₃(010) surface using a detailed ab-initio thermodynamic phase diagram under typical reaction conditions. Our analysis predicts an asymmetric oxygen vacancy with two subsurface hydrogen atoms as the most stable surface termination. We also observed that phases co-existing around the reaction conditions can transition from one phase to another through a single redox reaction involving the formation of a water and a surface oxygen defect in the presence of hydrogen gas.

Next, we used this most stable MoO₃(010) facet to investigate the elementary reaction steps for furan HDO. The potential energy diagram based on the thermodynamic stability of reaction intermediates suggests that the reaction pathways for thiophene HDS and furan HDO are similar. Activation barriers for elementary reactions further indicate that furan HDO on MoO₃ is facile. However, the oxygen vacancy formation on MoO₃(010) is slow. This scenario is reminiscent of HDS on MoS₂, where cobalt promotion facilitates sulfur vacancy formation, leading to increased activity of the industrially used CoMoS catalyst. Using this existing knowledge of HDS on metal-sulfides as rational design guidelines we investigated cobalt promoted MoO₃ and also observed an increase in active vacancy sites. These studies provide valuable insights regarding the extent to which existing HDS knowledge can be translated to HDO catalysis. We also investigated furan HDO on RuO₂, which is the oxide analogue of RuS₂, the most active HDS catalyst. Furan HDO on RuO₂ is facile but catalyst stability and cost might render it uneconomical.

Unraveling the similarities and differences between the two related hydrotreating processes (HDO and HDS) will ultimately let us rationally design a novel HDO catalyst for the commercial use of biomass for the production of chemicals and fuels.

[1] Prasomsri, T.; Nimmanwudipong, T.; Román-Leshkov, Y. Energy Environ. Sci. 2013, 6, 1732.
[2] Moses, P. G.; Hinnemann, B.; Topsøe, H.; Nørskov, J. K. J. Catal. 2007, 248, 188–203.