2025 AIChE Annual Meeting

(254e) Brønsted Acid Site Formation in Hydrodeoxygenation on Metal Oxide Surfaces: The Enigmatic Case of MoO3

Authors

Zeinab Hajali Fard - Presenter, Iowa State University
Kimberly Ung, Iowa State University
Brent H. Shanks, Iowa State University
Luke Roling, Iowa State University
Leveraging renewable biomass feedstocks offers a path to sustainable chemical production. Lignocellulosic biomass is abundant, though difficult to convert because after its pyrolysis the produced oil has high oxygen content, and so a hydrodeoxygenation (HDO) process is essential to improve its refinery compatibility. Mo-based catalysts offer better arene selectivity than transition metal catalysts under ambient hydrogen pressures.1 MoO₃ facilitates HDO via an oxygen-vacancy-driven mechanism, where the vacancy selectively cleaves C–O bonds. Oxygen vacancies act as Lewis acid sites, while hydroxyl groups from hydrogen chemisorption serve as Brønsted acid sites, both influencing catalytic activity.2-3

In this presentation, we share our recent computational results comparing acetone and anisole HDO mechanisms on α-MoO₃(010). Our DFT calculations highlight the role of oxygen vacancies in a reverse Mars-van Krevelen mechanism but initially overlooked the importance of surface hydroxyls on the mechanism. We therefore expanded our calculations with an in-depth study of H adsorption to better understand the true active site, showing that asymmetric oxygen favors hydroxyl formation both thermodynamically and kinetically. It was observed that hydrogen spillover to neighboring sites is facile due to low activation barriers (between 0.6 and 1.1 eV for hydrogen travel between terminal and bridge oxygens). These findings aid in better understanding the surface post-reduction and refining adsorption models for carbonyl-containing compounds, showing the importance of using the correct surface and improving the fidelity of our computational models with our experimental HDO kinetics experiments, enhancing the explanatory and predictive power of computations.

References:

  1. Kohler, A.J. et al. ACS Catal. 2023, 13, 22, 14813–14827.
  2. Ikeda, Y. et al. Phys. Chem. C 2022, 126, 17, 7728–7738.
  3. Prasomsri, T. et Energy Environ. Sci. 2013, 6 (6), 1732–1738.