(704a) Mechanistic Insights into the Direct Propylene Epoxidation over Au/TiO2/SiO2

Propylene oxide (C₃H₆O, PO) is a key intermediate for the industrial production of propene glycol ethers, propene glycol, and polyether polyols,¹ which are mainly applied to manufacture commercial products such as adhesives, solvents, and foams.^2-3 The currently available processes are not optimal, since the chlorohydrin process to produce PO suffers from environmental concerns, and the hydroperoxide process is challenged by poor economics because of the production of waste byproducts. The direct vapor-phase propylene epoxidation with molecular H₂ and O₂ catalyzed by nano-sized Au deposited on TiO₂ has been proved to be clean and highly selective.⁴

A lot of experimental and theoretical efforts have been dedicated to improving Au-based catalysts and understanding the fundamental reaction features of direct propylene epoxidation. It is generally proposed that the active oxygen species, hydrogen peroxide (H₂O₂), is readily synthesized on the anchored Au nanoparticles, and it subsequently migrates to the adjacent Ti sites to convert propylene into PO.^4-6 However, the mechanistic comprehension of direct propylene epoxidation is still under intense debate. There are several controversial issues: (i) the main factors that affect the catalytic activity of nanoscale Au; (ii) the origin of the activated oxidant (Ti-OOH), and the sites that are responsible for the direct PO reaction; and (iii) the mechanistic aspects of side reactions, as well as their influence on PO selectivity and catalyst deactivation.

In this work, kinetic Monte Carlo (KMC) simulations are performed to simulate the chemical kinetics of direct propylene epoxidation and explore the underlying mechanistic aspects. This spatially-resolved modeling approach can narrow the gap between DFT studies and experimental efforts, due to its ability to connect variations in the atomistic and electronic structure of a catalytic surface with its activity and selectivity. We show that KMC simulations in combination with experimental benchmarking provide valuable insight into the impact of reaction conditions (temperature, reactant concentration) on PO selectivity, hydrogen efficiency, and catalyst activity. The present study systematically probes the reaction mechanism, which involves acrolein formation on Au nanoparticles as well as the formation of PO and several byproducts (e.g., acetone, propanal, ethanal, etc.) at the Au/Ti dual interface sites, and it highlights the synergistic effect of interface sites in heterogeneous catalysis.

The KMC model takes into account the atomistic-level activity of low-coordinated Au sites and lateral interactions, and it also accounts for the correlations, fluctuations, and the spatial distribution of the reaction species on the catalyst surface. We find that the KMC model is able to quantitatively reproduce the majority of the experimental data (product distributions as a function of temperature, pressure, feed composition, etc.). Moreover, we are able to extract the influence of the local structure on the catalyst performance and the influence of neighbor-neighbor interactions on the reaction kinetics at high coverages. One of the key experimental challenges is improving the hydrogen efficiency of this system, so we use our model to identify optimal operating conditions (a low reaction temperature, a low hydrogen feed concentration, and a relatively high oxygen feed concentration). This work sets the stage for exploring more complex bimetallic catalysts for simultaneously achieving high selectivity and activity for PO synthesis.

References

(1) Huang, J.; Haruta, M. Res. Chem. Intermed. 2012, 38, 1-24.

(2) Nijhuis, T. A.; Makkee, M.; Moulijn, J. A.; Weckhuysen, B. M. Ind. Eng. Chem. Res. 2006, 45, 3447-3459.

(3) Yap, N.; Andres, R. P.; Delgass, W. N. J. Catal. 2004, 226, 156-170.

(4) Hayashi, T.; Tanaka, K.; Haruta, M. J. Catal. 1998, 178, 566-575.

(5) Stangland, E. E.; Stavens, K. B.; Andres, R. P.; Delgass, W. N. J. Catal. 2000, 191, 332-347.

(6) Wells, D. H.; Delgass, W. N.; Thomson, K. T. J. Am. Chem. Soc. 2004, 126, 2956-2962.