(632f) Not All Hydrogens Are Created Equal: Theoretical and Experimental Insight into How Protons and Hydrides Steer Different Hydrogenation Rates for Alkynes and O2 over Au/TiO2

            Preferential
oxidation of CO (CO PrOx) and selective alkyne
hydrogenation are two industrially important processes involving H₂
in their feed streams. Selective hydrogenation of alkynes in polyolefin
feedstocks to alkenes is a critical step in the plastics industry to prevent
deactivation of the downstream polymerization catalyst, and CO PrOx is potentially useful for purifying hydrogen streams
from methane steam reforming, due to its improved energy efficiency relative to
methanation. Both reactions are generally considered
to be hydrogen coverage limited; however, we have observed significantly lower
reaction rates for 1-octyne hydrogenation (relative to H₂ oxidation)
on the same Au/TiO₂ catalyst. Moreover, recent studies have shown that
the selectivity of Au catalysts towards CO oxidation can be enhanced by
poisoning the H₂ activation sites at the metal-support interface
(MSI).^(1,2)We employ density functional theory
(DFT) calculations and particle size study to understand these phenomena.

The DFT
calculations show heterolytic H₂ activation at the MSI, which
results in a support proton and a formal gold hydride as depicted in Figure 1, to
be the energetically preferred pathway for H₂ activation on Au/TiO₂.
The gold hydrides generated by H₂ activation near the interface were
shown to be less stable than the support protons, indicating that spillover to
the support is favorable.⁽²⁾ Regardless
of the heterolytic or homolytic
nature of H₂ activation near the interface, a low hydride coverage
on gold is expected. Oxygen hydrogenation benefits from the stability and
increased concentration of support protons near the MSI, as our calculated
activation barriers in Table 1 show facile proton transfer to O₂. In
contrast, the calculated barriers for proton addition to propyne
are about twice those of the barriers for hydride addition, suggesting that that
the hydride addition is the preferred pathway for alkyne hydrogenation. Large
differences in the availability of active hydrogen species (proton/hydride)
could explain the observed differences in hydrogenation rates. The relatively
strong binding reported for 1-octyne to the support and the Au/TiO₂
interface is another possible factor limiting the role of interface sites as
active sites for hydrogen activation during alkyne hydrogenation.⁽³⁾
To verify the DFT predictions we performed a particle size study, where we
measured the H₂ oxidation and 1-octyne hydrogenation rates on
catalysts with different gold particle sizes. For 1-octyne hydrogenation, the
turnover frequencies (TOFs) were the same between different particle sizes when
they were based on surface sites. Conversely, the TOF for O₂
hydrogenation was constant when normalized with the number of perimeter sites.

            Overall,
computational predictions and kinetic experiments suggest that hydrides on
surface gold sites are active for 1-octyne hydrogenation, while protons at the
Au/TiO₂ interface play a key role in O₂ hydrogenation. We
believe that our detailed insights into the hydrogenation kinetics over
supported gold catalysts will contribute to further improve selective
hydrogenation catalysis in industrially relevant processes.

References

(1) 
J.
Saavedra, T. Whittaker, Z. Chen, C. J. Pursell, R. M. Rioux and B. D. Chandler,
Nat. Chem. 8, 584(2016).

(2) 
T.
Whittaker, K. B. S. Kumar, C. Peterson, M. N. Pollock, L. C. Grabow and B. D.
Chandler, J. Am. Chem. Soc. 140,
16469(2018).

(3)  J. E. Bruno, K.B. S. Kumar, N. S. Dwarica, A. HŸther, Z. Chen, C. S.
Guzman IV, E. R. Hand, W. C. Moore, R. M. Rioux, L. C. Grabow, B. D. Chandler,
ChemCatChem. 11, 1650(2019).