(143c) The Proximity Effects and the Role of Molecular Shuttles in Water-Gas Shift and Methanol Synthesis

Methanol synthesis from synthetic gas mixtures (CO/CO₂/H₂) and water-gas shift (WGS) reactions that interconvert between CO/H₂O and CO₂/H₂ have been exploited in practice. Such routes have also received significant attention more recently for their potential use in converting unwanted CO₂. The oxides with hydration-dehydration reactivities (e.g., TiO₂, ZnO, CeO₂)^1,2 are well-known to promote the rates of WGS³ and methanol synthesis⁴ when used together with a metal function (e.g., Cu, Pt, Au), which catalyzes hydrogenation-dehydrogenation reactions.^5,6 These observations, taken together with the presence of formates during WGS and methanol synthesis,⁷ invoke the plausible role of HCOOH as the molecular shuttle that move the formate, formed on the oxide function from CO and H₂O, to the proximate metal function that catalyzes HCOOH hydrogenolysis (to CH₃OH) or dehydrogenation (to CO₂ and H₂). The low prevalent concentration of HCOOH requires the short inter-function distances but not atomic contact or periphery effects. This work combines kinetic, thermodynamic and transport assessments to elucidate the role of a molecular carrier, HCOOH, in WGS and methanol synthesis, which, in turn, provides insights into bifunctional cascade reactions that are kinetically coupled via reactive intermediates.

WGS turnover rates, defined as the rates per exposed Cu sites, remained the same (523 K; 3.3 kPa CO; 10 kPa H₂O) when Cu/ZnO/Al₂O₃ catalysts were physically mixed with the same ZnO/Al₂O₃ oxide to form pellets. The rates, in contrast, monotonically decreased with the extent of dilution with SiO₂. Such trends implicate that the separation of the two functions with SiO₂ causes the concentration gradient of the reactive intermediate (e.g., HCOOH) within the pellet, leading to lower WGS rates when they were limited by the metal function that dehydrogenates HCOOH to form CO₂ and H₂. More detailed assessments of WGS reactions via bifunctional catalysis require in-depth inquiries into each function that forms and uses reactive HCOOH intermediates and the transport of HCOOH between the two functions.

HCOOH formation elementary steps from CO and H₂O on oxides and the magnitude of kinetic and thermodynamic parameters are assessed from reverse HCOOH dehydration on model oxides, anatase and rutile TiO₂ (TiO₂(a), TiO₂(r)), at temperatures and reactant (and HCOOH) pressures relevant to WGS and methanol synthesis; the forward reaction is difficult to be assessed due to the detection limits of HCOOH. HCOOH dehydration rates on TiO₂(a) and TiO₂(r) increased monotonically with HCOOH pressure (0.1-0.5 kPa; 533 K), indicating the presence of HCOOH-derived species as minor surface species. The rates were independent of CO and CO₂ pressures, while inhibited by H₂O due to its competitive adsorption on the active Ti_5c-O_2c centers. The weak kinetic isotopic effects of HCOOD and DCOOD implicate that the rates are limited by the HCOOH adsorption elementary step, which was confirmed from density functional theory calculations. The measured free energy barriers were similar for TiO₂(a) and TiO₂(r) catalysts (ΔG_533K^Ç‚ = 132 vs. 130 kJ mol^-1) despite of their very different DFT-derived HCOOH binding energies (to form bidentate formate, ΔG_533K = -101 vs -148 kJ mol^-1). These trends result from the kinetically-relevant transition states that occur very early in the adsorption reaction coordinate and thus do not sense the stability of the adsorption product, bidentate formates. Such mechanistic interpretations, in turn, give the rate equation and parameters for reverse CO hydration reaction to form HCOOH, which are used to derive the prevalent HCOOH concentrations generated on TiO₂ oxide during WGS and methanol synthesis.

HCOOH formed on oxides can transport to the proximate metal function (Cu) to form WGS products (CO₂ and H₂). HCOOH dehydrogenation rates on Cu increased monotonically with HCOOH pressures (<0.5 kPa; 503 K), suggestive of the unoccupied Cu sites as the most abundant surface species. The rates decreased with increasing CO pressure (0-10 kPa; 503 K) due to its competitive adsorption on active Cu sites but were unaffected by H₂O (0-3 kPa; 503 K). At low acid coverages, the reaction is mediated by carboxylate intermediates, instead of bidentate formates, consistent with the insensitiveness of rates to H₂ pressure (0-30 kPa; 0.1 kPa HCOOH; 503 K). The turnover rates are limited by the HCOOH adsorption step that forms carboxylate, inferred by the strong kinetic isotopic effects with DCOOH but not with HCOOD.

These mechanistic interpretations of HCOOH dehydration and dehydrogenation reactions on oxide and metal functions are combined with the transport model to explain the WGS kinetics on catalytic systems with different inter-function distances and at the kinetic regime, limited by either oxide or metal function. The results of this work uncover the origin of proximity effects in WGS and methanol synthesis, the requirement imposed by a short-living reactive intermediate. In more general prospective, this work provides the analytic framework that can be used to understand many other cascade reactions via reactive molecular carriers that are ubiquitous in heterogeneous catalysis.

The authors acknowledge the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences (DE-AC05-76RL0-1830) for financial support. This work was accomplished by using computational resources at the Environmental Molecular Science Laboratory (EMSL), the National Energy Research Scientific Computing Center (NERSC), and the Extreme Science and Engineering Discovery Environment (XSEDE).

References

(1) Trillo, J. M.; Munuera, G.; Criado, J. M. Catal. Rev. 1972, 7, 51.

(2) Munuera, G. J. Catal. 1970, 18, 19.

(3) Shekhar, M.; Wang, J.; Lee, W.-S.; Williams, W. D.; Kim, S. M.; Stach, E. a; Miller, J. T.; Delgass, W. N.; Ribeiro, F. H. J. Am. Chem. Soc. 2012, 134, 4700.

(4) Spencer, M. S. Top. Catal. 1999, 8, 259.

(5) Ojeda, M.; Iglesia, E. Angew. Chemie 2009, 121, 4894.

(6) Scaranto, J.; Mavrikakis, M. Surf. Sci. 2016, 648, 201.

(7) Yang, Y.; Mims, C. A.; Mei, D. H.; Peden, C. H. F.; Campbell, C. T. J. Catal. 2013, 298, 10.