Quantum Monte Carlo Calculations on a Benchmark Molecule–Metal Surface Reaction: H₂ + Cu(111)

Abstract

Accurate modeling of heterogeneous catalysis requires the availability of highly accurate potential energy surfaces. Within density functional theory, these can—unfortunately—depend heavily on the exchange-correlation functional. High-level ab initio calculations, on the other hand, are challenging due to the system size and the metallic character of the metal slab. Here, we present a quantum Monte Carlo (QMC) study for the benchmark system H2 + Cu(111), focusing on the dissociative chemisorption barrier height. These computationally extremely challenging ab initio calculations agree to within 1.6 ± 1.0 kcal/mol with a chemically accurate semiempirical value. Remaining errors, such as time-step errors and locality errors, are analyzed in detail in order to assess the reliability of the results. The benchmark studies presented here are at the cutting edge of what is computationally feasible at the present time. Illustrating not only the achievable accuracy but also the challenges arising within QMC in such a calculation, our study presents a clear picture of where we stand at the moment and which approaches might allow for even more accurate results in the future.

Dissociative chemisorption of methane on Ni(111) using a chemically accurate fifteen dimensional potential energy surface

“…Recent QMC calculations were able to reproduce the semi-empirical reference value of the H 2 + Cu(111) barrier height to within no better than 1.6 kcal/mol, as a result of limitations imposed by fixed node, locality, and time step errors. 9 The accuracy of correlated wave function approaches with embedding is in practice limited by the size of the embedded cluster that can be modeled. 6 As a result, it is still challenging to predict surface reaction barrier heights and related potential energy surfaces (PESs) with chemical accuracy (<1 kcal/mol, or 4.2 kJ/mol, or 43 meV), which has been routinely realized in triatomic and tetratomic systems in the gas phase.…”

A new scheme for fixed node diffusion quantum Monte Carlo with pseudopotentials: Improving reproducibility and reducing the trial-wave-function bias

“…Many of these calculations are performed to evaluate a binding energy curve, which are needed for example in adsorption energy calculations of molecules on surfaces. [3][4][5][6]19,20 Whereas in this small system it is not overly burdensome to optimize J at every different geometry, in a larger and more complex adsorption system this would be tedious and time-consuming. Notwithstanding the variability of the quality of the optimization, due to its stochastic nature.…”

“…Nowadays there is an increasing amount of benchmark data for solids and surfaces obtained via FN-DMC. [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20] Data provided by DMC is of use in tackling interesting materials science problems and also to help the improvement of density functional theory (DFT) 21 and other cheaper computational approaches.…”

Selective hydrogen production at Pt(111) investigated by Quantum Monte Carlo methods for metal catalysis

“…The present work therefore represents a breakthrough in obtaining reliable information on the bond‐dissociation limitation of metal‐catalyzed reactions. QMC was benchmarked for hydrogen dissociation on copper, for which accurate molecular beam measurements are available . Our methodological advances are outlined here and the main results of an enabling 52 Mh supercomputer allocation are given here (the full data can be obtained from the repository FigShare 10.6084/m9.figshare.10293194), pending a full paper with more methodological detail.…”