Doblhoff-Dier Katharina, Meyer Jörg, Hoggan Philip E, Kroes Geert-Jan
Leiden Institute of Chemistry, Gorlaeus Laboratories, Leiden University , Post Office Box 9502, 2300 RA Leiden, The Netherlands.
Institute Pascal, UMR 6602 CNRS, University Blaise Pascal , 4 avenue Blaise Pascal, TSA 60026, CS 60026, 63178 Aubiere Cedex, France.
J Chem Theory Comput. 2017 Jul 11;13(7):3208-3219. doi: 10.1021/acs.jctc.7b00344. Epub 2017 Jun 9.
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 H + 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.
对多相催化进行精确建模需要高精度的势能面。遗憾的是,在密度泛函理论中,这些势能面很大程度上取决于交换关联泛函。另一方面,由于体系规模和金属平板的金属特性,高水平的从头算计算具有挑战性。在此,我们针对基准体系H + Cu(111)开展了量子蒙特卡罗(QMC)研究,重点关注解离化学吸附势垒高度。这些计算难度极大的从头算结果与化学精度的半经验值在1.6 ± 1.0千卡/摩尔范围内相符。详细分析了剩余误差,如时间步长误差和局域性误差,以评估结果的可靠性。这里展示的基准研究处于当前计算可行性的前沿。我们的研究不仅说明了可达到的精度,还展示了此类计算中量子蒙特卡罗方法所面临的挑战,清晰地呈现了我们目前的水平以及哪些方法可能在未来带来更精确的结果。