Descriptions of surface chemical reactions using a neural network representation of the potential-energy surface

Abstract

A neural network ͑NN͒ approach is proposed for the representation of six-dimensional ab initio potentialenergy surfaces ͑PES͒ for the dissociation of a diatomic molecule at surfaces. We report tests of NN representations that are fitted to six-dimensional analytical PESs for H 2 dissociation on the clean and the sulfur covered Pd͑100͒ surfaces. For the present study we use high-dimensional analytical PESs as the basis for the NN training, as this enables us to investigate the influence of phase space sampling on adsorption rates in great detail. We note, however, that these analytical PESs were obtained from detailed density functional theory calculations. When information about the PES is collected only from a few high-symmetric adsorption sites, we find that the obtained adsorption probabilities are not reliable. Thus, intermediate configurations need to be considered as well. However, it is not necessary to map out complete elbow plots above nonsymmetric sites. Our study suggests that only a few additional energies need to be considered in the region of activated systems where the molecular bond breaks. With this understanding, the required number of NN training energies for obtaining a high-quality PES that provides a reliable description of the dissociation and adsorption dynamics is orders of magnitude smaller than the number of total-energy calculations needed in traditional ab initio on the fly molecular dynamics. Our analysis also demonstrates the importance of a reliable, high-dimensional PES to describe reaction rates for dissociative adsorption of molecules at surfaces.

Generation and orientation of organoxenon molecule H–Xe–CCH in the gas phase

“…Such an interpolation of high-dimensional PESs is a tedious task, but in recent years, several techniques, like analytical fits 29,31,32,33,34 , tight binding representations 35,36,37,38 , genetic programming 39 and the modified Shephard method 40,41 , possibly combined with a corrugation-reduction scheme 42,43 , have been developed. In the present work we employ a very general neural network fitting scheme, which has already proven to be a powerful tool for the accurate representation of multi-dimensional PESs in similar applications 44,45,46,47 . Since the evaluation of the energy and forces from the neural network representation is about 5 to 6 orders of magnitude faster than direct ab initio calculations, a large number of MD trajectories can be calculated in the last step of the "divide and conquer" approach to obtain the sticking probabilities at various molecular kinetic energies.…”

“…70 In particular multilayer feedforward neural networks have already successfully been employed to provide accurate fits of potential-energy surfaces 44,45,46,47 . In the present work this NN type is applied to fit a smooth and continuous function to the DFT data by optimizing a set of parameters.…”

Stratified construction of neural network based interatomic models for multicomponent materials

“…Previously proposed strategies include running molecular dynamics (MD), creating regular meshes of crystal structure parameters, selecting specific desired geometries, etc. [62][63][64]. The MD-based approach has the advantage of sampling commonly accessible physical states.…”

Application of Artificial Neural Networks for Catalysis: A Review

“…To use an alternative potential energy, Sumpter and Noid first systemically showed that neural networks are able to map the vibrational motion determined from spectra onto a fully coupled potential energy surface (PES) with a high accuracy [87]. Subsequently, people started to use ANN and other machine learning techniques for mapping the PES of some simple chemical reactions (e.g., H2 dissociation [88]) and molecular interactions (e.g., water dimer [89]). In terms of the global optimization of catalytic systems, Ouyang et al used an ANN with 56 independent variables in the input layer to fit the PES of Au58 clusters [90].…”