Ephemeral data derived potentials for random structure search

Abstract

Structure prediction has become a key task of the modern atomistic sciences, and depends on the rapid and reliable computation of the energy landscape. First principles density functional based calculations are highly reliable, faithfully describing the entire energy landscape. They are, however, computationally intensive and slow compared to interatomic potentials. Great progress has been made in the development of machine learning, or data derived, potentials, which promise to describe the entire energy landscape at first principles quality. However, compared to first principles approaches, their preparation can be time consuming and delay searching. Ab initio random structure searching (AIRSS) is a straightforward and powerful approach to structure prediction, based on the stochastic generation of sensible initial structures, and their repeated local optimisation. Here, a scheme, compatible with AIRSS, for the rapid construction of disposable, or ephemeral, data derived potentials (EDDPs) is described. These potentials are constructed using a homogeneous, separable manybody environment vector, and iterative neural network fits, sparsely combined through non-negative least squares. The approach is first tested on methane, boron nitride, elemental boron and urea. In the case of boron, an EDDP generated using data from small unit cells is used to rediscover the complex $\gamma$-boron structure without recourse to symmetry or fragments. Finally, an EDDP generated for silane (SiH${}_4$) at 500 GPa enables the discovery of an extremely complex, dense, structure which significantly modifies silane's high pressure phase diagram. This has implications for the theoretical exploration for high temperature superconductivity in the dense hydrides, which have so far largely depended on searches in smaller unit cells.

17 pages, 12 figures

References in corpus (7)

• Quantum ESPRESSO: a modular and open-source software project for quantum simulations of materials
• Machine Learning Unifies the Modelling of Materials and Molecules
• High-pressure phases of silane
• Superconducting hydrides under pressure
• Symmetry-broken crystal structure of elemental boron at low temperature
• Stratified construction of neural network based interatomic models for multicomponent materials
• Revealing and exploiting hierarchical material structure through complex atomic networks

Cited by in corpus (2)

• How to validate machine-learned interatomic potentials
• Machine learning based modeling of disordered elemental semiconductors: understanding the atomic structure of a-Si and a-C