Data-driven material models for atomistic simulation

Abstract

The central approximation made in classical molecular dynamics simulation of materials is the interatomic potential used to calculate the forces on the atoms. Great effort and ingenuity is required to construct viable functional forms and find accurate parameterizations for potentials using traditional approaches. Machinelearning has emerged as an effective alternative approach to develop accurate and robust interatomic potentials. Starting with a very general model form, the potential is learned directly from a database of electronic structure calculations and therefore can be viewed as a multiscale link between quantum and classical atomistic simulations. Risk of inaccurate extrapolation exists outside the narrow range of time-and length-scales where the two methods can be directly compared. In this work, we use the Spectral Neighbor Analysis Potential (SNAP) and show how a fit can be produced with minimal interpolation errors which is also robust in extrapolating beyond training. To demonstrate the method, we have developed a new tungsten-beryllium potential suitable for the full range of binary compositions. Subsequently, large-scale molecular dynamics simulations were performed of high energy Be atom implantation onto the (001) surface of solid tungsten. The new machine learned W-Be potential generates a population of implantation structures consistent with quantum calculations of defect formation energies. A very shallow (< 2nm) average Be implantation depth is predicted which may explain ITER diverter degradation in the presence of beryllium.

Gaussian approximation potentials for body-centered-cubic transition metals

“…Well-established frameworks exist for interatomic potentials using artificial neural networks [3][4][5][6], Gaussian process regression and other kernel methods [7][8][9], and linear regression [10,11]. Although the field is still relatively new, it has already reached a level of maturity that high-quality machine-learning potentials are now routinely trained for a variety of materials and molecules [12][13][14][15][16].…”

Explicit Multi-element Extension of the Spectral Neighbor Analysis Potential for Chemically Complex Systems

“…In the original formulation of SNAP, the bispectrum descriptors only distinguish between neighbor atoms of different chemical elements based on their weighted contribution to the total atomic density. 24,25 This is similar in spirit to the construction of the density function within the embedded atom method for metal alloys. 26,27 For systems that show strong differences in bonding characteristics depending on the chemical identity of the atoms, this weighted-density (WD) approach is likely insufficient.…”

Machine‐learning‐based interatomic potentials for advanced manufacturing

“…114 A tungsten-beryllium potential is developed by SNAP and used to simulate highenergy Be atom implantation onto the (001) surface of solid tungsten. 115 Extensive atomic simulations were performed on the deposition and growth of amorphous carbon (a-C) thin films using a GAP model to describe the interaction between atoms. 116 Neural-network potentials (NNPs) are applied to simulation early-stage nucleation and growth of the Al-Cu system.…”