Atomicrex—a general purpose tool for the construction of atomic interaction models

Abstract

Abstract. We introduce atomicrex, an open-source code for constructing interatomic potentials as well as more general types of atomic-scale models. Such effective models are required to simulate extended materials structures comprising many thousands of atoms or more, because electronic structure methods become computationally too expensive at this scale. atomicrex covers a wide range of interatomic potential types and fulfills many needs in atomistic model development. As inputs, it supports experimental property values as well as ab initio energies and forces, to which models can be fitted using various optimization algorithms. The open architecture of atomicrex allows it to be used in custom model development scenarios beyond classical interatomic potentials while thanks to its Python interface it can be readily integrated e.g., with electronic structure calculations or machine learning algorithms.arXiv:1701.08864v1 [cond-mat.mtrl-sci] 30 Jan 2017A general purpose tool for the construction of atomic interaction models 2

Design, Parameterization, and Implementation of Atomic Force Fields for Adsorption in Nanoporous Materials

“…Atomicrex is a versatile tool for the construction of advanced atomistic models. [329] In a general sense, it allows one to develop models that describe a given scalar or vectorial property (e.g., total energies and forces) as a function of an atomic (or atom-like) configuration. GAFit is a software package based on a genetic algorithm that fits an analytic function to a given set of data points.…”

“…The user has the flexibility of changing the weights for fitting and uses relative error to judge the fitness of parameters. Atomicrex is a versatile tool for the construction of advanced atomistic models . In a general sense, it allows one to develop models that describe a given scalar or vectorial property (e.g., total energies and forces) as a function of an atomic (or atom‐like) configuration.…”

On the question of the applicability of the principle of thermodynamic similarity in liquid alkali metals

“…where the first term characterizes the short-range repulsion, the second term characterizes periodic oscillations in the electronic structure. The potential parameters for considered systems were obtained by machine learning: the algorithm is a controlled random search with local mutation [8]. The convergence criterion of the machine learning algorithm is the forces acting on the atoms obtained from the earlier ab-initio simulation in the VASP package (100 system configurations) [9,10].…”

Horizons of modern molecular dynamics simulation in digitalized solid freeform fabrication with advanced materials

“…It is pertinent to note that a good evaluation and assessment of the potential function is required and ideally a potential should be carefully selected for simulations involving large strain deformations such as those seen during contact loading operations like nanoindentation or nanoscratching [10]. A more recently growing trend in the area of development of potential functions is the use of the machine learning approach [72], although several codes exist, such as atomicrex [73] that allow parametrising the potential function based on the wisdom of the researcher. † https://www.cambridge.org/core/journals/mrs-bulletin/issue/three-decades-of-manybody-potentials-inmaterials-research…”