Abstract
This investigation presents a generally applicable framework for parameterizing interatomic potentials to accurately capture large deformation pathways. It incorporates a multi-objective genetic algorithm, training and screening property sets, and correlation and principal component analyses. The framework enables iterative definition of properties in the training and screening sets, guided by correlation relationships between properties, aiming to achieve optimal parametrizations for properties of interest. Specifically, the performance of increasingly complex potentials, Buckingham, Stillinger-Weber, Tersoff, and modified reactive empirical bond-order potentials are compared. Using MoSe2 as a case study, we demonstrate good reproducibility of training/screening properties and superior transferability. For MoSe2, the best performance is achieved using the Tersoff potential, which is ascribed to its apparent higher flexibility embedded in its functional form. These results should facilitate the selection and parametrization of interatomic potentials for exploring mechanical and phononic properties of a large library of two-dimensional and bulk materials.
Highlights
Molecular dynamics (MD) simulation based on force fields is a powerful tool for studying the temporal behaviors of materials at submicron scales
Among existing 2D materials, TMDC are one group of materials described by MX2, where M is a transition metal (Mo, W, etc.) and X is from the oxygen family (S, Se, etc.)
TMDCs with a focus on failure-related properties, which are critical to the stability and reliability of systems that require frequent mechanical deformation, e.g., flexible electronics
Summary
Molecular dynamics (MD) simulation based on force fields is a powerful tool for studying the temporal behaviors of materials at submicron scales. We propose a robust parametrization method built upon density parametrizations for TMDCs include SW potentials for the mechanical and thermal properties of MoS2(10,16,18,20), MoSe2(16,31), and WSe2(31); a Tersoff potential for the thermal properties of WSe221; a ReaxFF potential for the mechanical and transitional behaviors of MoS2(17,32); and a REBO-TMDC potential for the interfacial and mechanical properties of MoS2(4,5) Those interatomic potentials can be segmented into cluster potentials functional theory (DFT) data sets (considered as ground truth) and (SW), cluster functionals (Tersoff), and reactive cluster functionals the evolutionary multi-objective optimization algorithm, NSGAIII23. We further explore the parametrization flexibility of the selected interatomic potentials by conducting correlation and principal component analyses on their prediction errors, which reveals a positive correlation between the complexities of interatomic potentials, their flexibility, and their performances on MoSe2 Together, these results suggest a robust potential parametrization approach and a quantitative potential selection criterion, which may be generalized for a wide range of materials and materials properties beyond those explored in this study.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
