TY - JOUR
T1 - Multi-objective parametrization of interatomic potentials for large deformation pathways and fracture of two-dimensional materials
AU - Zhang, Xu
AU - Nguyen, Hoang
AU - Paci, Jeffrey T.
AU - Sankaranarayanan, Subramanian K.R.S.
AU - Mendoza-Cortes, Jose L.
AU - Espinosa, Horacio D.
N1 - Funding Information:
The authors acknowledge the support of the National Science Foundation, through award CMMI 1953806, and computational resources provided by the Center of Nanoscale Materials at Argonne National Laboratory, as well as the Quest High Performance Computing Cluster at Northwestern University. Use of the Center for Nanoscale Materials, an Office of Science user facility, was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. The authors acknowledge Dr. Henry Chan for the helpful discussions and suggestions.
Publisher Copyright:
© 2021, The Author(s).
PY - 2021/12
Y1 - 2021/12
N2 - 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.
AB - 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.
UR - http://www.scopus.com/inward/record.url?scp=85111172376&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85111172376&partnerID=8YFLogxK
U2 - 10.1038/s41524-021-00573-x
DO - 10.1038/s41524-021-00573-x
M3 - Article
AN - SCOPUS:85111172376
SN - 2057-3960
VL - 7
JO - npj Computational Materials
JF - npj Computational Materials
IS - 1
M1 - 113
ER -