Construction of High Accuracy Machine Learning Interatomic Potential for Surface/Interface of Nanomaterials—A Review

Abstract

The inherent discontinuity and unique dimensional attributes of nanomaterial surfaces and interfaces bestow them with various exceptional properties. These properties, however, also introduce difficulties for both experimental and computational studies. The advent of machine learning interatomic potential (MLIP) addresses some of the limitations associated with empirical force fields, presenting a valuable avenue for accurate simulations of these surfaces/interfaces of nanomaterials. Central to this approach is the idea of capturing the relationship between system configuration and potential energy, leveraging the proficiency of machine learning (ML) to precisely approximate high‐dimensional functions. This review offers an in‐depth examination of MLIP principles and their execution, and elaborates on their applications in the realm of nanomaterial surface and interface systems. We also discuss the prevailing challenges faced by this potent methodology.This article is protected by copyright. All rights reserved

Biomimetic fusion: Platyper's dual vision for predicting protein–surface interactions

“…One promising approach is the use of machine learning, particularly neural networks, which offer a potential solution to optimize the balance between computational efficiency and accurate dynamics prediction in SMD simulations. [11][12][13] Unlike the classical force-field based analytic function approach, machine-learning interatomic potentials, such as neural isotropic interatomic potential (NequIP) based on E(3) isotropic neural networks, are agnostic to the bonding topology of the system and treat all interactions equally based on the relative interatomic positions and interacting chemicals. 14 These rotationally invariant graph neural network interatomic potential (GNN-IP) models eliminate the need for handcrafted descriptors and instead learn geometrical data that represent graph-invariant features of atoms.…”

Minimizing Redundancy and Data Requirements of Machine Learning Potential: A Case Study in Interface Combustion

“…Interface sampling poses a significant challenge in realizing the machine learning potential, especially in the domains of catalysis and batteries, particularly in complex combustion reactions. 32 Solid propellants are typically composed of metal additives and an oxidizer, i.e., aluminum (Al) and ammonium perchlorate (AP), which account for approximately 20 and 60% of the composition, respectively. Although Al powder boasts high energy density, cost efficiency, and established safety, its practical implementation is impeded by various factors.…”

“…Interface sampling poses a significant challenge in realizing the machine learning potential, especially in the domains of catalysis and batteries, particularly in complex combustion reactions . Solid propellants are typically composed of metal additives and an oxidizer, i.e.…”