Implanted neural network potentials: Application to Li-Si alloys

Onat, Berk; Cubuk, Ekin D.; Malone, Brad D.; Kaxiras, Efthimios

doi:10.1103/physrevb.97.094106

Cited by 74 publications

(66 citation statements)

References 41 publications

(38 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…40,41 Generating the carbon structure in an ML-driven simulation bypasses the need for costly quantum-mechanical computations at runtime, making disordered anode materials such as "hard" carbons accessible to extended molecular-dynamics (MD) runs; in turn, once the carbon structures have been generated, they can be further treated with DFT to study chemical reactivity. 33 Beyond carbon, recent neural-network-type ML potentials for Li-Si phases 42,43 attest to the usefulness of such simulation tools. Even further, ML methods are beginning to be used in several other areas of energy materials research, such as the screening for suitable compositions.…”

Section: Introductionmentioning

confidence: 99%

First-principles study of alkali-metal intercalation in disordered carbon anode materials

et al. 2019

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

First-principles study of alkali-metal intercalation in disordered carbon anode materials

et al. 2019

View full text Add to dashboard Cite

show abstract

“…[26][27][28][29][30][31] Recent studies suggest that ML-based potentials are becoming viable tools for materials chemistry and physics. [32][33][34][35][36][37][38][39][40][41] We recently used such a potential for large-scale deposition simulations of ta-C films, describing the impact of thousands of individual atoms, one at a time, and achieving excellent agreement with experimental observables (including the count of fourfold-coordinated"sp 3 " atoms and mechanical properties). 38 Figure 1.…”

Section: Introductionmentioning

confidence: 99%

Computational Surface Chemistry of Tetrahedral Amorphous Carbon by Combining Machine Learning and Density Functional Theory

et al. 2018

View full text Add to dashboard Cite

Tetrahedral amorphous carbon (ta-C) is widely used for coatings due to its superior mechanical properties and has been suggested as an electrode material for detecting biomolecules. Despite extensive research, however, the complex atomic-scale structures and chemical reactivity of ta-C surfaces are incompletely understood. Here, we combine machine learning, density-functional tight-binding, and density-functional theory simulations to shed new light on this long-standing problem. We make atomistic models of ta-C surfaces, characterize them by local structural fingerprints, and provide a library of structures at different system sizes. We then move beyond the pure element and exemplify how chemical reactivity (hydrogenation and oxidation) can be modeled at the surfaces. Our work opens up new perspectives for modeling the surfaces and interfaces of amorphous solids, which will advance studies of ta-C and other functional materials.

show abstract

“…[153,164] Building on a computational description of pure disordered carbon, it was also shown how the insertion of Li ions in carbon can be described by a difference potential. [165] In this context, it is interesting to mention two separate studies that develop NN models for Li in amorphous silicon anodes: [166,167] these works provide a thematic link to the wide importance of silicon (Section 3) and underline the possibilities of ML potentials especially for energy materials modeling.…”

Section: Carbon Nanomaterialsmentioning

confidence: 99%

Machine Learning Interatomic Potentials as Emerging Tools for Materials Science

2019

View full text Add to dashboard Cite

In pursuit of this goal, atomic-scale computer simulations have long been a central approach, and two major families of methods are routinely used today. On the one hand, there are quantum-mechanical simulations, in which we solve Schrödinger's equation for the electronic structure of molecular and periodic systems, most widely based on density-functional theory (DFT). [6][7][8] These methods provide (largely) reliable results for structural models of materials that normally contain a few tens or hundreds of atoms. State-of-the-art DFT methods can be applied to many material classes, and they are increasingly used for high-throughput screening and "in silico" (computer-based) design of materials: new compositions and previously unknown structures have been identified in DFT searches and subsequently experimentally realized. [5,[9][10][11] On the other hand, interatomic potential models ("force fields"), parameterizing interactions between atoms with (relatively) simple functional forms, are widely used in materials science to describe matter in molecular dynamics (MD) simulations. These simulations grant access to larger time and length scales, reaching system sizes of up to hundreds of thousands of atoms. [12] In parameterizing these potentials, a certain physical form of the atomic interactions is assumed, often in terms of bond distances, angles, and so on, and physical properties such as equilibrium lattice parameters or elastic constants enter the fitting of the potential. For this reason, such potentials are often called "empirical." They are several orders of magnitude faster than DFT, but necessarily less accurate and less easily transferable.In this Progress Report, we highlight recent developments in "machine-learned" interatomic potentials, which represent a rapidly growing field that promises to do away with the aforementioned trade-offs. Over the last year, there has been a surge of interest in machine learning (ML) methodology: part of it is due to the dramatic growth of ML throughout the scientific disciplines, and part of it is due to tangible success stories of ML-based interatomic potentials that are now beginning to emerge. We will argue that this is an exciting development with very practical implications, currently on the verge of moving from a somewhat specialized new technology to everyday applicability, poised to enhance and complement the communities' existing strengths in computational materials modeling. We will show selected applications of ML potentials to problems in materials science, discuss the current limitations (and possible pitfalls), and outline what we expect to be interesting directions for the development of the field in the coming years.Atomic-scale modeling and understanding of materials have made remarkable progress, but they are still fundamentally limited by the large computational cost of explicit electronic-structure methods such as density-functional theory. This Progress Report shows how machine learning (ML) is currently enabling a new degree of realism in materi...

show abstract

Implanted neural network potentials: Application to Li-Si alloys

Cited by 74 publications

References 41 publications

First-principles study of alkali-metal intercalation in disordered carbon anode materials

First-principles study of alkali-metal intercalation in disordered carbon anode materials

Computational Surface Chemistry of Tetrahedral Amorphous Carbon by Combining Machine Learning and Density Functional Theory

Machine Learning Interatomic Potentials as Emerging Tools for Materials Science

Contact Info

Product

Resources

About