Accelerated Atomistic Modeling of Solid-State Battery Materials With Machine Learning

Guo, Haoyue; Wang, Qian; Stuke, Annika; Urban, Alexander; Artrith, Nongnuch

doi:10.3389/fenrg.2021.695902

Cited by 43 publications

(28 citation statements)

References 169 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine-learned force fields are an emerging class of simulation tools in the area of battery materials research, − and this has included initial applications of GAP models (Figure ). A long-term goal of such research would be to compute voltage curves that correspond to the experimental charging and discharging processes.…”

Section: Applications (I): Force Fieldsmentioning

confidence: 99%

Gaussian Process Regression for Materials and Molecules

et al. 2021

View full text Add to dashboard Cite

We provide an introduction to Gaussian process regression (GPR) machine-learning methods in computational materials science and chemistry. The focus of the present review is on the regression of atomistic properties: in particular, on the construction of interatomic potentials, or force fields, in the Gaussian Approximation Potential (GAP) framework; beyond this, we also discuss the fitting of arbitrary scalar, vectorial, and tensorial quantities. Methodological aspects of reference data generation, representation, and regression, as well as the question of how a data-driven model may be validated, are reviewed and critically discussed. A survey of applications to a variety of research questions in chemistry and materials science illustrates the rapid growth in the field. A vision is outlined for the development of the methodology in the years to come.

show abstract

Section: Applications (I): Force Fieldsmentioning

confidence: 99%

Gaussian Process Regression for Materials and Molecules

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Building ML potentials for solid-state electrode material simulations [20,24,252] is easier than for the SEI, as key properties can be obtained with an ML potential even if it is only accurate in the vicinity of the equilibrium structure and only captures short-range interactions. On the contrary, for effective SEI simulations, ML potentials need to handle long-range anisotropic electrostatic interactions and polarization, as well as be accurate in far-from-equilibrium structures that constitute reaction pathways.…”

Section: Machine Learning Potentialsmentioning

confidence: 99%

Modeling the Solid Electrolyte Interphase: Machine Learning as a Game Changer?

Diddens

Appiah

Mabrouk

et al. 2022

Adv Materials Inter

View full text Add to dashboard Cite

The solid electrolyte interphase (SEI) is a complex passivation layer that forms in situ on many battery electrodes such as lithium‐intercalated graphite or lithium metal anodes. Its essential function is to prevent the electrolyte from continuous electrochemical degradation, while simultaneously allowing ions to pass through, thus constituting an electronically insulating, but ionically conducting material. Its properties crucially affect the overall performance and aging of a battery cell. Despite decades of intense research, understanding the SEI's precise formation mechanism, structure, composition, and evolution remains a conundrum. State‐of‐the‐art computational modeling techniques are powerful tools to gain additional insights, although confronted with a trade‐off between accuracy and accessible time‐ and length scales. In this review, it is discussed how recent advances in data‐driven models, especially the development of fast and accurate surrogate simulators and deep generative models, can work with physics‐based and physics‐informed approaches to enable the next generation of breakthroughs in this field. Machine learning‐enhanced multiscale models can provide new pathways to inverse the design of interphases with desired properties.

show abstract

“…Batteries are the dominant source of energy for diverse applications and main work-horse for portable electronics [1,2]. Common examples where batteries are increasingly adopted are electric vehicles and grid energy storage [3,4].…”

Section: Introductionmentioning

confidence: 99%

Accurate Prediction of Voltage of Battery Electrode Materials Using Attention-Based Graph Neural Networks

Louis

Siriwardane

Joshi

et al. 2022

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

Performing first principle calculations to discover electrodes' properties in the large chemical space is a challenging task. While machine learning (ML) has been applied to effectively accelerate those discoveries, most of the applied methods ignore the materials' spatial information and only use pre-defined features: based only on chemical compositions. We propose two attention-based graph convolutional neural network techniques to learn the average voltage of electrodes. Our proposed method, which combines both atomic composition and atomic coordinates in 3D-space, improves the accuracy in voltage prediction by 17% when compared to composition based ML models. The first model directly learns the chemical reaction of electrodes and metal-ions to predict their average voltage, whereas the second model combines electrodes' ML predicted formation energy (E form ) to compute their average voltage. Our models demonstrates improved accuracy in transferability from our subset of learned metal-ions to other metal-ions.

show abstract

Accelerated Atomistic Modeling of Solid-State Battery Materials With Machine Learning

Cited by 43 publications

References 169 publications

Gaussian Process Regression for Materials and Molecules

Gaussian Process Regression for Materials and Molecules

Modeling the Solid Electrolyte Interphase: Machine Learning as a Game Changer?

Accurate Prediction of Voltage of Battery Electrode Materials Using Attention-Based Graph Neural Networks

Contact Info

Product

Resources

About