Unsupervised discovery of solid-state lithium ion conductors

Zhang, Ying; He, Xingfeng; Chen, Zhiqian; Bai, Qiang; Nolan, Adelaide M.; Roberts, Charles A.; Banerjee, Debasish; Matsunaga, Takuro; Mo, Yifei; Ling, Chen

doi:10.1038/s41467-019-13214-1

Cited by 198 publications

(196 citation statements)

References 57 publications

Supporting

Mentioning

173

Contrasting

Order By: Relevance

“…For instance, Tran et al applied active learning method to predict the electrocatalytic performance for CO 2 reduction and H 2 evolution. Zhang et al used unsupervised learning to develop a helpful method for investigations with small datasets and successfully propose potential solid electrolytes for Li batteries. Sun et al applied unsupervised method to classify different ternary nitrides.…”

Section: Challenges and Perspectivesmentioning

confidence: 99%

Machine learning: Accelerating materials development for energy storage and conversion

Chen

Zhou

2020

InfoMat

249

171

View full text Add to dashboard Cite

With the development of modern society, the requirement for energy has become increasingly important on a global scale. Therefore, the exploration of novel materials for renewable energy technologies is urgently needed. Traditional methods are difficult to meet the requirements for materials science due to long experimental period and high cost. Nowadays, machine learning (ML) is rising as a new research paradigm to revolutionize materials discovery.In this review, we briefly introduce the basic procedure of ML and common algorithms in materials science, and particularly focus on latest progress in applying ML to property prediction and materials development for energyrelated fields, including catalysis, batteries, solar cells, and gas capture. Moreover, contributions of ML to experiments are involved as well. We highly expect that this review could lead the way forward in the future development of ML in materials science. K E Y W O R D Sbig data, energy storage and conversion, machine learning, property prediction

show abstract

Section: Challenges and Perspectivesmentioning

confidence: 99%

Machine learning: Accelerating materials development for energy storage and conversion

Chen

Zhou

2020

InfoMat

249

171

View full text Add to dashboard Cite

show abstract

“…We note, therefore, that ML has also been proposed for the discovery of solid-state Li ion conductors without explicit simulation. Two examples of such materials discovery applications are a study based on unsupervised learning by Zhang et al [221] and a transferlearning approach applied to billions of candidate materials by Cubuk et al [222].…”

Section: Properties Of Battery Materialsmentioning

confidence: 99%

Machine learning-accelerated quantum mechanics-based atomistic simulations for industrial applications

Morawietz

Artrith

2020

J Comput Aided Mol Des

View full text Add to dashboard Cite

Atomistic simulations have become an invaluable tool for industrial applications ranging from the optimization of protein-ligand interactions for drug discovery to the design of new materials for energy applications. Here we review recent advances in the use of machine learning (ML) methods for accelerated simulations based on a quantum mechanical (QM) description of the system. We show how recent progress in ML methods has dramatically extended the applicability range of conventional QM-based simulations, allowing to calculate industrially relevant properties with enhanced accuracy, at reduced computational cost, and for length and time scales that would have otherwise not been accessible. We illustrate the benefits of ML-accelerated atomistic simulations for industrial R&D processes by showcasing relevant applications from two very different areas, drug discovery (pharmaceuticals) and energy materials. Writing from the perspective of both a molecular and a materials modeling scientist, this review aims to provide a unified picture of the impact of ML-accelerated atomistic simulations on the pharmaceutical, chemical, and materials industries and gives an outlook on the exciting opportunities that could emerge in the future.

show abstract

“…[20][21][22][23][24][25] Within the Li10GeP2S12 family, roomtemperature ionic conductivities have been reported in excess of 10 mS cm -1 , 19 and similarly high ionic conductivities have been reported for other lithium thiophosphates. 11,12,19 Understanding the factors that cause specific solid electrolytes to exhibit fast or slow ionictransport is a key research question, in part because such an understanding can inform the development of general "design rules" and support the identification and optimization of new fast-ion conducting materials, [26][27][28][29] thereby broadening the pool of candidate solid electrolytes for future solid-state battery applications. A partial answer to the question of what makes some solid electrolytes much faster ionic conductors than others comes from an understanding of favorable structural motifs-for example, fast-ion conduction is favored in materials that possess highly connected networks of lithium-diffusion pathways.…”

Section: Introductionmentioning

confidence: 99%

Evidence for a Solid-Electrolyte Inductive Effect in Superionic Conductors

Culver¹,

Squires²,

Minafra³

et al. 2020

Preprint

View full text Add to dashboard Cite

Identifying and optimizing highly-conducting lithium-ion solid electrolytes is a critical step towards the realization of commercial all–solid-state lithium-ion batteries. Strategies to enhance ionic conductivities in solid electrolytes typically focus on the effects of modifying their crystal structures or of tuning mobile-ion stoichiometries. A less-explored approach is to modulate the chemical-bonding interactions within a material to promote fast lithium-ion diffusion. Recently, the idea of a solid-electrolyte inductive effect was proposed, whereby changes in bonding within the solid-electrolyte host-framework modify the potential-energy landscape for the mobile ions, resulting in an enhanced ionic conductivity. This concept has since been invoked to explain anomalous conductivity trends in a number of solid electrolytes. Direct evidence for a solid-electrolyte inductive effect, however, is lacking—in part because of the challenge of quantifying changes in local bonding interactions within a solid-electrolyte host-framework. <a></a><a>Here, we consider the evidence for a solid-electrolyte inductive effect in the archetypal superionic lithium-ion conductor Li10Ge1−xSnxP2S12, using Rietveld refinements against high-resolution temperature-dependent neutron-diffraction data, Raman spectroscopy, and density functional theory calculations.</a> Substituting Ge for Sn weakens the {Ge,Sn}–S bonding interactions and increases the charge-density associated with the S2- ions. This charge redistribution modifies the Li+ substructure causing Li+ ions to bind more strongly to the host-framework S anions; which in turn modulates the Li-ion potential-energy surface, increasing local barriers for Li-ion diffusion. Each of these effects is consistent with the predictions of the solid-electrolyte inductive effect model. Density functional theory calculations further predict that this inductive effect occurs even in the absence of changes to the host-framework geometry due to Ge → Sn substitution. These results provide direct evidence in support of a measurable solid-electrolyte inductive effect and demonstrate its application as a practical strategy for tuning ionic conductivities in superionic lithium-ion conductors.

show abstract

Unsupervised discovery of solid-state lithium ion conductors

Cited by 198 publications

References 57 publications

Machine learning: Accelerating materials development for energy storage and conversion

Machine learning: Accelerating materials development for energy storage and conversion

Machine learning-accelerated quantum mechanics-based atomistic simulations for industrial applications

Evidence for a Solid-Electrolyte Inductive Effect in Superionic Conductors

Contact Info

Product

Resources

About