Abstract:Identifying new materials that combine high ionic conductivity with structural and electrochemical stability so far remains a slow trial and error search process. To rationally accelerate materials design and exploit the opportunities in the materials genome a dependable rapid screening of materials is required that can pre-select structures that merit higher level computational as well as experimental characterization. Here we report on the progress of our “softBV” bond-valence site energy-based automated pathway analysis utilizing our new Bond Valence Pathway Analyzer (BVPA) – with fast bond valence site energy calculations to quickly obtain suggested candidate materials for fast ionic conduction. BVPA provides rapid and simplified visualization, in order to bridge the gap between experimentalists and simulation [1,2]. Calculation of ion transport pathways can be done extremely quickly on the order of seconds or minutes on desktop PCs providing a speedup factor of 3 to 5 orders of magnitude compared to DFT-based NEB methods. Combined with a graphical user interface our software suite (that can be downloaded from [2] and is free for academic use) should enable experimentalists to quickly identify candidate solid electrolyte materials. We also aim to integrate the pre-screening into an automated workflow for subsequent DFT characterization [3]. Results will be benchmarked against both experimental and DFT NEB migration barriers. Besides the migration barriers the approach now also comprises an AI-based dopant predictor utilizing bond-valence-based crystal chemical descriptors to assist experimentalists in exploring favorable substitutional doping strategies. We will also compare the predictability of absolute room temperature conductivities from static energy landscape analysis, bond-valence based empirical MD simulations and ab initio molecular dynamics (AIMD) simulations. While for small fast-ion conductor structures at sufficiently high temperatures AIMD appears to be the gold standard, the less reliable but computationally empirical approaches have an advantage in modelling complex disordered interfaces at low temperatures over longer periods. This eliminates the hazards involved in extrapolations down to room temperature properties for the frequent cases of order-disorder phase transitions at intermediate temperatures. As an example we will discuss lithium and sodium compounds containing multiple anions, in particular the combination of thiophosphate and halide anions or various MS 4 polyanions. Based on computational screening using our bond valence site approach and DFT studies several thiophosphate halides along the A 3 PS 4 -LiX (Cl, Br, I; A = Li, Na) tie line [4] and the A x (MS 4 ) y (M’S 4 ) z phase space [5] have been explored and their properties discussed based on BVSE pathway models and molecular dynamics simulations in combination with experimental (X-ray and neutron) diffraction, solid state NMR and electrochemical characterisation. MD simulations e.g. show that Li 5 (PS 4 )Cl 2 is found to undergo an order-disorder phase transition and thus should, contrasting to earlier predictions, not be a fast Li + ion conductor. The newly predicted thermodynamically stable cubic solid electrolyte Li 15 (PS 4 ) 4 Cl 3 was successfully prepared and characterized. Though its conductivity does not reach a superionic level, it demonstrates that the computational approach can successfully predict a completely new classes of solid electrolytes and can predict its optimization by doping. The simplicity of the approach also facilitates the study of homogeneity ranges as exemplified for the solid solution systems Li 4-x PS 4 I x (0<x<0.67) and Na 9+x (MS 4 ) 3-x (SnS 4 ) x (M = P, Sb; x ≈ 2). References: [1] L.L. Wong, K.C. Phuah, H. Chen, W.S. Chew, R. Dai, S Adams; submitted. [2] http://www.dmse.nus.edu.sg/asn/software.html [2] B. He, S. Chi, A. Ye et al.; accepted in npj Scientific Data 2020. [3] R. Prasada Rao, H. Chen, S. Adams; Chemistry of Materials 31 (2019) 8649-8662. [4] A. Sorkin, S. Adams, accepted in Materials Advances 2020.

Discovering Virtual Na-based Argyrodites as Solid-State Electrolytes Using DFT, AIMD, and Machine Learning Techniques

Discovering Multi‐Compositional Li‐Argyrodite Solid‐State Electrolytes via Experimental Active Learning

Highly reliable and large-scale simulations of promising argyrodite solid-state electrolytes using a machine-learned moment tensor potential

Stability and ionic mobility in argyrodite-related lithium-ion solid electrolytes

Screening Platform for Promising Na Superionic Conductors for Na-Ion Solid-State Electrolytes

Lithium Localization by Anions in Argyrodite Solid Electrolytes from Machine‐Learning‐based Simulations

Optimizing ionic transport in argyrodites: a unified view on the role of sulfur/halide distribution and local environments

Optimizing ionic conductivity of lithium in Li$_7$PS$_6$ argyrodite via dopant engineering

Computational screening of sodium solid electrolytes through unsupervised learning

Predicting Reactivity and Passivation of Solid-State Battery Interfaces

Design of Multicomponent Argyrodite Based on a Mixed Oxidation State as Promising Solid-State Electrolyte using Moment Tensor Potentials

Electrolytomics: A unified big data approach for electrolyte design and discovery

Computational Investigation of Halide Solid Electrolytes for Na-Based Batteries

(Keynote) An Automatic Rapid Screening Tool for Fast Ionic Conductors

Argyrodite configuration determination for DFT and AIMD calculations using an integrated optimization strategy

Machine Learning-Aided Discovery of Superionic Solid-State Electrolyte for Li-Ion Batteries

Quantifying the search for solid Li-ion electrolyte materials by anion: a data-driven perspective

Accelerated discovery of solid-state electrolytes using Bayesian optimisation

Structural properties of amorphous Na$_3$OCl electrolyte by first-principles and machine learning molecular dynamics

Data-mining Fluoride-Based Solid-State Electrolytes for Monovalent Metal Batteries

A New Electrolyte Formulation for Securing High Temperature Cycling and Storage Performances of Na‐Ion Batteries