The ability to accurately predict the macroscopic properties of complex multicomponent solutions from the microscopic features of molecular species remains a grand challenge in chemistry, materials science, and engineering. Traditional approaches—ranging from experimental trial-and-error methods to brute-force and high-throughput computational techniques—often fall short due to the inherent complexity of functional properties across multiple length and time scales, as well as the vast chemical and parameter spaces that must be explored.
To address these challenges, we developed MISPR (Materials Informatics for Structure–Property Relationships), an open-source, high-throughput, multiscale computational infrastructure (https://molmd.github.io/mispr/html/index.html).1 MISPR seamlessly integrates density functional theory (DFT) calculations, classical molecular dynamics (CMD) simulations, and machine learning (ML) techniques to enable the automation of 100–1000s of parallelized calculations with minimal manual intervention. This platform generates high-fidelity databases of computational properties and includes automated workflows for evaluating electronic, thermodynamic, structural, and dynamical properties of liquid solutions.
In this talk, I will showcase MISPR’s unique capabilities through various applications in electrolyte design for next-generation energy storage devices. Specifically, I will highlight: (1) A novel DFT-CMD-DFT approach for accurately predicting stable species in Li-ion, Na-ion, and Mg-ion batteries by analyzing experimental nuclear magnetic resonance (NMR) spectra.2-5 (2) The use of high-throughput screening to optimize atomistic interactions in Li-S battery electrolytes, leading to the development of a publicly available database, ComBat (Computational Database for Li-S Batteries). ComBat contains ~2000 properties for solvents across 16 chemical classes and has facilitated the design of fluorinated electrolyte systems with high ionic conductivity, low viscosity, and low polysulfide solubility..6, 7 (3) The development of automated workflows for high-throughput electrode-electrolyte interface calculations. By combining MISPR's robust computational framework with data-driven insights, this approach aims to accelerate the discovery and optimization of advanced materials for energy applications.
Copyright. All Rights Reserved.
