Nav Nidhi Rajput, Stony Brook University: "Materials Informatics for Structure-Property Relationships (MISPR) for Liquid Solution"

10/31/2024
 | 
3 PM (US Eastern)
https://iu.zoom.us/meeting/register/tZYqd-2srD8tGtCXDem4Cka08rBz5

Liquid solutions are critical components of various chemical, materials science, engineering, and biological applications such as batteries, fuel, food industry, and drug discovery. Optimizing the performance of these technologies often necessitates exploring vast chemical and parameter spaces to avoid bias and derive trends only accessible by systematic datasets. In addition, understanding the structure, transport, and thermodynamic properties of chemical species comprising the solution is paramount to optimizing the performance of liquid applications. However, identifying the underlying correlations between the functional properties and the atomistic interactions can be a daunting task for multi-component complex liquid solutions, even by using advanced experimental and computational techniques. Driven by these needs, we developed an open-source high-throughput computational framework coined MISPR (Materials Informatics for Structure–Property Relationships) for guiding and accelerating materials discovery, optimization, and deployment of complex multicomponent liquid solutions.1-4 MISPR seamlessly integrates density functional theory (DFT) calculations with classical molecular dynamics (MD) simulations to robustly predict molecular and ensemble properties in complex multi-component liquid solutions. Functionalities of MISPR include (i) full automation of DFT and MD simulations, (ii) creation of computational databases for establishing structure-property relationships and maintaining data provenance and reproducibility, (iii) automatic error detection and handling, (iv) support for flexible and well-tested DFT workflows for computing properties such as bond dissociation energy, binding energy, and redox potentials, and (v) derivation of ensemble properties such radial distribution functions, ionic conductivity, and residence time. We demonstrate the unique features of MISPR by highlighting on one of its hybrid workflows that predicts stable species in liquid solutions from their nuclear magnetic resonance (NMR) chemical shifts.5 This workflow automatically extracts and categorizes hundreds of thousands of atomic clusters from MD simulations, identifies the most stable species in solution, and calculates their NMR chemical shifts via DFT calculations. The result is an output database of computed chemical shifts for liquid solutions across a wide chemical and parameter space. Our automated high-throughput framework overcomes various limitations of current NMR computational techniques,2 such as the Edisonian approach to building solvation structures, the difficulty in accounting for representative explicit solvation, and the substantial time required for manual job management. I will also discuss our recent work on using natural language processing (NLP) for extracting spectroscopy data of liquid solutions from literature.