(GR5) Machine Learning for Materials Informatics

Authors: Eric Vertina*, Emily Sutherland Drew Fitzgerald

Advisors:

Category: Graduate

Abstract:

MXenes are a hot topic in materials science research because of their expected unique properties and myriad applications, such as more efficient energy conversion in batteries and solar cells, environmental and water treatment, and many additional applications. This project aims to produce Machine Learning (ML) models that accurately predict certain MXene properties – like electrical conductivity, work function, carrier density, mobility, life-time, and sensitivity to disorder – based on standard elemental information (e.g., electronegativity of each constituent element of the MXene, atomic mass of a MXene molecule, etc.), with training data found from literature as well as data produced by our project’s Density Functional Theory (DFT) team.
*This project is part of the NSF Circular Economy and Data Analytics Engineering Research for Sustainability (CEDAR) grant WPI has received.

UN SDGs:

*This author is submitting separately, as each member contributes to vastly different aspects of the project