Development of a Metric Tool for Material Discovery through Informatics

REU: Materials REU | AUTHORS: Axel Garfio – University of Denver, Jack Marley – Texas State University, Daniela Dominguez – University of Georgia, Nancy Ortiz – University of Colorado – Boulder

MENTORS: Eric Toberer, Vanessa Meschke, Rachel Orenstein – Colorado School of Mines

ABSTRACT

Today’s society is pushed forward by a constant improvement of technology. Technology is pushed forward by a constant drive towards finding new and improved ways of developing materials. Programming has been a big part of this project in assisting in the manipulation of large amounts of data, primarily using Python and its libraries. The goal of the project is to help drive forward new and improved techniques for finding new materials. The metric tool is developed to compare compounds by using phase diagrams and assigning a similarity index of 0.0 – 1.0. The comparison is based on the similarity of the stoichiometry of different phase diagrams. Results based on the data analyzed demonstrate that carbon-based phase diagrams have the highest similarity index. Silicon-based phase diagrams have a consistent similarity index of 0.5, encouraging a deeper analysis to determine the suitability for element replacement. In addition, Carbon and Silicon are among the most common elements among the CIFs obtained from the Materials Project website. Overall, the project aims to accelerate the discovery of stable materials by developing a metric tool for judging the similarity of phase diagrams and creating a matrix with the most similar phase diagrams.