Magnetic and superconducting phase diagrams and transition temperatures predicted using text mining and machine learning

Abstract

Predicting the properties of materials prior to their synthesis is of great importance in materials science. Magnetic and superconducting materials exhibit a number of unique properties that make them useful in a wide variety of applications, including solid oxide fuel cells, solid-state refrigerants, photon detectors and metrology devices. In all these applications, phase transitions play an important role in determining the feasibility of the materials in question. Here, we present a pipeline for fully integrating data extracted from the scientific literature into machine-learning tools for property prediction and materials discovery. Using advanced natural language processing (NLP) and machine-learning techniques, we successfully reconstruct the phase diagrams of well-known magnetic and superconducting compounds, and demonstrate that it is possible to predict the phase-transition temperatures of compounds not present in the database. We provide the tool as an online open-source platform, forming the basis for further research into magnetic and superconducting materials discovery for potential device applications.npj Computational Materials (2020) 6:18 ; https://doi.

A database of battery materials auto-generated using ChemDataExtractor

“…This is because any ‘rogue data’ in the database is mitigated by the nature of the downstream analysis. For example, the aforementioned database of Curie and Néel temperatures with 73% precision has successfully reconstructed phase diagrams of magnetic materials and predicted phase-transition temperatures using machine-learning (ML) methods 85 . ML methods will naturally filter out erroneous data as outliers via the intrinsic nature of their data analytics procedure.…”

What Can Text Mining Tell Us About Lithium‐Ion Battery Researchers’ Habits?

“…Despite the lack of dedicated databases, TM (combined or not with AI/ML) has already proved its potential for leading to new discoveries and knowledge in the materials/energy field [10,54–59] . As a benchmark in the battery community, Huang et al .…”

Thermodynamic exploration of xenon/krypton separation based on a high-throughput screening

“…6,18,19 Thirdly, text and data mining are allowing to augment databases with content previously thought not being machine-readable and indexable, for example by identifying unreported properties of materials in older scientific papers. 20,21 Finally, the use of artificial intelligence techniques, such as statistical learning, 22 can 2 Current state of the art…”