Deep Learning Based Superconductivity: Prediction and Experimental Tests

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Kaplan, Daniel, Zhang, Adam, Blawat, Joanna, Jin, Rongying, Cava, Robert J., Oudovenko, Viktor, Kotliar, Gabriel, Sengupta, Anirvan M., Xie, Weiwei
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916581176508416
author Kaplan, Daniel
Zhang, Adam
Blawat, Joanna
Jin, Rongying
Cava, Robert J.
Oudovenko, Viktor
Kotliar, Gabriel
Sengupta, Anirvan M.
Xie, Weiwei
author_facet Kaplan, Daniel
Zhang, Adam
Blawat, Joanna
Jin, Rongying
Cava, Robert J.
Oudovenko, Viktor
Kotliar, Gabriel
Sengupta, Anirvan M.
Xie, Weiwei
contents The discovery of novel superconducting materials is a longstanding challenge in materials science, with a wealth of potential for applications in energy, transportation, and computing. Recent advances in artificial intelligence (AI) have enabled expediting the search for new materials by efficiently utilizing vast materials databases. In this study, we developed an approach based on deep learning (DL) to predict new superconducting materials. We have synthesized a compound derived from our DL network and confirmed its superconducting properties in agreement with our prediction. Our approach is also compared to previous work based on random forests (RFs). In particular, RFs require knowledge of the chem-ical properties of the compound, while our neural net inputs depend solely on the chemical composition. With the help of hints from our network, we discover a new ternary compound $\textrm{Mo}_{20}\textrm{Re}_{6}\textrm{Si}_{4}$, which becomes superconducting below 5.4 K. We further discuss the existing limitations and challenges associated with using AI to predict and, along with potential future research directions.
format Preprint
id arxiv_https___arxiv_org_abs_2412_13012
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Deep Learning Based Superconductivity: Prediction and Experimental Tests
Kaplan, Daniel
Zhang, Adam
Blawat, Joanna
Jin, Rongying
Cava, Robert J.
Oudovenko, Viktor
Kotliar, Gabriel
Sengupta, Anirvan M.
Xie, Weiwei
Machine Learning
Materials Science
Strongly Correlated Electrons
The discovery of novel superconducting materials is a longstanding challenge in materials science, with a wealth of potential for applications in energy, transportation, and computing. Recent advances in artificial intelligence (AI) have enabled expediting the search for new materials by efficiently utilizing vast materials databases. In this study, we developed an approach based on deep learning (DL) to predict new superconducting materials. We have synthesized a compound derived from our DL network and confirmed its superconducting properties in agreement with our prediction. Our approach is also compared to previous work based on random forests (RFs). In particular, RFs require knowledge of the chem-ical properties of the compound, while our neural net inputs depend solely on the chemical composition. With the help of hints from our network, we discover a new ternary compound $\textrm{Mo}_{20}\textrm{Re}_{6}\textrm{Si}_{4}$, which becomes superconducting below 5.4 K. We further discuss the existing limitations and challenges associated with using AI to predict and, along with potential future research directions.
title Deep Learning Based Superconductivity: Prediction and Experimental Tests
topic Machine Learning
Materials Science
Strongly Correlated Electrons
url https://arxiv.org/abs/2412.13012