{"title":"离子化合物非晶化焓的机器学习加速预测","authors":"Qian Yu, Guang Sun and Wei Luo*, ","doi":"10.1021/acsmaterialslett.5c0006610.1021/acsmaterialslett.5c00066","DOIUrl":null,"url":null,"abstract":"<p >Amorphous ionic materials hold promise for advanced energy storage, electrocatalysis, and optical devices, yet systematically evaluating their propensity to form amorphous phases remains underdeveloped. Here, we present a machine learning framework for efficiently predicting amorphization enthalpy, a universal measure of the thermodynamic cost of converting crystalline ionic compounds to amorphous phases. By combining a standardized DFT-based melt-and-quench protocol with machine learning, we build a training set of 407 compounds and identify key descriptors linked to the amorphization enthalpy. A random forest regressor highlights relevant features but shows a limited predictive power. To overcome this, we implement e3nn, one of the state-of-art graph neural networks (GNN), and adopt a transfer-learning approach. Applying this GNN to screen 12,123 ionic compounds reveals that nitrides and sulfides typically resist amorphization, while alkali- or halogen-rich and multi-cation compositions favor it. Our data-driven approach offers a practical guide for discovering novel amorphous ionic materials, accelerating experimental development across diverse applications.</p>","PeriodicalId":19,"journal":{"name":"ACS Materials Letters","volume":"7 4","pages":"1496–1502 1496–1502"},"PeriodicalIF":9.6000,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Machine Learning-Accelerated Prediction of Amorphization Enthalpy in Ionic Compounds\",\"authors\":\"Qian Yu, Guang Sun and Wei Luo*, \",\"doi\":\"10.1021/acsmaterialslett.5c0006610.1021/acsmaterialslett.5c00066\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p >Amorphous ionic materials hold promise for advanced energy storage, electrocatalysis, and optical devices, yet systematically evaluating their propensity to form amorphous phases remains underdeveloped. Here, we present a machine learning framework for efficiently predicting amorphization enthalpy, a universal measure of the thermodynamic cost of converting crystalline ionic compounds to amorphous phases. By combining a standardized DFT-based melt-and-quench protocol with machine learning, we build a training set of 407 compounds and identify key descriptors linked to the amorphization enthalpy. A random forest regressor highlights relevant features but shows a limited predictive power. To overcome this, we implement e3nn, one of the state-of-art graph neural networks (GNN), and adopt a transfer-learning approach. Applying this GNN to screen 12,123 ionic compounds reveals that nitrides and sulfides typically resist amorphization, while alkali- or halogen-rich and multi-cation compositions favor it. Our data-driven approach offers a practical guide for discovering novel amorphous ionic materials, accelerating experimental development across diverse applications.</p>\",\"PeriodicalId\":19,\"journal\":{\"name\":\"ACS Materials Letters\",\"volume\":\"7 4\",\"pages\":\"1496–1502 1496–1502\"},\"PeriodicalIF\":9.6000,\"publicationDate\":\"2025-03-21\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"ACS Materials Letters\",\"FirstCategoryId\":\"92\",\"ListUrlMain\":\"https://pubs.acs.org/doi/10.1021/acsmaterialslett.5c00066\",\"RegionNum\":1,\"RegionCategory\":\"化学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"MATERIALS SCIENCE, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"ACS Materials Letters","FirstCategoryId":"92","ListUrlMain":"https://pubs.acs.org/doi/10.1021/acsmaterialslett.5c00066","RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MATERIALS SCIENCE, MULTIDISCIPLINARY","Score":null,"Total":0}
Machine Learning-Accelerated Prediction of Amorphization Enthalpy in Ionic Compounds
Amorphous ionic materials hold promise for advanced energy storage, electrocatalysis, and optical devices, yet systematically evaluating their propensity to form amorphous phases remains underdeveloped. Here, we present a machine learning framework for efficiently predicting amorphization enthalpy, a universal measure of the thermodynamic cost of converting crystalline ionic compounds to amorphous phases. By combining a standardized DFT-based melt-and-quench protocol with machine learning, we build a training set of 407 compounds and identify key descriptors linked to the amorphization enthalpy. A random forest regressor highlights relevant features but shows a limited predictive power. To overcome this, we implement e3nn, one of the state-of-art graph neural networks (GNN), and adopt a transfer-learning approach. Applying this GNN to screen 12,123 ionic compounds reveals that nitrides and sulfides typically resist amorphization, while alkali- or halogen-rich and multi-cation compositions favor it. Our data-driven approach offers a practical guide for discovering novel amorphous ionic materials, accelerating experimental development across diverse applications.
期刊介绍:
ACS Materials Letters is a journal that publishes high-quality and urgent papers at the forefront of fundamental and applied research in the field of materials science. It aims to bridge the gap between materials and other disciplines such as chemistry, engineering, and biology. The journal encourages multidisciplinary and innovative research that addresses global challenges. Papers submitted to ACS Materials Letters should clearly demonstrate the need for rapid disclosure of key results. The journal is interested in various areas including the design, synthesis, characterization, and evaluation of emerging materials, understanding the relationships between structure, property, and performance, as well as developing materials for applications in energy, environment, biomedical, electronics, and catalysis. The journal has a 2-year impact factor of 11.4 and is dedicated to publishing transformative materials research with fast processing times. The editors and staff of ACS Materials Letters actively participate in major scientific conferences and engage closely with readers and authors. The journal also maintains an active presence on social media to provide authors with greater visibility.