{"title":"FusNet:通过机器学习解锁分子融合特性","authors":"Jiahui Chen, Yuxin Qiu, Wenyao Chen, Hongye Cheng, Xuezhi Duan, Zhiwen Qi, Zhen Song","doi":"10.1007/s11705-025-2593-7","DOIUrl":null,"url":null,"abstract":"<div><p>Accurate prediction of molecular fusion properties is critical for energy-efficient material design and sustainable process optimization, yet remains challenging due to data scarcity and complex thermodynamic interdependencies. This work introduces machine learning tools to address these gaps by combining expert-curated molecular descriptors with deep learning. By systematically evaluating statistical machine learning algorithms and attention-based architectures, optimized models are identified: a SMILES-augmented Transformer-Convolutional Neural Network for fusion temperature and a graph attention network for fusion enthalpy. Prediction power is further validated experimentally on four structure diverse compounds (<i>γ</i>-butyrolactone, methyl octanoate, <i>N</i>-phenylbenzenesulfonamide, and triethylene glycol dimethyl ether). Interpretability analyses reveal that these models prioritize key structures in molecules: attention in text-based models focuses on key atoms while that in graph models focuses on key chemical bonds, aligning with empirical thermodynamic evidences. By providing rapid, interpretable fusion property predictions, this framework can support the development of low-energy phase-change materials and sustainable solvent systems, advancing data-driven green chemistry.</p><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":571,"journal":{"name":"Frontiers of Chemical Science and Engineering","volume":"19 9","pages":""},"PeriodicalIF":4.5000,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"FusNet: unlocking molecular fusion properties through machine learning\",\"authors\":\"Jiahui Chen, Yuxin Qiu, Wenyao Chen, Hongye Cheng, Xuezhi Duan, Zhiwen Qi, Zhen Song\",\"doi\":\"10.1007/s11705-025-2593-7\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>Accurate prediction of molecular fusion properties is critical for energy-efficient material design and sustainable process optimization, yet remains challenging due to data scarcity and complex thermodynamic interdependencies. This work introduces machine learning tools to address these gaps by combining expert-curated molecular descriptors with deep learning. By systematically evaluating statistical machine learning algorithms and attention-based architectures, optimized models are identified: a SMILES-augmented Transformer-Convolutional Neural Network for fusion temperature and a graph attention network for fusion enthalpy. Prediction power is further validated experimentally on four structure diverse compounds (<i>γ</i>-butyrolactone, methyl octanoate, <i>N</i>-phenylbenzenesulfonamide, and triethylene glycol dimethyl ether). Interpretability analyses reveal that these models prioritize key structures in molecules: attention in text-based models focuses on key atoms while that in graph models focuses on key chemical bonds, aligning with empirical thermodynamic evidences. By providing rapid, interpretable fusion property predictions, this framework can support the development of low-energy phase-change materials and sustainable solvent systems, advancing data-driven green chemistry.</p><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>\",\"PeriodicalId\":571,\"journal\":{\"name\":\"Frontiers of Chemical Science and Engineering\",\"volume\":\"19 9\",\"pages\":\"\"},\"PeriodicalIF\":4.5000,\"publicationDate\":\"2025-07-16\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Frontiers of Chemical Science and Engineering\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://link.springer.com/article/10.1007/s11705-025-2593-7\",\"RegionNum\":3,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ENGINEERING, CHEMICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers of Chemical Science and Engineering","FirstCategoryId":"5","ListUrlMain":"https://link.springer.com/article/10.1007/s11705-025-2593-7","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, CHEMICAL","Score":null,"Total":0}
FusNet: unlocking molecular fusion properties through machine learning
Accurate prediction of molecular fusion properties is critical for energy-efficient material design and sustainable process optimization, yet remains challenging due to data scarcity and complex thermodynamic interdependencies. This work introduces machine learning tools to address these gaps by combining expert-curated molecular descriptors with deep learning. By systematically evaluating statistical machine learning algorithms and attention-based architectures, optimized models are identified: a SMILES-augmented Transformer-Convolutional Neural Network for fusion temperature and a graph attention network for fusion enthalpy. Prediction power is further validated experimentally on four structure diverse compounds (γ-butyrolactone, methyl octanoate, N-phenylbenzenesulfonamide, and triethylene glycol dimethyl ether). Interpretability analyses reveal that these models prioritize key structures in molecules: attention in text-based models focuses on key atoms while that in graph models focuses on key chemical bonds, aligning with empirical thermodynamic evidences. By providing rapid, interpretable fusion property predictions, this framework can support the development of low-energy phase-change materials and sustainable solvent systems, advancing data-driven green chemistry.
期刊介绍:
Frontiers of Chemical Science and Engineering presents the latest developments in chemical science and engineering, emphasizing emerging and multidisciplinary fields and international trends in research and development. The journal promotes communication and exchange between scientists all over the world. The contents include original reviews, research papers and short communications. Coverage includes catalysis and reaction engineering, clean energy, functional material, nanotechnology and nanoscience, biomaterials and biotechnology, particle technology and multiphase processing, separation science and technology, sustainable technologies and green processing.