Multi-model framework for intelligent research and development of super-hydrophilic coatings

IF 7.7 2区材料科学 Q1 MATERIALS SCIENCE, MULTIDISCIPLINARY

Science China Materials Pub Date : 2026-06-04 DOI:10.1007/s40843-026-4146-8

Lie Wu (, ), Shiyu Fu (, ), Yutang Li (, ), Zhanyuan Li (, ), Bing Wang (, ), Yiqi Li (, ), Fan Yang (, ), Ming Gao (, ), Xin Wang (, ), Wenhua Zhou (, ), Huijuan Ma (, ), Paul K. Chu (, ), Xue-Feng Yu (, )

{"title":"Multi-model framework for intelligent research and development of super-hydrophilic coatings","authors":"Lie Wu \n (, ), Shiyu Fu \n (, ), Yutang Li \n (, ), Zhanyuan Li \n (, ), Bing Wang \n (, ), Yiqi Li \n (, ), Fan Yang \n (, ), Ming Gao \n (, ), Xin Wang \n (, ), Wenhua Zhou \n (, ), Huijuan Ma \n (, ), Paul K. Chu \n (, ), Xue-Feng Yu \n (, )","doi":"10.1007/s40843-026-4146-8","DOIUrl":null,"url":null,"abstract":"<div><p>The conventional trial-and-error approach for the research and development (R&D) of high-performance super-hydrophilic coatings presents long-standing challenges, including data-scarce, unclear structure-activity relationships and design guidelines. In this study, an innovative multi-model framework that synergistically integrates a specialized polymer large language model (PolyLLM) and a polymer machine learning model (PolyML) for intelligent R&D of super-hydrophilic coating systems is designed and demonstrated. During the hydrophilic polymer research phase, the multi-model leverages the domain-specific insights and chemical tool capabilities of the fine-tuned PolyLLM to systematically screen 852 hydrophilic monomers and generate 3880 hypothetical polymer architectures. By utilizing the PolyML, the multi-model enables precise performance predictions and quantitative evaluations of feature importance to facilitate efficient high-throughput screening of optimal structures from the generated polymer database. Experimental validation confirms that the multi-model achieves a mean absolute error (MAE) below 10% across various polymer property prediction tasks compared to experimental data. This approach leads to the rapid discovery of hydrophilic polymers with exceptional anti-swelling and wear-resistance. In developing super-hydrophilic coatings, the multi-model effectively combines PolyLLM’s guidance for large-scale synthesis with PolyML’s capabilities for precise formulation optimization and curing parameter adjustments. The effectiveness of multi-model is demonstrated by the accelerated development of high-performance anti-fogging coatings for swim goggles and optical films with outstanding water- and wear-resistance that significantly outperform leading commercial products. This multi-model offers a flexible approach for advanced polymer coatings, leading to a seamless connection between laboratory research and industrial development.</p><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":773,"journal":{"name":"Science China Materials","volume":"69 9","pages":"5653 - 5664"},"PeriodicalIF":7.7000,"publicationDate":"2026-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Science China Materials","FirstCategoryId":"88","ListUrlMain":"https://link.springer.com/article/10.1007/s40843-026-4146-8","RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MATERIALS SCIENCE, MULTIDISCIPLINARY","Score":null,"Total":0}

引用次数: 0

Abstract

The conventional trial-and-error approach for the research and development (R&D) of high-performance super-hydrophilic coatings presents long-standing challenges, including data-scarce, unclear structure-activity relationships and design guidelines. In this study, an innovative multi-model framework that synergistically integrates a specialized polymer large language model (PolyLLM) and a polymer machine learning model (PolyML) for intelligent R&D of super-hydrophilic coating systems is designed and demonstrated. During the hydrophilic polymer research phase, the multi-model leverages the domain-specific insights and chemical tool capabilities of the fine-tuned PolyLLM to systematically screen 852 hydrophilic monomers and generate 3880 hypothetical polymer architectures. By utilizing the PolyML, the multi-model enables precise performance predictions and quantitative evaluations of feature importance to facilitate efficient high-throughput screening of optimal structures from the generated polymer database. Experimental validation confirms that the multi-model achieves a mean absolute error (MAE) below 10% across various polymer property prediction tasks compared to experimental data. This approach leads to the rapid discovery of hydrophilic polymers with exceptional anti-swelling and wear-resistance. In developing super-hydrophilic coatings, the multi-model effectively combines PolyLLM’s guidance for large-scale synthesis with PolyML’s capabilities for precise formulation optimization and curing parameter adjustments. The effectiveness of multi-model is demonstrated by the accelerated development of high-performance anti-fogging coatings for swim goggles and optical films with outstanding water- and wear-resistance that significantly outperform leading commercial products. This multi-model offers a flexible approach for advanced polymer coatings, leading to a seamless connection between laboratory research and industrial development.

查看原文本刊更多论文

超亲水性涂料智能研发的多模型框架

对于高性能超亲水性涂料的研究和开发，传统的试错方法面临着长期存在的挑战，包括数据匮乏、结构-活性关系不明确和设计指南。在本研究中，设计并演示了一个创新的多模型框架，该框架协同集成了用于超亲水性涂层系统智能研发的专用聚合物大语言模型（PolyLLM）和聚合物机器学习模型（PolyML）。在亲水聚合物的研究阶段，多模型利用领域特定的见解和精细调整的PolyLLM的化学工具功能，系统地筛选852个亲水单体，并生成3880个假设的聚合物结构。通过使用PolyML，多模型能够精确的性能预测和特征重要性的定量评估，从而促进从生成的聚合物数据库中高效、高通量地筛选最佳结构。实验验证证实，与实验数据相比，该多模型在各种聚合物性能预测任务中的平均绝对误差（MAE）低于10%。这种方法导致了具有优异的抗膨胀性和耐磨性的亲水聚合物的快速发现。在开发超亲水性涂料方面，多模型有效地将PolyLLM对大规模合成的指导与PolyML精确配方优化和固化参数调整的能力相结合。泳镜和光学薄膜的高性能防雾涂层的加速发展证明了多模型的有效性，这些涂层具有出色的耐水性和耐磨性，显著优于领先的商业产品。这种多模型为先进的聚合物涂层提供了一种灵活的方法，从而实现了实验室研究和工业发展之间的无缝连接。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Science China Materials Materials Science-General Materials Science

CiteScore

11.40

自引率

7.40%

发文量

949

期刊介绍： Science China Materials (SCM) is a globally peer-reviewed journal that covers all facets of materials science. It is supervised by the Chinese Academy of Sciences and co-sponsored by the Chinese Academy of Sciences and the National Natural Science Foundation of China. The journal is jointly published monthly in both printed and electronic forms by Science China Press and Springer. The aim of SCM is to encourage communication of high-quality, innovative research results at the cutting-edge interface of materials science with chemistry, physics, biology, and engineering. It focuses on breakthroughs from around the world and aims to become a world-leading academic journal for materials science.