A machine learning platform for polymer flammability prediction

IF 6.3 2区化学 Q1 POLYMER SCIENCE

Polymer Degradation and Stability Pub Date : 2025-05-21 DOI:10.1016/j.polymdegradstab.2025.111411

Duy Nhat Phan , Alexander B. Morgan , Lokendra Poudel , Rahul Bhowmik

{"title":"A machine learning platform for polymer flammability prediction","authors":"Duy Nhat Phan , Alexander B. Morgan , Lokendra Poudel , Rahul Bhowmik","doi":"10.1016/j.polymdegradstab.2025.111411","DOIUrl":null,"url":null,"abstract":"<div><div>Flammability index (FI) and cone calorimetry outcomes, such as maximum heat release rate, time to ignition, total smoke release, and fire growth rate, are critical factors in evaluating the fire safety of polymers. However, predicting these properties is challenging due to the complexity of material behavior under heat exposure. In this work, we investigate the use of machine learning (ML) techniques to predict these flammability metrics. We generated synthetic polymers using Synthetic Data Vault to augment the experimental dataset. Our comprehensive ML investigation employed both our polymer descriptors and those generated by the RDkit library. Despite the challenges of limited experimental data, our models demonstrate the potential to accurately predict FI and cone calorimetry outcomes, which could be instrumental in designing safer polymers. Additionally, we developed POLYCOMPRED, a module integrated into the cloud-based MatVerse platform, providing an accessible, web-based interface for flammability prediction. This work provides not only the predictive modeling of polymer flammability but also an interactive analysis tool for the discovery and design of new materials with tailored fire-resistant properties.</div></div>","PeriodicalId":406,"journal":{"name":"Polymer Degradation and Stability","volume":"240 ","pages":"Article 111411"},"PeriodicalIF":6.3000,"publicationDate":"2025-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Polymer Degradation and Stability","FirstCategoryId":"92","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S014139102500240X","RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"POLYMER SCIENCE","Score":null,"Total":0}

引用次数: 0

Abstract

Flammability index (FI) and cone calorimetry outcomes, such as maximum heat release rate, time to ignition, total smoke release, and fire growth rate, are critical factors in evaluating the fire safety of polymers. However, predicting these properties is challenging due to the complexity of material behavior under heat exposure. In this work, we investigate the use of machine learning (ML) techniques to predict these flammability metrics. We generated synthetic polymers using Synthetic Data Vault to augment the experimental dataset. Our comprehensive ML investigation employed both our polymer descriptors and those generated by the RDkit library. Despite the challenges of limited experimental data, our models demonstrate the potential to accurately predict FI and cone calorimetry outcomes, which could be instrumental in designing safer polymers. Additionally, we developed POLYCOMPRED, a module integrated into the cloud-based MatVerse platform, providing an accessible, web-based interface for flammability prediction. This work provides not only the predictive modeling of polymer flammability but also an interactive analysis tool for the discovery and design of new materials with tailored fire-resistant properties.

查看原文本刊更多论文

聚合物易燃性预测的机器学习平台

可燃性指数（FI）和锥量热法结果，如最大放热率、点火时间、总烟释放量和火焰生长速率，是评估聚合物防火安全性的关键因素。然而，由于材料在热暴露下的行为的复杂性，预测这些特性是具有挑战性的。在这项工作中，我们研究了使用机器学习（ML）技术来预测这些可燃性指标。我们使用合成数据库生成合成聚合物来增强实验数据集。我们全面的ML调查使用了我们的聚合物描述符和RDkit库生成的描述符。尽管实验数据有限，但我们的模型证明了准确预测FI和锥体量热结果的潜力，这可能有助于设计更安全的聚合物。此外，我们还开发了polycomppred，这是一个集成到基于云的MatVerse平台中的模块，为可燃性预测提供了一个可访问的基于web的界面。这项工作不仅提供了聚合物可燃性的预测建模，而且为发现和设计具有定制耐火性能的新材料提供了交互式分析工具。

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

Polymer Degradation and Stability 化学-高分子科学

CiteScore

10.10

自引率

10.20%

发文量

325

审稿时长

23 days

期刊介绍： Polymer Degradation and Stability deals with the degradation reactions and their control which are a major preoccupation of practitioners of the many and diverse aspects of modern polymer technology. Deteriorative reactions occur during processing, when polymers are subjected to heat, oxygen and mechanical stress, and during the useful life of the materials when oxygen and sunlight are the most important degradative agencies. In more specialised applications, degradation may be induced by high energy radiation, ozone, atmospheric pollutants, mechanical stress, biological action, hydrolysis and many other influences. The mechanisms of these reactions and stabilisation processes must be understood if the technology and application of polymers are to continue to advance. The reporting of investigations of this kind is therefore a major function of this journal. However there are also new developments in polymer technology in which degradation processes find positive applications. For example, photodegradable plastics are now available, the recycling of polymeric products will become increasingly important, degradation and combustion studies are involved in the definition of the fire hazards which are associated with polymeric materials and the microelectronics industry is vitally dependent upon polymer degradation in the manufacture of its circuitry. Polymer properties may also be improved by processes like curing and grafting, the chemistry of which can be closely related to that which causes physical deterioration in other circumstances.