Machine learning for polymeric materials: an introduction

Abstract

Polymers are incredibly versatile materials and have become ubiquitous. Increasingly, researchers are using data science and polymer informatics to design new materials and understand their structure–property relationships. Polymer informatics is an emerging field. While there are many useful tools and databases available, many are not widely utilized. Herein, we introduce the field of polymer informatics and discuss some of the available databases and tools. We cover how to share polymer data, approaches for preparing a dataset for machine learning and recent applications of machine learning to polymer property prediction and polymer synthesis. © 2021 Society of Industrial Chemistry.

Data‐Driven Materials Innovation and Applications

“…Key to the efficient use of ML in the field of chemical materials is the “descriptor selection” tool, which takes the entire descriptor set as an input, or combines it into a new reduced, but more reliable, descriptor set through correlation analysis while providing a mapping to a key performance indicator (KPI) fingerprint. [ 50 ] In this section, the strategy of transforming material data to ML through descriptors is introduced; descriptors can be divided into five main types: constitutional descriptors; [ …”

“…Comprehensive reviews have detailed the applicability of datadriven approaches to energy materials [9,30,38] structural materials, [39] polymeric materials, [40] and porous materials, [32] with the help of high-throughput approaches such as density functional theory (DFT) and ML. [5] The applications of ML in synthetic chemistry [41] and the prediction of material properties [29] have also been published.…”

A Machine Learning Method for Material Property Prediction: Example Polymer Compatibility

“…The properties of polymer materials are closely related to the multidimensional factors in the process of polymer synthesis and processing, which makes it complicated and challengeable to give accurate prediction [18,19,20,21]. Researchers have made active and effective attempts to apply ML method in exploring polymer syntheses and polymer materials [22]. The copolymer synthesis and defectivity [23,24,25], mechanical properties of polymer composites [26], liquid crystal behavior of copolyether [27], thermal conductivity [28], dielectric properties [29], glass transition, melting, and degradation temperature and quantum physical and chemical properties [30,31,32,33] have been applied with machine learning and good prediction accuracy is achieved.…”

Nurturing inclusivity: removing obstacles and valuing perspectives

“…Machine learning tools are relatively new to most chemists yet the expectation that these tools will be placed adjacent to legacy polymer characterization tools is an exciting addition to accelerate discovery and innovation. Assary, Moore, and Cencer provide a starting point for understanding the implications of machine learning on many research activities 13 . Attention should also be directed to Halden and coworkers, who continue the critical discussion of microplastics and nanoplastics in the ocean, and this review highlights the methods and challenges for detecting these contaminants in our environment 14 .…”

“…Assary, Moore, and Cencer provide a starting point for understanding the implications of machine learning on many research activities. 13 Attention should also be directed to Halden and coworkers, who continue the critical discussion of microplastics and nanoplastics in the ocean, and this review highlights the methods and challenges for detecting these contaminants in our environment. 14 These techniques strive to reduce the source of microplastics, improve detection methods, and ultimately lead to understanding their impact and opportunities for removal.…”