Machine learning-assisted development of conductive polymers

Abstract

Machine learning (ML) techniques are increasingly being used to predict and enhance the performance of new materials, including conductive polymers, which are valued for their unique electrical properties. These materials are crucial for a range of applications, such as electronics, energy storage, and sensors. This paper provides a comprehensive review of the properties and applications of major types of conductive polymers, including intrinsic, doped, and nanocomposite-based systems. The concept of “Face IDs” is introduced as an analogy for the key chemical features and properties of conductive polymers, helping to translate complex chemical structures, fabrication parameters, and performance indicators into machine-readable descriptors. This approach bridges experimental polymer science with advanced data-driven methodologies. Additionally, the paper explores the current progress of ML-assisted design in advancing conductive polymers, with a focus on optimizing properties such as electrical conductivity, mechanical strength, and thermal stability. However, challenges persist in applying ML for the development of new conductive polymers with desired properties, such as the limited availability of high-quality datasets, the complexity of polymer structures, and the need for better models for reverse design. This review aims to facilitate collaboration between researchers in the fields of polymer science and ML, highlighting the potential of interdisciplinary efforts to drive innovation in the development of next-generation conductive polymers.