Automated High-Throughput Screening of Polymers Using a Computational Workflow

TL;DR

This paper introduces an automated computational workflow for high-throughput screening of polymers, combining automated simulations with machine learning to efficiently predict properties like density and glass transition temperature.

Contribution

It presents a novel automated workflow that minimizes human intervention and computational cost, enabling large-scale polymer property prediction using machine learning.

Findings

Workflow successfully automates polymer screening process

Machine learning models accurately predict polymer density

Method reduces computational resources needed for screening

Abstract

High-throughput computational screening of polymers offers a powerful way to address the imbalance between the vast number of polymers synthesised for diverse applications and the relatively small subset that can be studied using atomistic simulations. This work presents an automatic workflow designed to enable the rapid and efficient screening of an extensive polymer library. The workflow integrates an automated annealing protocol with adaptive control, allowing for reproducible simulations with minimal human intervention and minimisation of the computational cost. The availability of a homogenous large set of simulations enables the adoption of machine learning approaches for a variety of tasks. We exemplify this possibility by proposing rapid machine-learning-based method to predict the (computed) polymer density and (experimental) glass transition temperature.