Designing degradable Polyethene with desirable physical properties via molecular dynamics simulation and machine learning

Abstract

Degradable materials has been a huge concern in the world due to poor recycling rates and pollution which contributes to the poor environmental conditions of our world today. Thus, the importance of degradable materials is essential to build a better world for our future. One of the commonly used materials which consumers use day to day would be Polyethene (PE).

The huge consumption combined with the poor recycling rate of PE materials in recent years has affected the current world environmental conditions negatively. Thus, it is crucial to design degradable PE materials with desirable physical properties. Machine learning techniques have shown great potential in accelerating the discovery of new materials and helping us build structure-properties relationships. Traditionally, the discovery of new materials and the exploration of their properties can be a timeconsuming and expensive process that involves a lot of trial and error experimentation. However, with machine learning, we can leverage large datasets of materials properties and use algorithms to identify patterns and make predictions about the behavior of new materials.

In this project, we have utilized Machine Learning (ML) techniques to evaluate the Glass Transition Temperature (Tg) as the Tg is well known to be one of the most important factors of a polymer as it helps to determine target physical properties of the specific polymers. We have also explored the field pf polymer informatics by utilizing ML techniques to evaluate Tg and other properties of polymers.