We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Feasibility of Utilizing Machine Learning to Identify a More Sustainable Alternative to Polyester in Textiles
Summary
Researchers used machine learning trained on the PI1M polymer dataset of over one million compounds to identify sustainable alternatives to polyester in textiles, predicting key material properties including glass transition temperature, density, melting temperature, oxygen permeability, and bulk modulus. The study found that artificial neural networks could screen polymer candidates for environmental compatibility while maintaining performance characteristics comparable to polyester.
The mass production of textiles has created several concerns for the environment and its future, with the increase of greenhouse gases that are released during the production of Polyester.Machine learning is a recent technology commonly used in polymer science to consider vast amounts of data, relatively quickly.The aim of the project is to find an alternative polymer that possesses similar characteristics to polyester with a considerably lower environmental impact.In this study, we used a publicly available dataset, PI1M, containing information on over one million polymers to train artificial neural net.We looked into predicting key features for these polymers, including glass transition temperature, density, melting temperature, oxygen permeability, compressibility, and bulk modulus.Through the process of machine learning modeling, several properties of new polymers were successfully predicted.An agreement was found between the true and predicted values for each of these features.The results show the presence of a possibly more sustainable alternative to Polyester in textiles.This study demonstrates the feasibility of using machine learning to discover new, more sustainable polymers.