Background/Objectives
A necessary transformation for a sustainable economy is the transition from fossil-derived plastics to polymers derived from biomass or waste resources. While renewable feedstocks can enhance material performance through unique chemical moieties, probing the vast material design space by experiment alone is not practically feasible. In addition, balancing performance across multiple material properties remains a challenge in polymer discovery and redesign.
Approach/Activities
We developed a machine learning-based tool called PolyID to reduce the design space of renewable feedstocks and to enable efficient discovery of performance-advantaged, bio-based polymers. PolyID is a multi-output graph neural network specifically designed to increase accuracy and to enable quantitative structure-property relationship (QSPR) analysis for polymers. It includes a novel domain-of-validity method to demonstrate how gaps in training data can be filled to improve accuracy.
Results/Lessons Learned
We used PolyID to identify poly(ethylene terephthalate) (PET) analogs with predicted improvements to thermal and transport performance and found five suitable candidates from among 1.4x106 accessible bio-based polymers. Experimental validation for one of the PET analogs demonstrated a glass transition temperature between 85 °C and 112 °C, which is higher than PET and within the predicted range from the PolyID tool. Overall, PolyID can aid the bio-based polymer practitioner to navigate the vast number of renewable polymers to discover sustainable materials with enhanced performance. We are actively working with several companies such as Kraft Heinz to redesign their products to be more sustainable.
