Designing Formulations of Bio-Based, Multi-Component Epoxy Resin Systems via Machine Learning

Petroleum-based epoxy resins are commonly used as matrix in fiber reinforced polymer composites. Bio-based epoxy resin systems might be a more environmentally friendly alternative to conventional epoxy resins. In this work, novel formulations of multi-component, amino acid-based resin systems exhibiting high or low glass transition temperatures (Tg ) were designed via Bayesian optimization and active learning techniques. After only five high-Tg experiments, thermosets with Tg already higher than those of the individual components were obtained, suggesting the existence of synergistic effects among the amino acids used and confirming the efficiency of the theoretical design. Linear and non-linear Machine Learning (ML) models successfully predicted Tg with a mean absolute error of 3.98 oC and R2 score of 0.91. A price reduction of up to 13.7 % was achieved while maintaining the Tg of 130 oC by using an optimized formulation. The LASSO model provided information about the dependence of Tg on the number of active hydrogen atoms and aromaticity. This study highlights the importance of<br/>Bayesian optimization and ML to achieve a more sustainable development of epoxy resin materials.