Design of Recyclable\nPlastics with Machine Learning\nand Genetic Algorithm

We present an artificial intelligence-guided approach\nto design\ndurable and chemically recyclable ring-opening polymerization (ROP)\nclass polymers. This approach employs a genetic algorithm (GA) that\ndesigns new monomers and then utilizes virtual forward synthesis (VFS)\nto generate almost a million ROP polymers. Machine learning models\nto predict thermal, thermodynamic, and mechanical propertiescrucial\nfor application-specific performance and recyclabilityare\nused to guide the GA toward optimal polymers. We present potential\nsubstitute polymers for polystyrene (PS) that achieve all property\ntargets with low estimated synthetic complexity.