AI-Guided Discovery of new CHIPs Comonomers

Creating easily processable and recyclable infrared (IR) optical materials is needed to make a myriad of applications, including optics and sensing sustainable. Currently, inorganic solid-state materials are predominately used for IR-optical applications, leading to high manufacturing and processing costs. Chalcogenide Hybrid Inorganic/Organic Polymers (CHIPs) have the potential to revolutionize the industry by combining the optical properties of solids, with the processability of plastics. CHIPs is a class of materials that copolymerize organic molecules with elemental sulfur to create high-sulfur content polymers with an index of refraction on the order of 1.6-2.1, and a high IR transmission. Here we present a gradient boosted tree model to predict the IR-optical properties of the comonomers and determines which ones merit further consideration as high-performing CHIPs materials. After training models on previously calculated IR-absorption data we apply them to a larger set 960,966 molecules from the GDB dataset and validate the predictions on a subset of the extrapolation set. We then focus on a subset of these molecules that are CHIPs-capable and find 2,942 possible comonomers we predict to have better optical properties than a recently discovered promising comonomer Stillene. Finally, we calculate optical properties of all 2,942 comonomers in the gas phase and in a configuration to approximate the polymer films to find a new set of target comonomers.