Creation of Polymer Datasets with Targeted Backbones for Screening of High-Performance Membranes for Gas Separation

• DOI: 10.1021/acs.jcim.3c01232
• Publication Date: 2024-01-31T16:26:20Z
• Authors: Surya Prakash Tiwari; Wei Shi; Samir Budhathoki; James Baker; Ali K. Sekizkardes; Lingxiang Zhu; Victor A. Kusuma; David P. Hopkinson; Janice A. Steckel
• Keywords: Polydimethylsiloxane; Peptoid

A simple approach was developed to computationally construct a polymer dataset by combining simplified molecular-input line-entry system (SMILES) strings of targeted backbone and variety molecular fragments. This method used create 14 datasets seven backbones molecules from two large (MOSES QM9). Polymer that were studied include four polydimethylsiloxane (PDMS) based backbones, poly(ethylene oxide) (PEO), poly(allyl glycidyl ether) (PAGE), polyphosphazene (PPZ). The generated can be for various cheminformatics tasks, including high-throughput screening gas permeability selectivity. study utilized machine learning (ML) models screen the polymers CO2/CH4 CO2/N2 separation using membranes. Several interest identified. results highlight employing an ML model fitted selectivities leads higher accuracy in predicting selectivity compared ratio predicted permeabilities.