Supplementary Data: Insights into the Defluorination of PFAS from Quantum Chemical Calculations and Machine Learning

This dataset accompanies the thesis titled "Insights into the Defluorination of PFAS from Quantum Chemical Calculations and Machine Learning." The thesis examines the use of computational chemistry methods and machine learning to accurately predict the C-F bond dissociation energies of per- and polyfluoroalkyl substances (PFAS). This dataset includes all optimised molecular structures and machine learning input data that were used to generate the figures, graphs, and quantitative analyses presented in the thesis. Specifically, optimised molecular structures are presented as XYZ coordinates obtained from quantum chemical calculations, and machine learning input data are presented as comma-separated values of numerical descriptors, string-representation of PFAS, and C-F bond dissociation energies. Files are separated based on their relevance to the discussion in each chapter of the thesis. Data contained in ch5_energies.tgz is reprinted with permission from J. Phys. Chem. A 2023, 127, 38, 7943–7953. Copyright 2023 American Chemical Society. Data provided in the remaining files are first published here.