Development of computational design for reliable prediction of dielectric strengths of perfluorocarbon compounds

Abstract The development of robust computational protocols capable of accurately predicting the dielectric strengths of eco-friendly insulating gas candidates is crucial; however, it lacks relevant efforts significantly. Consequently, a series of computational protocols are employed in this study to enable the computational prediction of polarizability and ionization energy of eco-friendly, perfluorinated carbon-based candidates, followed by the equation-based prediction of their dielectric strength. The validation process associated with the prediction of the afore-mentioned variables for selected datasets confirms the suitability of the B3LYP-based prediction protocol for reproducing experimental values. Subsequently, the validation of dielectric strength prediction outlines the following three conclusions. (1) The referenced equation adopted from a previous study is incapable of predicting the dielectric strengths of 137 organic compounds present in our database. (2) Parameterization of the coefficients in the referenced equation leads to the accurate prediction of the dielectric strengths. (3) Incorporation of a novel variable, viz. molecular weight, into the referenced equation combined with the parameterization of the coefficients leads to a robust protocol capable of predicting dielectric strengths with high efficiencies even with a significantly smaller fitting dataset. This implies the development of a comprehensive solution capable of accurately predicting the dielectric strengths of a substantially large dataset.