A novel simple correlation for predicting glass transition temperature of energetic azido-ester plasticizers through molecular structures

This study presents a new reliable correlation for the prediction glass transition temperature of energetic azido-ester plasticizers using their structural parameters. The methodology assumes that the glass transition temperature of azido-ester plasticizers can be expressed as a function of several structural parameters through multiple linear regression method. The predictive ability of the correlation is checked using a cross-validation method (R2 = 0.958, QLOO2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Q_{\text{LOO}}^{2}$$\end{document} = 0.964 and QLMO2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Q_{\text{LMO}}^{2}$$\end{document} = 0.968). The proposed correlation has the root-mean-square deviation and the average absolute deviations of 3.97 and 3.16 °C, respectively, for 21 azido-ester plasticizers with different molecular structures as training set. Also, this correlation gives good predictions for further six azido-ester plasticizers as test set. The proposed method can also apply for designing novel energetic azido-ester plasticizers.