Application of machine learning in developing a quantitative structure–property relationship model for predicting the thermal decomposition temperature of nitrogen-rich energetic ionic salts

Yunling Zhang,Liang Fan,Chao Su,Zhenyu Shu,Haijie Zhang
DOI: https://doi.org/10.1039/d4ra05875e
IF: 4.036
2024-11-28
RSC Advances
Abstract:While thermal decomposition temperature ( T d ) is one of the most important indexes for energetic materials, the most common way of determining and evaluating T d requires laboratory experiments that are complicated, time-consuming and expensive. In the present study, the quantitative structure–property relationship (QSPR) model of T d for 21 nitrogen-rich energetic ionic salts was built and used for T d prediction through 13 descriptors and principal component analysis. The relatively small dataset of 21 samples may lead to overfitting. In the case of small datasets, possible overfitting was reduced by the support vector machine to derive the non-linear QSPR model. The obtained correlation coefficient ( R 2 ) of 96.31% and root-mean-square error (RMSE) of 15.72 indicate the relative reliability of the QSPR model developed in this work. Moreover, T d values of 6 newly designed nitrogen-rich energetic ionic salts were predicted using the new QSPR model. The predicted T d values range from 194 to 225 °C, which are better than that of 4-amino-3,5-dinitro-1 H -pyrazole (LLM-116: 178 °C), and no. 1, 2 and 5 are comparable to that of the traditional explosive 1,3,5-trinitro-1,3,5-triazinane (RDX: 230 °C), indicating the excellent properties of the designed energetic ionic salts, which can be used for the preparation of potential energetic materials.
chemistry, multidisciplinary
What problem does this paper attempt to address?