Quantitative Structure-Property Relationship Modeling to Predict the Density of Energetic Materials Based on the Monte Carlo Optimization Method

The most recent generation of energetic materials is becoming defined by explosive properties, including increased energy, density, and insensitivity, as the concept of green chemistry keeps developing. Consequently, highly accurate molecular design and green and efficient synthesis of energetic materials will be among the most serious challenges. The structure-property relationship (QSPR) is suggested for accurately predicting the detonation performance of energetic materials. This relationship is competently employed to estimate the density index of energetic materials. The Simplified Molecular Input-Line Entry System (SMILES) is employed to illustrate the molecular structure of the compounds. Based on the Monte Carlo method, QSAR modeling is accomplished in the CORAL software via the balance of correlation method. The statistical validity of QSAR models is improved by the index of ideality of correlation (IIC). The SMILES attributes identified the most optimal descriptor for constructing the QSAR models. In this study, ۱۶۲ energetic compound molecules in the data set were arbitrarily divided into three categories. To further investigate, the performance of the three models shows that split one can achieve a better prediction result for the density index of energetic material with an R۲ value of ۰.۸۰. The CRp۲> ۰.۵ for all three constructed models in the Y-randomization test determines the reliability of the created models.

Keywords:

QSAR models , SMILES , CORAL software , energetic materials , density index