Which molecular properties determine the impact sensitivity of an explosive? A machine learning quantitative investigation of nitroaromatic explosives

We decomposed density functional theory charge densities of 53 nitroaromatic molecules into atom-centered electric multipoles using the distributed multipole analysis that provides a detailed picture of the molecular electronic structure. Three electric multipoles, ∑▒〖Q_0 (NO_2)〗 (the charge of the nitro groups), ∑▒〖Q_1 (NO_2)〗 (the total dipole, i.e., polarization, of the nitro groups), ∑▒〖Q_2 (C) 〗 (the total electron delocalization of the C ring atoms), and the number of explosophore groups (#NO_2) were selected as features for a comprehensive machine learning (ML) investigation. The target property was the impact sensitivity h_50 (cm) values quantified by drop-weight measurements. After a preliminary screening of 42 ML algorithms, four were selected based on the lowest root mean square errors: Extra Trees, Random Forests, Gradient Boosting, and AdaBoost. The predicted h_50 values of molecules having very different sensitivities for the four algorithms are in the range 19% - 28% compared to experimental data. The most important properties for predicting h_50 are the electron delocalization in the ring atoms and the polarization of the nitro groups with averaged weights of 39% and 35%, followed by the charge (16%) and number (10%) of nitro groups. A significant result is how the contribution of these properties to h_50 depends on its sensitivities: for the most sensitive explosives (h_50 up to ~ 50 cm), the four properties contribute to reducing h_50, and for intermediate ones (~ 50 cm ≲ h_50 ≲ 100 cm) #NO_2 and ∑▒〖Q_1 (NO_2)〗 contribute to increasing it and the other two properties to reducing it. For highly insensitive explosives (h_50≳ 200 cm), all four properties essentially contribute to increasing it. These results furnish a consistent molecular basis of the sensitivities of known explosives that also can be used for developing safer new ones.