The long-standing performance-stability contradiction issue of high energy density materials (HEDMs) is of extremely complex and multi-parameter nature. Herein, machine learning was employed to handle 28 feature descriptors and 5 properties of detonation and stability of 153 HEDMs, wherein all 21,648 data used were obtained through high-throughput crystal-level quantum mechanics calculations on supercomputers. Among five models, namely, extreme gradient boosting regression tree (XGBoost), adaptive boosting, random forest, multi-layer perceptron, and kernel ridge regression, were respectively trained and evaluated by stratified sampling and 5-fold cross-validation method. Among them, XGBoost model produced the best scoring metrics in predicting the detonation velocity, detonation pressure, heat of explosion, decomposition temperature, and lattice energy of HEDMs, and XGBoost predictions agreed best with the 1,383 experimental data collected from massive literatures. Feature importance analysis was conducted to obtain data-driven insight into the causality of the performance-stability contradiction and delivered the optimal range of key features for more efficient rational design of advanced HEDMs.

ReaxFF-MPNN, a reactive force field model that combines the high accuracy of machine learning and the high efficiency of the classical force field, is reported.

The orientation dependence of shock sensitivity in high explosive crystals was explored in this study. As a widely used wood explosive, 1,3,5-tri-amino-2,4,6-tri-nitrobenzene (TATB) is insensitive to thermal ignition and mechanical impact. Its typical anisotropic crystal structure suggests anisotropic shock sensitivity. Shockwaves were applied to an incised TATB crystal along three orthogonal directions using the multiscale shock technique (MSST) combined with the ReaxFF method to study the origin of anisotropic shock sensitivity. The physical and chemical responses of the TATB crystal during shock were investigated. The results show that the temperature, stress, volume compressibility, and decomposition rate of TATB are strongly dependent on the shockwave direction. In other words, the sensitivity of TATB to mechanical shock is strongly dependent on the crystal orientation. TATB is relatively sensitive along the directions parallel to the (001) crystal plane (X and Y directions) and is highly insensitive along the [001] direction (Z direction). We calculated the energy of intermolecular hydrogen bonds and the elastic constants of the TATB crystal using ab initio simulations, which also show anisotropy. We found that the unique structure of intermolecular hydrogen bonds and the difference in temperature rise induced by orientation-related compressibility are primarily responsible for the anisotropic shock wave sensitivity.

The thermal sensitivity difference between TATB and BTF is interpreted from pyrolysis response, initial decomposition mechanism, and clusters formation.