Extensive density functional theory (DFT) calculation and data analysis on molecular and crystal level features of 60 reported energetic materials (EMs) allowed us to define key descriptors that are characteristics of these compounds’ thermostability. We see these descriptors as reminiscent of “Lipinski’s rule of 5”, which revolutionized the design of new orally active pharmaceutical molecules. The proposed descriptors for thermostable EMs are of a type of molecular design, location and type of the weakest bond in the energetic molecule, as well as specific ranges of oxygen balance, crystal packing coefficient, Hirshfeld surface hydrogen bonding, and crystal lattice energy. On this basis, we designed three new thermostable EMs containing bridged, 3,5-dinitropyrazole moieties, HL3, HL7, and HL9, which were synthesized, characterized, and evaluated in small-scale field detonation experiments. The best overall performing compound HL7 exhibited an onset decomposition temperature of 341 °C and has a density of 1.865 g cm^–3, and the calculated velocity of detonation and maximum detonation pressure were 8517 m s^–1 and 30.6 GPa, respectively. Considering HL7’s impressive safety parameters [impact sensitivity (IS) = 22 J; friction sensitivity (FS) = 352; and electrostatic discharge sensitivity (ESD) = 1.05 J] and the results of small-scale field detonation experiments, the proposed guidelines should further promote the rational design of novel thermostable EMs, suitable for deep well drilling, space exploration, and other high-value defense and civil applications.

Combined density functional theory and empirical-potential calculations are performed to investigate the lowest-energy sites and migration mechanisms of He in various $$\alpha${}$-Fe grain boundaries (GBs). Before the defect calculations, we show that structural optimizations, including simulated annealing and atom removal, are crucial for locating the stable GB structure in a given temperature regime. Then, the He formation energies for all the substitutional and interstitial sites in two different GBs are evaluated, showing a strong He segregation tendency. At variance with the bulk Fe case, the formation energy of an interstitial He is either lower than or similar to that of a substitutional He in the GBs. Finally, both static and dynamic barriers for interstitial He diffusion in the GBs are determined. Although the diffusion details and precise paths are GB dependent, some common features are identified: (1) The He atom always remains confined to the GB region while diffusing; (2) the He diffusion is highly anisotropic along the GBs; (3) the GB diffusion of an interstitial He atom is found to be always slower than its bulk diffusion, but it can still be faster than the bulk diffusion of a substitutional He.

The machine learning-driven approach and sequential construction strategy enabled the design of 168 nitrogen–oxygen compounds. High-throughput calculations identified 106 stable candidates, some rivaling N ₂ O ₄ , with potential for space exploration.

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.