Thermal decomposition and shock response mechanism of DNTF: Deep potential molecular dynamics simulations

Abstract

A neural network deep potential (DP) is developed for the shock response and thermal decomposition mechanisms of 3,4-Bis(3-nitrofurazan-4-yl)furoxan (DNTF). This DP potential, trained on ab initio datasets, achieves the accuracy of density functional theory (DFT) and higher computational efficiency. The simulation results show that DNTF is anisotropic under shock loading. And the critical shock decomposition temperatures along the [100], [010], and [001] directions are 463.64 K, 451.85 K, and 486.69 K, respectively. For the first time, the low-temperature (<750 K) decomposition of DNTF is successfully simulated using molecular dynamics combined with DP model. The decomposition of DNTF begins with the O-N bond opening of the furoxan ring, followed by the breaking of the C-NO₂ bond and opening of the furazan rings, under thermal conditions. The critical thermal decomposition is between 458.3 K and 562.5 K at a heating rate of 13.5 K/ps. The final products are CO₂, N₂, and CO, and the main intermediates are NO₂, NO, and N₂O. The activation energy of decomposition is 142.4 kJ/mol obtained by Ozawa method. This study not only provides a powerful tool for investigating the performance of DNTF but also offers a feasible approach for other energetic materials, advancing the field significantly.