Reactive molecular dynamics simulations on thermal decomposition of 3-methyl-2,6-dinitrophenol

In this paper, we simulated the decomposition mechanism of 3-methyl-2,6-dinitrophenol (MDNP) based on reaction molecular dynamics using ReaxFF force field. In addition, the evolution of some main products over time at different heating rates (10, 15, and 20 K·ps−1) was studied. As indicated by the simulation results, with the elevation at different heating rates, the time required for the system to reach equilibrium was shortened, and more products were obtained. At three heating rates, C7H7O5N2, C7H6O4N2, C7H5O5N2, C7H5O4N2, HON, NO, and NO2 were the main intermediate products, and N2, CO2, H2O, H2, and NH3 were the primary final products. To be specific, C7H5O5N2 was the first produced intermediate product, and H2O was the first produced final product with the largest abundance. The intermediate products first increased and then decreased to zero. Moreover, the primary chemistry reactions in the MDNP pyrolysis were simulated through ReaxFF MD simulations.

On the battlefield, ammunitions are susceptible to unforeseen exterior stimulating factors, which may cause catastrophes when blasting materials are being transported, stockpiled, and used. Heat stimuli are one of the most common exterior stimulating factors, which may lead to the thermolysis of blasting materials and further cause explosion. In accordance with the thermolysis causal link, the thermal risks of MDNP can be reduced during usage. Accordingly, to ensure the safeness of MDNP during usage, it is urgent to study the thermolysis features and causal links of MDNP. In general, experiment data are limited to the information collected from TGA or TGA/MS, which can merely offer dynamic rate constants of the general mass loss and yield of the eventual gas products, and information on intermediates has not been available thus far.
ReaxFF force field refers to a brilliant technology for investigating the thermolysis features and causal links of materials [2], which can effectively compensate for the experimental flaws. Over the past few years, the ReaxFF force field has become a primary technology for studying the thermolysis of energy materials through reactive force fields [3][4][5][6][7][8]. The excellent instances above significantly contribute to the computation of other energy materials [9][10][11][12][13].
In this paper, ReaxFF force field was used to analyze the primary middle products, eventual products, and chemistry reaction processes at 10 K·ps −1 , 15 K·ps −1 , and 20 K·ps −1 , respectively. The primary middle products, eventual products, and chemistry reaction processes were obtained. Based on the simulated results, the thermolysis causal link of MDNP was determined.

Simulation methodologies
ReaxFF MD simulations were used to study the thermolysis of MDNP through ReaxFF-lg force field. ReaxFF, proposed by van Duin [14], has a nearly ab initio level of depiction of the surface for multiparticle systems. During ReaxFF MD simulations, the charge and bond orders were computed after the respective step. On that basis, the forming and breaking of covalent linkages were found. During ReaxFF simulations, the entire atoms were recognized as individual interactive centers, and the transient dot electric charge on every atom was found through the static electric field due to the entire charges around. Chaban et al. [15][16][17][18] elucidated the methods of ReaxFF MD, and these methods have been successfully used to address considerable intricate challenges. Accordingly, this paper used ReaxFF MD to describe the forces on the atoms within the C/H/O/N system.

Temperature-dependent decomposition simulations
In this paper, the LAMMPS was used to carry out the ReaxFF MD simulations of MDNP (  chemistry reaction processes to analyze the thermolysis path of MDNP. During ReaxFF MD simulations, the charge and bond orders were computed after the respective step. Thus, a comparatively small integral time should be used to efficiently cover the phase space that could enable the smooth occurrence of collision and reactions. During such simulation, every ReaxFF MD simulation had 0.1-fs timestep, and the kinetic path involving atom locations and speeds was documented per 200 fs. The data above were used to study the molecule species to deepen the understanding of chemistry reaction processes.

Evolution of potential energy (PE) and total number of MDNP
The PE development of MDNP can be exploited to verify if the chemistry reactions reach balance, and the sum of MDNP can be used to assess the thermolysis level of MDNP. Accordingly, the PE of MDNP was analyzed after simulations. The PE development of MDNP varied over time at different heating rates (Fig. 3). Fig. 3 presents the development of total PE over time at different heating rates. MDNP gradually absorbed heat to reach primary decomposition energy at different heating rates. With the increase in the temperature, MDNP was decomposed. Then, a series of middle and eventual products were generated, and considerable heat was released, so the potential energy decreased significantly. Moreover, if the temperature increased, the maximal results of PE and the PE reduction rate would be higher. With the increase in the temperature, the higher the stability of PE, the shorter the time for PE to stabilize would be. As indicated by the curves of MDNP, at 10 K·ps −1 , 15 K·ps −1 , and 20 K·ps −1 , MDNP molecules were converted into intermediate products through decomposition in 178.02 ps, 110.60 ps, and 100.00 ps, respectively, which demonstrated that faster heating rates could lead to more rapid thermolysis of MDNP. The time evolution of consumption of MDNP at different temperatures is presented in Fig. 4. According to Fig. 3 and Fig. 4, as MDNP continued to be consumed, PE increase further. When the MDNP reaction was nearly complete, PE began to decrease gradually.   demonstrated that the production of N 2 , CO 2 , H 2 O, H 2 , and NH 3 molecules was ongoing during the entire decomposition process.

Evolution of final products
The entire eventual products were small molecule products (e.g., N 2 , CO 2 , H 2 O, H 2 , and NH 3 ). Tables 4,5 and 6 list the generation times of the main eventual products produced by MDNP thermolysis. Fig. 6

Conclusion
In this paper, we used ReaxFF MD simulations to explore the thermolysis of MDNP. The primary middle products consisted of C 7 H 7 O 5 N 2 , C 7 H 6 O 4 N 2 , C 7 H 5 O 5 N 2 , C 7 H 5 O 4 N 2 , HON, NO, and NO 2 . In addition, N 2 , CO 2 , H 2 O, H 2 , and NH 3 were the primary eventual products. The occurrence of the middle products had a temporal order of C 7 H 7 O 5 N 2 , C 7 H 5 O 5 N 2 , C 7 H 5 O 4 N 2 , C 7 H 6 O 4 N 2 , NO 2 , NO, HON. The occurrence of the final products followed a temporal order of H 2 O, H 2 , N 2 , NH 3 , CO 2 . As revealed by the results above, ReaxFF can deepen the understanding of the intricate reactions in MDNP thermal decomposition, and it can be an attractive approach to clarify the chemistry reaction processes in the thermal decomposition of MDNP.

Data availability and material
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
Author contribution Jiaxiang Zhao performed the data analyses and wrote the manuscript; Yun Xiao contributed to the conception of the study; Jiayuan He performed the experiment; Jianlong Wang helped perform the analysis with constructive discussions.
Code availability In this study, the large-scale atomic/molecular massively parallel simulator (LAMMPS.2017) was adopted.