Kinetics of HMX Decomposition

We first discuss the overall chemical process predicted, followed by a discussion of reaction mechanisms. Under the simulation conditions, the HMX was in a highly reactive dense fluid phase. There are important differences between the dense fluid (supercritical) phase and the solid phase, which is stable at standard conditions. Namely, the dense fluid phase cannot accommodate long-lived voids, bubbles, or other static defects, since it has no surface tension. Instead numerous fluctuations in the local environment occur within a timescale of 10s of femtoseconds. The fast reactivity of the dense fluid phase and the short spatial coherence length make it well suited for molecular dynamics study with a finite system for a limited period of time. Under the simulation conditions chemical reactions occurred within 50 fs. Stable molecular species were formed in less than a picosecond. We report the results of the simulation for up to 55 picoseconds. Figs. 11 (a-d) display the product formation of H2O, N2, CO2 and CO, respectively. The concentration, C(t), is represented by the actual number of product molecules formed at the corresponding time (. Each point on the graphs (open circles) represents a 250 fs averaged interval. The number of the molecules in the simulation was sufficient to capture clear trends in the chemical composition of the species studied. These concentrations were in turn fit to an expression of the form C(/) = C(l- e ), where C is the equilibrium concentration and b is the effective rate constant. From this fit to the data, we estimate effective reaction rates for the formation of H2O, N2, CO2, and CO to be 0.48, 0.08,0.05, and 0.11 ps, respectively. 

It is not surprising that the rate of H2O formation is much faster than that of N2. Fewer reaction steps are required to produce a triatomic species like water, while the formation of N2 involves a much more complicated mechanism. [70] Further, the formation of water (Fig. 11-a) starts aroimd 0.5 ps and seems to have reached a steady state at 10 ps, with oscillatory behavior of decomposition and formation clearly visible. We expect this trend to continue until chemical equilibrium is reached, well beyond the current simulation time. The formation of N2 (Fig. 11-b), on the other hand, starts around 1.5 ps and is still progressing (slope of the graph is slightly positive) after 55 ps of simulation time, albeit at small variation. 

As can be noticed in Fig. 12, the results of our present simulation compare very well with the formation of H2O, N2, and HNCO. The relative concentration of CO and CO2, however, is reversed at the limited time of our simulation. No condensed carbon was found in the current simulation. Several other products and intermediates with lower concentrations, common to the two methods, have also been identified. These include HCN, NH3, N2O, CH3OH, and CH2O. It is hoped that interplay between the two vastly different approaches could be established at much longer simulation time. The goal will be to expand the product molecule set of the theromochemical code with important species determined from our ab nitio based simulations for kinetic modelling. 

Seven CH2N2O2 species have been formed at around 200 fs of simulation time. These results are similar to those identified in thermal decomposition experiments. [51,53] A further N-NO2 bond breaking then follows the decomposition (1) and (2) above. From (1), this leads to the formation of CH2N and NO2. These pathways are remarkably similar to those predicted previously by Melius from the decomposition of nitramines at fast heating rates. [70] 

As the radical CH2N is formed, the production of HCN occurs via the reaction  

---