Energy & Fuels, Vol.33, No.8, 7176-7187, 2019
Comparison of RP-3 Pyrolysis Reactions between Surrogates and 45-Component Model by ReaxFF Molecular Dynamics Simulations
This work compares pyrolysis reactions of 3- and 4-component surrogate models of RP-3 aviation fuel by a ReaxFF molecular dynamics (MD) simulation method. To evaluate the reactivity of the two RP-3 surrogate models, a multi component baseline model that consists of 45 components was constructed as a representative of real RP-3 fuel. Reactive MD simulations of RP-3 fuel pyrolysis were performed for the two simple surrogate models and the multi-component baseline model using the GPU code of GMD-Reax. Reaction pathways were analyzed with aid of the unique software of VARxMD. The main product yield and the initial reaction pathways in heat-up pyrolysis simulations of the two RP-3 surrogate models are found different from those in the 45-component model. In comparison to the 45-component baseline model, the weight fraction of C2H4 generated can be 15% higher for the 4-component surrogate model and 10% higher for the 3-component surrogate model. Because the C2H4 molecules in RP-3 pyrolysis are mainly produced through beta-scission reactions of normal paraffins, the overestimation of the C2H4 yield in heat-up pyrolysis simulations by the two RP-3 surrogate models can be attributed to the much higher normal paraffin of the 4-component surrogate model (86.5%) and 3-component surrogate model (73.0%) over the 45-component model (26.7%). Accordingly, the multiple branched methylalkane radicals generated in ring-opening reactions of 1,3,5-trimethylcyclohexane are responsible for the overestimation of C3H6 production found in simulations of the 3-component surrogate model. The differences of the major product generation between the two surrogate models and the baseline model indicate that both the fuel component composition and molecular structures in RP-3 surrogate models are closely associated with their reactivity differences in pyrolysis. ReaxFF MD simulations can be a useful approach to provide clues for further refinement of compositions and molecular structures of RP-3 surrogate fuel.