Large Eddy Simulation of High Reynolds Number Non-Reacting and Reacting JP8 Sprays With a Kerosene Surrogate and Detailed Chemistry

High-resolution single-plume JP-8 spray simulations have been performed to characterize detailed mixture formation process of high-pressure sprays for several common rail fuel injectors of interest to the Army. The first phase of the study involves examining the spray-induced turbulent mixing and global penetration parameters to present experimentally validated results across several computationally challenging length scales. Statistical convergence effects on the spray behavior and penetration profiles are presented by conducting several realizations for each injection case study. The second phase of the project adopts the grid-criteria approach developed for evaporating conditions to model turbulent combustion of a JP-8 reacting spray at compression-ignition engine conditions. A coupled Eulerian Lagrangian formulation is used to model the ensuing spray primary and secondary atomization regions using classical Kelvin Helmholtz - Rayleigh Taylor (KH-RT) wave type models. The flow turbulence subgrid scale microstructure is modeled via Dynamic Structure Large Eddy Simulation (DSLES) approach, largely resolving the anisotropic flow structures. The simulations are conducted across several fuel injector nozzle orifice dimensions ranging from 40–147 μm at a rail pressure of 1000 bar and typical compression-ignition engine operating condition of 900K and 60 bar, which is denoted as ECN Spray A. Liquid fuel physical properties are prescribed using a JP-8 surrogate mixture containing 80% n-decane and 20% trimethylbenzene (TMB) by volume. The reacting gas phase kinetics is modeled using the Aachen mechanism [26–27] and a detailed chemistry approach of a kerosene surrogate mixture. Measurements from the Army Research Laboratory (ARL) Constant Pressure Flow (CPF) chamber provide global spray and combustion parameters for comparison, including spray penetration profiles, ignition delay and flame lift-of-lengths (LOL) for JP-8 fuels. The simulation results present validated non-reacting and reacting spray simulations (ignition delay agreed within 4% and flame LOL agreed within 5% of measured data) and provide insights into the atomization and mixing characteristics across several orifice dimensions.

Comparison of Different Chemical Kinetic Models for Biodiesel Combustion

The current study compares the predictions by four different published mechanisms in literature which have been used for 3 dimensional compression ignition engine simulations. These four mechanisms use two different sets of surrogates: (a) methyl decanoate, methyl 9-decenoate, and n-heptane, (b) methyl butanoate and n-heptane. The mechanisms include: (1) 115 species and 460 reactions [1] using surrogate mixture (a); (2) 77 species and 209 reactions [2] using surrogate mixture (a); (3) 145 species and 869 reactions [3] using surrogate mixture (b); (4) 41 species and 150 reactions [4] using surrogate mixture (b). The different reduction techniques implemented to obtain the reduced mechanisms from the detailed mechanisms are briefly described. The surrogate mixture compositions are then modified to match the cetane number of the real biodiesel fuels. The experimental data for comparison include jet-stirred reactor data for species concentrations for biodiesel derived from rapeseed oil and 3 dimensional constant volume combustion data (for ignition, combustion, and emission characteristics), engine data (for pressure, heat release rate, and emission characteristics) for soy-derived biodiesel. 0-D and 3-D constant volume simulations with all the mechanisms can capture the general experimental trends quite well. Large surrogate models and mechanisms tend to provide better predictions at the expense of increased computational costs. The 115 species and 460 reaction mechanism was observed to perform the best among the mechanisms in predicting the jet-stirred reactor and 3-D constant volume data. It was observed that all the mechanisms are able to qualitatively capture the engine performance and emission characteristics.