Relations between MILD combustion and LTC strategies in diluted compression ignition conditions

This study investigates the applicability of the classical Moderate and Intense Low Oxygen Dilution (MILD) combustion definition to identify combustion regimes under high-pressure conditions relevant to compression ignition (CI) engines. Using n-heptane as a diesel surrogate fuel, a novel methodology has been developed that combines an experimental campaign, Stochastic Reactor Model (SRM) simulations, and well-stirred reactor (WSR) modeling to analyze Low-Temperature Combustion (LTC) strategies under both idealized and engine-relevant conditions. Initially, WSR simulations with detailed chemistry were used to generate $T_{\mathrm{in}}-X_{O_2}$ maps, analyzing the effects of pressure, dilution, and equivalence ratio on combustion regimes. The results show that elevated pressure lowers self-ignition temperature ($T_{\mathrm{si}}$), while increased dilution and lean mixtures promote flattening of the S-curve, facilitating the transition to the MILD regime. However, satisfying the MILD criterion $\Delta T<T_{\mathrm{si}}$ becomes increasingly difficult under high-pressure conditions, requiring higher levels of dilution and inlet temperature. Low-temperature oxidation (LTO) was found to enhance pre-ignition behavior when $T_{\mathrm{si}}$ exceeded approximately 525 K, with a more pronounced effect at elevated pressures. The SRM, calibrated against CFD-supported experimental data of the PCCI regime, was then used to replicate realistic in-cylinder conditions of both PCCI and HCCI, and to refine WSR input parameters. The comparative analysis revealed that HCCI enables LTC at lower oxygen dilution levels compared to PCCI, emphasizing the importance of mixture homogeneity and pre-ignition chemistry. The observed reduction in $T_{\mathrm{WSR}}$ at high pressures across varying oxygen concentrations is consistent with the primary goal of LTC strategies-minimizing NO${}_x$ emissions. Overall, this work demonstrates the limitations of idealized homogeneous reactors in identifying MILD regimes under realistic conditions and highlights the value of incorporating advanced yet computationally feasible reactor models, such as SRM, for regime mapping and emissions analysis in practical engine applications.