Experimental and kinetic study of diesel/gasoline surrogate blends over wide temperature and pressure.

Blending diesel and gasoline is a promising way to stabilize combustion process and reduce emissions of NOx and soot in the Homogeneous Charge Compression Ignition (HCCI) engine. In order to computationally study diesel/gasoline blends for their autoignition characteristics, a five-component diesel surrogate (21.6% n-hexadecane, 15.5% n-octadecane, 26.0% isocetane, 20.7% 1-methylnaphthalene, 16.2% decalin, by mol.) and a five-component gasoline surrogate (11.0% n-heptane, 30.8% isooctane, 38.2% toluene, 10.3% diisobutylene, 9.7% cyclohexane, by mol.) are blended in three volume ratios of 30%/70%, 50%/50%, and 70%/30% (denoted as SD30, SD50, and SD70) to compare measured ignition delay with published data of diesel/gasoline blends. A heated shock tube and a heated rapid compression machine (RCM) are used to measure the ignition delay of three surrogate blends under compressed pressures of 6, 10, and 20 bar for fuel-lean (ϕ = 0.5), fuel-rich (ϕ = 1.5), and stoichiometric mixtures. Experimental results indicate that surrogate blends match well with fuel-lean and stoichiometric diesel/gasoline/air mixtures while get longer ignition delays for the fuel-rich case, which is thought of as the consequence of fitting the surrogate to fuel-lean conditions in the engine. The kinetic simulation for ignition delays is performed with a comprehensive mechanism. Due to the mechanism overpredicting ignition delays in the intermediate-to-high temperature region, proper refinement is done to the original mechanism based on sensitivity analyses at 1000 and 660 K. A further comparison of ignition delay predictive ability for two mechanisms is conducted subsequently. The modified mechanism not only simulates ignition delays in the intermediate-to-high temperature region successfully but also maintains the good performance of the original mechanism in the negative-temperature-coefficient (NTC) and low-temperature region. Besides, species and temperature profiles after the compression stroke in RCM are simulated using the modified mechanism for three surrogate blends at equal equivalence ratio, compressed pressure, and compressed temperature. Sensitivity analysis is also conducted to identify reactions dominating the system reactivity. [ABSTRACT FROM AUTHOR]