Soot Volume Fraction Measurements in a Gasoline Direct Injection Engine by Combined Laser Induced Incandescence and Laser Extinction Method 2010-01-0346

In order to study the soot formation and oxidation phenomena during the combustion process of Gasoline Direct Injection (GDI) engines, soot volume fraction measurements were performed in an optical GDI engine by combined Laser-Induced Incandescence (LII) and Laser Extinction Method (LEM). The coupling of these two diagnostics takes advantages of their complementary characteristics. LII provides a two-dimensional image of the soot distribution while LEM is used to calibrate the LII image in order to obtain soot volume fraction fields. The LII diagnostic was performed through a quartz cylinder liner in order to obtain a vertical plane of soot concentration distribution. LEM was simultaneously performed along a line of sight that was coplanar with the LII plane, in order to carry out quantitative measurements of path-length-averaged soot volume fraction. The LII images were calibrated along the same path as that of the LEM measurement. A calibration factor was obtained for each pair of LII and LEM measurements. The analysis of the evolution of this calibration factor for varying measurement conditions was used to evaluate the validity of the methodology.

These techniques were then applied to study the in-cylinder soot concentration evolution in a GDI engine equipped with an outward-opening piezoelectric injector. The impact of injection strategy and EGR rate on soot concentration were studied in stratified operation. Results show that soot production is strongly correlated to hot fuel-rich regions. In competition with the soot production phenomena, soot oxidation processes are responsible for a dramatic decrease (1 to 2 orders of magnitude) of in-cylinder soot volume fraction during the expansion stroke until about 90 CAD after the combustion TDC. Soot oxidation is mainly governed by the mixing between soot and hydroxyl radicals. Beyond ~90 CAD after TDC in-cylinder mean temperature seems to be too low for soot-oxidation reactions to occur. Additional measurements were performed under homogeneous fuel rich conditions in order to further validate the techniques. The soot volume fraction evolution showed that no soot oxidation occurred. This result is consistent with the fact that almost no oxidant is left after the main combustion process under these conditions.