UNCERTAINTY QUANTIFICATION FOR LIQUID PENETRATION OF EVAPORATING SPRAYS AT DIESEL-LIKE CONDITIONS

Seeking to quantify the liquid volume fraction at the measured liquid penetration length for more forthright comparison to CFD results, we compared 10 different light-scatter and extinction diagnostics for measurement of the "liquid length" of an evaporating diesel spray. Results show that light-scatter imaging is sensitive to the orientation of the illumination source, producing different maximum intensity locations depending on the optical setup. However, the scattered intensity from different setups can be normalized to provide similar liquid length values if the appropriate reference intensity is known. Light-extinction diagnostics are more quantitative because of a built-in reference light intensity, but can be sensitive to beam-steering effects due to refractive index gradients. The most quantitative diagnostic in this study is a small laser beam with large collection optics to accommodate beam steering. Using a liquid length defined based on 3% of the maximum scatter intensity and the measured optical thickness at this same axial location, we estimate an expected range of liquid volume fraction at this position for the "spray A" conditions of the Engine Combustion Network. Even though there is a possibility that this condition has supercritical mixtures where distinct droplets do not exist, we apply Mie scatter theory with a range of droplet diameters (0.1−10 µm) to mimic how light may scatter at liquid surfaces where density gradients remain sharp and light effectively scatters as if there were a gas-liquid interface. With a measured liquid path length of 1.4 mm, the upper-bound estimate for the path-length-averaged liquid volume fraction is 0.15%.