Piston Fuel Films as a Source of Smoke and Hydrocarbon Emissions from a Wall-Controlled Spark-Ignited Direct-Injection Engine 2003-01-0547

Thin films of liquid fuel can form on the piston surface in spark-ignited direct-injection (SIDI) engines. These fuel films can result in pool fires that lead to deposit formation and increased hydrocarbon (HC) and smoke emissions. Previous investigations of the effects of piston fuel films on engine-out HC and smoke emissions have been hampered by their inability to measure the fuel-film mass in operating direct-injection engines. In this paper, a recently developed high-speed refractive-index-matching imaging technique is used for quantitative time- and space-resolved measurements of fuel-film mass on a quartz piston window of an optically-accessible direct-injection engine operating over a range of fully-warmed-up stratified-charge conditions with both a high-pressure hollow-cone swirl-type injector and with a high-pressure multihole injector. Measured fuel-film mass is a small percentage of the total fuel injected with the high-pressure swirl injector (maximum of ∼1% with gasoline fuel and ∼0.1% with isooctane fuel). Most of the piston fuel-film mass evaporates during the cycle and burns as a pool fire. These pool fires are observed by endoscopic and through-the-piston imaging, and the occurrence and location of the pool fires are consistent with the measured piston fuel films. The fuel-film data are also correlated with engine-out HC and smoke emissions measurements from a conventional all-metal single-cylinder engine of the same design. Smoke emissions from the engine with a high-pressure swirl injector increase linearly with the measured fuel-film mass. Fuel films are found to be the dominant source of smoke emissions with the swirl injector in this engine, with ∼10% of the wall-film mass converted to emitted smoke mass. Smoke emissions from the engine with a high-pressure multihole injector are very small or zero, consistent with the much smaller measured fuel-film mass (∼0.05% of the injected gasoline fuel volume). In contrast, engine-out HC emissions do not correlate with fuel-film mass. For optimum injection timings, the measured fuel-film mass is so small that even in the unlikely event that all of the fuel film mass was converted to engine-out HC emissions, fuel films could account for less than 15% of the total HC emissions for the swirl injector and less than 2% for the multihole injector. For off-optimum injection timings, the HC emissions are significantly larger, but wall films can account for at most 35% of the unburned HC emissions. This is contrary to some previous studies that claimed fuel films were the largest contributor to HC emissions (∼80%) in stratified SIDI engines. The data from this engine support overmixing as the dominant source of HC emissions for optimum engine operating conditions. However, fuel films may be a significant source of HC emissions for cold start or low-speed engine-operating conditions.