Shedding light on soot burnout in conventional diesel combustion by high-speed laser-induced incandescence imaging

Laser-induced incandescence (LII) has become a well-known technique for tracking soot in combustion environments. One of its applications is in the field of engine research, where it has been utilized to study in-cylinder soot behavior for many years. Almost all research up to date involves low-speed LII, allowing only one image to be taken per cycle. To obtain crank-angle-resolved data from a single combustion cycle, a high-speed system should be applied. The present work was performed to investigate the feasibility of employing such a highspeed system in engine experiments. Furthermore, high-speed LII was applied to study spatiotemporal behavior of soot in conventional diesel combustion (CDC) under varying test conditions. Attention was mainly directed to the burnout phase, i.e. after the injection event, for a fundamental understanding of this phase is still lacking. Initially, effects of repeated pulsing on the properties of soot were studied. Measurements were performed on a stable diffusion flame under atmospheric conditions. Sublimation and local gas heating, both occurring in a step-wise fashion, were identified as possible multiple exposure effects. Sublimation should be avoided to prevent excessive alterations to soot particles by employing a low fluence. Local gas heating, on the other hand, is expected to have a minor impact in engine experiments, because of elevated conduction rates. Subsequently, high-speed LII experiments were performed on an optically-accessible compression ignition (CI) engine. The technique was found to be feasible for visualizing soot in engine applications, although the balance between fluence and signal-tobackground ratio is extremely sensitive. To obtain useful images, the strong broadband background radiation present in CDC has to be effectively suppressed. Employing proper detection equipment is essential, for the 40 ns gating time used in this work proved to be a limiting factor in obtaining sufficient signal-to-background ratio. Also, filtering in the blue part of the spectrum assists in separating LII signals from background luminosity. Applying a low fluence poses additional problems in terms of interpretation of the LII signal, as pulseto-pulse fluctuations in laser power, irregularities in the spatial beam profile and beam steering effects cause variations in signal yield. Additionally, beam attenuation due to large amounts of absorption decreases fluence, causing downstream soot to be undetected. Measurement data suggest that burnout of soot is enhanced when injection pressure is increased. LII signals prevail longer at low injection pressure, which might indicate that soot concentrations are higher late in the combustion cycle. Moreover, signal intensity is considerably lower at high injection pressure, especially in the burnout phase. Incylinder fluid dynamics were shown to be consistent between measurements, but also comparable between test cases. The observed differences in LII signal can therefore not be ascribed to changes in flow behavior. Another influence factor is the extreme temperature dependency of the LII signal, which definitely plays a role in signal yield. However, it was argued that temperature variations in between test cases would actually strengthen the observed trends. Fuel composition has been shown to affect the sooting tendency during combustion also. Applying a gas-to-liquid (GTL) synthetic diesel, low in sulfur and aromatics, produces significantly less LII signal than operating on European diesel (EN590). In addition, this decreasing trend is observed for all applied injection pressures. It is extremely difficult to pinpoint the origin of the difference by LII experiments alone. To what extent temperature influences these results is yet unclear.