An analytical and quantitative analysis of the laser-induced incandescence of soot

The focus of this research was on the development of a diagnostic method called Laser-Induced Incandescence (LII) for the measurement of different parameters of combustion-generated carbon particulate, or soot. This non-intrusive technique involves the heating of soot particles in-situ with a pulsed laser source and detecting the subsequent visible emission from the particles. In this research, LII signals were acquired in both a well-documented diffusion flame and a simulated exhaust flow for the purpose of making concentration and particle size measurements. For both types of measurements, the effect on the LII signal of experimental parameters such as gas temperature, particle size, soot composition and morphology, laser fluence, laser wavelength, detection wavelength and timing are analyzed. Two particle sizing methods were considered, one based on the post-laser-heating signal decay rate (conductive cooling rate) and the other by measuring the peak temperatures attained (pyrometry).; A dependence of the LII signal per soot volume on particle size was found, with larger particles emitting more signal. This is in agreement with the numerical predictions of a modeled simulation of the LII process. Positive results are obtained for sizing measurements with each of the proposed techniques, However, limitations for both are also noted. Both model and experimental results indicate that nonuniform particle size distributions will make such size measurements more difficult. For an effective 1-D Gaussian laser distribution, the detected (integrated) LII signal magnitude, spectral signal profile (observed temperature, ±200 K), and signal decay rate are insensitive to changes in laser fluence above a certain threshold (0.3 to 0.5 J/cm2, depending on laser wavelength). At higher fluences with visible wavelength illumination, significant emission from laser-produced C ₂ is also noted. The LII signal was also detected at time intervals delayed from the laser illumination. Little difference (<10%) from ‘promptly’ detected signals was noted up to delays of 100 ns. The LII signal also shows sensitivity to laser beam shape, strong soot concentration gradients, and signal trapping in dense soot regions. In conclusion, the implications of these LII signal characteristics for applications in real-world combustion and exhaust environments are discussed.