The thesis describes the characterisation and application of the laser induced incandescence technique for making soot measurements in practical devices. Laser induced incandescence is the phenomenon whereby particulates such a soot absorb laser radiation and are heated to a temperature much higher than the bath gas. The broadband incandescence signal from the hot particles can be detected and the signal is proportional to volume fraction. The technique was used to study soot in partially premixed counterflow ethylene air flames, iso-octane explosion flames, and to image soot in a D. I. Diesel engine. Mie scattering, OH-LIF and absorption were used a complementary diagnostics. Appropriate ratios of LII and Mie images allowed the relative particle size and number density to be imaged. The counter flow burner measurements were used to study the effects of strain on soot formation while the bomb work demonstrated soot production in hydro-dynamically unstable cellular flames. The Diesel engine measurements are a demonstration of optical diagnostics in a real device. In order to characterise the LII signal behaviour two types of carbon aerosol generators were built. The liquid dispersion device produces almost spherical sub-micron carbon black particles. The device was used to characterise the soot field response, laser fluence response, signal decay and spectrum of the LII signal.

The steady coflow diffusion flame is a widely used configuration for studying combustion kinetics, flame dynamics, and pollutant formation. In the current work, a set of diluted ethylene-air coflow flames are simulated to study the formation, growth, and oxidation of soot, with a focus on the effects of pressure on soot yield. Firstly, we assess the ability of a high performance CFD solver, coupled with detailed transport and kinetic models, to reproduce experimental measurements, like the temperature field, the species’ concentrations and the soot volume fraction. Fully coupled conservation equations for mass, momentum, energy, and species mass fractions are solved using a low Mach number formulation. Detailed finite rate chemistry describing the formation of Polycyclic Aromatic Hydrocarbons up to cyclopenta[cd]pyrene is used. Soot is modeled using a moment method and the resulting moment transport equations are solved with a Lagrangian numerical scheme. Numerical and experimental results are compared for various pressures. Reasonable agreement is observed for the flame height, temperature, and the concentrations of various species. In each case, the peak soot volume fraction is predicted along the centerline as observed in the experiments. The predicted integrated soot mass at pressures ranging from 4-8 atm, scales as P2.1, in satisfactory agreement with the measured integrated soot pressure scaling (P2.27). Significant differences in the mole fractions of benzene and PAHs, and the predicted soot volume fractions are found, using two well-validated chemical kinetic mechanisms. At 4 atm, one mechanism over-predicts the peak soot volume fraction by a factor of 5, while the other under-predicts it by a factor of 5. A detailed analysis shows that the fuel tube wall temperature has an effect on flame stabilization.