Experimental And Numerical Studies On The Soot Formation Of Liquid Fuel Relevant Compounds

In the present thesis, fundamental experimental and numerical studies are performed for the soot formation of liquid fuel relevant compounds. The thesis is composed of four research studies. The first develops an improved data analysis approach for the combined laser extinction and two-angle elastic light scattering diagnostics to relate the various measured optical cross sections to soot aggregate properties. Compared to previously reported studies, the proposed approach can be applied to a wider range of soot sources by removing the assumption made to scattering regime or moment ratio of aggregate size distribution. The second study investigates the effects of n-propylbenzene addition to n-dodecane on soot formation and aggregate structure in a laminar coflow diffusion flame using the combined laser extinction and two-angle elastic light scattering method. It is shown that the relative importance of soot inception and surface growth affected by n-propylbenzene addition is different along the flame wing and centerline, with the aromatic fuel chemistry effect being stronger along the centerline. The third study extends the investigation on the same issue using a numerical model. The simulation results show that mixing n-propylbenzene into the liquid fuel mixture accelerates soot inception, and increases soot surface growth per unit surface area by PAH addition, while soot surface growth per unit surface area by HACA is shown to decrease modestly with n-propylbenzene addition. The fourth and final study investigates the soot formation from jet fuel in a laminar coflow diffusion flame using both numerical and experimental methods. The results demonstrate the robustness of the soot model to changes of fuel and also show that the HyChem model (i.e., lumped fuel breakdown approach, Xu et al., 2017) can be used to predict soot formation from real jet fuel combustion in laminar coflow diffusion flames by adding a PAH growth scheme to the model.