SUMMARY
Predictive engine simulations are key for rapidly exploring and optimizing the design of cleaner burning and more fuel efficient engines. However, the physics governing the breakup of an injected liquid fuel jet into droplets have not been well characterized to date. Three mechanisms are believed to contribute to primary atomization in diesel sprays, namely the aerodynamic growth of waves on the fuel jet surface, turbulence generated in the injector nozzle, and cavitation. If computational design tools are to be used to guide the use of direct injection strategies for cleaner and more fuel efficient engines, the physics underpinning the role of these primary atomization mechanisms must be better understood to ensure the development of predictive simulations of fuel-air mixing and vaporization within the engine. Thus, the central aim of this thesis is to improve the physical representation of spray breakup physics within today's engine simulation packages. The work presented in this thesis investigates the role of the proposed physical mechanisms on the primary atomization process in diesel sprays. In order to advance current understanding of spray breakup, the dynamic and geometric factors contributing to cavitation were suppressed so that primary atomization due to aerodynamics and nozzle-generated turbulence could be studied in isolation. A new experimental methodology was developed and applied in a high-pressure spray chamber to characterize the average size of droplets formed from the spray breakup process. This experimental data, in conjunction with x-ray measurements from the Argonne National Laboratory, have been used to assess predictions from existing spray breakup models. Comparison between predicted and measured drop size distributions revealed that an aerodynamic-induced spray breakup model could capture experimentally observed sensitivities under conventional diesel engine conditions. However, for injection into relatively lower ambient density environments, the model could not accurately predict the initial rate of droplet size decrease in the near-nozzle region, suggesting that other mechanisms, such as turbulence generated inside the nozzle, likely augment the primary breakup process. Evaluation of droplet sizes under low ambient density conditions allowed for the turbulence-induced breakup process to be studied, while minimizing the influence of aerodynamic inertial forces on the spray. Although several turbulence-induced breakup models have been proposed in the literature, empirical correlations describing droplets formed from eddies within the inertial sub-range of the turbulence spectrum were best able to capture the measured sensitivities in droplet size to changes in ambient and injection conditions. These findings informed recommendations for an improved hybrid spray breakup model, capable of representing both aerodynamic and turbulent breakup mechanisms in the atomization of non-cavitating diesel sprays.