Lucky Mulenga

Date of Award

January 2018

Document Type


Degree Name

Doctor of Philosophy (PhD)


Chemical Engineering

First Advisor

Gautham Krishnamoorthy


The goal of this dissertation was to obtain greater insights into detonation scenarios involving hydrogen-air/oxygen mixtures using computer simulations. A primary goal was to use coarse meshes to study detonations for realistic geometries and scales in a computationally efficient manner. We identified the chemistry model, kinetics model, and turbulence model that helped us during our investigation. We further studied the influence of equivalence ratio, viscosity and radiation on various detonation scenarios.

In the first chapter, we begin by introducing the pertinent experimental and theoretical background of detonations. This section serves the purpose of preparing the reader on the main ideas in the research area. In addition to this, we list our contributions to the research area.

In chapter 2, the impact of viscous and radiative losses and the point source approximation on detonation hydrodynamics were studied using a hydrocode and a computational fluid dynamics (CFD) framework for hydrogen-air mixtures. The hydrocode solved for the hydrodynamics using the non-reacting TNT equivalency method as well as the inviscid (Euler) equations and the JWL equation of state. For the CFD framework, we solved the hydrodynamics using the SRK equation of state, Large Eddy Simulations (LES) as well as the spectrally-averaged mean absorption coefficient for radiative properties. In addition, the CFD framework employed a 21-step detailed chemistry mechanism utilizing a hydrogen-air mixture

After validating our simulation methodology by comparing it to transient pressure profile measurements from a small-scale explosion study, the settings in the validation were then utilized to solve for a detonation in a larger domain. The study on the impact of equivalence ratio showed that rich and lean flames strengthened the acceleration and strength of the wave. While viscous losses were shown to weaken the detonation, the impact of radiation wasn’t appreciable due to the difference in the magnitude of the radiative source and chemical heat release term.

In Chapter 3, we report our findings on the feasibility of a coarse mesh (cell size ~ 2mm) finite volume solver to reproduce experimental research on hydrogen-air mixture detonations. The solver utilized: Large Eddy Simulation (LES) to model the turbulence, a 21-step detailed mechanism to model the combustion, estimated transport properties (binary diffusion coefficients) using kinetic theory and employs the Soave-Redlich-Kwong equation of state to account for compressibility effects associated with the high-pressure detonation wave.

Our solution methodology was first validated by comparing numerical predictions against experimental measurements of the interaction of a non-reacting shock wave against the walls of a cavity. We then carried numerical predictions of detonation at different blockage ratios (BR), BR = 0.3 and BR = 0, and equivalence ratios (0.5, 0.75, 1.0, 1.25 and 1.5). Using our approach, we showed that the trends of the detonation velocities with blockage ratios and equivalence ratios followed experimental trends. The methodology can therefore be extended to other detonation scenarios that have large dimensions or complex geometry and that require coarse computational cells (~2mm).

Finally, we finish the dissertation by discussing our contributions to the research and some thoughts on future work