Theoretical and Laboratory Modeling of Turbulent Jets in Low Aspect Ratio Cavities

This research is motivated by the need to understand and parameterize mixing processes that occur during degassing and refilling of oil in large underground strategic petroleum reserve SPR caverns. The oil is pumped from cavern bottom to a degassing station and then returned to the cavern by using a vertical turbulent jet located near the cavern top. This dissertation concerns a laboratory experimental and theoretical modeling program conducted to investigate: (a) mixing mechanisms and precession of turbulent jets in homogeneous fluids (b) mixing of turbulent jet in stratified fluids and (c) wall attachment of offset jet in a homogeneous fluid. In the first part, a round turbulent axisymmetric jet is discharged in a long cylinder. It is found that the flow does not reach a true steady state, but vacillates periodically. Digital video recordings and particle image velocimetry are used to map the flow structures and velocity/vorticity fields, from which the frequency of jet switching, jet stopping distance, mean flow, turbulence characteristics and the influence of end-wall boundary conditions are inferred. The results are parameterized using the characteristic length D and velocity J^ (1/2)/D scales based on the jet kinematic momentum flux J and cylinder width D. The scaling laws so developed could be used to extrapolate laboratory observations to SPR flows. In the second part, a turbulent (positively or negatively) buoyant jet is injected vertically into a slender cylinder containing a stratified fluid. The interest is the vertical density distribution in the container and its dependence on time and other parameters. For each case (lighter or heavier jet) the experimental data, when properly non-dimensionalized, could be collapsed into a 'universal' time dependent behavior. A theoretical model is advanced to explain the results. Finally, a round turbulent offset jet in a low-aspect ratio cylinder is investigated. Particle Image Velocimetry and flow visualization are used for flow diagnostics. The measurements include the jet penetration (mixing) depth l, jet spreading rate and the mean velocity/vorticity fields for different offset positions 'Delta'. With the introduction of offset, the flow patterns change drastically. For 0 < Delta/D < 0.2, the jet deflects toward the wall while precessing (as in the axisymmetric case), for 0.2 < Delta/D < 0.4 the jet hugs the wall but with an oscillating tail, and for 0.45 < Delta/D the jet appears as a wall jet. In all cases, the jet is destroyed at a certain distance (mixing or penetration depth) from the origin. This mixing depth takes its lowest value for 0 < Delta/D < 0.2, with l ≈ (3.2-3.6)D, becomes maximum at Delta/D equal to 0.4 with l ≈ 5.2D, and drops to l ≈ 4.5D when the jet is close to the wall. Based on derived results some recommendations are made for optimal operation of SPR degassing.