Joseph Katz, William F. Ward Sr. Distinguished Professor, Department of Mechanical Engineering, Johns Hopkins University
Abstract: A series of laboratory-scale experiments examine the generation of both subsurface and airborne crude oil droplets by breaking waves, bursting of bubbles, subsurface plumes and raindrop impact. For waves, premixing the oil with dispersant (Corexit 9500A) reduces the droplets sizes to the micron- and submicron-scales, and changes the slope of their size distribution. Without dispersant, the characteristic oil droplet diameters can be predicted based on the relevant turbulence scales. Once entrained, the temporal evolution of concentration and size distribution of these droplets can be modeled as a combined effect of turbulent diffusion and buoyant rise. With dispersant, the droplet sizes are much smaller than the turbulence scales, in part due to tip-streaming at the oil-water interface. Furthermore, the droplet fragmentation persists long after the wave breaking. Aerosolization of oil is caused both by the initial splash and by subsequent bubble bursting, as entrained bubbles rise back to the surface. Dispersants increase the airborne nano-droplet concentration by orders of magnitude, raising health concerns. The shape of subsurface crude oil plumes in cross flow is affected by the droplet sizes and their interaction with the plume’s counter-rotating vortex pair. Hence, dispersants modify the entire plume geometry and the spatial distribution of droplets in it. The near-field plume breakup processes, which determines the droplet sizes, are probed by matching the refractive index of silicon oil with that of sugar water. The measurements show that the droplet sizes, location of breakup, and even the plume scales are Reynolds- and Weber-number dependent. The frequent generation of compound oil droplets and ligaments, which contain smaller water droplets, has significant effect on the oil-water interfacial area.