| Summary: | Thesis (Ph.D.)--Massachusetts Institute of Technology, Dept. of Civil and Environmental Engineering, 2000. === Includes bibliographical references (p. 235-241). === Open-water disposal and capping is a promising solution for disposing of the 14 to 28 million m3 of contaminated sediment dredged annually in the United States (National Research Council, 1997). Such practice raises concerns about the feasibility of accurately placing the material in a targeted area and the loss of material to the environment during disposal. To better predict the fate of these materials, the objective of this research was to gain new insight into the physical processes governing the mechanics of their convective descent. Instantaneously released sediments form axisymmetric "clouds" resembling selfsimilar thermals. Current particle cloud models employ thermal theory and an integral approach using constant entrainment (a), drag (CD), and added mass (k) coefficients. The aim of this study was to investigate how real sediment characteristics (particle size, water content, and initial momentum) affect cloud behavior (i.e., velocity, growth rate, and loss of particles) and time variations in a, CD, and k. Flow visualization experiments were conducted using a glass-walled tank, special sediment release and capture (i.e., "trap") mechanisms, and various cohesive and non-cohesive particles. Particle sizes were scaled to real-world dimensions through the cloud number (Nc), defined as the ratio of the particle settling velocity to the characteristic cloud velocity. An "inverse" integral model was developed in which the conservation equations were solved for a and k using measured velocity and radius data. Based on the "inverse" model results, particle cloud experiments were simulated with an integral model using constant and time-varying a and k. The non-cohesive sediments evolved rapidly into "thermals" with asymptotic deceleration and large growth rates (a = 0.2 - 0.3). The particles eventually organized into "circulating thermals," with linear growth rates obeying buoyant vortex ring theory. In this phase, large particles (Nc > 10-) produced laminar-like vortex rings with small a (0.1 - 0.2). Compared to the cohesive sediments, which exhibited a wide range of growth rates, changes in water content and initial momentum of the non-cohesive particles produced 10 - 20 % variations in a. Material not incorporated into the cloud upon release formed a narrow "stem" behind the cloud, which contained as much as 30 % of the original mass depending on the release conditions. Much of the "stem" material either re-entrained into the cloud later in descent or reached the bottom shortly after it. Material not incorporated into the "stem," which may be advected by ambient currents, was found to be only a small fraction (< 1 %) of the original mass. Inverse integral model results suggest that CD and k are close to zero within the "thermal" phase. In the "circulating thermal" phase, the reduction in a caused by large particles (Nc > 10-4) increased k to a value similar to that of a solid sphere. Integral model results confirm the suitability of using constant coefficients for modeling particle clouds with Nc less than 10-. When Nc is greater than 10-4, time-varying a and k are required to properly simulate cloud behavior in the "circulating thermal" phase. === by Gordon J. Ruggaber. === Ph.D.
|
|---|
