Bulletin of the American Physical Society
65th Annual Meeting of the APS Division of Fluid Dynamics
Volume 57, Number 17
Sunday–Tuesday, November 18–20, 2012; San Diego, California
Session M2: Convection and Buoyancy-Driven Flows VII |
Hide Abstracts |
Chair: Aline Cotel, University of Michigan Room: 23A |
Tuesday, November 20, 2012 8:00AM - 8:13AM |
M2.00001: Turbulent acidic jets and plumes injected into an alkaline environment Hendrik Ulpre The characteristics of a strong acidic turbulent jet or plume injected into an alkaline environment comprising of a weak/strong base are examined theoretically and experimentally. A chemistry model is developed to understand how the pH of a fluid parcel of monoprotic acid changes as it is diluted and reacts with the ambient fluid. A standard fluid model, based on a top-hat model for acid concentration and velocity is used to express how the dilution of acid varies with distance from the point of discharge. These models are applied to estimate the point of neutralisation and the travel time with distance within the jet/plume. An experimental study was undertaken to test the theoretical results. These experiments involved injecting jets or vertical plumes of dilute nitric acid into a large tank containing a variety of base salts dissolved in water. The injected fluid contained litmus indicator dye which showed a change in colour from red to blue close to the point of neutralisation. In order to obtain a range of neutralisation distances, additional basic salts were added to the water to increase its pH buffering capacity. The results are applied to discuss the environmental implications of an acidic jet/plume injected into the sea off the South East coast of Great Britain. [Preview Abstract] |
Tuesday, November 20, 2012 8:13AM - 8:26AM |
M2.00002: Convection in a stratified atmosphere: from isolated thermals to the convective boundary layer Chiel van Heerwaarden, Juan Pedro Mellado We have used direct numerical simulations to study the transition between an isolated turbulent plume penetrating a stably stratified layer and the convective boundary layer, one of the archetypes of the atmospheric boundary layer. Our simulation setup consists of a stably stratified fluid that is heated from below by thermals that form over patches with high surface heat fluxes. The patches are surrounded by regions without surface flux. We have defined a non-dimensional system that allows for studying the transition by varying only one parameter: the ratio of the characteristic length scale of the largest turbulent motions to the distance between two individual patches of high heat flux. By varying its value from zero, which defines the isolated thermal, to the threshold at which the patches are so close that the statistics resemble those of the convective boundary layer, we span the entire transition. Our results show that when the thermals are sufficiently close, the presence of nearby thermals limits their lateral expansion. Rather than merging with neighboring thermals, each thermal organizes into a rising core surrounded by narrow regions of subsiding motions. Nonetheless, in this case many flow statistics resemble those of the convective boundary layer. [Preview Abstract] |
Tuesday, November 20, 2012 8:26AM - 8:39AM |
M2.00003: Turbulent Free Convection Over a Heated Plate Juan Pedro Mellado Temporally-evolving turbulent free convection above a heated plate is investigated by means of direct numerical simulations. This study complements previous work in Rayleigh-B{\'e}nard convection and in the convective boundary layer. Results show a vertical structure with distinct, overlapping inner and outer layers. Townsend scaling using the surface flux and the molecular diffusivity characterizes the former and Deardorff scaling using the convection scales does so in the latter. It is also observed that the inner layer approaches a quasi-steady behavior. Then, some statistics inside this region coincide with the corresponding ones in classical Rayleigh-B{\'e}nard convection if the plate is interpreted as half of the convection cell, which justifies the transfer of results between the two configurations. Despite the moderate Rayleigh numbers, of the order of $10^9$, the overlap region between the inner and outer layers starts to show the power-law scaling of the buoyancy variance predicted by the classical similarity theory and seen in the atmosphere. Results also indicate a growth rate of the convective boundary layer in neutral conditions larger than previous estimates derived by extrapolation from related configurations. [Preview Abstract] |
Tuesday, November 20, 2012 8:39AM - 8:52AM |
M2.00004: Numerical simulation of buoyant jet issuing from a seabed Akihiko Nakayama, Hamid Rahai, Jeremy Bricker A numerical simulation of effluent mixing in a coastal region has been conducted. The calculation method is a large eddy simulation using the filtered equations of motion and the equations for temperature and salinity without assumption of hydrostatic pressure distribution. The method is tested in a calculation of wall jet and applied to a real flow of effluent discharged from an outfall at the bottom of sea near a coast. The calculation region is about several hundred meters in the horizontal direction and tens of meters in the vertical direction and the time span of about one hour. The results can be fed into larger scale calculation that may not resolve turbulent mixing but calculates dispersion in larger area. [Preview Abstract] |
Tuesday, November 20, 2012 8:52AM - 9:05AM |
M2.00005: Asymmetric character of Rayleigh-Taylor and double-diffusive fingers in reactive systems L. Lemaigre, M.A. Budroni, L.A. Riolfo, P. Grosfils, A. De Wit Buoyancy-driven flows induced by the hydrodynamic Rayleigh-Taylor or double diffusive instabilities develop symmetrically around the initial contact line when two solutions of given solutes with different density are put in contact in the gravity field. If the solutes affecting the density of these solutions are involved in chemical reactions, changes in composition due to the underlying reaction-diffusion processes can modify the density profile in space and time and affect the hydrodynamic patterns. We show both experimentally and numerically that the resulting chemo-hydrodynamic patterns are asymmetric with regard to the initial contact line as fingers develop here upwards and not both up and downwards as in the non reactive situation. In some cases, the dynamical in-situ generation of the product of the reaction can even lead to the co-existence of different convective modes. The experiments are performed using an aqueous solution of an acid put in contact in a Hele-Shaw cell with a miscible aqueous solution of a base. The numerical simulations are made on the basis of a reaction-diffusion-convection model coupling Darcy's law to the evolution equations of the concentrations of the various chemicals involved. Good agreement between experimental results and theory is obtained. [Preview Abstract] |
Tuesday, November 20, 2012 9:05AM - 9:18AM |
M2.00006: Transport Processes in GaN Deposition in a Chemical Vapor Deposition Reactor Yogesh Jaluria, Jiandong Meng, Sun Wong A study has been carried out to characterize the metalorganic chemical vapor deposition (MOCVD) growth of Gallium Nitride (GaN). With trimethylgallium (TMGa) and ammonia (NH$_{3})$ carried by hydrogen (H$_{2})$, as precursors, the entire process which involves fluid flow, heat and mass transfer and chemical kinetics is modeled. A major objective of this work is to examine the dependence of the growth rate of GaN and film uniformity on the flow, as determined by various design parameters and operating conditions involved in the MOCVD process. The results are expected to provide a quantitative basis for the design and optimization of MOCVD system for the fabrication of GaN devices. The study focuses on techniques to guide the impinging flow, and the effect of buoyancy on the resulting flow. Based on the detailed mathematical model and the appropriate chemical mechanisms, a study on the effects of various critical parameters such as the reactor pressure, inlet velocity, susceptor temperature, inflow concentration and rotating speed on the flow and on the growth rate of GaN and thin-film uniformity is conducted for a 3D rotating reactor. The comparison between 3D modeling and previous 2D impinging vertical reactor modeling is presented. The flow and associated transport processes are discussed, indicating approaches to improve the uniformity of the film. In addition, the dependence of the film quality on the inflow profile is also examined, which makes an attempt to minimize the effect of reactor pressure on fluid loss and reduce flow recirculation. [Preview Abstract] |
Tuesday, November 20, 2012 9:18AM - 9:31AM |
M2.00007: Inertial effects in an incompressible stratified Euler fluid in a channel Giovanni Ortenzi, Roberto Camassa, Shengqian Chen, Gregorio Falqui, Marco Pedroni Inertial properties of an incompressible Euler fluid are discussed in the case of a stably stratified fluid confined between infinite rigid upper and lower horizontal plates in hydrostatic equilibrium at infinity. In this set-up, the possibility of non-conservation of horizontal momentum emerges, despite the fact that only vertical external forces act on the system (an apparent paradox seemingly first noticed by Benjamin, 1986). We show that this phenomenon is a consequence of the rigid lid constraint coupling through incompressibility with the infinite inertia of the far ends of the channel. When inertia is removed by eliminating the stratification or, remarkably, by using the Boussinesq approximation, horizontal momentum conservation is recovered. The pressure imbalance responsible for lack of horizontal momentum conservation and its comparison with direct numerical simulation is explicitly shown for: 1) a two-layer full Euler system with small velocities and density variations, and 2) a long wave one-dimensional model of the same set-up. A side effect detected by the model is the relevance of internal-wave dispersion at short times. Effects on the variation of physical quantities other than momentum, such us the total circulation of the fluid, will be discussed. [Preview Abstract] |
Tuesday, November 20, 2012 9:31AM - 9:44AM |
M2.00008: Characterization of Mixing Between Water and Biofuels Aline Cotel, Erica Green, Marina Acevedo, Margarita Otero, Avery Demond Currently, gasoline containing ethanol is considered to be among the best alternatives to gasoline. However, the potential environmental impact of a spill of ethanol-based biofuels on aquatic environments is an area of open discussion and research. Since these fuels are a combination of a miscible fluid (ethanol) and an immiscible fluid (gasoline), models used for traditional gasoline fuels (immiscible in water) are not applicable. Preliminary experiments show that when a solution of ethanol and glycol is mixed with water, a third mixed fluid is formed. Two distinct mixing regimes were observed. An exothermic reaction also occurred between ethanol and water. In the first regime, a turbulent wake is created between the ethanol/glycol and water layers causing the ethanol and glycol solution to entrain and mix into with the water phase. Because the mixed fluid is denser than either parent fluid, a dramatic overturning is possible. The amount of mixing was found to be dependent upon the initial ratio of ethanol to glycol in the parent fluid. The second regime begins when the turbulent wake has dissipated and the internal wave created by the plate has begun to settle, typically within the first minute. At this point, B\'{e}nard-like cells, similar to those typically seen in Rayleigh-B\'{e}nard convection, form at the interface and relatively slow mass transfer is evident. The cells at the interface show distinct features of interfacial turbulence, including small transverse waves, denoting that instabilities exist there. [Preview Abstract] |
Tuesday, November 20, 2012 9:44AM - 9:57AM |
M2.00009: Flux variation and layering in turbulent stratified Taylor-Couette flow Rosalind Oglethorpe, C.P. Caulfield, Andrew W. Woods We present new experimental measurements of turbulent mixing in stratified Taylor-Couette flow. We vary the inner radius and rotation rate, and consider both two layer and initially linear stable stratification. With a two layer stratification, we demonstrate that the flux of salt through the interface is a non-monotonic function of a Richardson number $Ri=g\Delta\rho D/(\rho_0u^2)$, where $D$ and $u$ are characteristic length and velocity scales of the turbulent flow. As predicted by Phillips (1972){\it DSR} {\bf19}, this behaviour implies that a linear stratification will spontaneously form layers of relatively well mixed fluid separated by relatively thin interfaces. We demonstrate the spontaneous development of such layers in experiments where the stratification is initially linear with constant buoyancy frequency $N$. The depth of these layers $h_L\propto U_H/N$, where $U_H$ is a horizontal velocity scale, and the flux through these layers is independent of $h_L$. In particular we demonstrate, consistently with Woods {\it et al.} (2010){\it JFM} {\bf663}, who considered the strongly stratified limit in two layer flow, that the salt flux tends to an asymptotic constant value when the flow is strongly stratified, even when multiple layers are present. [Preview Abstract] |
Tuesday, November 20, 2012 9:57AM - 10:10AM |
M2.00010: Entrainment and mixing dynamics of surface-stress-driven linearly stratified flow in a cylinder Georgy Manucharyan, C.P. Caulfield We consider experimentally a linearly stratified fluid (with constant buoyancy frequency $N$) in a cylinder of depth $H$ subject to surface stress forcing from a disk spinning at constant angular velocity $\Omega$. A turbulent mixed layer develops bounded by a sharp interface of constant thickness. Its depth $h/H \sim (N/\Omega)^{-2/3} (\Omega t)^{2/9}$. We argue this is a consequence of: the kinetic energy of the mixed layer staying constant with time (as previously observed in a two layer flow by Shravat et al. 2012) the entrainment at the interface being governed entirely by local processes; and the rate of increase of the total potential energy of the fluid being dependent only on the global dissipation rate and the ratio $N^2/\Omega^2$. Below the moving primary interface, we also observe in some circumstances the formation of another partially mixed layer, separated by a secondary interface from the linearly stratified fluid below. Depending on the local flow properties, the secondary interfaces can exhibit rich time-dependent dynamics including drift towards or away from the primary interface, merger and/or decay. The secondary interfaces appear to develop due to the non-monotonic dependence of buoyancy flux on stratification as originally argued by Phillips (1972). [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700