13:30
Transport and mixing 1
Chair: J Christos Vassilicos
13:30
15 mins
|
Simultaneous velocity and density measurements in variable density mixing of fully turbulent buoyant gas jets
John Charonko, Kathy Prestridge
Abstract: Fully turbulent buoyant jets with high and low density ratios (Atwood numbers) at an initial jet Reynolds number of ~19,000 are studied at several locations downstream of the orifice. Simultaneous planar PIV and acetone PLIF are used to compute density weighted velocity statistics to determine the evolution of the turbulent behavior in the flow. Buoyancy effects preserve or increase the turbulent behavior with distance in large density ratio flow, but not when the density ratio is small. Also, asymmetries in the PDFs appear for buoyant flows that are absent for non-buoyant conditions.
|
13:45
15 mins
|
Direct Numerical Simulations of Turbulent Mixing Layers Between Two Fluids of Large Density Difference
Jon Baltzer, Daniel Livescu
Abstract: In numerous practical applications, shear layers exist between fluids of strongly differing densities. At high Atwood numbers, the large variations in density introduce important effects that have recently been observed in other flows (e.g., Livescu and Ristorcelli, J. Fluid Mech., 605:145–180, 2008). To investigate the inertial variable density effects on the instability growth and structure of mixing layers, we perform very large Direct Numerical Simulations of planar mixing layers between two miscible fluids, each with different density. The DNS domain sizes accommodate large extents of mode pairings, based on the most unstable modes obtained from linear stability analysis. The results display the overall statistical effects on the turbulence and mixing, as well as the structural differences that occur as Atwood number is varied. In particular, significant asymmetries are introduced by the differences in the densities of the mixing layer streams.
|
14:00
15 mins
|
Long-range ordering of turbulent stresses in 2D turbulence
Nicholas Ouellette, Yang Liao
Abstract: We use filter-space techniques to study the geometric alignment of turbulent stresses and strain rates in an experimental quasi-two-dimensional weakly turbulent flow. When these stresses and strains are misaligned, the usual turbulent energy cascade can be suppressed; and more generally, the relative alignment of these two tensors determines the direction of the cascade. We show that as a function of length scale, the turbulent stress undergoes a transition to system-spanning order. However, by exploring analogously defined quantities in a field built from random Fourier modes, we see qualitatively similar behavior, suggesting that at least some of this ordering is purely kinematic. By comparing our results from the experiment and the random field, we highlight the role played by the orientation of the rate of strain tensor in the energy transfer process; additionally, our results allow us to pose several intriguing conjectures.
|
14:15
15 mins
|
WALL TO WALL OPTIMAL TRANSPORT
Charles R. Doering
Abstract: The calculus of variations is employed to find steady divergence-free velocity fields that maximize transport of a tracer between two parallel walls held at fixed concentration for one of two constraints on flow strength: a fixed value of the kinetic energy or a fixed value of the enstrophy (the mean square rate of strain in this situation). The optimizing flows realize upper limits on convective transport in this scenario. We interpret the results in the context of buoyancy-driven Rayleigh–Bénard convection problems that satisfy the flow intensity constraints, enabling us to investigate how optimal transport scalings compare with upper bounds on Nu expressed as a function of the Rayleigh number Ra.
|
14:30
15 mins
|
Effective and anomalous diffusion of inertial particles in flowing fluids
Marco Martins Afonso, Andrea Mazzino
Abstract: We perform an analytical study of the inertial-particle dynamics in the limit of small but finite inertia, in incompressible flows, exploring their diffusion process. By means of a multiscale expansion, we analyse the particle effective diffusivity, and in particular its dependence on relative inertia, Brownian diffusivity, gravity, and particle-to-fluid density ratio (i.e. added mass). We obtain forced advection-diffusion equations for auxiliary quantities in the physical space, thus simplifying the problem from the full phase space to a system which can easily be solved numerically. In the case of parallel flows with power-law velocity spectra, we identify some cases of anomalous diffusion, according to the values of the exponents connected to the possible presence of long-range spatio-temporal correlations.
|
14:45
15 mins
|
JOINT INVESTIGATION OF SETTLING AND PREFERENTIAL CONCENTRATION OF INERTIAL PARTICLES IN TURBULENCE
Sholpan Sumbekova, Alberto Aliseda, Alain Cartellier, Romain Volk, Mickael Bourgoin
Abstract: Particle laden flows are of a great interest in many industrial and natural systems. When the carrier flow is turbulent, a striking feature named preferential concentration takes place: particles denser than the carrier fluid tend to inhomogeneously distribute in space, forming clusters and depleted regions. The present study aims at giving new insight into the possible connection between preferential concentration and particles settling in homogeneous isotropic turbulence. Preferential concentration is quantified using Voronoi diagrams, while 3D Lagrangian Particle Tracking is used to access particles dynamics and in particular statistics of settling velocity. Independent and joint statistics of local concentration and settling velocity will be discussed. Besides, the dependence of particles settling rates on clustering and turbulence levels is further investigated with the use of Laser Phase Doppler Interferometry.
|
15:00
15 mins
|
Passive scalar mixing of a turbulent jet emitted into homogeneous, isotropic turbulence
Alejandro Perez-Alvarado, Laurent Mydlarski, Susan Gaskin
Abstract: Although most jets, whether they be natural or industrial in origin, are emitted into a turbulent environment, almost all previous research on turbulent jets has dealt with jets emitted into quiescent or laminar background flows. The present work extends the work of Khorsandi, Gaskin and Mydlarski, J. Fluid Mech., 2013 – who studied the effect of background turbulence on the velocity field of a turbulent jet emitted into turbulent surroundings – to the study of passive scalar mixing of a jet released into a turbulent flow. To this end, the experiments described herein use planar laser-induced fluorescence to study the mixing of a (high-Schmidt-number) passive scalar within a turbulent jet that is emitted into a quasi-homogeneous, isotropic, zero-mean-flow turbulent background. We examine herein statistics of the jet’s scalar field, and compare them to those of a jet emitted into a quiescent background.
|
15:15
15 mins
|
Entrainment temporal evolution across stably and unstably stratified vapor/clear air interfaces
Luca Gallana, Francesca De Santi, Silvio Di Savino, Renzo Richiardone, Michele Iovieno, Daniela Tordella
Abstract: Warm clouds as stratocumuli swathe a significant part of earth’s surface and play a major role in the global dynamics of atmosphere by strongly reflecting incoming solar radiation so that an accurate representation of their dynamics is important in large-scale analyses of atmoshperic flows [Wood 2012].The mixing and entrainment processes at the cloud top have been identified as fundamental to determine the internal structure of warm clouds, so that a clear and complete understanding of their physics is required [Gerber et al 2013]. The aim of this work is to study some of the basic phenomena which occur at a stratified interface focusing on the smallest scales of the flow which influence. These scales are important to understand the global dynamic of clouds, as pointed out by Malinowski et al (2013). To achieve the results, a campaign of high-resolution simulation of the local transport through a dry/moist air were performed by the means of Direct Numerical Simulations (DNS) using our home produced computational code that implements a de-aliased pseudospectral Fourier-Galerkin spatial discretization and an explicit low storage fourth order Runge-Kutta time integration scheme [Iovieno et al 2001].
We consider the interface between clear air and moist air in a 6m × 6m × 12m parallelepipedic domain coupling two homogeneous and isotropic turbulent regions with different kinetic energy that interact through a mixing layer. The energy ratio is of the same order of the ones measured in warm clouds (see, e.g., [Malinowski et al 2013]) and, furthermore, it allows us to compare our results with experiments on shearless mixing (see [Veeravalli & Warhaft 1989, Tordella & Iovieno 2011]) in absence of any stratification. For each simulation two interfaces have been obtained, one in highly stably stratified condition, and one in unstable condition. The dynamics of interfaces is analyzed through an initial temperature perturbation located across one of the vapor/clear air interfaces thus generating a local stable layer; the water vapor is treated as a passive scalar. The level of stratification is quantified with the Froude number.
For the stable cases, the Froude numbers considered ranges from 12.7 (weak stratification) to 0.6 (intense stratification), while for the unstable cases Fr^2 ranges from -250 to -16. In both stable and unstable cases the evolution of the system can be split in two different phases. In the first one, the buoyancy terms are negligible, and there are no significant differences with respect to a non-stratified case. As the system evolves, the effect of stratification becomes relevant (as soon as the stratification is intense).
About the unstable case layer we observe a high intermittency and an intense growth rate of the layer, which becomes overdiffusive in the case Fr^2 = −16. In particular, the entrainment, after an initial decay, asimptotically always shows a positive growth rate.
Here, for reason of space, we give details about the stably stratified layer which presents a more complex dynamics associated to the onset of a pocket very low turbulent kinetic energy. It can be observed the onset of a sub-layer characterized by the presence of low values of kinetic turbulent energy. At about 8 time scales, we observe the 8% of the energy in the wapor cloud and the 50% of the kinetic energy in the clear-air region. A similar trend was also observed in the LES cloud topped boundary layer simulations carried out by using Deardoff TKE model (NCAR group) and by using the ARAP TKE model (WVU group) [Moeng et al 1996]. The presence of such sublayer induces the formation of two local interfaces. Both of these interfaces present an intermittent behavior, and the entrainment (flux of dry air into the moist one) is blocked; the velocity of the moist air front can be considered a characteristic parameter, since the entrainment of clear air is responsible of the growth of the cloud [Mellado 2010, Moeng 2000]. As a consequence, the entrainment of clear air is confined to a thin interfacial layer. Also the dissipative terms inside the pit becomes relatively more important compared to the kinetic energy, making the pit deeper and deeper with respect to the external regions.
|
|