Session A28: Turbulence: Interfaces

Experimental investigation of entrainment processes of a turbulent jet

We implemented simultaneous, time-resolved, multi-scale-Particle Image Velocimetry (PIV) and Planar Laser-Induced Fluorescence (PLIF) to study entrainment processes in the far-field of a round, turbulent jet. The experiments were performed using water as the test medium and a passive dye with a Schmidt number $Sc \gg 1$ to identify the turbulent/non-turbulent (T/N-T) interface of the jet. The Reynolds number based on the nozzle exit is $Re = 25,300$ and is considerably higher than existing studies of entrainment in jets. Independent 2D PIV and PLIF measurements confirmed that the far-field flow characteristics agree well with the classical scaling laws of turbulent jets. We use the auto-correlation of entrainment velocity along the T/N-T interface to show that the interface is dominated by fluid motion of $\mathcal{O}(\lambda)$. We also show that there exists a balance between the mass flux across the interface calculated at the small-scales and the mass flux calculated at larger scales. The interface is more convoluted at smaller scales, which results in a larger interfacial surface area. The mass-flux balance therefore indicates that the entrainment velocity at the interface scales at a rate that is inversely proportional to the surface area.   

Effect of free stream turbulence on the entrainment characteristics of jets

Direct numerical simulations of turbulent planar jets are used to analyze the effects of free stream turbulence on the entrainment characteristics and enstrophy dynamics near the turbulent/turbulent interface (TTI) that separates strong turbulence (inside the jet shear layer) from weaker turbulence outside of the jet. The higher the integral scales and turbulence intensities in the free stream the more effects it has on the jet shear layer, and for strong free stream turbulence the viscous superlayer is absent from the jet edges.   

Flow topology inside the interface layer at the edge of a turbulent jet

The invariants of the velocity gradient tensor are analysed in the viscous superlayer and turbulent sublayer within the sharp interface layer that exists at the edges of wakes, jets, mixing layers and boundary layers (turbulent/non-turbulent interface). The invariants display marked differences depending on the characteristics of each one of these layers, and show the imprint of the small scale eddies from the nearby turbulent region.   

Flow Dynamics Near the Turbulent/Non-Turbulent Interface in Compressible Shear Layers

Direct numerical simulation (DNS) of compressible turbulent shear layers at varying convective Mach numbers are used to assess the flow dynamics in proximity of the turbulent/non-turbulent interface (TNTI) separating the turbulent and the irrotational regions. This interface is identified by using a certain threshold for the vorticity norm. For both incompressible and compressible mixing layers, the TNTI layer thickness is found to be approximately one Taylor length scale. The conditional flow statistics based on the normal distance from the TNTI are compared for different convective Mach numbers. The terms in total kinetic energy, turbulent kinetic energy, and vorticity transport equations are examined in order to determine the effects of compressibility on the transport mechanisms across the TNTI. For all the convective Mach numbers, different terms in these equations are scaled with the Taylor length and velocity scales in interface coordinates. It is observed that for compressible cases, the intense vortical structures (IVS) generate a baroclinic torque as they become close to the TNTI.   

Deformation of the turbulent/non-turbulent interface by large-scale motions in boundary layers

The relationship between large-scale motions (LSMs) and the shape of the turbulent/non-turbulent interface (TNTI) is examined using data from direct numerical simulation (DNS) of turbulent boundary layer (TBL) flow. The Reynolds number based on the momentum thickness and the free-stream velocity ranges from 1180 to 3500. Feature extraction techniques are used to identify cores of the large-scale motions in the perturbation fields. Since turbulence kinetic energy and enstrophy are different inside low- and high-speed LSMs, the wall-normal elevation of TNTI is correlated with the streamwise momentum of these structures. The large-scale crests and troughs of TNTI are matched to the locations of the wall-detached LSMs of low- and high-speed streaks, respectively. In addition, abrupt changes in turbulence statistics across the TNTI reported in previous studies are associated with population trends of the wall-detached LSMs near the TNTI.   

Some Characteristics of Entrainment in a Compressible Turbulent Mixing Layer

The results of direct numerical simulation (DNS) of temporally evolving compressible mixing layer are used to study the entrainment process across the turbulent/non-turbulent interface (TNTI) separating the turbulent and the irrotational regions. This interface is detected by using a certain threshold for the vorticity norm. The compressible form of the conservation equations for mass, momentum, energy, and conserved scalar are solved. The local entrainment velocity is calculated using the enstrophy transport equation. Conditional averages of the terms in this equation across TNTI are examined in order to gain a better understanding of the physical mechanisms contributing to entrainment. The entrainment process in turbulent flows can be associated with two different mechanisms. Nibbling, which is related with small scale motions, and engulfment, which is mostly due to large scale motions. The role of each mechanism is examined. The local entrainment velocity is also decomposed into an inviscid and a viscous part, and the contribution of each part is evaluated. The role of compressibility on the entrainment process is also studied.   

Local Flow Topology in Compressible Turbulent Shear Layers

The local flow topology is studied using the invariants of the velocity gradient tensor in compressible turbulent mixing layer via direct numerical simulation (DNS). The topological behavior of the flow is analyzed in two different regions: in proximity of the turbulent/non-turbulent interface (TNTI), and inside the turbulent region. The occurrence probability of different flow topologies conditioned by the dilatation level is presented and it is shown that the structures in the locally compressed regions tend to have stable topologies while in locally expanded regions the unstable topologies are prevalent. It is found that the distribution of various flow topologies in regions close to the TNTI differs from inside the turbulent region, and in these regions the most probable topologies are non-focal. At the distances farther than one Taylor microscale from the TNTI, the probability of various topologies is almost constant, and is equal to the values obtained for turbulent region in the mixing layer.   

On the entrainment dynamics of inergodic, non-stationary flows

Entrainment is typically studied through the conditional averaging along the turbulent non-turbulent interface (TNTI) of ergodic flows. However, this method is unsuitable for inergodic, non-stationary flows, as the TNTI is non-similar at different points in space and time. To understand how a TNTI's mean time dependence effects entrainment, the current study investigates the transport of irrotational fluid into a vortex forming behind an accelerating plate. The plate accelerates to a final velocity within a full-, half- and quarter-chord tow. Phase-averaged, planar, particle tracking velocimetry data is acquired and the forward finite-time Lyapunov exponent and vorticity fields are used to identify the TNTI. The TNTI is then represented by a contour, which is used to approximate the entrainment rate and investigate the transport mechanisms across the TNTI. Early results show that increasing acceleration suppresses vortex growth and entrainment. We hypothesize that shear-layer structure is integral to entrainment by altering the feeding rate of rotational fluid and the TNTI's convexity. The hypothesis is tested by altering plate-edge geometry and by varying the final chord-based Reynolds number from 5000 to 20 000.   

Interfacial phenomena in turbulent magnetohydrodynamic channel flows at low magnetic Reynolds number

Direct numerical simulations (DNS) are performed to examine whether interfacial phenomena can be observed in magnetohydrodynamic (MHD) turbulent channel flows under the influence of imposed magnetic field. The magnetic Reynolds number is assumed to be sufficiently low such that the quasi-static approximation can be applied. For high Hartmann number, the visualization of the vorticity field reveals flow structures consisting of turbulent boundary layers (TBL) near the walls and a quiescent channel core. The statistical analysis of the DNS data shows that physical quantities, such as the spanwise vorticity and streamwise velocity, possess sharp gradients at the edges of the TBL. The features of the sharp gradients are qualitatively similar to those of the turbulent/non-turbulent interfaces which have been observed in hydrodynamic turbulent shear flows, e.g. turbulent boundary layers, wakes and jets. The Joule dissipation rate, which is a characteristic quantity in MHD flow, is shown to have sharp gradients at the edges of the TBL. The average height of the edges is theoretically estimated and the estimation is assessed using the DNS results.