R. Delfos

Decay of dipolar vortex structures in a stratified fluid

In this paper the viscous decay of dipolar vortex structures in a linearly stratified fluid is investigated experimentally, and a comparison of the experimental results with simple theoretical models is made. The dipoles are generated by a pulsed horizontal injection of fluid. In a related experimental study by Flor and van Heijst [J. Fluid Mech. 279, 101 (1994)], it was shown that, after the emergence of the pancake‐shaped vortex structure, the flow is quasi‐two‐dimensional and decays due to the vertical diffusion of vorticity and entrainment of ambient irrotational fluid. This results in an expansion of the vortex structure. Two decay models with the horizontal flow based on the viscously decaying Lamb–Chaplygin dipole, are presented. In a first model, the thickness and radius of the dipole are assumed constant, and in a second model also the increasing thickness of the vortex structure is taken into account. The models are compared with experimental data obtained from flow visualizations and from digit...

Characteristics of the turbulent/non-turbulent interface of a non-isothermal jet

The turbulent/non-turbulent interface of a jet is characterized by sharp jumps (‘discontinuities’) in the conditional flow statistics relative to the interface. Experiments were carried out to measure the conditional flow statistics for a non-isothermal jet, i.e. a cooled jet. These experiments are complementary to previous experiments on an isothermal Re=2000 jet, where, in the present experiments on a non-isothermal jet, the thermal diffusivity is intermediate to the diffusivity of momentum and the diffusivity of mass. The experimental method is a combined laser-induced fluorescence/particle image velocimetry method, where a temperature-sensitive fluorescent dye (rhodamine 6G) is used to measure the instantaneous temperature fluctuations. The results show that the cooled jet can be considered to behave like a self-similar jet without any significant buoyancy effects. The detection of the interface is based on the instantaneous temperature, and provides a reliable means to detect the interface. Conditional flow statistics reveal the superlayer jump in the conditional vorticity and in the temperature.

Flows Through Real Porous Media: X-Ray Computed Tomography, Experiments, and Numerical Simulations

Two different direct-forcing immersed boundary methods (IBMs) were applied for the purpose of simulating slow flow through a real porous medium: the volume penalization IBM and the stress IBM. The porous medium was a random close packing of about 9000 glass beads in a round tube. The packing geometry was determined from an X-ray computed tomography (CT) scan in terms of the distribution of the truncated solid volume fraction (either 0 or 1) on a three-dimensional Cartesian grid. The scan resolution corresponded to 19.3 grid cells over the mean bead diameter. A facility was built to experimentally determine the permeability of the packing. Numerical simulations were performed for the same packing based on the CT scan data. For both IBMs the numerically determined permeability based on the Richardson extrapolation was just 10% lower than the experimentally found value. As expected, at finite grid resolution the stress IBM appeared to be the most accurate IBM. [DOI: 10.1115/1.4025311]

Drag reduction by surface treatment in turbulent Taylor-Couette flow

We use a Taylor-Couette facility to study the drag reducing effects of commercial surface products at high shear Reynolds numbers (Res) under perfect couter-rotating conditions (riwi=rowo). The correlation between torque contribution of the von Karman flow and shear Reynolds number is investigated. At this moment no significant drag changes are found for the commercial products. However, further research is needed to exclude uncertainties and errors from the torque measurements.

Measurements of liquid film thickness for a droplet at a two-fluid interface

Coalescence of a droplet at a two-fluid interface is studied at Bond numbers larger than one and at three different values of the viscosity ratio. Both the thickness of the liquid film between the rising droplet and the two-fluid interface, and the location of film rupture are measured using laser induced fluorescence. Particle image velocimetry was applied to the flow in the film. It is found that the film thins asymetrically, and that the time interval between collision and film rupture is shorter than predicted by commonly used models. The film ruptures at an off-center location. It can be concluded that asymmetric film drainage speeds up coalescence.

Experimental studies of liquid-liquid dispersion in a turbulent shear flow

We study liquid-liquid dispersions in a turbulent Taylor - Couette flow, produced between two counterrotating coaxial cylinders. In pure Water and in counterrotation, Reynolds numbers up to 1.4 10^5 are reached. We first characterize the single-phase flow, in terms of threshold for transition to turbulence, scaling of the torque and measurements of the mean flow and of the Reynolds stress by stereoscopic PIV. We then study the increase of the dissipation in the two-phase flows and find that the torque per unit mass can be twice the torque for a single-phase flow. Long-time behaviours are also reported.

Experimental investigation of torque scaling and coherent structures in turbulent Taylor–Couette flow

The effect of flow structures to the torque values of fully turbulent Taylor-Couette flow was experimentally studied using tomographic PIV. The measurements were performed for various relative cylinder rotation speeds and Reynolds numbers, based on a study of Ravelet et al. (2010). We confirmed that the flow structures are strongly influenced by the rotation number. Our analyses using time-averaged mean flow showed the presence of Taylor vortices for the two smallest rotation numbers that were studied. Increasing the rotation number initially resulted in the shape deformation of the Taylor vortices. Further increment towards only outer cylinder rotation, showed transition to the dominance of the small scale vortices and absence of Taylor vortex-like structures. We compared the transition of the flow structures with the curves of dimensionless torque. Sudden changes of the flow structures confirmed the presence of transition points on the torque curve, where the dominance of small and large scale vortical structures on the mean flow interchanges.

Scaling of torque in turbulent Taylor-Couette flow with background rotation

We did an experimental study on Taylor-Couette flow in between two coaxial cylinders of length L = 220 mm and radii ri = 110 and ro = 120, respectively, the fluid-filled ‘Taylor-Couette gap’ or TC-gap being h = 10 mm, thus gap ratio η = ri/ro = 0.917, and gap aspect ratio L/h = 22). Both cylinders are rotating independently, with angular frequencies ωi,o, The torque T on the inner cylinder is measured through the axis driving the inner cylinder with a co-rotating torque meter.

Dispersion of bubbles in fully developed channel flow

Dispersion and preferential concentration of small, low Stokes number bubbles in horizontal turbulent channel flow is studied by DNS and experiments. A DNS of turbulent channel flow at Re = 360 with Lagrangian tracking of one-way coupled bubbles (d < n, St = 1.3×10?3) shows that equilibrium bubble concentration profiles can be described by a gradient diffusion hypothesis in analogy to flows with suspended sediment as studied by Rouse (1937). The conditionally averaged flow around the bubbles is measured by simultaneous PIV and bubble shadowgraphy and confirms the finding of the DNS that bubbles are preferentially concentrated in large-scale downward flowing fluid regions, which compensates for the rise velocity of the bubbles. This clustering is not an inertia effect, but results from the combination of a concentration gradient and turbulent mixing.

Measurement of the recoalescence flux into the rear of a Taylor bubble

Most of the theoretical models on vertical slug flow assume the mass balance of a Taylor bubble to depend only on the incoming gas flux at the top of the Taylor bubble and on the outgoing entrainment flux at the bottom. This means that the recoalescence flux, which is defined as the fraction of the entrainment flux that coalesces back into the bubble, is neglected. Only in Fernandes et al. [AIChE J. 29, 981 (1983)] is a model proposed for this recoalescence flux but their model has never been verified by measurements. Therefore, we set out in the present research to measure and quantify the recoalescence flux. Our experiments have been carried out in a recirculating flow facility with a vertical cylindrical test section with inner diameter Di=100 mm. In this test section a Taylor bubble is kept at a fixed vertical position by a constant downward liquid flow ΨL. A continuous stream of small helium bubbles is injected into the wake of the Taylor bubble. The recoalescence flux is then determined by measuring...