Computational experiments based on direct numerical simulation of wall-bounded flow reveal that turbulence production can be suppressed by the action of a transverse travelling wave. Flow visualizations show that the near-wall flow structure is altered substantially, compared to other turbulence control techniques, leading to a large amount of shear stress reduction (i.e., more than 30%). The travelling wave can be induced by a spanwise force that is confined within the viscous sublayer, it has its maximum at the wall, and decays exponentially away from it. We demonstrate the robustness of this approach, and its application in salt water using arrays of electro-magnetic tiles that can produce the required travelling wave excitation. We also study corresponding results from spanwise oscillations using a similar force, which also leads to large drag reduction. Although the turbulence statistics for the two approaches are similar, the near-wall structures appear to be different: in the spanwise oscillatory excitation there is a clear presence of wall-streaks whereas in the travelling wave excitation these streaks have disappeared. From the fundamental point of view, the new finding of this work is that appropriate enhancement of the streamwise vortices leads to weakening of the streak intensity, as measured by the normal vorticity component, and correspondingly substantial suppression of turbulence production. From the practical point of view, our findings provide guidance for designing different surface-based actuation techniques including piezoelectric materials, shape memory alloys, and electro-magnetic tiles.

Three-dimensional velocity distributions of a turbulent flow in the core region of a square duct at ReH = 1.2 × 105 are measured using holographic particle image velocimetry (HPIV). Spatial filtering of the 136 × 130 × 128 velocity vector maps enables calculation of subgrid-scale (SGS) stresses and parameters based on the filtered velocity gradients, such as the filtered strain-rate tensor and vorticity vector. Probability density functions (p.d.f.) of scalar parameters characterizing eigenvalue structures confirm that the most probable strain-rate topology is axisymmetric extension, and show that the most probable SGS stress state is axisymmetric contraction. Conditional sampling shows that high positive SGS dissipation occurs preferentially in regions with these preferred strain-rate and stress topologies. High negative SGS dissipation (backscatter) occurs preferentially in regions of axisymmetric contracting SGS stress topology, but is not associated with any particular strain-rate topology. The nonlinear model produces the same trends but tends to overpredict the likelihood of the preferred stress state.

Joint p.d.f.s of relative angles are used to investigate the alignments of the SGS stress eigenvectors relative to the vorticity and eigenvectors associated with eddy viscosity and similarity/nonlinear models. The results show that the most extensive SGS stress eigenvector is preferentially aligned at 32° to the most contracting strain-rate eigenvector. This alignment trend persists, with some variations in angle and peak probability, during conditional samplings based on the SGS dissipation rate, vorticity and strain-rate magnitudes. The relative alignment of the other two stress and strain-rate eigenvectors has a bimodal behaviour with the most contracting and intermediate stress eigenvectors ‘switching places’: from being aligned at 32° to the most extensive strain-rate eigenvector to being parallel to the intermediate strain-rate eigenvector. Conditional sampling shows that one of the alignment configurations occurs preferentially in regions of high vorticity magnitude, whereas the other one dominates in regions where the filtered strain-rate tensor has axisymmetric contracting topology. Analysis of DNS data for isotropic turbulence at lower Re shows similar trends.

Conversely, the measured stress eigenvectors are preferentially aligned with those of the nonlinear model. This alignment persists in various regions of the flow (high vorticity, specific flow topologies, etc). Furthermore, the alignment between the strain-rate and nonlinear model tensors also exhibits a bimodal behaviour, but the alignment angle of both configurations is 42°. Implications of alignment trends on SGS dissipation are explored and conditions for high backscatter are identified based on the orientation of the stress eigenvectors. Several dynamical and kinematical arguments are presented that may explain some of the observed preferred alignments among tensors. These arguments motivate further analysis of the mixed model, which shows good alignment properties owing to the dominance of the Leonard stress on the alignments. Nevertheless, the data also show that the mixed model produces some unrealistic features in probability distributions of SGS dissipation, and unphysical eigenvector alignments in selected subregions of the flow.

We study the steady three-dimensional flow field and bed topography in a channel with sinusoidally varying width, under the assumptions of small-amplitude width variations and sufficiently wide channel to neglect nonlinear effects and sidewall effects. The aim of the work is to investigate the role of width variations in producing channel bifurcation in braided rivers. We infer incipient bifurcation in cases where the growth of a central bar leads to planimetric instability of the channel, i.e. when the given infinitesimal width perturbation is enhanced. Results of the three-dimensional model suggest that the equilibrium bottom profile mainly consists of a purely longitudinal component, uniformly distributed over the cross-section, which induces deposition at the wide section and scour at the constriction, and of a transverse component in the form of a central bar (wide sections) and scour (constrictions), with longitudinal wavelength equal to that of width variations. A comparison between the results of the three-dimensional model and those obtained by means of a two-dimensional depth-averaged approach shows that the transverse component is mainly related to three-dimensional effects. Theoretical findings display a satisfactory agreement with results of flume experiments. Transverse variations are responsible for the planimetric instability of the channel; we find that in the range of values of Shields stress typical of braided rivers, the incipient bifurcation is enhanced as the width ratio of the channel increases.

The singular effects of steady large-scale external strain on the viscous wake generated by a rigid body and the overall flow field are analysed. In an accelerating flow strained at a positive rate, the vorticity field is annihilated owing to positive and negative vorticity either side of the wake centreline diffusing into one another and the volume flux in the wake decreases with downwind distance. Since the wake disappears, the far-field flow changes from monopolar to dipolar. In this case, the force on the body is no longer proportional to the strength of the monopole, but is proportional to the strength of the far field dipole. These results are extended to the case of strained turbulent wakes and this is verified against experimental wind tunnel measurements of Keffer (1965) and Elliott & Townsend (1981) for positive and negative strains. The analysis demonstrates why the total force acting on a body may be estimated by adding the viscous drag and inviscid force due to the irrotational straining field.

Applying the analysis to the wake region of a rigid body or a bubble shows that the wake volume flux decreases even in uniform flows owing to the local straining flow in the near-wake region. While the wake volume flux decreases by a small amount for the flow over streamline and bluff bodies, for the case of a clean bubble the decrease is so large as to render Betz's (1925) drag formula invalid.

To show how these results may be applied to complex flows, the effects of a sequence of positive and negative strains on the wake are considered. The average wake width is much larger than in the absence of a strain field and this leads to diffusion of vorticity between wakes and the cancellation of vorticity. The latter mechanism leads to a net reduction in the volume flux deficit downstream which explains why in calculations of the flow through groups of moving or stationary bodies the wakes of upstream bodies may be ignored even though their drag and lift forces have a significant effect on the overall flow field.

Experimental results are reported on the structure of gravity-driven film flow along an inclined periodic wall with rectangular corrugations. A fluorescence imaging method is used to capture the evolution of film height in space and time with accuracy of a few microns. The steady flow is found to exhibit a statically deformed free surface, as predicted by previous asymptotic and numerical studies. Though usually unstable, its characteristics determine much of the subsequent non-stationary dynamics. Travelling disturbances are observed to evolve into solitary multi-peaked humps, and pronounced differences from the respective phenomena along a flat wall are noted. Finally, a remarkable stabilization of the flow at high Reynolds numbers is documented, which proceeds through the development of a three-dimensional flow structure and leads to a temporary decrease in film thickness and recession of solitary waves.

The instabilities in a sinusoidally oscillating non-separated flow over smooth circular cylinders in the range of Keulegan–Carpenter numbers, K, from about 0.02 to 1 and Stokes numbers, β, from about 103 to 1.4 × 106 have been observed from inception to chaos using several high-speed imagers and laser-induced fluorescence. The instabilities ranged from small quasi-coherent structures, as in Stokes flow over a flat wall (Sarpkaya 1993), to three-dimensional spanwise perturbations because of the centrifugal forces induced by the curvature of the boundary layer (Taylor–Görtler instability). These gave rise to streamwise-oriented counter-rotating vortices or mushroom-shaped coherent structures as K approached the Kh values theoretically predicted by Hall (1984). Further increases in K for a given β led first to complex interactions between the coherent structures and then to chaotic motion. The mapping of the observations led to the delineation of four states of flow in the (K, β)-plane: stable, marginal, unstable, and chaotic.

It is demonstrated that the growth of the mixing zone generated by Rayleigh–Taylor instability can be greatly retarded by the application of rotation, at least for low Atwood number flows for which the Boussinesq approximation is valid. This result is analysed in terms of the effect of the Coriolis force on the vortex rings that propel the bubbles of fluid in the mixing zone.

Linear stability analysis of flow in a channel bounded by wavy walls is considered. It is shown that wall waviness gives rise to an instability that manifests itself through generation of streamwise vortices. The available results suggest that the critical stability criteria based on the Reynolds number based on the amplitude of the waviness can be formulated.

Dissolution of a porous medium creates, under certain conditions, some highly conductive channels called wormholes. The mechanism of propagation is an unstable phenomenon depending on the microscopic properties at the pore scale and is controlled by the injection rate. The aim of this work is to test the ability of a Darcy-scale model to describe the different dissolution regimes and to characterize the influence of the flow parameters on the wormhole development. The numerical approach is validated by model experiments reflecting dissolution processes occurring during acid injection in limestone. Flow and transport macroscopic equations are written under the assumption of local mass non-equilibrium. The coupled system of equations is solved numerically in two dimensions using a finite volume method. Results are discussed in terms of wormhole propagation rate and pore volume injected.

A model for the deformation of thin viscous sheets of arbitrary shape subject to arbitrary loading is presented. The starting point is a scaling analysis based on an analytical solution of the Stokes equations for the flow in a shallow (nearly planar) sheet with constant thickness T0 and principal curvatures k1 and k2, loaded by an harmonic normal stress with wavenumbers q1 and q2 in the directions of principal curvature. Two distinct types of deformation can occur: an ‘inextensional’ (bending) mode when [mid ]L3(k1q22 + k2q21)[mid ] [Lt ] ε, and a ‘membrane’ (stretching) mode when [mid ]L3(k1q22 + k2q21)[mid ] [Gt ] ε, where L ≡ (q21 + q22)−1/2 and ε = T0/L [Lt ] 1. The scales revealed by the shallow-sheet solution together with asympotic expansions in powers of ε are used to reduce the three-dimensional equations for the flow in the sheet to a set of equivalent two-dimensional equations, valid in both the inextensional and membrane limits, for the velocity U of the sheet midsurface. Finally, kinematic evolution equations for the sheet shape (metric and curvature tensors) and thickness are derived. Illustrative numerical solutions of the equations are presented for a variety of buoyancy-driven deformations that exhibit buckling instabilities. A collapsing hemispherical dome with radius L deforms initially in a compressional membrane mode, except in bending boundary layers of width ∼ (εL)1/2 near a clamped equatorial edge, and is unstable to a buckling mode which propagates into the dome from that edge. Buckling instabilities are suppressed by the extensional flow in a sagging inverted dome (pendant drop), which consequently evolves entirely in the membrane mode. A two-dimensional viscous jet falling onto a rigid plate exhibits steady periodic folding, the frequency of which varies with the jet height and extrusion rate in a way similar to that observed experimentally.

In experiments performed aboard NASA's low-gravity KC-135 aircraft, it was found that rapid active control of radial electrostatic stresses can be used to suppress the growth of the (2,0) mode on capillary bridges in air. This mode naturally becomes unstable on a cylindrical bridge when the length exceeds the Rayleigh–Plateau (RP) limit. Capillary bridges having a small amount of electrical conductivity were deployed with a ring electrode concentric with each end of the bridge. A signal produced by optically sensing the shape of the bridge was used to control the electrode potentials so as to counteract the growth of the (2,0) mode. Occasionally the uncontrolled growth of the (3,0) mode was observed when the length far exceeded the RP limit. Rapid breakup from the growth of the (2,0) mode on long bridges was confirmed following deactivation of the control.

A theory is developed for the speed and structure of steady-state non-dissipative gravity currents in rotating channels. The theory is an extension of that of Benjamin (1968) for non-rotating gravity currents, and in a similar way makes use of the steady-state and perfect-fluid (incompressible, inviscid and immiscible) approximations, and supposes the existence of a hydrostatic ‘control point’ in the current some distance away from the nose. The model allows for fully non-hydrostatic and ageostrophic motion in a control volume V ahead of the control point, with the solution being determined by the requirements, consistent with the perfect-fluid approximation, of energy and momentum conservation in V, as expressed by Bernoulli's theorem and a generalized flow-force balance. The governing parameter in the problem, which expresses the strength of the background rotation, is the ratio W = B/R, where B is the channel width and R = (g′H)1/2/f is the internal Rossby radius of deformation based on the total depth of the ambient fluid H. Analytic solutions are determined for the particular case of zero front-relative flow within the gravity current. For each value of W there is a unique non-dissipative two-layer solution, and a non-dissipative one-layer solution which is specified by the value of the wall-depth h0. In the two-layer case, the non-dimensional propagation speed c = cf(g′H)−1/2 increases smoothly from the non-rotating value of 0.5 as W increases, asymptoting to unity for W → ∞. The gravity current separates from the left-hand wall of the channel at W = 0.67 and thereafter has decreasing width. The depth of the current at the right-hand wall, h0, increases, reaching the full depth at W = 1.90, after which point the interface outcrops on both the upper and lower boundaries, with the distance over which the interface slopes being 0.881R. In the one-layer case, the wall-depth based propagation speed Froude number c0 = cf(g′h0)−1/2 = 21/2, as in the non-rotating one-layer case. The current separates from the left-hand wall of the channel at W0 ≡ B/R0 = 2−1/2, and thereafter has width 2−1/2R0, where R0 = (g′h0)1/2/f is the wall-depth based deformation radius.

In this study, the limiting maximum drag-reduction asymptote for the moment coefficient of a rotating disk in a surfactant solution was obtained analytically. The analysis, which was based on the logarithmic velocity profile of turbulent pipe flow in the surfactant solution, was carried out using momentum integral equations of the boundary layer, and the moment coefficient results agreed with experimental results for maximum drag reduction in surfactant solution. Additionally, flow visualization was performed using the tracer and the tuft techniques, which revealed that the direction of flow of surfactant solution on the disk was turned towards the circumferential direction and the amplitude of the circular vortex on the rotating disk was reduced by addition of surfactant solution. The experimental results for flow angle on a rotating disk can be explained well with the analytical results.

The linear stability of the impulsively started flow through a pipe of circular cross-section is studied at high Reynolds number R. A crucial non-dimensional time of O(R7/9) is identified at which the disturbance acquires internal flow characteristics. It is shown that even if the disturbance amplitude at this time is as small as O(R−22/27) the subsequent evolution of the perturbation is nonlinear, although it can still be followed analytically using a multiple-scales approach. The amplitude and wave speed of the nonlinear disturbance are calculated as functions of time and we show that as t → ∞, the disturbance evolves into the long-wave limit of the neutral mode structure found by Smith & Bodonyi in the fully developed Hagen–Poiseuille flow, into which our basic flow ultimately evolves. It is proposed that the critical amplitude found here forms a stability boundary between the decay of linear disturbances and ‘bypass’ transition, in which the fully developed state is never attained.

A rigid-plastic Cosserat model for slow frictional flow of granular materials, proposed by us in an earlier paper, has been used to analyse plane and cylindrical Couette flow. In this model, the hydrodynamic fields of a classical continuum are supplemented by the couple stress and the intrinsic angular velocity fields. The balance of angular momentum, which is satisfied implicitly in a classical continuum, must be enforced in a Cosserat continuum. As a result, the stress tensor could be asymmetric, and the angular velocity of a material point may differ from half the local vorticity. An important consequence of treating the granular medium as a Cosserat continuum is that it incorporates a material length scale in the model, which is absent in frictional models based on a classical continuum. Further, the Cosserat model allows determination of the velocity fields uniquely in viscometric flows, in contrast to classical frictional models. Experiments on viscometric flows of dense, slowly deforming granular materials indicate that shear is confined to a narrow region, usually a few grain diameters thick, while the remaining material is largely undeformed. This feature is captured by the present model, and the velocity profile predicted for cylindrical Couette flow is in good agreement with reported data. When the walls of the Couette cell are smoother than the granular material, the model predicts that the shear layer thickness is independent of the Couette gap H when the latter is large compared to the grain diameter dp. When the walls are of the same roughness as the granular material, the model predicts that the shear layer thickness varies as (H/dp)1/3 (in the limit H/dp [Gt ] 1) for plane shear under gravity and cylindrical Couette flow.

Journal of Fluid Mechanics, vol. 443 (2001), pp. 69–99