The flow in a plane liquid jet from an aperture is obtained by direct simulation of the Navier–Stokes equations. The gas–liquid interface is tracked using the level set method. Flows are calculated for different Reynolds and Weber numbers. When We = ∞, the maximum value of the discharge coefficient appears around Re = O(100). The regions that are vulnerable to cavitation owing to the total stress are identified from calculations based on Navier–Stokes equations and viscous potential flow; the two calculations yield similar results for high Reynolds numbers. We prove that the classical potential flow solution does not give rise to a normal component of the rate of strain at the free streamline. Therefore, the normal component of the irrotational viscous stresses also vanishes and cannot change the shape of the free surface. The results of calculations of flows governed by the Navier–Stokes equations are close to those for viscous potential flow outside the vorticity layers at solid boundaries. The Navier–Stokes solutions for the axisymmetric aperture are also given for two values of Reynolds numbers. The results for axisymmetric and planar apertures are qualitatively similar, but the axisymmetric apertures have a lower discharge coefficient and less contraction.

Nonlinear mode selection, from initial random Gaussian field perturbations, in a model of trailing line vortices (swirling jets), in the breakdown regime, is addressed by direct numerical simulations with a Reynolds number equal to 1000. A new concept of mode activity in the nonlinear evolution is introduced. The selected modes, according to their activities, are reported and related to strain eigenvectors (with maximum eigenvalues) of the basic flow corresponding to the trailing line vortex under consideration. The selected modes are also related to results from the linear eigenmode (exponential growth) instability theory using the concept of dispersion relation envelope. It is found that the global mode hypothesis of the linear eigenmode theory is violated near the flow axis when the swirl number increases. However, far from the flow axis the linear eigenmode theory is in good agreement with the nonlinear evolution in the breakdown regime. The discrepancy between the nonlinear evolution and the linear eigenmode theory is related to the transient growth of optimal perturbations resulting from the non-normality of the linearized Navier–Stokes equations about shear flows. A clear distinction between an eigenmode, an optimal perturbation (non-modal) and a direct numerical simulation (DNS) mode is made. It is shown that the algebraic (transient) growth contributions from the inviscid continuous spectrum could trigger nonlinearities near the flow axis. The DNS mode selected in the nonlinear regime coincides with the long-wave eigenmode benefiting from the algebraic growth in the linear regime. This eigenmode is different from the short-wave eigenmode with the absolute maximum exponential growth. Although it is promoted by transients, in the linear regime, the long-wave component is selected nonlinearly.

Motivated by examples of erosive incision of channels in sand, we investigate the motion of individual grains in a granular bed driven by a laminar fluid to give us new insights into the relationship between hydrodynamic stress and surface granular flow. A closed cell of rectangular cross-section is partially filled with glass beads and a constant fluid flux Q flows through the cell. The refractive indices of the fluid and the glass beads are matched and the cell is illuminated with a laser sheet, allowing us to image individual beads. The bed erodes to a rest height hr which depends on Q. The Shields threshold criterion assumes that the non-dimensional ratio θ of the viscous stress on the bed to the hydrostatic pressure difference across a grain is sufficient to predict the granular flux. Furthermore, the Shields criterion states that the granular flux is non-zero only for θ > θc. We find that the Shields criterion describes the observed relationship hr ∝ Q1/2 when the bed height is offset by approximately half a grain diameter. Introducing this offset in the estimation of θ yields a collapse of the measured Einstein number q* to a power-law function of θ − θc with exponent 1.75 ± 0.25. The dynamics of the bed height relaxation are described well by the power-law relationship between the granular flux and the bed stress.

We simulate the moving contact line in two-dimensional chemically patterned channels using a diffuse-interface model with the generalized Navier boundary condition. The motion of the fluid–fluid interface in confined immiscible two-phase flows is modulated by the chemical pattern on the top and bottom surfaces, leading to a stick–slip behaviour of the contact line. The extra dissipation induced by this oscillatory contact-line motion is significant and increases rapidly with the wettability contrast of the pattern. A critical value of the wettability contrast is identified above which the effect of diffusion becomes important, leading to the interesting behaviour of fluid–fluid interface breaking, with the transport of the non-wetting fluid being assisted and mediated by rapid diffusion through the wetting fluid. Near the critical value, the time-averaged extra dissipation scales as U, the displacement velocity. By decreasing the period of the pattern, we show the solid surface to be characterized by an effective contact angle whose value depends on the material characteristics and composition of the patterned surfaces.

We report high-resolution measurements of the properties of the velocity boundary layer in turbulent thermal convection using the particle image velocimetry (PIV) technique and measurements of the temperature profiles and the thermal boundary layer. Both velocity and temperature measurements were made near the lower conducting plate of a rectangular convection cell using water as the convecting fluid, with the Rayleigh number Ra varying from 109 to 1010 and the Prandtl number Pr fixed at 4.3. From the measured profiles of the horizontal velocity we obtain the viscous boundary layer thickness δυ. It is found that δυ follows the classical Blasius-like laminar boundary layer in the present range of Ra, and it scales with the Reynolds number Re as δυ/H = 0.64Re−0.50±0.03 (where H is the cell height). While the measured viscous shear stress and Reynolds shear stress show that the boundary layer is laminar for Ra < 2.0 × 1010, two independent extrapolations, one based on velocity measurements and the other on velocity and temperature measurements, both indicate that the boundary layer will become turbulent at Ra ~ 1013. Just above the thermal boundary layer but within the mixing zone, the measured temperature r.m.s. profiles σT(z) are found to follow either a power law or a logarithmic behaviour. The power-law fitting may be slightly favoured and its exponent is found to depend on Ra and varies from −0.6 to −0.77, which is much larger than the classical value of −1/3. In the same region, the measured profiles of the r.m.s. vertical velocity σw(z) exhibit a much smaller scaling range and are also consistent with either a power-law or a logarithmic behaviour. The Reynolds number dependence of several wall quantities is also measured directly. These are the wall shear stress τw ~ Re1.55, the viscous sublayer δw ~ Re−0.91, the friction velocity uτ ~ Re0.80, and the skin-friction coefficient cf ~ Re−0.34. All of these scaling properties are very close to those predicted for a classical Blasius-type laminar boundary layer, except that of cf. Similar to classical shear flows, a viscous sublayer is also found to exist in the present system despite the presence of a nested thermal boundary layer. However, velocity profiles normalized by wall units exhibit no obvious logarithmic region, which is likely to be a result of the very limited distance between the edge of the viscous sublayer and the position of the maximum velocity. Compared to traditional shear flows, the peak position of the wall-unit-normalized r.m.s. profiles is found to be closer to the plate (at z+ = z/δw ≃ 5). Our overall conclusion is that a Blasius-type laminar boundary condition is a good approximation for the velocity boundary layer in turbulent thermal convection for the present range of Rayleigh number and Prandtl number.

The instability of a rotating, stratified flow with arbitrary horizontal cross-stream shear is studied, in the context of linear normal modes with along-stream wavenumber k and vertical wavenumber m. A class of solutions are developed which are highly localized in the horizontal cross-stream direction around a particular streamline. A Rayleigh–Schrödinger perturbation analysis is performed, yielding asymptotic series for the frequency and structure of these solutions in terms of k and m. The accuracy of the approximation improves as the vertical wavenumber increases, and typically also as the along-stream wavenumber decreases. This is shown to correspond to a near-inertial limit, in which the solutions are localized around the global minimum of fQ, where f is the Coriolis parameter and Q is the vertical component of the absolute vorticity. The limiting solutions are near-inertial waves or inertial instabilities, according to whether the minimum value of fQ is positive or negative.

We focus on the latter case, and investigate how the growth rate and structure of the solutions changes with m and k. Moving away from the inertial limit, we show that the growth rate always decreases, as the inertial balance is broken by a stabilizing cross-stream pressure gradient. We argue that these solutions should be described as non-symmetric inertial instabilities, even though their spatial structure is quite different to that of the symmetric inertial instabilities obtained when k is equal to zero.

We use the analytical results to predict the growth rates and phase speeds for the inertial instability of some simple shear flows. By comparing with results obtained numerically, it is shown that accurate predictions are obtained by using the first two or three terms of the perturbation expansion, even for relatively small values of the vertical wavenumber. Limiting expressions for the growth rate and phase speed are given explicitly for non-zero k, for both a hyperbolic-tangent velocity profile on an f-plane, and a uniform shear flow on an equatorial β-plane.

The homogenization of a heterogeneous mixture of two pure fluids with different densities by molecular diffusion and stirring induced by buoyancy-generated motions, as occurs in the Rayleigh–Taylor (RT) instability, is studied using direct numerical simulations. The Schmidt number, Sc, is varied by a factor of 20, 0.1 ≤ Sc ≤ 2.0, and the Atwood number, A, by a factor of 10, 0.05 ≤ A ≤ 0.5. Initial-density intensities are as high as 50% of the mean density. As a consequence of differential accelerations experienced by the two fluids, substantial and important differences between the mixing in a variable-density flow, as compared to the Boussinesq approximation, are observed. In short, the pure heavy fluid mixes more slowly than the pure light fluid: an initially symmetric double delta density probability density function (PDF) is rapidly skewed and, only at long times and low density fluctuations, does it relax to a Gaussian-like PDF. The heavy–light fluid mixing process asymmetry is relevant to the nature of molecular mixing on different sides of a high-Atwood-number RT layer. Diverse mix metrics are used to examine the homogenization of the two fluids. The conventional mix parameter, θ, is mathematically related to the variance of the excess reactant of a hypothetical fast chemical reaction. Bounds relating θ and the normalized product, Ξ, are derived. It is shown that θ underpredicts the mixing, as compared to Ξ, in the central regions of an RT layer; in the edge regions, θ is larger than Ξ. The shape of the density PDF cannot be inferred from the usual mix metrics popular in applications. For example, when θ, Ξ ≥ 0.6, characteristic of the interior of a fully developed RT layer, the PDFs can have vastly different shapes. Bounds on the fluid composition using two low-order moments of the density PDF are derived. The bounds can be used as realizability conditions for low-dimensional models. For the measures studied, the tightest bounds are obtained using Ξ and mean density. The structure of the flow is also examined. It is found that, at early times, the buoyancy production term in the spectral kinetic energy equation is important at all wavenumbers and leads to anisotropy at all scales of motion. At later times, the anisotropy is confined to the largest and smallest scales: the intermediate scales are more isotropic than the small scales. In the viscous range, there is a cancellation between the viscous and nonlinear effects, and the buoyancy production leads to a persistent small-scale anisotropy.

This paper concerns the manufacture of non-axisymmetric capillary tubing via the Vello process, in which molten glass is fed through a die and drawn off vertically. The shape of the cross-section evolves under surface tension as it flows downstream. The aim is to achieve a given desired final shape, typically square or rectangular, and our goal is to determine the required die shape.

We use the result that, provided the tube is slowly varying in the axial direction, each cross-section evolves like a two-dimensional Stokes flow when expressed in suitably scaled Lagrangian coordinates. This allows us to use a previously derived model for the surface-tension-driven evolution of a thin two-dimensional viscous tube. We thus obtain, and solve analytically, equations governing the axial velocity, thickness and circumference of the tube, as well as its shape. The model is extended to include non-isothermal effects.

The deformation of an elastic micro-capsule in an infinite shear flow is studied numerically using a spectral method. The shape of the capsule and the hydrodynamic flow field are expanded into smooth basis functions. Analytic expressions for the derivative of the basis functions permit the evaluation of elastic and hydrodynamic stresses and bending forces at specified grid points in the membrane. Compared to methods employing a triangulation scheme, this method has the advantage that the resulting capsule shapes are automatically smooth, and few modes are needed to describe the deformation accurately. Computations are performed for capsules with both spherical and ellipsoidal unstressed reference shape. Results for small deformations of initially spherical capsules coincide with analytic predictions. For initially ellipsoidal capsules, recent approximate theories predict stable oscillations of the tank-treading inclination angle, and a transition to tumbling at low shear rate. Both phenomena have also been observed experimentally. Using our numerical approach we can reproduce both the oscillations and the transition to tumbling. The full phase diagram for varying shear rate and viscosity ratio is explored. While the numerically obtained phase diagram qualitatively agrees with the theory, intermittent behaviour could not be observed within our simulation time. Our results suggest that initial tumbling motion is only transient in this region of the phase diagram.

It has recently been shown that the instability characteristics of planar jets and wakes change when the flows are confined between two flat plates. This is due to constructive interaction between modes with zero group velocity in the inner and outer flows. In this theoretical study, a linear spatio-temporal analysis is performed on unconfined and confined round jets and wakes in order to discover whether the same effect is observed. There are several similarities between the planar case and the round case as well as some significant differences. Nevertheless, the effect of confinement on round flows is found to be very similar to that on planar flows and to act via the same physical mechanism. This paper examines density ratios from 0.001 to 1000 and has important implications for the design of fuel injectors, which often employ confined shear flows at high Reynolds numbers and large density ratios to generate strong mixing in combustion chambers.

Computation of viscous flow over a circular disk of aspect ratio 10 (thickness/diameter) in the Reynolds number (Re) range of 10 to 300 was performed. The following flow regimes were observed: (I) steady axisymmetric flow when Re < 135, with the presence of a toroidal vortex behind the disk; (II) regular bifurcation with loss of azimuthal symmetry but with planar symmetry and a double-threaded wake, for 135 ≤ Re < 155; (III) three-dimensional flow with periodic shedding of double-sided hairpin-shaped vortex structures and periodic motion of the separation region for 155 ≤ Re < 172; (IV) regular shedding of double-sided hairpin-shaped vortex structures with planar and spatio-temporal symmetry for 172 ≤ Re < 280; (V) periodic three-dimensional flow with irregular rotation of the separation region when Re = 280–300. This transition process for the disk differs from that for the sphere as we observe a loss of the symmetry plane in Regime III due to a twisting motion of the axial vorticity strands in the wake of the disk. The periodic flow was characterized by double-sided hairpin structures, unlike the one-sided vortex loops observed for the sphere. This resulted in the drag coefficient oscillating at twice the frequency of the axial velocity. In Regime IV, the vortex loops were shed from diametrically opposite locations and with equal strength, resulting in the lift coefficient oscillating symmetrically about a zero mean. These results imply the presence of spatio-temporal symmetry.

The flow pattern generated by a sphere accelerated from rest by a small constant applied forceshows scaling behaviour at long times, as can be shown from the solution of the linearized Navier–Stokes equations. In the scaling regime the kinetic energy of the flow grows with thesquare root of time. For two distant settling spheres starting from rest the kinetic energy ofthe flow depends on the distance vector between centres; owing to interference of the flowpatterns. It is argued that this leads to relative motion of the two spheres. Thecorresponding interaction energy is calculated explicitly in the scaling regime.

A condition is derived for the hydraulic criticality of a 2-layer flow with transverse variations in both layer velocities and thicknesses. The condition can be expressed in terms of a generalized composite Froude number. The derivation can be extended in order to obtain a critical condition for an N-layer system. The results apply to inviscid flows subject to the usual hydraulic approximation of gradual variations along the channel and is restricted to flows in which the velocity remains single-signed within any given layer. For an intermediate layer with a partial segment of sluggish flow, the long-wave dynamics of the overlying and underlying layers become decoupled.

We consider laminar displacement flows in narrow eccentric annuli, oriented horizontally, between two fluids of Herschel–Bulkley type, (i.e. including Newtonian, power-law and Bingham models). This situation is modelled via a Hele-Shaw approach. Whereas slumping and stratification would be expected in the absence of any imposed flow rate, for a displacement flow we show that there are often steady-state travelling wave solutions in this displacement. These may exist even at large eccentricities and for large density differences between the fluids. When heavy fluids displace light fluids, annular eccentricity opposes buoyancy and steady states are more prevalent than when light fluids displace heavy fluids. For large ratios of buoyancy forces to viscous forces we derive a lubrication-style displacement model. This simplification allows us to find necessary and sufficient conditions under which a displacement can be steady, which can be expressed conveniently in terms of a consistency ratio. It is interesting that buoyancy does not appear in the critical conditions for a horizontal well. Instead a competition between fluid rheologies and eccentricity is the determining factor. Buoyancy acts only to determine the axial length of the steady-state profile.

Numerical simulations of laser energy deposition in air are conducted. Local thermodynamic equilibrium conditions are assumed to apply. Variation of the thermodynamic and transport properties with temperature and pressure are accounted for. The flow field is classified into three phases: shock formation; shock propagation; and subsequent collapse of the plasma core. Each phase is studied in detail. Vorticity generation in the flow is described for short and long times. At short times, vorticity is found to be generated by baroclinic means. At longer times, a reverse flow is found to be generated along the plasma axis resulting in the rolling up of the flow field near the plasma core and enhancement of the vorticity field. Scaling analysis is performed for different amounts of laser energy deposited and different Reynolds numbers of the flow. Simulations are conducted using three different models for air based on different levels of physical complexity. The impact of these models on the evolution of the flow field is discussed.

In order to investigate the physical mechanism of the energy cascade in homogeneous isotropic turbulence, the internal energy and its transfer rate are defined as a function of scale, space and time. Direct numerical simulation of turbulence at a moderate Reynolds number verifies that the energy cascade can be caused by the successive creation of smaller-scale tubular vortices in the larger-scale straining regions existing between pairs of larger-scale tubular vortices. Movies are available with the online version of the paper.

We have performed self-consistent computations of the interactions between a superfluid vortex-ring and a solid particle for two different vortex-ring sizes and over a wide range of temperatures. In all cases, the particle and the vortex eventually separate. For temperature T = 0 K, larger rings tend to trap the particle more effectively than smaller rings. Trying to escape the vortex, the particle follows a spiralling trajectory that could be experimentally detected. The dominant dynamical process is the excitation and propagation of Kelvin waves along the vortices. For T > 0 K, particle–vortex collision induces particle vibrations that are normal to the particle's direction of motion and might be experimentally detectable. In contrast to the T = 0 K case, smaller rings induce larger particle oscillation velocities. With increasing temperature, enhanced mutual friction damping of Kelvin waves leads to the damping of both the intensity and frequency of post-collision particle vibrations. Moreover, higher temperatures increase the relative impact of the Stokes drag force on particle motion.

This paper is concerned with the stability of fibre coatings at large Reynolds numbers. Both single- and double-layer coatings are considered; no restriction is placed on the coating thicknesses. Calculations for the maximum growth rate, together with the corresponding length scale of the instability, are presented. Rescaling with respect to the maximum growth rate generates universal dispersion relations over the unstable wavenumber range. For double-layer composite coatings, modifications are required when the density ratio becomes large.

Acoustic resonances of simple three-dimensional finite-length structures in an infinitely long cylindrical pipe are investigated numerically by solving an eigenvalue problem. To avoid unphysical reflections at the finite grid boundaries placed in the uniform cross-sections of the pipe, perfectly matched layer absorbing boundary conditions are applied in the form of the complex scaling method of atomic and molecular physics. Examples of the structures investigated are sound-hard spheres, cylinders, cavities and closed side branches. Several truly trapped modes with zero radiation loss are identified for frequencies below the first cutoff frequency of the pipe. Such trapped modes can be excited aerodynamically by coherent vortices if the frequency of the shed vortices is close to a resonant frequency. Furthermore, numerical evidence is presented for the existence of isolated embedded trapped modes for annular cavities above the first cutoff frequency and for closed side branches below the first cutoff frequency. As applications of engineering interest, the acoustic resonances are computed for a ball-type valve and around a simple model of a high-speed train in an infinitely long tunnel.

The stability behaviour of a low-Reynolds-number recirculation flow developing in a curved channel is investigated using a global formulation of hydrodynamic stability theory. Both the resonator and amplifier dynamics are investigated. The resonator dynamics, which results from the ability of the flow to self-sustain perturbations, is studied through a modal stability analysis. In agreement with the literature, the flow becomes globally unstable via a three-dimensional stationary mode. The amplifier dynamics, which is characterized by the ability of the flow to exhibit large transient amplifications of initial perturbations, is studied by looking for the two- and three-dimensional initial perturbations that maximize the energy gain over a given time horizon. The optimal initial two-dimensional perturbations have the form of wave packets localized in the upstream part of the recirculation bubble. It is shown that they are first amplified while travelling downstream along the shear layer of the recirculation bubble and then decay when leaving the recirculation bubble. Maximal energy gain is thus achieved for a time horizon approximately corresponding to the propagation of the wave packet along the whole recirculation bubble. The resonator and amplifier dynamics are associated with different types of structures in the flow: three-dimensional steady structures for the resonator dynamics and nearly two-dimensional unsteady structures for the amplifier dynamics. A comparison of the strength of the two dynamics is proposed. The transient energetic growth of the two-dimensional unsteady perturbations is large at moderate time, compared to the very weak exponential growth of the three-dimensional stationary mode. This suggests that, as soon as there is noise in the system, the amplifier dynamics dominates the resonator dynamics, thus explaining the appearance of unsteadiness rather than the emergence of stationary structures in similar experimental flows.