In active materials, turbulent flows emerge spontaneously from collective motion, yet identifying meaningful signatures of their transitions remains challenging. Here, we use an information-theoretic divergence measure, based on computable information density (CID), to reveal a robust activity threshold that marks a qualitative shift in defect dynamics, flow correlations, and mixing behavior. This threshold offers a new spatiotemporal order parameter for characterizing nonequilibrium transitions in active systems. Our findings open avenues for linking information flow to physical organization in living materials.

Active particles often navigate through inhomogeneous environments. Here, we analyze the dynamics of active particles in inhomogeneous viscoelastic fluids and demonstrate that spatial variations in fluid relaxation time give rise to a novel mechanism of taxis, which we refer to as a form of durotaxis in fluids.

We introduce a method to stabilize large-eddy simulations of compressible, vortex-dominated flows using a minimal yet sufficient amount of artificial diffusivity. By restraining additional shear viscosity to unstable low-pressure vortex cores, numerical stability is preserved without smearing key turbulent flow features. This strategy enables accurate, stable, and cost-effective large eddy simulations of complex flows such as bluff-body wakes and separation regions.

Since David Ruelle’s 1979 estimate, the maximal Lyapunov exponent of turbulence was thought to scale with the inverse of the smallest Lagrangian time-scale (the Kolmogorov time-scale). It actually also depends on the smallest Eulerian time-scale via random sweeping of small-scale uncertainty fluctuations. This leads to a sweeping relation which involves the integral length scale of the uncertainty field which, in turn, tends towards sub-Kolmogorov scales with increasing Reynolds number. The resulting maximal Lyapunov exponent scales with the Taylor length time scale divided by the square of the Kolmogorov time scale.

Fluid-structure interactions are central to many biological and engineering systems but remain challenging to model. This study combines precision experiments and simulations to explore how an elastic tube responds to pulsatile flow. At small forcing, the system behaves like a damped harmonic oscillator, but at higher amplitudes, complex nonlinear and asymmetric deformations emerge, including tube collapse. These findings offer new insights for advancing simulation models of elastic vessels.

How do non-Newtonian shear-thickening fluids behave under turbulent conditions? Using direct numerical simulations, this study uncovers how increasing shear-thickening (Carreau number, Cu) progressively alters flow dynamics, reshaping near-wall structures and disrupting the self-sustaining processes of turbulence. Remarkably, even in regimes where Newtonian flows would relaminarize, these fluids sustain unsteady motion, exposing a complex interplay between rheology and turbulence.

We have examined the validity and/or applicability of the usual Boussinesq approximation to ideal gas flows in light of the kinetic theory of gases. Asymptotic analysis for small Knudsen numbers reveals that there are missing terms in the usual Boussinesq approximation. Our results support the so-called thermodynamic Boussinesq approximation rather than the usual Boussinesq approximation. The impact of the missing terms has been presented numerically as well in the thermal convection problem.

A laminar duct flow is considered based on the kinetic theory of gases. An asymptotic analysis for small Knudsen numbers reveals that the thermal stress, which is missing in the Navier-Stokes equation, must be included in the momentum balance in the duct’s cross-sectional directions. This stress arises from the nonuniform temperature field developed by viscous dissipation of the main axial flow. It induces a slow secondary flow in the cross-sectional plane, which, through convection, produces finite effects on the axial velocity and temperature fields.

Rayleigh-Bénard convection in liquid metals under a strong magnetic field is a fundamental physical process in nature and engineering, yet its complex dynamics when subjected to external shear are not well understood. Through direct numerical simulations, this study reveals three distinct flow regimes governed by the competition between shear and buoyancy, observing unique phenomena such as “vortex reconnection” and “vortex proliferation.” Counterintuitively, under very strong shear, the flow does not become more chaotic but instead reorganizes into a stable large-scale structure, significantly enhancing heat transfer efficiency.

A charge-neutral drop in a uniform DC electric field does not migrate in free space — but a nearby insulating wall breaks this symmetry. We show that the wall induces lateral migration, with direction and magnitude governed by the conductivity and permittivity ratios (R and P) of the drop to the medium. For R/P < 1, electrohydrodynamic flow repels the drop from the wall; for R/P > 1, it attracts. Experiments and boundary integral simulations quantify this lift force and reveal that the dominant effect is the long-range flow from the image stresslet.

A carefully designed wave form orchestrates sequential 2-fiber interactions on a soft substrate planted with a fiber array. Between identical neighboring fibers, a droplet goes through a strongly asymmetrical breakup as the wave pushes across, ensuring a near-complete liquid transfer from one fiber to the next in the direction of wave propagation.

It has been reported in previous studies that a highly viscous droplet could not bounce on non-wetting solid surfaces. In this work, we promote viscous droplet bouncing by replacing the non-wetting solid surface with a curved soap film. The highly viscous droplet that cannot rebound from the solid surface is observed to easily bounce off the curved soap film. The bouncing promotion mechanism can be elucidated by a spring-mass-damper system in which the viscous droplet and the curved soap film are modeled in analogy with a damper and a spring connected in series.

The yield stress of a plastic matrix embedding bubbles can give rise to distinct coarsening regimes, depending on the balance between plastic and capillary forces. These include: (i) classical coarsening, with bubble growth similar to that in simple liquids; (ii) damped coarsening, where growth slows progressively; and (iii) complete arrest of coarsening once a critical threshold is exceeded.

This experimental investigation examines water droplet impacts on hydrophobic or super-hydrophobic surfaces across moderate Weber numbers (3 ≤ We ≤ 120) and size ratios (1.04 ≤ Ω ≤ 2.08). Through high-speed imaging, we observe distinct regime transitions as Ω increases, identifying the maximum spreading angle (θ_max > 90°) as the critical transition criterion. An energy-based theoretical model is developed for predicting regime boundaries and maximum spreading behavior.

As paramagnetic manganese oxide particles in aqueous suspension undergo magnetophoresis under nonuniform magnetic fields, the resulting concentration gradients induce a bulk fluid motion that hastens the removal of these particles from suspension. We systematically study this fluid motion using both experiments and simulations, and we hypothesize that this induced bulk fluid motion occurs for magnetic Grashof numbers of 1 and greater. We also find that where magnetophoresis opposes sedimentation under gravity, regions of particle depletion form where magnetophoresis is stronger. Finally, we predict field-induced particle aggregation for particles of radii 130 nm and greater.

Tacking is a sailing maneuver that is necessary for upwind navigation. In this work, using a sail membrane and vortex-sheet model, we systematically characterize how a wide range of sail material parameters and tacking motions affects the sail dynamics during the tacking maneuver. We focus on whether a given set of parameters will result in a successful tack, meaning that the sail will flip around to adopt its mirror-image shape, or if it will remain stuck in a metastable state that is close to its initial shape.

A linear stability analysis is performed to investigate the double diffusive convection stability of a density-stratified fluid contained in an inclined slot subjected to a lateral temperature gradient. The results reveal an asymmetrical behavior in stability with respect to the vertical slot configuration. The conditions of small, intermediate, and large solute gradients are respectively considered to explore the effects of inclination angle, and thermal and solute gradients on stability characteristics, especially the behaviors of multiple transitions in instability modes.

We conducted a direct numerical simulation on the flow dynamics within a flat-plate boundary layer, characterized by a transition under favorable–adverse pressure gradients and a three-dimensional rough surface. The skewness and streamwise position of the rough surface was varied. The simulation accurately resolved the rough surface using the embedded boundary method. This research bridges a critical gap in understanding how a three-dimensional rough surface impacts transition. Notably, placing the rough surface within the laminar separation bubble can substantially delay the onset of downstream transition, an effect that remains robust to variations in skewness.

The well-known phenomenon of circular rolls in the rotor-stator flow is analyzed using a homotopy approach. Decreasing the curvature effects changes the transition scenario from subcritical to supercritical, leading to a nonlinear branch of saturated axisymmetric rolls. Direct numerical simulations performed on this branch, together with analysis of the base flow eigenspectrum, lead to a qualitative scenario for the roll merging observed experimentally.

Laminar separation plays a crucial role in wing aerodynamics during the transition to turbulence and, depending on the flow parameters and conditions, can lead to multiple stable states. In the presence of bounding walls, the flow is inherently three-dimensional, with strong spanwise motions observed within the boundary layer up to the wing midspan. These effects persist even in time-averaged views and have important implications for potential control strategies.

The constitutive law relating the viscous stress tensor of a two-phase flow to the local strain-rate tensor in the framework of the one-fluid formulation is derived by combining principles of continuum mechanics with exact kinematic and dynamic matching conditions at the interface. Predictions of this viscous stress model and those of usual ad hoc models are assessed against experimental results in a three-dimensional exchange flow generated by superimposing unstably two fluids in a long circular pipe sealed at both ends. It is shown that the model derived from ground principles and conditions is the only one capable of correctly reproducing the observed flow evolution at a reasonable computational cost.

When a drop impacts a solid surface, it can acquire an electrical charge—a phenomenon reminiscent of triboelectric charging, which occurs when two surfaces, like a balloon and hair, are rubbed together. Here, we consider the charging of bouncing drops on a superhydrophobic surface, focusing in particular on salty drops. Surprisingly, our experiments reveal that adding ionic charges to the liquid reduces the final charge acquired by the drop. More broadly, by testing drops with various compositions and properties, we show that the resulting static electricity is governed by a competition between a friction-based charging mechanism and a bulk discharge process, modeled by a simple RC circuit.

When a thin film of sticky liquid recedes from a surface—such as during evaporation or suction—it can split into regularly-spaced thin filaments. Experiments and theory suggest that the onset of this filament formation depends on surface tension and the Hamaker constant. Varying liquid viscosities and angles of inclination confirmed that the film thickness falling below a critical threshold triggers the onset of filament, which is confirmed by bifurcation analysis and numerical analysis in Mathematica. This explains why sticky fluids, such as the pitcher plant fluids and PEO-water solutions, sometimes form such symmetric and geometrically pleasing patterns.

Flow and solute transport within a cylindrical domain with randomly distributed walls’ topology is solved by means of a stochastic mapping which converts the corrugated flow domain into another one of regular boundaries. The impact of the former corrugated surface is transferred into the coefficients of the transformed flow equations, therefore allowing the computation of the effective flow and transport properties.

We considered a Taylor’s swimming sheet under a finite Brinkman layer. This finite layer thickness changes the established swimming dynamics relative to single and two fluid region models. Unlike swimming in a Newtonian bubble surrounded by an infinite Brinkman fluid, the swimming speed does experience any enhancement, relative to swimming in an infinite Newtonian fluid, as the distance to the Brinkman fluid increases. Even so an increase in the maximum swimming occurs for non-unity porosities in thin highly permeable Brinkman layers.

The flow over a circular cylinder at Re = 220 exhibits mode-A instability, characterized by a spanwise wavelength of around 4d, where d is the cylinder diameter. In the present study, the spanwise domain is extended up to 252d to investigate if the flow field indeed exhibits a periodic pattern with such a spanwise wavelength of 4d. The results reveal that, in addition to the parallel shedding of approximately 4d, very long oblique shedding up to 45d and vortex dislocation are observed in the cylinder wake. In contrast, at Re = 300, where the flow exhibits mode-B instability, such oblique shedding or vortex dislocation does not occur even with a long spanwise domain.

Measurement of suspension properties in a viscometrical device as a Couette cell yield dynamic data due to migration of particles in the shear field. Ultimately, the system stabilizes and properties as particle concentration, particles size distribution, suspension effective viscosity, velocity distribution become stationary and position dependent. The paper displays the stationary properties for a suspension having, initially, a Gaussian particle size distribution. The figure displays the profiles of stationary average particle size in the Couette device.

This paper is part of a study related to the crystallization of drug molecules on low-Earth-orbit satellites. Here, we model the coupled crystallization and sedimentation of a particle in a solvent. We find that by varying the strength of gravity from hypergravity to microgravity, we can control whether the particle will sediment quickly and then grow, or grow while sedimenting. Asymptotic analysis allows us to determine and efficiently model the regimes through which a particle evolves during its growth and sedimentation.

Our high-resolution simulations demonstrate that sediment plume heights from deep-sea mining operations follow consistent, self-similar scaling patterns with vehicle speed, discharge rate, and downstream distance. This finding establishes a robust physical foundation for scaling up to large-scale sediment dispersion models across ocean basins.

In this paper we numerically characterize the stresses experienced by low-Reynolds-number aggregates of various sizes and fractal dimensions. Our work extends previous studies focussed on the stresses felt by settling aggregates, relaxing the low-fractal-dimension assumption through proper accounting for the presence of neighboring particles in the aggregates. We also consider aggregates subjected to a shear background flow, a case experimentally relevant but numerically understudied. Our findings provide insights on the distribution of the stresses felt by aggregates, which could be used to develop refined dynamical models of aggregation that include breakup mechanisms.

At the start of a race or in light winds, windfoil athletes use intermittent propulsion by pumping the sail to get or keep the board in foiling mode, for example after a tack change. This involves periodically changing the angle of the sail relative to the wind by moving the center of mass up and down. We experimentally investigated the impact of the incidence angle on the aerodynamic forces using a pitching foil at a reduced scale, as well as different frequency and amplitude combinations within a certain range of Strouhal numbers. Our measurements revealed aerodynamic behaviors that enabled us to explore sailing race strategies.

We examine the spontaneous emergence of rotating, regular, and peaked polygonal patterns in Leidenfrost rings (first observed in 2007) with a model incorporating surface tension, poloidal rolling vortices, and the interplay of buoyancy-driven and thermocapillary flows. Fluid velocity is decomposed into potential and rotational components, and a vortex Reynolds number is used. A nonlinear equation, solved with Helmholtz–Hodge boundary conditions, produces periodic cnoidal wave solutions consistent with experiments. A parallel model, based on capillary pressure averaging, leads to a KdV-type equation yielding cnoidal and trigonometric solutions, closely matching the observed peaked polygons.

The present work explores how elastic vibrations of a panel can impact the shock-induced separation dynamics. We apply advanced dynamical systems tools and high fidelity experimental datasets to glean new insights into the mechanisms that drive fluid/structure feedback interactions. This expands our understanding of computational aeroelastic models and improves control of structural vibrations in supersonic and hypersonic platforms.

Richtmyer-Meshkov (RM) and Rayleigh-Taylor (RT) instabilities are expected to play a crucial role in the explosion-driven post-detonation mixing. They can control mixing within an explosive fireball and that of an external agent with the background air and products. The conventional understanding and models of RMI and RTI primarily stem from channel studies; however, this numerical study considers three-dimensional and reactive simulations of a spherical explosion to demonstrate that they can be applied for post-detonation mixing under certain conditions.

Nutrient mass transport to a spherical squirmer whose motion causes recirculating regions is examined. Mass transport across the dividing streamline III must equal mass transport to the squirmer along II. Nutrient concentration boundary layers form at large Peclet numbers along I and II with thicknesses diverging at stagnation point B. These boundary layers emerge out of the stagnation region with a discontinuity in nutrient mass flux along III. New boundary layers form alongside III to compensate for this discontinuity. We conduct a detailed analysis of this mass transfer process to obtain an analytical expression for the rate of nutrient mass transport at large Peclet numbers.

The work investigates the turbulent mixing of passive substances for Schmidt numbers $Sc\gtrsim 1$ without and with phase changes. We compare direct numerical simulations with the Lagrangian diffuselet model of filament aggregation and demonstrate an agreement of the scalar statistics for the initial stage of the mixing process.

The understanding and prediction of coherent flow structures are crucial for drag reduction and flow control of wall-bounded turbulence. A white-noise forcing model for predicting the flow structures is proposed for the eddy-viscosity-based linear analysis of turbulent channel flow, based on the self-similar features of the optimized ones within a wide range of Reynolds numbers. Regardless of the specific eddy-viscosity model adopted, the relative errors of the predicted covariance tensor and the estimated flow state with the forcing model are notably reduced compared to those without such a model, highlighting its universality in the linear analysis of wall-bounded turbulence.

Longitudinal and oblique acoustic waves within the channel flow were accurately captured under the constraints of impedance walls. The wavenumber-frequency relationships for all reflected waves and surface waves subjected to the specified time-domain impedance boundary conditions were theoretically developed, and the accuracy of these theoretical predictions was validated through high-fidelity direct numerical simulation results. The wavenumber spectrum on the wall was further explored, revealing the presence of supersonic ridges distributed along a series of concentric ellipses, but with modified wavenumber relationships.

In this study, we perform wall-resolved large-eddy simulations to investigate flow-acoustic resonances in deep cavities (D/L = 2.632) at three inclination angles and two Mach numbers. We discover that inclined cavities generate acoustic responses more than 30 dB stronger at a surprisingly low frequency (St=0.276) than the orthogonal cavity. Through modal and resolvent analyses, we identify the distinctive vortex dynamics mechanism at play and reveal the primary factors that contribute to the enhanced aeroacoustic responses in the inclined cavities. Finally, we propose a predictive criterion for the onset of deep cavity resonance associated with the distinctive vortex dynamics identified.

The Lamb vector appears in the rotational form of the Navier–Stokes equations and governs vortical nonlinearity and energy cascade in turbulence. This study elucidates its three-dimensional spatial structure in the separated near wake of a NACA0018 wing. The Helmholtz decomposition reveals distinct patterns of the potential and solenoidal components. Integration of the cross-stream component of the Lamb vector demonstrates that the Taylor–Sears condition is satisfied at the trailing edge.

This study is aimed at investigating the influence of Euler acceleration on the transient dynamics over a rotating wing. A new dimensionless acceleration parameter, defined as the ratio of Euler acceleration to convective acceleration, is introduced. Using this parameter, the transient dynamics are classified into quasi-steady and acceleration-dominated regimes, with each regime exhibiting distinct lift generation and vortex evolution mechanisms. The proposed scaling and regime classification offer a valuable framework for studying transient aerodynamic behavior and guiding future investigations in selecting appropriate parameters to better understand the role of Euler acceleration.

We investigate how vortex rings, generated inside a water tank, behave when they interact with a water-air interface. We find that strong vortex rings exhibit near-perfect reflections when approaching the interface at sufficiently large incident angles. Through systematic experiments and simulations of a vortex-sheet-vortex-pair model, we establish the first phase diagram of vortex ring dynamics at the free surface, for a wide range of Froude numbers and incident angles. A simplified flux/momentum conservation model offers additional insight into the physical mechanism behind the vortex-ring reflection.

When elongated, flexible floaters such as dead leaves, drifting nets or agglomerated microplastic blobs drift on surface waves, they spontaneously align with the direction of wave propagation. We investigate this phenomenon through theoretical analysis and laboratory experiments. We demonstrate that a thin, flexible strip experiences a mean second-order moment that induces a slow angular drift, analogous to the Stokes drift mechanism for the linear motion. This drift arises from an imbalance between the slightly stronger accelerations on the wave crests, that favor longitudinal orientation, and the weaker accelerations in the troughs, that favor transverse orientation.

Inhomogeneous rotating turbulence is ubiquitous but rarely investigated. The unevenly distributed Coriolis force causes a difference in the relative intensity between inertial force and Coriolis force, resulting in distinguished scaling laws in the rapid and slow rotating regions. In our study, we conducted direct numerical simulations (DNS) of inhomogeneous rotating turbulence, which reveals how the Coriolis force influences coherent structures and, in turn, the scaling laws. That is, the Coriolis force accounts for the formation of dumbbell-shaped columnar structures and the generation of helicity.

This study revisits the perturbation approach to develop a consistent continuum framework for a variable-coefficient acoustofluidic solver. By distinguishing between the fluid’s Lagrangian and mass transport velocities, we reveal two analytically equivalent but numerically distinct forms of the second-order mass source. These velocities are systematically related to demonstrate that zero Lagrangian and mass transport velocity boundary conditions are generally not equivalent. Numerical accuracy is verified via the Method of Manufactured Solutions, and several test cases highlight sensitivity to boundary conditions and spatial variations in density.

Fluid flows involve multiple time scales and averaging over fast time scales is often key to their analysis. Averaging is usually carried out at fixed positions, but averaging along particle trajectories – Lagrangian averaging – has advantages. The numerical computation of Lagrangian averages from simulation data is a challenge, however.

This work develops a new, fast method for the computation of Lagrangian averages. The method is effective at filtering out fast waves and extracting turbulent flow features as the shallow-water simulation shown in the image illustrates (top half, full vorticity field; bottom half, Lagrangian-averaged vorticity field).

Traditional cardiovascular CFD simulations use time integration to capture the transient behavior of blood flow. In this manuscript, we leverage the periodic nature of cardiovascular flow and use a finite element method to solve the harmonic balance form of the Navier–Stokes equations. The harmonic balance method transforms the governing equations into the frequency domain, thereby significantly reducing the computational cost of time integration. We demonstrate that this approach yields cost savings of one to three orders of magnitude, with minimal impact on solution accuracy.

A convolutional neural network (CNN) is constructed to predict the shape and location of arbitrarily positioned bluff obstacles in a two-dimensional channel flow, trained using either velocity or concentration fields obtained from Lattice Boltzmann simulations. We analyzed multiple cases to explore various input types and degrees of data sparsity, testing adaptability and robustness of the CNN with limited data input. This approach can be extended to three-dimensional flows, experiments, or even $in$ $vivo$ biological systems that are optically accessible, leading to an accurate predictive framework for determining complex geometry in a variety of biomedical, geophysical, and other systems.

Typically, flow visualization uses specialized particles, such as silver-coated glass microbeads, illuminated with laser light; however, synthetic particles might cause potential health risks or environmental concerns. In our new study, we characterize how starch – a safe, plant-based material – can be used as tracers in underwater experiments with foils, jellyfish, and brine shrimp. Starch particles are effective for particle image velocimetry, pose fewer health risks for humans and animals, are environmentally friendly, and cost a fraction of commercial-grade options. This work is intended to promote more sustainable and ethical research practices in biology and traditional fluid dynamics.