Capillary suspensions are defined by liquid-bound particle clusters, yet despite decades of study, their stability in strong flows remains incompletely understood. Here, we establish a universal set of stability criteria for a liquid-bound particle doublet in extensional flow. A critical capillary number governing stability is identified through a combination of analytical theory and experiments. Below this threshold, stability depends sensitively on initial particle separation, whereas above it, clusters are unconditionally unstable. These results provide a quantitative framework for predicting and controlling flow-induced breakup in capillary suspensions.

Peristaltic pumping drives flow in diverse biological and engineered systems. Previous models focus solely on radius-imposed contraction waves or neglect the elastic coupling between radial and longitudinal strains. By describing the fluid using the material coordinates of the deforming wall, we analyze flow in a tube undergoing simultaneous transverse and longitudinal contractions. We show that the elastic coupling between circumferential and longitudinal strains strongly alters net transport, reflux, and trapping, as compared to the case with radial contractions alone.

Many animals leverage stereo inhalation for respiration and olfaction, drawing fluid and odors into a spatially separated pair of nares. Olfaction efficacy is known to be enhanced by the structure and dynamics of flow exterior and interior to the nares. We characterized stereo inhalation flow kinematics and capture volumes using numerical models of an idealized, dual siphon geometry, providing further context for sensory adaptations. We find capture volumes that are modulated by Reynolds number and siphon geometry, suggesting that organisms may alter morphology and inhalation dynamics at behavioral or evolutionary timescales to increase fitness.

Bicuspid valves with crescent-shaped leaflets in veins and lymphatics ensure unidirectional flow to the heart by preventing reflux. While longer leaflets increase hydrodynamic resistance and excessive stiffness hinders proper valve closure, a key question remains: why is the leaflet crescent-shaped, and to what extent should it be creased to optimize performance? This study isolates geometry by varying only leaflet length under backward flow, revealing a transition from reflux to full blockage. The threshold and, thus, valve competency depend strongly on cusp shape, explaining reflux in short, immature, or abnormal valves.

When the worldlines of inertial particles in background flows cross, they generate low-dimensional structures with diverging particle number-density, formally similar to optical caustics. We show that orientable motile particles in flows can form caustics even when their mechanical inertia is neglected. Singular perturbation analysis of self-propelled particles around a point vortex and numerical simulations of their motion in a turbulent flow uncover the various regimes of caustics, demarcating the necessary conditions for their formation. Active caustics greatly enhance encounters between Stokesian swimmers, and an order-of-magnitude estimate points to their ecological relevance.

Cavitation bubbles near free surfaces are known to move rapidly away from the liquid surface and generate high-speed microjets. We demonstrate that viscoplastic fluids can fundamentally change this outcome by arresting the bubble and trapping it beneath the surface once a critical yield number is exceeded. A regime map classifies the flow into four regimes (swelling, trapped, bullet jet, and vapor jet), and a force-balance model provides a predictive framework for the bubble penetration depth.

Smaller gas-solid contact area or smaller bifurcation angle? What happens when inlet and outlet conditions are exchanged? This study employs topology optimization to investigate optimal manifold configurations across gas rarefaction regimes and uncovers two counterintuitive phenomena: a wetted-area minimum in the slip regime and inlet–outlet reciprocity in free-molecular flows. These findings provide crucial guidance for microfluidic and vacuum system applications.

Hypersonic boundary layers in the transition–continuum regime (Knudsen number Kn ≈ 0.1–10) challenge Navier–Stokes solvers due to the breakdown of transport laws and slip/jump wall conditions. We embed physics-constrained neural closures for viscous stress and heat flux directly in the partial differential equations and train them using adjoint-computed gradients to match direct simulation Monte Carlo target data. A distribution-function wall model built from mixtures of skewed Gaussians replaces empirical slip-velocity models, substantially improving bulk flow and boundary-layer predictions and generalizing across unseen Mach numbers, Knudsen numbers, and geometries.

Wall-attached Rayleigh-Bénard convection arises in the presence of a damping body force, e.g. the Coriolis or Lorentz force, when this force is less effective near a lateral boundary than in the bulk. The shape of the container is important in this context. A numerical linear stability analysis of Rayleigh-Bénard magnetoconvection in a wide rectangular container with a vertical magnetic field and electrically insulating walls shows that the least stable mode of convection becomes localized in the corners rather than spread out over the whole circumference of the container. This corner mode has a similar dependence on the magnetic field strength as the ordinary wall-attached mode.

Droplet transport in deformable constrictions is important in microfluidics, enhanced oil recovery, biomedical systems, and surface-acoustic-wave-driven platforms. We study how actuation controls droplet motion in a constricted deformable tube using high resolution multicomponent fluid-structure interaction simulations. We show that oscillatory traction on the tube walls exhibits resonance, i.e., the droplet’s mobilization time is minimized when the traction frequency nears the tube’s natural frequency. However, actuation via oscillatory fluid forcing exhibits no such resonance. Our findings provide design guidelines for tunable, actuation-controlled droplet motion in soft confined geometries.

We investigate the nonlinear axisymmetric oscillations of non-viscous droplets with a focus on superimposed initial deformations, in which an even mode with large amplitude is superimposed on an odd mode at small amplitude. Our analysis of modal coupling shows that the even mode with large amplitude strongly influences the odd mode with small amplitude due to asymmetries in local curvature and restoring forces between flattened and elongated shapes (as in the image), while the odd mode with small amplitude has only a minor influence in return.

Drying of complex fluids is crucial in many natural and technological processes, yet predicting it and probing associated transport phenomena remain challenging. We introduce an original interferometry‑based method for confined two-dimensional droplets in a humidity‑controlled chamber, enabling simultaneous high‑precision, high‑resolution measurement of drying kinetics and internal concentration fields, providing a powerful tool to accurately characterize drying dynamics and transport in complex fluids.

Existing research on Worthington jets has paid most attention to those forced by cavity collapse. Focusing on the cavity-free regime across various sphere-liquid impact configurations, we derive a universal scaling law for the maximum jet height from first principles and identify three distinct pinch-off modes governed by Rayleigh–Plateau instability. Self-similar analysis accurately captures the evolution of jet shape and height, revealing gravity-dominated jet dynamics. These findings confirm that this jet undergoes a fundamentally different physical process from that in air-entrainment scenarios, thereby significantly enriching the classical Worthington jet phenomenon.

A water droplet hanging from the tip of a vertical sugar fiber seems destined to fall as gravity overcomes capillarity. However, the droplet is also gradually dissolving its own support. Will the droplet fall? Sometimes. However, it may instead suddenly jump upward and climb the very fiber it consumes. We show how this counterintuitive motion emerges from the interplay of surface tension, gravity, and dissolution, and identify the conditions that determine whether the droplet falls or rises. This process may again occur until the droplet is full.

Acoustic radiation force can be used to steer ultrasound contrast microbubbles toward the desired clinical target, but the link between their oscillations, displacement, and stability has remained unclear. By tracking single lipid-coated microbubbles in free space, we show that their displacement is accurately captured only when history drag is included in the force balance. A simple linear scaling connects volumetric expansion to transport distance. Above a critical radial expansion, however, shape-mode oscillations emerge and dissolution rises sharply, revealing a trade-off between transport efficiency and bubble integrity.

This paper addresses the problem of a two-dimensional (2D) droplet under shear. We introduce an analytical approach, utilizing a 2D Lamb solution to derive an expression for the apparent viscosity of a dilute 2D emulsion and to develop a deformation theory for small capillary numbers. Validated through direct numerical simulations, our findings establish benchmarks for computational fluid dynamics methods and for interpreting 2D droplet behavior.

Large-eddy simulations of the Navier–Stokes equations under the Boussinesq approximation investigate the hydrodynamics of a three-dimensional hyperpycnal plume over a sloping bed. In the unconfined configuration, plunging is governed by lateral slumping rather than by a critical densimetric Froude number, producing a characteristic triangular surface pattern and a distinct downstream wake. The wake extent increases significantly with increasing inflow densimetric Froude number and agrees with field observations of the Rhone River inflow into Lake Geneva. Total entrainment increases with decreasing densimetric Froude number due to enhanced lateral spreading and increased underflow velocity.

We demonstrate that asymmetric ice blocks floating in water can self‑propel while melting. Experiments with triangular ice prisms show that melting generates a directed, buoyancy‑driven gravity current along the inclined face, producing steady translation. A momentum‑balance model quantitatively predicts the propulsion velocity as a function of ice geometry and bath temperature. This mechanism persists in saltwater at sufficiently warm temperatures, highlighting melting as a generic propulsion mechanism in buoyancy‑driven flows and a possible secondary contributor to iceberg drift.

The recent literature on wall blowing flow control was here adapted to suction-based laminar flow control (LFC) to study efficient control scenarios. These are evaluated by means of a numerical framework deploying fast and relatively accurate CFD and flow transition solvers embedded in an efficient optimization routine. More than 700 combinations of LFC parameters were explored showcasing the flow sensitivity to the deployed control and yielding to power savings overcoming the value of 30%.

While classical Hele-Shaw models focus on single interface dynamics, the mechanisms driving instabilities in multi-connected fluid domains remain largely unexplored. We reveal that the spatial configuration and viscosity of internal fluid domains fundamentally break radial symmetry, triggering viscous fingering on the outer boundary. By strategically arranging these inner interfaces under a time dependent injection flux, one can suppress unfavorable instabilities and actively promote preselected, self-similar limiting shapes.

Controlling capillary fingering via patterned porous media (PPM) optimizes industrial processes (e.g., fuel cells). We introduce a 2D Zoned Sequential Deposition method to fabricate PPM with tunable porous media features. Direct numerical simulations across varying capillary numbers and heterogeneity factors show that highly heterogeneous PPM (having larger pore-diameter contrasts between different zones) promotes structured drainage: flow follows underlying porous microstructure, draining through large pores with less than 10% occupancy of finer spaces. This coupling of fabricated morphology and flow behavior provides a framework for designing porous materials with predictable flow patterns.

Wedge-shaped confinements are increasingly relevant in low-Reynolds-number microfluidics, yet existing Green’s functions describe only flows driven by point forces. We derive the flow induced by localized torques using a Fourier–Kontorovich–Lebedev framework combined with the Papkovich–Neuber representation. The resulting solutions reveal how geometric asymmetry couples rotation and translation and yield the full torque– mobility tensor. These analytical results provide predictive tools for controlling particle motion in confined microfluidic systems.

The motion of rigid bodies in viscous fluids at vanishing Reynolds number is governed entirely by its geometry. While highly symmetric particles like spheres and flat circular disks sediment without reorientation and chiral bodies follow helical trajectories, the dynamics of weakly asymmetric shapes remain difficult to predict. Here, we combine numerical simulations and experiments to investigate “pinched” U-shaped disks with a single plane of symmetry. By varying the degree of pinching, we demonstrate that this class of achiral particles can realize the complete spectrum of inertialess sedimentation behavior, from straight settling to robust quasiperiodic spiraling.

Classic orifice flow models, originally developed for low-Reynolds-number liquids, use the decoupling between velocity and concentration fields to simplify the analysis. This simplification fails for gaseous mixing, where composition changes directly alter the velocity field. Our study addresses the coupling in the Sc $\sim$ Pe $\sim 1$ regime typical of gas-delivery systems. We introduce a unified framework that combines new analytical solutions for low Pe with simulations. This approach provides quantitative predictions for mass-transfer rates and pressure drops, and can help design the restrictive orifices critical to semiconductor manufacturing and precision gas-delivery technology.

Singular solutions in the stationary Stokes regime are reported for fluids confined by cylindrical walls. The stokeslet, stresslet, couplet, point source, and point source dipole are obtained internally, externally, and within the annular region between cylindrical walls using bitensorial calculus. Beyond providing hydrodynamic solutions relevant to particle transport in confined environments, this work highlights the strength of bitensorial calculus in handling curved geometries and systematically yielding hydrodynamic singularities within a unified framework. Forces on sedimenting particles and active microswimmers near cylindrical walls are investigated as an application.

Present continuum based modified electrokinetic model capture the nonclassical pattern of the electric double layer arises in the strong coupling regime i.e., layered structure of ions, counterion saturation, overscreening of surface charge, and reversal in electroosmotic flow. Based on the present modified model we have established qualitative agreement with several experimental observations, which the mean-field based models fails to envisage. The short-range effects on ion transport and their impact on membrane polarization are quantified in this study, which has not been addressed in previous studies. It may provide useful insights on tuning the electroosmosis and particle trapping.

We investigated experimentally for the first time the rotational dynamics of flexible fibers in homogeneous and isotropic turbulence. The rotation and the deformation of the fibers have been measured thanks to an efficient three-dimensional reconstruction algorithm from three simultaneous images. We show that an increase of fiber flexibility enhances the rotation rate. We show that this phenomenon is directly related to the amplitude of deformation.

Classical pressure-drop correlations for packed beds often yield inconsistent predictions across different geometric scales and flow rates. By applying a residual-driven sensitivity analysis to an extensive experimental dataset, this work reveals that while geometric wall effects initially dominate prediction errors, the superficial velocity overwhelmingly dictates residual behavior once these are minimized. This finding indicates that future model improvements should prioritize flow-regime-dependent corrections over further geometric refinement.

Understanding droplet dynamics under detonation loading is vital for advanced propulsion technologies like rotating detonation engines. This study reveals fundamental differences between detonation and shock wave interactions with water droplets using high-resolution simulations. We demonstrate that the rapid post-wave pressure attenuation in detonations accelerates cavitation collapse and suppresses the Rayleigh-Taylor forward jet typical of shock impacts, leading instead to unique leeward-side flattening.

The vortex shedding phenomenon has long fascinated researchers, yet only recently has the behavior of unstable growth and saturation been placed on firm mathematical footing through the Stuart–Landau and Ginzburg–Landau equations. While these models capture key aspects of the dynamics, a complete understanding of vortex shedding remains elusive. Here, we use data-driven methods trained on coarse-grained numerical flow data to learn distinct local models at multiple stations in the downstream wake. This approach yields new insight into the nature of the instability, the wavemaker region, and vortex shedding itself.

How many degrees of freedom characterize turbulent flow? We investigate the dimensionality of two-dimensional Kolmogorov flow across Reynolds numbers and forcing scales using convolutional autoencoders and Lyapunov analysis. Two dynamical transitions are identified, first associated with periodic-orbit destabilization and later with large-scale saturation. The resulting saturation dimension scales linearly with the forcing wavenumber rather than with the total number of available Fourier modes.

Simulations of interfacial mass transfer often interpolate the diffusion flux at the interface, making concentration predictions artificially sensitive to the chosen interface width. In this study, we propose a source-free phase-field-lattice-Boltzmann model that ensures bulk concentration profiles remain completely independent of interface thickness. We also introduce an additional free parameter to the evolution equation of the model, which significantly improves its numerical stability under low Henry constants. The proposed framework is highly beneficial for investigating complex multiphase systems, such as those involving surfactants or Marangoni effects.

We introduce a simple method to objectively identify the axes of coherent vortices in turbulence using only the velocity-gradient tensor. The method is readily applicable to experimental data. As an example, applying it to stereo-PIV measurements of a wind-tunnel turbulent boundary layer, we quantitatively show that boundary-layer-scale vortices form hairpin shapes.

In rotating geophysical flows, turbulence can either drive small-scale mixing or build large-scale coherent eddies. Here, a very large rotating-tank jet experiment with planar particle imaging velocimetry (PIV) is used to map the local energy flux across scales. The results show that stress–strain alignment and a local jet Rossby number organize where and how long inverse energy-cascade patches appear. This identifies a local control parameter for steering energy pathways in jet-like environmental flows.

Temporal decay of rotating turbulent thermal counterflow of He II is probed by second sound and found to display interesting new features. Two effects are observed, acting against each other and affecting the late temporal decay of vortex line density, $L(t)$. The first one is gradual decrease of the decay exponent of the power law $L(t)$, confirming that turbulent thermal counterflow under rotation acquires 2D features. The second one is the influence of the effective Ekman layer built within the effective quantum Ekman time. For increasing rotation rates, $L(t)$ gradually ceases to display a clear power law. Instead, rounded and ever steeper decays occur, gradually shifted toward shorter times.

In the presence of linear friction, the properties of two-dimensional turbulence deviate from the classical Kraichnan phenomenology. The enstrophy flux in the direct cascade is suppressed resulting in a steeper energy spectrum. The spectral exponent can be predicted in terms of the statistics of the Lagrangian finite time Lyapunov exponent. We numerically verify this prediction and propose a simple phenomenological model for the dependence of the Lyapunov exponent on friction intensity.

Spontaneous stochasticity—persistent randomness in the limit of vanishing noise and viscosity—has been observed in turbulence models, yet its universality lacked a theoretical explanation. We develop a renormalization-group (RG) framework for the fluctuating Sabra shell model, showing that the ideal turbulent dynamics is governed by a fixed-point RG attractor. This approach explains universality across dissipation and noise mechanisms and predicts a complex RG eigenvalue responsible for the slow, oscillatory convergence observed numerically.

Broadband trailing edge noise is generated by the scattering of three-dimensional hydrodynamic structures, but the role of spanwise wavenumber selection in acoustic radiation for a finite spanwidth airfoil remains unresolved. Wall resolved compressible large eddy simulation of a NACA0012 airfoil shows that nonzero spanwise modes become dominant above the acoustic cut-on frequency associated with obliquely convecting wavepackets identified via spectral proper orthogonal decomposition (SPOD). A reduced-order model based on extended SPOD reproduces far-field noise using only a small number of modes, providing a compact and control oriented framework for noise prediction and mitigation.

Turbulent flow through sudden pipe expansions is widely studied, yet the role of expansion angle in shaping turbulence structure remains poorly understood. Using high-resolution stereo-Particle-Imaging-Velocimetry in a refractive-index-matched facility, we directly compare abrupt (90°) and gradual (45°) axisymmetric expansions. We show that slope fundamentally reorganizes the return flow, amplifying shear-layer interaction, turbulence production, and anisotropy in gradual expansions. The results provide a mechanistic explanation for the higher losses long observed in sloped geometries.

A key open question in turbulence research concerns the nature of fluid motions that can produce a self-similar energy cascade consistent with Kolmogorov’s statistical theory of turbulence. We approach this problem by considering the one-dimensional viscous Burgers equation as a toy model, and frame the question in terms of a family of partial-differential-equation-constrained optimization problems which are solved numerically. Our results represent a successful effort to construct time-dependent solutions of this model characterized by self-similar energy transfers, providing a framework that may be used to search for self-similar behavior in three-dimensional turbulence.

Helicity, the volume integral of the velocity-vorticity scalar product, is a key dynamical invariant encoding flow topology; however, measuring it in turbulence is a significant challenge due to the requirement for high-resolution velocity gradients. We introduce chirality tomography, a Lagrangian method that reconstructs three-dimensional helicity maps from the entanglement of particle trajectories. By establishing a robust proxy between trajectory linking and local helicity, we provide the first spatially resolved maps of chiral structures in fully developed turbulence. The approach bridges trajectory-level topology with fundamental physics, with a practical diagnostic for complex flows.

In turbulent flows of dilute gases, the amplitudes and correlations of the turbulent fluctuations of the thermodynamic variables (pressure, density and temperature), satisfy exact nonlinear compatibility relations and inequalities. These define realizability constraints on the thermodynamic turbulence structure, valid from the quasi-incompressible-flow limit to hypersonic Mach numbers. Furthermore, the ratios between fluctuation intensities define the signs of correlations between the thermodynamic fluctuations, and define bivariate mappings of the thermodynamic turbulence structure.

While drag reduction in fully developed viscoelastic flows is widely studied, the combined influence of polymer additives and strong spatial acceleration remains largely unexplored. This study employs high-resolution velocimetry to examine near-wall turbulence in semidilute polymer solutions subjected to varying favorable pressure gradients. The results demonstrate that the interplay of viscoelasticity and acceleration profoundly suppresses Reynolds shear stresses, driving the boundary layer toward a distinct quasi-relaminarized state dominated by elastic effects.

Intermittent swimming boosts efficiency—but at what cost? Using two-dimensional simulations of an unconstrained pitching foil, we show that burst-and-coast motion induces pronounced lateral drift and a sharp, amplitude-driven reversal in trajectory direction. The deflection is governed by Strouhal-controlled scaling and wake realignment across burst and glide phases. These results uncover how intermittent kinematics trade stability for energetic advantage.

Rotating stall in reversible pump-turbines operating in the S-shaped region degrades stability and drives strong torque and pressure oscillations, yet quantitative inception prediction remains limited. Based on small-disturbance theory, we couple runner perturbation dynamics with the external system characteristic to analyze resonance and stability of disturbance waves and predict stall onset, wave speed, and cell number. Unsteady CFD ramp-downs from runaway to low flow under four guide-vane openings, with wavelet analysis of vaneless area pressure, validate the model’s accuracy in predicting stall onset.

Flexible vortex generators enhance heat and mass transport, but most studies focus on solid, non-porous panels or passive flexible reeds. Inspired by porous, compliant fish-gill filaments, we demonstrate the mixing benefit of pitching flexible perforated panels. Pitching drives unsteady entrainment; perforation yields spatially discretized vortices without a conventional leading-edge contribution, while chord-wise flexibility sustains mixing via wake-mode transitions. We examined the Lagrangian coherent structures to link vortex dynamics to convective–diffusive transport, and proposed three indices to quantify mixing by uniformity, mean increase, and spatial dispersion.

Traditional simulations of large-scale fully nonlinear free-surface wave–body interactions remain computationally demanding. We present a spectral–fundamental solution (SFS) method that combines global spectral bases with local fundamental solutions, achieving high efficiency across large domains while maintaining accuracy near the body surface. Using this method, we investigate nonlinear ship-wave dynamics in extensive domains. The simulations reveal the physical origins of distinct energy bands in ship wakes, the effects of acceleration on wake evolution, and the mechanism behind wake-angle narrowing at high speeds.

When tracers are scarce in flow measurement using particle imaging velocimetry (PIV), experimenters must wait for tracers to come in the test section, making measurements inefficient. To address this ill-suited case for PIV, the paper introduces optical trapping of a tracer; briefly let it go so the flow moves it, then pull it back at the same initial position. The method, termed as optically-trapped particle tracking velocimetry, is validated by measuring a slow microflow in a square microchannel. The authors also demonstrate an application to optothermal flows where tracers are depleted near the heat source, showing how ot-PTV can still probe tracer motion in an ill-suited case for PIV.

In a recent study of the formation of electric charge caps on a leaky dielectric droplet, it has been shown that, in the limit of high electric Reynolds number, the electric field problem satisfies a non-standard boundary value problem of mixed type. By using novel techniques involving consideration of a so-called prime function, this paper shows that it is possible to solve this mixed boundary value problem analytically.

We developed a new method for high-resolution measurement of CO${}_2$ transport across the air–water interface, addressing a critical need in physical oceanography studies and in modern carbon sequestration applications. This method forgoes conventional probe-based sampling and statistical correlation, and exploits precision laser spectroscopy with rapid spatial beam scanning to directly quantify gas-phase CO${}_2$ distribution near the interface at millimeter and millisecond resolutions. To our knowledge, this is the first time such a level of resolution has been achieved. This method should prove useful in both field measurements and laboratory studies of cross-interface gas transport.

The transient growth of disturbances is typically investigated using the matrix exponential of the linearized Navier-Stokes operator. We introduce a data-driven algorithm that computes optimal initial conditions, response modes, and their associated energy growth directly from snapshots of flow data. Our method simplifies and broadens the application of transient growth analysis, eliminating the need for access to the linearized operator and enabling application to experimental data. We demonstrate the method, including a regularization to mitigate the sensitivity to noise, using a Ginzburg-Landau equation, Poiseuille flow, and a transitional boundary layer.

Resolving turbulent flows pushes both computations and algorithms to their limits. As a result, high-fidelity turbulence simulations increasingly rely on GPU-accelerated solvers that adapt to massive parallelism and memory constraints to overcome the computational limits of CPU-based solvers. This review maps the rapidly expanding ecosystem of GPU-accelerated DNS and LES codes for single- and multiphase turbulence for both compressible and incompressible flow, analyzing algorithmic strategies, porting challenges, and performance bottlenecks. By linking numerical methods to hardware evolution and memory constraints, we outline the path toward efficient, exascale turbulence simulations.