Can we predict how mixed grains unmix themselves? A continuum model couples particle-level segregation forces with mixture rheology to capture the inherently linked evolution of flow and segregation in bi-disperse granular systems. It reveals the crucial role of composition-dependent packing arising from size disparity in accurately predicting segregation dynamics. The model successfully reproduces segregation evolution across diverse configurations, compositions, and size ratios, closely matching Discrete Element Method simulations.

Compressible turbulence exhibits complex mechanisms of energy exchange between kinetic and internal modes, yet their dependence on Mach number remains poorly understood. This work investigates two-dimensional compressible Taylor–Green vortex flows and reveals how Mach number governs the pathway of internal-kinetic energy transfer. We identify three stages of energy evolution and derive an analytical model that predicts the initial growth of dilatational kinetic energy. The results provide new insight into compressible-flow energy exchange and inform future turbulence modeling strategies.

We investigate the impact of permeability heterogeneity on a bimolecular A + B → C reaction that creates a nonmonotonic fluid density profile which can trigger or suppress convective instabilities. The intensity and structure of heterogeneity control how mixing and reaction scale. We find opposing trends for the mixing of reactants and the reaction depending on the media stratification and the density profile. In horizontally stratified media, reaction is hindered by heterogeneity, while in vertically stratified ones it favors reaction. A similar reaction-favoring behavior is also observed for anisotropic multi-Gaussian log-permeability fields.

Basal ice melting under shear flow is a key process in polar and lacustrine environments, where it is influenced by both external forcing and water’s nonlinear equation of state. While previous studies have examined buoyancy-driven convection and shear separately, their interplay with density inversion remains poorly understood. By using direct numerical simulations, we reveal distinct flow regimes and interface morphologies controlled by the inversion stability ratio and effective Richardson number, providing scaling laws for heat transport and a theoretical framework for regime transitions.

We measure the efficiency of convective heat transfer (Nusselt number) by turbulent convection in a rapidly rotating Rayleigh-Bénard convection experiment. Series of measurements are done at two constant values of the Rayleigh number. Using water at different mean temperatures, we change the Prandtl number. Raising the Prandtl number leads to a reduction of the Nusselt number with a significantly stronger dependence than without rotation. We hypothesize that this dependence is caused by the changing ratio of the thermal and kinetic boundary layer thicknesses.

As a bubble oscillates close to a water surface, a new phenomenon of partial ventilation which the bubble’s exposure to the surrounding air is observed induced by the Rayleigh-Taylor instability penetrating the bubble wall. Depending on the exposure, three distinct types of bubble behaviors with decreasing dimensionless stand-off distance are summarized: (i) nonventilation; (ii) partial ventilation; and (iii) complete ventilation. The boundaries for ventilation time and stand-off distance are obtained by solving the analytical model comprising a small-amplitude model and a bubble-oscillation model, which agrees well with the experimental observations and numerical results.

Early bubble release is vital in electrochemical and boiling systems. Coalescence can trigger bubble departure, but it can also leave bubbles pinned to the electrode surface. By observing the contact-line dynamics and quantifying adhesion and dissipation energies, this work identifies the threshold conditions governing this transition. The developed general energy-balance framework predicts bubble detachment in agreement with available experimental and numerical data.

Growth of bubbles nucleated on a wall appear in various industrial applications (gas/liquid contactors, evaporators, hydrogen production), in which their detachment radius is an important parameter for performance optimization. Thanks to gas injection experiments in a shear flow, a force balance model for predicting the detachment radius is built using the relevant forms of the forces extracted from recent literature. A new detachment criterion, based on a maximum advancing contact, allows to determine the receding contact angle and detachment radius with good agreement for both hydrophilic and hydrophobic surfaces.

When bubbles burst on the ocean surface, small drops are emitted. In this work, we find that when there are surfactants, a particularly common type of contamination found on fluid surfaces, small bubbles can emit drops that are significantly larger than expected without contamination. For the bubbles considered in our experiment and simulations, we see up to a 5 times increase in the size of the emitted drop. Previous studies involving large bubbles saw the opposite effect, where the drop size decreased in the presence of surfactants. This finding therefore has fundamental implications for our understanding of aerosol production from bubble bursting.

The paper presents a study of the evolution of two-dimensional gravity currents, propagating against constant and pulsating counter flow. The effect of mean and oscillatory velocity amplitude on the gravity current evolution, the onset of interfacial shear instabilities such as Kelvin-Helmholtz billows, unstable near-bed density stratification, and the resulting density redistribution is addressed. Two non-hydrostatic processes, shear-driven Kelvin-Helmholtz billows and Rayleigh-Taylor-like overturning induced by differential advection, enhance vertical mixing and horizontal transport of dense fluid within the gravity current and are expected to affect salt intrusion in estuaries.

Understanding how intermolecular forces govern nanoscale wetting and phase change requires tools that can link molecular and continuum scales. This work develops a continuum framework that integrates non-local solid-fluid interactions into the Navier-Stokes-Korteweg model, enabling the study of adsorption and thin film dynamics with thermodynamic consistency from the continuum level. By incorporating these interactions directly into the free energy, the model captures near-wall density reorganization and thin film stabilization during phase change, thus providing a consistent framework for examining nanoscale interfacial dynamics while maintaining a clear connection to molecular-level physics.

Viscoelastic swirling jets hold significant potential for engineering applications, such as atomizers and combustors. This study carries out theoretical analysis and numerical simulations to investigate the role of elasticity in the instability of swirling jets. It is found that the elastic force could suppress jet instability. However, the suppressing effect is weakened as elasticity gradually increases, thereby enhancing the influence of centrifugal force. This results in modes with the higher azimuthal wavenumbers dominating the jet breakup.

This study explores the transport of microparticles in a three-dimensional stationary vortex generated in a microfluidic cross-slot geometry. Experiments and numerical simulations show that, when increasing particle diameter, particles are progressively excluded from the vortex core. Initially, small particles follow a Burgers vortex-like self-similar motion, but for larger particle diameters, deviations from this trend emerge due to fluid inertia and finite-size effects.

Shock-induced breakup of liquid droplets is critical to the development of novel detonation-based propulsion devices, but the fundamental breakup processes are difficult to quantify experimentally. To address this need, this work presents three-dimensional multiphase simulations of shock-induced catastrophic droplet breakup, analyzing the droplet deformation, displacement, and distributions of secondary droplet sizes. The results agree well with existing experimental data and provide insights into how instabilities along the droplets’ surfaces help to facilitate their atomization.

Liquid jets injected into crossflows characterized by very low gaseous Reynolds numbers, low momentum flux ratios, and extreme density ratios experience intensified bending, rapid surface stripping, and early column fracture compared with classical air-flow conditions. In this regime, ligament formation becomes strongly aligned with the crossflow, and instability waves wrap around the entire jet circumference rather than remaining on the windward side. Our results show that Kelvin–Helmholtz, rather than Rayleigh–Taylor, controls the breakup dynamics, with surface wavelengths remaining independent of the Weber number.

The pronounced particle streaks will be observed when particle-laden turbulent boundary layers sweep over a concave surface. Its underlying dynamical mechanism provides new insights into the interaction between turbulence and particles.

Dense suspensions of finite-size particles exhibit strongly heterogeneous modulation of wall-bounded turbulence. Using interface-resolved Direct Numerical Simulations over a wide range of particle inertia, we show that low-inertia particles homogenize near-wall flow topology and restore a symmetric tear-drop structure in the Q–R plane, whereas high-inertia particles generate asymmetric wake-driven topology, suppress Reynolds stresses, and reorganize the momentum and energy budgets. These results establish a clear physical link between particle-migration-induced flow topology and the regional modulation of turbulence in dense suspensions.

Understanding how emulsions transport heat under buoyancy forcing is essential in many natural and industrial flows, yet the role of interfacial physics remains poorly explored. Using mesoscale lattice Boltzmann simulations, we compare stabilized and non-stabilized liquid–liquid dispersions in Rayleigh–Bénard convection. While their global heat transfer is similar, stabilized emulsions sustain stronger small-scale heat-flux fluctuations, revealing how interfacial stabilization reshapes convective dynamics at the droplet scale.

This work seeks to understand the importance of nonlocality in modeling scalar transport in turbulent Rayleigh-Taylor instability (RTI) at different Atwood numbers. We apply the Macroscopic Forcing Method to determine moments of the eddy diffusivity from high-fidelity numerical simulations of RTI. We additionally present a framework for incorporating nonlocality for modeling RTI at different Atwood numbers. We find that nonlocality is important for modeling RT and appears to increase in importance with Atwood number.

A nonlocal expression for the Reynolds stress and passive vector flux was investigated using a direct numerical simulation (DNS) of homogeneous isotropic turbulence with an inhomogeneous passive vector. The Green’s function for the passive vector was evaluated to obtain the nonlocal eddy viscosity. The nonlocal expression for the passive vector flux agreed with the DNS data, and the nonlocal effects accounted for the overestimation by the local expression, as well as the phenomenon of counter-gradient diffusion. A model for the nonlocal eddy viscosity was also proposed and validated using the DNS data.

A method for determining the friction velocity in wall bounded flows has been extended for use with rough surfaces. The original method (Dixit et al.) is applicable to smooth wall flows, and requires only the mean streamwise velocity profile at a single location. It does not rely on any assumptions about the shape of the profile, and is applicable in both zero and non-zero pressure gradients. The new method is the same as the original, with the exception that the kinematic viscosity is replaced with an effective viscosity that is proportional to the product of the roughness height and flow velocity.

The isosurfaces of zero fluctuating streamwise velocity within the turbulent region of spatially developing compressible mixing layers are defined as the interfaces of high- and low-speed large-scale structures (LSSs). This study advances the knowledge of local flow characteristics at the interfaces of LSSs for different streamwise regions and compressibility ($M_c=0.3$ and $M_c=0.8$). For the first time, enstrophy transfer is systematically analyzed at LSS interfaces, and local topology conditioned on interface orientation is used to explain flow patterns near these interfaces.

In turbomachinery and aerodynamics, wake-induced transition can trigger turbulence in boundary layers, yet how wake-separated boundary layer interactions drive transitions is not well understood. We use direct numerical simulations to reveal a hybrid transition pathway — a fusion of classical separated boundary layer instability and wake-driven bypass transition — in boundary layers at moderate gap ratios. Unlike abrupt classical transitions, this process unfolds in distinct stages — linear amplification, nonlinear saturation, and turbulent breakdown — driven by coherent Λ-vortices synchronized with the wake. Our findings offer new insights for transition control in engineering systems.

Two new modeling components for the energy equation in ordinary-differential-equation-based wall-models are proposed in order to improve the accuracy of heat flux predictions in high-speed turbulent boundary layers: a model for the diffusion of turbulence kinetic energy, and an altered near-wall damping of the thermal eddy diffusivity. These modeling additions reduce the $a$ $priori$ error in the heat flux to within 5% across the tested conditions. A decomposition of the $a$ $posteriori$ error shows the presence of two other error sources in wall-modeled large eddy simulations, which are found to contribute up to 5-10% additional error.

We present a novel experimental technique based on Diffusing-Wave Spectroscopy that, for the first time, provides high-resolution spatiotemporal maps of energy dissipation at the boundary of a turbulent flow. The flow, generated by an impeller, exhibits Reynolds numbers in the range $1.5\times {10}^4$ to $6\times {10}^5$. Measurements were performed at the boundary of a square container, within a 5×5cm² region at the impeller height. This method enables a direct experimental characterization of the statistical properties of turbulent structures near flow boundaries, offering new insights into boundary-layer dynamics in highly turbulent regimes.

A wind-tunnel experiment mimicking an airplane in cruise flight reveals how a wingtip vortex interacts with the nearby parallel jet depending on their mutual spacing. By tracking the wake up to 20 wingspans, the work clarifies how jet placement governs entrainment, spiraling mixing, or even trapping by the vortex core, while the vortex itself remains largely resilient. The findings show increased vortex motion downstream and the influence of the deteriorated central part of the wake on this dynamic. Besides the jet is shown to sustain increased dispersion with vortex proximity. The dataset also offers valuable benchmarks for validating numerical simulations of vortex wakes.

The pair correlation function of the velocity field is calculated analytically in a rotating three-dimensional coherent turbulent vortex at distances much larger than the scale of wave forcing and below the vortex size. The function demonstrates anisotropic behavior caused by the relatively large shear flow produced by differential rotation in the vortex. The diagonal elements decrease logarithmically with distance in the streamwise direction, and power-like in radial and vertical directions, that manifest upscale energy transfer. The radial-azimuthal component, which turns into the Reynolds stress for zero distance, is short-correlated and is determined by the forcing correlation function.

Streamlines of an exact solution of the Hall-MHD equation for a spherical vortex containing toroidal magnetic flux, wrapped by a vortex sheet. The magnetic tension acts to accelerate the traveling speed of a spherical vortex. The Hall effect, originating from relative fluctuations of electrons and ions, provides further acceleration. Besides, a singular solution specific to Hall-MHD is found for the spherical vortex, which vanishes when the Hall effect is switched off.

Subsurface variations of water density enable internal waves, which play a key role in oceanic mixing and energy transport. Internal waves are affected by resonant three-wave interactions, which, in the ocean, form only when different interfaces (i.e. vertical modes) interact. For confined basins, however, a new paradigm emerges: resonant triads may form between different “sloshing” modes at a single interface, so include only the lowest vertical mode. We characterize this abundant new class of triads for the case of two-layer flows in basins of arbitrary cross section with vertical walls and discuss the implications on inverse energy cascades and internal seiching in lakes and harbors.

Turbulent zonal (east–west) jets are ubiquitous in atmospheres, oceans, and planetary interiors. Predicting their long-term nonlinear equilibration is challenging due to complex feedback effects with the underlying turbulence and waves. We address this by deriving a nonlocal closure model that parameterizes Rossby waves and their interaction with large-scale jets. Our new quasilinear model, based on a WKB expansion of the wave field, is the first purely zonal closure to naturally produce zonal jets separated by a Rhines scale. It also reproduces a transition between locally driven and globally driven jets observed in laboratory experiments.

For a stratified fluid in a closed geometry, high amplitude internal waves can be generated by forcing global modes of the domain. These waves can then destabilize through triadic resonant instability (TRI), generating secondary waves. This experimental study, together with an interpretation based on a weakly nonlinear analysis, shows that the combination of box resonance conditions and nonlinear resonance conditions governs the nature of the secondary waves, causing in some cases a significant deviation from the internal waves dispersion relation and in general complex nonlinear interaction dynamics, with multiple pairs of secondary waves appearing dynamically.

When surface gravity waves propagate through the Marginal Ice Zone, they interact with numerous floating ice floes and affect the evolution of the polar environment. This work develops a method capable of modeling wave interaction with large arrays of arbitrarily shaped ice floes, which remains highly efficient even when the number of floes becomes exceedingly large. The results show how floe geometry and spatial arrangement influence hydrodynamic forces and scattered wave energy, offering new physical insights into realistic wave–ice interactions.

Using large-eddy simulation with overset grids, this work reveals how launch angle shapes the water-exit dynamics of a neutrally buoyant sphere. The sphere exhibits nonlinear velocity attenuation, angle-dependent rotation switching, and systematic lateral deflection driven by shed vortices and image-induced pressure asymmetry. Two exit phases with distinct hydrodynamic signatures are identified. The findings deepen mechanistic understanding of water exit and aid the control of trans-medium bodies.

This paper presents a lattice Boltzmann model based on the vielbein formalism for simulating fluid flows on spherical surfaces. By capturing the underlying geometry of spherical surfaces, the model enables Cartesian treatment of velocity space while ensuring fluid trajectories stay confined to the manifold. Validated against exact solutions for sound and shear waves, and tested with shockwave and vortex dynamics, this approach enhances the study of geophysical flows and provides a robust framework for future turbulence modeling on curved manifolds.

This work presents a real-time, model-free control framework that utilizes single-step deep reinforcement learning to suppress Tollmien-Schlichting waves in a two-dimensional boundary layer. By learning an opposition-control strategy directly from sensor feedback, the controller outperforms a classical adaptive Filtered-x Least Mean Square (FXLMS) approach in both effectiveness and robustness. The results demonstrate a compact and experimentally feasible AI-based solution for real-time control of convective instabilities.

Numerical simulations of gas-liquid-solid (GLS) interactions play a significant role in many essential areas; however, several challenges related to boundary conditions and mass conservation remain. We present a GLS contact condition-enforced immersed boundary method for simulating multiphase flow problems with curved and moving boundaries. This method has been validated by simulating many challenging GLS problems, indicating that it can accurately enforce Dirichlet and Neumann boundary conditions and efficiently restrain nonphysical liquid/mass penetrations near the solid surfaces. The present method is also applied to more complex GLS problems at large density ratios O(${10}^3$).

Spectral proper orthogonal decomposition (SPOD) finds orthogonal space-time modes that best capture the 2nd order statistics of a data set. However, SPOD is limited to uniform time-resolved data. Many experimental systems are limited by camera speed. Leveraging the relaxed requirement of pairwise data for dynamic mode decomposition (DMD), we propose a pairwise SPOD algorithm that estimates SPOD modes of nonuniformly sampled data. Using 2 PIV setups in tandem, the time delay within a data pair satisfies the Nyquist criterion, but the time between pairs does not, allowing time to process the last snapshot. Computational data is used for validation and the algorithm is of general use.