The structure of liquid foams follows simple geometric rules formulated by Plateau 150 years ago. On smooth surfaces, the foam liquid channels, also called pseudo Plateau borders, are straight between vertices. We demonstrate experimentally that on rough surfaces and under some conditions that we establish, the bubble footprint exhibits a morphological transition. The footprint can adopt a zigzag shape between vertices. We rationalize the number of zigzag segments by a geometric distribution describing the observations made with the footprint perimeter and the mesh size of the asperities.

A self-aligning particle (SAP) attains near perfect alignment with the fluid lamellae of a low Reynolds number simple shear flow without application of external torques in contrast with the continuous rotation exhibited by most rigid bodies including thin fibers and disks. We characterize the robustness of the flow alignment of SAPs to secondary perturbations such as flow disturbances, Brownian motion, inter-interparticle interactions, and the presence of a wall using dynamic simulations of ring-shaped SAP geometries. The robust flow alignment of SAPs provides an alternative route to access highly aligned microstructures that are inaccessible to suspensions of traditional particle geometries.

The flow rate–pressure drop relations for laminar flow of Newtonian fluids in common geometries are well understood. However, a complete understanding of how the interplay between shear-thinning rheology and wall compliance sets the flow rate–pressure drop relation for a deformable configuration is still lacking. Here, we provide detailed quantitative comparisons between theory and experiments for the flow rate–pressure drop relation for Newtonian and shear-thinning fluids in two common deformable configurations: a rectangular channel and an axisymmetric tube. Such a comparison is of fundamental importance since it provides insight into the adequacy of the constitutive model used.

In stars and planets natural processes heat convective flows in the bulk of a convective region rather than at hard boundaries. Internally heated convection has been studied extensively in incompressible fluids, but the effects of stratification and compressibility have not been examined in detail. In this work, we study fully compressible convection driven by a spatially uniform heating source in a suite of two- and three-dimensional Cartesian, hydrodynamic simulations. We characterize how Mach, Reynolds, and Nusselt numbers scale with the characteristic strength of the internal heat source. We also measure kinetic energy power spectra and discuss the flow morphologies.

The dynamics of a liquid drop attached to a solid and subjected to arbitrary accelerations can be qualitatively reproduced with a low-dimensional model in which the drop is replaced by a cuboid. The cuboid deforms, vibrates and slides similarly to a drop, while being much simpler to describe mathematically.

Cavitation bubbles are present in a plethora of industrial and medical applications, and their understanding proves crucial to the development and tuning processes. In this study, we perform three dimensional direct numerical simulations (DNS) of multiple cavitation bubbles driven by pressure waves. Similarly observed in previously conducted experiments, it is found that these bubbles can exhibit an asymmetry in the direction of wave propagation. We show that this asymmetry is a consequence of the force induced by the wave on the bubbles, due to the finite speed of sound in the liquid medium.

This work investigates the interaction of a viscoelastic solid and a nearby collapsing bubble, using numerical simulations. The Deborah number, which quantifies the relative time scales of solid deformation and bubble dynamics, is found to effectively characterize the bubble-solid interaction. Analyses on the temporal distribution of energy components and the imbalance in pressure distribution surrounding the bubble reveal the rationale behind changes in bubble behaviors with respect to the viscosity and elasticity of the solid. Furthermore, the solid deformation induced by an expanding and collapsing bubble and the shape of the crater formed by a bubble jet are examined.

This numerical simulation investigates the vertical convection of liquid metal with varying magnetic fields and wall conductivities. The applied horizontal magnetic field alters plume dynamics and topology, leading to a more coherent large-scale flow structure but weakening convection through Joule dissipation. This competition between rectification and magnetic damping determines the magnetic field’s impact on heat transfer, with the quasi-two-dimensional state being the threshold. Our analysis demonstrates that while the plume area remains constant, condensation of coherent structures enhances horizontal heat transport per unit area, significantly improving overall heat transfer.

The existing hybrid-scale micro-continuum approach faces difficulties in numerical diffusion issues at the gas-liquid interface and the solid boundary. This article proposes a multiscale compressive Continuum Species Transfer (MC-CST) scheme and a concentration extrapolation algorithm to improve the accuracy of two-phase reactive flow simulations. Furthermore, a case study simulating calcite dissolution in a porous medium is presented to underscore the importance of multiscale fluid-rock interactions for an in-depth comprehension of the dissolution regime.

We examine how fluids mix in three-dimensional (3D) porous materials due to differences in density which represent one mechanism of underground carbon dioxide storage. The experiment closely matched the simulation in terms of the patterns and speed of mixing. Interestingly, the experiment reveals columnar plumes self-organizing into a reticular pattern, previously seen only in 3D simulations. Results demonstrate quantitative matching over time in concentration, variance, scalar dissipation rate, and dissolution flux. A new relation between dissipation rate and flux is established, highlighting a 30% higher flux in 3D versus 2D systems, affirming prior estimations.

A three-dimensional boundary layer is the seat of a spatially amplified instability, the crossflow instability, which is encountered for example on swept wings. Beyond a certain amplitude, the flow destabilizes again through a secondary instability. The objective of this work is to show that by using a discrete spatial Floquet method (Block method) from a certain transversally periodic elementary pattern, there can exist a large-scale modulation linked to amplification mechanisms involving detuned modes.

We present the automated discovery of control strategies for drag reduction in cylinder flows. Reinforcement Learning algorithms discover control strategies for two-dimensional configurations that generalize to three dimensional flows. We discuss the physical processes involved in the drag reduction mechanisms along with their generalization capabilities. This work demonstrates a practical approach to handling the computationally intensive task of deploying Reinforcement Leaning for bluff body flow control problems: namely train in 2D and control in 3D.

A physical mechanism has been identified through which a small surface add-on to a swept (angled-back) aircraft wing can delay the development of turbulent disorganized airflow. To reduce the aerodynamic drag of an aicraft, it is important to maximize the extent of smooth laminar airflow around its wings. The mechanism, termed “reverse lift-up effect”, dampens out the critical instability waves leading to turbulent airflow over swept aircraft wings under certain conditions. This may be achieved through particular design of a surface pattern on the aicraft wings. Overall, the present findings aim to contribute to novel design methodologies for future generations of laminar aicraft wings.

We systematically examine how heat conduction between two fluids at different temperatures, large contrasts in transport properties, and sudden changes in transport properties can affect the fully compressible Rayleigh-Taylor instability (RTI) using direct numerical simulations. These variations cause departures from the classical self-similar development of the RTI, along with misalignment between regions of mixing and regions of most intense turbulent activity. Under certain conditions, dynamical quantities such as vorticity and dissipation appear to depend only on the transport properties and not on past flow history.

Shear flows in contact with compliant boundaries exhibit a rich dynamics involving traveling-wave-flutter, Tollmien-Schlichting, as well as divergence instabilities. In this context, maximum transient growth effects are the result of optimal energy exchanges during the fluid-structure interaction process. The present investigation studies the detailed contribution of the different interacting modes. In particular, it is found that the optimal gain may be associated with a large-amplitude oscillatory behavior and that wall-compliance enhances this phenomenon.

Unsteady flows in curved pipes are ubiquitous in science and engineering but their stability characteristics are not well understood in most cases of practical interest. We study the linear stability of pulsatile flow in the archetypal configuration of a toroidal pipe, which appears e.g. in aortic blood flow. The Floquet stability analysis of the harmonically forced system reveals that the curvature leads to nonlinear interactions in the baseflow andconsiderable stabilization that can be orders of magnitude larger than in the corresponding planar case. The figure shows a typical snapshot of the streamwise velocity field in a torus subject to a pulsating pressure gradient.

Seed dispersal strategies exemplify the role of morphology in defining falling paths. Here, macroscopic geometry effects are systematically studied by considering trajectories of annular disks. Via linear stability analysis, we identify the stability boundary for vertical fall, with a non-monotonic behavior of the critical falling velocity with the hole size, before increasing for large holes. Nonlinear simulations confirm linear analyses and suggest strategies for annular seed release at different heights, with the emergence of paths possibly beneficial for controlled positioning or advantageous for covering large lateral distances, depending on the hole size and disk weight.

The impact and freezing process of a water droplet are studied experimentally, incorporating the presence of frost and salt. A distinct transition of the spreading dynamics is observed, altering from the well-known 1/2 inertial scaling law to a 1/10 capillary-viscous scaling law. By considering the effect of impact inertia, partial-wetting behavior, and salinity, the mechanism of this transition is elucidated, and a unified model for predicting the droplet arrested diameter is proposed.

Electrocapillary flows in a classical experiment of Melcher and Taylor are studied. The computed streamlines qualitatively represent the experimental image. With the increase of electrocapillary forcing, the main circulation localizes near a boundary with a larger electric potential. When a dielectric liquid is replaced by a poorly conducting one, the system becomes non-isothermal owing to the Joule heating, and the flow is driven also by buoyancy and thermocapillary convection. The results show that consideration of the two-phase model is mandatory. The Lippmann equation, connecting electrically induced surface tension with nonuniform surface electric potential, is numerically verified.

Stokes flow driven by rotation of two parallel cylinders inside a cylinder of large radius $R_0$ is investigated, and the flow in this triply-connected domain is determined numerically, with particular focus on the narrow-gap situation, when the local behavior is well described by lubrication theory. The asymptotic situation for infinite $R_0$ is inferred (i) when the cylinder axes are unconstrained, and (ii) when they are held fixed. Contributions to the far-field are identified: a torquelet and a radial quadrupole in the counter- and co-rotating cases, respectively. In the former case, when the cylinders make contact (zero gap) a contact force acting on the cylinder pair is identified.

A flow progression from laminar flow slippage to rotational vortices in cavitied microchannels is shown to cause a reversal of flow resistance, i.e. a reduced flow resistance at low Reynolds numbers compared to an unmodified microchannel but a comparatively higher flow resistance at high Reynolds numbers. Furthermore, an earlier transition of initial laminar flow to turbulent flow is suggested to be triggered by instabilities generated along shear layers, formed between the mainstream flow and rotational vortices in each cavity that modifies the sidewalls of microchannels.

We develop a model that describes the permeance of simple fluids as well as small hydrocarbon molecules through nanoporous, atomically thin membranes. The model is in agreement with molecular dynamics simulations for a wide range of pore sizes, including pores approaching the steric exclusion limit, as needed for understanding separation processes using such membranes.

The diffusiophoretic velocity of a particle depends nonlinearly on solute concentration. We examine the implications of this nonlinearity in a dead-end pore, which is a geometry found both in nature and in microfluidic devices. We define the efficiency of injection and withdrawal processes, then describe particle dynamics with a step-like change in solute concentration at the pore inlet (commonly used in experiments) and with time-dependent boundary solute concentration. We show that particle migration becomes linear with slow transitions in solute concentration and demonstrate changes in particle dynamics with oscillatory solute concentration at the pore inlet.

The concept of apparent coefficient of restitution is extended, to describe the wall-bouncing of a cylinder falling in a viscous fluid. When the Stokes number (St) is increased, the coefficient of restitution increases from 0 to 1 over two decades of St, with a critical St${}_c$=75±25 (compared to St${}_c\approx 10$ for spheres). While rationalizing results from experiments and numerical simulations, we evidence the relevance of lumping the complex details of physical phenomena involved during contact into a simple concept based on the contact apparent roughness and elasticity. The increase of dissipation associated with the contact elasticity is well captured by a model based on solid contact time scale.

We combine experiments and numerical simulations to investigate the interaction between a rigid fiber and a triangular obstacle in a microfluidic channel. We find different dynamics depending on the initial position and orientation of the fiber. We show that these dynamics are dictated by the fiber configuration in the vicinity of the obstacle. Some dynamics induce a cross-stream migration which grows with the fiber length. Our findings could in the future be used to design and optimize microfluidic sorting devices to sort rigid fibers by length.

Acoustic fields are used in numerous applications to induce movement of suspended particles. We develop a rigorous theory that systematically unifies inviscid acoustophoresis with viscous streaming. The theory connects particle motion to a generalized form of the secondary radiation force, which depends on the Stokes layer thickness around the particle, and the contrast of density and compressibility between the particle and the fluid. We identify a reversal of particle motion when inertial and viscous forces are comparable, validated with numerical solutions. This has significant implications for applications involving particle sorting or focusing based on size or material properties.

The presence of a dispersed phase substantially modifies small-scale turbulence. Here we present a comprehensive mechanistically based model to predict turbulence modulation, the predictions of which, compared with particle-resolved simulations and experiments, is shown.

This study focuses on how Lagrangian coherent structures (LCSs) control solute dispersion in heterogeneous poroelastic media (HPM). We show how interactions between medium compressibility, conductivity heterogeneity, and periodic forcing give rise to complex flows and diverse LCS types (KAM islands, chaotic saddles, etc) that have profound impacts on diffusive solute transport (main image) that do not arise in the steady counterpart (inset). Strongly anomalous transport impacts both spatial moments and residence time distributions and persists at low Péclet numbers. This study reveals the complex transport phenomena that can arise in HPM and shows how LCSs govern solute dispersion.

We report experiments on the dissolution of a single particle during its sedimentation in a quiescent aqueous solution in the regime of low Reynolds and high Péclet numbers. We use butyramide, a chemical which does not change the density of water when it dissolves. The particle shrinks at a rate independent of its initial radius, in agreement with the model that we derive assuming Stokes drag and a mass transfer rate given by Levich (1962). This model becomes quantitative when including two correction factors to account for the non-sphericity of the particle and for the inclusions of air bubbles inside the particle.

Two possible and physically meaningful scalings for near-wall streamwise turbulence intensity are derived using velocity spectra. First, if viscous wall effect on the wall-attached inactive motions from the log and outer layers is negligible enough so that the related spectrum does not diminish with Reynolds number, the near-wall turbulence intensity grows indefinitely and scales as ln Re${}_{\tau }$. Whereas if it is strong enough so that the near-wall turbulence intensity remains finite at all Reynolds numbers, it scales as 1/ln Re${}_{\tau }$. The existing measurement data favors the latter scaling, but the issue will be settled by the availability of such data at higher Reynolds numbers.

Taylor’s hypothesis of frozen flow is revisited in homogeneous turbulent shear flow by examining the cancellation properties of Eulerian and convective accelerations at different spatial flow scales. The anti-alignment in the joint pdf given in the figure shows that the Eulerian and convective accelerations cancel for small scales of the turbulent motion, indicating that Taylor’s hypothesis holds. This is not so for the large-scale motion of homogeneous turbulent shear flow.

In this work, we developed a novel framework for incorporating the near-wall non-overlapping domain decomposition (NDD) method with the machine learning technique. It allows the solution to be calculated with a Robin-type (slip) wall boundary condition on a relatively coarse mesh and then be corrected in the near-wall region by solving the thin boundary-layer equations on a fine subgrid. Through an estimated turbulent viscosity profile provided by a neural network, the proposed method can be easily extended to different turbulence models and achieve commendable accuracy for the test cases of turbulent wall-bounded flows at various Reynolds numbers.

We propose a novel approach to accurately model complex flow, ultimately predicting the behavior of a turbulent jet by molding a set of velocity signals input from an idealized flow (readily available from numerical databases online). The model uses fundamental properties of the jet, such as velocity means and standard deviations, easily accessible from textbooks, experiments, or low-order simulations (RANS, LES). The modeled jet reproduces many subtle and intricate properties of the turbulent flow, including the intermittent extreme events known to exist in turbulent flows, which have been typically thought of as not capable of being captured with current modeling techniques.

Turbulent flows comprise a multitude of scales which make them extremely difficult to analyze and simulate. In this work, we study homogeneous isotropic turbulence using a novel approach which solves the Navier-Sokes equations on a reduced set of modes that are sampled stochastically. The complementary set are solved using trivial dynamics. This method, called Selected Eddy Simulations, is able to capture broad dynamics of turbulence with just 10% of resolved modes, suggesting the turbulence attractor may be smaller or more robust to modeling than previously thought. This result also holds promise for developing alternative low-cost numerical approaches to study turbulent flows.

In this work, we experimentally investigate turbulence ingesting rotor noise under various thrusting states. The broadband noise caused by turbulence ingestion is found to dominate at the normalized frequency range of $fR/U_{\mathrm{\infty }}$ = 20 to 80 when the rotor is under low-thrusting conditions. Results also suggest that the turbulence ingestion broadband noise can be scaled by Mach number scaling of $M_{\mathrm{\infty }}^2$$M_c^4$, where $M_{\mathrm{\infty }}$ is the freestream Mach number, and $M_c$ is the corresponding blade tip Mach number.