2010 Image Gallery

Corrie White
Igor Kliakhandler

Let us mix regular bubble liquid (bought at any store), slight amount of glycerin, and green food color (from any store). Then we play a little with lighting, camera, angle of shooting, zoom, etc. as to prepare the best shooting environment. We blow a bubble on the surface of the mixed colored bubble liquid, and let liquid drops fall from above on the bubble. After all that, let us record a very large number of pictures, and repeat that experiments, while shooting, many, many times. (By the way, shooting cannot be done continuously, as it is necessary to wait, each time, for liquid to become still...)

We hope for lucky coincidence that, when something interesting will be happening at collision of bubble and drops, we will record that. And this picture is the result of such patience, luck and experimentation.

As could be seen from the image, the bubble is serene, but the burst front is unstable, and spreads many small drops around it - a known fact for anybody who ever burst a bubble... Many serious applications are related to the collapse of liquid sheets and jets, including fuel in combustion engines.

Many more fascinating images may be found on Liquid Drop Art.

Reporters and Editors
This image may be freely used in any media. Please credit the authors. Use of images from Liquid Drop Art is described on the website.

Ching-Yao Chen
W.-L. Wu
National Chiao Tung University
Hsinchu, Taiwan

Jose A. Miranda
Universidade Federal de Pernambuco
Recife, Brazil

We report an experimental study of an interfacial pattern formation which occurs during the spreading of an immiscible thin ferrofluid drop subjected to a radial magnetic field. A magnetically induced selection mechanism is illustrated by typical time evolutions (from left to right) of a drop of various initial diameters (demonstrated on different rows).First, a perpendicular field is applied generating an array of numerous sub-scale droplets as shown in the first column. Soon after this, the perpendicular field is turned off, and the radial field is immediately switched on. This leads to the collapse of the tiny peaks, followed by the coalescence of the sub-scaled droplets, creating a symmetrically perturbed structure. Subsequently, fingers start to spread out due to the action of a radial magnetic force. This demonstrates the capability of tuning the ultimate number of growing fingers by prescribing a proper initial perturbation.

This work is funded by the National Science Council of Taiwan (Republic of China).

Reporters and Editors
Reporters may freely use these images. Credit: Chen and Wu, National Chiao Tung University, Taiwan; Jose A. Miranda Universidade Federal de Pernambuco.

Navish Wadhwa
Sunghwan Jung
Department of Engineering Science and Mechanics
Virginia Polytechnic Institute and State University
Blacksburg, Virginia

Intuition tells us that two or more jets of the same fluid impinging into each other will readily coalesce to form a single mass of fluid. Fluid chains, sheets and fish-bone structures resulting from oblique collision of fluid jets are well studied phenomena [1, 2]. Here, we report the non-coalescence and rebounding of fluid jets off each other- something that completely defies common intuition.

A sample picture (Figure 1) shows two jets of silicone oil (viscosity 10 cSt at 25 °C) with diameter roughly equal to 500 µm, impinging obliquely onto a vertical jet of same fluid and same diameter. We observe that instead of coalescing, the jets from the sides rebound off the middle jet. This is due to lubrication effect of a thin film of air maintained between the jets. The motion of the jets keeps replenishing the air resulting in indefinitely sustained non-coalescence of the jets of same fluid [3]. Similar phenomenon can also be observed in the interaction between a jet and drops of the same fluid, an example of which is shown in Figure 2.

In this case we observed that the drops plunged into the jet from two sides without coalescing into it, causing it to bend at two locations.

References
[1] J. W. M. Bush and A. E. Hasha, J. Fluid Mech. 511, 285 (2004).
[2] G. Taylor, Proc. R. Soc. A 259, 1 (1960).
[3] P. DellAversana, V. Tontodonato, and L. Carotenuto, Phys. Fluids 9, 2475 (1997).

Reporters and Editors
These images may be used with the permission of BIF lab in the department of Engineering Science and Mechanics at Virginia Tech. Please credit the source.

Sandeep Rana
Marcus Herrmann
Arizona State University
Tempe, Arizona

Visualization of the breakup of an aircraft engine turbulent liquid fuel jet injected into a compressed turbulent gaseous cross-stream.

We present a visualization of the primary atomization of a turbulent liquid fuel jet injected into a compressed turbulent gaseous cross-stream representative of aircraft gas turbine engines and augmentors. Detailed numerical simulation results were obtained using a finite volume, balanced force, incompressible LES/DNS flow solver. Grid resolution in the primary atomization region is a constant 64 grid points per injector diameter in the flow solver, and 128 grid points per injector diameter in the level set solver, resulting in grid sizes of 110 million control volumes for the flow solver and a theoretical maximum of 6.7 billion nodes for the level set solver. We employ a hybrid Eulerian/Lagrangian approach for the liquid in that broken off, small, nearly spherical liquid drops tracked by the Eulerian level set approach are transferred into Lagrangian point particles to capture the evolution of the liquid spray downstream of the primary atomization region.

The simulation results clearly show the simultaneous presence of two distinct breakup modes. While the main column of the jet is subject to a wavy instability mode, resulting in the formation of bags that break under the influence of the cross stream flow at the end of the liquid core, ligaments are formed on the sides of the jet near the injector exit that stretch and break. The flow in the wake of the bending liquid jet is characterized by strong turbulence. Comparison of the simulation results to experimental data show that mean jet penetration is in excellent agreement to experimental correlations and drop size distributions converge under grid refinement (M. Herrmann, J. Eng. Gas Turb. Power, 132(2), 2010).

This work is funded in part by Cascade Technologies Inc under NavAir SBIR N07-046 and supported by Arizona State University's HPC Initiative.

References
M. Herrmann, J. Eng. Gas Turb. Power, 132(2), 2010.

Reporters and Editors
Reporters may freely use these images. Credit: S. Rana & M. Herrmann, Arizona State University (2010).

Jonathan Varkovitzky
Chris Svedman
Peter Mitrano
Jean Hertzberg
Department of Mechanical Engineering
University of Colorado, Boulder
Boulder, Colorado

Students in a course on Flow Visualization (http://flowvis.colorado.edu/) attempted to create interaction between two vortex lines by putting two holes in the bottom of a round bucket. Adding food dye to the flow showed that only the hole in the center of the bucket created the swirling flow of the common ‘bathtub’ vortex. Flow went straight down the off-center hole without swirl. Even when the off-center hole was made larger than the centered hole the vortex continued to form only over the centered hole. The ‘bathtub’ or ‘drain’ vortex is most famous for swirling counter clockwise in the northern hemisphere, and clockwise south of the equator, but in the 1960s this was shown to be a weak effect (Shapiro 1962), and you can make a drain vortex flow either direction by swishing the flow as you like.

More recently, studies of bathtub vortexes use a rotating bucket to study details of the flow and make mathematical models predicting what it will look like (Andersen et al. 2003; Tyvand and Haugen 2005; Yukimoto et al. 2009). However, all of these studies only have one hole centered in the bucket. This image raises the question of why only one vortex forms when the drains are close together.

References
Andersen, A., Bohr, T., Stenum, B., Rasmussen, J.J., Lautrup, B.,( 2003). Anatomy of a Bathtub Vortex. Physical Review Letters 91: 1045021-1045024.

SHAPIRO, A.H.,( 1962). Bath-Tub Vortex. Nature 196: 1080-1081.

Tabeling, P., Zocchi, G., Libchaber, A.,( 1987). An Experimental Study of the Saffman-Taylor Instability. Journal of Fluid Mechanics Digital Archive 177: 67-82.

Tyvand, P., Haugen, K.,( 2005). An impulsive bathtub vortex. Physics of Fluids, Phys. Fluids (USA) 17: 62105-1.

Yukimoto, S., Niino, H., Noguchi, T., Kimura, R., Moulin, F.Y.,( 2009). Structure of a bathtub vortex: importance of the bottom boundary layer. Theor. Comput. Fluid Dyn. 24: 323-327.

Reporters and Editors
This image may be freely used for educational purposes as long as the authors are credited.

Matthieu Roché¹, Zhenzhen Li¹, Ian M. Griffiths^{1, 2}, Arnaud Saint-Jalmes³, Howard A. Stone¹

¹Department of Mechanical and Aerospace Engineering,
Princeton University Princeton, New Jersey

²Oxford Centre for Collaborative Applied Mathematics,
University of Oxford
Oxford, United KIngdom

³Institut de Physique de Rennes
Rennes
Cedex, France

Spreading of liquids on solid or liquid surfaces is of interest in many applications such as surface coating and to help neonates who suffer from respiratory distress syndrome. The addition of surface-active molecules, such as soap, in the liquid to be spread influences dramatically the spreading behavior. Indeed, these molecules induce a strong flow (known as a Marangoni flow) on the expanding surface of the spreading liquid because of their inhomogeneous concentration.

In the case pictured here, an olive-oil-in-water emulsion containing soap is constantly supplied through a steel needle on the surface of a layer of ultra-pure water. A front of oil droplets spreads until it reaches a characteristic position that depends on the chemical affinity between water and soap molecules. Oil droplets move with speeds on the order of 0.5 m/s across the transparent area (black), which has a diameter of approximately 60 mm. When droplets reach the white oil-droplet-rich external zone, they slow down dramatically and induce spectacular vortices similar to what is seen in turbulent soap films. These vortices exist as long as emulsion is flowing. Once the flow stops, the transparent area collapses to cover the surfactant-poor zone in the middle.

This experiment shows that a flow induced by an inhomogeneous distribution of surface active molecules can transport objects. This image also provides evidence that such a transport is efficient only on a characteristic length scale.

Reporters and Editors
Reporters may freely use this image. Credit: Matthieu Roché, Zhenzhen Li, Ian M. Griffiths, Arnaud Saint-Jalmes and Howard A. Stone (2010).

Terry Tuk
Julie Macdonald
Igor Kliakhandler

Have you ever wondered how water strider is able to move, glide, and even jump on the water surface? And there are good reasons to wonder. Conventionally, it is thought that it is wax on the body of water strider that gives him the floatability and mobility. However, the best water-repellent substances (similar to organic oils or waxes) cannot nearly account for that.

Close investigation shows that hairy legs, and very small hairs on strider’s body are in fact responsible for its remarkable floatability and water-repellent properties. This is a vivid manifestation of Cassie’s law for surface wettability of heterogeneous materials. In fact, presence of small hairs is universal for insects and many plants' leaves. It is plausible that Nature used this mechanism to repel ubiquitous water and protect insects and leaves from infestation and rotting. Future generations of smart small floating devices will most likely use similar phenomena to move on the water surface.

Reporters and Editors
Reporters may freely use this image. Credit: Terry Tuk, Julie Macdonald, Igor Kliakhandler (2010).

K. K. Varanasi
A. T. Paxson
Massachusetts Institute of Technology
Cambridge, Massachusetts

T. Deng
M. Hsu
N. Bhate
GE Global Research
Niskayuna, New York

Superhydrophobic surfaces are extremely water repellent, which makes them valuable for applications in energy conversion and water desalination. Some examples of superhydrophobic surfaces can be found in nature, such as the leaf of a Lotus plant and the skin of a Brazilian pygmy gecko.

In order to shed water so easily, these surfaces have multitudes of sharp microscopic bumps that support a droplet, as if resting on a bed of nails. If a water drop is able to sink into the valleys between these bumps, it becomes stuck. This becomes a big problem during condensation onto these superhydrophobic surfaces. Usually, microscopic water droplets condense randomly (Figure A) and begin growing in the valleys between the bumps, so by the time it grows into a large drop, it is strongly stuck to the surface (Figure B).

Namib beetles have developed a strategy to overcome this problem in order to drink water in the desert. Their mostly waxy bodies are patterned with tiny wax-free patches. Condensation occurs only at these patches, so when the drops grow larger, they are only stuck at very few places and can roll off into the beetle's mouth.

Similarly, if we pattern a surface with hydrophilic (water-loving) and hydrophobic (water-hating) regions, preferential condensation occurs on the hydrophilic regions (Figure C). By placing these hydrophilic regions on the tops of hydrophobic bumps, we can preferentially condense on bump tops, prevent microscopic drops from condensing between the bumps (Figure. D) and overcome condensation-related limitations of superhydrophobic surfaces.

The research is partially supported by General Electric and the MIT Energy Initiative.

References
[1] Quere, D (2005) Non-sticking drops, Rep. Prog. Phys. 68:2495-2532
[2] Varanasi, K. K., Hsu M., Bhate N., Yang W., Deng T. (2009) Spatial control in the heterogeneous nucleation of water vapor, APL, 95(9):094101-094101-3
[3] Parker, A. R. and Lawrence, C. R. (2001) Water capture by a desert beetle, Nature, 414:33-34

Reporters and Editors
Reporters may freely use these images. Credit: K. K. Varanasi, A. T. Paxson, T. Deng, M. Hsu, N. Bhate (2010).

Alex Felce
Thomas Cubaud
Stony Brook University
Stony Brook, New York

This figure displays the shape instability of a viscous drop (silicone oil, viscosity: 50 cSt) falling in a liquid (isopropanol) having a low interfacial tension (scale bar: 1 mm). Dye is mixed with silicone oil to help visualize slender capillary structures. To form the drop, oil is initially deposited at the surface of the solvent. As most of the oil sinks, the droplet remains connected to the surface by a small viscous thread. The subtle balance between viscous and capillary forces leads to the formation of multiple vortex rings joined by complex and regular catenoid-like thin films. This ephemeral flow architecture is observed for a narrow range of oil viscosity.
The research is supported by the National Science Foundation.

Reporters and Editors
Reporters may freely use this images. Credit: Alex Felce and Thomas Cubaud, Stony Brook University (2010).

Leopold Grinberg
Dmitry Fedosov
Bruce Caswell
George Em Karniadakis
Brown University,
Division of Applied Mathematics
Providence, Rhode Island

Joseph A. Insley
Mike Papka
Argonne National Laboratory
Argonne, Illinois

The blood flow dynamics in small arteries is strongly affected by the presence of red blood cells (RBC). Healthy red blood cells have the ability to deform and even squeeze through arterioles significantly smaller than undeformed cells. However, sick red blood cells loose the ability to deform, they become stiffer and may clot small arteries and stop local blood circulation. To better understand the behavior of blood flow containing healthy and sick red blood cells, researches from Brown University modeled the cells using dissipative particle dynamics method (DPD), where a membrane of each cell is modeled as a collection of particles connected by visco-elastic spring[1]. The blood plasma was also modeled by such virtual particles.

In this figure the red cells represent the healthy RBCs, while the blue cells represent sick (malaria-infected) RBCs. Each blood cell is represented by 500 DPD-particles; small dots – DPD particles representing the blood plasma, the particles are colored with respect to their velocity. Instantaneous streamlines and contours (shown in four slices) represent the average velocity of the blood flow (red - high, and blue - low velocity).

Blood flow modeling is work is funded by NIH, grant NIH R01HL094270.
NSF funds developing visualization tools, grant NSF OCI-0904288.

References
[1] D. Fedosov, B. Caswell and G.E. Karniadakis, Biophysical Journal, Volume 98, Issue 10, 2215-2225

Reporters and Editors
Reporters may freely use this image. Credit: D. Fedosov, B. Caswell and G. E. Karniadakis (Red blood cell modeling) and L. Grinberg, J. A. Insley and M. Papka (Visualization) (2010).

S. P. M. Bane
R. Mével
J. E. Shepherd
California Institute of Technology

An explosion is ignited in a gaseous mixture of hydrogen (H2) and nitrous oxide (N2O) at 40 kPa by an electric spark between two electrodes. The flame front propagates outwards and initially maintains an approximately spherical shape. The flame front is perturbed as it passes over the electrodes, resulting in “cracks” that grow across the flame surface as the perturbation is amplified by the thermo-diffusive instability.

As the flame continues to grow it becomes subject to hydrodynamic instability as well. By approximately 24 ms after the flame was initiated, the flame surface is completely covered by a cellular structure of instabilities.

Thermo-diffusive and hydrodynamic instabilities and the effects of flame stretch all contribute to the unstable nature of the flame. The development and appearance of the unstable flame front is affected by the mixture composition and initial conditions. Flame instabilities cause a flame to accelerate, which can lead to larger peak pressures or possible transition to a detonation. This H2-N2O flame is an example of an extremely unstable explosion.

Reporters and Editors
Reporters may freely use this image. Credit: S. P. M. Bane, R. Mével, and J. E. Shepherd, California Institute of Technology (2010).

Jens Kasten
Ingrid Hotz
Hans-Christian Hege
Jan Reininghaus
Zuse-Institut Berlin
Berlin, Germany

Kilian Oberleithner
Technische Universität Berlin
Berlin, Germany

Bernd R. Noack
Institut Prime - CNRS
Université de Poitiers
Poitiers, France

We consider a numerical simulation of a weakly compressible 2D flow over a cavity (courtesy: M. Samimy). The flow over the cavity (yellow) is from left to right, the time is represented by the third dimension. Focus is placed on the temporal evolution of the vortices. These vortices (blue vortex cores) are identified as minima of the acceleration magnitude following the feature extraction concept of finite- time topology. The spiraling curves represent fluid particle paths in the vortical regions. The volumetric smoke-like regions indicate a range of large acceleration magnitudes. The vortices originate at the leading edge and move through the cavity over the trailing edge. The halo of each vortex consists of spiraling particles with large acceleration values.

This work is funded by the German Research Foundation (DFG): Emmy Noether Programm, SFB 557 and NO 258/2

References
Kasten et al., 2010, in Pascucci et al., Springer

Reporters and Editors
This image can be used freely by reporters so long as they credit all authors and their affiliations.

Sean C. Garrick
Jun Liu
Department of Mechanical Engineering
University of Minnesota - Twin Cities
Minneapolis, Minnesota

The rising need for clean, renewable energy sources has led to recent studies on hydrogen production via hydrolysis. The idea is to utilize metal nanoparticles to split water into hydrogen and oxygen. The flows consist of hot metal (zinc) vapor issuing into cooler, inert argon gas. As the zinc vapor cools, crystals nucleate, grow and are transported throughout the flow-field. Further downstream, these particles come into contact with water vapor and undergo a surface reaction to produce zinc-oxide and hydrogen gas. We utilize direct numerical simulation and high-performance computing to assess the effects of fluid mixing on crystal formation and growth, and ultimately hydrogen production.

The image shows contours of the nanocrystal nucleation rate are shown at different downstream locations across the jet. When the jet becomes turbulent, the rate of mixing increases significantly and nanocrystals form wherever the hot zinc vapor mixes with the cooler argon gas. The crystals are between 0.7 and 1.5 nanometers in diameter, with a large number of the zinc molecules located on the surface of the crystal. The fluid vortices act to increase the rate of mixing and the number of crystals produced. Turbulent flows are very sensitive to small-scale perturbations and the patterns observed due to the non-linear nature of the underlying physics and chemistry. The use of high-performance supercomputers helps us to learn how the crystals form and potentially control their size and structure.

Funding is provided by the Initiative for Renewable Energy & the Environment and the Institute on the Environment at the University of Minnesota. Computational resources are provided by Minnesota Supercomputing Institute.

Reporters and Editors
Reporters may freely use these images. Credit: Professor Sean Garrick, Computational Transport Phenomena Laboratory, University of Minnesota (2010).

Enrique Soto
Centro de Ciencias Aplicadas y Desarrollo Tecnológico,
Universidad Nacional Autónoma de México

Alicia Aguilar-Corona
Universidad Michoacana de San Nicolás de Hidalgo
Morelia, Michoacán,
México

Amós Benjamín Domínguez Gómez
Instituto Tecnológico de Ciudad Madero
Tamaulipas, México

Roberto Zenit
Instituto de Investigaciones en Materiales
Universidad Nacional Autónoma de México

An air stream is injected into a packed bed immersed in water. The refractive index of the water an the packed bed are quite similar and the edges of the spherical particles can be seen. Two distinctive regimens can be observed. The first one, for low air flow rates, which is characterized by the percolation of the air thought the interstitial space among particles. And the second one, for high air flow rates, which is characterized by the accumulation of air inside the packed bed without percolation, it can be observed that the bubble pull apart the particles apart. Furthermore, for the first case the position of the particles remains constant while for the second one a circulation of particles is induced by the bubbles flow.

References
Gostiaux, L., Gayvallet,H. and Géminard, J.-C. 2002 Dynamics of a gas bubble rising through a thin immersed layer of granular material: an experimental study. GranularMatter, 4,39-44.

Reporters and Editors
Reporters may freely use this image. Credit: Universidad Nacional Autonoma de Mexico (UNAM) (2010).

Carlos Caicedo-Carvajal
Troy Shinbrot
Rutgers University
Newark, New Jersey

In 1855, James Thomson -- better known as Lord Kelvin's brother - wrote a paper describing "certain curious motions" on liquid surfaces. These "curious motions" were later studied by Italian scientist, Carlo Marangoni, and can be reproduced easily by carefully depositing a droplet of food coloring into a shallow (about 4 mm) dish of water.

One of several intriguing motions that can be seen is the pulsation of the coloring. This pulsation leads to stretching and folding of small quantities of food coloring that spread on the surface of the water, shown in this false color image. Stretching and folding is the hallmark of chaotic mixing", which is vital to the manufacture of pharmaceuticals, cosmetics, and fine chemicals.

Remarkably, after more than 150 years, the effects described by Thomson still remain mysterious - for example, why does the coloring pulsate? Why are the patterns seen sometimes chaotic, and at other times nearly uniform? And, as the reader can verify, why do fine regular striations appear on the surface after about a minute? This simple experiment illustrates that the days when a novice in her home can uncover new and unexplained physics are far from over.

This work is funded by the National Science Foundation.

Reporters and Editors
Reporters may freely use this image. Credit: Carlos Caicedo-Carvajal & Troy Shinbrot (2010).

Jeffrey S. Guasto
Karl A. Johnson
J.P. Gollub
Haverford College
Haverford, Pennsylvania

Chlamydomonas are single-celled green algae that use a pair of thin flagellar appendages (250 nanometers diameter, 10 micrometers long) to swim through aqueous environments. The flagella beat like a breast-stroking swimmer to pull the cell forward. Flagella, and related structures called cilia, are present on many organisms, including some cells in the human body.

In this work [1], the algae were mixed with small tracer particles and suspended in a thin liquid film under a microscope. A high speed camera was used to track the movements of the tracer particles around swimming cells. Many examples were gathered, aligned and averaged to compile detailed information regarding the movements of the fluid around a typical cell (black disk). In the resulting analysis, the cell speed (lower left inset) and flagellar position (lower right inset) were correlated with the fluid flows throughout the breast-stroke cycle. The resulting patterns are read like a changing topological map (red contour lines) with strength and directionality indicated by the length and orientation of the blue arrows. Jet-like flows, whorl-like vortices and dead zones (also called stagnation points) evolve in a cyclical pattern as the cell strokes it flagella throughout the beat cycle, which takes just 20 thousandths of a second.

Developing a better understanding of the biophysics of how Chlamydomonas swim will shed important light on a variety of processes, from biologically-driven mixing, cell migration, predator-prey relationships and mating interactions. Because flagella and cilia are also important in human biology, our results offer insight into the natural processes that move fluids in our lungs, kidneys and reproductive systems as well as the diseases that result from defects in the functions of flagella and cilia (called ciliopathies).

This work was supported by NSF Grant DMR-0803153.

References
[1] Jeffrey S. Guasto, Karl A. Johnson, and J.P. Gollub, Physical Review Letters 105, 168102 (2010)*

Reporters and Editors
Reporters may freely use these images. Credit: J.S. Guasto, K.A. Johnson, & J.P. Gollub, Haverford College (2010).

S. Protière
CNRS/IJLRDA-UPMC
Paris, France

M. Abkarian
CNRS/LCVN-Montpellier 2
Montpellier, France

J. Aristoff
H. Stone
MAE-Princeton University
Princeton, New Jersey, USA

We sprinkle dense small particles into an oil layer, where the particles sediment until they straddle the oil-water interface as a consequence of surface tension forces. Due to the weight of a single particle there is a long-range attraction that gathers nearby particles into capillary rafts while simultaneously bending the interface downwards on a much larger scale than the particle radius. When the number of particles in the raft is large enough, the raft is unstable and sinks (see figure), thus encapsulating the upper oil phase in water. The sinking raft breaks into stable centimeter-size droplets. We thus form a new object called «armored droplets» which encapsulate and retain the oil. This could be a chemistry-free solution to the threat of oil spills on practical time scales necessary to mitigate spilling catastrophes.

Reporters and Editors
Reporters may freely use this image. Credit: Protiere, Abkarian, Aristoff and Stone (2010).

T. J. O'Hern
E. F. Romero
C. F. Brooks
B. Shelden
Sandia National Laboratories

J. R. Torczynski
A. M. Kraynik
L. A. Romero
G. L. Benavides
Sandia National Laboratories

Gas-liquid interfaces can be forced to break up when subjected to vibrations within critical ranges of frequency and amplitude. This breakup mechanism was examined experimentally using deep layers of silicone oils over a range of viscosity and sinusoidal, primarily axial vibration conditions that can produce dramatic disturbances at the gas-liquid free surface.Although small-amplitude vibrations produce standing Faraday waves, large-amplitude vibrations produce liquid jets into the gas, droplets pinching off from the jets, gas cavities in the liquid from droplet impact, and bubble transport below the interface.

Experiments used several different silicone oils over a range of pressures and vibration conditions. Applications include liquid fuel rockets, inertial sensing devices, moving vehicles, mixing processes, and acoustic excitation.

Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000.

Reporters and Editors
Reporters may freely use this image. Credit: Courtesy Sandia National Laboratories.

Roberto Camassa
Mark Hemphill
David Holz
Richard M. McLaughlin
Keith Mertens
Cameron Moseley
Casey Smith
RTG Fluids Group
The Carolina Center for Interdisciplinary Applied Mathematics
The University of North Carolina at Chapel Hill

Close up view of a heavy, green-dyed salt water droplet falling in fresh water. Because of the friction in the fluid, the droplet develops into a vortex ring which shows up visually through the sharp change in refractive index due to the different densities of the fluids. Streaks of salt and dyed fluid are left behind in the motion of the droplet towards the bottom giving a rendition of a crystal pendant chandelier. This phenomenon is a member of a class of interactions occurring in stratified fluids which can play an important role on global scales for processes such as the carbon budget in the oceans and atmosphere. Further interactions of this type in rising fluids under their buoyancy likely played a role in the formation of underwater oil plumes in the recent Deepwater Horizon Gulf Oil Spill.

This work is funded by the National Science Foundation through NSF RTG DMS-0502266, NSF RTG DMS-0943851, NSF RAPID CBET-1045653, NSF CMG ARC-1025523, and NSF DMS-1009750.

Reporters and Editors
Reporters may freely use this image. Credit: The Carolina Center for Interdisciplinary Applied Mathematics, UNC Joint Fluids Lab, The University of North Carolina at Chapel Hill (2010).

M.J. Burin
C.J. Czarnocki
Department of Physics
California State University, San Marcos

Fluid flow bound between rotating cylinders, or Taylor-Couette flow, is one of the most studied and celebrated experimental arrangements in fluid dynamics for the investigation of flow instabilities and the transition to turbulence. Typically the inner cylinder is spun – this leads to a transition to turbulence that is analogous to heat convection and is well studied. When only the outer cylinder is spun the flow also may transition to turbulence, but through an entirely different manner. Such flows become turbulent from the presence of shear, as also happens in pipe flow. Between the laminar and turbulent state a wealth of patterns may be observed, as the three examples pictured here illustrate. A. If the flow is rapidly accelerated, and if the gap between cylinders is narrow enough, roll waves may develop – this appears to be the rotating version of the waves that develop in a thin layer of water as it flows down a sidewalk. B. When the flow is accelerated, but to a mild speed, turbulence initially develops from the end-boundaries, but is not sustained; it eventually dies out. C. If the flow becomes turbulent however, but is then slowed down a bit, it may develop ‘spiral turbulence’. This peculiar intermediate state resembles a classic barbershop pole, featuring inclined alternating bands of turbulent and smooth flow. The origins of this pattern are still not fully understood.

This research is partially funded by the Research Corporation for Science Advancement,a foundation dedicated to science since 1912

Reporters and Editors
Reporters may freely use this image. Credit: M.J. Burin, CSU San Marcos (2010).

Philipp A. Boettcher
Brian Ventura
Joseph E. Shepherd
California Institute of Technology
Pasadena, California

In a mixture of hexane (4.32 %) and air containing an excess amount of the fuel hexane, a flame is ignited at the tip of the glow plug seen at the bottom of the image. The flame then propagates upwards and outwards and is captured in these image using color schlieren visualization, which allows us to see gradients in the density resulting from increase in temperature from the combustion. Additionally, we can observe a horizontal line across the flame, which is the initial instability resulting from the difference in flame speed in the hot plume above the glow plug and the much colder surrounding gas. The instability grows and forms a cellular flame as shown in the second photograph from the same experiment at a later time.

Ignition temperature and behavior of the flame propagation are important characteristics in safety analysis. In industrial and transportation applications combustible mixtures may come in contact with hot surfaces, simulated here by the glow plug, and ignite. The development of instabilities creates the wrinkled flame we see in the second frame, which has a larger surface area than a smooth flame. This increase in the flame area leads to acceleration of the flame. The flame speed is a factor in the peak pressure rise and thus potential structural damage resulting from an ignition event.

A homogeneous mixture of gaseous hexane (6.48 %) in air is heated locally by a glow plug until the mixture ignites. Unlike mixtures with lower concentrations of hexane this flame does not propagate outward and consume the reactants. The flame is instead swept upward by buoyancy and the reactants immediately following in its wake ignite. The two images show snapshots of this cyclic behavior which has a frequency of 12 Hz for this mixture. This behavior only ceases when the hot products of the reaction, which accumulate the top, fill up the 2 liter vessel up to the glow plug and prevent reactant to come in contact with the hot surface. Puffing behavior is generally associated with situations where fuel and air are not mixed beforehand such as a fires above a pool of liquid fuel [1].

To our knowledge, this is the first observation of a cyclic flames in homogeneous mixtures of gaseous fuel and air from a continuous hot surface. The most common situation is a single flame propagating through the mixture and consuming the reactants. The pressure rise seen here is very mild in comparison to that created by a single flame propagating through the mixture.

[1] Cetegen, BM and TA Ahemd. Experiments on the periodic instability of buoyant plumes and pool fires. Combustion and Flame 93, 1-2: 157-184, 1993.

This work was funded by The Boeing Company under Strategic Research and Development Research Agreement CT-BA-GTA-1.

References
Philipp A. Boettcher, B. Ventura , G. Blanquart, and J. E. Shepherd. "Hot Surface Ignition of Hydrocarbons in Air - A Comparison of Experimental and Computational Results." Eighth International Symposium on Hazards, Prevention, and Mitigation of Industrial Explosions (8th ISHPMIE). Manuscript under review.

Reporters and Editors
Reporters may freely use this image. Credit: P. A. Boettcher, B. Ventura, and J. E. Shepherd, California Institute of Technology (2010).

D. Plocher
F.K. Browand
University of Southern
Los Angeles, California

Spray produced by motor vehicle tires rolling through water on a roadway is a familiar sight. Here we simulate that phenomenon in the laboratory by running a single-grooved tire (the upper one) against a smooth tire (the lower) simulating the road. Water is supplied from the left, flowing at a speed matched to the tire speed (6 m/s). The region immediately downstream of the tire contact patch is observed with high speed video.

As the water in the tire groove exits the contact patch it forms a thin sheet between the two tires. This sheet of water develops holes. As the margins of two neighboring holes collide, the margins break up, forming the very smallest droplets whose diameter is comparable to the sheet thickness, of the order of 100 microns. Much of the sheet remains intact, however, and begins to form thicker, more-or-less arc shaped structures or ligaments.

These ligaments are unstable (the Rayleigh instability) and pinch off to form the largest droplets whose diameter is roughly twice the diameter of the ligaments.

This work is funded in part by the U.S. Department of Energy and Michelin Americas Research and Development Corporation.

Reporters and Editors
Reporters may freely use this image. Credit: Dennis Plocher, Fred Browand, and Charles Radovich (University of Southern California) (2010).

F. Celestini, R. Kofman
J. Jean Rajchenbach
LPMC, CNRS UMR662
Université de Nice-Sophia Antipolis
France

When a water jet impinges upon a solid surface it produces the well-known circular hydraulic jump that everyone can observe in the sink of its kitchen. . It is characterized by a central thin liquid sheet bounded by a circular rise of the water height due to the combined effect of capillary and gravitational forces. In this experiment we study the interaction between two circular hydraulic jumps as a function of their distance and the flux supplied by the two water jets. For a sufficiently small distance and/or large fluxes we observe the structure represented in the picture. One can observe the merging of two hydraulic jumps on the substrate, and the consecutive formation of a stable fluid arch supported by a thin liquid sheet. For larger velocities the arch is broken in two rims and the fluid is ejected at the top of it.

Reporters and Editors
Reporters may freely use this image. Credit: F. Celestini, R. Kofman and J. Jean Rajchenbach, LPMC, CNRS UMR662, Université de Nice-Sophia Antipolis (2010).

Sanjay Kumar
Benito Gonzalez
Department of Engineering
The University of Texas at Brownsville

Flow past a uniformly rotating circular cylinder placed in a uniform flow is a classical fundamental fluid mechanics problem with the fluid region behind the cylinder (wake) exhibiting interesting changes in flow features as the cylinder rotation rate is increased. The figure represent the wake structure behind two uniformly rotating cylinders with both cylinders having the same rotation rates and the cylinder surfaces in between the two cylinders moving in a direction opposite to the main flow direction. This problem has not been investigated in as much detail as the problem of flow past single rotating cylinder. The image represents symmetrical wake structure at Reynolds number of 200. The cylinders are separated such that ratio of center-to-center distance to cylinder diameter is 2.5. The ratio of surface speed of cylinder to free stream velocity is 1.35.

It is found that there is symmetry breaking occurring near this rotation rate at fixed cylinder spacing and Reynolds number, i.e, by changing (decreasing) the rotation rate by very small increment to say 1.30 the wake structure (flow pattern) changes dramatically to a very different pattern showing no symmetry at all. This change has no hysterisis with respect to rotation rate, i.e, one can switch the flow pattern by very small changes in rotation rate. The study increases our understanding of the fundamental fluid physics of this not-so-well studied problem and can have impact on several practical applications involving flow control, on design of devices such as mixers based on this technique, and possibly many more. The flow visualization is done using hydrogen bubble technique.

The research on this problem also used Particle-Image-Velocimetry system, the purchase of which was made possible by NSF grant CMMI - 0723094.

References
S.Kumar, B.Gonzalez, and O.Probst, "Flow past two rotating cylinders," Physics of Fluids (accepted).

Reporters and Editors
Reporters may freely use this image. Credit: Sanjay Kumar and Benito Gonzalez, Department of Engineering, The University of Texas at Brownsville (2010).

Rene Sanjuan-Galindo
Enrique Soto, Gabriel Ascanio
Roberto Zenit
Universidad Nacional Autonoma de Mexico

In the process to disperse a viscous oil in water in a stirred tank, filaments are produced before droplets became formed. The impeller blades drive the oil and pushed it radially into the liquid bulk. In this step, filaments suffer elongation due to the fluid forces that make them thinner. This set of photographs were obtained using a high speed camera, it can be noticed that filaments are highly unstable and easily deformed, what induces that breakage occurs unexpectedly. The broken filaments recoil at the time they are dispersed and finally droplets are produced.

Reporters and Editors
Reporters may freely use this image. Credit: Universidad Nacional Autonoma de Mexico (UNAM) (2010).

Ching-Yao Chen
Jia-Fen Liu
National Chiao Tung University
Taiwan

When a ferrofluid is subjected to a sufficiently strong vertical magnetic field, visually striking peaks are formed on its free surface, known as the Rosensweig instability. The images show an interesting pattern-forming instability occurring when a ferrofluid droplet is immersed in a thin layer of a nonmagnetic fluid [1]. Evolution of the Rosensweig instability in the formation of a newly discovered fluid annulus is first observed to exceed the interface as shown at t=0.06 s. As the instability proceeds, a central peak is formed at the middle of the annulus at t=0.1 s and 0.2 s. The instability develops continuously and results in a dominant peak in a state of equilibrium at t=3 s and 8 s.

This work is funded by the National Science Council of Taiwan (Republic of China).

References
[1] Chen, Ching-Yao, Liu, J.-F. and Wang, L.-C., to appear in Magnetohydrodynamics, 2010.

Reporters and Editors
Reporters may freely use this image. Credit: Chen and Liu, National Chiao Tung University, Taiwan (2010).

Silas Alben
School of Mathematics
Georgia Institute of Technology

We have used computers to solve for the motions of flexible sheets falling in fluid. We have studied sheets with various mass densities and bending rigidities. Show here are sheets with a single mass density and eight bending rigidities (the numbers in the upper left of the panels), released from rest at different angles. The sheet trajectories show a wide range of intricate structures as they fall. The basic behavior is a repeated series of accelerations to a critical speed at which the sheet flexes and rapidly decelerates, corresponding to sharp turns in the paths. The sheet trajectories also show persistent circling (the heavy diagonal lines in panels labeled 10, 14, and 24), periodic flapping (the zig-zag downward lines in the panel labeled 1), and more complex repeated patterns. 

This work is funded by the National Science Foundation under Grant No. DMS-0810602.

References
S. Alben, Flexible sheets fallling in an inviscid fluid, 22, 061901-1-12 (2010). 

Reporters and Editors
Reporters may freely use this image. Credit: Silas Alben and Physics of Fluids (2010).

H. C. Mayer
R. Krechetnikov
University of California at Santa Barbara

Almost all of us at some point in our lives have had the experience of popping a soap bubble and marveling at the amazing speed with which the bubbles disappears. It is relatively easy to rupture soap films from a point: a gentle touch with a fingertip will do the trick. But is it possible too 'cut a soap film with scissors' as we can cut paper or fabric? And if so, will we something different? This was a basic question that motivated this research.

Although extensively explored in the past, there are still many questions regarding the retraction of liquid films that remain unanswered and many experiments that stay unperformed because it had been thought impossible to release a liquid film uniformly along an edge. Indeed, all prior experimental works have focused on measurements of the retraction of thin liquid films using rapture from a point. Our idea was to pass a high-voltage microsecond impulse through the wire frame supporting the soap film, which led to the film edge boiling and its uniform release. Experiments were performed for different frame shapes (square, rectangle, and circular) and sizes: the resulting patterns captured with a high speed digital camera and shown on still images are amazing and make one to think about non-trivial interaction of the film retraction and wave propagation in the soap film. This new approach for freeing soap films allows one to study features in their behavior which were not accessible before. 

Reporters and Editors
If any reporters are interested in using this image, please contact Dr. Rouslan Krechetnikov, Department of Mechanical Engineering, University of California at Santa Barbara, Santa Barbara for permission. 

Patrick Wessels
Jean Hertzberg
Department of Mechanical Engineering
University of Colorado, Boulder

A famous fingering instability is found when air is injected into a thin layer of water (or any more viscous or stiffer fluid), sandwiched between two flat sheets. Here, we are looking down on a sandwich of two sheets of Plexiglas with a mixture of water and vegetable oil dyed green as the filling of the sandwich. When the top sheet is lifted a little bit, air is pulled into the middle of the sandwich. It doesn't enter uniformly; instead it forms fingers pushing into the mixture. You may have noticed a similar pattern when peeling apart any kind of gooey sandwich. In a laboratory, the sandwich is called a Hele Shaw cell, and the fingering instability is the Saffman Taylor (Tabeling et al. 1987). 

References
Tabeling, P., Zocchi, G., Libchaber, A., (1987). An Experimental Study of the Saffman-­‐Taylor Instability. Journal of Fluid Mechanics Digital Archive 177: 67-­‐82.

Reporters and Editors
This image may be freely used for educational purposes as long as the authors are credited.