Posts Tagged ‘Fluid dynamics’

How to Plug an Oil Leak with Corn Starch

February 1, 2011 3 comments

NASA'S Terra Satellite captured this image of the Deepwater Horizon oil slick on May 24, 2010. (Photo: NASA/GSFC)
One of the biggest, if not THE biggest news story of 2010 was the Deepwater Horizon oil spill. The spill released over 200 million gallons of crude oil into the Gulf of Mexico and is one of the biggest man-made natural disasters in history.

The spill lasted for nearly 3 months due to the high difficulty of plugging the well. One such attempt to plug the leak was the “top-kill” method, which involves releasing a heavy fluid (“mud”) into the well in hopes that it would sink to the bottom and stem the flow of oil. BP’s attempt at this failed, and a paper published online yesterday in Physical Review Letters entitled “Viscoelastic Suppression of Gravity-Driven Counterflow Instability” explains why, and how they could have done better.

So you can imagine the problem like this: you have a well gushing oil upwards. You want to slow down this flow of oil, so you release a dense fluid down into the well and hope it will flow downwards, against the pressure of the oil.

When two different fluids come into contact at different velocities it generates turbulence at the interface. This is called the Kelvin-Helmholtz instability. You have seen it before when you look out onto the ocean or a body of water and the wind creates ripples and small waves on the water. Turbulence is being generated at the interface of the water and the air causing those ripples.

A similar thing happens when oil meets a dense fluid. If the turbulence generated at the interface of the two fluids is high enough to break the “mud” into tiny droplets, then the top-kill method will fail because the mud simply breaks apart and won’t plug the well.

In this paper, the authors demonstrate a fluid which overcomes the Kelvin-Helmholtz instability under conditions similar to those at the site of the Deepwater Horizon spill.

First, however, they theoretically study the effect of using a typical water-based mud in a top kill method for conditions similar to those at Deepwater. What they found was that a typical mud would not have descended quickly enough into the mud to be effective. In other words, the velocity of the oil shooting up the well was much greater than the rate at which the mud would have descended into the oil, resulting in the mud simply being washed out of the well. This would explain why BP’s top-kill attempt had failed.

The authors then tested their own recipe for a mud which may have worked. They introduced a “dilatant polymer with shear-thickening and viscoelastic properties”. What this means is they added a material which would actually get harder under stress (shear-thickening) and resist breaking apart when in contact with the oil. In fact, the force of the oil moving upwards is what would cause it to get harder.

A fluid like this is made of a corn-starch water emulsion. Under high shear stress this fluid gains a “tramponlinelike” behaviour, helping it resist the turbulent flow of the oil.

To test their fluid, they “filled a transparent column 1.6 m tall and 63 mm in internal diameter with a transparent light mineral oil”. They then released their mud into the oil and observed the effects. Using a plunger they simulated the movement of oil over the mud. They found that their corn-starch mixture did not break apart and descended as a coherent “slug”.

The also found that the slug descended at a rate fast enough that it would have overcome the upward velocity of the oil from the Deepwater well. However, they did note that the experiment would have to be repeated at a larger scale to gain a better idea if this would be a useful approach for an actual oil spill.

Beiersdorfer, P., Layne, D., Magee, E., & Katz, J. (2011). Viscoelastic Suppression of Gravity-Driven Counterflow Instability Physical Review Letters, 106 (5) DOI: 10.1103/PhysRevLett.106.058301

The Physics of Coffee Rings

November 24, 2010 Leave a comment

In keeping with the abstract on the physics of jump rope, the 63rd meeting of the American Physical Society has yielded yet another fascinating study.

63rd Annual Meeting of the APS Division of Fluid Dynamics

Volume 55, Number 16 

Abstract: RU.00007 : Coffee ring deposition in bands


  Shreyas Mandre
    (Brown University)

  Ning Wu
    (Colorado School of Mines)

  Joanna Aizenberg
    (Harvard University)

  Lakshminarayanan Mahadevan
    (Harvard University)

Microscopic particles suspended in a liquid are transported and deposited at a contact line, as the contact line recedes due to evaporation. A particle layer of uniform thickness is deposited if the particle concentration is above a threshold; below this threshold the deposit forms periodic bands oriented parallel to the contact line. We present a model for the formation of these bands based on evaporation leading to the breakup of the thin liquid film near the contact line. The threshold results from a competition between evaporation speed and deposition speed. Using this model, we predict the thickness and length of the bands, making the control of patterned deposition possible.

[My comments: The authors used glass particles in a liquid to mathematically model how rings form. They can make these predictions using parameters such as evaporation rate and surface tension of the liquid. Aside from just being interesting, this study may have some practical implications for working at small scales.

Controlling the ring deposition process would be useful for creating such things as new microphysics tools operating at a scale where pliers or other traditional tools for moving particles cannot operate,” notes Mandre. (From]

The Physics of Jumping Rope

November 23, 2010 Leave a comment

63rd Annual Meeting of the APS Division of Fluid Dynamics Volume 55, Number 16 

Abstract: CX.00008 : The aerodynamics of jumping rope


  Jeffrey Aristoff
    (Department of Mechanical and Aerospace Engineering, Princeton University)

  Howard Stone
    (Department of Mechanical and Aerospace Engineering, Princeton University)

We present the results of a combined theoretical and experimental investigation of the motion of a rotating string that is held at both ends (i.e. a jump rope). In particular, we determine how the surrounding fluid affects the shape of the string at high Reynolds numbers. We derive a pair of coupled non-linear differential equations that describe the shape, the numerical solution of which compares well with asymptotic approximations and experiments. Implications for successful skipping will be discussed, and a demonstration is possible.

[My comments: The authors built a robot jump rope device and controlled parameters of rope rotation rate, rope density, diameter, length etc. using high speed cameras, they developed equations to describe the motion of the jump rope.

“Our main discovery is how the air-induced drag affects the shape of the rope and the work necessary to rotate it,” says Princeton researcher Jeff Aristoff. “Aerodynamic forces cause the rope to bend in such a way that the total drag is reduced.” (Leaves do this too when they bend out of the wind.) This deflection or twisting is most important in the middle of the rope and the least at the ends. If the rope is too light it might not clear the body of the jumper. (From

I hope they did a demonstration. My experience is that physics conferences can be a bit stuffy.]