In 1963, the British engineer Arthur Kaye reported that when he poured a complex organic solution in a thin stream onto a patch covered in the same solution, a jet of it would suddenly eject from the surface. He could offer no explanation for this behavior.

In 1976, A. A. Collyer and P. J. Fisher published an article in the magazine Nature called "The Kaye effect revisited". Their conclusions were that the fluid had to be "highly shear-thinning as well as elastic and 'pituitous' (slimy or sticky), and that a rigid surface is required to back the reflected liquid stream"1. Recently (2006), however, researchers from the University of Twente in the Netherlands have found an explanation for the Kaye effect, and have shown that many of the 'requirements' stated by A. A. Collyer and P. J. Fisher are incorrect. In fact, this phenomenon can be observed with simple handsoap or shampoo, which means you can test it for yourself in the shower!

One of the things which is required though, is that the fluid needs to be shear-thinning. Shear-thinning is the phenomenon that a fluid is stable at rest, but shows a great reduction in viscosity (the reverse of fluidity) when a shear stress is applied. Fluids which show this behavior are shampoo, hand soap, blood, the smearing fluid between your joints, no-drip paints, and lots of other fluids.

But what exactly happens? If you pour a thin stream of shampoo onto a surface, the fluid will curl and form a heap. At some point, the stream of fluid will slide off of the side of the heap. This will create a shear stress, which will create a small layer of fluid with a low viscosity. The incoming stream will keep sliding off this 'slippery hill', and because of the vertical force it exerts on the heap, a dimple will form, creating something in the likes of a ramp used in ski jumping. This will cause a jet of fluid from the ramp. The dimple keeps deepening, until the jet is nearly vertical. The jet will then hit the incoming stream, disturbing it and ending the Kaye effect.

To prevent the jet from hitting the incoming stream, you can tilt the bottom surface. You need to keep a steady flow of liquid on the surface though, so you need a reservoir at the top. If you then pour liquid on top, the jet will orient itself towards the lower end of the surface, while it never rises to vertical. At the spot where the jet hits the surface again, another jet can form. Like this, a cascading Kaye effect can be formed. The number of jets is limited however, because at every dimple, a small bit of the energy is lost.

Do it yourself! Next time you take a shower, try pouring the shampoo onto your facecloth from about 20 cm height in a small stream. By small I mean about a millimeter in diameter or smaller. If it doesn't work, try varying a bit. Experiment. I've seen mentioned that fluids with higher viscosity work better, yet my shampoo made better jets than my hand soap, even though that's thicker. It probably depends quite a bit on the type of soap you use, so try some different ones. In a Dutch article2 from the same researchers two types of soap are mentioned which should work well: Sanex Demo Protector douche gel and Kruidvat Nature's choice fluid soap. The Debutante tells me the former is available in the UK. The latter is probably a Dutch-only product, since Kruidvat is a Dutch store. One thing that helped me is to create a small stream from the shampoo bottle, than squeeze a tiny bit so it's just a little thicker for a second. Then let it go back to its original thickness.

Things to note:

  • The stream of fluid doesn't mingle with the surface from which it leaps. This was shown by shining a laser through the stream like in a fiber-optic cable, and also by having an opaque soap leap from a transparent one. Both experiments showed there was no mixing of the fluids.
  • This phenomenon doesn't need a rigid surface. They have tested this by creating the Kaye effect on a deep container full of liquid, and also by creating it on a thin film of the soap suspended in a metal frame. Think bubble blower, but without the water.
  • A nice video explaining this effect and showing nice high-speed recordings is available at: http://pof.tnw.utwente.nl/3_research/files/Gallery_shampoo_audio.mpg

    |
    |
    |
    |
    |
    |
 _  |
| | | .-'''-.
  '-./       \
     '-.      \ 
        '-.    | ,--,
           '-. |/    \
              '-.     
                 '-.  | _
                    '-.' \
                       '-.-.
                          '-.
                             '-.


  1. Michel Versluis, Cor Blom, Devaraj van der Meer, Ko van der Weele and Detlef Lohse, Leaping shampoo and the stable Kaye effect, Journal of statistical mechanics: theory and experiment, 20 July 2006, Full text available at: http://www.iop.org/EJ/abstract/1742-5468/2006/07/P07007
  2. http://pof.tnw.utwente.nl/5_publications/Leukje%20NTvN%2072-9%20(2006).pdf