Cavitation hots up

Gas inside collapsing bubbles can become very hot and, as a result, emit light. It turns out that temperatures of more than 15,000 kelvin can be reached — as hot as the surface of a bright star.

In 1917, Britain's Royal Navy had problems with bubble cavitation. This is a process in which tiny bubbles grow in size and then collapse as a result of pressure variations in the turbulent water around ships' propellers. The process is so violent that it was causing considerable damage to the propellers1, so the navy asked the renowned physicist Lord Rayleigh to analyse the problem2. His research led to what is now called the Rayleigh equation, which describes the dynamics of the collapsing bubble walls1,2. However, the solution to the equation produced a singularity. It implied that, during collapse, the gas inside the bubble is compressed so fast that it cannot equilibrate with the surrounding liquid, leading to energy focusing and an infinite temperature increase. In reality, of course, this cannot happen, so the question is: what limits the temperature increase, or, in other words, how hot does the bubble get? On page 52 of this issue3, Flannigan and Suslick report a study of light emission from single bubbles during cavitation, and provide a direct answer to this question.

The temperature reached by the collapsing bubble depends on how much of the focused energy is lost by sound emission at the collapse and how much is consumed by internal processes such as vibrations, rotations, dissociation and eventually ionization. If there are many collapsing bubbles, they disturb each other, which leads to a less-spherical collapse and therefore less-efficient energy focusing. Nonetheless, temperatures can rise so high that the bubbles start to glow. This phenomenon has already been investigated intensively by using sound waves to drive bubble production in liquids and then detecting the light emitted; the sound waves cause a temporarily reduced pressure in the liquid, which makes the bubbles grow and eventually collapse again (Fig. 1). So far, emission spectra with a detailed line structure have only been observed for many transient bubbles together (so-called multi-bubble sonoluminescence). Analysis of the emitted spectral lines4 indicates that the temperature reached inside these bubbles is around 5,000 kelvin.

Figure 1: Bubble sonoluminescence — bubbles are driven by sound waves to emit light.

a, At low sound-wave pressure, a gas bubble expands until (b) an increase in pressure triggers its collapse. Flannigan and Suslick3 find that, during collapse, temperatures can soar to 15,000 K, as the authors observed from spectra of light emitted from the bubble (c). Analysis of the emission spectra also provides direct evidence for the existence of a plasma inside the collapsing bubbles.

In single-bubble sonoluminescence5,6, an isolated and stable bubble is studied; disturbances from other bubbles are absent. The light emission from such a bubble can be more than 107 photons per flash7. As the bubble is driven periodically with sound waves at frequencies of typically 20–40 kHz, the emitted light is visible to the naked eye. However, it has previously been difficult to deduce the temperature reached, as the emission spectra from single bubbles were basically featureless.

But Flannigan and Suslick3 have obtained well-resolved spectral lines for the single-bubble case. They use xenon- and argon-filled bubbles in sulphuric acid, a set-up that has various advantages6. First, the high fluid viscosity of sulphuric acid ensures a stable spherical shape for relatively large bubbles. Second, monoatomic gases such as argon and xenon do not consume energy in rotational and vibrational degrees of freedom, and so more of the focused energy ends up as thermal energy. Third, because of the low vapour pressure of sulphuric acid, hardly any (polyatomic) vapour molecules invade the bubble at expansion; that would also eventually lead to additional energy absorption. In this way, Flannigan and Suslick are able to observe a thousand times more photons than observed from xenon and argon bubbles in water. As a result, they obtain good spectral details, from which a temperature of 15,000 kelvin is deduced — as high as is found at the surface of bright stars.

Perhaps an even more remarkable finding is that the emission spectra indicate the existence of plasma (ionized matter) inside the collapsing bubbles. Flannigan and Suslick observe that there are highly excited emissive states, which is inconsistent with thermal processes. Instead, some of the emitted light must originate from high-energy electrons and ions that are decelerated owing to collisions inside the gas bubble.

The presence of a weakly ionized plasma and the origin of the light emission, as well as the high temperatures in single bubbles, have been predicted theoretically6,8,9,10, but experimental evidence has been indirect. In previous work, the deduction of the bubble temperature from observable parameters required modelling assumptions (Fig. 2). Flannigan and Suslick's experiments are a milestone in single-bubble sonoluminescence, as they constitute the first direct measurement of the temperature and the state of matter in a single bubble at collapse.

Figure 2: Indirect evidence for the temperature reached inside a collapsing bubble.

Hitherto, the temperature T(t) (as a function of time t) in single collapsing bubbles could only be deduced indirectly, using modelling steps to link observable parameters (blue circles) such as the chemical reaction rates11, bubble radius R(t), and the spectral radiance Pλ(t). Flannigan and Suslick3 have measured the temperature directly from light-emission spectral lines.


  1. 1

    Brennen, C. E. Cavitation and Bubble Dynamics (Oxford Univ. Press, 1995).

    Google Scholar 

  2. 2

    Rayleigh, L. Phil. Mag. 34, 94–98 (1917).

    Article  Google Scholar 

  3. 3

    Flannigan, D. J. & Suslick, K. S. Nature 434, 52–55 (2005).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Flint, E. B. & Suslick, K. S. Science 253, 1397–1399 (1991).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Crum, L. A. Phys. Today 47, 22–29 (1994).

    CAS  Article  Google Scholar 

  6. 6

    Brenner, M. P., Hilgenfeldt, S. & Lohse, D. Rev. Mod. Phys. 74, 425–484 (2002).

    ADS  CAS  Article  Google Scholar 

  7. 7

    Barber, B. P. et al. Phys. Rep. 281, 65–144 (1997).

    ADS  CAS  Article  Google Scholar 

  8. 8

    Moss, W. C. et al. Science 276, 1398–1401 (1997).

    CAS  Article  Google Scholar 

  9. 9

    Hilgenfeldt, S., Grossmann, S. & Lohse, D. Nature 398, 402–405 (1999).

    ADS  CAS  Article  Google Scholar 

  10. 10

    Toegel, R. & Lohse, D. J. Chem. Phys. 118, 1863–1875 (2003).

    ADS  CAS  Article  Google Scholar 

  11. 11

    Didenko, Y. T. & Suslick, K. S. Nature 418, 394–397 (2002).

    ADS  CAS  Article  Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Lohse, D. Cavitation hots up. Nature 434, 33–34 (2005).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.