Inside a micro-reactor

Gas bubbles in a liquid can convert sound energy into light. Detailed measurements of a single bubble show that, in fact, most of the sound energy goes into chemical reactions taking place inside this 'micro-reactor'.

'Single-bubble sonoluminescence' is the remarkable phenomenon that describes how a gas bubble in liquid, exposed to a strong, standing sound wave, collapses and emits light. First observed 12 years ago1, the basic physics of the process seems to be understood2. That there is strong and crucial chemical activity inside the sonoluminescing bubble had already been hypothesized3 and indirectly confirmed4,5. Now Didenko and Suslick6 (page 394 of this issue) have performed the first direct measurements of the reaction rates inside an individual bubble as it sonoluminesces. Energy-wise, it seems that a sonoluminescing bubble should be viewed not as a light bulb, but rather as a high-temperature, high-pressure, miniature reactor.

The process of sonoluminescence is shown in Fig. 1. First, at low sound pressure, the micrometre-size bubble expands, increasing its volume by a factor of 1,000. When the pressure increases again, the bubble eventually collapses dramatically, shrinking to a radius that corresponds to solid-state densities. The compression drives up the temperature of the gas inside the bubble — through this 'adiabatic' heating the bubble interior is thought to reach around 10,000–20,000 K. Consequently, the gas becomes partly ionized and the recombination of electrons and ions leads to the emission of light.

Figure 1: Glowing bubbles: a sound wave in liquid causes sonoluminescence.

a, At low pressure, a gas bubble expands dramatically, until b, an increase in sound-wave pressure triggers its collapse. As the temperature inside the bubble soars to over 10,000 K, the gas becomes partly ionized, forming a plasma. Finally, c, recombination of electrons and ions results in light emission. But according to Didenko and Suslick6, more energy goes into chemical reactions in the bubble gas than is released as light.

As the bubble expands, gas dissolved in the liquid enters the bubble. At the point of adiabatic collapse, some of these gases are trapped inside the hot bubble and start to react. For example, nitrogen molecules dissociate into nitrogen radicals and then react to form gases such as NH and NO. These highly soluble gases re-dissolve in the surrounding water when the bubble cools down. As the bubble expansion begins again, the next reaction cycle starts. Didenko and Suslick's calculations of the energy budget of sonoluminescence show that the amount of energy going into endothermic chemical reactions inside the bubble is two orders of magnitude higher than that going into light emission.

However, one complication that still remains is that the temperature inside the bubble cannot be measured directly. It has to be deduced either from the bubble dynamics (for example, by Mie scattering1,7,8) or from the properties of the light emitted (spectral information, intensity and widths of the light pulses9). Either way, assumptions have to be made, whether in the modelling of the bubble dynamics10 and the thermodynamics of the heat and mass exchange between the bubble and its surroundings, or in the modelling of the plasma physical processes to predict the observable light properties.

The information obtained in these two ways should obviously be consistent in a viable theory of sonoluminescence. Even then, we can't be certain, as errors arising in the modelling of the bubble interior and light emission could compensate for each other. But Didenko and Suslick's measurements of the chemical reaction rates open up a third experimental window on the process. This extra constraint reduces the freedom in modelling, leading towards further convergence of the models.

It is astounding how many sub-disciplines of physics and chemistry have played a role in disentangling what happens in single-bubble sonoluminescence. They range from acoustics, fluid dynamics, plasma physics, thermodynamics, atomic physics and spectroscopy, to physical and analytical chemistry, chemical kinetics, dynamical-system theory and applied mathematics in general. But nuclear and fusion physics are not on the list: the final conclusion from Didenko and Suslick's results is that it is the chemical reaction rate within the bubble that limits the efficiency of bubble collapse. So 'bubble fusion' — an energy-generating fusion reaction in the high-density, high-temperature interior of the collapsing bubble11 — is most unlikely.

Although fusion may be out of reach, there are other uses for sonoluminescent bubbles. The extreme conditions inside the bubble are adjustable through external parameters such as forcing pressure or water temperature, so the bubble can be considered as a controlled high-temperature reaction chamber, offering opportunities to measure reaction rates in extreme temperature and pressure regimes. But understanding single bubbles is not enough. Before this knowledge can be applied to sonochemistry12 — the enhancement of chemical reactions through ultrasound in a bubbly fluid — a better understanding of bubble–bubble interactions will be needed.

Just as the hydrogen atom was the basic model for larger atoms and molecules, so the single bubble is the simplest building block in the physics of a sound-driven bubbly fluid. With the detailed understanding of the hydrogen atom, atomic physics began to flourish. By analogy, now that there is a basic understanding of single-bubble sonoluminescence and the chemical activity inside the bubble, I expect also a flourishing of cavitation physics.


  1. 1

    Gaitan, D. F. An Experimental Investigation of Acoustic Cavitation in Gaseous Liquids. Thesis, Univ. Mississippi (1990).

    Google Scholar 

  2. 2

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

    ADS  CAS  Article  Google Scholar 

  3. 3

    Lohse, D., Brenner, M. P., Dupont, T. F., Hilgenfeldt, S. & Johnston, B. Phys. Rev. Lett. 78, 1359–1362 (1997).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Matula, T. J. & Crum, L. A. Phys. Rev. Lett. 80, 865–868 (1998).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Ketterling, J. A. & Apfel, R. E. Phys. Rev. Lett. 81, 4991–4994 (1998).

    ADS  CAS  Article  Google Scholar 

  6. 6

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

    ADS  CAS  Article  Google Scholar 

  7. 7

    Weninger, K. R., Barber, B. P. & Putterman, S. J. Phys. Rev. Lett. 78, 1799–1802 (1997).

    ADS  CAS  Article  Google Scholar 

  8. 8

    Matula, T. J. Phil. Trans. R. Soc. Lond. A 357, 225–249 (1999).

    ADS  CAS  Article  Google Scholar 

  9. 9

    Gompf, B., Günther, R., Nick, G., Pecha, R. & Eisenmenger, W. Phys. Rev. Lett. 79, 1405–1408 (1997).

    ADS  CAS  Article  Google Scholar 

  10. 10

    Plesset, M. S. & Prosperetti, A. Annu. Rev. Fluid Mech. 9, 145–185 (1977).

    ADS  CAS  Article  Google Scholar 

  11. 11

    Taleyarkhan, R. P. et al. Science 295, 1868–1873 (2002).

    ADS  CAS  Article  Google Scholar 

  12. 12

    Suslick, K. S. Science 247, 1439–1445 (1990).

    ADS  CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Detlef Lohse.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Lohse, D. Inside a micro-reactor. Nature 418, 381–383 (2002).

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.