Characterisation of the acoustic cavitation cloud by two laser techniques
Introduction
In a sonicated liquid medium, when the ultrasound wave amplitude is greater than a threshold value, the acoustic cavitation appears. Several models of single bubble dynamics ([1] and references cited therein) and of bubble cloud dynamics [2], [3], [4] have been developed. However, experimental data concerning the bubble population (size of the cavitation bubbles and void fraction) are still lacking. Until now, few experimental studies have been devoted to the characterisation of the acoustic cavitation bubble field. High speed photography, e.g. high speed holocinematography [5], has allowed a qualitative description of the spatial dynamics of the bubble cloud. At low ultrasound frequency (20 kHz or less), a complex filamentary or branched structure composed of tiny bubbles, also called ‘streamers’ or ‘acoustic Lishtenberg figures’, can be observed inside a cylindrical piezoelectric transducer submerged in water. A series of holograms showed that the bubble field oscillates as a function of the acoustic intensity periodically, subharmonically or chaotically [6]. Other authors [7] also showed by high speed photography, at 10 kHz in the same acoustic system as before, the identification of stable cavitation and unstable or transient cavitation. They also examined the physical phenomena of coalescence and fragmentation which proved to be an explanation of the observed bubble size heterogeneity. High speed photography [8] has also revealed that in a standing wave field, established in a cup-horn reactor operating at 28 kHz and a low acoustic intensity of 3.2 W cm−2, clouds composed of stable cavitation bubbles were trapped in the pressure nodes, whereas clouds formed of transient bubbles appeared at the pressure antinodes. These latter moved due to the Bjerknes forces between pulsating bubbles to pressure nodes. Another technique using an optic fibre laser knife and an optic fibre laser Doppler anemometer simultaneously [9] has investigated the acoustic bubble behaviour in a cup-horn reactor at 95 kHz. In a standing wave field the initial bubble diameter was found to be 40 μm and in the pressure nodes this bubble could grow up to 500 μm. Finally, a hyperfrequency method [10] has been used to determine the void fraction and the size distribution of the bubble population. A maximum void rate of 3×10−4 was obtained at 308 kHz and an acoustic pressure of 0.2 bar.
In the present paper, two laser techniques, laser diffraction (LD) and phase Doppler (PD), are implemented and compared. After describing the two principles of measurement, the results will show essentially the effect of the local acoustic power on the mean diameters and the void fraction of a bubble cloud.
Section snippets
Set-up
Tests are carried out in a rectangular glass vessel (150×100×270 mm3) containing bi-distilled water (liquid height 190 mm). An immersion horn of diameter 13 mm, operating at 20 kHz, is vertically introduced into the vessel such that the distance between the tip of the horn and the bottom is close to the water ultrasound wavelength (70 mm). The ultrasound power was the electric power supplied to the generator Pin. This can vary up to maximum power of 320±5 W. The corresponding acoustic intensity IA
Measuring techniques [12–15]
The LD instrument is a Malvern 2600 Particle Sizer. As shown in Fig. 1, a parallel monochromatic light beam of diameter 9 mm emitted by a low power helium–neon laser passes through the vessel. The probe volume is defined as the intersection between the bubble cloud and the laser beam. Light scattered in the near forward direction by the bubbles is focused by a Fourier transform lens onto a series of 31 concentric semi-circular diodes placed in the focal plane of the lens; the diode ring areas
Implementation of the LD technique under ultrasound
First of all, a few experimental precautions must be taken. During the irradiation, large degassing bubbles can be observed with the naked eye. If some of them travel through the light beam, the signal due to the bubble cloud does not seem to be influenced; however, those that coalesce on the vessel wall at the probe volume location bias the signal. Moreover, the effects of multiple scattering [12], [19] must be avoided. This can be checked by verifying that the obscuration rate is less than
Conclusion
Quantitative information on the acoustic bubble population generated at 20 kHz was obtained by the LD and PD techniques. Their comparison shows similarities. A study of the effect of ultrasound power on the radial and axial distributions showed that at low local ultrasound power the spatial void rate is low and large bubbles are measured. This observation is pronounced with the PD instrument: the Sauter mean diameter is assessed as about 50 μm, whereas it is approximately 10 μm at high local
References (28)
- et al.
Ultrasonics Sonochemistry
(1997) - et al.
Ultrasonics
(1989) - et al.
Ultrasonics
(1998) Ultrasonics Sonochemistry
(1995)- et al.
J. Acoustic. Soc. Am.
(1994) - et al.
Int. J. Bifurcation Chaos
(1993) - et al.
Phys. Rev.
(1987) - et al.
SPIE Laser Anemometry Advances and Applications
(1993) - S. Labouret, Ph.D. thesis, University of Valenciennes, France,...