Ultrasonics Sonochemistry 17 (2010) 3033Contents lists available at ScienceDirect

Ultrasonics Sonochemistry

journal homepage: www.elsevier .com/locate /u l tsonchShort Communication

Direct observation of cavitation fields at 23 and 515 kHzq

Gareth J. Price *, Naomi K. Harris, Alison J. StewartDepartment of Chemistry, University of Bath, BATH, BA2 7AY, UK

a r t i c l e i n f o a b s t r a c tArticle history:Received 9 February 2009Received in revised form 19 April 2009Accepted 22 April 2009Available online 3 May 2009

Keywords:SonoluminescenceCavitationBubble coalescenceFrequency1350-4177/$ - see front matter 2009 Elsevier B.V. Adoi:10.1016/j.ultsonch.2009.04.009

q Part of this work was presented at the 11thSonochemistry Society meeting at La Grande Motte, F

* Corresponding author. Tel.: +44 1225 386504.E-mail address: chsgjp@bath.ac.uk (G.J. Price).Direct observation of cavitation fields using photography, sonoluminescence and luminol mapping isreported for a 23 kHz horn sonicator and a 515 kHz plate transducer system. The effect of sound intensityand added surfactant on the cavitation fields is described. The observations support previously reportedresults suggesting significant differences in the cavitation fields between the two sonication systems.

2009 Elsevier B.V. All rights reserved.1. Introduction

Over recent years, ultrasound has become a highly useful meth-od for performing a wide range of chemical reactions and processesincluding chemical synthesis, materials production and watertreatment [13]. Most of the effects arise from cavitation [4,5].One of the factors limiting the wider adoption of sonochemistryand sonoprocessing on a larger, industrial is the difficulty in pre-dicting both the occurrence and consequences of cavitation andthe effect of various experimental parameters on individual cavita-tion bubbles and bubble fields.

Among the parameters that affect cavitation, the effect of theultrasound frequency is perhaps the least well understood. Fre-quency effects are influenced by both the number of bubble col-lapses per unit time and the product of each bubble collapse.Maximum bubble sizes are larger at lower frequencies and morereactive intermediates may be formed in each bubble collapse. Anumber of studies of the frequency effect have been published,some with conflicting results. However, most have generally fol-lowed the observations of Henglein [6] who reported that for arange of solutes in water, solvolysis (i.e. pyrolysis) was the mainpathway at 1 MHz but that radical production and mechanical ef-fects predominate at 20 kHz. Theron and co-workers [7] observedin the degradation of phenyltrifluoromethylketone in water at 30and 500 kHz a similar difference in mechanism. Several other re-ports have appeared of different products attributed to an alterna-tive mechanism operating at different frequencies. [810]. Okitsull rights reserved.

meeting of the Europeanrance, June 2008.and co-workers [11] also showed that the frequency used influ-enced the size and shape of sonochemically synthesised goldnanoparticles.

In an attempt to clarify these effects, Price et al. [1214] haveused a number of methods to investigate cavitation bubbles andtheir products when produced using a 20 kHz horn sonicator or a515 kHz plate transducer system. On the basis of different chemi-cal products of sonolysis reactions and differences in acousticemission and sonoluminescence quenching during sonication, theyconcluded that there were significant differences in the types ofcavitation produced by a 515 kHz plate transducer system and a20 kHz horn system of the type widely used for sonochemistry.At 515 kHz, stable cavitation was mainly observed where a bubbleundergoes many oscillations during its lifetime. This can lead tosignificant amounts of solute adsorbing to the bubble interfaceand evaporating into the bubble so that a major reaction at this fre-quency is from pyrolysis of solutes. At lower frequencies usinghorn type sonicators, transient cavitation is the predominant ef-fect. Here, bubbles undergo few oscillations before finally collaps-ing so limiting the amount of solute entering the bubble. Sincebubbles grow to a larger size and collapse more violently, largerquantities of free radicals can be produced by sonolysis andmechanical effects and attack by radical transfer into solution arethe major mechanisms. Both transient and stable cavitation occurunder each set of conditions but it is the balance between themthat is dependent on the conditions used.

A number of studies have identified different cavitation bubbledistributions arising from different types of ultrasound generatorsand configurations [1519]. For example, as illustrated by Suslicket al. [15] a 20 kHz horn system usually gives strong cavitation ina limited zone near the horn tip while Petrier and co-workers[1618] used the emission from luminol solutions to show that

mailto:chsgjp@bath.ac.ukhttp://www.sciencedirect.com/science/journal/13504177http://www.elsevier.com/locate/ultsonch

G.J. Price et al. / Ultrasonics Sonochemistry 17 (2010) 3033 31higher frequency emitters tend to give a more diffuse, widely dis-tributed zone of cavitation. In particular, Petrier et al. showed [18]that the nature of the cavitation field as well as the rates of hydro-xyl radical production were different under different sonicationconditions with a low frequency horn giving a smaller, more in-tense cavitation zone. In this work, we report direct observationof cavitation bubbles at the two frequencies using video and sono-luminescence methods to provide confirmatory evidence for thepreviously reported effects as well as reporting how the ultrasoundintensity and added surfactants change the cavitation bubble field.

2. Experimental

Sonication at 23 kHz was carried out with a Sonics and Materi-als VC 600 fitted with a 1 cm diameter titanium horn. 150 cm3 ofthe solution under investigation was measured into a beaker fittedwith a water jacket to allow for temperature regulation. All the re-sults reported were recorded at 20 4 C. The tip of the horn waspositioned 1.5 cm below the surface of the solution; care was takento ensure the horn and camera were placed in the same position forall experiments. A fresh sample of solution was used for eachexperiment when changing intensity. For higher frequency,515 kHz sonication, an Undatim UL03/1 reactor employing a5 cm diameter plate transducer was used. 150 cm3 of solutionwas contained in a jacketed cylinder over the transducer. Theintensity of ultrasound used was measured by calibrated calorim-etry in the usual manner [20].

High resolution video images were obtained using a SonyDCR106 video camera. In order to record sonoluminescenceimages, the apparatus was contained in a light-proof box. After sat-uration of the solution with Argon gas, images were recorded on anArtemis CCD camera with a 35 mm focal length lens capable of anf2.8 aperture and incorporating a Sony ICX285AL lowlight CCD sen-sor. The camera has an imaging resolution of 1392 1040 pixels(1.4 megapixels). The total intensity of the emission was calculatedafter subtraction of background levels using ImageJ software [21]which was also used for further image manipulation. Unless indi-Fig. 1. CCD images for luminol solutions sonicated at the indicated intensities (W cm2)right-hand image.

Fig. 2. CCD images for luminol solutions sonicated at the indicated intensities (W cm2) uright-hand image.cated below, images were collected for 60 s sonication. For someexperiments, enhanced images were obtained by sonicating a solu-tion of chemiluminescent luminol. This was prepared by dissolving1 mmol of luminol (3-aminophthalhydrazide, 97%), 0.1 mol hydro-gen peroxide and 0.1 mol EDTA (ethylenediaminetetraacetic acid)in 1 dm3 of 0.1 M sodium carbonate. The solution was adjustedto pH 12 by adding sodium hydroxide.

All chemicals were obtained from Aldrich (UK). Aqueous solu-tions were prepared in deionised water from a MilliQ system thathad a resistance >10 MX.

3. Results and discussion

In order to visualise the cavitation field under the different son-ication conditions, initial experiments recorded the chemilumines-cent emission from luminol solutions. Luminol mapping relies onemission from luminol that has captured a hydroxyl radical pro-duced by sonolysis of water during cavitation collapse. It thus givesa good indication of where chemically active cavitation bubbles oc-cur in a system. Fig. 1 shows the results for a 23 kHz horn system.As expected, the cavitation activity is largely concentrated in acone just under the horn. As the intensity increases, the field ofactivity gets larger indicating a larger population of active bubbles.

The corresponding results for the 515 kHz system are shown inFig. 2. The differences from the horn system are clear. The cavita-tion field is much larger and more diffuse. This is consistent withprevious observations [18] of Petrier and co-workers that high fre-quency transducers give luminol emission from a much larger vol-ume of a sonicated solution than the small, intense region aroundthe tip of a 20 kHz horn. In the current work, the cavitation field islayered, indicating that there is a standing wave field; the spacingbetween the bright layers corresponds to the wavelength of soundin water at this frequency. It is also consistent with suggestions[1214] that mainly stable cavitation is produced in this type ofsystem since it would predominate in a standing wave field.

Although not as apparent as in the 23 kHz horn system, theamount of light emission and hence cavitation increases withusing a 23 kHz horn. The position of the horn and the container are indicated on the

sing the 515 kHz plate transducer. The position of the transducer is indicated on the

Fig. 3. The effect of ultrasound intensity on the total integrated sonochemilumi-nescence emission from sonicated luminol solutions.

Fig. 6. Effect of SDS concentration on the SL emission intensity during sonication(23 kHz horn at 29 W cm2, 515 kHz at 0.31 W cm2).

32 G.J. Price et al. / Ultrasonics Sonochemistry 17 (2010) 3033the ultrasound intensity. This can be seen in Fig. 3 which plots thetotal emission integrated over the exposure time. Note that theluminescence is plotted on the same scale but the sound intensitiesare very different. This is partly a consequence of the larger emit-ted area of the 515 kHz plate but shows that higher cavitationactivity as measured by hydroxyl radical production is producedat the higher frequency. The sound energy emitted correspondedto 1.26.0 W compared with 1260W into the same volume ofsolution with the 23 kHz horn.

The light emission patterns recorded in Figs. 1 and 2 arise fromsecondary reactions with the products of cavitation collapse. Tofurther investigate the occurrence of cavitation and to eliminateany effect due to the trapping efficiency of the luminol or the life-time of the luminol excited state or to other reactions, images oftrue sonoluminescence (SL) from argon-saturated pure water wererecorded and are shown in Fig. 4. These measurements are at thelimit of our CCD camera and longer exposure times (5 min) wereneeded to obtain satisfactory images. They are similar in form tothose from the luminol solutions and showed a similar increaseFig. 4. Sonoluminescence from sonicated water at the indicated inte

Fig. 5. SL images of SDS solutions with the indicated concentrations (of emission with rising intensity. The difference in the nature ofthe sound field between the two systems is thus confirmed. How-ever, it is noticeable that the volume of solution from which emis-sion occurs at 20 kHz in Fig. 1 is larger than that in Fig. 4aindicating that chemical effects (in this case reaction with sonolyt-ically generated radicals) occurs around bubbles that are not nec-essarily sonoluminescent. This in part arises from the lifetime ofthe luminol excited state but also suggests that chemical effectsoccur in and around bubbles that do not reach the very high tem-peratures needed for sonoluminescence.

Previous work looking at the effects of additives such as surfac-tants on SL and acoustic emission led Ashokkumar et al. [14,22] tosuggest that, in a stable cavitation field, the number of active bub-bles is largely influenced by adsorption at the bubble-solutioninterface preventing bubble coalescence. Fig. 5 shows the effectof adding the surfactant sodium dodecyl sulfate (SDS) on the SLemission at 515 kHz. The emission intensity is plotted as a functionof SDS concentration in Fig. 6. There is a significant increase innsities (W cm2) (a) 23 kHz horn, (b) 515 kHz plate transducer.

mM) sonicated with a 515 kHz plate transducer at 0.31 W cm2.

Fig. 7. Photographs of 515 kHz sonication at 0.31 W cm2. (a) pure water (b) aqueous 1 mM SDS solution (c) aqueous 1 mM SDS + 0.1 M sodium perchlorate.

G.J. Price et al. / Ultrasonics Sonochemistry 17 (2010) 3033 33emission from solutions with low concentrations of SDS before itreturns to levels similar to that from water at higher concentra-tions. The standing wave nature of this cavitation field is still read-ily apparent. In contrast, with the 20 kHz horn, there was littlediscernable effect on the field of bubbles. The plume of activityemanating from the horn appeared somewhat larger at higherSDS concentrations although the total measured emission, alsoshown in Fig. 6, remained fairly constant and fell at high concen-trations. Previous observations of this type [14] have been attrib-uted to electrostatic repulsion between small, SL (andsonochemically) active bubbles preventing their coalescence intolarger, inactive bubbles. This effect is lessened when using a23 kHz horn since the sound field produces much greater turbu-lence and bubble motion so that small changes in inter-bubblerepulsions have much smaller influence. Interestingly, when theemission from luminol solutions was investigated as a functionof added SDS, a significant decrease in emission was observed forboth sonication systems. We interpret this as SDS adsorbed atthe bubble-solution interface trapping some of the hydroxyl radi-cals produced inside the bubble so that they cannot react withluminol in solution.

This bubble coalescence can in fact be observed visually in thesonicated solutions. Fig. 7 shows photographs taken from a videoof the system being sonicated. In pure water (Fig. 7a) a range ofbubble sizes can be observed. These are not individual cavitationbubbles but are gas bubbles which arise from coalescing bubbles.This is demonstrated by their behaviour when the sound field isswitched off when the bubbles remain intact for some 3060 s dur-ing which time they drift to the surface of the solution. On switch-ing the sound field back on, large bubbles immediately reappear. Incontrast, these larger bubbles are not visible during sonication of1 mmol dm3 solution of SDS (Fig. 7b). The electrostatic repulsionbetween cavitation bubbles prevents their coalescence into the lar-ger gas bubbles. Hence there are more active bubbles in the systemso that greater SL emission is observed. The cloud of cavitationbubbles is just visible in the video although difficult to discern inthe still photographs. In this case, switching off the sound resultsin the cavitation bubbles disappearing instantaneously. Signifi-cantly, addition of salt to this solution eliminates the surfactant ef-fect and larger gas bubbles are again observed (Fig. 7c). The systembehaves in an identical manner to water since the salt screens andhence eliminates the electrostatic shielding of the surfactant.4. Conclusions

The optical and SL photographs presented here confirm previ-ously reported results that the cavitation fields produced by a515 kHz plate transducer and a 23 kHz horn sonicator are signifi-cantly different. The higher frequency apparatus produces a stand-ing wave field. This further emphasises that when comparingresults in the literature, not only the frequency of ultrasound usedbut also the type of apparatus used must be considered. It is knownthat standing wave fields can be generated at low frequenciesaround 20 kHz [2326] and the nature of cavitation under theseconditions is currently being investigated.

References

[1] M. Ashokkumar, T. Mason, Sonochemistry, Kirk-Othmer Encyclopedia ofChemical Technology, John Wiley and Sons, 2007.

[2] K.S. Suslick, G.J. Price, Ann. Rev. Mater. Sci. 29 (1999) 295.[3] G. Cravotto, P. Cintas, Angew. Chem. Int. Ed. 46 (2007) 5476.[4] T. Leighton, The Acoustic Bubble, Academic Press, London, 1994.[5] F.R. Young, Cavitation, Imperial College Press, London, 1999.[6] A. Henglein, Ultrason. Sonochem. 2 (1995) S115.[7] P. Theron, P. Pichat, C. Guillard, C. Petrier, T. Chopin, Phys. Chem. Chem. Phys. 1

(1999) 4663.[8] L.K. Weavers, N. Malmstadt, M.R. Hoffmann, Environ. Sci. Technol. 34 (2000)

1280.[9] M. Beckett, I. Hua, J. Phys. Chem. A 105 (2001) 3796.[10] C. Petrier, A. Francony, Water Sci. Technol. 35 (1997) 17.[11] K. Okitsu, M. Ashokkumar, F. Grieser, J. Phys. Chem. B 109 (2005) 20673.[12] G.J. Price, M. Ashokkumar, F. Grieser, J. Am. Chem. Soc. 126 (2004) 2755.[13] G.J. Price, M. Ashokkumar, M. Hodnett, B. Zequiri, F. Grieser, J. Phys. Chem. B

109 (2005) 17799.[14] M. Ashokkumar, M. Hodnett, Z. Zeqiri, F. Grieser, G.J. Price, J. Am. Chem. Soc.

129 (2007) 2250.[15] K.S. Suslick, Y. Didenko, M. Fang, T. Hyeon, K. Kolbeck, W. McNamara, M.

Mdleleni, M. Wong, Phil. Trans. Roy. Soc. A 357 (1999) 335.[16] C. Petrier, A. Francony, Ultrason. Sonochem. 4 (1997) 295.[17] A. Benahcene, P. Labbe, C. Petrier, G. Reverdy, New J. Chem. 19 (1995) 989.[18] C. Petrier, M.-F. Lamy, A. Francony, A. Benahcene, B. David, V. Renaudin, N.

Gondrexon, J. Phys. Chem. 98 (1994) 10514.[19] A. Brotchie, M. Ashokkumar, F. Grieser, J. Phys. Chem. C 112 (2008) 8343.[20] T.J. Mason, D. Peters, Practical Sonochemistry Power Ultrasound Uses and

Applications, second ed., Ellis Horwood Publishers, 2002.[21] M.D. Abramoff, P.J. Magelhaes, S.J. Ram, Biophotonics Intl. 11 (2004) 36.[22] M. Ashokkumar, F. Grieser, Phys. Chem. Chem. Phys. 9 (2007) 5631.[23] F. Ratoarinoro, F. Contamine, A.-M. Wilhelm, J. Berlan, H. Delmas, Ultrason.

Sonochem. 2 (1995) S43.[24] H. Mitome, Jpn. J. Appl. Phys. 40 (2001) 3484.[25] L.A. Crum, G.T. Reynolds, J. Acoust. Soc. Am. 78 (1985) 137.[26] W. Lauterborn, T. Kurtz, R. Mettin, C.D. Ohl, Adv. Chem. Phys. 110 (1999) 295.

Direct observation of cavitation fields at 23 and 515kHzIntroductionExperimentalResults and discussionConclusionsReferences