Ultrasonics Sonochemistry 18 (2011) 864–872

Contents lists available at ScienceDirect

Ultrasonics Sonochemistry
journal homepage: www.elsevier.com/locate/ultsonch

The characterization of acoustic cavitation bubbles – An overview

Muthupandian Ashokkumar
School of Chemistry, University of Melbourne, VIC 3010, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Acoustic cavitation, in simple terms, is the growth and collapse of preexisting microbubbles under the
Received 7 October 2010 influence of an ultrasonic field in liquids. The cavitation bubbles can be characterized by the dynamics
Accepted 26 November 2010 of oscillations and the maximum temperatures and pressures reached when they collapse. These aspects
Available online 4 December 2010
can be studied both experimentally and theoretically for a single bubble system. However, in a multibub-
ble system, the formation of bubble streamers and clusters makes it difficult to characterize the cumu-
Keywords: lative properties of these bubbles. In this overview, some recently developed experimental procedures
Cavitation bubbles
for the characterization of acoustic cavitation bubbles have been discussed.
Bubble size
Bubble temperature
Ó 2010 Elsevier B.V. All rights reserved.
Transient cavitation
Stable cavitation

1. Introduction bubbles. An overview of these experimental procedures and their

Ultrasound, when passing through a liquid medium causes
mechanical vibration of the liquid. In addition to this effect, ultra-
sound also generates acoustic streaming within the liquid. If the li-
quid medium contains dissolved gas nuclei, which will be the case
under normal conditions, they can be grown and collapsed by the
action of the ultrasound. The phenomenon of growth and collapse
of microbubbles under an ultrasonic field is known as ‘‘acoustic
cavitation’’ [1]. When cavitation bubbles oscillate and collapse,
several physical effects are generated, namely, shock waves, micro-
jets, turbulence, shear forces, etc. The physical effects of ultrasound
have been used for a number of applications that include emulsifi-
cation, extraction, cleaning, etc. [2]. The collapse of the acoustic
cavitation bubbles is also near adiabatic and generates tempera-
tures of thousands of degrees within the bubbles for a short period
of time [3]. Under this extreme temperature conditions, highly
reactive radicals are generated. For example, if water is the med-
ium, H and OH radicals are generated by the homolysis of water.
These radicals have been used to achieve chemical reactions that
include the synthesis of nanomaterials, polymers, degradation of
organic pollutants, etc. [1,2,4–7]. In addition, acoustic cavitation
is found to be useful in diagnostic and therapeutic medicine [8].
Despite the use of ultrasonics and sonochemistry in a variety of
applications [4–12], the studies devoted to the fundamental under-
standing of the characteristics of cavitation bubbles are limited.
We have been, for the past 15 years, developing a number of exper-
imental procedures for the characterization of acoustic cavitation
E-mail address: masho@unimelb.edu.au
significance in understanding and controlling of acoustic cavitation
bubbles has been provided. In particular, the experimental tech-
niques for the characterization of the growth, temperature, lifetime
and capability (chemistry vs. light emission) of cavitation bubbles
have been briefly discussed in this overview. For detailed informa-
tion on these techniques, appropriate references that have been ci-
ted can be referred.
2. Bubble growth
Unlike in a single bubble system, the bubble growth in a multi-
bubble system involves two processes, namely, rectified diffusion
and bubble coalescence [13]. While it is possible to use experimen-
tal techniques such as light scattering, stroboscopic or fast video
recording to monitor the rectified diffusion growth of the bubbles
in a single bubble system [3,14], these techniques cannot be easily
adapted for multibubble systems. Added to this complexity is the
contribution from bubble coalescence to the growth of the bubbles.
Despite the complexities involved in the bubble growth process, a
simple experimental technique based on the initial growth of mul-
tibubble sonoluminescence (MBSL) has been developed by us to
qualitatively monitor the bubble growth by rectified diffusion
and bubble coalescence.
It has been shown that several acoustic pulses are required to
attain a steady-state MBSL in aqueous solutions under pulsed son-
ication conditions, in particular at high ultrasound frequencies
[15,16]. The reason for this induction period to reach a steady-state
MBSL is the absence of resonance sized bubbles at the initial stages
of sonication. The bubble nuclei present in the solution need to
grow to reach the resonance size range (see Fig. 1 [17]).

1350-4177/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.ultsonch.2010.11.016
 M. Ashokkumar / Ultrasonics Sonochemistry 18 (2011) 864–872
Fig. 1. Schematic representation of the growth of cavitation bubbles in an acoustic field and the corresponding experimental data (515 kHz, pulse on = 4 ms, pulse
off = 12 ms) on the growth of MBSL. Adapted from Ref. [17].
As mentioned above, the bubble growth in a multibubble filed
can occur by either rectified diffusion or bubble coalescence or
by a combination of both processes. We have used the initial MBSL
growth data to qualitatively understand the bubble growth under
pulsed sonication conditions. Fig. 2 shows that about fifteen 4 ms
pulses (duration between pulses is 12 ms) are required to reach a
steady-state bubble population of sonoluminescence bubbles in
water. Compared to water, the number of pulses required to reach
steady-state increases to about 30 in aqueous solution containing
1 mM of a surfactant (sodium dodecyl sulfate, SDS). Let us return
to this observation after some discussion on the role of surfactants
in bubble coalescence.
Surfactants are known to adsorb at bubble solution interface
and reduce the coalescence between bubbles [13]. We have used
a capillary technique to monitor bubble coalescence among cavita-
tion bubbles in the absence and presence of surface active solutes
[18]. The capillary technique involves the attachment of a capillary
that can measure the change in volume (DVT) that occurs due to
the formation of large inactive bubbles formed by coalescence
among cavitation bubbles. This is schematically shown in Fig. 3
along with a photograph of a 500 kHz ultrasound generator and
capillary setup. Thus, a large DVT means more coalescence and a
small DVT means less coalescence between bubbles.
Fig. 3 also shows that the relative coalescence between cavita-
tion bubbles in 1 mM SDS solution is significantly reduced to about
20% compared to that observed in water. The photographs shown
Fig. 2. The initial growth of MBSL as a function of acoustic pulse number. Frequency = 515 kHz, duty cycle = 4 ms on, 12 ms off.
865
in Fig. 4 also support the above discussion. In water, large coa-
lesced bubbles are clearly visible whereas, they are completely ab-
sent in 1 mM SDS solution.
Returning back to the discussion regarding the initial growth of
MBSL (Fig. 2), it is possible to interpret the data by bubble coales-
cence and rectified diffusion processes. In water, bubble coales-
cence occurs as shown by the capillary technique data (Fig. 3).
Hence, the initial growth of the cavitation bubbles may occur by
bubble coalescence pathway. However, rectified diffusion also con-
tributes to this growth. The support for this argument comes from
the SDS data in Fig. 3, which shows that SDS retards bubble coales-
cence. Hence, in the presence of SDS, the growth of cavitation bub-
bles by bubble coalescence pathway is significantly reduced. The
initial growth of MBSL observed in Fig. 2 must then be due to the
growth of cavitation bubbles primarily by rectified diffusion path-
way. Based on the number of acoustic pulses required for steady-
state MBSL in water, it can be suggested that the contributions
for the bubble growth form bubble coalescence and rectified diffu-
sion are approximately equal. In the absence of bubble coales-
cence, almost double the number of acoustic pulses is required in
SDS solution where the bubbles primarily grow by rectified diffu-
sion pathway. It should be clearly understood that the observed re-
sults are for a given ultrasound frequency under specific
experimental conditions (power, duty cycle, etc.). The number of
acoustic pulses required may be different under different experi-
mental conditions.
866 M. Ashokkumar / Ultrasonics Sonochemistry 18 (2011) 864–872
Fig. 3. A 500 kHz ultrasound generator (Undatim Ultrasonics) along with the capillary setup. A schematic diagram showing the volume change in water and surfactant
solution and the experimental coalescence data are also shown. Adapted from Ref. [17].
Fig. 4. Photographs taken during sonication of water and 1 mM SDS solution at
515 kHz. In water the formation of large bubbles due to coalescence is clearly seen.
In 1 mM SDS solution, no large bubbles are observed.
3. Bubble temperature
The bubble temperature is another important characteristic of
cavitation bubbles. Based on a simple thermodynamic model for
bubble collapse and assuming adiabatic compression takes place,
the maximum theoretical temperature within the bubble (Tmax)
can be calculated using Eq. (1) [19]:
 
Pm ðc  1Þ
T max ¼ T 0
Pv
T0 is the ambient solution temperature, Pm is the pressure in the
liquid (a sum of the hydrostatic and acoustic pressures), c ¼ ccPv is
the specific heat ratio of the gas/vapor mixture. Pv is the pressure
in the bubble at its maximum size; usually assumed to be equal
to the vapor pressure of the liquid, although it is known that gas
molecules originally dissolved in the fluid will also be present.
For example, if we assume that pm is 2 atm, the gas inside the bub-
ble is equal to the equilibrium water vapor pressure (Pv = 0.031
atm), the ambient solution temperature is 298 K and c for water
vapor is 1.32, then the theoretical estimate of Tmax will be
6150 K. It should be noted that Eq. (1) over estimates the Tmax,
because it does not take into account the heat leaking from the
bubble or thermal conductivity of the gases or the energy con-
sumed in the decomposition of the vapor/gas within the bubble.
Experimental determinations of the temperature within a cav-
itation bubble have been made by a number of research groups.
By fitting the experimentally recorded single bubble sonolumi-
nescence spectra using the spectrum radiated by a blackbody,
temperatures in the range of 5000 K – >50,000 K [3,19] have been
estimated. However, experimental estimates of the temperature
within the collapsing bubbles based on multibubble sonochemis-
try and sonoluminescence are reported to be between 750 K and
6000 K. Misík et al. [20], using the kinetic isotope effect in an EPR
spin-trapping study of the sonolysis of H2O/D2O mixtures found
that the cavitation temperature determined was dependent on
the specific spin trap used and fell in the range of 1000–4600 K.
They attributed the difference in the values obtained by the spin
traps as being due to the sampling of different regions of the cav-
itation ‘‘hot spot’’. Misík and Riesz [21] made use of the kinetic
isotope effect in the ultrasound induced production of radicals
in organic liquids to estimate the temperatures during cavitation.
The temperature region where hydrogen radical abstraction oc-
curred in n-dodecane was estimated to be about 750 K, whereas
the region where benzyl radical formation occurred in toluene
was about 6000 K. An interesting aspect of this work, as in the
H2O/D2O study described above, was the determination of a mean
temperature in different regions of a ‘‘hot spot’’ depending on
which radicals were detected. Using comparative rate thermome-
try in alkane solutions, Suslick et al. [22] postulated that there are
two regions of sonochemical reactivity: a gas phase zone within
the collapsing cavity with an estimated temperature and pressure
of 5200 ± 650 K and 500 atm, respectively, and a thin liquid layer
immediately surrounding the collapsing cavity with an estimated
ð1Þ temperature of 1900 K.
Sonolysis of methane in argon saturated water has been used by
Henglein and coworkers [23] to estimate the bubble core temper-
ature. Depending upon the percentage of methane and argon pres-
ent in water, the temperature was estimated to be in the range of
1930–2720 K (the temperature decreased with an increase in the
percentage of the methane). Tauber et al. [24] used a similar tech-
nique in their study of the sonolysis of t-butanol in water, and esti-
mated the temperature to be in the range 2300 K and 3600 K. The
lower temperature was obtained for solution with the highest con-
centration of t-butanol (0.5 M). Both these studies show that the
core temperature of the bubble is strongly dependent upon the
amount of organic solute in the system.
Ashokkumar and Grieser [25] have shown that there exists two
bubble temperatures: a time and volume averaged temperature
that is responsible for the chemical reactions and the peak temper-
ature that is responsible for the SL. In Fig. 5, it can be seen that the
bubble temperatures measured by the methyl radical recombina-
tion (MRR) method [23–26] is not significantly affected in the pres-
ence of up to 50 mM of ethanol whereas the SL intensity is almost
 M. Ashokkumar / Ultrasonics Sonochemistry 18 (2011) 864–872 867

Fig. 5. It is shown in this figure that the relative MBSL intensity decreases to more than 90% of that observed in water in the presence of ethanol whereas the bubble
temperature measured by the MRR method is not significantly affected by the same amount of ethanol. Frequency: 356 kHz. Adapted from Ref. [25].

90% quenched. It has been discussed that the volatile ethanol and
its decomposition products can decrease the peak temperature
reached upon bubble collapse without affecting the average tem-
perature, measured by the MRR method.
We have also measured the cavitation bubble temperatures
using the MRR method in argon saturated alcohol solutions at
20 kHz, 363 kHz, and 1056 kHz and the results are shown in
Fig. 6 [27]. It can be seen at 20 kHz that the bubble temperatures
did not significantly change in the presence of the alcohols. The
reason for almost no change in the bubble temperature may be
due to the transient nature of the cavitation bubbles, which is
discussed later. It can also be noticed that the average bubble tem-
peratures (chemical temperature discussed above) do decrease at
higher alcohol concentrations at higher frequencies: the cavitation
bubble temperatures measured decrease with an increase in the
concentrations of the alcohols and for a given concentration of
the alcohol, the longer chain alcohol shows a greater effect. The
reason for the lowering of the bubble temperatures at high fre-
quencies is the hydrocarbon product accumulation as a result of
the evaporation and decomposition of volatile solutes under the
high temperature conditions generated during bubble collapse
[28]. The higher surface activity of butanol compared to that of

Fig. 6. Cavitation bubble temperatures measured using the MRR [23–26] method at 20 kHz, 363 kHz and 1056 kHz as a function of the concentrations of ethanol and butanol.
Adapted from Ref. [27].
868 M. Ashokkumar / Ultrasonics Sonochemistry 18 (2011) 864–872
ethanol causes more butanol molecules to evaporate into the bub-
ble during the expansion phase of the bubble oscillation leading to
the production of a relatively larger amount of hydrocarbon prod-
ucts. A similar trend in MBSL quenching has been reported and dis-
cussed based on the surface activity of the alcohols [28].
The temperature data shown in Fig. 6 can also be used to obtain
the cavitation bubble temperature in the absence of any solute by
extrapolating the temperature data to zero concentration of the
solutes. By extrapolating to zero, both alcohols data yield a ‘‘zero’’
concentration temperature in the range 3700–4200 K at 20 kHz.
This variation in the cavitation bubble temperature might be due
to experimental variations. The bubble temperature obtained at
363 kHz (Fig. 6) by extrapolating to zero solute concentrations is
in the range of 4200–4700 K. A similar treatment at 1056 kHz
yields a bubble temperature in the range of 5000–6200 K. The tem-
perature trend thus observed for water at different frequencies is,
T1056 kHz > T363 kHz > T20 kHz. This trend is opposite to what can be
expected theoretically. At 20 kHz, the theoretical resonance size
of the bubble is about 150 lm; this is very large compared to the
bubble size at 1056 kHz, which is about 3 lm. A 20 kHz bubble
would have a large dV (change in volume) term on collapse leading
to a higher collapse temperature compared to that of a 1056 kHz
bubble. However, the observed trend can be explained, if the
amount of water vapor that evaporated within one acoustic cycle
was taken into account.
A 20 kHz bubble has a relatively longer expansion cycle time
than a bubble at a higher frequency; this would result in the evap-
oration of relatively more water molecules at 20 kHz than that at a
higher frequency. These molecules can effectively consume more
heat during the bubble collapse. Owing to a very short expansion
cycle time, a 1056 kHz bubble would not have as many water mol-
ecules within the bubble as at a lower frequency to consume the
heat energy.
Tronson [29] has calculated the total mass of water vapor that
could evaporate into an expanding bubble by using Eq. (2).
Z t¼t
mvap ¼ 4pJ obs R2 ðtÞdt
t¼0
where mvap is the upper bound for the total mass of water vapor
evaporated into the bubble, Jobs is the rate of efflux of water vapor
per unit area; =178 g s1 m2, R is the bubble radius and t is the
expansion time (approximately equal to the expansion cycle time
of the sound wave). This quantity can be compared to the mass
(in grams) of water present in a monolayer (mmon) at the surface
of the bubble at Rmax [29] using Eq. (3).
4pR2max Mw
mmon ¼
N A Aw
where NA is the Avagadro number, Mw is the molar mass of water
(=18 g mol1) and Aw is the cross sectional area of a molecule of
water at a surface, taken to be 11 Å2 in this case. Table 1 summa-
rises the mvap and mmon data calculated using Eqs. (2) and (3) for
the three frequencies used in this investigation.
The data in Table 1 clearly shows that the amount of water va-
por evaporated into the bubble is equivalent to several monolayers
of water for 20 kHz, whereas for the higher frequencies the expan-
sion time is not sufficient for the evaporation of one monolayer of
Table 1
mvap and mmon data along with Rmax and Ro data used in the calculations [27].
Frequency (kHz) Expansion cycle time (s)
6
20 25  10
360 1.4  106
1060 0.47  106
water from the surface of the bubble. Thus, the relatively lower
amount of water vapor present inside the collapsing bubbles at
higher frequencies results in a lower heat energy consumption that
leads to relatively higher temperatures on bubble collapse. It
should also be noted that the amount of primary radicals generated
per collapsing bubble is larger at lower frequencies. However, the
overall radical yield, which is the product of the radicals per bubble
and the total number of bubbles, is higher at higher frequencies.
This is due to the increase in the number of bubbles at higher fre-
quencies. In fact, the overall radical yield seems to be the maxi-
mum in the frequency range 200–600 kHz [1]. The radical yield
decreases at very high frequencies (for example, 1 MHz) due to
the lack of time available for evaporation of water as discussed
above.
4. Bubble size
While it is relatively convenient to monitor the cavitation bub-
ble size in a single bubble system using either light scattering or
stroboscopic imaging, it is difficult to monitor the bubble size in
a multibubble environment due to bubble clustering and interfer-
ence caused by bubble movements. A few experimental techniques
that are available to measure the bubble size are laser light diffrac-
tion [30], active cavitation detection [31] and phase-Doppler [32].
We have introduced a relatively simple technique based on pulsed
MBSL [33]. The method is based on the dissolution of the cavitation
bubbles during the ‘‘off’’ time of the ultrasound between pulses as
shown in Fig. 7. During the pulse ‘‘on’’ time, cavitation bubbles
grow as discussed earlier. During the pulse ‘‘off’’ time, bubbles dis-
solve. For a given duty cycle, the system continues to maintain a
steady-state population of active bubbles and approximately con-
stant MBSL intensity as shown in Fig. 2 [15,16].
It can be seen in Fig. 7 that the average MBSL intensity does not
change significantly when the ‘‘off’’ time is increased from 12 ms to
200 ms. However, beyond this value, the MBSL intensity decreases
and almost reaches zero beyond 290 ms. What this indicates is that
ð2Þ the active cavitation bubble population has reached a size range
that is below a critical value to grow the bubbles to that required
for SL to occur during the next pulse on period. This critical size
can be considered as equal to the resonance size of the bubbles:
bubbles that reach the resonance size range grow to a maximum
and collapse to emit light. It can be assumed that the bubbles are
just below the resonance size when the average MBSL intensity
reaches zero at this critical pulse off time. Using the Epstein–Ples-
set equation shown in Fig. 7 Caption and the procedure discussion
by Lee et al. [33], the resonance size of the cavitation bubbles can
be estimated. The data provided in Table 2 shows that the bubble
ð3Þ
size measured using the pulsed SL technique is in agreement with
that measured by other techniques.
5. Bubble lifetime (transient vs. stable)
There is always confusion with the terminology referring to
transient and stable bubbles. Stable bubbles are considered to be
weakly and symmetrically oscillating bubbles, whereas transient
bubbles are considered to be ‘‘active cavitation’’ bubbles. Leighton
[19] has provided an appropriate discussion to clarify this
Rmax (m) Ro (m) mvap (g) mmon (g)
6 6 10
150  10 15  10 6.2  10 7.7  1011
8  106 0.8  106 9.8  1014 2.2  1013
2.8  106 0.28  106 4.1  1015 2.7  1014
 M. Ashokkumar / Ultrasonics Sonochemistry 18 (2011) 864–872
Fig. 7. The growth and dissolution of cavitation bubbles under pulsed sonication conditions (left); MBSL observed during the 4 ms pulse on time at various pulse off times
(right top); Epstein and Plesset equation (right bottom) [33] for a dissolving bubble: D – diffusion coefficient, C – gas concentration, qg – gas density, Ro – initial bubble radius
(assumed to be equal to the resonance radius), M – molar mass of the gas, c – surface tension of the liquid, R – gas constant, T – temperature. Using the pulse off time for ‘‘t’’, Ro
is calculated.
confusion. He mentions, quote: ‘‘the terms ‘‘stable’’ and ‘‘transient’’
are perhaps unfortunate, since they suggest a temporal differenti-
ation, when in fact they are most useful in describing the energy
and effects of the collapse’’. For example, single bubble cavitation
can be considered as ‘‘stable’’ cavitation in a sense that the same
bubble undergoes continuous oscillations emitting light on col-
lapse during each cycle. The energy of the collapse is significant
in that it can also be classified as ‘‘transient collapse’’. In order to
overcome this confusion, Leighton further classified the high-en-
ergy collapse (transient cavitation) into ‘‘Fragmentary Transient
Cavitation’’ and ‘‘Repetitive Transient Cavitation (High-energy sta-
ble cavitation)’’. When we first investigated [28,34] cavitation at
low and high ultrasound frequencies, we have referred to fragmen-
tary transient cavitation as simply transient cavitation and the
high-energy stable cavitation as just stable cavitation. From the
collapse energy point of view, they both refer to high-energy col-
lapse. The transient cavitation bubbles then refer to high-energy
collapse followed by fragmentation and generation of new cavita-
tion nuclei and the stable cavitation bubbles refer to high-energy
collapse where same bubbles repetitively collapse and grow sev-
eral times.
We have provided experimental evidence based on MBSL
quenching where transient bubbles existed only for one or a few
Fig. 8. MBSL quenching at 20 kHz and 515 kHz as a function of alcohol concentra-
tion. Adapted from Ref. [34].
869
Table 2
Experimentally measured resonance bubble radii at various frequencies [33].
Experimental technique Frequency (kHz) Experimental Ro (lm)
Pulsed MBSL 515 2.8–3.7
Active cavitation 1100 0.9, 1.38
Laser diffraction 20 3.8
Phase-doppler 20 5.0
acoustic cycles whereas stable bubbles existed for hundreds of
acoustic cycles [34]. We observed that the sonoluminescence
intensity is decreased significantly in the presence of volatile sol-
utes at higher ultrasound frequencies (e.g., 515 kHz) whereas SL
intensity is not affected by the same solute at 20 kHz (Fig. 8). We
proposed that the accumulation of hydrocarbon products within
cavitation bubbles is important. At 515 kHz, the solution volume
(and height) used was sufficient to establish standing waves in
the reaction cell. Under standing wave conditions, bubbles undergo
‘‘stable’’ cavitation. At 20 kHz, the solution in the reaction cell was
not sufficient for the establishment of standing waves and under
these circumstances, bubbles underwent transient cavitation. Sta-
ble bubbles are considered to undergo a large number of oscilla-
tions during their existence and in doing so exist for a sufficient
time for surface active solutes to adsorb to the bubble/solution
interface, accumulate gaseous decomposition products, and to al-
low a significant amount of the gases solubilised in solution to en-
ter the bubble. In contrast, for transient bubbles these events can
be significantly reduced.
Ashokkumar et al.’s recent work [35] on MBSL at low and high
frequencies introduced another complexity to the behavior of the
cavitation bubbles. It was thought that the reason for the genera-
tion of transient bubbles might be the absence of a standing wave
in a horn system and the stable bubbles can only be generated un-
der standing wave conditions. In order to clarify this aspect, MBSL
quenching experiments were carried out at four ultrasound fre-
quencies keeping the sonication cell configuration the same. The
solution height and volume were chosen to make sure that stand-
ing waves could be formed at all frequencies. In the first system, a
20 kHz horn was positioned closer to the water/air interface (see
Fig. 9). When the MBSL was photographed using this arrangement,
active cavitation was observed closer to the tip of the horn only.
Despite the possibility for a standing wave, there were no active
cavitations observed in any other region of the reactor. However,
when a 25 kHz plate transducer was attached at the bottom of
the sonication cell (reactor), standing waves were visible in the
870 M. Ashokkumar / Ultrasonics Sonochemistry 18 (2011) 864–872

Fig. 9. (a) Photographs showing the sonication cell used at 20 kHz (horn; left), MBSL at 20 kHz (horn) and MBSL at 37 kHz (right); (b) MBSL quenching data at different
frequencies. Adapted from Ref. [35].

Fig. 10. (a) Sonoluminescence from water and (b) sonochemiluminescence from an aqueous solution containing luminol; frequency = 170 kHz; power = 12 W [39].

MBSL photograph. A similar pattern was observed at 37 kHz (Fig. 9)
and at 440 kHz. The MBSL quenching data observed in the presence
of various concentrations of propanol is also shown in Fig. 9.
One interesting aspect that can be seen immediately is a rela-
tively small SL quenching (about 20%) observed with 20 kHz horn
system compared to almost no quenching reported by Tronson
et al. [34]. It may be speculated that there might be some active
stable bubbles present in the 20 kHz system used by Ashokkumar
et al. [35] that were not visible in the photograph. However, com-
pared to the SL quenching observed with the horn system, about
80% SL quenching was observed at 25 kHz and 100 mM propanol
when a plate transducer was used. A similar SL quenching was also
observed at 37 kHz. Considering that the SL quenching observed at
25 kHz and 37 kHz were almost similar to that observed at
447 kHz, it was suggested that these systems consisted of primar-
ily stable bubbles. The reason for the generation of primarily tran-
sient bubbles by the 20 kHz horn is the high energy density
delivered at the tip of the horn compared to the more diffused en-
ergy density delivered in plate type transducers. It has been esti-
mated that the stable bubbles may have lifetimes of up to 5000
acoustic cycles depending upon the acoustic frequency [36]. Using
bubble coalescence data, Sunartio et al. [37] have estimated the
bubble lifetimes at different ultrasound frequencies, which range
from 100 to 350 ls that correspond to 70–100 acoustic cycles.
6. SL bubbles vs. SC bubbles
The data shown in Fig. 5 indicates that the lower MBSL does not
directly correlate with the cavitation activity of a system. When
MBSL intensity is decreased by more than 90% in the presence of
small amounts of ethanol, there are still hydrocarbon products pro-
duced in the bubbles indicating the existence of cavitation bubbles
generating sonochemical reactions. As indicated previously, the
cavitation bubbles may not attain the temperatures required for
SL to occur, however, the temperatures may be sufficient to
decompose molecules. In following reports, we have used sonolu-
minescence and sonochemiluminescence photographs to show the
existence of two groups of cavitation bubbles in a given reactor
[38,39]. One group of bubbles reaches higher temperatures for SL
to occur and the 2nd group causing chemical reactions. When
water was sonicated at 170 kHz, sonoluminescence was observed
primarily closer to the water/air interface as can be seen by the
weak emission in Fig. 10a. However, when luminol was present
 M. Ashokkumar / Ultrasonics Sonochemistry 18 (2011) 864–872
Fig. 11. Bubble radius distribution of SL and SC bubbles. Frequency: 575 kHz.
Adapted form Ref. [40].
in the same reactor, chemiluminescence was observed from more
than 90% of the reactor under the same experimental conditions
(Fig. 10b). It is well known that the luminol chemiluminescence
is initiated by the reaction between cavitation generated OH radi-
cals and luminol molecules.
It was also shown that the SC bubbles appeared at relatively
lower acoustic power levels compared to the SL bubbles at a given
frequency and the power required to observe SL bubbles increased
with an increase in the ultrasound frequency [39]. In this study, it
was speculated that the sonoluminescence bubbles may reach a
relatively larger size and generate higher Tmax compared to those
bubbles that are sonochemically active. Further experimental sup-
port to this speculation was later obtained using the bubble size
measurement procedure discussed earlier. The bubble size distri-
butions were measured using pulsed SL and pulsed SCL and the re-
sults are shown in Fig. 11 [40].
It is clear from this data that the size distribution of SL bubbles
is relatively larger than that of SC bubbles and supports the spec-
ulation that two groups of bubbles exist in a reactor. Those bubbles
located in high energy regions can grow to relatively larger sizes
and attain relatively higher temperatures to generate SL. These
bubbles are relatively lower in numbers compared to the sono-
chemically active bubbles as can be seen in Fig. 10.
7. Final remarks
This overview is focused on only a few aspects of acoustic cav-
itation bubbles. It is known that the presence of solutes [41] affect
the cavitation bubbles in a number of ways. One important issue
that still remains is the estimation of bubble populations. Iida
and coworkers [42] have recently used pulsed laser diffraction
method to estimate the bubble population. A number of other
groups [43–50] have studied the cavitation structures and other
aspects that have not been discussed in this overview. However,
the aspects that are covered in this article expose some simple
experimental techniques that have been developed to characterize
the cavitation bubbles, which will be highly beneficial to the son-
ochemistry community.
References
[1] M. Ashokkumar, T. Mason, ‘‘Sonochemistry’’, Kirk-Othmer Encylcopedia of
Chemical Technology, John Wiley and Sons, 2007, doi:10.1002/
0471238961.1915141519211912.a01.pub2, Article Online Posting Date:
October 19, 2007.
[2] T.J. Mason (Ed.), Advances in Sonochemistry, vol. 1–6, Elsevier, Amsterdam,
The Netherlands, 1990–2001.
[3] F.R. Young, Cavitation, McGraw-Hill, London, 1994.
871
[4] K.S. Suslick, Ultrasound: Its Chemical, Physical and Biological Effects, VCH
Publishers, New York, 1988.
[5] G. Price (Ed.), Current Trends in Sonochemistry, RSC, Cambridge, UK, 1992.
[6] M. Ashokkumar, Sonochemical synthesis of inorganic nanoparticles, in: P.D.
Cozzoli (Ed.), Advanced Wet-Chemical Synthetic Approaches to Inorganic
Nanostructures, Transworld Research Network, 2008, pp. 107–131 (Chapter
4).
[7] S. Muthukumaran, S.E. Kentish, M. Ashokkumar, G.W. Stevens, Application of
ultrasound in membrane separation processes: a review, Rev. Chem. Eng. 22
(2006) 155–194.
[8] F. Cavalieri, M. Zhou, M. Ashokkumar, The design of multifunctional
microbubbles for ultrasound image-guided cancer therapy, Curr. Top. Med.
Chem. 10 (2010) 1198–1220.
[9] R. Bhaskaracharya, S.E. Kentish, M. Ashokkumar, Selected applications of
ultrasonics in food processing, Food Eng. Rev. 1 (2009) 31–49.
[10] T.J. Mason, C. Petrier, in: S. Parsons (Ed.), Advanced Oxidation Processes for
Water and Wastewater Treatment, IWA Publishing, 2004, pp. 185–208.
[11] J.-L. Luche (Ed.), Synthetic Organic Chemistry, Plenum Press, New York, 1998.
[12] M.J. Povey, T.J. Mason (Eds.), Ultrasound in Food Processing, Springer, New
York, 1998.
[13] M. Ashokkumar, J. Lee, S. Kentish, F. Grieser, Bubbles in an ultrasonic field: an
overview, Ultrason. Sonochem. 14 (2007) 470–475.
[14] J. Lee, S. Kentish, M. Ashokkumar, Effect of surfactants on the rate of growth
of an air bubble by rectified diffusion, J. Phys. Chem. B 109 (2005) 14595–
14598.
[15] A. Henglein, R. Ulrich, J. Lilie, Luminescence and chemical action by pulsed
ultrasound, J. Am. Chem. Soc. 111 (1989) 1974–1979.
[16] M. Ashokkumar, R. Hall, P. Mulvaney, F. Grieser, Sonoluminescence from
aqueous alcohol and surfactant solutions, J. Phys. Chem. B 101 (1997) 10845–
10850.
[17] J. Lee, The Behavior of Ultrasound Generated Bubbles in the Presence of
Surface Active Solutes, Ph.D. Thesis, University of Melbourne, 2005.
[18] J. Lee, S. Kentish, M. Ashokkumar, The effect of surface-active solutes on bubble
coalescence in the presence of ultrasound, J. Phys. Chem. B 109 (2005) 5095–
5099.
[19] T.G. Leighton, The Acoustic Bubble, Academic Press, London, 1994.
[20] V. Misík, N. Miyoshi, P. Reisz, EPR spin-trapping of the sonolysis of H2O/D2O
mixtures: probing the temperatures of cavitation regions, J. Phys. Chem. 99
(1995) 3605–3611.
[21] V. Misík, P. Reisz, EPR study of free radicals induced by ultrasound in organic
liquids II. Probing the temperatures of cavitation regions, Ultrason. Sonochem.
3 (1996) 25–37.
[22] K.S. Suslick, D.A. Hammerton, R.E. Cline, Sonochemical hotspot, J. Am. Chem.
Soc. 108 (1986) 5641–5642.
[23] E.J. Hart, C.-H. Fischer, A. Henglein, Sonolysis of hydrocarbons in aqueous
solution, Radiat. Phys. Chem. 36 (1990) 511–516.
[24] A. Tauber, G. Mark, H.-P. Schuchmann, C. Von Sonntag, Sonolysis of tert-
butyl alcohol in aqueous solution, J. Chem. Soc., Perkin Trans. 2 (1999)
1129–1135.
[25] M. Ashokkumar, F. Grieser, A comparison between multibubble
sonoluminescence intensity and the temperature within cavitation bubbles,
J. Am. Chem. Soc. 127 (2005) 5326–5327.
[26] J. Rae, M. Ashokkumar, O. Eulaerts, C. Von Sonntag, J. Reisse, F. Grieser,
Estimation of ultrasound induced cavitation bubble temperatures in aqueous
solutions, Ultrason. Sonochem. 12 (2005) 325–329.
[27] T. Vu, Effect of Organic and Inorganic Solutes on Multibubble
Sonoluminescence in Aqueous Solutions, M.Sc. Thesis, University of
Melbourne, 2004.
[28] M. Ashokkumar, P. Mulvaney, F. Grieser, The effect of pH on multibubble
sonoluminescence from aqueous solutions containing simple organic weak
acids and bases, J. Am. Chem. Soc. 121 (1999) 7355–7359.
[29] R. Tronson, The Effect of Surface Active Solutes and Ultrasound Frequency on
Sonoluminescence in Aqueous Solutions, University of Melbourne, Ph.D.
Thesis, 2004.
[30] F. Burdin, N.A. Tsochatzidis, P. Guiraud, A.M. Wilhelm, H. Delmas,
Characterization of the acoustic cavitation cloud by two laser techniques,
Ultrason. Sonochem. 6 (1999) 43–51.
[31] W.S. Chen, T. Matula, L.A. Crum, The disappearance of ultrasound cavitation
bubbles: observation of bubble dissolution and cavitation nuclei, Ultrasound
Med. Biol. 28 (2002) 793–803.
[32] N.A. Tsochatzidis, P. Guiraud, A.M. Wilhelm, H. Delmas, Determination of
velocity, size and concentration of ultrasonic cavitation bubbles by phase-
Doppler technique, Chem. Eng. Sci. 56 (2001) 1831–1840.
[33] J. Lee, M. Ashokkumar, S. Kentish, F. Grieser, Determination of the size
distribution of sonoluminescence bubbles in a pulsed acoustic field, J. Am.
Chem. Soc. 127 (2005) 16810–16811.
[34] R. Tronson, M. Ashokkumar, F. Grieser, Comparison of the effects of water
soluble solutes on multibubble sonoluminescence generated in aqueous
solutions by 20- and 515-kHz pulsed ultrasound, J. Phys. Chem. B 106
(2002) 11064–11068.
[35] M. Ashokkumar, J. Lee, Y. Iida, K. Yasui, T. Kozuka, T. Tuziuti, A. Towata, The
detection and control of stable and transient cavitation bubbles, Phys. Chem.
Chem. Phys. 11 (2009) 10118–10121.
[36] R. Tronson, M. Ashokkumar, F. Grieser, Multibubble sonoluminescence from
aqueous solutions containing mixtures of surface active solutes, J. Phys. Chem.
B 107 (2003) 7307–7311.
872 M. Ashokkumar / Ultrasonics Sonochemistry 18 (2011) 864–872
[37] D. Sunartio, M. Ashokkumar, F. Grieser, Study of the coalescence of acoustic
bubbles as a function of frequency, power and water soluble additives, J. Am.
Chem. Soc. 129 (2007) 6031–6036.
[38] D. Sunartio, K. Yasui, T. Tuziuti, T. Kozuka, Y. Iida, M. Ashokkumar, F. Grieser,
Correlation between Na⁄ emission and chemically active acoustic cavitation
bubbles, ChemPhysChem 8 (2007) 2331–2335.
[39] M. Ashokkumar, J. Lee, Y. Iida, K. Yasui, T. Kozuka, T. Tuziuti, A. Towata, Spatial
distribution of acoustic cavitation bubbles at different ultrasound frequencies,
ChemPhysChem 11 (2010) 1680–1684.
[40] A. Brotchie, F. Grieser, M. Ashokkumar, Effect of power and frequency on
bubble size distributions in acoustic cavitation, Phys. Rev. Lett. 102 (2009)
0843021–0843024.
[41] M. Ashokkumar, F. Grieser, The effect of surface active solutes on bubbles in an
acoustic field, Phys. Chem. Chem. Phys. 9 (2007) 5631–5643.
[42] Y. Iida, M. Ashokkumar, T. Tuziuti, T. Kozuka, K. Yasui, A. Towata, J. Lee, Bubble
population phenomena in sonochemical reactor: I estimation of bubble size
distribution and its number density with pulsed sonication – laser diffraction
method, Ultrason. Sonochem. 17 (2010) 473–479.
[43] W. Lauterborn, Cavitation bubble dynamics-new tools for an intricate
problem, Appl. Sci. Res. 38 (1982) 165–178.
[44] W. Lauterborn, T. Kurz, Physics of bubble oscillations, Rep. Prog. Phys. 73
(2010) 1065011–1065088.
[45] J. Lee, K. Yasui, T. Tuziuti, T. Kozuka, A. Towata, Y. Iida, Spatial distribution
enhancement of sonoluminescence activity by altering sonication and solution
conditions, J. Phys. Chem. B 112 (2008) 15333–15341.
[46] J.F. Guan, T.J. Matula, Understanding the pulse-to-pulse evolution of
microbubble clouds using light scattering, IEEE Ultrason. Symp. 1–3 (2004)
218–221.
[47] N. Segebarth, O. Eulaerts, J. Reisse, L.A. Crum, T.J. Matula, Correlation between
acoustic cavitation noise, bubble population and sonochemistry, J. Phys. Chem.
B 106 (2002) 9181–9190.
[48] A. Otta, T. Nowak, R. Mettin, F. Holsteyns, A. Lippert, Characterization of a
cavitation bubble structure at 230 kHz: bubble population, sonolumines-
cence and cleaning potential, Solid State Phenomena 145–146 (2009) 11–13.
[49] R. Mettin, A.A. Doinikov, Translational instability of a spherical bubble in a
standing ultrasound wave, Appl. Acoustics 70 (2009) 1330–1339.
[50] M. Arora, O.D. Ohl, D. Lohse, Effect of nuclei concentration on cavitation cluster
dynamics, J. Acoust. Soc. Am. 121 (2007) 3432–3436.