17282
J. Phys. Chem. B 2006, 110, 17282-17285
Effect of Alcohols on the Initial Growth of Multibubble Sonoluminescence Judy Lee,‡ Muthupandian Ashokkumar,*,‡ Sandra Kentish,† and Franz Grieser‡ Particulate Fluids Processing Centre, Department of Chemical and Biomolecular Engineering, School of Chemistry, UniVersity of Melbourne, VIC 3010, Australia ReceiVed: May 29, 2006; In Final Form: July 11, 2006
The effect of alcohols on the initial growth of the multibubble sonoluminescence (MBSL) intensity in aqueous solutions has been investigated. With increasing concentrations of the alcohols, the number of pulses required to grow the MBSL intensity to a steady state (Ncrit) increases (relative to that of water) initially to a maximum for all the alcohols used in this study, followed by a decrease for methanol and ethanol. The cause of the initial increase in Ncrit is attributed to the inhibition of bubble coalescence in the system. This inhibition in bubble coalescence results in a population of bubbles with a smaller size range and thus a larger number of pulses is required to grow the bubbles to their sonoluminescing size range. It is suggested that the decrease in the Ncrit at higher alcohol concentrations may be caused by an increase in the bubble growth by rectified diffusion.
Introduction Ultrasound waves passing through a liquid medium can cause the occurrence of a phenomenon known as acoustic cavitation. This term is used to describe the creation of cavities (bubbles) or the expansion of preexisting cavities in the liquid in the presence of ultrasound.1 The term cavitation is also used loosely to encompass other ultrasound induced effects such as bubble growth by rectified diffusion and bubble collapse leading to the emission of light, known as sonoluminescence. The rectified diffusion process involves the diffusion of gas/vapor molecules in and out of a cavitation bubble. When a cavitation bubble expands, gas/vapor molecules diffuse into the bubble. The gas/ vapor molecules are forced out of the bubble during the compression phase of the bubble. However, due to the difference in the surface area during the expansion and compression cycles, the amount of gas/vapor that diffuses into the bubble is more than that diffuses out of the bubble. As a result of this, the amount of material inside the bubble increases with an increase in the number of bubble oscillations leading to an increase in the size of the bubble (bubble growth). Figure 1 provides an overview of the different events the bubbles could undergo in an ultrasonic field. The coalescence between cavitation bubbles may lead to the formation of “over-sized” bubbles (above the resonance size range), which may exit the pressure antinode regions due to acoustic field gradients. The coalescence between the bubble nuclei may also generate bubbles in the resonance size range, more rapidly than if the bubbles were to grow by a rectified diffusion pathway alone. By measuring the intensity of the SL emitted under different solution conditions, one can gain an insight into the bubble dynamics. In addition to this, by applying the ultrasound in a pulsed mode, it is possible to capture changes in the SL intensity with time. Henglein et al.2-4 reported the effect of pulsed ultrasound on the initial chemiluminescence (CL) growth in luminol solutions. The initial CL growth required to reach the * Address correspondence to this author. E-mail:
[email protected]. ‡ School of Chemistry. † Department of Chemical and Biomolecular Engineering.
Figure 1. Various events that bubbles in an acoustic field may undergo.
maximum (equilibrium) CL intensity was thought to be brought about by the existence of two types of cavitation nuclei, stable (preexisting) nuclei and short-lived (transient) nuclei generated by the ultrasound. The finite amount of time required to build up the CL intensity was associated with an increase in the ultrasound generated bubble nuclei. The time required to reach the steady-state CL was measured by operating the ultrasound unit in a pulsed mode and recording the number of pulses required to reach a stable level of CL (Ncrit). The number of pulses required to reach this pulse intensity increased with increasing pretreatment (sonication of the solution and then left to stand for a fixed period) time, decreasing air concentration, and a decrease in the on/off ratio of the pulse modulation. This was explained in terms of the removal of the stable nuclei by the pretreatment and deaeration processes.3 It has been reported in the literature that the presence of surface active solutes has a significant effect on the SL intensity.5-11 Ashokkumar et al.9 investigated the effect of surface active solutes on the number of pulses or Ncrit required to reach a steady-state MBSL intensity. This work differs from that of Henglein2-4 in that the SL was directly monitored instead of CL. They found that the addition of 1 mM of SDS increased
10.1021/jp063320z CCC: $33.50 © 2006 American Chemical Society Published on Web 08/10/2006
Effects on the Growth of Multibubble Sonoluminescence
J. Phys. Chem. B, Vol. 110, No. 34, 2006 17283
both the steady-state SL intensity and Ncrit relative to that of water. Upon the addition of aliphatic alcohols, a decrease in the steady-state SL intensity relative to that of water was observed. No effect on the Ncrit was observed when a low concentration of alcohol was added. However, a limited set of results were obtained and only one alcohol concentration was examined. In their report on MBSL at 20 kHz, Price et al.12 have shown that the addition of alcohol increases the transient cavitation bubble activity. They have suggested that this increase in the number of cavitation bubbles might be due to a hindrance in bubble coalescence by the interfacially adsorbed alcohols. However, at higher frequencies the importance of bubble coalescence in increasing the number of active bubbles is difficult to realize due to the dominating SL quenching activity by the alcohols.9 The aim of this study was to investigate the effect of different alcohols with varying concentrations on Ncrit and to use this information to gain a better understanding of the behavior of bubble clusters under the influence of an acoustic field. The Ncrit data have also been compared with the bubble coalescence data for the alcohols. Experimental Details Methanol (99.8%) and ethanol (99.5%) were purchased from Ajax Chemicals. Butan-1-ol (99.5%) and pentan-1-ol (98.5%) were purchased from BDH. The solutions were prepared with Milli-Q water (conductivity of less than 10-6 S cm-1 at 20 °C) that had been left overnight to allow air saturation equilibrium to be reached. The setup used by Tronson et al.10 was adopted in this investigation for measuring the SL intensity. A cylindrical Pyrex cell, filled with 50 mL of solution, was mounted over a 35 mm, 515 kHz flat plate unfocused transducer. The transducer was connected to a 515 kHz Undatim Ultrasonics D-reactor generator with an adjustable power setting between 0 and 100 W. The generator was modified in-house to allow for pulsed energy delivery where both pulse on and off time could be varied. For this work, the on and off time of the acoustic pulse “train” was fixed at 4 and 12 ms, respectively. The acoustic power used was 15.7 W, measured calorimetrically. A Hamamatsu end-on photomultiplier tube (model No. R64704) was used to record the SL intensity and the output from the photomultiplier tube was fed to a digital oscilloscope (Tektronix, model No. TDS 320). The oscilloscope, triggered by the ultrasonic generator, recorded the sequence of signals at a set time interval. The data were then displayed on the oscilloscope and saved on a computer. The cell and the photomultiplier tube were both housed in a light-proof enclosure to minimize background light being detected by the photomultiplier. The solution was first sonicated for 30 s. The solution was then allowed to stand for 5 min to enable any bubbles formed within the 30 s sonication to dissolve away. The initial pulse train capturing the growth of the SL intensity was then recorded. This preliminary sonication procedure allowed reproducible results to be obtained. Total bubble volume measurements (∆VT) were obtained by using the experimental methodology described in detail by Lee et al.5 at a calorimetric power of 9.3 W. In this method, the volume changes caused during 5 s sonication of aqueous solutions were measured using a capillary technique. This volume change corresponds to the total volume of bubbles produced by bubble-bubble coalescence.5
Figure 2. Relative SL intensity observed for (a) water, (b) 1.0 M methanol, and (c) 1.5 M methanol as a function of initial ultrasound pulses.
Results and Discussion The SL intensity generated from a “train” of 4 ms pulses of 515 kHz ultrasound, with 12 ms interval between pulses, for water and two methanol concentrations, is shown in Figure 2. Figure 2a demonstrates that a finite number of pulses is required for the majority of the bubbles to reach the active size range (i.e., the size range where bubble collapse produces sonoluminescence and sonochemistry) and thus to produce a steady-state SL intensity. In this study, approximately 10 pulses were observed for water under the experimental system/conditions used. This number of pulses required to grow the active bubble population to the steady-state population is referred to as Ncrit. It can also be noted that there appears to be a fluctuation in the intensity of the SL pulses once the steady-state level is reached. It is likely that this effect is due to the change in the acoustic field due to reflections from the bubble clouds in the system.4 Panels b and c in Figure 2 show the effect of 1.0 and 1.5 M methanol, respectively, on the initial growth of the MBSL. As reported by Ashokkumar et al.,9 the intensity of the steadystate SL decreases with increasing alcohol concentrations: this decrease in SL intensity is due to the decomposition of the alcohol that evaporates into the cavitation bubble, which leads to the formation and accumulation of hydrocarbon products within the cavitation bubbles.9 Irrespective of this quenching,
17284 J. Phys. Chem. B, Vol. 110, No. 34, 2006
Figure 3. Effect of methanol, ethanol, butan-1-ol, and pentan-1-ol on Ncrit relative to water.
Figure 4. Relative change in total bubble volume and Ncrit as a function of alcohol concentration. The changes in total bubble volume have been normalized relative to the average change in total volume for water. The filled symbols indicate the Ncrit values and the open symbols indicate the total bubble volume values.
a qualitative comparison of data presented in Figures 2a-c reveals that the presence of the alcohol also affects the Ncrit. The influence of alcohol on Ncrit as a function of methanol, ethanol, butan-1-ol, and pentan-1-ol concentrations is shown in Figure 3. For methanol and ethanol, the Ncrit increases to a maximum and then decreases. For butan-1-ol and pentan-1-ol, Ncrit increases but a decrease is not obvious within the concentration range used. For these longer chain alcohols the use of higher concentrations was restricted by their solubility in water. Another difficulty encountered was that the MBSL quenching by these alcohols was significant, which restricted the accuracy of the initial SL growth measurements at high concentrations. To understand the observed increase in the Ncrit by alcohols, the coalescence between acoustic cavitation bubbles must be considered. Lee et al.5 have measured the change in total bubble volume, using a capillary system, during the sonication of aqueous solutions containing surface active solutes and have shown that this volume change can be correlated with the level of bubble-bubble coalescence in a multibubble system. This capillary system has been adopted to examine the effect of alcohols on bubble-bubble coalescence under similar experimental conditions to that of the Ncrit measurements and the results are shown in Figure 4. It can be observed that with increasing alcohol concentration the ∆VT decreases. This decrease in ∆VT can be attributed to the inhibition of bubble coalescence of acoustic cavitation bubbles restricting the formation of larger bubbles by coalescence.5 A restriction to the bubble growth by bubble-bubble
Lee et al. coalescence would require additional acoustic pulses for the growth of the bubbles (by rectified diffusion7,13) to reach the active (resonance) size range14 and hence to reach a steadystate MBSL intensity. Further support for the postulate that the reduction in the bubble size is responsible for the rise in Ncrit is the observation that the maximum Ncrit occurs at approximately the same alcohol concentration as that for minimum ∆VT.5 This minimum ∆VT indicates that no further inhibition to bubble coalescence by the adsorbed alcohols can occur. The second observation from Figure 3 is that the concentration at which Ncrit reaches a maximum decreases with an increase in the alkyl chain length of the alcohols. This can be clearly observed in Figure 4. Ncrit reaches a maximum at ∼1.0, ∼0.5, ∼0.1, and ∼0.05 M for methanol, ethanol, butan-1-ol, and pentan-1-ol, respectively. A similar trend in ∆VT with the alkyl chain length of the alcohols can also be observed in Figure 4. The concentration at which a minimum in ∆VT occurs decreases with an increase in the alkyl chain length of the alcohols. It is known that the surface activity of the alcohols increases with an increase in the alkyl chain length of the alcohols. For a given concentration, there would be a greater number of longer chain length alcohol molecules (e.g., pentan-1-ol) adsorbed at the bubble/solution interface compared to shorter chain length alcohol molecules (e.g., methanol). The greater the number of alcohol molecules at the bubble/solution interface, the stronger will be the hindrance to coalescence between the cavitation bubbles. This hindrance to coalescence at a relatively lower concentration for the longer alkyl chain alcohols means that the bubbles would require a greater number of acoustic pulses to grow to the resonance size range, hence an increase in Ncrit and a decrease in ∆VT. To understand the decrease in Ncrit (Figures 3 and 4) observed at the higher range of alcohol concentrations, the effect of the alcohols on the growth of the individual bubbles needs to be considered, since it is highly unlikely that at the higher alcohol concentrations there would be an increase in bubble-bubble coalescence. The observed increase in Ncrit may be caused by an increase in the rate of bubble growth by rectified diffusion. Crum13 and Lee et al.7 have shown that surface active solutes can increase the rate of bubble growth by rectified diffusion, but there are no experimental reports on whether alcohols have the same effect. However, the ∆VT data (Figure 4) show that there is an increase in total bubble volume at higher concentrations of the alcohols. In the absence of bubble-bubble coalescence, the only other possibility for an increase in ∆VT is by rectified diffusion. This scenario has been depicted in Figure 5, which shows how ∆VT is affected by different concentrations of alcohol. For water (∆VTa), larger bubbles are generated by bubble-bubble coalescence leading to a large ∆VT. At low alcohol concentrations (∆VTb), the surface activity of the alcohols inhibits bubble-bubble coalescence. This leads to a bubble population with a smaller size range, which tends to dissolve by Laplace pressure effects once the ultrasound source is switched off, reducing the total bubble volume of the system (∆VTa > ∆VTb). At high alcohol concentrations (∆VTc), an increase in the growth rate by rectified diffusion can grow bubbles resulting in an increase in ∆VT (∆VTa > ∆VTc > ∆VTb). While some of these relatively larger bubbles would be “active” contributing to a decrease in Ncrit, a significant number of them may exceed the “resonance” size during the ultrasound pulse and remain inactive at pressure nodes. The presence of clusters of these relatively larger bubbles at the pressure nodes might be responsible for the observed increase in ∆VT at higher solute concentrations.
Effects on the Growth of Multibubble Sonoluminescence
J. Phys. Chem. B, Vol. 110, No. 34, 2006 17285 in the rate of bubble growth by rectified diffusion. The results shown in this paper are also significant in understanding the effect of alcohols on MBSL intensity. The previous MBSL reports have always interpreted the high-frequency SL results exclusively in terms of quenching by the alcohols. The outcome of the present study indicates that the extent of SL quenching by alcohols might be significantly higher than previously realized if the increase in the active bubble population due to coalescence hindrance is taken into account. Acknowledgment. Judy Lee acknowledges the receipt of “The Albert Shimmins Postgraduate Writing-up Award” granted by the Faculty of Science, University of Melbourne. References and Notes
Figure 5. The effect of alcohol concentrations on the total bubble volume (∆VT). (a) For water, bubble-bubble coalescence can occur, which give rise to ∆VTa. (b) At low concentrations of alcohol, bubblebubble coalescence is inhibited: ∆VTa > ∆VTb. (c) At high concentrations of alcohol, a possible increase in bubble growth by rectified diffusion can result in ∆VTa > ∆VTc > ∆VTb.
Conclusions An examination of the growth in the initial MBSL intensity in the presence of alcohols has revealed additional information that otherwise would not be obtained by steady-state MBSL experiments alone. The increase in the Ncrit with increasing alcohol concentration parallels the decrease in the ∆VT. This indicates that inhibition of bubble-bubble coalescence is the dominating effect. It is proposed that the observed decrease in the Ncrit at high alcohol concentrations is caused by an increase
(1) Young, F. R. CaVitation; Imperial College Press: London, UK, 1999. (2) Henglein, A.; Ulrich, R.; Lilie, J. J. Am. Chem. Soc. 1989, 111, 1974-1979. (3) Henglein, A.; Herburger, D.; Gutierrez, M. J. Phys. Chem. 1992, 96, 1126-1130. (4) Henglein, A.; Gutierrez, M. J. Phys. Chem. 1993, 97, 158-162. (5) Lee, J.; Kentish, S. E.; Ashokkumar, M. J. Phys. Chem. B 2005, 109, 5095-5099. (6) Lee, J.; Kentish, S.; Matula, T. J.; Ashokkumar, M. J. Phys. Chem 2005, 109, 16860-16865. (7) Lee, J.; Kentish, S.; Ashokkumar, M. J. Phys. Chem B 2005, 109, 14595-14598. (8) Lee, J.; Ashokkumar, M.; Kentish, S.; Grieser, F. J. Am. Chem. Soc. 2005, 127, 16810-16811. (9) Ashokkumar, M.; Hall, R.; Mulvaney, P.; Grieser, F. J. Phys. Chem. B 1997, 101, 10845-10850. (10) Tronson, R.; Ashokkumar, M.; Grieser, F. J. Phys. Chem. B 2002, 106, 11064-11068. (11) Sunartio, D.; Ashokkumar, M.; Grieser, F. J. Phys. Chem. B 2005, 109, 20044-20050. (12) Price, G. J.; Ashokkumar, M.; Grieser, F. J. Am. Chem. Soc. 2004, 126, 2755-2762. (13) Crum, L. A. J. Acoust. Soc. Am. 1980, 68, 203-211. (14) Yasui, K. J. Acoust. Soc. Am. 2002, 112, 1405.
