Influence of Degree of Gas Saturation on Sonochemiluminescence

Oct 3, 2013 - Influence of Degree of Gas Saturation on Sonochemiluminescence. Intensity Resulting from Microfluidic Reactions. Toru Tuziuti*. National...
2 downloads 13 Views 1MB Size
Download PDF
Article pubs.acs.org/JPCA

Influence of Degree of Gas Saturation on Sonochemiluminescence Intensity Resulting from Microfluidic Reactions Toru Tuziuti* National Institute of Advanced Industrial Science and Technology (AIST), 2266-98 Shimoshidami, Moriyama-ku, Nagoya 463-8560, Japan ABSTRACT: This work examined the effects of dissolved gas degree of saturation (DOS) on sonochemical reaction yields in both a one-dimensional (1D) microspace and a threedimensional (3D) millimeter-sized space. The extent of each reaction was monitored by measuring sonochemiluminescence (SCL) intensity at 213 kHz. The results demonstrated that, at relatively low levels of power density, selecting a solution DOS in the supersaturation range at atmospheric pressure resulted in higher yields per unit volume in the 1D space compared to that obtained from the 3D space. This effect is attributed to a decrease in the cavitation threshold of the 1D reaction system since, at low power density, the 1D space represents a more homogeneous reaction volume. Comparing the highest SCL intensity levels obtained from the 3D and 1D reactions shows that enhancing the reaction yield in the 1D space requires higher DOS values than are required to generate elevated yields in the 3D space. The 3D space contains a greater concentration of bubbles than the 1D space, but many of these are ineffective at promoting the reaction. Thus, reactions in the 3D environment require not only the application of higher power density levels but also a lower DOS, so as to allow the bubbles to undergo the violent pulsations necessary to facilitate the sonochemical reaction.

1. INTRODUCTION The use of microfluidic devices with dimensions of tens to hundreds of micrometers to perform chemical reactions is a promising green chemistry technique since it reduces both the quantity of raw materials required and the resultant waste.1 Because of their small dimensions, microfluidic devices also feature short diffusion paths and high specific surface areas that increase mass and heat transfer, leading to acceleration of multiphase reactions.2 The application of ultrasound to chemical reaction processes in microfluidic devices has been increasingly studied.2−10 The chemical effects of ultrasound11,12 are based on the formation of radicals and oxidants, such as hydroxyl radicals, ozone, and hydrogen peroxide,13 by dissociation of water vapor subjected to extreme conditions. These conditions involve temperatures of several thousand degrees Kelvin and pressures of several hundred atmospheres within a nonlinearly pulsating cavitation bubble.14 The chemical reactions involving species generated by ultrasound are referred to as sonochemical reactions,11,12 and it has been found that both the mass transfer and mixing induced by streaming around bubbles undergoing pulsating motions15−17 assist in accelerating multiphase reactions. Iida et al. demonstrated a difference in radical formation rates between one- and two-dimensional (1D and 2D) spaces with a characteristic length of 200 μm and a three-dimensional (3D) space of a few tens of millimeters.3 Rivas et al. generated microbubbles from micropits on a micromachined silicon substrate and found a correlation between radical formation and the number of pits where the bubble structures form, based on images obtained by luminol sonochemiluminescence (SCL).4 Ohl et al. showed that a microfluidic system operating at the optimal acoustic amplitude generates small bubbles in water near the air−water interface, which subsequently become © 2013 American Chemical Society

unstable and oscillate volumetrically while moving away from the interface.18 Bubbles such as these can possibly contribute to sonochemical reactions. During reactions in a microfluidic system, the reaction solution is delivered into the system via a syringe pump or similar mechanism and so it is reasonable to assume that the static pressure in the solution is above ambient pressure. If the solution delivered into a microfluidic system at above ambient pressure contains a dissolved gas at a high degree of saturation (DOS) could conceivably decrease the cavitation threshold and thus increase the yield of the sonochemical reaction. The author and co-workers have in fact demonstrated that sonochemical yield increases through a maximum and then decreases as the solution DOS is increased through the range in which cavitation bubbles occur under ambient atmospheric conditions.19 In a microfluidic system, therefore, applying appropriate conditions of DOS and acoustic amplitude should enable us to obtain reaction yields superior to those obtained without adjusting these parameters. To the best of the author’s knowledge, however, there have been no reports concerning investigations of sonochemical yields under optimized conditions of DOS and acoustic power in a microfluidic system. The differences in yields obtained from a 1D microreactor and a 3D millimeter-sized system at various DOS values has also not been examined to date. In the present study, the author primarily investigated the power dependence of the intensity of luminol-based SCL in both a 1D microspace and a 3D millimeter-sized space, with the aim of determining the optimal DOS values associated with Received: July 16, 2013 Revised: September 12, 2013 Published: October 3, 2013 10598

dx.doi.org/10.1021/jp407068n | J. Phys. Chem. A 2013, 117, 10598−10603

The Journal of Physical Chemistry A

Article

Figure 1. Experimental apparatus.

to be proportional to the dissolved air DOS, with a DOS value of 1 equiv to complete saturation. Solutions at five different DOS values were prepared just prior to loading the syringe (DOS = 1.001, 1.048, 1.097, 1.150, and 1.200 at atmospheric pressure). The DOS values of these same solutions when pressurized to 2.02 atm (explained below) in the syringe were estimated to be 0.500, 0.524, 0.549, 0.575, and 0.599. The 1D space had a 100 μm × 100 μm cross-section and was contained in a glass plate (25.4 × 50.8 mm, 1 mm thickness). The SCL intensity of emissions from the 1D space was monitored along a 22 mm section of the longer side of the plate. The 3D space consisted of two tubular vessels, each 30 mm in internal diameter and 8.8 mm in internal height. The volumes of the 1D and 3D spaces were 0.22 and 124.4 mm3, respectively. One of the 3D vessels connected to the syringe through a transfer tube played an adaptor to 1D space. The second 3D vessel was used to allow the solution to flow out of the 1D space. The static pressure of the solution in the pressurized syringe was calculated based on the displacement of two identical springs (TRUSCO TCS 1000-120-120) inserted in-line between the cylinder and the syringe pump. Each spring had a spring constant of 6.963 N/mm and the force applied by the cylinder to the solution was calculated to obtain 33.1 N. Calculations based on the continuity equation and Bernoulli’s theory of fluid mechanics22 determined that the static pressure experienced by the solution was 2.02 atm in both the 1D and 3D spaces. Note that the frictional force between the cylinder and the syringe was taken into consideration and the speed of cylinder movement (0.64 mm/s) was relatively low. The cylinder speed applied in this work resulted in a solution flow rate through the microfluidic device of 0.489 mL/min. Both the experimental solution and a 50 mm diameter PZT transducer were temperature controlled at 19.5 ± 0.5 °C using a circulator

high yields during sonochemical reactions. On the basis of the results, a mechanism explaining the varying relationships of SCL intensity with DOS values in a 1D microspace and a 3D millimeter-sized space is proposed. Luminol-based SCL was employed in this study based on work by Henglein et al., which showed that there is a correlation between iodine yield and luminescent intensity under sonication.20 The intensity of SCL resulting from luminol was therefore used as a means of quantifying the sonochemical reaction yield under various conditions in the present study.

2. EXPERIMENTAL DETAILS Figure 1 shows the experimental apparatus used to measure the intensity of SCL resulting from luminol (3-aminophthalhydrazide). Luminol reacts with OH radicals generated in cavitation bubbles during intense ultrasound irradiation to yield aminophthalate anions, which subsequently undergo the emission of blue fluorescence.21 A solution consisting of 10.0 mM NaOH (Wako) and 56.4 μM luminol (Wako) was prepared in distilled water, and the concentration of dissolved air in this solution was adjusted by bubbling air through the solution as well as by temperature control. Luminol solutions having different dissolved oxygen degree of saturation (DOS) values were loaded into a syringe (NORM-JECT 20 mL) and pressurized by the motion of its piston (20.2 mm in diameter) under the impetus of a syringe pump (YMC YSP-101). The motion of the syringe piston delivered the solution to a microfluidic device (Translume SSC-100-100-L-1) incorporating both 1D and 3D spaces. The DOS of each test solution was measured with a dissolved-oxygen sensor (YSI ProODO). It should be noted that the dissolved oxygen DOS was assumed 10599

dx.doi.org/10.1021/jp407068n | J. Phys. Chem. A 2013, 117, 10598−10603

The Journal of Physical Chemistry A

Article

(AS ONE CH-202) to maintain a flow of cooling water through a copper coil surrounding the syringe as well as an aluminum plate (23.4 × 54.5 × 10.0 mm) to which the transducer and the microfluidic device had been affixed on opposite sides with epoxy resin. A gap was purposely maintained between the aluminum plate and the regions of the microfluidic device directly beneath each of the two 3D vessels. The ultrasonic transducer was powered to propagate surface waves through the plate and thus emit a sound wave into the adjacent fluid in the microfluidic system.18,23,24 A CW sinusoidal signal from a function generator (NF 1942) was amplified using a 55 dB power amplifier (ENI 1140LA) and the input power was measured using a power meter (Towa TAW60A). The combined sonochemiluminescent light emissions from the solutions in the 1D and both 3D spaces was detected with a photomultiplier tube (Hamamatsu R928) and, in addition, light emitted solely from the 1D space was measured by temporarily shielding the 3D space emission with aluminum foil. The output voltage from the photomultiplier was received by a multimeter (ADVANTEST TR6847), and the resulting data were acquired by computer (NEC PC9821Xc16). Each SCL intensity measurement was repeated three times. The SCL intensity component resulting from the 3D space was estimated by subtracting the measured emission from the 1D space from the combined emission detected from the 1D and both 3D spaces. Since the 3D vessel was not transparent, its light attenuation was evaluated using a spectrometer (Hamamatsu C10082CAH) and a light source (LED LENSER V2), from which it was determined that light was attenuated by the vessel by a factor of 9.15. Accordingly, the SCL emission intensity calculated for the 3D system was multiplied by 9.15 to allow the data to be directly compared to the emissions from the 1D system.

Figure 3. Power density dependence of the SCL intensity at different DOS values, showing (a) combined emission from 1D and 3D spaces and (b) emission solely from the 1D space. The vertical bars represent average relative errors.

3. RESULTS AND DISCUSSION 3.1. Frequency Dependence of SCL Intensity. Figure 2 shows the frequency dependence of SCL intensity, normalized

once a threshold value of power density is exceeded, after which it continues to increase as the power density increases. Interestingly, the relationship between power-density and SCL intensity is different between solutions with different DOS values. Note that each DOS value is the initial one measured just before sonication and that degassing during sonication has a possibility to change its value. A detail investigation on the influence of the change in DOS value with such degassing on the yield of sonochemical reaction in a microspace is left to a future study. 3.3. Detailed Power-Density Dependence of SCL Intensity. In order to investigate in detail the SCL intensities obtained at varying values of DOS, the SCL intensity emitted by the 3D space alone at each DOS value was estimated from the combined 1D and 3D emission data minus the 1D emission. Figure 4 shows an example of the estimated SCL intensities obtained from the 3D space alone, along with the combined and 1D emissions at a DOS value of 0.500. The inset of the figure presents a magnified plot of the SCL intensity from the 1D space over the range of power density where the 1D SCL intensity begins to increase and shows that the powerdensity threshold associated with the onset of 1D emission is significantly above that for 3D emission. This is not unexpected since Iida et al.3 encountered a similar phenomenon whereby the power-density threshold increased when transitioning from 1D to 3D space. Rivas et al.1 have also noted that high power density is required to induce cavitation in confined spaces since

Figure 2. Frequency dependence of the SCL intensity (values normalized by maximum intensity).

to the maximum intensity value. These data demonstrate that the maximum SCL peak intensity is obtained at 213 kHz. As a result, the driving frequency employed during sonication in subsequent trials was fixed at this value. 3.2. General Power-Density Dependence of SCL Intensity. Figure 3 shows the power-density dependence of SCL intensity measured at different DOS values for (a) the combined 1D and 3D spaces and (b) the 1D space alone. In each case, it is evident that the SCL intensity exhibits an initial stage over which emission is essentially zero, then increases 10600

dx.doi.org/10.1021/jp407068n | J. Phys. Chem. A 2013, 117, 10598−10603

The Journal of Physical Chemistry A

Article

Figure 4. Variations in SCL intensity with power density at a DOS value of 0.500. The inset presents a magnified view of data from the 1D space. The vertical bars represent average relative errors.

bubble dynamics and hence the extent of collapse are constrained by the microspace. 3.4. Threshold Power Density for SCL Intensity. Figure 5 plots the relationship between the DOS value and the SCL

Figure 5. Variations in the SCL threshold power density with DOS.

intensity threshold power density, incorporating both 1D and 3D data. This plot confirms that the 1D threshold is higher than that of the 3D space within the range of DOS values applied in these trials. It is worth noting that, at higher DOS values (approaching the maximum of 0.599), the difference in the thresholds of the 1D and 3D spaces becomes relatively small. The use of a high DOS solution apparently makes it possible to create chemical effects through the action of cavitation bubbles in confined spaces at power density values, which are significantly below those required at relatively low DOS values. At a DOS value of 0.549, the threshold power density of the 1D space exhibits a maximum, while, at the same DOS value, a minimum appears in the 3D data. This likely occurs because the solution experiences a certain amount of degassing during effective sonication as the result of the lowered threshold in the 3D space on the side of the syringe while being transported to the 1D space, leading to an increase in the 1D power density threshold. 3.5. DOS Dependence of SCL Intensity. Figure 6a plots the DOS dependence of the 3D SCL intensity at different levels of power density. At a relatively low power density (0.87 W/ cm2), the SCL intensity exceeds its value at DOS = 0.500 only at DOS = 0.524. At power densities above 1.40 W/cm2, however, the SCL intensity is greater than the value at DOS =

Figure 6. DOS dependence of SCL intensity at different power density levels, obtained from (a) the 3D space only, (b) the 1D space only, and (c) a comparison between 1D and 3D spaces (normalized by the intensity at DOS = 0.500).

0.500 at all levels of DOS above 0.500. The range of DOS values at which the SCL intensity is enhanced is thus enlarged and the maximum emission intensity recorded at DOS = 0.549 is increased remarkably as the power density is increased. The appearance of maxima peaks at intermediate DOS values is also seen in this figure. This occurs because overly high DOS levels do not as efficiently generate increasing quantities of the pulsating bubbles, which promote the sonochemical reaction, since the presence of a large number of bubbles decreases the acoustic amplitude in the solution. 10601

dx.doi.org/10.1021/jp407068n | J. Phys. Chem. A 2013, 117, 10598−10603

The Journal of Physical Chemistry A

Article

reaction than the 3D space. In this context, an effective volume means that there is only a small region in which the sonochemical reaction does not occur. Figure 7 shows that, in all cases except at a DOS value of 0.549 in conjunction with relatively low power density, the normalized SCL ratio is greater than the volume ratio. It is interesting to observe that the highest DOS value of 0.599, when applied with a reduced level of power density of approximately 1 W/cm2, results in the highest SCL intensity ratio. This ratio is about 15 times greater than the volume ratio indicated on the corresponding fitted curve. The trend observed for the normalized ratio at the highest DOS value of 0.599 occurs because the increased quantity of bubbles in the 3D space lowers the acoustic amplitude in the solution. This in turn tends to prevent the violent collapse of bubbles required to promote SCL at relatively low levels of power density. In contrast, the 1D space contains fewer bubbles throughout its volume, but most of these are capable of actively facilitating the sonochemical reaction. For this reason, the normalized ratio at the highest DOS value exhibits its maximum at a relatively low power density. Within a limited range of power density, the number of active bubbles in the 3D space increases with increasing power to a greater extent than in the 1D space since there are more bubbles in the 3D space, which have the potential to be active. Accordingly, the normalized ratio decreases with increasing power density. The above discussion suggests that, for the reactive to proceed more effectively in the 3D space than in the 1D space, both a high level of power density and an optimal DOS are required. From Figure 7, it can be seen that the normalized ratio is below 1 when applying a DOS of 0.549 and a power density in the range of 1.8−1.9 W/cm2, which is higher than the power density of approximately 1 W/cm2 and is effective for the 1D space. Iida et al. determined the production rate of fluorescent hydroxyterephthalate (HTPA) ions from the sonochemical reaction of OH radicals with terephthalate anions in both 1D and 3D spaces.3 The rates that they found were 1.5 × 1010 HTPA molecules/s in a 2.4 mm3 1D volume and 1.0 × 1013 HTPA molecules/s in a 4000 mm3 3D volume, which corresponds to a volume-normalized rate ratio of 2.5. This value is somewhat lower than the data obtained from the present work, as plotted in Figure 7, if one compares it to the present normalized ratio obtained at a DOS of 0.500. One believes the present results differ because the 1D volume in the experimental apparatus used by Iida is larger than that of the apparatus used in the present study and thus may have a larger ineffective volume.

Figure 6b plots the DOS dependence of the 1D SCL intensity at different power density levels. These data show that the SCL intensity initially decreases, moves through a minimum, and then increases at each power density. In contrast to the 3D data, the 1D SCL intensity at DOS = 0.500 is greater than the intensities recorded at DOS = 0.524 and 0.549. The reaction intensities at the latter two DOS values might have been influenced by changes to the solution provided from the 3D space. The 3D space at DOS = 0.500 contains relatively few bubbles and undergoes minimal degassing before it is transferred to the 1D space, which could account for the higher SCL intensity observed at DOS = 0.500 for the 1D reaction. In this same figure, the highest SCL intensity values are found at the highest value of DOS, which is 0.599. Comparing the highest SCL intensities of the 3D and 1D spaces, it is noteworthy that enhancement of the SCL intensity in the case of the 1D space required the application of a higher DOS. The plot in Figure 6c shows the DOS dependence of the normalized SCL intensity at different power density values for both the 3D and 1D spaces. In this graph, the intensity of SCL emission at each data point has been normalized to that at DOS = 0.500. These results confirm that the SCL intensity resulting from reaction in the 3D space is remarkably high at intermediate DOS levels near 0.549, while emission from the 1D reaction peaks at high DOS levels in the vicinity of 0.599. 3.6. Difference in Effective Volume between 1D and 3D Spaces. Figure 7 shows the power dependence of the 1D/

Figure 7. Variations in the 1D/3D SCL intensity ratio with power density, normalized by the 1D/3D volume ratio. Dotted lines extending from the horizontal axis indicate the power density threshold at each DOS. The solid line parallel to the horizontal axis is the 1D/3D volume ratio, defined as 1. Continuous, broken, dotted, dot and dash, and two-dot fitted lines indicate normalized SCL intensity ratios at DOS values of 0.500, 0.524, 0.549, 0.575, and 0.599, respectively.

4. CONCLUSIONS This study demonstrated that, at relatively low levels of power density, tuning the reaction solution DOS within the range of supersaturation at atmospheric pressure allows higher yields per unit volume from a sonochemical reaction in a 1D space compared to that obtained from a 3D space. This effect results from a decrease in the cavitation threshold of the 1D space such that, at low levels of power density, the 1D space provides a more homogeneous reaction volume than the 3D. A comparison of the maximum SCL intensities obtained from the 3D and 1D spaces showed that increasing the yield in the 1D space requires higher DOS values than those required to enhance the reaction in the 3D space. The 3D space contains a greater number of bubbles than the 1D space, many of which

3D SCL intensity ratio, normalized by the volume ratio, at various values of DOS. In this figure, the dotted lines extending from the horizontal axis indicate the power density threshold associated with each level of DOS, while the flat line parallel to the horizontal axis is the 1D/3D volume ratio, defined as 1. The fitted curves indicated by continuous, broken, dotted, dot and dash, and two-dot patterns correspond to the normalized SCL intensity ratios at DOS = 0.500, 0.524, 0.549, 0.575, and 0.599, respectively. A normalized ratio above 1 indicates that the 1D space provides a more effective volume for the sonochemical 10602

dx.doi.org/10.1021/jp407068n | J. Phys. Chem. A 2013, 117, 10598−10603

The Journal of Physical Chemistry A

Article

(18) Ohl, T. S.-W.; Ow, D. S. W.; Klaseboer, E.; Wong, V. V. T.; Camattari, A.; Ohl, C.-D. Creation of Cavitation Activity in a Microfluidic Device through Acoustically Driven Capillary Waves. Lab Chip 2010, 10, 1848−1855. (19) Tuziuti, T.; Yasui, K.; Iida, Y.; Sivakumar, M.; Koda, S. LaserLight Scattering from a Multibubble System for Sonochemistry. J. Phys. Chem. A 2004, 108, 9011−9013. (20) Henglein, A.; Ulrich, R.; Lilie, J. Luminescence and Chemical Action by Pulsed Ultrasound. J. Am. Chem. Soc. 1989, 111, 1974− 1979. (21) Pétrier, C.; Lamy, M.-F.; Francony, A.; Benahcene, A.; David, B.; Renaudin, V.; Gondrexon, N. Sonochemical Degradation of Phenol in Dilute Aqueous Solutions: Comparison of the Reaction Rates at 20 and 487 kHz. J. Phys. Chem. A 1994, 98, 10514−10520. (22) Krause, E. Fluid Mechanics I. In Fluid Mechanics; Springer: Berlin, Germany, 2005. (23) Haake, A.; Dual, J. Positioning of Small Particles by an Ultrasound Field Excited by Surface Waves. Ultrasonics 2004, 42, 75− 80. (24) Tuziuti, T.; Masuda, Y.; Yasui, K.; Kato, K. Two-Dimensional Patterning of Inorganic Particles in Resin Using Ultrasound-Induced Plate Vibration. Jpn. J. Appl. Phys. 2011, 50, 088006.

are ineffective, and requires the application of both high power density and a suitable DOS (lower than the optimal 1D DOS) to induce the violent pulsations of bubbles necessary to promote the sonochemical reaction.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The author would like to thank Dr. Kyuichi Yasui for helpful discussions. This work was supported by a Grant-in-Aid for Scientific Research (23560917) from the Japan Society for the Promotion of Science.

■

REFERENCES

(1) Rivas, D. F.; Cintas, P.; Gardeniers, H. J. G. E. Merging Microfluidics and Sonochemistry: Towards Greener and More Efficient Micro-Sono-Reactors. Chem. Commun. 2012, 48, 10935− 10947. (2) Hübner, S.; Kressirer, S.; Kralisch, D.; Bludszuweit-Philipp, C.; Lukow, K.; Jänich, I.; Schilling, A.; Hieronymus, H.; Liebner, C.; Jähnisch, K. Ultrasound and Microstructures: A Promising Combination? ChemSusChem 2012, 5, 279−288. (3) Iida, Y.; Yasui, K.; Tuziuti, T.; Sivakumar, M.; Endo, Y. Ultrasonic Cavitation in Microspace. Chem. Commun. 2004, 2280−2281. (4) Rivas, D. F.; Prosperetti, A.; Zijlstra, A. G.; Lohse, D.; Gardeniers, H. J. G. E. Efficient Sonochemistry through Microbubbles Generated with Micromachined Surfaces. Angew. Chem., Int. Ed. 2010, 49, 9699− 9701. (5) Rivas, D. F.; Ashokkumar, M.; Leong, T.; Yasui, K.; Tuziuti, T.; Kentish, S.; Lohse, D.; Gardeniers, H. J. G. E. Sonoluminescence and Sonochemiluminescence from a Microreactor. Ultrason. Sonochem. 2012, 19, 1252−1259. (6) Ohl, T. S.-W.; Ow, D. S. W.; Klaseboer, E.; Wong, V. V.; Dunke, R.; Ohl, C.-D. Sonochemistry and Sonoluminescence in Microfluidics. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 5996−5998. (7) Ahmed-Omer, B.; Barrow, D.; Wirth, T. Effect of Segmented Fluid Flow, Sonication and Phase Transfer Catalysis on Biphasic Reactions in Capillary Microreactors. Chem. Eng. J. 2008, 135S, S208− S283. (8) Aljbour, S.; Yamada, H.; Tagawa, T. Ultrasound-Assisted Phase Transfer Catalysis in a Capillary Microreactor. Chem. Eng. Process. 2009, 48, 1167−1172. (9) Aljbour, S.; Tagawa, T.; Yamada, H. Ultrasound-Assisted Capillary Microreactor for Aqueous-Organic Systems-Hydrodynamics Study. J. Chem. Eng. Jpn. 2009, 42, 400−406. (10) Aljbour, S.; Tagawa, T.; Yamada, H. Ultrasound-Assisted Capillary Microreactor for Aqueous-Organic Multiphase Reactions. J. Ind. Eng. Chem. 2009, 15, 829−834. (11) Mason, T. J. “1. An Introduction to the Uses of Power Ultrasound in Chemistry. In Sonochemistry; Oxford University Press: New York, 1999. (12) Suslick, K. S. Sonochemistry. Science 1990, 247, 1439−1445. (13) Yasui, K.; Tuziuti, T.; Sivakumar, M.; Iida, Y. Theoretical Study of Single-Bubble Sonochemistry. J. Chem. Phys. 2005, 122, 224706. (14) Matula, T. J. Inertial Cavitation and Single-Bubble Sonoluminescence. Philos. Trans. R. Soc., A 1999, 357, 225−249. (15) Liu, R.; Yang, J.; Pindera, M. Z.; Athavale, M.; Grodzinski, P. Bubble-Induced Acoustic Micromixing. Lab Chip 2002, 2, 151−157. (16) Marmottant, P.; Hilgenfeldt, S. Controlled Vesicle Deformation and Lysis by Single Oscillating Bubbles. Nature 2003, 423, 153−156. (17) Iida, Y.; Tuziuti, T.; Yasui, K.; Towata, A.; Kozuka, T. Bubble Motions Confined in a Microspace Observed with Stroboscopic Technique. Ultrason. Sonochem. 2007, 14, 621−626. 10603

dx.doi.org/10.1021/jp407068n | J. Phys. Chem. A 2013, 117, 10598−10603