Article pubs.acs.org/crystal
Cite This: Cryst. Growth Des. XXXX, XXX, XXX−XXX
Quantification of Cavitation Activity by Sonoluminescence To Study the Sonocrystallization Process under Different Ultrasound Parameters Judy Lee,*,†,∥ Kyuichi Yasui,‡ Muthupandian Ashokkumar,§ and Sandra E. Kentish† †
Department of Chemical Engineering, University of Melbourne, Melbourne, Victoria 3010, Australia National Institute of Advanced Industrial Science and Technology (AIST), 2266-98 Anagahora, Shimoshidami, Moriyama-ku, Nagoya 463-8560, Japan § School of Chemistry, University of Melbourne, Melbourne, Victoria 3010, Australia ∥ Chemical and Process Engineering, University of Surrey, Guildford, Surrey GU2 7XH, United Kingdom
Downloaded via KAOHSIUNG MEDICAL UNIV on July 25, 2018 at 19:17:51 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
‡
ABSTRACT: In this study, both the antisolvent sonocrystallization process of sodium chloride and cavitation activity were investigated as a function of frequency (22−1080 kHz) and acoustic calorimetric power (0−30 W). For frequencies between 20 and 139 kHz, the size of the sodium chloride crystals decreased sharply with increasing power. For frequencies 647 and 1080 kHz, a certain power threshold needs to be exceeded before a decrease in the crystal size was observed. This power threshold coincided with the power threshold for sonoluminescence emission from cavitation bubbles. It was found that the onset of cavitation bubble activity, irrespective of the magnitude (measured in terms of sonoluminescence), enhanced the crystal nucleation rate and decreased crystal size. The minimum crystal size obtained was found to decrease with increasing maximum total integrated sonoluminescence intensity. The results suggest sonoluminescence could be used as a measure to evaluate the sonocrystallization process and that a greater collapse intensity would yield the smallest crystals. In addition, photographs of the sonocrystallization process are reported, suggesting a link between nonsymmetrical transient cavitation activity and crystal nucleation.
1. INTRODUCTION Acoustic cavitation refers to the ultrasound induced growth and collapse of a bubble in a liquid. When the applied acoustic pressure surpasses a certain threshold, the existing gas nuclei are forced to oscillate and grow in size.1 Once the bubble reaches a certain critical size range (active bubbles), they begin to oscillate nonlinearly and undergo violent transient collapse, generating extreme temperatures and pressures within the bubble core.2 It is these intense conditions that have allowed cavitation bubbles to facilitate a number of applications,3−6 including crystallization. The application of ultrasound in crystallization processes is known as sonocrystallization, which has been shown to exhibit positive effects on crystal nucleation rate, metastable zone width, crystal size distribution, and crystal growth.7−14 Reports have shown that ultrasound can also selectively crystallize specific polymorphs that are otherwise unstable,15−18 as well as reduce striation in crystals.19,20 Although various explanations have been proposed,10,13,14,21−26 the exact mechanism behind sonocrystallization is difficult to resolve. This is largely due to the chaotic and dynamic nature of the system where complex bulk and localized effects produced by cavitation, such as temperature, pressure, fluid shear, and oscillating bubble surfaces, could have significant influence on crystal nucleation. © XXXX American Chemical Society
This influence could also vary depending on the type of crystallization such as antisolvent, cooling, melts, and inorganic and organic systems. Nevertheless, practically all theories reach the same consensus that cavitation bubbles are the direct cause of sonocrystallization effects.22−24,27−29 To the authors’ knowledge, only one report30 has claimed the onset of fat crystallization by 1.5 MHz ultrasound at a power level where no transient cavitation signals were detected by the hydrophone and the calculated mechanical index was below the transient cavitation threshold. The intensity of cavitation activity can be quantified by varying both acoustic power and frequency. It has been shown that cavitation activity, which can be quantified by either sonoluminescence or sonochemical activity, increases with increasing frequency and power but decreases at high frequencies and excessive powers.31−35 If sonocrystallization is related to cavitation activity, then one would expect to observe a similar dependence of sonocrystallization upon frequency and power, but there appears to be conflicting reports in the current literature. Kordylla et al.9 examined the Received: April 12, 2018 Revised: June 26, 2018
A
DOI: 10.1021/acs.cgd.8b00547 Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
Article
sodium chloride. Cavitation activity, measured using SL, will be used as a measure to probe the influence of ultrasound on sonocrystallization.
effect of 355.5 and 1046 kHz on the crystallization of dodecandioic acid in different solvents. They found the metastable zone width decreased with increasing power, with 355.5 kHz giving the narrowest metastable zone width. Wohlgemuth et al.36 studied the crystallization of adipic acid and found 355.5 kHz to give only a slightly better metastable zone width reduction and smaller crystal size distribution compared to 204 and 610 kHz. Nii et al.12 have reported, for a sonication time up to 20 min, the enhancement in the crystal growth of glycine with 1.6 MHz ultrasound and found the growth rate (and crystal size) to be proportional to power, whereas sonication with a 20 kHz plate transducer led to a slight decrease in the crystal size. This is in contrast to the results reported by Louhi-Kultanen et al.37 where 20 kHz horn irradiation between 30 to 90 min resulted in a significant reduction to the glycine crystal size. It is important to note that Nii et al.12 used antisolvent crystallization, whereas LouhiKultanen et al.37 used a cooling crystallization process to induce supersaturation. From these investigations, the reported frequency effects appear to vary significantly from study to study. This might be attributed to the limited frequency range investigated and also difficulties in comparing results produced by different crystallization conditions, sonication systems/ configurations, sonication time, and powers. Lee et al.11 reported a systematic study comparing sodium chloride crystals produced by an ultrasonic horn to that produced by plate transducers with frequencies ranging from 22 to 647 kHz at a fixed calorimetric power of 20 W. It was found that high frequency ultrasound of up to 647 kHz was capable of inducing crystallization and reducing crystal size comparable to that obtained by an ultrasonic horn at a lower frequency,11 but no significant frequency effect was observed, whereas a report by Jordens et al.,38 using a much lower power of 8 W, revealed an optimum frequency of 41 kHz for maximum reduction in the metastable zone width of paracetamol. At frequencies higher than 41 kHz, the reduction in the metastable zone width was much lower, and a significant degradation of the paracetamol was observed. The degree of degradation correlated with the amount of radical production at elevated frequencies. Although a broad range of frequencies were studied in these two reports, only one power level was used and different crystallization systems were investigated. The importance of power has been demonstrated by Kurotani et al.39 where it was reported that at low acoustic powers primary nucleation rates were inhibited and induction time increased, whereas at high acoustic powers the nucleation rates increased. It is often difficult to compare the effect of frequency on the crystallization process. A measure of cavitation activity would be a better comparison between frequencies than the acoustic power. Rossi et al. 29 has correlated photographs of sonochemiluminescence (SCL), which is the emission of light from the reaction of luminol with OH radicals produced from transient cavitation, with the onset of sonocrystallization inside a microfluidic channel. In that report, the SCL was varied by adjusting the distance of the microfluidic channel away from the surface of a 20 kHz sonotrode operated at a fixed power. To the authors’ knowledge, such a quantitative comparative study between sonocrystallization and sonoluminescence (SL), which is purely from transient cavitation bubbles, for different frequencies and powers has not been reported. Therefore, this study aims to provide a comprehensive and systematic investigation on the influence of both frequency and power on the antisolvent crystallization of
2. MATERIALS AND METHODS 2.1. Materials. Sodium chloride (NaCl, 99.5%) was purchased from Merck, and 100% absolute ethanol was purchased from ChemSupply. These were used as received without any further purification. All solutions were prepared using purified water with resistivity greater than 18.2 MΩ cm. Antisolvent crystallization experiments were performed in a cylindrical Pyrex cell with an approximate internal diameter of 6.2 cm and height of 11.2 cm. For experiments using ultrasound plate transducers, the transducer was fixed at the bottom of the Pyrex cell. A Langevin-type multifrequency transducer of diameter 4.5 cm (resonance frequencies: 22, 44, 98, and 139 kHz, Honda Electric) and a piezo-electric ceramic plate transducer of 5.5 cm (647 and 1080 kHz, L-3 Communications ELAC Nautik GmbH) were used. The transducers were powered by a T&C Power Conversion, Inc. Amplifier (AG series). For experiments with an ultrasonic horn, a 20 kHz horn (Branson Sonifier 450) with a tip diameter of 12 mm diameter was placed at the top of the Pyrex cell, with the tip positioned 1 cm below the liquid surface. The actual power dissipated into the solution was determined using a standard calorimetric technique.40 2.2. Crystallization Setup and Procedures. The crystallization setup and procedures are illustrated in Figure 1. At first, 150 g of
Figure 1. Schematic of the crystallization procedure employed in this investigation.
100% ethanol (17.1 M) was weighed and transferred into the Pyrex cell. This was followed by the addition of 15 g of 200 g/L (3.4 M) NaCl aqueous solution to create a supersaturation ratio of 6.9. The addition of the NaCl was performed by simply pouring the solution into the cell containing the antisolvent (addition time was less than 2 s). This procedure was performed independently at least three times to ensure reproducibility. The solution was allowed to crystallize for 90 s before it was transferred into a particle size analyzer for particle size distribution analysis. Turbidity measurements confirmed that the system reached equilibrium within 30 s where the system had completely desupersaturated and an equilibrium saturation concentration had been reached. For all sonication experiments, the ultrasound was switched on just before the addition of NaCl, and continuous sonication was applied for 90 s. 2.3. Induction Time. For induction measurements an overhead stirrer operating at 100 rpm was used to ensure that the solution was well mixed and that NaCl crystals remained in suspension. All experiments were performed at an initial temperature of 20 ± 0.5 °C. The turbidity of the solution was recorded per second using a fiberoptic turbidity probe (Crystal eyes, H.E.L), and the induction time was determined by the time lapse between the addition of the NaCl stock solution and the onset of turbidity. The turbidity can be affected by the initial degassing of ethanol with the addition of concentrated salt solution and cavitation bubbles from ultrasound. Therefore, the onset of turbidity was determined by extrapolating the slope of the initial linear rise in the turbidity back to the initial turbidity level of the ethanol solution prior to the addition of salt solution. B
DOI: 10.1021/acs.cgd.8b00547 Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
Article
Figure 2. Volume mean particle size for various ultrasound frequencies. For all frequencies, the applied calorimetric power was fixed at 20 W. 2.4. Particle Size Distribution. The particle size distribution was determined using a Malvern Mastersizer 2000 Laser diffraction system (Malvern Instruments Ltd., Malvern, UK). A refractive index of 1.52 for NaCl crystals and 1.36 for ethanol was used. Particle size distribution was taken as an average of three samples obtained from three independent experiments and the volume mean particle diameter (D4,3) was obtained. 2.5. Sonoluminescence (SL) Experiments. In another sequence of experiments, the SL intensity at each power and frequency was determined. Ethanol is known to quench SL;41 therefore, water was used as the sonication medium in these experiments. Cavitation activity has been shown to vary with liquid height;42 therefore, to ensure that the liquid height remained the same as that used in the crystallization experiments, an equivalent volume of 200 mL of water was used. The intensity of the SL was measured using a Hamamatsu end-on photomultiplier tube (PMT; model: R647-04), connected to a Canberra high voltage supply (model: 2003) set at 500 V, positioned at a fixed height of 4.5 cm above the liquid surface. Both the cell and the PMT were housed in a light-proof enclosure to avoid interference from external light. The output signal from the PMT was displayed on a digital oscilloscope (Tektronix, model No. TDS 320). The SL intensity was averaged over 256 acquisitions to increase signal-tonoise ratio and recorded once the SL intensity reached a steady state, which was less than a minute. For the 1080 kHz transducer, the SL intensity for acoustic powers less than 35 W fell below the detection limit of the system. In this case, the high voltage supply to the PMT was increased to 1000 V, and the detection point was brought closer, to 2 cm above the liquid surface.43 The amplification at this new setting was determined by repeating experiments above 35 W to obtain SL intensity at both settings. The SL intensity for 1080 kHz at low powers was then corrected by this amplification factor to allow a direct comparison to other frequencies.
size, only 61% of that obtained in the absence of ultrasound. This suggests that at high frequencies the influence of the cavitation activity on the crystallization process is weaker compared to lower frequencies. Despite the very small changes in the volume mean particle size observed, the size distribution profiles (Figure 3) reveal a noticeable trend as ultrasound frequency is increased. The crystal size distribution shifts to smaller sizes as the ultrasound frequency is increased from 22 to 44 kHz (Figure 3a). Indeed, the crystal size distribution for 44 kHz has a very similar particle size distribution as that obtained by a 20 kHz horn operating at the same calorimetric power (20 W). Although the calorimetric power is the same, the ultrasonic horn is in fact emitting at a higher acoustic intensity of 9.8 W/cm2 compared to 2.8 W/cm2 for the plate transducers. This is due to the plate transducers having a larger emitting surface area. The result differs from a previously published report for filtered crystals, which exhibited no significant change in the particle size distribution profile for different frequencies up to 647 kHz,11 whereas in this study increasing the frequency from 98 to 1080 kHz revealed a reduction in the effect of ultrasound with the crystal size distribution shifting toward the crystal distribution obtained under no ultrasound. As indicated in ref 11, the filtering process resulted in the aggregation of crystals, making it difficult to redisperse them in ethanol for particle size analysis. In this study, the sample sizes were analyzed immediately without filtering and drying, thereby allowing small differences in the particle size to be detected. This frequency effect is believed to be attributed to the change in the cavitation bubble dynamics as frequency is varied. In light of this new information, the acoustic power, which also has an effect on cavitation bubble dynamics, was varied for each frequency. 3.2. Influence of Power. Figure 4 shows the variation in the volume mean particle size with increasing power for different frequencies. For 22, 48, and 98 kHz, this graph shows an initial sharp decrease in the crystal size with increasing power, followed by a plateau where the crystal size becomes invariant to any further increment in power. In addition, for
3. RESULTS 3.1. Influence of Frequency. The volume mean particle size of NaCl crystals obtained under sonication at a calorimetric power of 20 W across a range of frequencies is shown in Figure 2. The crystal size formed under sonication is 83−95% smaller compared to that without sonication. As previously reported, there appears to be no obvious trend for frequencies up to 647 kHz.11 However, a further increase in the frequency to 1080 kHz resulted in a smaller mean particle C
DOI: 10.1021/acs.cgd.8b00547 Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
Article
minimum size obtained is tabulated in Table 1. The data reveal an optimum frequency between 44 and 98 kHz where the power to achieve 50% size reduction and the minimum size reached is the lowest. Jordens et al.38 also report a similar optimum frequency of 41 kHz from a range of 41−1140 kHz investigated. 3.3. Cavitation Activity. There are various theories proposed in the literature as to the mechanism behind ultrasound initiated crystallization. However, most agree that the effect of ultrasound during crystallization is not due to the sound waves, per se, but to the effect caused by cavitation bubbles. Sonoluminescence (SL) can be used as a probe for detecting the onset of cavitation activity. Shown in Figure 6 are the SL intensities as a function of power under the four selected frequencies. The SL from frequencies 44 and 139 kHz were not investigated as they gave similar crystal size as 98 and 647 kHz, respectively. Although the SL intensities were obtained for water, it is assumed that the observed relative changes in the cavitation activity as a function of power for a given frequency in water would apply for ethanol. The dependence of SL intensity on power and frequency in Figure 6 are consistent with previous reports.31,34,35 The particle volume mean diameters are also shown in this figure for comparison. There exist a strong alignment between the power threshold for SL activity and the power threshold for particle size reduction for frequencies 647 and 1080 kHz. Furthermore, for all frequencies, the decrease in the particle size coincided with the observed increase in the SL intensity as a function of power. However, as illustrated in Figure 7, there does not appear to be a correlation between the magnitude of SL intensity and the reduction in the mean particle size, with the exception of the final particle size obtained at each frequency, which decreases with increasing SL intensity as indicated by the dotted line. The relative rates of nucleation and growth of crystals is inversely proportional to the induction time,44 which can be determined by measuring the time interval between creation of supersaturation condition and occurrence of detectable crystals. Figure 8 shows that the induction time under various acoustic powers (indicated by the labels next to the data) at a fixed frequency of 98 kHz is strongly correlated with a decrease in the volume mean diameter of the crystals. Increasing acoustic power will elevate the bulk solution temperature and, in this case, raise the solubility concentration of NaCl in ethanol. This will in turn decrease the supersaturation ratio and increase induction time. However, for a calorimetric power of 5.5 W and 90 s sonication, the rise in the temperature of ethanol would only be 1.34 °C, and rather than increasing induction time, Figure 8 is showing a decrease. Therefore, the increase in the nucleation rate and decrease in the average crystal must be related to the increase in the cavitation activity in the solution.
Figure 3. Particle size distribution for (a) frequencies below 44 kHz and (b) frequencies from 44 to 1080 kHz. For all frequencies, the applied calorimetric power was fixed at 20 W. The size distribution for no US is shown for comparison.
4. DISCUSSION It is well-known that bulk fluid mixing has a significant effect on the crystallization process. Although ultrasound can generate bulk and localized mixing from pressure gradients and cavitating bubbles, its effect on the crystallization process is different to mechanical mixing. This has been demonstrated in a previous report by Lee et al.11 that, although crystals obtained under bulk mechanical mixing are smaller compared to no mixing, the crystal sizes are irregular in shape, whereas crystals obtained under sonication were all uniform and cubic
Figure 4. Effect of ultrasound frequency and power on volume mean diameter of NaCl crystals.
frequencies 647 and 1080 kHz, there appears to be an initial threshold power that needs to be overcome before there is any reduction in the crystal size (Figure 5). To compare the effectiveness of various frequencies at reducing the crystal size, the power to achieve 50% reduction in crystal size and the D
DOI: 10.1021/acs.cgd.8b00547 Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
Article
Figure 5. Particle size distribution with increasing ultrasound power for frequencies 22, 44, 647, and 1080 kHz.
during the collapse phase to generate SL.2,45 The ratio of Rmax/ Ro and SL intensity will both rise with increasing power. The range of bubble sizes that can generate SL also widens with increasing power.45 As frequency increases, the wavelength of ultrasound wave becomes shorter, which can effectively increase the number of antinodes and hence increase the population of SL bubbles.31 However, the period of the ultrasound wave is shorter, causing the bubble to expand less (resulting in smaller ratio of Rmax/ Ro) and hence a weaker collapse intensity, which leads to a milder emission of SL.45 As a consequence of this, higher frequencies have a greater inertial cavitation threshold as illustrated in Figure 6 by the power threshold for sonoluminescence emission of 2.5 and 4.3 W observed for 647 and 1080 kHz, respectively. The range of bubble sizes capable of undergoing SL also becomes narrower with increasing frequencies.45,46 It is this balance between the increase in the number of SL bubbles and the decrease in the ratio of Rmax/Ro and the range of bubble sizes capable of SL or transient inertial collapse45 that gives rise to the maximum SL intensity observed at 98 kHz. It is important to note that the SL measurements were performed in water and the crystallization experiments in ethanol (after the addition of aqueous salt solution). It is not possible to quantify the cavitation activity in ethanol via SL measurements due to quenching effects;41 however, it is assumed that the variations in transient cavitation activity as a function of power and
Table 1. Power To Achieve 50% Reduction in Crystal Size and Minimum Crystal Size Obtained for Various Frequencies frequency [kHz]
power at 50% size reduction [W]
minimum size [μm]
22 44 98 139 647 1080
1.8 0.8 0.8 1.6 3.0 10.8
15.0 ± 1.3 8.3 ± 1.4 7.2 ± 0.1 11.3 ± 1.1 13.3 ± 0.6 30.2 ± 1.7
in shape. This suggests that the sonication affects the crystal nucleation rate rather than irregular crystal growth as indicated in Figure 8. As the crystal nucleation rate increases with acoustic power, the consumption of the supersaturation by this higher number of nuclei results in the overall smaller crystals. To understand the different effects that frequency and power have on the nucleation rate, one needs to consider the magnitude of cavitation activity occurring under these conditions. The cavitation activity can be quantified by the total SL intensity, which is related to both the collapse intensity (proportional to the ratio of maximum bubble size reached to the ambient bubble radius, Rmax/Ro) and the number of the individual SL bubbles. In general, as the acoustic power increases, bubbles at the pressure antinodes will grow to a critical size for the bubbles to gain sufficient inertial energy E
DOI: 10.1021/acs.cgd.8b00547 Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
Article
Figure 6. Sonoluminescence intensity (black) as a function of power for different frequencies. The change in the volume mean diameter (gray) is added for comparison.
Figure 7. Plot of volume mean particle diameter as a function of SL intensity. The dotted line indicates the minimum particle size and SL intensity reached for each frequency.
Figure 8. Induction time plotted as a function of volume mean diameter, obtained for different acoustic powers (indicated by the labels next to the data) at a frequency of 98 kHz.
frequency will be proportional to the relative variations observed for SL intensity in water. Although 44 and 98 kHz are most favorable in terms of energy input according to Table 1, Figure 7 shows that a 50% reduction in the particle size can be achieved at SL intensities from a low of 0.05 mV at 1080 kHz to a high of 500 mV for 98 kHz. This result indicates that it is the onset of cavitation activity, irrespective of the intensity of collapse, that enhances nucleation rate and decreases the crystal size. However, the final crystal size reached for each frequency (indicated by the dotted line) decreases with increasing SL intensity, in the order
of 1080 kHz > 647 kHz > 22 kHz > 98 kHz. Furthermore, as shown in Figure 8, at any particular frequency the crystal size is strongly correlated to the acoustic power and the induction time. This suggests that the number and collapse intensity of the cavitating bubbles do play a role in determining this crystal size. SL can be classified as single bubble sonoluminescence (SBSL) and multibubble sonoluminescence (MBSL) and is normally associated with symmetrical bubble collapse. Cairos and Mettin49 have demonstrated that multibubble SL emission is largely attributed to nonspherical bubble geometries brought about by spatial translation, interaction, and collision of F
DOI: 10.1021/acs.cgd.8b00547 Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
Article
tion, and from the sonocrystallization photographs, suggests the importance of nonsymmetrical transient cavitation as the cause of sonocrystallization.
bubbles with each other that leads to transient SL bubble dynamics.47,48 In this study, attempts were made to drive a single SL bubble during the antisolvent crystallization process, but no crystallization was observed. Due to the use of ethanol, it is difficult to confirm if single transient bubble (symmetrical) collapse was established. However, what was readily observed was crystallization emanating from a cluster of cavitation bubbles as shown in Figure 9a,b, suggesting the importance of
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Judy Lee: 0000-0001-5808-2983 Sandra E. Kentish: 0000-0002-4250-7489 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors would like to acknowledge the financial support through the Australian Research Council for the DECRA (Discovery Early Career Research Award, DE120101567), Mr. James Dyck for repeating some of the crystallization experiments, and Mr. Junjie Jiao for the imaging.
■
REFERENCES
(1) Brenner, M. P.; Hilgenfeldt, S.; Lohse, D. Single-bubble sonoluminescence. Rev. Mod. Phys. 2002, 74, 425−484. (2) Matula, T. J. Inertial cavitation and single-bubble sonoluminescence. Philos. Trans. R. Soc., A 1999, 357, 225−250. (3) Mason, T.J. Practical Sonochemistry: User’s Guide to Applications in Chemistry and Chemical Engineering; Ellis Horwood: New York, 1991. (4) Et Taouil, A.; Lallemand, F.; Hallez, L.; Hihn, J. Electropolymerization of pyrrole on oxidizable metal under high frequency ultrasound irradiation. Application of focused beam to a selective masking technique. Electrochim. Acta 2010, 55, 9137−9145. (5) Ashokkumar, M.; Bhaskaracharya, R.; Kentish, S.; Lee, J.; Palmer, M.; Zisu, B. The ultrasonic processing of dairy products - An overview. Dairy Sci. Technol. 2010, 90, 147−168. (6) Nakabayashi, K.; Amemiya, F.; Fuchigami, T.; Machida, K.; Takeda, S.; Tamamitsu, K.; Atobe, M. Highly clear and transparent nanoemulsion preparation under surfactant-free conditions using tandem acoustic emulsification. Chem. Commun. 2011, 47, 5765− 5767. (7) Lyczko, N.; Espitalier, F.; Louisnard, O.; Schwartzentruber, J. Effect of ultrasound on the induction time and the metastable zone widths of potassium sulphate. Chem. Eng. J. 2002, 86, 233. (8) Bund, R. K.; Pandit, A. B. Sonocrystallization: Effect on lactose recovery and crystal habit. Ultrason. Sonochem. 2007, 14, 143−152. (9) Kordylla, A.; Koch, S.; Tumakaka, F.; Schembecker, G. Towards an optimized crystallization with ultrasound; Effect of solvent properties and ultrasonic process parameters. J. Cryst. Growth 2008, 310, 4177−4184. (10) Harzali, H.; Baillon, F.; Louisnard, O.; Espitalier, F.; Mgaidi, A. Experimental study of sono-crystallisation of ZnSO4·7H2O, and interpretation by the segregation theory. Ultrason. Sonochem. 2011, 18, 1097−1106. (11) Lee, J.; Ashokkumar, M.; Kentish, S. Influence of mixing and ultrasound frequency on antisolvent crystallisation of sodium chloride. Ultrason. Sonochem. 2014, 21, 60−68. (12) Nii, S.; Takayanagi, S. Growth and size control in anti-solvent crystallization of glycine with high frequency ultrasound. Ultrason. Sonochem. 2014, 21, 1182−1186. (13) Nguyen, T. T.; Khan, A.; Bruce, L. M.; Forbes, C.; O’Leary, R. L.; Price, C. J. The Effect of Ultrasound on the Crystallisation of Paracetamol in the Presence of Structurally Similar Impurities. Crystals 2017, 7, 294. (14) Gielen, B.; Claes, T.; Janssens, J.; Jordens, J.; Thomassen, L. C.; Gerven, T. V.; Braeken, L. Particle size control during ultrasonic
Figure 9. Images of antisolvent crystallization of sodium chloride crystals nucleated by cavitation bubbles: (a) a cluster of cavitating bubbles leaving a trail of fine NaCl crystals (33.1 kHz), (b) close-up of the bubble cluster in (a), (c) collision between approaching bubbles (indicated by the arrows) in frame 1 and collision of the bubbles leading to nucleation in frame 7 (22 kHz), and (d,e) nucleation at the pressure antinodes for 44 and 139 kHz, respectively.
nonspherical bubble collapses leading to sonocrystallization. Further evidence of spatial translation, interaction, and collision of bubbles leading to sonocrystallization can be observed in Figure 9c from the collision of three acoustically driven bubbles. Similarly, crystallization from bubble streamers and strong transient cavitation bubbles at pressure antinodes can be observed in Figure 9d,e, respectively. What is difficult to elucidate is what aspects of crystallization cavitation are enhanced, i.e., primary or secondary nucleation.
5. CONCLUSION In this study, both the size of NaCl crystals and sonoluminescence (SL) intensity were measured as a function of ultrasound frequency and power. The results demonstrated that the nucleation rate can be enhanced by ultrasound at very low acoustic powers and is most effective at frequencies between 44−98 kHz. The mean particle size was found to decrease sharply as a function of power for frequencies 22−139 kHz. At frequencies 647 and 1080 kHz, particle size decreased only when the power level exceeded a certain threshold. This threshold coincided with the power threshold for SL activity. However, no correlation between the SL intensity and the reduction in the crystal size was observed. It is thus concluded that the onset of cavitation activity, irrespective of its magnitude, is capable of enhancing the crystal nucleation rate and causes a decrease in the size of crystals formed. However, the minimum crystal size obtained was found to be proportional to the intensity of cavitation activity generated. The association between the SL intensity and sonocrystallizaG
DOI: 10.1021/acs.cgd.8b00547 Cryst. Growth Des. XXXX, XXX, XXX−XXX
Crystal Growth & Design
Article
cooling crystallization of paracetamol. Chem. Eng. Technol. 2017, 40, 1300−1308. (15) Gracin, S.; Uusi-Penttila, U.; Rasmuson, A. C. Influence of ultrasound on the nucleation of polymorphs of p-Aminobenzoic acid. Cryst. Growth Des. 2005, 5, 1787−1794. (16) Price, G.; Mahon, M. F.; Shannon, J.; Cooper, C. Composition of calcium carbonate polymorphs precipitated using ultrasound. Cryst. Growth Des. 2011, 11, 39−44. (17) Mori, Y.; Takahashi, Y.; Ikeda, K.; Yamada, T.; Nishimura, K.; Maruyama, M.; Yoshikawa, H.; Okada, S.; Adachi, H.; Sugiyama, S.; Matsumura, H.; Takano, K.; Inoue, T.; Murakami, S.; Yoshimura, M.; Mori, Y. Polymorphism Control of Pharmaceutical Compound Acetaminophen by Ultrasonic Irradiation. In 17th International Conference on Crystal Growth and Epitaxy; Sobanska, M., Zytkiewicz, Z. R., Eds.; Pielaszek Research: Warsaw, Poland, 2013; p 124. (18) Bhangu, S. K.; Ashokkumar, M.; Lee, J. Ultrasound assisted Crystallization of Paracetamol: Crystal Size Distribution and Polymorph Control. Cryst. Growth Des. 2016, 16, 1934−1941. (19) Kozhemyakin, G. Influence of ultrasonic vibrations on the growth of semiconductor single crystals. Ultrasonics 1998, 35, 599− 604. (20) Kozhemyakin, G. N.; Zolkina, L. V.; Inatomi, Y. Influence of Ultrasound on Crystal Growth from Solution and Related Flow Visualization. Cryst. Growth Des. 2006, 6, 2412−2416. (21) Hem, S. L. The effect of ultrasonic vibrations on crystallization processes. Ultrasonics 1967, 5, 202−207. (22) Ruecroft, G.; Hipkiss, D.; Ly, T.; Maxted, N.; Cains, P. W. Sonocrystallization: The use of ultrasound for improved industrial crystallization. Org. Process Res. Dev. 2005, 9, 923−932. (23) Louisnard, O.; Gomez, F.; Grossier, R. Segregation of a liquid mixture by a radially oscillating bubble. J. Fluid Mech. 2007, 577, 385−415. (24) Zeiger, B. W.; Suslick, K. S. Sonofragmentation of Molecular Crystals. J. Am. Chem. Soc. 2011, 133, 14530−14533. (25) Sander, J. R.; Zeiger, B. W.; Suslick, K. S. Sonocrystallization and sonofragmentation. Ultrason. Sonochem. 2014, 21, 1908−1915. (26) Cogné, C.; Labouret, S.; Peczalski, R.; Louisnard, O.; Baillon, F.; Espitalier, F. Theoretical model of ice nucleation induced by inertial acoustic cavitation. Part 2: Number of ice nuclei generated by a single bubble. Ultrason. Sonochem. 2016, 28, 185−191. (27) Hickling, R. Nucleation of freezing by cavity collapse and its relation to cavitation damage. Nature 1965, 206, 915−917. (28) Chow, R.; Blindt, R.; Chivers, R.; Povey, M. A study on the primary and secondary nucleation of ice by power ultrasound. Ultrasonics 2005, 43, 227−230. (29) Rossi, D.; Jamshidi, R.; Saffari, N.; Kuhn, S.; Gavriilidis, A.; Mazzei, L. Continuous-flow sonocrystallization in droplet-based microfluidics. Cryst. Growth Des. 2015, 15, 5519−5529. (30) Arends, B.J.; Blindt, R.A.; Janssen, J.; Patrick, M. Crystallisation Process Using Ultrasound; Lipton, Division of Conopco, Inc.: United States, 2002; pp 1−7. (31) Hatanaka, S.; Yasui, K.; Tuziuti, T.; Kozuka, T.; Mitome, H. Quenching mechanism of multibubble sonoluminescence at excessive sound pressure. Jpn. J. Appl. Phys. 2001, 40, 3856−3860. (32) Ashokkumar, M.; Lee, J.; Iida, Y.; Yasui, K.; Kozuka, T.; Tuziuti, T.; Towata, A. Spatial distribution of acoustic cavitation bubbles at different ultrasound frequencies. ChemPhysChem 2010, 11, 1680−1684. (33) Ashokkumar, M.; Sunartio, D.; Kentish, S.; Mawson, R.; Simons, L.; Vilkhu, K.; Versteeg, C. Modification of food ingredients by ultrasound to improve functionality. Innovative Food Sci. Emerging Technol. 2008, 9, 155−160. (34) Beckett, M. A.; Hua, I. Impact of Ultrasonic Frequency on Aqueous Sonoluminescence and Sonochemistry. J. Phys. Chem. A 2001, 105, 3796−3802. (35) Kanthale, P.; Ashokkumar, M.; Grieser, F. Sonoluminescence, sonochemistry (H2O2 yield) and bubble dynamics: Frequency and power effects. Ultrason. Sonochem. 2008, 15, 143−150.
(36) Wohlgemuth, K.; Ruether, F.; Schembecker, G. Sonocrystallization and crystallization with gassing of adipic acid. Chem. Eng. Sci. 2010, 65, 1016−1027. (37) Louhi-Kultanen, M.; Karjalainen, M.; Rantanen, J.; Huhtanen, M.; Kallas, J. Crystallization of glycine with ultrasound. Int. J. Pharm. 2006, 320, 23−29. (38) Jordens, J.; Gielen, B.; Braeken, L.; Van Gerven, T. Determination of the effect of the ultrasonic frequency on the cooling crystallization of paracetamol, Chem. Eng. Process. Chem. Eng. Process. 2014, 84, 38−44. (39) Kurotani, M.; Miyasaka, E.; Ebihara, S.; Hirasawa, I. Effect of ultrasonic irradiation on the behavior of primary nucleation of amino acids in supersaturated solutions. J. Cryst. Growth 2009, 311, 2714− 2721. (40) Hallez, L.; Touyeras, F.; Hihn, J.; Klima, J. Energetic balance in an ultrasonic reactor using focused or flat high frequency transducers. Ultrason. Sonochem. 2007, 14, 739−749. (41) Ashokkumar, M.; Hall, R.; Mulvaney, P.; Grieser, F. Sonoluminescence from aqueous alcohol and surfactant solutions. J. Phys. Chem. B 1997, 101, 10845−10850. (42) Asakura, Y.; Nishida, T.; Matsuoka, T.; Koda, S. Effect of ultrasonic frequency and liquid height on sonochemical efficiencyof large-scale sonochemical reactors. Ultrason. Sonochem. 2008, 15, 244− 250. (43) This might explain the report by Arends et al.,30 where fat crystallization occurred even though no cavitation emission was detected. It is plausible that the cavitation events were too small to be detected by the hydrophone. As indicated, the SL intensity at 1080 kHz was extremely low, and a higher amplification had to be used. (44) Söhnel, O.; Mullin, J. W. Interpretation of crystallization induction periods. J. Colloid Interface Sci. 1988, 123, 43−50. (45) Yasui, K. Influence of ultrasonic frequency on multibubble sonoluminescence. J. Acoust. Soc. Am. 2002, 112, 1405−1413. (46) Brotchie, A.; Grieser, F.; Ashokkumar, M. Effect of power and frequency on bubble-size distributions in acoustic cavitation. Phys. Rev. Lett. 2009, 102, 084302. (47) Thiemann, A.; Holsteyns, F.; Cairós, C.; Mettin, R. Sonoluminescence and dynamics of cavitation bubble populations in sulfuric acid. Ultrason. Sonochem. 2017, 34, 663−676. (48) Nowak, T.; Mettin, R. Unsteady translation and repetitive jetting of acoustic cavitation bubbles. Phys. Rev. E 2014, 90, 033016. (49) Cairós, C.; Mettin, R. Simultaneous high-speed recording of sonoluminescence and bubble dynamics in multibubble fields. Phys. Rev. Lett. 2017, 118, 064301.
H
DOI: 10.1021/acs.cgd.8b00547 Cryst. Growth Des. XXXX, XXX, XXX−XXX
