Response of interfacial nanobubbles to ultrasound irradiation

Abstract

The behaviour of nanobubbles and nanodroplets at solid–liquid interfaces in an acoustic field is of interest in terms of both fundamental research and the various applications of nanofluids coupled with ultrasound. Herein, we show by in situatomic force microscopy imaging that nanobubbles experienced growthvia rectified gas diffusion yet did not nucleate cavitation. Nanodroplets were remarkably immobile after a period of initial mobility in the sound field. The stability of the interfacial nanofluids towards ultrasound may be attributed to pinning on the three-phase contact line.

---

The presence of nanobubbles at solid–water interfaces was first proposed to explain the origin of the long-range attractive force between hydrophobic surfaces in water.^1–3 Their existence, however, has been controversial due to thermodynamic arguments.^1,3–5 Recent studies have shown that nanoscale gas bubbles can be reproducibly formed at a solid–water interface where a gas supersaturation exists.^6–9 The lifetime of interfacial nanobubbles can be several days and their morphology, as observed by atomic force microscopy (AFM), resembles a spherical cap with a typical height of 10–20 nm and lateral diameter of 1–2 µm.⁹ The gaseous nature of this interfacial state has been confirmed by quartz crystal microbalance,¹⁰ and surface plasmon resonance and de-gassing studies. Moreover, the chemical identity of carbon dioxide nanobubbles has been elucidated through infrared spectroscopy.⁹

Interfacial nanobubbles are implicated in various interfacial phenomena such as surface forces, biomolecule adsorption and slippage boundary conditions,^3,11–13 and have been exploited in applications such as electroless templating¹⁴ and nanomotors.¹⁵ However, solid understanding of their properties remains elusive. An interesting question pertains to the response of nanobubbles to acoustic waves. When bubbles are subjected to ultrasound, the flux of gas across the bubble interface will be affected by the oscillating pressure field, such that the bubble may grow through a slow rectified gas diffusion.^16,17 Mechanical agitation from sound waves and cavitation events can induce liquid flow (acoustic streaming), which increases the convective mass transport of dissolved gases, acting to accelerate gas (and solute) diffusion. It is well established that free spherical microbubbles in the bulk solution can nucleate, grow, shrink or collapse in an acoustic field, depending on the size of the bubbles and the intensity of the sound wave.¹⁷ The question therefore is, does the confinement of the nanobubbles to the solid surface have a significant influence on their physical response to a sound field? Various models for bubble nucleation have been proposed, such as spontaneous nucleation from thermal spikes caused by cosmic radiation^18,19 and nucleation from stabilised gas pockets entrained in tiny crevices.^20–22 Recent research concerning interfacial nanobubbles suggests that they are ‘super-stable’ towards nucleation even under an enormous reduction of the liquid pressure down to −6 MPa.²³

In this work we have shown the response of interfacial nanobubbles to the application of a low pressure ultrasound field (for a period of 40 s) by atomic force microscopy (AFM) imaging before, and after, ultrasound exposure. The behaviour of nanobubbles is compared to that of interfacial nanodroplets in the acoustic field where rectified gas diffusion cannot occur due to the incompressible nature of the liquid inside the droplets.

Nanobubbles were produced on the surface of freshly cleaved highly ordered pyrolytic graphite (HOPG) (ZYB, SPI Supplies, PA) by the exchange of ethanol with water.^6,9HOPG has a layered structure with each layer being atomically smooth. The advancing angle of water on HOPG is 72° and receding angle is 66°. The exchange procedure has been established in our previous work.^6,9,10 In brief, the interfacial nanobubbles are generated by exposing HOPG in one solvent (ethanol) and exchanging that solvent with a second solvent (water). The gas solubility (of N₂, O₂ and CO₂) is much higher in ethanol than water. The hypothesis is that during the exchange gases that were originally dissolved in ethanol become supersaturated, and form nanobubbles on the surface.

The solvent exchange was performed in a closed fluid cell. The bottom of the fluid cell was attached firmly to a piezoceramic ultrasound transducer (Undatim Ultrasonics, 515 kHz). A schematic diagram of the setup is shown in Fig. 1a. Following the exchange of ethanol by water, 100 mM n-propanol aqueous solution (surface tension ∼60 mN m⁻¹) was used to displace the water. The solution was equilibrated with air before use. This step was introduced as surface-active species enhance the process of gas rectified diffusion into an oscillating bubble in the acoustic field.^24,25 It should be noted that similar results were observed in pure water but with less growth. The nanobubbles were imaged by tapping mode AFM (MFP-3D, Asylum Research, Santa Barbara, CA). The cantilever (0.32 N m⁻¹, NP, Veeco, USA) was cleaned by UV irradiation for 20 min. The ultrasound was delivered in situ, permitting the same nanobubbles on the surface to be imaged before and after the application of ultrasound. Sonication was terminated after 35–40 s, immediately upon the formation of micrometre-sized cavitation bubbles. These microbubbles arose from adventitious nuclei on the surface and/or solution and formed similarly in the absence and presence of the nanobubbles. The ultrasonic intensity transmitted to the fluid cell was determined calorimetrically to be 2.2 W. These sonication conditions could cause a temperature rise of about 4 °C. AFM imaging was conducted after the system had cooled to the same temperature as before sonication, and the images were taken using the identical tip and imaging conditions.

Fig. 1 (a) Schematic representation of the modified closed fluid cell for in situ ultrasound application and tapping-mode AFM imaging; (b) the cross-sectional profiles of three pairs of nanobubbles, as marked in (c). Tapping mode AFM height images of nanobubbles are shown in (c) prior to sonication and (d) after sonication for 40 s. The scale bars for (c) and (d) are identical and the scan size is 5 × 5 µm².

Nanobubble images before and after sonication are shown in Fig. 1c and d. It can be observed that nanobubbles neither disappeared nor appeared after the ultrasound application and remained at the same location, reflecting their immobility on the surface. Individual nanobubbles, however, became noticeably larger after sonication. This is particularly evident in the cross-sectional profiles corresponding to the three bubbles marked in the height image shown in Fig. 1b. The height of the nanobubbles increased markedly, whereas the lateral size did not change appreciably, indicating that the bubbles were pinned on the three-phase contact line.

These results also suggest that the interfacial nanobubbles do not serve as nucleation sites for the formation of free acoustic cavitation bubbles. If they were to act as nucleation sites, one would expect a much greater change in volume after sonication (if the bubble did not detach) or removal from the surface (if the bubble detached). Furthermore, we have observed that free cavitation bubbles readily remove all interfacial nanobubbles. However, the nanobubbles shown in the AMF images have likely experienced gentle radial oscillations and only modest growth in the acoustic field. This is not entirely unexpected behaviour. The Blake threshold defines the radius below which a bubble will oscillate in a smooth and stable manner, and above which the bubble will undergo explosive growth and subsequent inertial collapse.^17,26 We calculate that with the acoustic pressure applied in the present experiment (∼1.2 bar, estimated based on the calorimetric power), the threshold is about 2.3 µm. The radii of curvature of nanobubbles observed on HOPG ranged between 1 and 2.6 µm with only a few larger than 2.5 µm in our experiments. Therefore, explosive growth and nucleation is not expected for the majority of the surface nanobubbles. We note that the experiment was conducted under ambient pressure of ∼1 atm. Despite being experimentally challenging, a future study into the effect of ambient pressure on nanobubble dissolution and growth under ultrasound may provide further insight into our understanding of these processes.

Rectified diffusion of gas is the principal mechanism of (single) bubble growth in an acoustic field, provided the acoustic pressure is sufficiently high to overcome bubble dissolution.¹⁶ Briefly, this arises because in one acoustic cycle the amount of gas that diffuses into a bubble (during the expansion phase) is greater than the amount that diffuses out (during the compression phase). Considering that interfacial nanobubbles are quasi-stable towards dissolution, they are susceptible to growth in an acoustic field. Although this is a complex phenomenon, one should expect a strong relationship between volume change and surface area of the gas–liquid interface. The plot of the absolute increase in the nanobubble volume as a function of the initial nanobubble surface area is shown in Fig. 2a and indeed, an almost linear relationship is exhibited. The relative increase in the height, lateral size and volume as a function of the initial surface area of the nanobubbles is shown in Fig. 2b. The relative increases in height and volume are much more pronounced for the smaller bubbles, as would be expected for free cavitation bubbles in the bulk solution. Importantly, this result provides further proof that the soft domains in the AFM images were indeed gas bubbles. In addition to spherical cap nanobubbles, another type of morphology of interfacial gases, “micropancakes” (i.e. flat (quasi-two-dimensional gaseous layers)), can also be formed on HOPG by solvent exchange.²⁷ We observed that the height of the micropancakes increased while the lateral size decreased following the application of ultrasound (see Fig. S1 in ESI†).

Fig. 2 Change of nanobubble size after sonication. (a) Absolute change in nanobubble volume as a function of the initial surface area of the gas–liquid interface; (b) relative change in nanobubble height, lateral diameter and volume as a function of the initial surface area; (c) the change in contact angle as a function of the initial surface area. Bubble volumes, surface areas and contact angles were calculated using a spherical cap model with the height profile of the nanobubbles from AFM data.

It is also possible that the heating effect of the ultrasound could have contributed to the growth of the nanobubbles. However, we calculated a negligible thermal expansion (less than 2%) for the temperature rise during the experiment. Furthermore, the correlation between initial size and the extent of growth was not observed in a separate experiment when we heated the solution rapidly and then let it cool to room temperature.²⁸ This indicates that ultrasound exposure is the main factor responsible for the bubble growth. It is conceivable that rectified gas diffusion in the sound field might have been reinforced by the temperature effect.

It can also be noted that the contact angle of the nanobubbles increased significantly, by about 10–50% (2–5°) through the gas phase, after sonication. The increase of the contact angle with the initial surface area is shown in Fig. 2c. As the contact angles did not revert to their initial values prior to sonication, the nanobubbles were clearly quasi-stable at the two different angles (i.e., they were not in equilibrium).^29,30 The increase in the contact angle of nanobubbles after the growth indicates that the three-phase contact line of the nanobubbles was pinned on the surface. Despite the mechanical agitation, nanobubbles were neither dislodged nor moved laterally following sonication, providing further evidence that they were anchored on the substrate.

To aid our understanding of the behaviour and stability of nanobubbles, the response of oil nanodroplets to the mechanical perturbation imparted by the acoustic field was investigated. Unlike nanobubbles, rectified diffusion does not occur for nanodroplets due to the incompressible nature of a liquid droplet. Nanoscale decane droplets were generated on the HOPG surface by the exchange of 40% ethanol aqueous solution with water,³¹ and subsequently exposed to ultrasound. Both solutions were saturated with decane and the ultrasound parameters were the same as those applied to the nanobubbles.

The images taken before and after sequential sonication periods are shown in Fig. 3 (larger scan area images are provided in the ESI†). After the initial sonication period of 20 s, the distribution of nanodroplets became significantly more non-uniform. Some of the droplets increased in size appreciably whereas most became smaller and, in fact, many new small droplets appeared. It is apparent that many droplets were mobile on the surface. In the vicinity of those that had undergone a large increase in size, there were regions with droplets that had shrunk or that were devoid of droplets, indicating the occurrence of coalescence between the neighbouring droplets and/or Ostwald ripening. The appearance of new small droplets may also have arisen through splitting of droplets due to ultrasound induced fragmentation and subsequent re-adsorption of the small oil fragments. In any case, these results show that sonication can lead to a dramatic change in the interfacial nanodroplets.

Fig. 3 Tapping mode AFM height images of decane droplets on a HOPG surface taken prior to sonication (a) and after sonication for a period of 20 s (b) and 60 s (c). The images were taken at the same location, as shown by the steps on the HOPG. The scan size is 20 × 20 µm².

After the second application of ultrasound for a longer period of 60 s, there was much less change in the size and spatial distribution of the nanodroplets. Only a few droplets appeared to merge into bigger droplets and several little droplets disappeared, indicating the occurrence of dissolution.³² The fact that there was less change between 20 and 60 s sonication eliminates the possibility that the changes observed in the oil droplets were predominantly temperature mediated. At this point, with respect to mobility on the surface, this population of nanodroplets behaved similarly to the nanobubbles under sonication. These remaining nanodroplets were located near the steps of the graphite, observed from the images. Such steps of the surface may have pinned the nanodroplets within this population and prevented their dislodgement and coalescence.

Contact line pinning from the heterogeneity on the surface should be a general occurrence for both nanodroplets and nanobubbles on the same surface, although the total pinning force exerted on the contact line could be slightly different from the difference in their interfacial tension.³³ The question that arises is why the nanobubbles were not mobile at all in the acoustic field whereas some nanodroplets were mobile during the initial sonication period. This may be explained by the difference in the formation processes of nanobubbles and nanodroplets. Nanobubbles can only be formed through nucleation on the solid surface during the solvent exchange process,³⁴ which would preferentially occur at defects. Nanodroplets, however, can be formed both from nucleation on the surface and from the adsorption of oil drops floating in the bulk emulsion.^31,35 For the latter case, nanodroplets could be positioned on relatively smooth areas of the substrate. The interfacial nanodroplets born on the surface would therefore be more likely to have a pinned three-phase contact line and show high immobility under sonication. Although diffusion related processes of growth and dissolution are dramatically different between the gas and liquid, the similar observations made for the spatial stability of air bubbles and oil droplets alike under ultrasound indicate that the interaction with the solid surface defines the stability of both nanofluids. The strong pinning on the surface may have prevented coalescence with neighbouring nanobubbles thereby restricting the growth of the nanobubbles by rectified diffusion.³⁶

In summary, we have shown that interfacial nanobubbles grow moderately in an ultrasound field. The growth was mainly achieved through an increase in height and consequently an increase in contact angle. Interfacial nanodroplets exhibited stability after initial facile mobility in the sound field. The stability of both nanofluids may be attributed to the pinning from the heterogeneity on the surface. The results of this work provide insight into the stability of interfacial nanofluids and also have potential implications in several applied fields. For instance, in therapeutic applications, small gas-filled bubbles and oil filled droplets have been used as effective ultrasound contrast agents and ultrasound enhanced drug delivery vehicles.³⁷ These nanofluids often adhere onto the target site in the ultrasound field. Nanobubbles may also find application in microfluidic devices where acoustic waves are used to facilitate liquid transport through narrow channels. One of the fundamental questions remaining in the area of nanobubbles pertains to the mechanism of the stability responsible. While the present work shows that nanobubbles and nanodroplets alike are highly stable in an ultrasonic field, further understanding will require a systematic investigation of the influence of various environmental parameters (e.g. acoustic frequency and pressure, ambient pressure, gas type and surface hydrophobicity).

Acknowledgements

We greatly thank Professor Franz Crieser for inspiring discussion and valuable advice. We gratefully acknowledge Professor Muthupandian Ashokkumar for the initial involvement in this project, the advice on the manuscript and the final proof reading. We acknowledge an Australian Research Council Discovery Project grant (DP0880152). A.B. acknowledges the receipt of an Australian Postgraduate Award and a David Lachlan Hay Post-graduate Writing-up Award.

Notes and references

1. P. Attard, Adv. Colloid Interface Sci., 2003, 104, 75.
2. A. Carambassis, L. C. Jonker, P. Attard and M. W. Rutland, Phys. Rev. Lett., 1998, 80, 5357.
3. H. K. Christenson and P. M. Claesson, Adv. Colloid Interface Sci., 2001, 91, 391.
4. J. C. Eriksson and S. Ljunggren, Colloids Surf., A, 1999, 159, 159.
5. P. Ball, Nature, 2003, 423, 25.
6. S. T. Lou, Z. Q. Ouyang, Y. Zhang, X. J. Li, J. Hu, M. Q. Li and F. J. Yang, J. Vac. Sci. Technol., B, 2000, 18, 2573.
7. J. W. Yang, J. M. Duan, D. Fornasiero and J. Ralston, J. Phys. Chem. B, 2003, 107, 6139.
8. S. J. Yang, P. C. Tsai, E. S. Kooij, A. Prosperetti, H. J. W. Zandvliet and D. Lohse, Langmuir, 2009, 25, 1466.
9. X. H. Zhang, A. Khan and W. A. Ducker, Phys. Rev. Lett., 2007, 98, 131601.
10. X. H. Zhang, Phys. Chem. Chem. Phys., 2008, 10, 6842.
11. S. M. Dammer and D. Lohse, Phys. Rev. Lett., 2006, 96, 206101.
12. G. M. Liu, Z. H. Wu and V. S. J. Craig, J. Phys. Chem. C, 2008, 112, 16748.
13. C. Neto, D. R. Evans, E. Bonaccurso, H.-J. Butt and V. S. J. Craig, Rep. Prog. Phys., 2005, 68, 2859.
14. C. Huang, J. Jiang, M. Lu, L. Sun, E. I. Meletis and Y. Hao, Nano Lett., 2009, 9, 4297.
15. W. F. Paxton, K. C. Kistler, C. C. Olmeda, A. Sen, S. K. St Angelo, Y. Y. Cao, T. E. Mallouk, P. E. Lammert and V. H. Crespi, J. Am. Chem. Soc., 2004, 126, 13424.
16. L. A. Crum, Ultrasonics, 1984, 22, 215.
17. T. G. Leighton, The Acoustic Bubble, Academic Press Inc., London, 1994.
18. F. Seitz, Phys. Fluids, 1958, 1, 2.
19. R. D. Finch, J. Acoust. Soc. Am., 1964, 36, 2287.
20. D. E. Yount, J. Acoust. Soc. Am., 1982, 71, 1473.
21. A. A. Atchley and A. Prosperetti, J. Acoust. Soc. Am., 1989, 86, 1065.
22. B. M. Borkent, S. Gekle, A. Prosperetti and D. Lohse, Phys. Fluids, 2009, 21, 102003.
23. B. M. Borkent, S. M. Dammer, H. Schonherr, G. J. Vancso and D. Lohse, Phys. Rev. Lett., 2007, 98, 204502.
24. J. Lee, S. Kentish and M. Ashokkumar, J. Phys. Chem. B, 2005, 109, 14595.
25. L. A. Crum, J. Acoust. Soc. Am., 1980, 68, 203.
26. S. Hilgenfeldt, M. P. Brenner, S. Grossmann and D. Lohse, J. Fluid Mech., 1998, 365, 171.
27. X. H. Zhang, X. Zhang, J. Sun, Z. Zhang, G. Li, H. Fang, X. Xiao, X. Zeng and J. Hu, Langmuir, 2007, 23, 1778.
28. Nanobubbles were formed on HOPG and the temperature was raised from 28 °C to 38 °C in 2 min by a heating element in the absence of ultrasound. This induced a detectable size increase of nanobubbles, however, the correlation between initial size and the extent of growth was not observed as with the ultrasound treatment.
29. J. N. Israelachvili, Intermolecular and Surface Forces, Academic Press Limited, San Diego, 2nd edn, 1992.
30. R. D. Deegan, Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top., 2000, 61, 475.
31. X. H. Zhang and W. Ducker, Langmuir, 2007, 23, 12478.
32. It should be noted that it was difficult to obtain the precise cross-sectional profiles of the droplets in water due to the deformation by the imaging force.
33. P. G. de Gennes, Rev. Mod. Phys., 1985, 57, 827.
34. A. Habich, W. Ducker, D. E. Dunstan and X. H. Zhang, J. Phys. Chem. B, 2010, 114, 6962–6967.
35. X. H. Zhang and W. Ducker, Langmuir, 2008, 24, 110.
36. When a large cavitation bubble formed during sonication and detached from the surface, all nanobubbles present in that vicinity disappeared, indicating the occurrence of dislodgement and/or coalescence.
37. H. Tang, C. C. J. Wang, D. Blankschtein and R. Langer, Pharm. Res., 2002, 19, 1160.

---

Footnote

† Electronic supplementary information (ESI) available: Morphological changes of micropancakes after sonication are shown in Fig. S1 and large scan area (50 µm × 50 µm) AFM images of oil nanodroplets before and after 20 s and 60 s of sonication are provided in Fig. S2. See DOI: 10.1039/c0sm00731e

---