Jetting from cavitation bubbles due to multiple shockwaves

We present experimental observations of microjets formed by cavitation microbubbles. An underwater electric discharge, applied beneath a flat free surface, produces a primary compression wave, which undergoes several phase inversions upon reflections from the free surface and spark-bubble interface. The first reflection yields a tension wave, which produces a cloud of secondary cavitation bubbles in the liquid, some of which form microjets upon collapse. The tuning of these reflections enables an effective control of the microjet direction in the bubble cloud. All of the jets of the microbubbles between the spark bubble and free surface are directed radially away from the spark bubble. The mechanical response of an alumina plate placed between the electrodes and free surface generates a quasi-planar compression wave, which, following its multiple reflections from the free surface and plate, orients the microjets in the same direction toward the plate. These observations imply that the jet direction is determined mainly by the secondary compression wave, which is the first and thus most energetic compression wave acting on a sufficiently grown cavitation bubble.

We present experimental observations of microjets formed by cavitation microbubbles. An underwater electric discharge, applied beneath a flat free surface, produces a primary compression wave, which undergoes several phase inversions upon reflections from the free surface and sparkbubble interface. The first reflection yields a tension wave, which produces a cloud of secondary cavitation bubbles in the liquid, some of which form microjets upon collapse. The tuning of these reflections enables an effective control of the microjet direction in the bubble cloud. All of the jets of the microbubbles between the spark bubble and free surface are directed radially away from the spark bubble. The mechanical response of an alumina plate placed between the electrodes and free surface generates a quasi-planar compression wave, which, following its multiple reflections from the free surface and plate, orients the microjets in the same direction toward the plate. These observations imply that the jet direction is determined mainly by the secondary compression wave, which is the first and thus most energetic compression wave acting on a sufficiently grown cavitation bubble. V C 2018 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). https://doi.org/10.1063/1.5060645 In recent years, cavitating bubbles have attracted significant attention for medical applications owing to the powerful phenomena occurring during their violent collapse and ensuing mechanical and chemical effects. The erosive power of these bubbles is mostly attributed to the shock wave emission and formation of high-speed microjets as the bubbles collapse in a nonspherical manner. Shock waves from extracorporeal lithotripters (ESWL) transfer proteins into cells, 1 which is likely attributed to the formation of small cavities produced by the tensile phase of the ESWL or by their reflections from bodily tissues. 2 Ultrasonic irradiation of cells in the presence of microbubbles enhances the membrane permeabilization; this phenomenon is referred to as sonoporation, which assists targeted cellular drug or gene delivery. [3][4][5] Many mechanisms for cell penetration by these bubbles have been suggested, one of which is the formation of directed microjets. 6 Therefore, the control of the microjet direction could promote such localized targeting.
Extensive theoretical, experimental, and numerical studies have been carried out to investigate the formation of microjets due to shock waves acting on bubbles, with pioneering work having started a few decades ago. 7,8 The directions of microjets have been controlled using a planar shock wave 9 or focused ultrasound 10 on existing bubbles. The microjet is generally oriented in the direction of the shock wave propagation. However, in confined environments, such as in-between different tissues in the body, the subsequent reflections of the pressure waves from surfaces make the control of the microjetting more complex. In particular, it is important to understand the behavior of a cloud of gas or vapor bubbles when complex interactions between pressure waves and surfaces are involved, which typically occurs in most medical applications.
In this study, we experimentally investigate the jetting in secondary microcavities produced in water by the first shock wave reflection from the free surface. In order to tune the jetting direction, the interactions between a near free surface and two types of shock waves are analyzed: one with a spherical and one with a quasi-planar propagation.
A schematic of the experimental setup is shown in Fig. 1. It consists of generation of a cavitation bubble using an electric-discharge-induced plasma in a small cuvette (25 Â 30 Â 60 mm 3 ) filled with distilled and air-saturated water. The pulsed high-voltage circuit consists of a DC power source, spark gap switch, resistances, and capacitors; it can produce a voltage of up to 30 kV. 11 The spherically propagating shock wave, denoted as primary compression wave, is generated by the vapor bubble expansion from the spark plasma, denoted as the primary bubble (potential energy at the maximum radius: $1.4 mJ). The quasi-planar wave is produced by placing a 0.2-mm-thick alumina plate above and in contact with the electrodes. The mechanism for the primary compression wave generation is different: as an electric discharge is applied, the mechanical response of the excited plate generates a quasi-planar compression wave upward into the liquid. Shadowgraphs of the pressure wave a) E-mail:sato@ifs.tohoku.ac.jp 193703 (2018) propagations and bubble dynamics are recorded with an ultra-high-speed intensified charge-coupled device (ICCD) camera (ULTRA Neo, NAC Image Technology Co. Ltd.), which could capture 12 subsequent images with a speed of up to 200 Â 10 6 frames/s and an exposure time of 5 ns, and with a CMOS camera (HPV-X, Shimadzu) capturing 256 frames with a speed of up to 10 Â 10 6 frames/s and an exposure time of 60 ns. A continuous laser is used for backlight illumination. The image is magnified up to a magnification of 300 by a large-distance microscopic objective (VHZ-50L, Keyence). The pressures of the various compression and tension waves are measured using a calibrated fiber optic probe hydrophone (FOPH 2000, RP Acoustics) with a bandwidth of up to 100 MHz and rise-time of 3 ns. The hydrophone tip is placed under the free surface of the water, above the electrodes. The optical hydrophone is connected to an oscilloscope (WaveSurfer MXs-B, Lecroy) sampling at 1 GHz. Figure 2(a) illustrates a typical interaction of the spherical primary compression wave emitted at the electric discharge with a free surface. Secondary cavitation bubbles (maximum radius range: 20-120 lm) appear in a stochastic manner between the primary bubble and free surface after the passage of the reflected tension wave [ Fig. 2(b)]. The microjets emerge during the rebound of these microbubbles; they have the same direction as that of the spherically propagating primary compression wave. Figure 2(c) visualizes such microjets and shows their directions for bubbles in different positions with respect to the center of the electrodes. An example of the life of a jetting bubble is presented in Fig. 2(d). Further reflection of the tension wave from the primary bubble interface causes another phase inversion, 12 turning the tension wave back into a compression wave, denoted as the secondary compression wave. Such a wave is visible at approximately 1.5 ls after the electric discharge in the fiber optic probe hydrophone signal in Fig. 2(e). The severe noise in the beginning of the signal is caused by the electric discharge. The directions of all of the observed microjets projected onto the 2D image plane are presented in Fig. 2(f) as angles between the jet and vertical axis as a function of the bubble position with respect to the center of the electrodes. The true jet angle includes another component corresponding to the angle with respect to the image plane, which has not been considered owing to the exact 3D locations of the bubbles being unknown. However, assuming this additional component to be negligible as the image is initially focused onto the electrodes and therefore on the bubbles above them (the depth-of-field is approximately 100 lm), all of the jets are clearly oriented radially away from the center of the primary bubble.
The generation of a quasi-planar primary compression wave by the plate is used to identify the compression wave that determines the microjet direction. The primary compression wave is also confined in-between the free surface and plate, generating multiple reflections. The interaction of such a wave with the free surface is visualized in Fig. 3(a). This compression wave is reflected from the free surface as a tension wave, which forms secondary cavitation between the plate and free surface, as shown in Fig. 3(b). The tension wave reflects from the plate keeping its phase, leading to a
Appl. Phys. Lett. 113, 193703 (2018) further growth of the secondary cavitation. It becomes a secondary compression wave after reflection from the free surface again [fourth frame in Fig. 3(a)]. It is worth noting that the direction of the secondary compression wave is now opposite to that of the primary compression wave. Microjets are observed as the secondary cavitation bubbles collapse and rebound, as shown in Figs. 3(c) and 3(d); approximately 95% of the 29 observations are directed downward, toward the plate. Figure 3(e) shows the signal of the fiber optic probe hydrophone capturing the various shock wave passages. Figure 3(f) shows the microjet directions for bubbles in different positions with respect to the center of the electrodes, as observed on the image plane. In order to exclude the possibility that the free surface and rigid boundary act as dominant microjet drivers in this case, only bubbles sufficiently far away from these surfaces are considered, i.e., s=R max > 7 (where s is the distance to the surface and R max is the maximum bubble radius), where microjets should be insignificant in the absence of other jet drivers. 13 The results suggest that the passage of the secondary compression wave determines the microjet direction, rather than the primary compression wave or later reflections. As the formation of the cavities occurs stochastically in distinct locations and with bubbles with different maximum radii, the pressure wave passages occur at different stages during the life of each individual bubble. The bubbles in the current setup are subject to the passages of multiple pressure waves, whose reflections occur at time intervals of 0.9 ls, while the bubble oscillation times are in the range of 4 to 20 ls. However, as these bubbles rebound, the majority form downward jets. Figure 4(a) shows an example of secondary cavitation bubble-jetting occurring exceptionally in the opposite direction to the secondary compression wave. The neighboring bubbles R 1 and R 2 form microjets oriented upward and downward, respectively. It should be noted that the images that only show the directions on the 2D image plane are considered, excluding the possible directions away from the plane (3D). The bubbles are located at approximately equal distances of $530 lm from the free surface and plate. For each bubble, the collapse time T C , i.e., the measured time between the instants when the bubble reaches its maximum and minimum sizes, is significantly different from the natural collapse time in the absence of pressure waves predicted by the Rayleigh model, where q is the water density, p 1 is the pressure in the liquid, and p v is the vapor pressure. 14 For R 1 and R 2 , T C =T R are 0.65 and 1.17, respectively. This can be attributed to multiple pressure waves interacting with these bubbles during their growth and collapse, as shown in Fig. 4(b), which presents the passage of each pressure wave in time. Although R 1 and R 2 have very similar maximum bubble radii, R max ¼ 65:3 lm and 63.4 lm, respectively, the measured collapse time of R 2 is almost 55% longer than that of R 1 . The bubble's radial