Skip to main content
NCBI home page
The Journal of the Acoustical Society of America logoLink to The Journal of the Acoustical Society of America
. 2014 Dec;136(6):3422–3430. doi: 10.1121/1.4898422

The influence of distance between microbubbles on the fluid flow produced during ultrasound exposure

Carolyn E Schutt 1,a), Stuart D Ibsen 2, William Thrift 3, Sadik C Esener 3
PMCID: PMC4257971  PMID: 25480086

Abstract

The collapse dynamics of lipid monolayer-coated microbubbles in the clinically-relevant size range under 6 μm in diameter have not been studied directly due to their small size obscuring the collapse visualization. This study investigates the influence of inter-microbubble distance on the shape of lipid debris clouds created by the collapse of the microbubble destroying the microbubble lipid monolayer. The shape was highly influenced by the fluid motion that occurred as the microbubbles collapsed. It was observed that at inter-microbubble distances smaller than 37 μm the microbubbles began to interact with one another resulting in distorted and ellipsoid-shaped debris clouds. At inter-microbubble distances less than 10 μm, significantly elongated debris clouds were observed that extended out from the original microbubble location in a single direction. These distortions show a significant distance-dependent interaction between microbubbles. It was observed that microbubbles in physical contact with one another behaved in the same manner as separate microbubbles less than 10 μm apart creating significantly elongated debris clouds. It can be hypothesized that small inter-microbubble distances influence the microbubble to collapse asymmetrically resulting in the creation of fluid jets that contribute to the formation of debris fields that are elongated in a single direction.

I. INTRODUCTION

The ability of lipid-coated microbubbles to undergo inertial cavitation when exposed to focused ultrasound has made them important particles for use in a variety of therapeutic medical applications including drug delivery (Ibsen et al., 2013b; Kheirolomoom et al., 2007; Klibanov, 2006; Lentacker et al., 2006; Liu et al., 2006; Unger et al., 1998), histotripsy and high intensity focused ultrasound (Roberts et al., 2006; Yu et al., 2008), and enhancing drug extravasation from blood vessels by facilitating microcapillary damage (Bohmer et al., 2010; Caskey et al., 2011; Skyba et al., 1998; and Stieger et al., 2007). Studying how the ultrasound-induced collapse of microbubbles is influenced by nearby surfaces and adjacent microbubbles is important in order to understand these therapeutic mechanisms. The collapse of microbubbles under static pressure was first described by Lord Rayleigh (Rayleigh, 1917). Under the influence of sufficiently high peak negative pressure ultrasound the microbubbles can undergo expansion and then adiabatic collapse, allowing the microbubbles to concentrate the acoustic energy (Apfel, 1997; Kodama and Tomita, 2000; Neppiras and Noltingk, 1951; Noltingk and Neppiras, 1950). This process is known as inertial cavitation and is different from non-inertial cavitation where the microbubble oscillates in diameter with the compression and rarefaction phases of the ultrasound pulse (Flynn, 1964). Non-inertial cavitation can result in microstreaming creating fluid motion around the microbubble (Collis et al., 2010; Elder, 1959; Tho et al., 2007).

There are two main types of microbubble collapse during inertial cavitation. The first is a symmetric collapse of the microbubble with a resulting radial expansion of the shockwave (Kodama and Tomita, 2000). The second is an asymmetric collapse of the microbubble resulting in a jet of fluid (Miller et al., 1996; Postema et al., 2005). The asymmetric collapse of microbubbles concentrates the energy into a smaller volume and can project that energy for longer distances (Brujan et al., 2005).

Much of the previous work done to visualize how microbubble collapse is influenced by other nearby microbubbles has been done with bubbles that are either uncoated, or too large to pass into the microcirculation. Microbubbles that can pass through the majority of the microcirculation in medical applications need to be smaller than 6 μm in diameter (Hogg, 1987). The average size of microbubbles used medically as ultrasound contrast agents is 2.5 μm in diameter (Schneider, 1999). The microbubbles also need to be stabilized with a lipid coating because uncoated microbubbles have extremely short half-lives, making them less attractive for clinical use (Ferrara et al., 2007). These coatings can affect how the microbubbles respond to the ultrasound (van der Meer et al., 2007).

Laser pulses are often used to create uncoated microbubbles that have diameters on the millimeter scale to visualize their collapse dynamics. Two millimeter scale uncoated bubbles created in close proximity with this method were shown to asymmetrically collapse forming jets pointed toward each other (Lauterborn and Kurz, 2010). Uncoated microbubbles that are 500 μm in diameter have been observed to change their oscillation behavior as they approach one another, producing chaotic-type oscillations (Leighton, 1995). A few studies have looked at lipid-coated microbubbles in the 10–20 μm diameter range and have observed them to undergo asymmetric collapse with jet formation when exposed to ultrasound (Chen et al., 2011a; Chen et al., 2011b; Postema et al., 2005).

Most of the work done to characterize how the microbubbles interact with surfaces has also been done with large uncoated microbubbles. When laser-created bubbles in the 2–3 mm diameter range come into direct contact with a rigid surface they undergo an asymmetric collapse pointed directly at the surface. This dissipates the energy of the jet in a radial pattern across the surface causing the bubble to appear to flatten out over the rigid surface (Lauterborn and Kurz, 2010; Vogel et al., 1989). Observations of microbubble jet formation near a rigid surface have been conducted with uncoated laser-generated microbubbles as small as 100 μm in diameter (Brujan et al., 2005; Crum, 1979). These types of microbubbles have also been created near the air–water interface and observed to undergo jet formation (Wang et al., 1996). When these laser-created bubbles come into direct contact with a flexible surface, such as a gel or a vessel wall, they also have the ability to create jets that are oriented in different directions depending on the orientation of the bubble and surface (Chen et al., 2011a; Gracewski et al., 2005; Kodama and Tomita, 2000).

These types of collapse phenomena have not been directly observed with microbubbles less than 6 μm in diameter because this size range is too small to allow for reliable direct observation of an asymmetric collapse. The optical distortions created by the index of refraction difference between the gas and the surrounding water as well as the microbubbles' small size obscure the actual dynamics of collapse and involution that is the hallmark of jet formation. This makes it a challenge to directly observe the collapse dynamics and determine the fluid jet direction in a statistically significant number of microbubbles less than 6 μm in diameter.

Fluorescence imaging was used to study the fluid flow that occurred as a result of the microbubble interaction with ultrasound to overcome the limitations of direct observation at this clinically relevant size scale. A fluorescent dye was incorporated into the lipid monolayer that surrounded each microbubble. The ultrasound-induced collapse of the microbubbles fragmented the lipid monolayer into a fine debris cloud that was larger than the original microbubble and did not suffer from any optical distortions from the gas. The shape of the fluorescent debris cloud revealed details about the direction of the fluid flow. The microbubbles studied here were positioned against a glass cover slip which served as a rigid surface. Studying the shape of the debris field created by groups of microbubbles at varying separation distances can yield information about how the microbubbles influenced one another as they collapsed.

II. MATERIALS AND METHODS

A. Materials

Distearoyl phosphatidylcholine (DSPC) was purchased from Avanti Polar Lipids, Inc. (Alabaster, AL) and distearoyl phosphatidylethanolamine-methyl poly(ethylene glycol) MW5000 (mPEG-DSPE 5k) was purchased from Laysan Bio, Inc. (Arab, AL). Dulbecco's phosphate buffered saline (DPBS) was purchased from Hyclone Laboratories Inc. (Logan, UT). 3,3′-dioctadecyloxacarbocyanine, perchlorate (DiO) was purchased from Biotium, Inc. (Hayward, CA). Perfluorohexane (PFH) was purchased from Sigma-Aldrich (St. Louis, MO).

B. Microbubble fabrication

To fabricate lipid-coated microbubbles, stock solutions of lipid and lipophilic dye were prepared in chloroform at 20 mg/ml DSPC, 50 mg/ml mPEG-DSPE 5k, and 1 mM DiO. Fifty μl of the DSPC solution, 20 μl of the mPEG-DSPE 5k solution, and 30 μl of the DiO were successively added to a 4 ml glass vial while vortexing. The amount of DiO in the combined solution was 2 mol % of the total lipid and dye content. The chloroform was evaporated under an argon gas stream while the solution was under vortex. This created a lipid film along the inner surface of the vial. Five hundred μl of DPBS was added to the film and the lipids and dye were resuspended by vortexing the vial for 15 s followed by heating at 75 °C for 1 min. The cycle of vortexing and heating was repeated until the lipids were well suspended and no lipid residue was left on the vial walls. The sample was left to cool to ambient temperature.

To create the gas microbubbles, the top of the vial was first covered with parafilm to create a barrier between the gas in the vial and outside the vial. A 5 ml syringe equipped with a 22 gauge needle was used to draw up 1 ml of liquid PFH. With the plunger fully drawn back, the syringe was rotated to coat the walls with the liquid PFH and left to sit for at least 3 min to encourage vaporization of PFH into the air within the syringe creating an air/PFH vapor mixture. The syringe needle was then bent at a 130° angle into a hook shape and was inserted through the parafilm cover into the vial headspace. Next, the syringe was pumped 65 times into the vial headspace to inject the PFH/air mixture into the vial headspace. Care was taken to prevent any liquid PFH from entering the vial. An XL-2000 probe sonicator (QSonica LLC., Newtown, CT) tip was immediately inserted through the parafilm cover and positioned 1 mm below the gas/liquid interface. The probe sonicator was then operated at 25 W for 3 s to create microbubbles. The resulting bubbles were left to sit for at least 5 min before further processing.

Excess lipid and dye were removed from the microbubble sample by a washing procedure. The microbubbles was moved to a microcentrifuge tube and centrifuged at 1000 rpm for 3 min which caused the bubbles to float to the top of the liquid. The subnatant was partially removed and replaced with additional DPBS. This process was repeated 1 to 3 additional times and helped to reduce fluorescence background in the images from dye that was not associated with bubbles.

C. Ultrasound exposure

The ultrasound experiments were carried out using a modified version of a custom-designed system that combined fluorescence imaging with ultrasound (Ibsen et al., 2013a). The basic system is shown in Fig. 1. A ten gallon tank of water was used to allow coupling between the ultrasound transducer and the microbubble sample. The fluorescent microbubbles in MilliQ purified water were placed in 10 μl samples on a glass microscope slide and then covered with a glass cover slip for imaging. The glass slide was then placed at the air–water interface such that just the bottom of the slide was in contact with the water. By preventing the water level from reaching the top of the glass slide, the microbubble sample was prevented from being washed into the tank water. Positioning the microbubble sample at the air–water interface allowed a 100× oil immersion objective to collect the images. The use of a high numerical aperture oil immersion objective allowed more fluorescent light to be collected from the microbubbles in order to achieve higher frame rates and allow debris cloud resolution.

FIG. 1.

FIG. 1.

Open in a new tab

(Color online) Schematic diagram of the microscope system used to observe and record the interaction of the fluorescent microbubbles with the focused ultrasound (figure adapted from Ibsen et al., 2013a).

The contact between the bottom side of the glass slide and the water allowed the ultrasound to travel through the water and hit the bottom of the glass slide. The ultrasound intensity was attenuated by the glass but not enough to prevent microbubble cavitation from occurring in the sample.

A 3 min time delay between sample preparation and exposure to ultrasound allowed all the microbubbles to settle up against the glass cover slip due to their buoyancy. This ensured that there were no microbubbles in different focal planes and allowed all the microbubbles to be visible to the optical system.

Ultrasound was generated with a submersible 2.25 MHz transducer (V305-Su, 0.75 in. element diameter, 1 in. spherical focal length, Panametrics, Waltham, MA) using a waterproof connector cable (BCU-58 - 6 W, Panametrics, Waltham, MA). The ultrasound was focused to a 1 mm2 focal cross-sectional area allowing microbubbles within this region to be affected by the ultrasound pulse. The samples were scanned for microbubble pairs at different distances from one another. Once a group was identified it was centered into the ultrasound focal zone and hit with an ultrasound pulse that consisted of a 10-ms long 2.25-MHz sine wave with a peak negative pressure of 0.8 MPa. Inertial cavitation of microbubbles has been shown to occur at ultrasound peak negative pressure amplitudes as low as 0.58 MPa (Miller and Thomas, 1995). The interaction was recorded at 60 frames per second and saved for subsequent analysis.

The 0.8 MPa peak negative pressure does account for the attenuation of the glass slide. This was determined by placing the glass slide between the transducer and a needle hydrophone (HNP-0400 Broadband Needle Hydrophone AH - 2020–100 with 2 mm sensitive element and hydrophone pre amp, 50 kHz to 100 MHz, 0 + 20 dB, Onda Corporation, Sunnyvale, CA).

The needle hydrophone was used to measure the sound field of the transducer and find the focal zone. The hydrophone was mounted on an XYZ micromanipulator stage and the hydrophone was positioned so that the measured peak pressure from the transducer was optimized for the highest value. This was defined as the focal zone of the transducer.

D. Video analysis

The videos were analyzed using ImageJ software (version 1.46r, National Institutes of Health, Bethesda, MD) to measure the starting edge-to-edge distance of each microbubble to its nearest neighbor microbubble (“inter-bubble distance”) before ultrasound exposure. Isolated microbubbles with no other microbubbles visible in the microscope field of view were assigned the distance between the bubble edge and the edge of the field of view. The ultrasound-induced collapse of the microbubble caused the fluorescent lipid monolayer on the microbubbles to fragment, leaving behind a debris cloud. The dimensions of the fluorescent debris cloud for each microbubble were then measured and assigned to one of three categories. The first category was radial, where a circular pattern of debris was seen, resulting from conditions where the fluid jet was pointed normal to and directly toward the cover glass surface with little or no horizontal directional component. Debris clouds where there was less than a 1.3-fold difference between length (the longer dimension) and width were assigned to the radial category. The second category was ellipsoidal where a debris cloud was produced that was between 1.3 to 2 times longer than its width, resulting from a detectable horizontal directional component in the fluid jet. The third category, denoted as “elongated,” was where the debris cloud took on a longer shape such that its length was greater than 2 times the width.

E. Passive inertial cavitation detection

A passive detection method similar to that described by Chen et al. (2003) was used here to detect the occurrence of microbubble inertial cavitation in this experimental setup. The microbubbles were held inside of a Rose chamber which consisted of a circular rubber septum compressed between two glass coverslips to create a fluid space that allowed the microbubbles to be monitored both optically and acoustically as they were exposed to ultrasound (Ibsen et al., 2014). The tip of a submersible broadband needle hydrophone (HNP-0400) from Onda Corporation in connection with their AH - 2020–100 hydrophone pre amp (50 kHz to 100 MHz, 0 +20 dB) was placed inside the Rose chamber through the bottom opening created by removing the lower cover slip. No microbubbles were present. The ultrasound pulse was the same as that used in the experiments described in Sec. II C and consisted of a 10 ms 2.25 MHz sine wave with peak negative pressure of 0.8 MPa. The transducer was run for 20 separate pulses. Each pulse was recorded by the hydrophone. The frequency spectra of these 20 signals were averaged together for analysis.

A 50 μl sample of microbubbles at a concentration of 3.8 × 107 microbubbles per ml was then introduced into the Rose chamber without making any change to the setup using a hypodermic needle through the bottom opening of the Rose chamber. The microbubbles floated to the upper glass cover slip which is the same position that the microbubbles are located in the experiments described in Sec. II C. The same ultrasound pulse sequence was run 20 more times with the microbubbles being replenished after each ultrasound exposure pulse. The frequency spectra from these 20 signals were averaged together for analysis.

III. RESULTS

A. Passive inertial cavitation detection

Ultrasound insonation of the sample holder in the presence of the microbubbles increased the recorded signal strength compared to insonation in the absence of microbubbles as shown by the differential signal strength in Fig. 2. This increase in broadband frequency content is due to the broadband signal production typical of microbubble inertial cavitation. This is similar to previously published data on the passive acoustic detection of microbubbles (Chen et al., 2003) demonstrating that inertial cavitation was taking place under the experimental conditions of our study. The signal strength difference in Fig. 2 is shown to decrease at 4.5 and 6.75 MHz because the transducer produced harmonics at these frequencies with enough intensity to overwhelm the signal from the microbubble cavitation. The 2.25 MHz fundamental frequency produced by the transducer overwhelmed the microbubble cavitation signal out to 3 MHz.

FIG. 2.

FIG. 2.

Open in a new tab

(Color online) Passive detection of inertial cavitation in the experimental setup. The graph shows the relative difference in signal strength between signals recorded with and without microbubbles in the ultrasound focal zone. The observed increase in broadband frequency content in the presence of microbubbles shows that the microbubbles are undergoing inertial cavitation creating a detectable broadband acoustic signal.

B. Debris cloud observation

A series of still frames from the collected videos showing different microbubble debris cloud forms is shown in Fig. 3. Figures 3(A), 3(B), and 3(C) show two microbubbles at an edge-to-edge distance of 36 μm. These microbubbles did not appear to influence one another's collapse behavior and both created radial debris clouds. Figures 3(D), 3(E), and 3(F) show microbubbles at a starting distance of 8.0 μm. Some interaction appears to have occurred between the bubbles and distorted their fluorescent debris clouds into ellipsoidal shapes. Figures 3(G), 3(H), and 3(I) show microbubbles at a starting distance of 9.3 μm. The microbubbles appear to have interacted with each other to distort both of their debris clouds into highly elongated forms that extend well beyond the original locations of the microbubbles. It is also important to note that the debris fields extend in a single straight-line direction from the original microbubble locations.

FIG. 3.

FIG. 3.

Open in a new tab

(Color online) Still frames from videos showing examples of the three categories of debris cloud expansion. (A) Two fluorescent microbubbles are shown before ultrasound exposure with an edge-to-edge distance of 36 μm. (B) The two microbubbles did not appear to interact with each other during their collapse and the resulting debris clouds are both radial in shape as shown in (C). (D) Here the microbubbles have an edge-to-edge distance of 8.0 μm before ultrasound exposure. (E) The microbubbles appear to have interacted with one another and distorted the shape of the resulting two debris clouds to be ellipsoidal in shape as shown in (F). (G) Here the two microbubbles have an edge-to-edge distance of 9.3 μm before ultrasound exposure. (H) The microbubbles appear to have interacted with each other during collapse resulting in long distortions of both of the lipid debris clouds that extend in a single direction from the original microbubble locations. (I) The dimensions depicted indicate two elongated debris clouds with lengths greater than 2 times their heights.

Figure 4 shows three microbubbles that are in physical contact with one another before exposure to ultrasound. The resulting debris cloud took an elongated shape that extended out from the original microbubble location in a single straight-line direction.

FIG. 4.

FIG. 4.

Open in a new tab

Three microbubbles are in physical contact with one another before ultrasound exposure. Exposure to ultrasound showed a resulting elongated lipid debris cloud that extends in a single direction from the original microbubble location. All other microbubble pairs or triplets in physical contact prior to ultrasound exposure were also observed to form elongated debris clouds.

C. Analysis of debris cloud shape dependence on inter-microbubble distance and microbubble size

A total of 76 microbubbles were analyzed for debris cloud shape, microbubble size, and inter-bubble distance. The data are summarized in Table I.

TABLE I.

Summary of microbubble data acquired from video analysis.

95% Confidence interval for the mean
N Mean Std. deviation Std. error Lower Bound Upper Bound Minimum Maximum
Distance between microbubbles in μm Radial 25 55.6 41.1 8.2 38.6 72.5 17.9 169.2
Ellipsoidal 11 22.1 9.7 2.9 15.6 28.6 9.4 38.5
Elongated 40 10.7 9.7 1.5 7.6 13.7 0.0 35.9
Total 76 27.1 31.9 3.7 19.8 34.4 0.0 169.2
Microbubble diameter in μm Radial 25 2.6 0.7 0.1 2.3 2.9 1.4 4.9
Ellipsoidal 11 2.7 1.0 0.3 2.0 3.4 1.9 5.7
Elongated 40 4.2 2.0 0.3 3.6 4.9 1.4 11.4
Total 76 3.5 1.7 0.2 3.1 3.9 1.4 11.4
Open in a new tab

Box plots of the inter-microbubble distances for the three different debris cloud shape categories are shown in Fig. 5. The data were not normally distributed so a two-tailed nonparametric one-way analysis of variance by ranks Kruskal-Wallis Test was performed at the 0.05 significance level. A significant difference was found in inter-microbubble distance between the three debris cloud shape categories (p < 0.001). Post hoc comparisons using the Dunn-Sidak adjustment to counteract the problem of multiple comparisons showed a significant difference between the inter-microbubble distances that resulted in radial debris clouds and those that resulted in elongated debris clouds (p < 0.001). There was also a significant difference between the distances that resulted in ellipsoidal debris clouds and elongated debris clouds (p = 0.041). The difference between the microbubble distances that resulted in radial debris clouds and those that resulted in ellipsoidal debris clouds was not significantly different (p = 0.067). All statistical calculations were performed using IBM® SPSS® Statistics Version 21 software (Armonk, New York).

FIG. 5.

FIG. 5.

Open in a new tab

(Color online) Box plots of the distance between the microbubble and its nearest neighbor in microns for each of the three debris cloud expansion categories. Whiskers denote maxima and minima and the medians are represented as horizontal lines. The small circles denote data points that are statistical outliers in the data set which were defined as any points that were two standard deviations away from the mean. The radial and elongated debris cloud categories and the ellipsoidal and elongated categories were found to be statistically significantly different from each other (p < 0.001 and p = 0.041, respectively) with regard to inter-microbubble distance. The difference in inter-bubble distances resulting in radial and ellipsoidal debris clouds was not statistically significant (p = 0.067).

Box plots of the diameter of the microbubbles in the three different debris cloud expansion categories are shown in Fig. 6. There was no significant difference between the microbubble diameters of bubbles that produced radial and ellipsoidal debris clouds (p = 1) using the same statistical test described for the above comparisons of inter-bubble distance. There was a significant difference between the diameters of bubbles that produced radial debris clouds and those that resulted in elongated debris clouds (p < 0.001). This difference was likely a result of the larger microbubbles being generally closer to one another than the smaller bubbles in the study population, perhaps caused by how these bubbles settled at the liquid-coverslip interface. There was also a significant difference between the diameters of bubbles that produced ellipsoidal debris clouds and elongated debris clouds (p = 0.009) which was likely also caused by the closer inter-bubble distance seen with larger microbubbles.

FIG. 6.

FIG. 6.

Open in a new tab

(Color online) Box plots for the diameter of the microbubbles in each of the three debris cloud expansion categories. Whiskers denote maxima and minima and the medians are represented as horizontal lines. The small circles denote data points that are statistical outliers in the data set which were defined as any points that were two standard deviations away from the mean.

A scatter plot of the diameter of the microbubble versus the inter-microbubble distance is shown in Fig. 7 and is color-coded to the three different debris cloud categories. A clear pattern can be seen where radial debris cloud expansion occurs at the larger inter-microbubble distances and ellipsoidal and elongated debris clouds occur only at the smaller distances. Ellipsoid formation was observed to begin occurring at inter-microbubble distances of 37 μm. Below 10 μm in inter-microbubble distance, bubbles were seen to exclusively produce elongated debris clouds.

FIG. 7.

FIG. 7.

Open in a new tab

(Color online) Scatter plot showing the diameters of the microbubbles plotted against distance between the microbubble and its nearest neighbor. Points are color-coded to the shape of the debris cloud. A trend is observed where radial expansion is seen at larger inter-microbubble distances and ellipsoidal and elongated debris cloud formation are seen at the shorter distances.

IV. DISCUSSION

The morphology of the debris cloud was significantly influenced by the inter-microbubble distance as can be seen in Figs. 3 and 4. When the microbubbles were more than 37 μm apart the resulting debris clouds were circular. The shape of the debris cloud began to distort at inter-microbubble distances shorter than 37 μm resulting in ellipsoidal debris clouds. At an inter-microbubble distance of around 25 μm the debris clouds began to have elongated forms. Microbubbles that had less than 10 μm of separation or were in physical contact with one another prior to ultrasound exposure were always seen to result in elongated debris clouds. This supports the conclusion that the microbubbles interacted with other nearby bubbles to influence the fluid flow.

The formation of elongated and ellipsoidal debris clouds did not appear to be influenced by the size of the microbubble below 4 μm in diameter. Above 4 μm in diameter the microbubbles all had inter-microbubble distances below 30 μm except for one microbubble that was isolated and had a radial debris cloud. The population subset of microbubbles below 4 μm in diameter shows a statistically significant difference in inter-microbubble distance between those bubbles that formed elongated debris clouds and those that had radial debris clouds (p < 0.001) using the same statistical test as described above for comparisons of inter-microbubble distance. However, for this same population of bubbles below 4 μm, there was no statistically significant difference in bubble size among all three groups (p = 0.169). This supports the conclusion that for microbubbles less than 4 μm in diameter, which is the relevant size range for clinically-used microbubbles, the dominant factor for the formation of elongated debris clouds was the inter-microbubble distance and not their size.

Two types of interactions were likely occurring between microbubbles in this study when ensonified with ultrasound. The first was the influence of microstreaming fluid flow around each microbubble that occurred as the microbubbles started to oscillate just before collapse. The second were the secondary Bjerknes forces that caused microbubble attraction. This microbubble attraction is shown in Figs. 3(D) and 3(E). In Fig. 3(D) the microbubbles start at a center-to-center distance of 11.0 μm but the resulting debris clouds in Fig. 3(E) have a center-to-center distance of 6.5 μm indicating that the microbubbles attracted each other before collapse. If the microbubbles were attracted to one another just before their collapse then the microstreaming occurring around each microbubble could also influence the individual microbubble oscillations. This influence would grow in intensity as the inter-microbubble distance decreased.

The elongated debris clouds were observed to project in a single direction from the original microbubble location. Work done by Tho et al. (2007) has shown the fluid flow patterns created by microstreaming in single and multi-bubble groups. Bubbles imaged in those experiments were not seen to create flow patterns that project in a single straight-line direction. The flow patterns were always balanced by a stream moving in the opposite direction or displayed a form that flared out at wide angles creating circular eddy current type formations (Tho et al., 2007). For example, a stream moving away from the bubble to the right would be balanced by a stream moving away from the bubble to the left.

The observation that the debris fields created by the microbubbles in our experiments extend in a single direction from the original microbubble location without a balanced component in the opposite direction indicates that another contributing factor other than microstreaming was involved in shaping the debris cloud. One hypothesis is that this extra component could be created by an asymmetric collapse of the microbubble. Asymmetric collapse creates fluid jets that result in a single-direction straight line fluid flow that is not balanced by a diametrically opposed jet. This collapse behavior would be similar to what is seen in the literature with laser generated bubbles that were close to one another (Lauterborn and Kurz, 2010). The results of the passive acoustic detection showed that inertial cavitation of microbubbles was occurring in the experimental setup under these conditions which indicates that the microbubbles could be producing fluid jets.

Bubbles will jet toward a rigid surface, flattening out along the surface in a circular expansion (Lauterborn and Kurz, 2010; Vogel et al., 1989). This could account for the circular debris fields created by the microbubbles when separated by greater than 37 μm. As the inter-microbubble distance decreased, the microbubbles could have begun to influence one another, causing the asymmetric collapse to not just point at the glass slide, but also have a component that pointed parallel along the glass. This is similar to what has been observed previously with two laser-generated microbubbles in close proximity that are seen to influence each other and cause asymmetric collapse of both bubbles (Lauterborn and Kurz, 2010). In the case of our study, the jetting direction would be influenced by both the glass slide and by the neighboring microbubble creating horizontal components to what would usually be a perpendicular jet direction toward the rigid surface. Having these horizontal components in the direction of the asymmetric collapse could account for the elongated debris cloud expansion.

The therapeutic consequences of the inter-microbubble distance effect on fluid flow during microbubble collapse will be studied in future experiments. Of particular interest is understanding how this distance-dependent fluid flow influences the adhesion of cells to glass cover slips, and how it influences microcapillary damage. The influence of this effect can also be studied for sonoporation (Bao et al., 1997; Li et al., 2011; Zarnitsyn et al., 2008) and can also have implication for histotripsy and high intensity focused ultrasound tissue damage.

V. CONCLUSIONS

The starting distance between microbubbles prior to ultrasound exposure had a significant effect on the shape of the resulting lipid debris cloud. In our experimental configuration, inter-microbubble distances greater than 37 μm resulted in circular debris fields. This indicates that at this distance the microbubbles were not interacting with one another during collapse. At inter-microbubble distances less than 37 μm microbubble microstreaming and secondary Bjerknes forces likely began to influence and distort nearby microbubble oscillations resulting in ellipsoid-shaped debris clouds. At inter-microbubble distances less than 10 μm significant elongation of the debris clouds was observed in every instance, indicating a strong interaction with nearby bubbles. Microbubbles that were in physical contact with one another exclusively formed these elongated debris clouds. One hypothesis for the creation of the debris clouds that were elongated in a single direction is the asymmetric collapse of the microbubbles resulting in fluid jets.

ACKNOWLEDGMENTS

Support was provided by Grant Nos. T32 CA121938 and R25 CA153915 from the National Cancer Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute or the National Institutes of Health.

References

  • 1.Apfel, R. E. (1997). “ Sonic effervescence: A tutorial on acoustic cavitation,” J. Acoust. Soc. Am. 101(3), 1227–1237 10.1121/1.418130 [DOI] [Google Scholar]
  • 2.Bao, S., Thrall, B. D., and Miller, D. L. (1997). “ Transfection of a reporter plasmid into cultured cells by sonoporation in vitro,” Ultrasound Med. Biol. 23(6), 953–959 10.1016/S0301-5629(97)00025-2 [DOI] [PubMed] [Google Scholar]
  • 3.Bohmer, M. R., Chlon, C. H. T., Raju, B. I., Chin, C. T., Shevchenko, T., and Klibanov, A. L. (2010). “ Focused ultrasound and microbubbles for enhanced extravasation,” J. Controlled Release 148(1), 18–24 10.1016/j.jconrel.2010.06.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Brujan, E. A., Ikeda, T., and Matsumoto, Y. (2005). “ Jet formation and shock wave emission during collapse of ultrasound-induced cavitation bubbles and their role in the therapeutic applications of high-intensity focused ultrasound,” Phys. Med. Biol. 50(20), 4797–4809 10.1088/0031-9155/50/20/004 [DOI] [PubMed] [Google Scholar]
  • 5.Caskey, C. F., Hu, X., and Ferrara, K. W. (2011). “ Leveraging the power of ultrasound for therapeutic design and optimization,” J. Controlled Release 156(3), 297–306 10.1016/j.jconrel.2011.07.032 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Chen, H., Brayman, A. A., Kreider, W., Bailey, M. R., and Matula, T. J. (2011a). “ Observations of translation and jetting of ultrasound-activated microbubbles in mesenteric microvessels,” Ultrasound Med. Biol. 37(12), 2139–2148 10.1016/j.ultrasmedbio.2011.09.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Chen, H., Kreider, W., Brayman, A. A., Bailey, M. R., and Matula, T. J. (2011b). “ Blood vessel deformations on microsecond time scales by ultrasonic cavitation,” Phys. Rev. Lett. 106(3), 034301. 10.1103/PhysRevLett.106.034301 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Chen, W.-S., Matula, T. J., Brayman, A. A., and Crum, L. A. (2003). “ A comparison of the fragmentation thresholds and inertial cavitation doses of different ultrasound contrast agents,” J. Acoust. Soc. Am. 113(1), 643–651 10.1121/1.1529667 [DOI] [PubMed] [Google Scholar]
  • 9.Collis, J., Manasseh, R., Liovic, P., Tho, P., Ooi, A., Petkovic-Duran, K., and Zhu, Y. (2010). “ Cavitation microstreaming and stress fields created by microbubbles,” Ultrasonics 50(2), 273–279 10.1016/j.ultras.2009.10.002 [DOI] [PubMed] [Google Scholar]
  • 10.Crum, L. A. (1979). “ Surface oscillations and jet development in pulsating bubbles,” J. Phys. (Paris) 40(11), C8-285–C8-288 10.1051/jphyscol:1979849 [DOI] [Google Scholar]
  • 11.Elder, S. A. (1959). “ Cavitation microstreaming,” J. Acoust. Soc. Am. 31(1), 54–64 10.1121/1.1907611 [DOI] [Google Scholar]
  • 12.Ferrara, K., Pollard, R., and Borden, M. (2007). “ Ultrasound microbubble contrast agents: Fundamentals and application to gene and drug delivery,” Annu. Rev. Biomed. Eng. 9, 415–447 10.1146/annurev.bioeng.8.061505.095852 [DOI] [PubMed] [Google Scholar]
  • 13.Flynn, H. G. (1964). “ Physics of acoustic cavitation in liquids,” in Physical Acoustics, edited by Mason W. P. ( Academic Press, New York: ), Vol. 1, Pt. B, pp. 57–172. [Google Scholar]
  • 14.Gracewski, S. M., Miao, H., and Dalecki, D. (2005). “ Ultrasonic excitation of a bubble near a rigid or deformable sphere: Implications for ultrasonically induced hemolysis,” J. Acoust. Soc. Am. 117(3), 1440–1447 10.1121/1.1858211 [DOI] [PubMed] [Google Scholar]
  • 15.Hogg, J. C. (1987). “ Neutrophil kinetics and lung injury,” Physiol. Rev. 67(4), 1249–1295. [DOI] [PubMed] [Google Scholar]
  • 16.Ibsen, S., Benchimol, M., and Esener, S. (2013a). “ Fluorescent microscope system to monitor real-time interactions between focused ultrasound, echogenic drug delivery vehicles, and live cell membranes,” Ultrasonics 53(1), 178–184 10.1016/j.ultras.2012.05.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Ibsen, S., Schutt, C. E., and Esener, S. (2013b). “ Microbubble-mediated ultrasound therapy: A review of its potential in cancer treatment,” Drug Des., Dev. Ther. 7, 1–15 10.2147/DDDT.S31564 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Ibsen, S., Shi, G., Schutt, C., Shi, L., Suico, K.-D., Benchimol, M., Serra, V., Simberg, D., Berns, M., and Esener, S. (2014). “ The behavior of lipid debris left on cell surfaces from microbubble based ultrasound molecular imaging,” Ultrasonics. 54(8), 2090–2098 10.1016/j.ultras.2014.06.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Kheirolomoom, A., Dayton, P., Lum, A., Little, E., Paoli, E., Zheng, H., and Ferrara, K. (2007). “ Acoustically-active microbubbles conjugated to liposomes: Characterization of a proposed drug delivery vehicle,” J. Controlled Rel. 118, 275–284 10.1016/j.jconrel.2006.12.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Klibanov, A. (2006). “ Microbubble contrast agents targeted ultrasound imaging and ultrasound assisted drug-delivery applications,” Invest. Radiol. 41(3), 354–362 10.1097/01.rli.0000199292.88189.0f [DOI] [PubMed] [Google Scholar]
  • 21.Kodama, T., and Tomita, Y. (2000). “ Cavitation bubble behavior and bubble shock wave interaction near a gelatin surface as a study of in vivo bubble dynamics,” Appl. Phys. B 70(1), 139–149 10.1007/s003400050022 [DOI] [Google Scholar]
  • 22.Lauterborn, W., and Kurz, T. (2010). “ Physics of bubble oscillations,” Rep. Prog. Phys. 73(10), 106501. 10.1088/0034-4885/73/10/106501 [DOI] [Google Scholar]
  • 23.Leighton, T. G. (1995). “ Bubble population phenomena in acoustic cavitation,” Ultrason. Sonochem. 2(2), S123–S136 10.1016/1350-4177(95)00021-W [DOI] [Google Scholar]
  • 24.Lentacker, I., Geest, B. G. D., Vandenbroucke, R. E., Peeters, L., Demeester, J., Smedt, S. C. D., and Sanders, N. N. (2006). “ Ultrasound-responsive polymer-coated microbubbles that bind and protect DNA,” Langmuir 22(17), 7273–7278 10.1021/la0603828 [DOI] [PubMed] [Google Scholar]
  • 25.Li, Z. G., Luo, K. Q., Ohl, C. D., Zhang, J. B., and Liu, A. Q. (2011). “ A single-cell membrane dynamic from poration to restoration by bubble-induced jetting flow,” in 15th International Conference on Miniaturized Systems for Chemistry and Life Sciences, pp. 94–96. [Google Scholar]
  • 26.Liu, Y., Miyoshi, H., and Nakamura, M. (2006). “ Encapsulated ultrasound microbubbles: Therapeutic application in drug/gene delivery,” J. Controlled Release 114, 89–99 10.1016/j.jconrel.2006.05.018 [DOI] [PubMed] [Google Scholar]
  • 27.Miller, D. L., and Thomas, R. M. (1995). “ Ultrasound contrast agents nucleate inertial cavitation in vitro,” Ultrasound Med. Biol. 21(8), 1059–1065 10.1016/0301-5629(95)93252-U [DOI] [PubMed] [Google Scholar]
  • 28.Miller, M. W., Miller, D. L., and Brayman, A. A. (1996). “ A review of in vitro bioeffects of inertial ultrasonic cavitation from a mechanistic perspective,” Ultrasound Med. Biol. 22(9), 1131–1154 10.1016/S0301-5629(96)00089-0 [DOI] [PubMed] [Google Scholar]
  • 29.Neppiras, E. A., and Noltingk, B. E. (1951). “ Cavitation produced by ultrasonics: Theoretical conditions for the onset of cavitation,” Proc. Phys. Soc., London, Sect. B 64(12), 1032–1038 10.1088/0370-1301/64/12/302 [DOI] [Google Scholar]
  • 30.Noltingk, B. E., and Neppiras, E. A. (1950). “ Cavitation produced by ultrasonics,” Proc. Phys. Soc., London, Sect. B 63(9), 674–685 10.1088/0370-1301/63/9/305 [DOI] [Google Scholar]
  • 31.Postema, M., van Wamel, A., Folkert, J., and de Jong, N. (2005). “ High-speed photography during ultrasound illustrates potential therapeutic applications of microbubbles,” Med. Phys. 32, 3707–3711 10.1118/1.2133718 [DOI] [PubMed] [Google Scholar]
  • 32.Rayleigh, L. VIII. (1917). “ On the pressure developed in a liquid during the collapse of a spherical cavity,” The London, Edinburgh, and Dublin Philos. Mag. J. Sci. 34(200), 94–98 10.1080/14786440808635681 [DOI] [Google Scholar]
  • 33.Roberts, W. W., Hall, T. L., Ives, K., Wolf, J. S., Jr., Fowlkes, J. B., and Cain, C. A. (2006). “ Pulsed cavitational ultrasound: A noninvasive technology for controlled tissue ablation (histotripsy) in the rabbit kidney,” J. Urol. 175(2), 734–738 10.1016/S0022-5347(05)00141-2 [DOI] [PubMed] [Google Scholar]
  • 34.Schneider, M. (1999). “ Characteristics of SonoVueTM,” Echocardiography 16(s1), 743–746 10.1111/j.1540-8175.1999.tb00144.x [DOI] [PubMed] [Google Scholar]
  • 35.Skyba, D. M., Price, R. J., Linka, A. Z., Skalak, T. C., and Kaul, S. (1998). “ Direct in vivo visualization of intravascular destruction of microbubbles by ultrasound and its local effects on tissue,” Circulation 98(4), 290–293 10.1161/01.CIR.98.4.290 [DOI] [PubMed] [Google Scholar]
  • 36.Stieger, S. M., Caskey, C. F., Adamson, R. H., Qin, S., Curry, F.-R. E., Wisner, E. R., and Ferrara, K. W. (2007). “ Enhancement of vascular permeability with low-frequency contrast-enhanced ultrasound in the chorioallantoic membrane model 1,” Radiology 243(1), 112–121 10.1148/radiol.2431060167 [DOI] [PubMed] [Google Scholar]
  • 37.Tho, P., Manasseh, R., and Ooi, A. (2007). “ Cavitation microstreaming patterns in single and multiple bubble systems,” J. Fluid Mech. 576, 191–233 10.1017/S0022112006004393 [DOI] [Google Scholar]
  • 38.Unger, E., McCreery, T., Sweitzer, R., Caldwell, V., and Wu, Y. (1998). “ Acoustically active lipospheres containing paclitaxel: A new therapeutic ultrasound contrast agent,” Invest. Radiol. 33(12), 886–892 10.1097/00004424-199812000-00007 [DOI] [PubMed] [Google Scholar]
  • 39.van der Meer, S. M., Dollet, B., Voormolen, M. M., Chin, C. T., Bouakaz, A., de Jong, N., Versluis, M., and Lohse, D. (2007). “ Microbubble spectroscopy of ultrasound contrast agents,” J. Acoust. Soc. Am. 121, 648–656 10.1121/1.2390673 [DOI] [PubMed] [Google Scholar]
  • 40.Vogel, A., Lauterborn, W., and Timm, R. (1989). “ Optical and acoustic investigations of the dynamics of laser-produced cavitation bubbles near a solid boundary,” J. Fluid Mech. 206(1), 299–338 10.1017/S0022112089002314 [DOI] [Google Scholar]
  • 41.Wang, Q. X., Yeo, K. S., Khoo, B. C., and Lam, K. Y. (1996). “ Nonlinear interaction between gas bubble and free surface,” Comput. Fluids 25(7), 607–628 10.1016/0045-7930(96)00007-2 [DOI] [Google Scholar]
  • 42.Yu, T., Hu, D., and Xu, C. (2008). “ Microbubbles improve the ablation efficiency of extracorporeal high intensity focused ultrasound against kidney tissues,” World J. Urol. 26(6), 631–636 10.1007/s00345-008-0290-z [DOI] [PubMed] [Google Scholar]
  • 43.Zarnitsyn, V., Rostad, C., and Prausnitz, M. (2008). “ Modeling transmembrane transport through cell membrane wounds created by acoustic cavitation,” Biophys. J. 95(9), 4124–4138 10.1529/biophysj.108.131664 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The Journal of the Acoustical Society of America are provided here courtesy of Acoustical Society of America

RESOURCES