Poster R7 - The Effects of Frequency and Peak Negative Pressure (PNP) on Acoustic Droplet Vaporization of HC/FC Endoskeletal Drops

Introduction: Vaporizable droplets are 0.1-10 µm diameter perfluorocarbon (FC) liquid-filled, lipid-stabilized particles suspended in water. The FC droplets can be vaporized into highly echogenic microbubbles, a process known as acoustic droplet vaporization (ADV) [1]. Once activated, the acoustically active echogenic microbubbles can be imaged and used for drug delivery with ultrasound for biomedical applications [1]. A limitation of FC droplets involves balancing their colloidal stability and acoustic threshold of vaporization. Shorter FC chains vaporize more easily, but they are less colloidally stable owing to their higher fugacity. Conversely, longer FC chains have greater colloidal stability, but they require a higher acoustic intensity to vaporize, which may damage tissue. Shakya et al. engineered an alternative approach: to create endoskeletal droplets using a longer chain (n-C5F12) liquid FC droplet with a solid hydrocarbon (HC) core (Figure 1) [2]. When heated, the HC core melts and mixes with the FC phase, thereby increasing the vapor pressure and lowering the spinodal (limit of superheat), causing vaporization by homogeneous nucleation. This approach offers a unique ability to lower the acoustic threshold of vaporization, while maintaining colloidal stability of the droplets. Previously, no study has characterized the ultrasound parameters involved in the ADV of vaporizable endoskeletal droplets (VEDs) in the absence of heating (i.e., for biomedical applications). We hypothesized that VEDs would vaporize at lower frequencies (F) and higher peak negative pressures (PNP), corresponding to a higher mechanical index (MI = PNP/(F^0.5)), consistent with previous literature findings for FC droplets [3].

Materials and

Methods: Fabrication of VEDs was adapted from Shakya et al. [1]. The droplet size distribution and concentration were measured in a Multisizer 3 Coulter Counter. An agar phantom with a tubeless flow channel was made following Martinez et al. [4]. A syringe pump was used to pull the droplets through the phantom at a rate of 0.5 mL/min. An ultrasound imaging probe connected to an open-platform research ultrasound system was used to vaporize and image the droplets (Figure 2). B-mode videos (10 s) were taken of the vaporizing droplets under combinations of PNP values ranging from 0.2 to 0.8 MPa and F ranging from 3-9 MHz at both room temperature (~25 °C) or physiological temperature (37 °C). The process was repeated for HC chain lengths from n-C¬20H42 to n-C¬23H48. Using the videos, a small region of interest inside (signal) and outside (noise) the channel was assessed for mean pixel (MP) intensity in MATLAB. Using the MP values, the signal-to-noise ratio (SNR in dB) was determined as shown in Equation 1.
SNR=20log(10)*(MP_signal)/(MP_noise) (1)

A two-way ANOVA was run to each PNP and F to determine if there was a significant difference between HC core composition (row factor) or F (column factor).

Results, Conclusions, and Discussions: Figure 3 shows size distributions of the VEDs, demonstrating that a plurality were less than 10 µm diameter by number weighting (left), although there were some larger droplets detected by volume weighting (right). Figure 4 depicts representative ultrasound images of pre- and post-vaporization. Contrast enhancement following vaporization at body temperature was observed, compared to the room temperature case. Figure 5 depicts SNR vs PNP for room and body temperature at each HC core composition and F tested. From the ANOVA tests, the p-values were all less than 0.05, showing that all row and column factors were significant. We observed an increase in SNR with PNP, supporting our hypothesis. However, the SNR also increased with F, indicating that ADV is not a simple monotonic relationship with MI. HC composition appeared to affect the MI vaporization threshold, although the trend was not clear. Following Kripfgans et al. [3], we determined the PNP threshold for ADV of our VEDs. Due to the use of the same FC component, we hypothesized that the threshold would be ~0.8 MPa. Interestingly, no clear abrupt threshold was seen, rather we observed a gradual increase in SNR with increasing PNP and F. We conclude that both increasing temperature and F facilitated ADV.

Acknowledgements (Optional): [1] Borden, M. A., Shakya, G., Upadhyay, A., & Song, K. H. (2020). Acoustic nanodrops for biomedical applications. Current opinion in colloid & interface science, 50, 101383. Israelachvili, J. N. (2022). Surface forces. In The Handbook of Surface Imaging and Visualization (pp. 793-816). CRC Press.
[2] Shakya, G., Hoff, S. E., Wang, S., Heinz, H., Ding, X., & Borden, M. A. (2020). Vaporizable endoskeletal droplets via tunable interfacial melting transitions. Science Advances, 6(14), eaaz7188.
[3] Kripfgans, O. D., Fowlkes, J. B., Miller, D. L., Eldevik, O. P., & Carson, P. L. (2000). Acoustic droplet vaporization for therapeutic and diagnostic applications. Ultrasound in medicine & biology, 26(7), 1177-1189.
[4]. Martinez, P., Bottenus, N., & Borden, M. (2022). Cavitation characterization of size-isolated microbubbles in a vessel phantom using focused ultrasound. Pharmaceutics, 14(9), 1925.