Acoustic microbubble propulsion, train-like assembly and cargo transport

The study addresses how to achieve controlled mobility, trapping, and transport of micro- and nanoparticles in viscous and gel-like media using acoustics. Prior manipulation approaches typically rely on external field gradients (optical, magnetic, electric, acoustic) or nonreciprocal motion enabled by microstructures and are mostly demonstrated in simple Newtonian fluids like water. Navigating complex, gel-like fluids remains challenging for artificial microswimmers despite biological analogs doing so effectively. There is limited prior work on acoustic manipulation in gels and on enabling propulsion and cargo handling under such constraints. The purpose here is to discover and characterize acoustic-field-driven behaviors of microbubbles confined in a shallow slit within a shear-thinning gel, and to demonstrate propulsion, self-assembly into trains, and cargo trapping/transport, thereby opening avenues for biotechnologies and life-science applications in viscous environments.

The paper situates its contribution among techniques that manipulate microparticles via external gradient fields (optical tweezers; magnetic, electric, and acoustic methods) and nonreciprocal propulsion strategies (light, electric, magnetic actuation). Most artificial microswimmers operate in water; only a few designs navigate viscous or gel-like media (e.g., magnetic micro-scallop, swimmers in bodily fluids and vitreous humor). Acoustics has enabled trapping/manipulation in gels (e.g., acoustic vortex trapping of bubbles in agarose) but controlled propulsion, train-like assembly, and cargo handling in viscous gels are largely unexplored. The work also references classic and recent theories of acoustic radiation forces (Bjerknes forces), bubble dynamics (Rayleigh–Plesset), shape modes, and standing-wave interactions as foundations for the observed microbubble behavior.

Experimental platform: A piezoelectric disc transducer bonded to a glass slide is driven by a function generator (square wave, ~20 Vp, frequencies ~22.3–23 kHz). A shear-thinning aqueous KY gel (water/glycerol; viscosity 0.5–100 Pa·s range discussed; density ~1000–1056 kg/m³; sound speed ~1450–1500 m/s) is applied to the slide. A glass capillary (outer diameter 1.5 mm; inner diameter 1.3 mm) is pressed atop the gel, creating a narrow slit (gap on order 10–50 µm) between the capillary and slide. The setup is mounted on an inverted microscope with high-speed imaging (ZEISS Axiovert 200 M; up to 40,420 fps). High-speed imaging synchronization: with ~22.3 kHz acoustic excitation (period ~45 µs), frames at 40,420 fps (25 µs exposure) are chronologically rearranged to reconstruct oscillation cycles at an effective 220,100 fps.

Theoretical modeling: The glass slide and capillary are approximated as two closely spaced cylinders in a fluid. The larger cylinder (slide; radius → ∞) vibrates radially, generating an acoustic field; the smaller cylinder (capillary; radius R) scatters it. Acoustic pressure is obtained via a velocity potential satisfying the Helmholtz equation, with boundary conditions at cylinder surfaces yielding pressure distributions that show strong amplification in the interstice (normalized maxima at the mid-gap, z=0). The viscous boundary layer δ ≈ (2ν/ω)^{1/2} ≈ 6 µm is small and does not significantly affect the scattered field. Laser doppler vibrometry informs pressure estimates (~100 kPa at the slit). Radiation forces on bubbles are computed by coupling the acoustic pressure field to bubble dynamics (Rayleigh–Plesset for instantaneous radius) to estimate time-averaged forces and components along axes in the slit.

Procedures and observations: Ultrasound activation nucleates microbubbles at the gas–liquid interface within the slit; bubbles of varying sizes form (smallest near center). Under continuous exposure, bubbles grow, become elliptical, and merge into aligned trains. When ultrasound is turned off, bubbles revert to spherical and exit laterally; reactivation drives them back into the slit where they rapidly flatten (discoidal) and trap. Propulsion experiments use single and multiple ellipsoidal microbubbles within the slit, tracking motion along the y-direction. Cargo experiments mix 10 µm polystyrene beads (and yeast cells) with the gel, then observe trapping between adjacent bubbles in a moving train and release upon deactivation.

Data acquisition: Imaging sequences quantify oscillation amplitudes (~10 µm volume-mode oscillation) and surface capillary waves (shape modes), translational velocities, inter-bubble spacing, and cargo trajectories/approach speeds as functions of acoustic parameters (frequency bands near transducer resonances; voltages ~16.8–40 Vpp).

• Acoustic focusing in a narrow slit between glass boundaries amplifies pressure (~100 kPa estimated), nucleating and cavitating microbubbles at the slit.
• Microbubbles exhibit reversible shape transformations: spherical when field is off; discoidal/ellipsoidal within milliseconds upon reactivation, enabling trapping in the slit.
• Controlled propulsion of single ellipsoidal microbubbles occurs along the slit (y-direction), driven by superposed volume oscillations (~10 µm amplitude) and high-amplitude surface (shape) modes; confinement prevents lateral deviation.
• A standing wave with wavelength ~5.6 cm is predicted in a 20 µm gap at 23 kHz; classic Bjerknes force alone cannot explain propulsion—interaction of shape modes generates additional radiation forces that direct motion away from the velocity node toward capillary ends.
• Microbubbles self-assemble into train-like chains via secondary Bjerknes interactions, then translate in unison at uniform velocity. Reported speeds reach up to ~0.6 mm/s.
• Train composition: bubble count typically 5–6 across tested voltages (20–40 Vpp at 2.6 kHz) and frequencies (near resonances), showing no strict monotonic dependence on voltage/frequency. Inter-bubble spacing decreases as voltage increases; a strong short-range repulsion maintains ~10 µm gaps preventing coalescence. Inner bubbles become more rectangular and oscillate less than leading/trailing bubbles.
• Robust cargo trapping: 10 µm polystyrene beads and yeast cells are attracted by bubble oscillations, slide along bubble surfaces to the equator, and become trapped at midpoints between adjacent bubbles. Approach velocity scales inversely with distance to the bubble; minimal oscillations (~2 µm) at trap sites aid stability. Cargo is transported with the train and reliably released when ultrasound is turned off.
• The platform operates in shear-thinning gels (e.g., KY gel; viscosity 0.5–100 Pa·s), suggesting applicability to biological gels (sputum, mucus).

The findings demonstrate that acoustic wave concentration in confined slits can be harnessed to nucleate, trap, propel, and self-assemble microbubbles in viscous, shear-thinning media. The propulsion mechanism relies on the interplay of volume and shape oscillation modes, which alters net radiation forces relative to classical weak-field Bjerknes predictions, enabling directed translation along the slit. Once assembled, bubble trains maintain stable inter-bubble spacing via repulsive interactions, forming repeatable traps at the midpoints between neighbors. These traps capture a variety of microparticles and cells and can transport and release them on demand by toggling the acoustic field. The approach provides a reagent-free, low-cost route for parallel particle manipulation in complex fluids, potentially enabling enrichment, separation, and spatial patterning tasks in microfluidics and biomedicine. The work addresses the challenge of micromanipulation in viscous gels where many artificial swimmers fail, and suggests that acoustic microbubble trains can serve as microscale cargo trains for positioning bioparticles. Open questions include determining precise control over propulsion direction and understanding dynamics in more complex slit geometries.

This work introduces an acoustofluidic platform that concentrates acoustic energy in a narrow slit to nucleate and trap microbubbles, enabling their controlled propulsion, spontaneous train-like self-assembly, and reliable trapping/transport of microparticles and cells in viscous, shear-thinning gels. Key contributions include: (i) experimental and theoretical evidence of acoustic amplification in slits and associated bubble nucleation; (ii) discovery of shape-mode-assisted propulsion of ellipsoidal microbubbles under confinement; (iii) demonstration of stable bubble trains that act as microscale cargo carriers to capture, transport, and release particles on demand. Potential applications span particle enrichment/separation, biomarker extraction from viscous fluids (sputum, mucus), targeted cell introduction/patterning in ECM-like gels, and studies of chemotaxis at single-cell resolution. Future research should focus on precise control of propulsion directionality, modeling radiation forces for discoidal bubbles, and extending the approach to arbitrary slit geometries and integrated microfluidic architectures.

• The radiation force theory used assumes spherical bubbles; no complete theory currently predicts forces for discoidal bubbles, so spherical approximations are employed for qualitative insight.
• Theoretical geometry simplifies the setup to two cylinders (capillary and effectively infinite-radius slide), which may not capture all complexities of the actual slit and boundary conditions.
• Control over the direction of propulsion remains incompletely understood and is identified for future study.
• Some reported parameter values (e.g., material properties, boundary layer effects) are approximations; detailed viscoelastic properties of gels and their influence on bubble dynamics are not fully resolved.
• Increased acoustic power can destabilize bubbles (erratic motion, merging), indicating a limited operational window for stable propulsion and train formation.