Researchers have built what they describe as an acoustic tractor beam that uses sound waves, not light or static electricity, to remotely rewrite how stiff or soft a material feels from the inside. The breakthrough, published in Nature Communications in March 2026, points toward adaptive structures, shape-shifting soft robots, and medical implants whose rigidity can be tuned on demand.
The new study is part of a small wave of acoustic manipulation advances in 2026. Just five months later, a separate team reported a complementary result in Physical Review Letters, showing that sound can also levitate and translate physical objects in midair at roughly six times the working distance of any previous single-sided acoustic trap. Taken together, the two papers suggest that sound-based tractor beam technology is moving from clever lab demos toward a practical engineering toolkit.
The stiffness-reprogramming work was co-led by the University of California San Diego, the University of Michigan, and the French National Center for Scientific Research (CNRS) at the Laboratory of Acoustics of Le Mans University, which showed for the first time that specific frequencies of acoustic waves can reliably move localized features called mechanical kinks, tiny internal boundaries that decide whether a given region of a material is soft or stiff.
“The idea here is that we’ve essentially made an acoustic tractor beam that moves a kink and changes the way a material feels, while creating gradients of stiffness, on demand,” said co-corresponding author Nicholas Boechler, a professor in the Department of Mechanical and Aerospace Engineering at the UC San Diego Jacobs School of Engineering.
The work builds on a long line of tractor beam research. Earlier efforts on the technology have largely focused on moving physical objects, including optical tractor beams that tug cells and DNA and electrostatic systems designed to clear orbital debris. The new study instead aims the beam at the internal structure of a material itself.
What the Researchers Actually Did
Mechanical kinks are the boundaries between two distinct internal states of a material. On either side of a kink, the material may be made of the same atoms or building blocks, but those blocks are oriented differently in three dimensions. That subtle reorientation produces very different mechanical properties, so where a kink sits directly controls where a material feels soft and where it feels stiff.
Kinks show up in many places in nature. They mark where metals permanently bend and where DNA strands separate. Materials scientists have long wanted to control them, but two stubborn problems have stood in the way. In most materials, kinks are pinned in place by internal energy barriers, and earlier attempts to manipulate them acoustically produced chaotic, unpredictable trajectories.
The new study solves both problems at once. The researchers designed a topological metamaterial in which moving the kink costs essentially no energy, a rare property achieved by engineering a structure whose behavior is dictated by its geometry rather than by the chemistry of its building blocks. With no energy barriers in the way, acoustic waves can pull the kink predictably and step by step.
How an Acoustic Tractor Beam Moves a Kink
“We showed that if you send acoustic waves in from one side, they actually pull the kink toward where the sound came from,” Boechler said. “You can send a small pulse, and the kink moves a little. Send another pulse, and it moves a little more. It’s basically remote control for the material’s internal state.”
To demonstrate, the team built a life-sized experimental model: a chain of stacked, rotating disks connected by springs, in which each disk plays the role of an atom and the springs mimic atomic bonds. One disk, rotated differently from the rest, represents the kink.
- Short acoustic pulses pull the kink toward the sound source, moving it a few disks at a time.
- Each additional short burst nudges the kink a little farther.
- Longer, continuous vibrations pull the kink across the entire chain, effectively flipping which side of the structure is soft and which is stiff.
- Only certain sound frequencies cause the kink to move. Others pass through with no effect.
Computer simulations also revealed that when an acoustic wave packet reaches the kink, part of the wave reflects and part passes through. Even so, the interaction transfers momentum to the kink and keeps it moving, an important detail for any future device that needs precise, repeatable control.
Why the Stiffness Profile Flips
The team designed the model so that wherever the kink sits, the material around it is soft, while the rest of the structure grows progressively stiffer. Moving the kink changes the entire stiffness map.
Park the kink at one end of the chain, and that end becomes soft while stiffness ramps up exponentially toward the opposite end. Move it to the middle, and the material turns soft in the center and stiff at both ends. Send a long pulse, and the stiffness profile flips end to end. The effect is a continuously tunable, position-controlled gradient, set by nothing more than the timing and frequency of sound pulses from outside the material.
Why This Is Different From Past Tractor Beams
Previous acoustic levitation work has shown that sound can hold or move small physical objects. Earlier research even produced force fields inside fluids using sound waves. The March 2026 study goes a step further. The acoustic beam is not moving a bead, a cell, or a particle. It is reaching inside a solid metamaterial and reorganizing its internal state without any physical contact, cuts, or wiring.
“This opens the door to future technologies where you can remotely tune configurations and functionalities deep inside a material without cutting it open,” said co-corresponding author Xiaoming Mao, a professor of physics and leader of the University of Michigan cohort.
Co-corresponding author Georgios Theocharis of CNRS at the Laboratory of Acoustics of Le Mans University rounded out the leadership team. Additional co-authors include Kai Qian (the first author), Nan Cheng, Francesco Serafin, Kai Sun, and Nicolas Herard.
What the Researchers Could Not Yet Do
The team is the first to admit the limits of the demonstration. The system can currently pull the kink toward the sound source, but not push it away, a one-way interaction the authors describe as already surpassing anything shown before, but still incomplete. The “toy model” is also a one-dimensional chain rather than a full 3D structure, and the kinks in the experiment are macroscopic stand-ins for atomic-scale features.
Next steps include building three-dimensional versions of the metamaterial and exploring whether the same kind of remote, acoustic control could exist at much smaller, eventually atomic, scales. The work is also being extended to more disordered metamaterials, where kink behavior gets richer and harder to predict.
A Sister Breakthrough: Long-Range Acoustic Levitation in Midair
While the metamaterial team was reworking the inside of materials, a separate group at the University of Tsukuba in Japan, the University of Bristol in the United Kingdom, and Pixie Dust Technologies in Tokyo was pushing the boundaries of the more familiar side of acoustic tractor beams, the ability to levitate and move physical objects in midair. Their results, published in Physical Review Letters in late August 2026, deliver the longest working distance ever achieved by a single-sided acoustic levitator.
Conventional acoustic levitators trap particles at the pressure nodes of a standing wave, the quiet spots surrounded by high pressure. For a decade, researchers have tried to trap particles at the opposite kind of location, the high-pressure core of a focused beam, because that geometry is what makes single-sided, long-distance manipulation possible in principle. The Tsukuba-Bristol team is the first to make it work in three dimensions in midair.
The trick was to use a zero-order Bessel beam, a kind of acoustic wave whose central core stays tightly focused over a much longer distance than a regular focused beam, because Bessel beams are diffraction-free and self-healing. A phased array of ultrasound transducers shaped the wavefront, and below a critical cone angle, the velocity-gradient contribution to the Gor’kov potential overpowered the pressure-gradient term. The result is a transverse restoring force that pulls particles toward the beam axis even though the axis is the high-pressure region.
- Working distance: up to 397 mm, about 40 cm, roughly 6 times farther than the previous 67 mm ceiling for single-sided traps.
- Particle: a 1.5 mm expanded-polystyrene sphere, levitated stably for the full 60-second observation window across 15 trials.
- Multi-particle support: a Dammann grating in the array generates multiple parallel Bessel beams that levitate particles side by side, or stacked one above the other in a bottle-trap configuration.
- Robustness: because Bessel beams reconstruct themselves after passing an obstacle, a particle was successfully levitated above a 50 mm cube placed in the beam path.
The work also lifts a long-standing experimental ceiling first flagged by Marzo and colleagues in 2015, who characterized levitation at an acoustic focus as “an experimentally unstable theoretical possibility.” The Tsukuba-Bristol result is the first clean experimental demonstration that the prediction holds, in midair, in three dimensions.
Bruce Drinkwater, a co-author of the new PRL paper, was also a co-author of the 2015 Bristol-Sussex acoustic-hologram paper that helped launch the modern acoustic tractor beam era. The new result extends that lineage by nearly an order of magnitude in working distance.
Why It Matters: Adaptive Materials, Soft Robotics, and Beyond
Looking past the physics, the practical promise of these two complementary acoustic tractor beam results is wide.
- Adaptive protective gear. Helmets, pads, or body armor whose stiffness can be turned up on impact and softened again afterward.
- Robotic muscles and soft robotics. Actuators whose rigidity is dialed in by sound rather than by bulky motors or pneumatics.
- Medical implants. Devices such as stents or supports that start out flexible for insertion and stiffen once in place.
- Vibration control. Structural elements that change how they absorb shocks on demand, useful in aerospace, vehicles, and buildings.
- Long-range contactless handling. The PRL Bessel-beam approach points to single-sided levitators that can manipulate pharmaceutical powders, fragile droplets, or printed components across a lab bench without enclosing chambers.
- Robust signal transmission. Tunable mechanical waveguides for acoustic or ultrasonic signal routing.
“Right now, this is a toy model,” Boechler said of the metamaterial result. “If something like this could be made into a real material, you could imagine structures that adapt on the fly, materials you can reprogram using sound.”
Funding and Where to Read the Studies
The stiffness-reprogramming work was supported by the U.S. Army Research Office (grant W911NF-20-2-0182) and the U.S. Office of Naval Research (MURI N00014-20-1-2479). The full study, “Observation of mechanical kink control and generation via acoustic waves,” by Kai Qian, Nan Cheng, Francesco Serafin, Nicolas Herard, Kai Sun, Georgios Theocharis, Xiaoming Mao, and Nicholas Boechler, appears in Nature Communications 17, article 2428 (DOI: 10.1038/s41467-026-68688-7). A reader-friendly account is available on the UC San Diego Today news site.
The long-range levitation work, “Midair Single-Sided Acoustic Levitation in High-Pressure Regions of Zero-Order Bessel Beams,” by Yusuke Koroyasu, Christopher Stone, Yoichi Ochiai, Takayuki Hoshi, Bruce W. Drinkwater, and Tatsuki Fushimi, appears in Physical Review Letters 137, 094001 (DOI: 10.1103/pfkh-4x7j). It was supported by the Japan Science and Technology Agency (JST ASPIRE, JPMJAP2330), JST SPRING, the Japan Society for the Promotion of Science (JSPS KAKENHI 25KJ0673), and the UK Engineering and Physical Sciences Research Council (EPSRC EP/Z534171/1).
