The July 2022 issue of IEEE Spectrum is here!

Close bar

Bendable Sound Waves Can Skirt Objects, Trap Particles

Someday, scientists will sort molecules with sound, and bones won't photobomb ultrasound images

2 min read
Bendable Sound Waves Can Skirt Objects, Trap Particles
Image: Xiang Zhang group

Scientists have developed a method to bend sound waves as they travel through open air, and can even create an acoustic “bottle” that can trap and hold tiny particles. The method could improve cell sorting, sharpen ultrasound images, even lead to a sonic cloaking device.

The method, developed in the laboratory of Xiang Zhang, professor of mechanical engineering at the University of California in Berkeley, requires adjusting the phase and amplitude of the sound coming from each of dozens of closely spaced speakers. With precise calculations, the differences among the speakers produce an interference pattern that causes a beam of sound to bend in a desired way. That includes bending the waves so the sound travels around an object and then continues on its original path. The waves can be manipulated to whatever extent is required, says Jie Zhu, who coauthored the Nature Communications paper describing the work when he was a postdoc in Zhang’s lab.

“First you have to know what you want to achieve, then, using our method, you can construct the shape at the source,” says Zhu, now assistant professor of mechanical engineering at the Hong Kong Polytechnic University.

With a linear array of 40 speakers, each 1.5 centimeters in diameter, spaced 2.5 cm apart, and operating at a frequency of 10 kilohertz, the researchers shaped the sound waves in two dimensions. When they set up a two-dimensional array of over 100 such speakers, they were able to curve the waves in three dimensions, creating an acoustic bottle, with high acoustic pressure forming the walls. They placed a hard plastic ball in the center of the bottle, and found that the pressure of the sound waves pushed at it to hold it in place. Zhu says the bottle was fairly weak, but using more powerful speakers would strengthen it. Such a bottle could be used to carry a small particle, creating acoustic tweezers that could be used, for example, to sort cells or molecules in a solution.

The ability to direct sound around objects in its path could improve ultrasound imaging by allowing sound waves to skirt bones that would otherwise be in the way of what they were trying to see.

The technique grew out of work Zhang’s lab has done on metamaterials, which contain periodic structures that cause light or sound waves to move in unusual ways. Zhang and others have used metamaterials to create cloaking devices to hide objects from specific frequencies of light or sound. Zhu says that using the same principles, but without the metamaterial, allows them to achieve similar feats in free space, making it much more attractive for practical use. “Most applications happen in the air or in liquid,” he says. “Most applications don’t allow us to change the medium.”

The Conversation (0)

The Future of Deep Learning Is Photonic

Computing with light could slash the energy needs of neural networks

10 min read
Image of a computer rendering.

This computer rendering depicts the pattern on a photonic chip that the author and his colleagues have devised for performing neural-network calculations using light.

Alexander Sludds

Think of the many tasks to which computers are being applied that in the not-so-distant past required human intuition. Computers routinely identify objects in images, transcribe speech, translate between languages, diagnose medical conditions, play complex games, and drive cars.

The technique that has empowered these stunning developments is called deep learning, a term that refers to mathematical models known as artificial neural networks. Deep learning is a subfield of machine learning, a branch of computer science based on fitting complex models to data.

Keep Reading ↓Show less