Phonon Lasers Make a More Practical Sound

Image: Dane Wirtzfeld/iStockphoto

Ever since the laser saw the light of day a half century ago, researchers have been playing with the idea that something similar could be created using sound rather than light. But the concept made little headway in the ensuing decades. In 2009, the situation changed abruptly, when scientists at Caltech and the University of Nottingham, in England, using tiny drums and stacked semiconductors, respectively, employed conventional lasers to stimulate or probe the emission of a stream of “phonons”—the quasiparticles of sound—proving that phonon lasers, or “sasers,” were indeed a sound idea.

Now researchers at NTT Basic Research Laboratories, in Japan, have taken a significant step forward by fabricating an entirely electromechanical resonator on a chip that also eliminates the need for the lasers that previous devices required. This advance makes integration with other devices easier, and applications like extremely high-resolution medical imaging and compact, low-power, high-frequency clock-pulse generators are now within reach, say its inventors.

The word laser is an acronym for “light amplification by stimulated emission of radiation.” A laser works by exciting electrons around an atom to higher levels, which then shed the extra energy in the form of photons. This activity takes place in an optical resonator, which is essentially an enclosed chamber, typically with mirrors at either end. The trapped photons bounce back and forth, stimulating the emission of more photons of the same wavelength, some of which are allowed to escape in a controlled beam of laser light.

“In our approach to the saser, we replaced the optical resonator with a microelectromechanical resonator, or oscillator, that moves up and down and produces a spectrum of discrete sonic vibrations, or phonon modes,” says Imran Mahboob, a researcher at NTT. “Simply put, we’re creating an electromechanical atom that we then jiggle to produce the phonons.”

The resonator consists of a micrometer-scale gallium arsenide bar (250 x 85 x 1.4 micrometers) called a beam, which is suspended above a gap in a semiconductor chip and whose oscillations are controlled with piezoelectric transducers. An alternating voltage applied to the beam’s terminals induces alternating expansion and compression. In this scheme, the bar plays the part of an optical resonator, while three levels of oscillating tones or modes (high, middle, and low) mimic the changing of the electron energy levels of the atoms in a specific type of optical laser, generating phonons in the process. When the high state is excited, it generates phonon emissions in the middle and low states. With some fine-tuning of the system, so that the sum frequency of the middle and low states matches the high mode, emission in the low mode is resonantly enhanced, and a precise, highly stable phonon beam is produced, with fluctuations limited to one part in 2 million.

Because the mechanical oscillations are extremely tiny, existing at the subnanometer level, “we place everything into a cryogenic environment with a temperature of around 2 kelvin to make them easier to observe,” says Mahboob. “This also ensures that the different resonance modes are precise, because if [the device is] hot, their frequencies would broaden and overlap so that the sum frequency of the middle and low states wouldn’t always match the high state.”

Well-Balanced Beam: A gallium arsenide resonator is the heart of NTT's phonon laser.Image: NTT

Notably, an output signal is observed only when the input voltage exceeds a specific figure. This threshold voltage is a signature feature of optical lasers, as is a large improvement in the beam’s frequency precision when phonon lasing is triggered. “So we’re convinced we have phonon lasing,” says Mahboob.

As for how such a laser could be used, he says that the device’s compactness, low energy consumption, and the possibility of high frequency give it the potential to replace the relatively bulky quartz-crystal resonators used to provide stable frequencies for synchronized operations and precise timekeeping in computers and other electronic equipment. Superior medical ultrasound imaging is another possible application, and Mahboob speculates that one day the laser might be used as a medical treatment.

Hiroshi Yamaguchi, an NTT senior distinguished researcher, also points out that by increasing the frequency of the oscillating states, the resonator could potentially be manipulated to store a discrete number of phonons. “This could open up new avenues to explore quantum cryptography and quantum computing,” he says, “as well as having the potential of enabling us to study quantum effects at the macro level.”

But before such speculations can be seriously investigated, the researchers admit they must first overcome a major challenge. Whereas optical lasers can travel through a vacuum, a phonon beam requires a medium. In this research, the sound propagates through the semiconductor crystal, and the researchers are now working out how to handle this limitation.

“On the other hand, the technology does have the advantage [in] that it’s a compound semiconductor,” points out Yamaguchi. “So it could, for example, easily be integrated with an optical device and an electrical device all on the same chip and, of course, integrated with a variety of systems. We believe this is a major advantage of our device.”

Other phonon laser researchers have been improving their devices, too. Tony Kent, a professor of physics at the University of Nottingham who is working with semiconductor stack devices to realize sasers, has been working on using them for applications that need frequencies in the hundreds of gigahertz or even terahertz frequencies. “Our main focus is exploring applications for a terahertz saser as a stable, low-noise reference or local oscillator, for use in communications, medical imaging, and security screening, and as a source for acoustic sensors of nano-objects,” he says.

Kent says that while he expects the NTT research to have a major impact on the fundamental science of micromechanical systems, he questions the practicality of some of the applications that are being suggested.

Putting aside the problem of having to work with low temperatures, and the difficulty getting the sound out of the resonator and into the semiconductor crystal, the reported beam device works at a frequency of only around 1 megahertz,” says Kent. “Yet there are already technologies generating acoustic signals for ultrasound measurement available now with frequencies higher than 1 GHz.”