Art Courtesy of Annli Zhu.
Communication is a natural part of life. Humans talk, birds chirp, and even trees interact through their root networks. To maintain this essential aspect of life, we adapt our methods to overcome communication challenges: a team meeting that once had to be held in a boardroom can now be effectively held on a Zoom call. For quantum computers—a system that looks to advance past the capabilities of classical computing—communication occurs by leveraging quantum particles and their properties, components of sub-atomic interactions that have historically been challenging to harness.
Recently, physicist Mo Li and his colleagues at the University of Washington were able to overcome one such challenge: dealing with unpredictable photon emitters. They achieved this by designing a deterministic emitter––one where they can determine where the photon is emitted––and in doing so, they discovered that their emitter produced a strong interaction between two important quantum quasiparticles: photons and phonons. Now, Li is hopeful that further research can use this interaction to advance communication in quantum computing systems and overcome some challenges in the field.
In classical computers, information is stored in bits: either 0 or 1. Quantum computers use quantum bits—called “qubits”—which can exist in a “superposition” state of being both 0 and 1 at the same time, like Schrödinger’s cat. This allows them to consider many possibilities simultaneously. Through a process called entanglement, qubits can be connected in a way such that the state of one qubit instantly influences the state of another, no matter how far apart they are, enabling quantum computers to perform complex calculations literally faster than light can travel. This means quantum computers have the potential to revolutionize fields like cryptography, drug discovery, and more. However, because they are highly sensitive to environmental conditions, require extremely low temperatures, and use extensive space, they are expensive to build and difficult to scale.
Although many subatomic particles can be used for quantum computers, scientists prefer to use photons—tiny, massless particles of light—to transmit quantum information because they travel at, well, the speed of light. But photons are difficult to reliably produce, control, and capture. Traditional methods of photon generation—through so-called “quantum emitters”—involve taking advantage of defects in various atomic lattices, which are patterned arrays of bound atoms. However, these defects often emit photons unpredictably, which is undesirable for highly precise quantum computers.
To address this problem, the team of scientists at the University of Washington set out to build a “deterministic” quantum emitter. “We want to engineer it in such a way that we can say ‘we want an emitter here’ and it indeed emits there,” said Li, Professor of Electrical & Computer Engineering and Physics and leader of the research team.
To achieve this goal, the team used two single-atom layers of tungsten and selenium, similar to existing quantum emitters. Then, they draped these layers over hundreds of nanoscopic pillars, creating tiny bumps in the 2D lattice that isolated the target regions. By shining a precise pulse of laser light at an electron in the material, they were able to free it for a very short period of time. Each time an electron returned to its place, it emitted a single photon encoded with quantum information—a successful quantum emitter.
Amidst their successes with the deterministic emitter, Li and his colleagues noticed something intriguing in their data. “The emitter ideally is supposed to generate a very sharp peak in energy at one wavelength associated with the photon, but when we looked a little bit closer, there [was] a group of satellite peaks on the sides, and we wondered where that [came] from,” Li said.
As they analyzed the data, they came to an exciting conclusion: phonons—quantum quasiparticles that are a unit of vibrational energy—may be responsible for these satellite peaks. The energy is caused by the vibration between two atomic layers, and such motion has been described as “atomic breaths.”
“It’s not uncommon,” Li said. “It’s called phonon replica, and it appears in other systems as well, but in our system, it’s very pronounced.” Normally, the phonon replica will appear as a group where intensity is strongest at the shortest wavelengths, and then rapidly decays. In their system, however, the phonon replica is the strongest in the middle and weaker at the side peaks. This indicated that the “coupling”—the phonon interaction with the emitter or the mechanical vibration between the two atomic layers—is very strong and overwhelms the emission that has no interaction with the phonon, creating this irregular array of peaks.
“Every time [the emitter] takes a breath, it emits one phonon and that phonon is taken out of one photon. So, the optical photon that is emitted is reducing energy by exactly one phonon,” Li said.
Phonons have been historically difficult to leverage for quantum computation, but they have great potential when coupled with photons. While photons are very popular for communication due to their speed, storing information on them is difficult. On the other hand, because phonons vibrate at a much lower frequency, future advancements in quantum technology may allow them to live much longer than photons, acting as temporary information storage. This is where the phonon-photon interaction comes in.
“They can exchange information. When you want to stall the quantum information there, you convert them into phonons. The information will stay there; a little while later you can come back and read it out. But if you’re ready to send that information out of the system to another system, then you convert it into a photon,” Li said.
Leveraging this deterministic emitter and strong coupling activity could advance quantum computing systems by improving inter-computer communication. Excited, Li shared some of his ideas for future research that may be able to translate his findings into something specifically useful for quantum computing. One idea is the possibility of building a similar system with more than one emitter. But what would this achieve?
“Because the phonons are localized, if the two emitters are close enough, the vibrations will interact with each other. This is a way to make two emitters talk to each other,” Li said.
Unlike photons, which don’t couple with each other, the phonon’s properties suggest a possibility of coordinating two or three emitters—by coupling the phonons instead. Since the photons cannot interact, they can instead “talk through” the phonons before flying off to their destinations. If an effective two- or three-emitter system is achieved, it could revolutionize the way quantum computers communicate with each other.
This theory may also be able to address the issue of scaling in quantum computers— something that has greatly challenged researchers in the field. While larger computers with more qubits are more powerful in completing tasks, they are difficult and expensive to build and maintain. Currently, IBM’s 433-qubit computer is the largest in the world. “But [433 qubits] isn’t enough to do any realistic quantum computing,” Li said. “Maybe some toy models, but nothing to the level of what quantum computers promise in theory.”
Instead, just like in classical computers, tasks would benefit from being modularized, split up to be completed in parallel by multiple smaller computers. But while classical computers can operate at room temperature, most quantum computers require extremely low-temperature and low-noise environments in order to facilitate the precise manipulation of qubits. On the other hand, any communication between computers, achieved by sending photons through fiber-optic cables, happens at a frequency five orders of magnitude higher than that at which quantum calculations are performed. “We need something to bridge this energy gap,” Li said, “This is where our emitters have their potential.”
The team’s breakthrough in photon-phonon coupling would allow these spatially separate quantum computers to solve the problem of effective transduction: converting signals between mediums without loss of information. This gives researchers the potential to build scalable, modularized quantum computing networks.
“The holy grail of this research would be to make two, maybe three, or more, emitters talk to each other,” Li said. “This will allow us to realize the full potential of quantum computing.”