Dec 3 2018
Computers will operate only when bits—the zeros and ones that constitute digital information—interact and provide data for processing. The same holds true in the case of quantum bits, or qubits, that constitute quantum computers.
Piecing together the process: The microscope objective (the big metallic barrel coming down from the top of the image), the diamond sample (the small plate that looks like glass in the center of the image), and the optical fiber that couples to the sample (glowing green point just above the sample). (Image credit: Denis Sukachev)
However, that interaction also presents an issue—in any type of system where qubits interact with one another, they are also likely to interact with their environment, leading to qubits that rapidly lose their quantum nature.
In order to overcome that issue, Ruffin Evans, PhD student of Graduate School of Arts and Sciences, studied photons—particles that are largely known for their lack of interactions.
Evans is the study’s lead author and works in the laboratory of Mikhail Lukin, the George Vasmer Leverett Professor of Physics and co-director of the Quantum Science and Engineering Initiative. Reported in the journal Science, the study describes a technique for creating an interaction between a pair of qubits with the help of photons.
It’s not hard to engineer a system with very strong interactions, but strong interactions can also cause noise and interference through interaction with the environment. So you have to make the environment extremely clean. This is a huge challenge. We are operating in a completely different regime. We use photons, which have weak interactions with everything.
Ruffin Evans, PhD Student, Graduate School of Arts and Sciences, Harvard University.
Along with his colleagues, Evans began by producing two qubits with the help of silicon-vacancy centers, which are atomic-scale impurities present in diamonds, and placing them within a nanoscale device called a photonic crystal cavity, which acts similar to two facing mirrors.
“The chance that light interacts with an atom in a single pass might be very, very small, but once the light bounces around 10,000 times, it will almost certainly happen,” he stated. “So one of the atoms can emit a photon, and it will bounce around between these mirrors, and at some point, the other atom will absorb the photon.”
However, the transfer of that specific photon doesn’t go only in a single direction.
“The photon is actually exchanged several times between the two qubits,” Evans stated. “It’s like they’re playing hot potato; the qubits pass it back and forth.”
While producing an interaction between qubits is not a novel concept—scientists were able to manage the feat in several other systems—there are a couple of factors that make the latest research unique, stated Evans.
The key advance is that we are operating with photons at optical frequencies, which are usually very weakly interacting. That’s exactly why we use fiber optics to transmit data—you can send light through a long fiber with basically no attenuation. So our platform is especially exciting for long-distance quantum computing or quantum networking.
Ruffin Evans, PhD Student, Graduate School of Arts and Sciences, Harvard University.
Evans informed that while the system was able to work only at very low temperatures, it is less complicated when compared to methods that need elaborate systems of optical traps and laser cooling to keep atoms in place. He added that since the system has been developed at the nanoscale, a number of devices could be possibly accommodated on a single chip.
“Even though this sort of interaction has been realized before, it hasn’t been realized in solid-state systems in the optical domain,” he stated. “Our devices are built using semiconductor fabrication techniques. It’s easy to imagine using these tools to scale up to many more devices on a single chip.”
Evans foresees two key directions for upcoming study. In the first direction, new ways can be developed to apply control over the qubits and consequently a complete suite of quantum gates can be built that would enable them to operate as a workable quantum computer.
The other direction is to say we can already build these devices, and take information, read it out of the device and put it in an optical fiber, so let’s think about how we scale this up and actually build a real quantum network over human-scale distances. We’re envisioning schemes to build links between devices across the lab or across campus using the ingredients we already have, or using next-generation devices to realize a small-scale quantum network.
Ruffin Evans, PhD Student, Graduate School of Arts and Sciences, Harvard University.
The study may eventually have wide-reaching effects on the future of computing, Evans said.
“Everything from a quantum internet to quantum data centers will require optical links between quantum systems, and that’s the piece of the puzzle that our work is very well-suited for,” he stated.
The study was supported with funding from the NSF, the CUA, the DoD/ARO DURIP program, the AFOSR MURI, the ONR MURI, the ARL, the Vannevar Bush Faculty Fellowship program, the DoD NDSEG, and the NSF GRFP.