Duke University and University of Maryland researchers used frequency measurements to gain insight into phase changes in quantum computers. This is similar to water turning to steam.

Researchers measured the number of operations that could be performed on a quantum computing platform without triggering its quantum state’s collapse. This gave them insight into the tipping points of other systems, both natural and computational. These results provide computer scientists with guidance as they attempt to correct quantum errors in order to enable quantum computers to reach their full potential.

The results were published online in Nature Physics on June 3, 2013.

The movement of molecules changes as water boils, until it reaches a critical point where it turns to steam. Similar to this, a quantum computing device can be controlled in discrete steps until it reaches a single solution.

Crystal Noel, an assistant professor of electrical, computer engineering, and physics at Duke, stated that there are deep connections between quantum theory and phases of matter. “The quantum computing system behaves in the same manner as quantum systems found throughout nature, like liquid changing to steam–even if it is digital.”

Quantum computers’ power lies in their ability to have both a 1 or 0 at once, with exponential increases in system complexity as more qubits become available. They can tackle large parallelism problems, such as putting together a puzzle by assembling all the pieces at once. However, the qubits must be able maintain their quantum indecisiveness up to the solution.

Error correction is one of the many problems this creates. The system must be capable of detecting and correcting errors. Some qubits may lose some information. It is difficult to keep an eye on errors in quantum systems because they lose their “quantumness” after being measured. Even with additional qubits, it’s difficult to keep an eye on quantum algorithms for errors.

Noel stated that quantum computers can only withstand a certain number of measurements before they lose their quantum information. “And that number is an analogy to how many errors the computer could withstand while still functioning correctly.”

Noel and her coworkers investigate the transition threshold and system state in the new paper.

The group worked closely with Christopher Monroe (the Gilhuly Family Presidential-Distinguished Professor of Engineering and Physics, Duke), Marko Cetina (assistant professor of Physics at Duke), Michael Gullans and Alexey Gorshkov at University of Maryland and National Institute of Standards and Technology to create software that would run random quantum circuits that were tailored to the capabilities of their quantum systems. The experiment was conducted on the Duke Quantum Center’s powerful quantum computers, the ion trap.

Noel stated that “the number of qubits within the system, its fidelity in operations and the level automation of the system are unique to this quantum computer systems.” While other systems are capable of achieving each individually, they have never been able to do all three simultaneously in an academic system. This is what enabled us to conduct these experiments.”

The team was able see how frequency of measurement affected qubits by averaging many random circuits. The critical point at which the system lost its coherence, quantum information and was inevitably reached was as predicted. Researchers will be able, by studying how the system behaved at each end of this phase transition, to develop better error-correction codes.

Researchers have never had the opportunity to look at how other phases change in nature before.

Monroe stated, “This demonstration is an excellent example of what we do unique at the Duke Quantum Center.” Our quantum computers are composed of atoms under precise control using electromagnetic traps lasers and optics. However, these systems can be used to do something entirely different, such as probing the quantum nature of phase transformations. The same quantum computer can be used to solve vexing problems in a variety of fields, including chemical reactions, DNA sequence, and astronomy. This requires expertise in not only atomic physics but also systems engineering and computer science, as well as in the field that will be used to run the application.