Mohammad Hafezi, Eugene A. Demler, Mikhail D. Lukin, Jacob M. Taylor [photos courtesy of Joint Quantum Institute (JQI) and Harvard University]

The scientists  say the work not only may lead to more efficient information processors on our desktops, but also could offer a way to explore a particularly strange effect of the quantum world known as the quantum Hall effect in which electrons can interfere with themselves as they travel in a magnetic field.

The corresponding physics is rich enough that its investigation has already resulted in three Nobel Prizes, but many intriguing theoretical predictions about it have yet to be observed.

The advent of optical fibers a few decades ago made it possible for dozens of independent phone conversations to travel long distances along a single glass cable by, essentially, assigning each conversation to a different color—each narrow strand of glass carrying dramatic amounts of information with little interference.

Ironically, while it is easy to send photons far across a town or across the ocean, scientists have a harder time directing them to precise locations across short distances—say, a few hundred nanometers—and this makes it difficult to employ photons as information carriers inside computer chips.

"We run into problems when trying to use photons in microcircuits because of slight defects in the materials chips are made from," says Jacob Taylor, a theoretical physicist at NIST and JQI. "Defects crop up a lot, and they deflect photons in ways that mess up the signal."

These defects are particularly problematic when they occur in photon delay devices, which slow the photons down to store them briefly until the chip needs the information they contain. Delay devices are usually constructed from a single row of tiny resonators, so a defect among them can ruin the information in the photon stream.

But the research team saw that using multiple rows of resonators would build alternate pathways into the delay devices, allowing the photons to find their way around defects easily.

As delay devices are a vital part of computer circuits, the alternate-pathway technique may help overcome obstacles blocking the development of photon-based chips, which are still a dream of computer manufacturers. While that application would be exciting, lead author Mohammad Hafezi says the prospect of investigating the quantum Hall effect with the same technology also has great scientific appeal.

"The photons in these devices exhibit the same type of interference as electrons subjected to the quantum Hall effect," says Hafezi, a research associate at JQI. "We hope these devices will allow us to sidestep some of the problems with observing the physics directly, instead allowing us to explore them by analogy."

Advances in experimental design, physicists at the National Institute of Standards and Technology (NIST) have achieved a record-low probability of error in quantum information processing with a single quantum bit (qubit)—the first published error rate small enough to meet theoretical requirements for building viable quantum computers.

THE LOW ERROR QUBIT
A quantum computer could potentially solve certain problems that are intractable using today's technology, even supercomputers. The NIST experiment with a single beryllium ion qubit, described in a forthcoming paper, is a milestone for simple quantum logic operations. However, a working quantum computer also will require two-qubit logic operations with comparably low error rates.

"One error per 10,000 logic operations is a commonly agreed upon target for a low enough error rate to use error correction protocols in a quantum computer," explains (right) Kenton Brown, who led the project as a NIST postdoctoral researcher Brown,  now works at the Georgia Institute of Technology

"It is generally accepted that if error rates are above that, you will introduce more errors in your correction operations than you are able to correct. We've been able to show that we have good enough control over our single-qubit operations that our probability of error is 1 per 50,000 logic operations."

The NIST experiment was performed on 1,000 unique sequences of logic operations randomly selected by computer software. Sequences of 10 different lengths, ranging from one to 987 operations, were repeated 100 times each. The measured results were compared to perfect theoretical outcomes. The maximum length of the sequences was limited by the hardware used to control the experiment.

The record low error rate was made possible by two major changes in the group's experimental setup. First, scientists manipulated the ion using microwaves instead of the usual laser beams. A microwave antenna was incorporated into the ion trap, with the ion held close by, hovering 40 micrometers above the trap surface.

The use of microwaves reduced errors caused by instability in laser beam pointing and power, as well as spontaneous ion emissions. Second, the ion trap was placed inside a copper vacuum chamber and cooled to 4.2 K with a helium bath to reduce errors caused by magnetic field fluctuations in the lab.

Co-author Christian Ospelkaus contributed to the research while at NIST and is now at research institutions in Germany. The research was supported in part by the Intelligence Advanced Research Projects Activity, the National Security Agency, the Defense Advanced Research Projects Agency and the Office of Naval Research.