20 September 2012—Building on some groundbreaking research from a couple of years ago, physicists led by Andrea Morello and Andrew Dzurak of the University of New South Wales, in Australia, are reporting in this week’s issue of Nature that they have managed to create a quantum bit in silicon using just a single atom. The experimental technique is, in principle, compatible with existing semiconductor technology and is being heralded as a big step toward the development of silicon-based quantum computers.
Quantum computers—as opposed to the computers we use daily—rely on the laws of quantum mechanics to speed up calculations. They could theoretically process complex calculations in a fraction of the time conventional computers require. Quantum computers are thought to be ideally suited to data-intensive problems, such as cracking modern ciphers, needle-in-a-haystack searches of huge databases, and modeling biological molecules and drugs.
However, it is extremely difficult for a quantum computer to hold information for long periods of time, as quantum systems tend to become unstable and “decohere,” losing data when they interact with their environments. As a result, only rudimentary quantum computers have been built, and the field of quantum computing has never quite left the laboratory. (A Canadian company called D-Wave Systems announced a quantum computer in 2011, but most physicists are skeptical about whether it is a true quantum computer.)
Some of the popular approaches to building quantum computers include confining atoms in cryogenic gases to create quantum bits (or qubits) and using superconducting circuits. In 1998, Bruce Kane, a physicist at the University of Maryland, in College Park, suggested an approach based on solid-state silicon semiconductors that are doped with phosphorus. He proposed using the quantum characteristic of spin in the nucleus of the phosphorus donor atom as the qubit. Many scientists were enamored with this idea, because it showed the potential to integrate quantum computers with conventional silicon processing.
Morello and Dzurak were among the physicists impressed by Kane’s proposal, but they chose to investigate electron spins instead, because electron spins in silicon have very long coherence times—that is, it takes a relatively long time for such a qubit to lose its information. In 2010, they demonstrated the ability to read the state of an electron’s spin. Basically, they managed to get the spin state of the electron to control the flow of electrons in a nearby circuit and produce a digital readout.
Now, Morello and Dzurak have discovered how to write the spin state. This completes the two-stage process required to operate a quantum bit. They managed to do this by using a microwave field to gain unprecedented control over an electron bound to a single phosphorous atom, which was implanted next to a specially designed silicon transistor.
Jarryd Pla, a Ph.D. student on their team, whose thesis will be based on this result, says the biggest challenge was “of an engineering nature.” The researchers needed to blast the system with high-power microwaves in order to control the spin, but to observe or read the state of the spin, they needed very specific conditions. They cooled their device to around 0.1 kelvin (just above absolute zero) to reduce thermal noise. But the high-power microwaves still posed a problem.
“The issue was that you cause decoherence,” says Pla. “We put a lot of effort into the engineering of a nanosize transmission line that was fabricated just 100 nanometers away from the qubit. This transmission line was good at delivering the high-frequency microwaves needed for control of the spin state and minimized unwanted signals like stray electric fields.” Researchers at the same university have done related work in single-atom transistors.
In a commentary that accompanied publication of the research in Nature, physicists Lee Bassett and David Awschalom praised the work highly, saying it “effectively combines the advantages of both atomic and solid-state implementations by using the spin of an electron bound to a single phosphorus atom that is implanted in a silicon substrate near a microfabricated transistor.”
So what’s next?
“The next natural step is to demonstrate a 2-qubit logic gate, by placing two such atoms close to each other,” says Morello. “We are already working on this project; prototype 2-qubit devices are being fabricated as we speak.”