THE first ever quantum computer chip has been built by researchers in Australia. But there鈥檚 a slight snag. They鈥檙e not sure whether it works.
Most efforts to build quantum computers focus on exotic technologies such as ultracold atoms. But in 1998, Bruce Kane at the University of Maryland suggested storing data in the spin of the nuclei of phosphorus atoms buried in a standard silicon chip. The spin can switch its orientation between two states.
Kane showed it would be possible to read the state of the nuclei in two neighbouring phosphorus atoms by measuring movement of an electron from one to the other. 鈥淧eople said 鈥榞reat, brilliant idea, but can you build it?'鈥� says Robert Clark of the Centre for Quantum Computer Technology at the University of New South Wales in Sydney.
Advertisement
Now Clark and his colleagues have built the basics of a quantum chip. They etched deep, parallel, nanometre-wide trenches into a silicon slice and then fired a single phosphorus atom into each one. Once the atom was in place they covered over the trenches with silicon.
But the team hit a problem when they tried Kane鈥檚 scheme for storing quantum information. When they applied an electric field to the phosphorus atoms to try to read the spins of their nuclei, the electron whose movement they were hoping to detect disappeared.
Clark is now working on a different way to read the quantum information. He says the missing electron would effectively leave a hole that oscillates between neighbouring phosphorus atoms. Charge detectors could track this oscillation. 鈥淭his really is the central processing element of a quantum computer,鈥� he says.
Paul Horn, head of IBM鈥檚 research programme in New York, says tricky manufacturing methods like the one Clark鈥檚 group is using could be hard to scale up. But Clark points out that a chip containing just 30 phosphorus atoms could outperform today鈥檚 supercomputers. 鈥淓ven if it does take a while to build, who cares?鈥� he says.
