Atomic circuitry and quantum computing

Conventional supercomputers have limitations: they are logical and fast, certainly, can be run in parallel grids across the globe, but when it comes down to solving problems with no logical answer, such as cracking sophisticated encryption, working out the travelling sales-rep problem of logistics and deliveries, or modelling the climate, they have serious limitations.

A quantum computer, on the other hand, could find all the answers almost instantaneously and pluck out the most appropriate based on probabilities and quantum mechanics. Building such a quantum computer is not proving simple. Now, US researchers have demonstrated that they can exert delicate control over a pair of atoms within a mere seven-millionths-of-a-second window that suggests the necessary atomic circuitry for a quantum computer might one day be possible.

“At some point in time you get to the limit where a single transistor that makes up an electronic circuit is one atom, and then you can no longer predict how the transistor will work with classical methods,” explains physicist Mark Saffman of the University of Wisconsin-Madison. “You then have to use the physics that describes atoms – quantum mechanics.” In the quantum realm, new possibilities for processing information emerge that mean certain types of problems could be solved exponentially faster on a quantum computer than on any foreseeable classical computer.

Mark Saffman
Mark Saffman

Working with colleague Thad Walker, Saffman and co-workers have successfully used atoms to create a controlled-NOT (CNOT) gate, a basic type of circuit that will be an essential element of any quantum computer. They describe details of the work in the journal Physical Review Letters and explain that this is the first demonstration of a quantum gate formed between two uncharged atoms.

The use of neutral rubidium atoms chilled to a fraction of a degree above absolute zero, rather than charged ions or other materials, distinguishes this achievement from previous work. “The current gold standard in experimental quantum computing has been set by trapped ions … People can run small programs now with up to eight ions in traps,” explains Saffman. However, to be useful for computing applications, systems must contain enough quantum bits, or qubits, to be capable of running long programs and handling more complex calculations. An ion-based system presents challenges for scaling up because ions are highly reactive, which makes them difficult to control.

Thad Walker
Thad Walker

“Neutral atoms have the advantage that in their ground state they don’t talk to each other, so you can put more of them in a small region without having them interact with each other and cause problems,” Saffman says. “This is a step forward toward creating larger systems.” The team is now working towards arrays of up to 50 atoms to test the feasibility of scaling up the system.

LINKS

Phys. Rev. Lett. 2010, 104, 010503 http://prl.aps.org/abstract/PRL/v104/i1/e010503

Mark Saffman
http://hexagon.physics.wisc.edu/marksaffman.htm

Thad Walker
http://www.physics.wisc.edu/people/faculty/twalker/