Anthony Doesburg: Arms race to help computers take a quantum leap

An LGM-30 Minuteman III intercontinental missile launches from the Vandenberg Air Force base in California. Photo/Wikimdedia Commons.

It's been characterised as an arms race - the efforts going on in physics labs around the world to be first to build a working quantum computer.

It's seen that way in some quarters because a quantum computer, a device that exploits the weird world of quantum mechanics, would make short work of deciphering the cryptographic codes used to secure internet banking - and numerous other things.

But hang on - there's no need to give up the convenience of paying your bills online just yet. No one believes quantum computers will emerge from the research labs for at least a decade. And even before they do, cryptography based on the same quantum principles will ensure state secrets stored on government computers - not to mention the money in your savings account - will remain secure.

So calling it an arms race is probably overstating things. But there's no denying numerous groups of researchers are vying to be first to make the breakthroughs - and many are needed - that will help bring quantum computers to reality.

The very concept of reality goes out the window when the world is reduced to atoms and their constituents, the components from which quantum computers are being built. To the layperson, when physicists start going on about atoms being in more than one place at a time - by inhabiting multiple universes, according to one theory - it begins to sound like make-believe.

Even Andrew Dzurak, who leads quantum computing research efforts at the University of New South Wales, calls it a "spooky" realm - although one suspects that's for the benefit of his lay audience.

Dzurak and team, who are funded by the Australian Research Council and the United States National Security Agency, have been working for about a decade on trying to develop the basic transistors of a quantum computer.

It seems they may be on the brink of one of those key breakthroughs. In an experiment late last year, team member Andrea Morello thinks he got close to succeeding in taking a real-time measurement of the spin direction - either "spin up" or "spin down" - of a phosphorous electron.

In the kind of computer they are trying to build - and different researchers are going down different paths - the electrons bound to phosphorous atoms are the quantum bits, or qubits, that will represent the ones and zeros at the core of the machine. Once able to tell whether in a "spin up" or "spin down" state, where up is one and down is zero, the next step is to be able to change from one state to the other.

It sounds simple enough until you remember this is taking place on a scale where measurements are in nanometres, or billionths of a metre. Dzurak and Morello are working with a single phosphorous atom implanted in silicon.

They have opted for phosphorous because it holds the direction of electron spin for long enough - an hour at -272C - to be useful. But it is painstaking work to ensure the electron is not interfered with by stray electrical charges in the silicon.

How does a single qubit advance the computing cause when equated with the billions of transistors on one of today's Pentium chips? Not much. But if you had 300 qubits, say, the spookiness really kicks in.

The Pentium processor's billions of transistors perform calculations in a sequential fashion, which for many tasks it does at an adequate speed.

Quantum computers, literally, will make a quantum leap.

Through the baffling process of "entanglement", those 300 qubits will be able contain the equivalent information of 2 to the power of 300 classical computer bits. That's a big number.

"It's actually comparable to all the atoms that make up the entire universe," Dzurak says. "It's the power of exponentials that quantum computing makes use of.

"For certain types of problems, it could solve in minutes what would take all the existing computers in the world put together thousands of years to calculate. So we're talking billions of times faster - as many zeros as you can think of."

Using all that power to crack cryptography codes is one eye-catching application. Dzurak sees others - like modelling the way biological structures, such as genes, are put together. But that is decades away.

And the arrival of quantum computers in offices and homes is years beyond that. "As they become easier and cheaper to make, and software is written for them, they'll get disseminated into multipurpose commercial products," Dzurak says.

In the meantime, Morello presented preliminary results of his spin measurement experiment, which he expected to "blow away" other phosphorous qubit researchers, at a semiconductor conference this week in Kyoto.

Not that he sees them as competitors. "You're all trying to help each other," Morello says. "But you're more helpful when you're in a good position."

Anthony Doesburg is an Auckland technology journalist.