A partnership between Microsoft and University of Sydney is confident its approach to building a quantum computer will see the first such machine switched on in Australia.
Microsoft’s multinational Station Q project to build a quantum computer began over a decade ago. At the project’s Sydney hub, announced last November and formalised with a signing at an event in July, things are shifting from pure science to engineering.
This comes in two components: breaking the project into parts and attacking them, and scaling up, says Professor David Reilly, the centre’s director.
“We're now using engineering to solve scientific problems,” he tells create.
“So take an engineering mindset to addressing systematically the challenges of quantum computing.
“It's knocking this problem from both directions. Coming at it from thinking about big picture stuff, continuing on the other hand to advance fundamental open frontier research.”
Scientific breakthroughs are still needed, to be sure, but the “multi-year global partnership agreement” signed on July 25 is confident it can get there first.
The diabolically tricky task of building a universal quantum computer is being attacked so vigorously, and with great financial investment, due to the predicted awesome usefulness of such a device.
Various ways of creating qubits - the basic unit of information in a quantum system - are being attempted. These include using single phosphorous atoms, electron spins, and photons. Station Q’s 'topological qubit' approach attempts to encode information in the braided paths of quasiparticles. These qubits haven’t been built, but Microsoft is confident that once this happens they will scale quickly and the qubits will have superior stability.
All qubits built so far are very fragile, and lose their quantum nature (or decohere) easily. Thermal vibrations can be one source of interference, and for this reason all current approaches use dilution refrigerators operating at temperatures near to absolute zero.
“You think about it with an analogy of being like the deep solid foundation for building a house,” Doug Carmean, partner architect at Microsoft Research and a former Intel Fellow, tells create.
“You would prefer that as possible as you can make it robust against earthquakes, noise and so forth. The topological qubit that we're using has that robustness that allows us to build an overall computer system on top of it.”
Station Q was launched in 2005 at University of California, Santa Barbara, and now includes eight hubs. Outside of the US these are at Sydney, Delft, Copenhagen and Zurich. All are concerned with the topological approach.
To realise this, advancements are needed into both building quantum devices and in electronic classical interfaces, for the control and readout of what’s happening at the quantum level. Currently required is no less than a “dual revolution” in these areas, believes Reilly.
He also believes success would revolutionise many other areas. One would be the creation of new materials, tailored by computer as required.
“We can figure out what the atoms are that make up that thing you picked up, whether it's a magnet, whether it's an optical device, what its electrical properties are; we can analyse that to a tremendous extent. What we can't do at the moment with our technology is go in reverse,” he explains.
“[What if] I want a magnet that is incredibly strong and can lift whatever… Having a quantum computer gives us the computational power to start to attack quantum chemistry problems and quantum materials problems that reverse that whole process.”
Oft-cited possibilities include answers to incredibly complex optimisation problems, and accurately simulating actual quantum systems, such as the interactions of molecules in chemicals (for example in drug research).
“David talked about designing materials,” adds Carmean.
“We think about one of the things that I'm excited about using the first one for is to actually design the next quantum computer.”
The possibilities are wide-ranging, but so are the difficulties. However, Reilly believes the first-ever such computer in the world could be switched on in Sydney.
Due to the complexity of the area, abstractions and analogies are regular. We ask why the topological approach is the most logical one to pursue. David Pritchard, Chief of Staff for Microsoft's Artificial Intelligence and Research Group, turns to a fable, The Tortoise and The Hare.
“Everyone else is doing a number of things that they're getting a lot of acclaim for, but we're really thinking about how we scale this and taking qubit - that doesn't last very long - and making and adding a lot more to it and having it last for much longer than what they're talking about right now,” he tells create.
“And so in the race, we're the slow pokes. But we're very determined in our approach, whereas they're much more out there in terms of more flash.”