Andrew Dzurak likes to think in terms of the big picture, even if not all of it has come into focus just yet.
An engineering professor at the University of New South Wales in Sydney and the director of the Australian National Fabrication Facility, Dzurak’s goal is to create the first practical quantum computer. And he believes his team’s newly revealed design for a quantum computer chip might represent a crucial first step to that destination — and it’s all building on the silicon chip architecture that powers today’s computers.
“It provides a ‘vision’ or a ‘pathway’ to building a large scale quantum processor, with the millions of qubits that will be needed to solve a range of important problem,” he tells Inverse.
Quantum computing is an advance that would arguably be the defining technological achievement of the 21st century — assuming, of course, we can pull it off in the next 83 years. That’s no certainty, considering a fully operational quantum computer would need to have millions of quantum bits, or qubits, on every chip. The ones under development at places like Google top out at about 50 qubits, with no guarantee those designs can scale upward.
But as Dzurak and his fellow researchers explain in a paper published Friday in Nature Communications, they believe their design can be built up to include the required menagerie of qubits, each taking advantage of quantum weirdness to transcend the limitations of binary and rapidly solve problems that would take traditional computers millions of years.
“This is a very simplistic analogy, but I guess you could say it’s a bit like when the moonshot team had a complete design for the entire mission, including the rocket engines, the timing of the stages, the landing module, the space suits, etc,” he says. “In order to realize a big project you need to have a vision of how everything fits together, and that’s what we’ve aimed to do with this paper.”
Dzurak’s team focuses on silicon quantum chips, one of the five main candidates for quantum computer architecture. Its overriding advantage is that it’s an extension of the silicon chip technology already in use, which offers a rough guide for how to make the qubits small enough to fit millions on a single chip.
“I guess it’s fair to say that I wouldn’t be devoting most of my life’s work to silicon qubits if I didn’t think they were the right way to go,” he says, though he acknowledges the ability to miniaturize those qubits can create further issues. “This is actually an important advantage over other qubits, because it means you can pack many more qubits onto a single chip, but it also creates some challenges in getting so many control lines into a small volume. In part that is one of the key challenges that our paper aims to address.”
The fact these chips share so many features with today’s chips also means they can be built using materials already available and in use. The paper further details how this design solves more technical issues like correcting errors in the qubit’s calculations and building the circuitry needed to control and read all those millions of quantum components.
So how much closer does this get us to a real, practical quantum computer?
“We want to start using the silicon chip manufacturing processes to make a small (say 10-qubit) system first – that’s goal number one – which we hope to achieve in 3 to 5 years,” Dzurak says. “Then we want to build up to a greater level of integration, aiming for say 100 qubits in around 6-10 years. At around 100 qubits we would have a prototype that could then continue to scale up further over time, but which could already be applied to some interesting problems.”
Dzurak says those timescales are highly dependent on how much investment his group receives. Realizing the team’s vision of a true quantum computer will take substantial resources. But at least that vision has never been clearer.
“When I began this design work I wanted to have a visualization of what a complete’ quantum computer chip might look like,” he says. “It’s been very important, since it’s highlighted both the benefits of using silicon, and also the challenges of making an entire quantum processor. There remain very real engineering challenges, which are going to take brainpower and determination to solve, but now we have a real target to aim at.”