PsiQuantum, the company which was awarded a billion-dollar contract by the Australian and Queensland governments to build what might be one of the world’s first quantum computers, has published detailed content about the science behind its proposal.

The US-based company released results which it says show that it is ready to begin construction of the world’s first “useful” quantum computer in the Australian city Brisbane.

The results are detailed in an unedited manuscript published overnight in Nature.

Quantum computers promise to be extremely powerful. These devices harness quantum effects such as superposition and entanglement to perform calculations that can’t be done on classical computers because they are not fast or powerful enough.

That’s the theory anyway. While the potential of quantum computing is undeniable, the reality is that the very same quantum effects which make them such a tantalising prospect also mean they are expensive, error prone and very hard to build.

Physicists and engineers around the world have spent decades experimenting with different materials to make the quantum bits (qubits) for these devices. Some use trapped ions, single electron spins or photons.  

Geoff Pryde, Chief Technical Director–Australia at PsiQuantum, tells Cosmos that the company is building on decades of photonic quantum computing which has its origins in Australia.

“Jeremy [O’Brien, CEO and founder of PsiQuantum] and some of his coworkers demonstrated the first photonic qubit on a chip,” Pryde says, referring to research done in the early 2000s. “That was an academic demonstration at the time, but it really laid the pathway towards the idea of using integrated photonics in silicon – wave guides carrying photons – as a way to reach a large-scale quantum computer.

“And this ‘Omega’ chipset is the demonstration of the maturity of that idea.”

Pryde is referring to the subject of the Nature paper – a set of silicon photonic chips developed by PsiQuantum which they are calling ‘Omega’.

“It contains all the components, all of the functionality demonstrated in a consistent, scalable, manufacturable platform that can enable you to build a modular system that in turn will lead to a million qubits,” Pryde says.

“This hardware platform is now in place. While there might be advances and improvements along the way, there are no fundamental changes to any of the technology that are required now to head towards a scalable quantum computer.”

There are quantum devices based on an array of different base pieces. Some of them even work very well with minimal error. The problem that quantum computer engineers face is scaling those devices to do things which classical supercomputers can’t.

Pryde says that “commercially useful” quantum computers will be able to do “computations in chemistry for pharmaceutical drug design, or catalysts for environmental tech applications or other things in material science”.

PsiQuantum is building 2 facilities the size of data centres in Brisbane and in Chicago, US to bring the large-scale quantum computer to life. The plan is to have the facilities built by 2027.

In issue 105 of Cosmos Magazine, I spoke with other physicists involved in quantum computing. They believed that quantum computing would be able to do something “useful” by the late 2020s or early 2030s.

PsiQuantum’s timeframe sounds too good to be true. Pryde says the key is in the method.

“The quantum system of interest for us is photons which are single particles of light.”

Like in all quantum devices, the qubits can be in a state 0, 1 or a superposition (mixture) of 0 and 1. This sets qubits apart from classical bits which can only be in 0 or 1 states.

“Wave guides running on an optical chip are just the optical versions of wires. You have a pair of these wave guides running side by side. And if the photons in the left hand one, then it’s a 0. That’s in the right hand one, then it’s a 1. And can be in a superposition – in some sense, both at the same time until you measure where it is,” Pryde says.

Pryde says there are advantages to working in photonic quantum computing.

One is that guides can be made with high precision and fabricated virtually identically.

Another is that optical chips can be easily connected to regular computers. “We send information by light, because that’s just a very natural way to transmit it from one place to another. It’s simple, it’s fast, it’s flexible, and it’s a well-established technology,” he says.

Because the photonic quantum computer is based on particles of light, there is no need to convert the information in the quantum device into light before transmitting through optical fibres to a regular computer.

“If you’re working with a different qubit technology then you have to convert from whatever kind of qubit you’ve got, a single atom for example, into light. That’s a really, really hard process to do,” Pryde says.

Finally, Pryde says that, unlike other qubit technologies, photonic quantum computing doesn’t require being cooled to within a few thousandths of a degree above absolute zero. They still require cooling to just above liquid helium – about 2 or 3 degrees above absolute zero.

“This is really kind of the evidence of the maturity of where quantum computing is approaching at a very fast rate,” Pryde comments. “And the idea is to build the world’s first useful quantum computer here in Australia, and then to use it to do useful things.”