Oxford Ionics scores funding for scalable quantum chip technology
Hermann Hauser of Acorn fame among the investors providing cash injection
A UK quantum startup has secured £30 million ($36.3 million) in Series A funding to help advance its technology, which it claims uses trapped ions as qubits but does not need lasers to control them, making it scalable through existing silicon manufacturing processes.
Oxford Ionics is based near Oxford, as the name implies, and Reg readers may be interested to hear that one of its investors is none other than Hermann Hauser, of Acorn Computers fame. Others include Oxford Science Enterprises, Braavos Investment Advisers, Lansdowne Partners, and 2xN.
The company's technology is based on trapped ions, using individual charged atoms as qubits. But while other quantum outfits with this technology such as IonQ typically use lasers to manipulate the atoms, Oxford Ionics does away with this and instead uses electric and magnetic fields produced by its silicon chip to control the qubits.
This is actually a more natural approach, according to co-founder Dr Chris Ballance.
"If you use electronic currents flowing in the chip, you can directly interact with the spin with the qubit," he told The Register.
However, controlling qubits with electronics has previously been found to involve very weak interactions, slowing down the whole process, and so Ballance and fellow co-founder Dr Tom Harty have spent years developing the technology to improve the strength of these interactions.
"We had the challenge to make the fields intense enough, which requires lots of semiconductor fabrication knowledge and antenna design knowledge to build the chip," he said.
This involved developing complex three-dimensional structures on the silicon chips that form resonant antennas, and then building up techniques to localize the electric and magnetic fields produced.
"Once you've done that, you can then make these chips very large, very fast, by just copying these repeating motifs," Ballance said, in the same way that standard semiconductor chips can be scaled up by copying and pasting design elements to make a more complex integrated circuit.
"We can do the same with quantum computing chips," he said.
According to Oxford Ionics, this has already been demonstrated via its partnership with semiconductor manufacturer Infineon Technologies, using its standard production line. However, the design of the chip means it can be manufactured in a variety of ways in lots of different foundries, Ballance told us.
The atoms that make up the qubits are held by these fields about 50 microns above the surface of the chip so that any small imperfection in the surface will not affect the qubits.
- Chinese researchers' claimed quantum encryption crack looks unlikely
- Quantum entanglement discovery could enable futuristic comms tech, Nuclear physicists say
- Quantum computing is a different kind of computing, says AWS
- Spooky entanglement revealed between quantum AI and the BBC
One advantage of this technology, according to the company, is that it produces higher quality qubits than some other quantum approaches, which means there is less need for large numbers of physical qubits lumped together to make a logical qubit to allow for error correction when performing quantum calculations.
"Useful problems start creating value at a few hundred qubits, and we believe we can get to a few hundred qubits with just a few hundred physical qubits," Balance told us.
"Pretty much everyone else in the market has error rates that are so high they can't run the few hundred qubit problems on a few hundred qubit machine. They need error correction. But because of how high quality our qubits are, we can get there with just a few hundred physical qubits without requiring warehouse-sized machines with millions of qubits," he claimed.
But the company admits that its technology, like other quantum gear, is still at an early stage of development. It aims to spend the next year or so building chips and intentionally breaking them as it pushes the technology to see how dense and compact it can make the features, and understanding what kind of emerging problems get thrown up and work out how to scale up successfully.
Currently, on a chip about the size of a postage stamp, Oxford Ionics could deliver thousands of qubits without any changes to its design, Ballance said, and that is using a 1 micron production process, way behind the cutting-edge 5nm or 3nm production nodes.
Ballance said that Oxford Ionics expects to be raising more funding in about another 18 to 24 months. "The current funding gives us a good two years of growth, plus a bit of padding to get to the next generation of technology, and then we'll be raising again," he said.
But for the near future, the company is aiming to stand up a quantum computing cloud service that will allow customers to try out its technology.
"At the moment, quantum computers are advancing so fast that if you order some new hardware, by the time it turns up and you have it installed, it is out of date," Ballance said. "So it is better value for customers to be able to dial into our infrastructure, then for us to continuously upgrade it to keep it at the highest performance level."
The service will be initially available only to alpha customers for the next year or so while the company works at validating the technology.
"We want to really focus on getting to the generation of devices which will be able to really meet the markets, rather than just allowing people to test out their assertions," Ballance said.
And how did Hermann Hauser get involved?
"I had a dinner at the Royal Society and Hermann Hauser was there," Balance said. "I spent a while telling him about how this is the way you should build a quantum computer, and he looked me in the eye and said, if you're so confident, why don't you start a company? And when you do, come to me, and I'll back you. And that's what we did. And that's what he did." ®