The race to build a quantum computer that works in the real world is heating up. Literally.
Many of the quantum computing technologies around the world operate at around 0.1 Kelvin — just a fraction above absolute zero (-273 degrees Celsius).
To keep the technology that cold, scientists need to use special cryogenic refrigeration techniques.
But now, two teams of physicists — one in Australia, the other in the Netherlands — have developed quantum technology in silicon that operates up to 1.5 Kelvin, they report in the journal Nature.
That jump in temperature might not seem like much to us, but that’s hot at the quantum scale and makes a huge difference, said Andrew Dzurak of the University of New South Wales and co-author of the Australian research.
Not only could it significantly reduce the cost and size of a quantum computer, he said, but it puts the technology on the path of being able to be integrated with existing technology and scaled up.
“Everyone in the community accepts it’s at least a decade away until we get quantum computers that are powerful enough to really solve real world problems,” Professor Dzurak said.
“This all of a sudden takes away one of the biggest stumbling blocks to achieving that and could shave years off achieving that goal.”
Why is temperature such a big deal?
In traditional computing, bits are either 1 or 0, while in a quantum computer, qubits can be both numbers at the same time.
With the ability to perform multiple calculations at once, the hope is that quantum computers can rifle through huge databases in the quest to quickly solve problems for things such as drug development and complex algorithms.
Last year Google announced it had achieved “quantum supremacy” by linking together 53 qubits on a superconducting chip.
But, Professor Dzurak said, quantum computers need millions of qubits on one chip to solve real-world problems.
And the more qubits there are, the more heat is generated, resulting in a loss of information.
Qubits have been traditionally refrigerated using two different types of helium isotopes, one of which can only be produced in nuclear reactors, to keep them close to absolute zero.
Not only is this process expensive, but all of the cables connecting the qubits to conventional chips like those in your mobile or laptop need to be cooled down to this temperature.
“It’s physically impossible and an engineering nightmare,” Professor Dzurak said.
“The idea is now we can get a much more miniaturised system, that’s much more engineerable and cheaper.”
What did they do?
In 2015, Professor Dzurak and colleagues miniaturised and cooled a metal oxide semiconductor — or transistor — used in current electronic devices down to 100 milliKelvin to create a quantum dot that contained electrons.
While they demonstrated they could write information on the spin of electrons trapped in the bottom layer of silicon on each dot, they did not have the technology at the time to read the information.
“A quantum bit is only useful if you can read its state and say now it’s zero, now it’s one.”
Over the next three years, Henry Yang, a post-doctoral scientist on the team, developed a technique that forced the electrons to tunnel towards each other between the quantum dots, enabling the qubits to be read.
“Henry realised that by doing that we could operate up at high temperatures,” Professor Dzurak said.
Late last year, the second team led by Menno Veldhorst at Delft University of Technology in the Netherlands conducted a similar experiment with similar results.
“What this shows is that the technology is very reproduceable, very reliable. It’s not just a one-off,” Professor Dzurak said.
How would a quantum computer work?
In this vision of a quantum computer, the qubits would sit alongside, and be connected to, regular silicon chips, which would control the quantum processor.
Not only would this make them smaller but, because both types of chips operate above 1 Kelvin, they could be cooled using conventional refrigeration techniques.
While it’s way too early to say whether or not this technology will win the quantum race — there are many other approaches to making qubits in development — the research tackles a “neglected problem”, said David Reilly, a quantum physicist at the University of Sydney and Microsoft.
“It’s fantastic to see research efforts now turning to the challenges that will limit scalability, and temperature is one of them,” Professor Reilly said.
The ability to demonstrate the technology can operate at much higher temperatures opens the pathway to scale up from a handful of qubits to millions of qubits.
But although both experiments showed the qubits could operate above 1 Kelvin, he said the results also showed that the quality of information on the qubits was affected by the higher temperatures.
“It’s unclear or uncertain yet as to whether or not that degradation of quality will be so precious that operating them at higher temperature won’t be viable,” Professor Reilly said.
But, he added, the scientific progress made in the experiments was important for ongoing research, not just into quantum computing, but other areas of physics as well.
“Even if the performance is degraded that just rapidly speeds up the research cycle.”