Although no one knows yet what the future quantum computer will look like or exactly what it will be able to do, the radically different way in which these futuristic machines calculate is fueling the hope among experts that they will soon be able to crack encrypted messages or develop medicines tailored to the individual. .
That superpower starts with the so-called qubit, the counterpart of the traditional bit. It can be, dictated by the laws of physics, not only ‘0’ or ‘1’, but also both at the same time. The question is what physical appearance such qubits will take in the future quantum computer.
One of the options are so-called quantum dots, calculation units that use the familiar computer chip material silicon. In an article in the magazine Nature Electronics Researchers from the quantum institute QuTech in Delft and computer chip giant Intel describe how they produced these parts with existing machines from the company’s factories.
This gives the quantum dot good credentials to become the definitive quantum successor to the transistor, the calculation unit that in traditional computers, billions of them simultaneously sit on computer chips and perform calculations. A full-fledged quantum computer, one that can perform every imaginable quantum calculation, must in the future contain millions to billions of such computing units.
From pen to printer
In the experiment, flat discs with a diameter of 30 centimeters, on which tens of thousands of quantum dots are located, rolled out of the machines. At least 98 percent of these worked properly. The researchers also analyzed one quantum dot more extensively, and they showed that it indeed exhibited the desired qubit behaviour.
‘In the past, the flag was raised in the lab when half of our quantum dots worked well,’ says physicist Anne-Marije Zwerver (QuTech), lead author of the article. ‘It’s like taking the step from writing with a pen to printing something out with a printer. The first is slow manual work, the second is uniform production, large numbers, but more rigid.’
Physicist Floris Zwanenburg (University of Twente), who also conducts research into quantum dots, is enthusiastic about the result. ‘This is great work. The group started working with Intel years ago, and that is now producing the desired effect,” he says. He praises the successful merging of the research cultures of the academy and the business community. ‘When we as scientists make quantum dots in a cleanroom, we are happy when one works. When Intel makes a thousand transistors and one doesn’t, they immediately wonder why. You have to combine those two methods if you want to be able to make quantum computers on an industrial scale in the future.’
The quantum dot currently lags somewhat behind the lab-produced variants that tech giants such as Google and IBM use in their quantum processors, but experts expect that they can overtake the competition in the long run due to two advantages. They are much smaller, which makes building machines with millions or billions of qubits easier. And they use silicon, which allows them to rely on the existing infrastructure and manufacturing techniques of chip giants such as Intel.
‘But the latter was really only theory before,’ says Zwerver. ‘We have now proved for the first time that it also works out that way in practice, on existing machines in a real factory.’
