Researchers in Finland have unveiled a new type of microwave controller that solves one of the big obstacles to scaling up quantum computers to the size where they can be usable.
The controller, developed mainly by Aalto University and VTT Technical Research Centre of Finland, can work at close to absolute zero, the extremely low temperatures used in quantum computers. The tiny — less than a millimetre across — device can therefore be placed inside the same supercooled space as the quantum processor it is sending the signals to.
It eliminates the need to run huge numbers of wires into the quantum processor.
Cabling may sound like a trivial issue, but it is about to become a big bottleneck for scaling up quantum computers, says Mikko Möttönen, professor at Aalto University and VTT, who led the team.
The cost of the wiring alone can be €1m on a quantum computer
“Each qubit currently needs three wires to carry the microwave signals in and out. It means the refrigerators cooling increasingly more qubits have to become larger to accommodate all the wires, and have to work harder because there is heat leakage from the cables,” he says. “It also adds to the cost, the cost of the wiring alone can be €1m on a quantum computer.”
👉 Read: The different types of quantum computer startups, explained
This is rapidly becoming a limiting factor in scaling up quantum computers, at least those that are based around superconducting qubits cooled to near absolute zero. The biggest existing computers of this type have about 100 qubits. A few hundred qubits could create an advantage in some practical problems but a general-purpose quantum computer may need a thousand-fold that number.
Superconducting quantum computers with just a handful of qubits are already huge and bulky — golden chandeliers of endless wiring with an accompanying refrigeration unit the size of a small car. Scaling them beyond thousands of qubits with the same design is unfeasible, so any incremental shrinking of components is an important step.
There are competing types of quantum computers that don’t rely on supercooled qubits. PsiQuantum, for example, is developing a photonics-based quantum computer that can operate at room temperature, although it needs supercooled detectors. Similarly, Quantum Brilliance, an Australian-German startup that recently came out of stealth, is developing a diamond-based quantum computer that could one day be miniaturised to the size of a desktop computer.
Superconducting quantum computers will risk losing out in the long run if they cannot shrink as well. Möttönen says his vision is to one day have a superconducting quantum computer that will have only one power cable and a single line bringing the data in and out. However, he says it is “not likely to happen in this decade”.
Several challenges still remain, even when it comes to the microwave controller. Currently the controller Möttönen’s team has developed can only emit a single continuous wave, but to really control the qubit it will need to be able to emit pulses. That's the next step for his team.