You won’t find a quantum computer in the computer section of Fnac anytime soon. But a new advance proves that this line of research can lead to high-precision calculations, which makes it possible to envisage the future usefulness of these machines.
When can quantum computers actually hit the market? These machines no longer use bits, but qubits, by mobilizing quantum mechanics, thus increasing computing power tenfold. Such a revolution will only happen when this new computing technology has proven its stability, making it possible to actually use quantum computing. Three papers published simultaneously in Nature, on January 19, 2021, demonstrate that a new step has been taken.
These three teams have each achieved historic precision in a quantum system:
- First study: Andrea Morello and his team, in the United States, achieved 99.95% accuracy on 1-qubit operations and 99.37% accuracy on 2-qubit operations.
- Second study: Seigo Tarucha and his team in the Netherlands achieved an accuracy of 99.87% for 1-qubit operations and 99.65% for 2-qubit operations.
- Third study: Akito Noiri and his team, in Japan, achieved 99.84% for 1-qubit and 99.51% for 2-qubits.
Why is accuracy so important?
If these three results are so decisive, it is because quantum computing cannot be envisaged without excellent stability. This comes from its… quantum specificities. Conventional computing relies on binary bits — 1s and 0s — as units of information storage. In your computer, the information takes either the state of a 1 or the state of a 0.
But in a quantum computer, information is stored in qubits (quantum bits): a qubit can be both a 1 and a 0. In a qubit, the 1 and the 0 are therefore potentially superimposed. This is quantum superposition. The same is true for certain particles on the quantum scale of the infinitely small: like Schrödinger’s cat which is dead and alive before both before we look in the box, the object is in two contradictory states before we measure it.
In terms of computing speed and power, the use of qubits is beyond measure. Storage in bits limits the taking into account of variables: each modified variable implies a new calculation. So even supercomputers end up facing a wall in quickly solving complex mathematical problems. A quantum computer, thanks to the superposition of 1s and 0s, is able to calculate all the probabilities at the same time. This makes it possible to obtain an answer to calculations extremely quickly on problems of infinite complexity.
This is all great, but there is a problem to be solved for the system to be functional (because: science!). The quantum world is unstable, difficult to control and to measure. Obtaining a reliable and precise result with qubits is perilous.
Stabilizing quantum bits, user guide
But the scientists whose work is published here in Nature came up with ingenious solutions. For example, Andrea Moreno’s team achieved a great success in 2014 by holding quantum information in silicon for 35 seconds in a two-qubit system — two nuclei of phosphorus atoms. This success was possible thanks to a “trick” based on the isolation of qubits to extract them from their environment and make them more stable over time. Except that this isolation made it difficult for the two qubits to communicate.
They had the idea of introducing an electron to connect the two nuclei. This connection through the electron seems to solve the problem by allowing communication. ” If you have two nuclei that are connected to the same electron, you can make them do a quantum operation says one of the authors, Dr. Mateusz Mądzik. ” As long as you don’t operate the electron, these nuclei securely store their quantum information. But now you have the ability to make them talk to each other via the electron, to perform universal quantum operations that can be adapted to any computational problem. »
The other two teams used another stabilization system. What is important is above all what all of this work published simultaneously demonstrates: high precision of almost 100% is possible. ” When errors are so rare, it becomes possible to detect them and correct them when they occur. This shows that it is possible to build quantum computers that have enough scales, and enough power, to handle meaningful computations. », Relates Andrea Morello.