German meticulousness played a decisive role in the discovery of a number of problems with the stability of superconducting qubits. To do this, we had to re-examine the data from hundreds of scientific papers and studies. As a result of the work done, an article by 30 authors was published in the journal Nature Physics, which explains how the likelihood of errors in quantum computing can be reduced by at least one order of magnitude.
It all started in 2019, when two graduate students from the Jülich Research Center (Forschungszentrum Jülich) and the Karlsruhe Institute of Technology, Dennis Willsch and Dennis Rieger, became entangled in data from measurements of the states of superconducting qubits using Josephson tunnel junctions. This model earned Brian Josephson the Nobel Prize in Physics in 1973. It is well represented mathematically and has been widely used to work with superconducting qubits based on transitions for about 15 years. The measurement data was outside the scope of the model, and this forced scientists to look for the root of the problems.
Under the guidance of the professor, the researchers collected data from similar studies by scientists at the École Normale Supérieure in Paris, work with a 27-qubit quantum computer from IBM, and others. As it later turned out, similar deviations in experimental and theoretical data were also discovered by researchers from the University of Cologne. Both groups joined forces and recruited more scientists, re-analyzing hundreds of papers on the topic. The result was surprising. It turned out that in the standard model the description of the operation of Josephson junctions does not take into account a number of important factors, and this leads to calculation errors.
The standard model assumes that the aggregate oscillations (mode) in the Josephson junction system have the form of an ideal sinusoid. In practice, we have achieved such a degree of measurement accuracy that we can notice deviations from the ideal curve. This is all due to harmonics, the strongest of which, as it turns out, affect the measurement results. Previously they were not taken into account in any way. A team of 30 authors has collected so much “compromising evidence” on harmonicas that it is no longer possible to brush them aside. And this is good. Refined formulas for calculating the states of superconducting qubits can lead to quantum bits becoming 2–7 times more stable, which will at least reduce the likelihood of errors by an order of magnitude.
“As an immediate consequence, we believe that the Josephson harmonic will help in the development of better and more reliable quantum bits by reducing errors by an order of magnitude, bringing us one step closer to the dream of a fully universal superconducting quantum computer,” the paper’s authors conclude.
If you notice an error, select it with the mouse and press CTRL+ENTER.