Triplet Superconductivity and Quantum Qubits

Most current prototypes of quantum computers are using superconducting materials to carry quantum computation, as these materials are able to keep quantum properties more stable, with the major alternative being the so-called ”trapped-ion quantum computer”.

So far, only trapped-ion models have proven to be sufficiently reliable, but they are very limiting in the number of useful qubits they can contain (the quantum computer equivalent of an ordinary computer’s bit).

Of course, the ideal option would be to improve superconducting materials so they become fit for quantum calculations. And some effort has been done in that direction, notably with chirurgie du treilliset with longer-lasting qubits. But still, this is proving not enough to create commercial, scalable superconducting quantum computers.

Another advanced field of computing science is spintronics, which uses the quantum characteristics of particles, spin, instead of electrical charges like in classical electronic computing. So far, quantum computing and spintronics have been somewhat related, but not directly joined, as superconducting materials do not have a spin. At least until now.

(You can learn more about spintronics in our article dedicated to this technology)

A team of researchers at the Norwegian University of Science and Technology and the Università degli Studi di Salerno (Italy) might have discovered a triplet superconductor, a type of superconductor with unique spin properties.

This new type of superconducting material could be a game-changer for the building of superconducting quantum computers. They published their findings in Physical Review Letters, under the title “Unveiling Intrinsic Triplet Superconductivity in Noncentrosymmetric NbRe through Inverse Spin-Valve Effects ».

“A triplet superconductor is high on the wish list of many physicists working in the field of solid-state physics. Materials that are triplet superconductors are a kind of ‘holy grail’ in quantum technology, and more specifically, quantum computing.”

Professor Jacob Linder – Norwegian University of Science and Technology

Meanwhile, another team of researchers at the Niels Bohr Institute at the University of Copenhagen, the Norwegian University of Science and Technology, the Leiden Institute of Advanced Computer Science (The Netherlands), Chalmers University of Technology (Sweden), University of Regensburg (Germany), and the company Machines quantiques have discovered how to defects, a key issue plaguing superconducting material with a new form of efficient fluctuation detection.

They published their findings in Physical Review X2, under the title “Real-Time Adaptive Tracking of Fluctuating Relaxation Rates in Superconducting Qubits ».

Triplet Superconductors

Glissez pour faire défiler →

Pourquoi est-ce important?

In theory, spin could be a perfect medium of transfer of quantum information between qubits and between different quantum computers.

The issue is that in its current form, the technology is just too unstable and the transfer of information too complex to be of practical use.

This might, however, not be true if we got access to triplet superconductors. This is because they can transfer spin without energy loss, so the superconducting particles now carry spin with them.

“Triplet superconductors make a number of unusual physical phenomena possible. These phenomena have important applications in quantum technology and spintronics.”

Professor Jacob Linder – Norwegian University of Science and Technology

So while a more ordinary singlet superconductor can carry power without resistance, a triplet superconductor could also carry spin currents with absolutely zero resistance. As a result, a quantum or spintronic computer could be ultra-fast while also operating while using almost no electricity at all!

Niobium–Rhenium Alloy’s

In their work, the researchers discovered that NbRe, a niobium–rhenium alloy, displays behavior characteristic of a triplet superconductor.

More precisely, they found “inverse spin-valve effect”, a special case of magnétorésistance géante, a magnetic property of multilayer materials, which discovery won the 2007 Nobel Prize.

This is not, in itself, proof that NbRe is a triplet superconductor, but it definitely proves that it does not behave the way a conventional singlet superconductor should.

Long Term Potential

This discovery has extra potential as NbRe is readily available in thin-film form, and the simplicity of the heterostructure makes it especially viable as a potential scalable platform for superconducting spintronics.

In addition, the material works as a superconductor at relatively high temperature (at least by superconducting materials standards), or just 7 degrees Celsius above absolute zero at -273.15 °C (−459.67 °F), while most other candidate materials need as little as one degree above absolute zero.

However, both niobium and rhenium are expensive and rare metals, so they will not directly make quantum computers cheaper.

The next step will be to have other researchers confirm these findings and conduct further tests pointing to triplet superconductivity.

Triplet superconductors can also be used to create a very exotic type of particle called a “Majorana particle”, which is its own antiparticle. Therefore, it can perform calculations in a quantum computer in a stable way.

As other researchers are also getting closer to leveraging Majorana particles and Microsoft already has a chip with Majorana Zero Modes (MZMs), this seems to be an increasingly promising direction for the future advancement of quantum computing.

Detecting Quantum Material Defects

Too Quick Changes

The materials in which qubits are embedded often display defects that are responsible for the unreliability of the qubit. These defects can spatially fluctuate extremely fast, sometimes hundreds of times per second.

So the current method of detection of these defects, which can take up to a minute, is completely insufficient to catch them. In fact, nobody exactly knew how fast this happened until now.

Instead, researchers are forced to measure an average energy-loss rate, which often gives an incomplete picture of the qubit’s true performance.

As a result, quantum computers relying on superconductivity need to rely on many “tricks” to still manage to perform their computation, even when, a lot of times, the qubit has suffered decoherence, without the user being able to detect it.

Using Classic Computers To Help

To speed up the detection of defects, the researchers used a Field-Programmable Gate Array (FPGA), a specialized controller. These specialized chips are not as flexible as the ones used in CPUs or GPUs, but they are ultra-specialized, much faster at a specific task, and less energy demanding.

By running the experiment directly on the FPGA, they could form a “best guess” of how quickly the qubit would lose its energy based on only a handful of measurements.

While this seems like an obvious solution, programming the FPGA correctly was very challenging, especially if the FPGA needs to be a little flexible.

The method they used is that the chip updates its internal “knowledge”, called a Bayesian model, after every single qubit measurement.

Source: Examen physique X

This allowed the system to continuously adapt how it learned about the qubit’s state as efficiently as possible.

“The controller enables very tight integration between logic, measurements, and feedforward: these components made our experiment possible.”

Associate Professor Morten Kjaergaard – Niels Bohr Institute

Toward Real-Time Calibration

Until now, the quantum computing industry had to just “hope” that their qubits were still working, and worked hard at reducing the probability and speed of decoherence.

But this new approach opens the way for calculation actively picking reliable qubits, even with less than perfect materials.

“With our algorithm, the fast control hardware can pinpoint which qubit is ’good’ or ’bad’ basically in real time. We can also gather useful statistics on the ’bad` qubits in seconds instead of hours or days.”

Associate Professor Morten Kjaergaard – Niels Bohr Institute

In the long run, this will open a new field of investigation, where a better understanding of what makes an individual “bad” qubit, instead of relying on averages and guesses.

Conclusion

As at the dawn of electronics, quantum computing progress will come from a multitude of directions.

One important aspect will be the production of better superconducting materials, able to create more stable and durable qubits. And maybe also transport information in the form of a superconducting spin current at the same time.

Meanwhile, improved detection of the decoherence of a given qubit could provide a sensor & software-driven method to radically improve performance without relying on more complex or hard-to-manufacture materials.

Investir dans l'innovation en informatique quantique

Microsoft

Si Microsoft est surtout connu pour sa très forte présence dans les systèmes d’exploitation avec Windows, il est également un mastodonte dans de nombreux autres domaines technologiques.

Par exemple, elle est leader en solutions d'entreprise, notamment Office (Outlook, Word, Excel et PowerPoint), mais aussi les appels d'entreprise (Teams), le stockage partagé dans le cloud (OneDrive), Visio (diagrammes, graphiques), Loop (espace de travail collaboratif) et Access (base de données).

Bien qu'il ne soit pas le leader des services cloud (dominés par AWS d'Amazon), Microsoft représente 20 % de l'infrastructure cloud mondiale via sa plateforme Azure, soit une part aussi importante que les parts combinées de Google + Alibaba + Oracle.

Source: Statesman

Microsoft est également propriétaire de LinkedIn, GitHub, Xbox et de plusieurs des plus grands studios de jeux vidéo au monde.

En matière d'IA, Microsoft s'est davantage concentré sur les cas d'utilisation techniques et les applications commerciales que sur les applications grand public, notamment avec Programme AI4Science, sur les IA utiles à la recherche scientifique.

Cela comprend, par exemple, l’accélération du travail des scientifiques des matériaux pour concevoir de nouvelles molécules ou électrodes de batterie en ayant une IA réduit 32 millions de matériaux potentiels à 500,000 800 candidats, puis à 80 en moins de XNUMX heures.

Source: Microsoft

Des entreprises comme Unilever utilisent déjà cette « chimie générative » pour accélérer leurs découvertes scientifiques.

Jusqu'à présent, en matière d'informatique quantique, Microsoft semblait être à la traîne par rapport à Google ou IBM ; il proposait des services cloud d'informatique quantique avec Azur Quantum. Le service peut également offrir « informatique hybride », mélangeant l'informatique quantique avec un service de supercalculateur traditionnel basé sur le cloud.

Source: Microsoft

As Microsoft released its own Majorana particle-based chip in early 2025, the company has become one of the global leaders in quantum computing.

With new materials like triplet superconductors or new possibilities of real-time calibration, it is likely that Microsoft will be able to keep progressing and integrate these new tools into its own quantum computers.

(Vous pouvez également lire our article putting the spotlight on Microsoft as a whole in more detail to better understand the company).

Dernières actualités et développements concernant l'action Microsoft (MSFT)

Étude référencée

<sup>1. F. Colangelo et al, Unveiling Intrinsic Triplet Superconductivity in Noncentrosymmetric NbRe through Inverse Spin-Valve Effects. Phys. Rev. Lett. 135, 226002 – Published 25 November, 2025. DOI: https://doi.org/10.1103/q1nb-cvh6
2. Fabrizio Berritta, et al. Real-Time Adaptive Tracking of Fluctuating Relaxation Rates in Superconducting Qubits. Phys. Rev. X 16, 011025 – Published 13 February, 2026. DOI: https://doi.org/10.1103/gk1b-stl3</sup>