Microsoft Invents A New State Of Matter For Majorana-1 Quantum Computing Chip

New State Of Matter To Scale Quantum Computer

It has been an eventful few months for progress in quantum computing and the companies we covered in “5 Best Quantum Computing Companies of 2025”.

It started with Google’s Willow in December 2024, maybe the first-ever scalable quantum chip. It then followed with the news of the first distributed quantum computing across an optical network link, opening the way to quantum computers being networked like normal ones in dedicated servers.

Now it is Microsoft’s turn to make a big splash with the presentation of Majorana 1, a chip using an entirely new state of matter to perform quantum computing: topoconductors.

Microsoft claims topoconductors can produce more reliable and scalable qubits, the building blocks for quantum computers.

This entirely new path for quantum computing radically changes how Microsoft is aiming to build its future quantum computers, with “a clear path to fit a million qubits on a single chip that can fit in the palm of one’s hand.”

What Are Topoconductors?

It seems the origin of the breakthrough stems from a new approach, looking at creating a transistor dedicated to quantum computing, going beyond what has been done until now.

“We took a step back and said ‘OK, let’s invent the transistor for the quantum age. What properties does it need to have?

And that’s really how we got here – it’s the particular combination, the quality and the important details in our new materials stack that have enabled a new kind of qubit and ultimately our entire architecture.”

Chetan Nayak, Microsoft technical fellow

Topological superconductors, described in the corresponding paper published in Nature¹, under the title “Interferometric single-shot parity measurement in InAs–Al hybrid devices“, are a state of matter different from the more familiar one of solid, liquid, or gas, or even the more exotic ones like plasma of Bose-Einstein condensate.

Topological state was only theorized until now, first by Ettore Majorana (1906-1938), but suddenly it seems it is not only observable but even controllable. This Majorana particle (also called Majorana fermion), a particle that is its own antiparticle, was observed by Microsoft’s researchers for the first time in 2024.

Majorana particles are similar to electrons in some way and could be used to preserve quantum data useful for quantum computation.

This was not at all an overnight success and has apparently been the result of more than 17 years of research, Microsoft’s longest-running research project, and until now a very well-kept secret.

To simplify things (a lot), a topoconductor is a semiconductor sharing some of its behavior at the atomic and sub-atomic level with superconductor materials.

The way this was achieved is by merging together in a wire indium arsenide (a semiconductor) and aluminum (a superconductor).

Source: Physical Review

When cooled to near absolute zero and tuned with magnetic fields, these devices form topological superconducting nanowires, containing so-called Majorana Zero Modes (MZMs) at the wires’ ends.

Source: Microsoft

Ultra-Stable Quantum State

In a “normal” superconductor, any unpaired electron can be detected because its presence requires extra energy. This makes its measurement simple, but also makes it very sensitive to noise and perturbations from the environment, making any quantum calculation difficult.

MZMs are radically different, as an unpaired electron is shared between a pair of MZMs, making it invisible to the environment. This unique property of Majorana particles protects the quantum information, making it ultra-stable and reliable.

Invisible Electrons

Of course, while this is ideal for preserving the quantum state in a stable and useful condition, it also makes any actual measurement of it extremely difficult, hence why Majorana particles have been theoretical only for a century until extremely recently.

While this makes our topoconductors ideal candidates for qubits, it also presents a challenge: How do we read quantum information that is so well hidden? How can we distinguish between, say, 1,000,000,000 and 1,000,000,001 electrons?

Microsoft’s solution to the issue leveraged quantum dots, a unique material we discussed extensively in “Investing in Nobel Prize Achievements – Quantum Dots & Nanocolors”. It can be described as a tiny semiconductor device that can store electrical charge.

The quantum dot is put at the end of the topological nanowire. This connection increases the dot’s ability to hold charge. Crucially, the exact increase depends on the parity of the nanowire.

Source: Microsoft

So, by measuring the state of the quantum dots, a well-understood process using microwaves, the system can also measure the otherwise invisible quantum state of the Majorana particle.

Source: Nature

Not only is the measurement possible, but it is also extremely reliable, even with a first prototype, before any further optimization is done.

We designed our devices so these changes are large enough to measure reliably in a single shot. Our initial measurements had an error probability of 1%, and we’ve identified clear paths to significantly reduce this.

Ultra-Reliable Qubits

It completely changes the approach to quantum state measurement used in quantum computing.

Until now, this required the rotation of quantum states through precise angles, requiring complex analog control signals customized for each qubit. It made error correction, relying on the same method extremely complex, expensive, and overall less reliable.

Instead, the method discovered by Microsoft can simply correct errors by connecting and disconnecting quantum dots from the nanowires, using a digital pulse.

Source: Microsoft

If you are interested, you can learn more about the details of how topoconductors have been developed in this long interview with Dr. Chetan Nayak, the leader behind this Microsoft’s project.

Inherently Scalable Architecture

Because the system is so much simpler in terms of engineering, if not in terms of particle physics, and also more reliable and stable, it is naturally easier to scale up.

The basic component would be a “tetron”, made of 2 nanowires, 4 MZM, and 4 quantum dots, creating a 2-qubit device.

When paired with another one, it can form a basic two-qubit device supporting a method used for quantum computation called “measurement-based braiding transformations”.

A 4×2 array of tetrons could perform error detection on two logical qubits.

Source: Microsoft

This block could then be replicated dozens, hundreds, or ultimately thousands or millions of times to build a massive quantum computer, much larger than anything envisioned until now.

It’s perhaps not surprising that quantum computation would require us to engineer a new state of matter specifically designed to enable it.

What’s remarkable is how accurate our readout technique already is, demonstrating that we are harnessing this exotic state of matter for quantum computation.

You can see how the Majorana chip looks from this render, starting at the Majorana particle and ending with the whole chip held in a person’s hand.

Finally, the key part that makes this technology especially scalable, is how small the physical components are. As a result, more than a million physical qubits can be inserted into a small chip that can hold in a hand.

Source: Microsoft

Further Improvement

As explained, even the 1% error probability in the measurement of the quantum dots and MZM can be decreased further. Microsoft’s researchers are already seeing a way to do it.

As errors in computing compound on each other, decreasing by 10x the error rate could increase much more the ultimate useful computation potential.

Another thing that can be improved is the overall quantum state stability of the Majorana particles.

“External energy—such as electromagnetic radiation—can break Cooper pairs, creating unpaired electrons that can flip the qubit’s state from even to odd parity. However, our results show that this is rare, occurring only once per millisecond on average.”

As the system developed by Microsoft has demonstrated impressive stability, it indicates the shielding is already doing its job well. However, there are likely more ways to reduce interference further, which are also already under investigation.

Accelerating The Quantum Computing Revolution

The next step for Microsoft is to build the 4×2 tetron array beyond the initial prototype and test it at scale.

And then use the entire eight-qubit array to implement quantum error detection (QEC) on two logical qubits.

Because the topological qubits have built-in error protection, it simplifies QEC greatly. In addition, Microsoft claims its custom QEC codes reduce overhead roughly 10x compared to the previous state-of-the-art approach.

So not only are Majorana-based qubits more reliable, but they can run faster and require fewer physical qubits to produce one logical qubit.

(logical qubits are the useful measurement units for practical applications, the same way a processor is measured in operation per second and not only by how many transistors it contains)

“We believe this breakthrough will allow us to create a truly meaningful quantum computer not in decades, as some have predicted, but in years.”

Satya Nadella – Microsoft CEO

Applications

Science

Microsoft’s communication around Majorana 1 is mostly focused on the scientific results it could create, especially in biology and material sciences.

These are very important fields for quantum computing, as from protein folding to complex materials for batteries, the calculation of how they behave at the atomic level is extremely greedy in terms of computation, pushing to the limit the current supercomputers.

Scalable quantum computers would be able to simulate these problems millions or even trillions of times more efficiently, likely allowing for a massive stream of new discoveries.

This will be done not only by performing previously impossible calculations, but also by avoiding billions of dollars in exhaustive experimental searches and wet-lab experiments.

Among the endless possibilities, a few were mentioned by Microsoft:

• Self-healing materials that repair cracks in bridges.
• Sustainable agriculture.
• Safer chemical discovery.

In a 1-hour long interview, the CEO of Microsoft even explained that from his perspective, the way to see if quantum computing and AI perform adequately is global economic growth.

“The real benchmark is if the world is growing at 10%” more than any other metrics.”

Satya Nadella – Microsoft CEO

Cyptography & Defense

Quantum computers have the potential to break almost all the currently used encryption methods, including for military communications, nuclear codes, banking transfers, etc.

So it is of course an absolutely crucial question for the US government (and all other major powers) to not be caught off guard by progress in quantum computing.

The Defense Advanced Research Projects Agency (DARPA) has selected Microsoft as one of two companies to advance to the final phase of their rigorous benchmarking program known as Underexplored Systems for Utility-Scale Quantum Computing (US2QC), (the other company being photonic quantum computing PsiQuantum).

US2QC brought together experts from DARPA, Air Force Research Laboratory, Johns Hopkins University Applied Physics Laboratory, Los Alamos National Laboratory, Oak Ridge National Laboratory, and NASA Ames Research Center.

Quantum Computing And Majorana Particle Company

Microsoft

While Microsoft is most known for its very strong presence in operating systems with Windows, it is also a juggernaut in many other tech fields.

For example, it is the leader in business solutions, including Office (Outlook, Word, Excel, and PowerPoint), but also company calls (Teams), cloud-shared storage (OneDrive), Visio (diagrams, charts), Loop (collaborative workspace), and Access (database).

While it is not the leader in cloud services (dominated by Amazon’s AWS), Microsoft is making up 20% of global cloud infrastructure through its Azure platform, as large as the combined shares of Google + Alibaba + Oracle.

Source: Statista

Microsoft is also the owner of LinkedIn, GitHub, Xbox, and many of the world’s largest videogame studios.

Source: Microsoft

When it comes to AI, Microsoft has been more focused on technical use cases and business applications than consumer apps, notably with the AI4Science program, on AIs useful for scientific research.

This includes, for example, speeding the work of material scientists to design new molecules or battery electrodes by having an AI narrow down 32 million potential materials to 500,000 candidates, and then to 800 in less than 80 hours.

Source: Microsoft

Companies like Unilever are already using this “Generative Chemistry” to speed up their scientific discoveries.

Until now, when it comes to quantum computing, Microsoft had seemed to be lagging compared to Google or IBM; it was offering quantum computing cloud services with Azure Quantum. The service can also offer “hybrid computing”, mixing quantum computing with traditional cloud-based supercomputer service.

Source: Microsoft

Now that the groundbreaking hardware leveraging Majorana particles has been revealed, it completely changes the actual position of the company.

Far from lagging, it was simply polishing its great announcement, and building up the software and use cases that its industrial clients will use with its scalable quantum computers.

This is a radical departure from a company mostly focused on software over hardware, with the exception of the Xbox console.

And one that could prove extremely profitable if it turns out the Majorana particle is the key to scalable, and ultra-powerful million-qubit quantum computers, providing extra growth to an already massive company.

(You can also read our article putting the spotlight on Microsoft as a whole in more detail, beyond quantum computing, to better understand the company).

Study Reference:

^{1. Microsoft Azure Quantum., Aghaee, M., Alcaraz Ramirez, A. et al. (2025) Interferometric single-shot parity measurement in InAs–Al hybrid devices. Nature 638, 651–655. https://doi.org/10.1038/s41586-024-08445-2}