"We want to develop a low-error quantum computer with excellent performance data"

Mr. Wilhelm-Mauch, what is it that fascinates you about quantum computers?

I am fascinated by the enormous scope of the topic, which ranges from the fundamentals of quantum physics to the development of concrete technologies. I find it particularly exciting to combine different fields of application, from novel connectors for microwave technology to superconducting components for very low temperatures. Building quantum computers also provides new insights into quantum physics, such as the so-called measurement problem: it is not possible to determine exactly what happens when a particle is measured. The theory is that the state of the particle is destroyed in the process. In building quantum computers, it is possible to study exactly how this works. You could say that we quantum computer researchers are a cross between philosophers and car mechanics.

Did you always want to be involved in this exciting field of research?

Not directly. I came to it through a crisis of meaning. I originally did my Ph.D. on microscopic modeling of Josephson junctions, components that function in superconducting quantum computers in a similar way to transistors in classical computers. At the time, however, my research wasn't focused on quantum computers, and I didn't see any long-term prospects in it, so I wanted to change direction. When I heard about quantum computers, I was immediately fascinated. In 1999, I had the opportunity to work as a postdoc with Hans Mooij in Delft, where I was able to research a new type of superconducting qubit. Qubits are the basic elements of a quantum computer, similar to bits in classical computers. I have been fascinated by quantum computing ever since.

So you witnessed the early years of quantum computing?

Yes, the field was still in its infancy. It was unclear whether quantum computing would be a fad or a long-term area of research. Although the theoretical foundations had been laid in computer science, the experimental side was still very new. I remember giving one of only two talks on superconducting quantum computers at the important American Physical Society meeting in March. That year, there were four or five talks on the same topic going on at the same time. That shows how much has happened in the meantime.

But superconducting qubits aren't the only option. Qubits can also be represented by neutral atoms, ions or light particles. Or do you think that superconducting qubits have won the race?

No, that's still up in the air. However, superconducting qubits are in a good position because their performance can be continuously improved through engineering advances without the need for groundbreaking new discoveries. We know what needs to be done, and there are no showstoppers in sight. But rapid progress is also being made with other platforms, most recently with neutral atoms.

In the QSolid project you want to build a superconducting quantum computer. What exactly will it look like?

We want to develop a low-error quantum computer with excellent performance. The goal is a system with up to 30 qubits that can be integrated into existing computing infrastructures at Forschungszentrum Jülich. Another goal is to establish the entire supply chain in Germany as far as possible in order to be more independent and to strengthen the quantum computing expertise in Germany. The project involves numerous partners, some of them highly specialized, who will tackle the various technical challenges.

Can you give us an example?

A good example is Rosenberger, a company based in the Traunstein district. They are a hidden champion. They develop compact and maintenance-friendly connectors and connection solutions for high-frequency signals - an important innovation that makes superconducting quantum computers more practical.

Is coordinating the many partners the biggest challenge?

The Forschungszentrum Jülich is by far the largest partner, but coordination is also very important. We approach this with a similar mindset to developing large scientific equipment, such as a particle accelerator. We look at the system as a whole. It is very important for us to have a professional approach to systems engineering. To that end, we employ someone with extensive experience in the aerospace industry.

In the space industry? Why is that?

In the world of systems engineering, space engineers are the people you turn to when there are many unknowns in a technology development. And that is the case with quantum computing. They can deal with the fact that once a system is built, it cannot be retrieved. This is also interesting for us.

Since you mention uncertainties in quantum computing research, one of them is how to deal with the error-prone nature of this new type of computer. Google, which is also working on superconducting qubits, recently announced a breakthrough in error correction - not the first headline Google has made with its quantum computers. How do you plan to catch up?

Our approach so far is different from Google's, where errors are not only avoided by good hardware, but actively corrected by software. We focus on hardware with very low error rates. The idea is that as many computations as possible can be done in the time it takes to get to zero errors. We want to achieve low-error qubits through optimized materials, for example.

What kind of optimizations are you talking about?

For example, we are looking at how to improve the fabrication of aluminum nanostructures in superconducting qubits. Purifying aluminum could lead to more stable qubits. We are also looking at new materials for qubit fabrication, such as tantalum.

How far along is the project?

We have already developed a working 10-qubit demonstrator and plan to expand the system to 30 qubits by the end of 2026. At the same time, we are integrating it into the supercomputing infrastructure at Forschungszentrum Jülich so that it can also be accessed by people from outside the center.

The quantum computer will be available to researchers on request. Is it already clear what it will be able to calculate?

The focus is on applications in chemistry, materials research and medical technology - areas where classical computers reach their limits.

Can you give a concrete example?

Sure. We want to test a new method for simulating certain exotic materials, such as high-temperature superconductors.

Superconductors that don't need to be cooled down nearly as much as conventional superconductors, and that would be interesting for everyday technology, like lossless power lines.

Exactly. This new method is particularly suitable for small, low-error quantum computers like the ones we want to build. We are optimistic about its success.

Helmholtz Quantum