All tests passed: DLR QCI accepts 4-qubit demonstrator SQ-RT with Princess QPU from SaxonQ

All tests have been completed, the acceptance protocols signed and the industrial partner informed: The first quantum computer demonstrator has thus passed our acceptance process and can be used in future by our research and industry partners for the development of novel quantum applications.

The system – the SQ-RT with Princess QPU from SaxonQ’s SuNQC project – is now being integrated into the Hamburg Innovation Centre and made available for use by our research and development teams.

But there is no time to rest, neither for us nor for SaxonQ. Work has long since begun on the next milestone: the first scaling stage with 8 qubits by coupling two NV centres. A system with up to 32 qubits will follow later.

Expert eye for detail

Max Kneiss and Robert Karsthof from SaxonQ with DLR QCI project manager Maximilian Kögl start a measurement on the SQ-RT with Princess QPU.

We attach great importance to our acceptance process. Quantum hardware in the DLR QCI has to pass a large number of requirements and tests. For example, the submitted system had to achieve the required gate qualities, gate durations and coherence times on all four qubits. This means that first the development team and then our project manager had to spend a long time repeatedly measuring gates on all four spins – under the conditions under which the demonstrators will later be operated and made available to the research and development teams.

Specifically, the first major milestone we demanded was a mobile room temperature system with four qubits each: one NV electron spin qubit and three core qubits. The quality criteria we defined for this were a gate quality of more than 95 per cent on one-qubit gates and more than 90 per cent on two-qubit gates. The verification of the two-qubit gates in particular was a very complex process with dozens of test routines and hundreds of thousands of gates with which SaxonQ had to verify the gate quality on all four qubits. In this way, however, we were able to set a lower fidelity limit for the entire register instead of cherry-picking the fidelity numbers. These measurement results were then scientifically verified by us and checked for plausibility using our own extensive test series.

For industry-ready products and competitive start-ups

Maximilian Kögl, DLR QCI project manager responsible for SuNQC, during the acceptance of the SQ-RT with Princess QPU at the Ulm Innovation Centre. The system will soon be transported to Hamburg.

We don’t just want to know that the hardware works as specified and achieves a sufficiently high gate quality, for example. It is just as important to us that it can hold its own as a product. This is particularly important for quantum computing technology such as NV centres: The acceptance test is not only passed by our industrial partners and their projects, but it also makes a statement about the platform as a whole. In this way, we are also signalling a high level of technological development to potential sponsors. Such signals from a competent and trustworthy organisation can be decisive for financing.

Commissioning DLR QCI as an anchor customer enables start-ups to transfer the latest research into the industrial production of quantum computers. But it also enables – and requires – the start-ups to undergo a maturing process, both for the technology and for the growing start-up and as a market player. This also includes, for example, useful and complete manuals, clean cable routing and compliance with the necessary technical standards for electronic products. This means rethinking product safety in particular: In contrast to university set-ups, our systems must comply with strict product safety standards for electromagnetic compatibility, protection against electric shocks, laser safety and so on.

With the acceptance of the 4-qubit system, it is clear to us that the foundations for quantum computing based on NV centres have been laid. We can already control small quantum registers at room temperature. The next – very important – step will now be to add several such registers together to prove scalability. Never before has a large quantum computer been built with NV centres. If the step from 4 to 8 and later to 32 qubits succeeds, this would be a clear signal: a quantum computer for industry-relevant applications can be built on this platform.

What to do with four qubits?

“The commissioning by DLR QCI has helped to advance the development of this critical mobile quantum technology.” – Prof. Dr. Marius Grundmann, SaxonQ, at the acceptance test with Maximilian Kögl, DLR QCI.

Four qubits cannot be used to solve mathematical tasks that provide an economic or scientific advantage. However, they can be used to gain important experience with real quantum hardware that a simulator or cloud access cannot provide. And this applies both to our industrial partners, who gain access to and control over real qubits through us, and to our research teams from the DLR institutes: QLearning from the DLR Institute of Quantum Technologies will use it to test the suitability of the quantum processors for various approaches to reinforcement learning, BASIQ from the DLR Institute of Engineering Thermodynamics will use the noise of the quantum computers to simulate molecules in an environment or simulate an H2 molecule on a qubit as an example. In this way, our doctoral students, the projects and companies will gain experience with real qubits, which will give them a head start in later, larger systems.

On the other hand, the experience gained on the application side also flows back to the hardware start-ups and provides them with important feedback on the application-orientated design of their systems. The fact that the application side learns from the supply side and vice versa, and that this knowledge remains in the ecosystem and can be utilised, is what we see as one of the key benefits of the DLR QCI.