MATQu – A Value Chain for Europe’s Quantum Computing Chips

Materials for Quantum Computing (MATQu) was a pan-European project to pilot the capabilities of a quantum hardware and materials value chain for manufacturing superconducting transmon qubits in Europe on the industrial scale. The project aimed at establishing approaches for producing superconducting qubit elements, such as resonators and Josephson Junctions structures, by using 300 mm wafer-scale CMOS-compatible processes. By addressing the complementing aspects of substrate technology, process flow, materials, 3D packaging and scalable cryogenic test and characterization capability, the consortium has identified pathways to accelerate the time-to-market and roll-out of improved superconducting qubits for quantum computing.

After 3 years the project came to a closing in the final review session last week. The consortium consisting of 17 partners and Fraunhofer Institutes in the coordinating role, gathered in Delft to present the result to the reviewers from Brussels. Next to the evaluation of results, the delegation was invited to see the facilities of the two Delft based project partners OrangeQS and Delft Circuits.

When talking about CMOS-compatibility and producing quantum chips in foundry-compatible processes, one must keep in mind that the large-scale market-introduction of novel materials, devices, and characterization represents a great challenge in this industry. Transistors have a history of more than half a century of fabrication optimization and successive generations are built upon each other. Breaking out of this path dependency is difficult, yet it presents an opportunity for Europe to play a role in the High-Performance Computing (HPC) industry. The motivation behind MATQu was to establish a complete European value chain for building superconducting qubit elements such as Josephson junctions and resonators for superconducting qubit chips.

One key difference to classical CMOS fabrication is the dominant role that electrical inspection plays in the characterization of quantum chips. While wafer-probers and in-line metrology can be used to validate the functionality of transistors, it is currently not possible to retrieve the performance and quality metrics of quantum chips using these methods. Due to the lack of correlations between materials and the required functionality of these devices, the quantum-mechanical behavior needs to be observed at temperatures close to absolute zero. Thus, a new end-of-line test system is needed to keep cryogenic characterization capabilities on par with increasing fabrication volumes. Addressing this need, OrangeQS is the first company to launch an integrated HW/SW solution for high-throughput characterization of superconducting qubits.

Quantum chip testing tools GRACE SCQT Quantify graph-based automated calibration execution OrangeQS

Subsystems that are integrated by OrangeQS to build automated qubit test equipment. The Automated Test Routines subsystem as a whole has been the focus of the MATQu work, specifically graph-based automation routines, where a 5 qubit tune up graph is shown as example.

In the figure above we depict the subsystems needed for an automated test solution, where the software and automation frameworks are an essential part. In the context of the MATQu project OrangeQS developed a framework of qubit test and diagnostics routines up to the level of benchmarking qubit performance with a simple VQE algorithm. In addition to this main task and in order to accelerate the testing of qubits further, a development took place to adapt and improve the graph-based automated calibration execution software (GRACE).

This project has been an important step towards a mature quantum chip industry and has proven that all required capabilities and resources are available within Europe. More information can be found on the project’s website.