Building Better Bridges on Quantum Chips
As superconducting quantum computers become larger and more complex, it gets harder to arrange the waveguides and capacitors that are used to control and stabilize the qubits. One solution to the problem is to lift these components above the rest of the chip. Niklas Bruckmoser and Leon Koch at the Technical University of Munich and their colleagues have now demonstrated a fabrication approach that enhances these elevated structures [1].
In most superconducting quantum computers, the “air bridges” that enable this architecture are made of aluminum—the same as the Josephson junctions that form the qubits. Usually, the Josephson junctions are fabricated first, followed by the air bridges atop soft supports that are later removed. Bruckmoser, Koch, and colleagues altered both the materials and the fabrication order. They constructed the bridges from niobium. But first they deposited a temporary layer of aluminum on the soft support. Without that protection, the niobium atoms would penetrate the soft support, spawning impurities that would lodge in the air bridges’ undersides. These steps were performed before building the other components, which meant that the niobium air bridges could be cleaned using methods that would have damaged preexisting aluminum structures.
The new process yielded waveguides with extremely low signal loss, attaining an internal quality factor greater than 8.2 × 106, which is comparable to that of nonelevated waveguides. When used as capacitors in qubits, the structures enabled median qubit lifetimes of 51.6 µs, which is not record-breaking but is long enough to sustain complex quantum circuits. Furthermore, niobium’s superconducting state is sturdier than aluminum’s, meaning niobium-based devices could operate at higher temperatures and in stronger magnetic fields.
–Marric Stephens
Marric Stephens is a Corresponding Editor for Physics Magazine based in Bristol, UK.
References
- N. Bruckmoser et al., “Niobium air bridges as low-loss components for superconducting quantum hardware,” Phys. Rev. Appl. 25, 024007 (2026).