Warwick-led chip design uses phonons to let distant qubits talk to each other

[Nicola47]

Researchers led by the University of Warwick have proposed a new way for distant qubits to exchange quantum information, Live Science reports. The approach, called Quantum Phononic Links, uses phonons as a quantum bus. Phonons are quasiparticles that carry vibrational energy through a material, a bit like packets of sound. The study was published in APL Quantum

The problem it tackles is a big one. In many quantum chips, qubits struggle to share information unless they sit right next to each other, which limits how circuits can be laid out and how far systems can scale. Phononic links could connect qubits that are far apart on the same chip. The researchers say the approach avoids the complex surface acoustic wave designs used in some earlier ideas

The prototype uses compressively strained germanium on silicon, a material system that Warwick has worked on for years. Qubit separations in the design range from under one micrometre to 300 millimetres. Lead researcher Maksym Myronov, an associate professor of semiconductor materials at Warwick, said: "Our work introduces a new concept in which phonons act as a quantum bus, enabling distant qubits to exchange quantum information."

The team believes the approach could support machines with more than one million semiconductor hole spin qubits, as well as hybrid quantum and classical systems. Silicon and germanium based qubits are attractive because they could use existing chip manufacturing, as Quantum Motion is doing in London. Connecting large numbers of them is one of the main hurdles. If phononic links work in practice, they could help with that

Is connecting qubits over long distances the real bottleneck for silicon based quantum computers? And is it good to see more UK university research in this area?

New quantum chip taps into weird quasiparticles to get qubits to communicate over long distanceswww.livescience.com

[NoMercyMatthew89]

Connectivity is a huge issue for spin qubits. They are tiny and packed close together, but only talk to neighbours. A quantum bus that reaches further would make error correction codes much easier to implement. Error correction codes like the surface code assume nearest neighbour links. Longer range links open up more efficient codes

[CosmicRay65]

Germanium is having a moment in quantum research. Hole spin qubits in germanium have shown some impressive results over the last few years

[GradientHydra]

Good to see Warwick in the news. It has a strong semiconductor group that does not get as much attention as Oxford or Cambridge. Their germanium on silicon work has been going for well over a decade

[Maya98]

This is a proposal and prototype design, not a working million qubit machine. Worth keeping that in mind

Lots of elegant ideas never survive contact with real fabrication. The paper was published in June, so it has had a few months of scrutiny already. Still early days