Adapting existing transistor designs for mass-produced quantum computing devices

Quantum computing has the potential to revolutionize nearly every scientific and technological field, but functional devices are currently difficult to acquire or produce. Some researchers have access to dedicated facilities for producing quantum computing hardware, but these can be of inconsistent quality. Alternatively, ordering quantum devices from a manufacturer can be expensive.

Michniewicz and Kim outlined potential approaches to modify commercial transistors to become effective quantum devices. With a few modifications, large industrial fabs could produce low-cost, high-quality qubits.

“Something like a single-triplet qubit or a single-spin qubit will require adding extra components like an antenna or a nanomagnet,” said author John Michniewicz. “For exchange-only qubits and hole-based qubits, the promise is that manufacturers barely need to adapt anything, all they need to do is to make sure the design and geometry is qubit-friendly.”

Of course, these mass-produced quantum computing devices can never match the capabilities of custom hardware, and they will likely not be suited for niche applications that require specialized designs. However, they will still be broadly useful for basic research, education, and testing.

The authors hope that by demonstrating the feasibility of these simple quantum devices, large industrial fabs will be encouraged to begin producing them. With easier access to quantum computing devices, far more research groups will be able to contribute to the field.

“What I would like to see is more people getting more devices quickly and cheaply,” said Michniewicz. “They will not need to write a grant proposal or make their own devices. Furthermore, this would enable the groups that are not already involved in quantum computing to get involved and do experiments.”

Source: “Leveraging off-the-shelf silicon chips for quantum computing,” by John Michniewicz and Myungshik Kim, Applied Physics Letters (2024). The article can be accessed at https://doi.org/10.1063/5.0207162 .