Fabricating single-photon detectors from superconducting aluminum nanostrips

In quantum computers, information is often carried by single photons and picked up by structures named superconducting nanostrip single-photon detectors (SNSPDs). In principle, traditional type-I superconductors would be easier to integrate into existing quantum computing architectures than the type-II materials more widely used today. So far, however, this possibility hasn’t been widely explored.

New research published in Superconductivity shows how Lixing You and colleagues at the Chinese Academy of Sciences, Shanghai, China have for the first time successfully fabricated an SNSPD using thin films of the type-I superconductor, aluminum, and used the structure to detect single photons of visible light with extremely high efficiency.

Compared with the type-II superconductors more commonly used in SNSPDs so far, aluminum is more compatible with the latest quantum computing architectures.

“This potential compatibility facilitates the integration of aluminum-SNSPDs into existing quantum computing systems, paving the way for more complex and sophisticated chip-based devices for processing quantum information,” explains co-author Hui Zhou.

When a superconductor is cooled to below its critical temperature—often very close to absolute zero—its resistance to electrical current abruptly drops to zero. This property makes superconducting nanostrips ideally suited to detecting extremely low light levels.

When it absorbs a photon’s energy, a nanostrip will temporarily heat up above its critical temperature at a local scale, increasing its resistance and thus creating a drop in the electrical current passing through. By detecting this drop, a quantum computer can then register that a photon has been detected. Afterwards, the excess heat will quickly dissipate, and the nanostrip’s superconducting properties return.

A key indicator of an SNSPD’s efficiency is its count rate: the proportion of incoming photons it manages to detect, without incorrectly registering a detection. Since type-II superconductor nanostrips are both easier to produce, and more robust under varying conditions, they have so far remained the dominant choice for SNSPDs. But as Zhou explains, type-I superconductors such as aluminum come with their own advantages.

“Compared with type-II materials, thin aluminum films have a lower resistivity, which is expected to improve photon counting rates,” she says.

“If it were possible to develop aluminum-based SNSPDs and integrate them with the current aluminum-based superconducting quantum computing architecture onto a chip, they would be better suited for processing electronic and optical information.”

To tackle these challenges, You and Zhou’s team looked to a unique property of aluminum nanostrips: as their thickness decreases, their critical temperature increases. This means that less energy is needed to cool them to maintain their performance, making them easier to integrate into quantum computing architectures.

While these ultra-thin type-I strips are less robust and more difficult to fabricate, You and Zhou’s team has developed a technique to overcome these hurdles.

“Since thin aluminum films are highly susceptible to oxidation, we grew an aluminum nitride layer on top of them for protection,” Zhou explains. “We also grew a layer of silicon dioxide to prevent the film from being corroded by the alkaline materials used in subsequent steps of the process.”

In addition, the team took careful steps to avoid the unwanted effect of ‘latching’: where a superconductor becomes permanently resistive after absorbing a photon. By fabricating nanostrips of just the right length, the researchers found that they could stabilize the current flowing through them during photon absorption, preventing them from latching.

With this approach, the team’s nanostrips could detect single photons in the visible light range with close to 100% quantum efficiency: an unprecedented performance for any type-I superconductor. By expanding further on their fabrication technique, they now hope their results could soon unlock exciting new capabilities in quantum computers.