Single photon camera with 400 times more pixels!

Now, a team at the National Institute of Standards and Technology (NIST) in Boulder, Colorado, has created a 400,000-pixel single-photon "camera" - 400 times larger than the largest camera of its kind before.

One of the most significant advantages of single-photon cameras is their superior sensitivity. By detecting individual photons, they can capture extremely weak signals and operate in near darkness. This sensitivity is particularly useful in applications such as astrophotography, night vision and microscopy, where capturing fine detail in low-light environments is critical.

At this point, single-photon cameras can reveal hidden information and provide high-quality images even under challenging lighting conditions.

In addition to sensitivity, single-photon cameras offer high temporal resolution. By precisely measuring the arrival time of each photon, these cameras are able to capture events quickly and with remarkable accuracy. This capability is valuable in applications such as fluorescence lifetime imaging and time-dependent single photon counting: these applications require precise time measurements.

This imaging technology is not new; in fact, it has been around for decades, yet the resolution of these devices has always been low: at around 1000 pixels, it is a far cry from the tens of megapixels found in today's commercial equipment with conventional digital cameras.

The main reason for this limitation is that single-photon cameras must be supercooled to work, and putting all the wiring required for a high-resolution camera into a cryostat has so far proven to be unfeasible.

The scientists report the results in the arXiv preprint paper A superconducting-nanowire single-photon camera with 400,000 pixels on June 15.

The first ever large-scale chip to detect single photons with superconducting nanowires.

As a result, this achievement has been highly evaluated by academia and industry.

For example, Stefan Carp, associate professor of radiology at Harvard Medical School, said, "From a scientific point of view, this certainly opens up a new avenue for optical imaging of the brain. Other methods for optically mapping cortical flow may be less costly, but they all have drawbacks that affect signal quality and often require complex signal processing. From a performance perspective, there are no compromises with this nanowire."

The superconducting nanowire detector captures virtually every photon, operates at visible, ultraviolet and infrared frequencies, and outputs results in just a few picoseconds for high frame rate detection.

The detector's sensitivity stems from the fact that a large enough current running through the superconductor will destroy its superconducting properties. Each pixel of the camera is a superconducting wire with a current setting just below the threshold, so that a single photon colliding with the wire will destroy its superconductivity. The result of the breakage is an increase in resistance on the wire, which is almost immediately detectable.

The performance of a single pixel is amazing, but putting many pixels on a single chip, in close proximity to each other, has been a long-term challenge.

To achieve superconductivity, the device must be cooled to low temperatures, and connecting many pixels to the cooling system is difficult. "I certainly can't put a million lines into my cryostat." Adam McCaughan, a physicist and NIST staff member who led the work, said, "It would be an incredible amount of engineering, not to mention getting it read out."

To overcome these difficulties, the team drew inspiration from other detector technologies. They borrowed the idea of a common readout bus that collects detector information from an entire row or column of pixels at a time; however, the direct application of the bus introduced crosstalk between pixels, destroying the sensitivity of the device.

"The problem with the usual way readout buses are made is that they are symmetric: anything that can go out can come in. So we wanted to start thinking about how can we connect the detector to the bus in an asymmetrical way?"

The key was figuring out the asymmetric solution - that is, the signal from the detector would be transferred to the bus, rather than the other way around.

To do this, the team designed an intermediate step next to each detection pixel, in which they connected a heating element in parallel with a superconducting nanowire. Photons hitting the nanowire will break the superconductivity and deflect the current to the heating element. The heating element will then naturally warm up and in turn break superconductivity in the bus, which is also made of superconducting wire. This will not interfere with the adjacent heating element, thus creating the desired asymmetric coupling.

(a) Imaging at 370 nm with an 800 × 500 array. Raw time-lapse data are shown as individual dots in red, and binned 2D histogram data are shown in black and white. (b) Count rate as a function of bias current for various wavelengths of light, and dark count rate. (c) False-color scanning electron micrograph of the lower right corner of the array, highlighting the interlaced row and column detectors. (Inset, lower left) Schematic showing detector-bus connections. (Inset, bottom right) Close-up showing the 1.1 μm detector width and effective 5 × 5 μm pixel size.

This design proved to be very productive.

The dramatic size improvement has opened up many applications, especially in biomedical imaging. For example, scientists such as Roarke Horstmeyer, assistant professor of biomedical imaging at Karp and Duke, are developing techniques to image the brain by shining light into it and detecting the tiny amount of light scattered back.

A more ambitious vision to follow, Horstmeyer believes, is to make a portable MRI.

For shining light into human tissue, near-infrared frequencies are ideal. They penetrate deeper into the tissue, are less destructive and allow higher intensities; commercially available silicon-based detectors do not perform well at these frequencies. As a result, Horstmeyer says, "This nanowire technology is really well suited to light for bio-optical devices. Having such a large device opens up the possibility of imaging the entire brain in real time, for example."

The NIST team is now working closely with several bioimaging groups to adapt the device to their specific needs, such as improved timing sensitivity. The researchers believe these improvements are achievable. "In terms of applications for this technology, the potential is, to some extent, endless: applications will be in areas ranging from biomedical imaging to particle and quantum physics research."

Reference links:

[1] https://spectrum.ieee.org/single-photon-camera

[2]https://arxiv.org/abs/2306.09473

[3]https://www.hackster.io/news/a-quantum-leap-in-imaging-414050903a5f