Wide-field mid- to long-wave infrared imaging with undetected photons
This paper demonstrates wide-field mid- to long-wave infrared imaging (6–10 µm) using quantum imaging with undetected photons via non-collinear phase-matching in an AgGaS₂ crystal, achieving background-noise-free, room-temperature operation with resolution and acquisition times significantly superior to conventional detectors.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Seeing the world in infrared light reveals a hidden layer of reality, one where the heat of a living cell or the chemical signature of a molecule becomes visible. This part of the spectrum is a powerful tool for scientists, offering a way to identify materials by their unique "fingerprints" without needing to label them with dyes or markers. However, capturing these images has long been a difficult task. Infrared light is dominated by the thermal glow of the environment itself; at room temperature, everything radiates a background noise that drowns out faint signals. To see clearly, traditional cameras often require expensive, specialized sensors that must be cooled to cryogenic temperatures, and even then, the process is slow and limited by this overwhelming background heat.
A different approach has emerged, one that relies on a strange property of light known as quantum interference. Instead of trying to detect the infrared photons that bounce off an object, this method uses them only as a messenger. The technique involves splitting a beam of light into two linked pairs: one part travels in the visible spectrum, which our eyes and standard cameras can see, while the other travels in the infrared. These two beams are born together and remain connected in a way that makes them indistinguishable from one another. If the infrared beam interacts with an object, it changes the behavior of its visible partner, even though the visible beam never touched the object. By measuring the visible light, scientists can reconstruct an image of what the invisible infrared light encountered, effectively bypassing the need to detect the infrared photons directly. This means the image is immune to the thermal noise that usually plagues infrared cameras.
Researchers at Imperial College London have now taken this concept and expanded it significantly, demonstrating a way to create wide-field images in the mid- to long-wave infrared range, specifically between 6 and 10 micrometers. This is a crucial region for scientific analysis, covering the wavelengths where many important chemical bonds absorb light. In their experiment, the team used a crystal made of silver thiogallate to generate these linked pairs of light. They arranged the setup so that the infrared beam, which they call the "idler," passed through a sample, while the visible "signal" beam was directed toward a standard silicon camera. Because the infrared light was never detected by the camera, the system avoided the background noise that typically limits such measurements. The researchers found that this method allowed them to detect infrared light at levels at least one hundred times fainter than what is possible with the best standard infrared detectors operating at room temperature.
The team achieved a high level of detail in their images, resolving more than 8,000 distinct points across the field of view. At a wavelength of 8 micrometers, they could distinguish features as small as 297 micrometers. To test the system, they placed metal masks with specific patterns in the path of the infrared beam. The camera, looking only at the visible light, successfully reconstructed the shapes of these masks, revealing the holes and solid areas with clarity. The researchers showed that they could tune the system to image at different wavelengths within the 6 to 10 micrometer range simply by adjusting the angle of the crystal and the color of the laser used to pump it. This flexibility suggests the method could be adapted to target specific chemical signatures in various materials.
Speed was another major factor in their success. The team demonstrated two ways to process the data. The first method involved taking a series of images while slowly shifting the phase of the light, a process that took about ten seconds per frame and resulted in high-quality images. The second method used a single snapshot to reconstruct the image, cutting the acquisition time down to just ten seconds for the entire image, though with a slight trade-off in image quality. This speed is a dramatic improvement over traditional infrared spectroscopy methods, which can take hours to produce a similar image. The results indicate that it is possible to perform fast, clear, and spectrally selective imaging in the infrared without the need for cooled detectors or complex equipment. This advancement opens the door to new applications in biology and chemistry, where the ability to see molecular structures quickly and without background noise could transform how scientists analyze samples.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.