× near
Artistic rendering of a butterfly above the bio-inspired image sensor. Credit: The Grainger College of Engineering at the University of Illinois Urbana-Champaign
There are many creatures on our planet with more developed senses than humans. Turtles can sense the Earth’s magnetic field. Mantis shrimps can detect polarized light. Elephants can hear much lower frequencies than humans. Butterflies can perceive a wider range of colors, including ultraviolet (UV) light.
Inspired by the improved visual system of the Papilio xuthus butterfly, a team of researchers has developed an image sensor capable of “seeing” in the UV range inaccessible to the human eye. The sensor design uses stacked photodiodes and perovskite nanocrystals (PNCs) capable of imaging different wavelengths in the UV range. Using the spectral signatures of biomedical markers such as amino acids, this new imaging technology is even able to distinguish cancer cells from normal cells with 99% confidence.
This new research, led by University of Illinois at Urbana-Champaign electrical and computer engineering professor Viktor Gruev and bioengineering professor Shuming Nie, was recently published in the journal Scientific progress.
Small variations
“We were inspired by the visual system of butterflies, which are able to perceive multiple areas in the UV spectrum, and designed a camera that reproduces this functionality,” says Gruev. “We did this by using new perovskite nanocrystals combined with silicon imaging technology, and this new camera technology can detect multiple UV regions.”
UV light is electromagnetic radiation with wavelengths shorter than those of visible light (but longer than X-rays). We are most familiar with UV radiation from the sun and the dangers it poses to human health. UV light is categorized into three different regions—UVA, UVB, and UVC—based on different wavelength ranges. Since humans cannot see UV light, it is challenging to capture UV information, especially recognizing the small differences between each region.
However, butterflies can see these small variations in the UV spectrum, just as humans can see shades of blue and green. Gruev notes, “It’s intriguing to me how they manage to see these small variations. UV light is incredibly difficult to capture, it’s just absorbed by everything, and butterflies do it extremely well.”
The imitation game
Humans have tricolor vision with three photoreceptors, where each perceived color can be made from a combination of red, green, and blue. Butterflies, however, have complex eyes with six (or more) classes of photoreceptors with different spectral sensitivities. In particular, Papilio xuthus, a yellow Asian swallowtail butterfly, has not only blue, green, and red receptors, but also violet, ultraviolet, and broadband receptors. In addition, butterflies have fluorescent pigments that allow them to convert ultraviolet light into visible light, which can then be easily sensed by their photoreceptors. This allows them to perceive a wider range of colors and details in their environment.
In addition to the increased number of photoreceptors, butterflies also display a unique multi-layered structure in their photoreceptors. To replicate the UV sensing mechanism of the Papilio xuthus butterfly, the UIUC team emulated the process by combining a thin layer of PNC with a multilayer array of silicon photodiodes.
PNCs are a class of semiconductor nanocrystals that exhibit unique properties similar to those of quantum dots—changing the size and composition of the particle changes the absorption and emission properties of the material. In the past few years, PNCs have emerged as an interesting material for various sensing applications, such as solar cells and LEDs. PNCs are extremely good at detecting UV (and even lower) wavelengths that traditional silicon detectors are not. In the new image sensor, the PNC layer is able to absorb UV photons and re-emit light in the visible (green) spectrum, which is then detected by the multilayer silicon photodiodes. Processing these signals allows UV signatures to be mapped and identified.
Healthcare and more
Cancer tissues have various biomedical markers in higher concentrations than healthy tissues – amino acids (building blocks of proteins), proteins and enzymes. When excited with UV light, these markers glow and fluoresce in the UV and part of the visible spectrum, in a process called autofluorescence. “Imaging in the UV region is limited, and I would argue that this is the biggest obstacle to making scientific progress,” Ni explains. “Now we’ve come up with this technology where we can image ultraviolet light with high sensitivity and we can also distinguish small differences in wavelength.”
Since cancer and healthy cells have different concentrations of markers and therefore different spectral signatures, the two classes of cells can be differentiated based on their fluorescence in the UV spectrum. The team evaluated their imaging device for its ability to distinguish cancer-related markers and found that it was able to distinguish between cancer and healthy cells with 99% confidence.
Gruev, Nie and their collaborative research team envision being able to use this sensor during surgery. One of the biggest challenges is knowing how much tissue to remove to ensure clear margins, and such a sensor could help facilitate the decision-making process when a surgeon removes a cancerous tumor.
“This new imaging technology allows us to distinguish cancerous from healthy cells and opens up new and exciting applications beyond just health,” says Nie. There are many species other than butterflies that can see in the ultraviolet, and having a way to detect this light will provide interesting opportunities for biologists to learn more about these species, such as their hunting and mating habits. Placing the sensor under water can also help to better understand this environment. While many UV rays are absorbed by water, there is still enough that gets through to have an effect, and there are many animals underwater that also see and use UV light.
More info:
Cheng Chen et al, Bioinspired, vertically aligned and perovskite nanocrystal-enhanced CMOS image sensors for resolving UV spectral signatures, Scientific progress (2023). DOI: 10.1126/sciadv.adk3860. www.science.org/doi/10.1126/sciadv.adk3860
Log information:
Scientific progress