Researchers at Chung-Ang University in South Korea have developed an innovative new design for color image sensors. The new image sensor technology uses quantum dots that are tailored to specific light frequencies and could pack more pixels onto an image sensor, which may pay dividends for certain camera applications, such as for wearables, biomedicine equipment and self-driving vehicles.
Before diving into the new research, it’s worth pausing to discuss quantum dot technology in general. Quantum dots are tiny semiconductor particles. They’re only a few nanometers in size. When quantum dots are illuminated by ultraviolet light, an electron in an individual quantum dot is excited and enters a state of higher energy. This technology has recently been incorporated into displays. Quantum dots can produce pure red, green and blue light, which, compared to traditional display technology, results in a more vivid display with more accurate colors. Quantum dots can be tuned to specific wavelengths of light to ensure accurate color reproduction with no crosstalk or contamination. The primary takeaway is that quantum dots are tiny crystals that react to light and electricity and can glow in specific wavelengths depending upon the number of atoms in an individual dot.
|The number of atoms in a quantum dot determines the wavelength of light it emits when excited. Image credit: Samsung|
How does this relate to the new color image sensor technology developed by the researchers at Chung-Ang University? By using a quantum dot pixel structure constructed with vertically stacked quantum dots, the innovative pixel structure uses ‘much less area per pixel than conventional image sensors.’ Further, quantum dots have high photosensitivity, durability and efficiency.
Being able to fit more pixels onto an image sensor could be a significant development for certain specialized applications. CMOS image sensor technology has limitations within some contexts. The research team writes, ‘Many rising fields of application, such as self-driving cars, flexible electronics, and healthcare and medical imaging, demand even higher resolutions and levels of integration. This is difficult to achieve because of the way each pixel of a color image is captured. In most image sensors, the red, green, and blue components of a given pixel are captured independently using a dedicated photodetector “cell” for each color. While the three cells of each pixel are arranged laterally and as close to each other as possible to use the available area efficiently, this design takes at least thrice as much space as each individual cell. In addition, the manufacture and processing costs for these photodetector arrays can be high due to their complexity.’
If instead, each pixel on a sensor could capture color information for multiple colors at once, that would open up the door for much denser sensor design and improved resolution, ignoring potential improvements in color sensitivity. Many doors are opened if you no longer require a color filter on top of photo diodes.
The team of scientists, including Professor Sung Kyu Park of Chung-Ang University, wrote a paper titled ‘Vertically Stacked Full Color Quantum Dots Phototransistor Arrays for High-Resolution and Enhanced Color-Selective Imaging‘ that outlines the new technology. Quantum dots can be stacked vertically in each pixel, which is a key advantage over typical image sensors, as they rely upon lateral pixel arrangement. The higher quantum dots don’t block the quantum dots in a lower position within a stack. Rather, the photons that are not absorbed by specific quantum dots, each of which are tuned to specific wavelengths, penetrate through to the bottom.
Professor Park notes that ‘The device density of our photodetector array is 5500 devices per square centimeter, which is remarkably larger than that reported for previous solution-processed flexible photodetectors, which reaches up to 1600 devices.’ That’s a significant difference. Further, during testing, vertically stacked quantum dot pixels achieved better color selectivity and photosensitivity. In the long term, the team believes that quantum dot sensors could replace CMOS image sensors in ‘many applications’ due to the simple fabrication processor, low power consumption, durability and overall imaging capabilities.
‘We think our design is a great advancement towards establishing a low-cost, high-resolution and integrated image sensor system that goes beyond conventional ones,’ said Professor Park. ‘It should be widely applicable in fields such as wearable sensory systems, biomedicine, and autonomous driving.’