Pixel sensors are one of the most ubiquitous and essential types of technologies in use today with applications ranging from optical communication systems to security surveillance, remote sensing, biomedical imaging, environmental monitoring, etc. The performance of any pixel sensor is benchmarked based on 1) Responsivity, which is defined as the sensitivity of the sensor to incident light 2) spectral uniformity, which is defined as the uniformity in the response of the sensor to different wavelengths of light 3) specific detectivity, which is defined as the signal-to-noise of the sensor normalized to the sensor area and 4) dynamic range, which is defined as the ratio of the maximum and minimum detection limit of the sensor.
Silicon (Si) complementary metal oxide semiconductor (CMOS) based active pixel sensor (APS) technology is the backbone of most modern imaging systems with each pixel comprising a photodiode and a triad of transistors (3T cell). The photodiode integrates the photocurrent in response to optical stimuli into charge and the triad of transistors converts the accumulated charge into a voltage for readout, resets the photodiode, and relays the voltage through a communication bus to other parts of the chip for further processing and/or storage.
As the scale and diversity of applications for pixel sensors have grown over time, information conversion and transmission of data through conventional architecture in complementary metal-oxide-semiconductor (CMOS) sensors have become extremely area and energy inefficient. Steady progress has been made toward the development of high-performance and energy-efficient pixel sensors through material discovery and innovation in device engineering and the development of in-sensor or near-sensor computing architectures.
Due to their atomically thin nature, tunable bandgap, and strong light-matter interaction when confined to their monolayer limit, two-dimensional (2D) transition metal dichalcogenides (TMDCs) have invoked tremendous excitement in the optoelectronics community. Although significant advancements have been made toward the development of high-performance 2D photodetectors, reducing the hardware burden, high yield, large-area growth, device-to-device variation, and developing energy-efficient pixel sensors for in-sensor processing remain a serious concern.
Acknowledging the aforementioned limitations, we have developed a programmable 2D active pixel sensor (2D APS) consisting of 900 phototransistors based on atomically thin molybdenum di sulfide. The 2D APS uses only one programmable phototransistor (1T cell), which significantly reduces the area overhead allowing one to fit 900 pixels in ~0.09 cm2. Commonly encountered problems in the field of 2D material-based vision sensors are also resolved by showing high yield and low device-to-device variation in photoresponse owing to high-quality growth and clean transfer of large area 2D material. To validate these claims, we have performed extensive material characterization techniques that include scanning transmission electron microscopy (STEM), atomic force microscopy (AFM), Raman and Photoluminescence (PL) maps, X-ray photoelectron spectroscopy (XPS), polarization-resolved second harmonic generation (SHG) of both as grown and transferred films for structure-property correlation.
We exploit the phenomenon of gate-tunable persistent photoconductivity in MoS2 phototransistors to achieve spectral uniformity for blue, green, and red wavelengths of light. Furthermore, we also exploit the photogating effect in the MoS2 phototransistors to exhibit high responsivity (~3.6×107 A/W), high specific detectivity (~5.6×1013 Jones), and high dynamic range (~80 dB). The amount of energy consumed in sensing this information was observed few tens of picojoules.
In addition to the optical sensing, the programmable gate stack in our 2D APS technology allows us to reset the visual information immediately and enable us to perform in-sensor computing operations. For example, by combining gate-induced optical sensing and electrical erase through a programable gate stack we have demonstrated de-noising capability in our 2D APS platform. We believe that our low-power 2D APS technology with in-sensor image processing capabilities can be transformative for many edge applications.
To know more about the work, please refer to the paper “Active Pixel Sensor Matrix based on Monolayer MoS2 Phototransistor Array” published in Nature Materials following the link https://www.nature.com/articles/s41563-022-01398-9.
Please sign in or register for FREE
If you are a registered user on Springer Nature Materials Community, please sign in