Optoelectronic Applications using Two-Dimensional (2D) Semiconductors
Restricted (Penn State Only)
- Author:
- Jayachandran, Darsith
- Graduate Program:
- Engineering Science and Mechanics
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- October 25, 2023
- Committee Members:
- Albert Segall, Program Head/Chair
Mark Horn, Major Field Member
Saptarshi Das, Chair & Dissertation Advisor
Larry Cheng, Major Field Member
Joan Redwing, Outside Unit & Field Member - Keywords:
- Two-dimensional materials
Optoelectronics
semiconductors
sensors - Abstract:
- The human species is currently witnessing a technological revolution at various forefronts including robotics and autonomous systems. In a typical autonomous system, sensors enable data collection from their environment. This data is then sent to logic and memory modules for data analysis and decision making. In traditional system architectures, there exists a gap between “sense”, “logic” and “memory” modules. The physical separation between these modules is a severe limitation for incorporating complex computational tasks associated with promising concepts like machine vision and artificial neural networks (ANNs). This is the motivation for developing devices capable of collocating “sensing”, “logic” and “memory” functionalities. For this, 1) the approach of performing computational tasks in memory is called in-memory computing and, 2) the approach of combining “sensing” and “computing” functionalities to develop sensors that can process sensory information for reducing redundant data collection is called in-sensor computing. For the last several decades, silicon has been the most efficient platform for developing processing chips, memory devices, and visible spectrum sensors. However, the implementation of concepts like in-memory computing and in-sensor computing necessitates the development of material platforms beyond silicon. Along these lines, two-dimensional (2D) materials like transition metal dichalcogenides (TMDs) have been extensively studied due to their huge promise towards continued logic scaling for increased computational power, beyond silicon. In addition, several recent demonstrations showcase the potential of 2D materials for in-memory and/or in-sensor computing. This dissertation focus on two such demonstrations using 2D materials, 1) a bio-inspired, low-power, nighttime collision detector that utilizes in-memory and in-sensor computing and, 2) an active pixel sensor (APS) with in-sensor image pre-processing capabilities. Accurately detecting a potential collision and triggering a timely escape is critical in the field of robotics and autonomous vehicle safety. In the case of cars, detecting a potential collision at night is a challenging task owing to the lack of discernible features that can be extracted from the available visual stimuli. To alert the driver or, alternatively, the maneuvering system of an autonomous vehicle, current technologies utilize resource draining and expensive solutions such as light detection and ranging (LiDAR) or image sensors coupled with extensive software running sophisticated algorithms. In contrast, insects perform the same task of collision detection with frugal neural resources. Even though the general architecture of separate sensing and processing modules is the same in insects and in image-sensor-based collision detectors, task-specific obstacle avoidance algorithms allow insects to reap substantial benefits in terms of size and energy. For example, the lobula giant movement detector (LGMD) neuron in locusts can perform non-linear mathematical operations on visual stimuli to elicit an escape response with minimal energy expenditure. Collision avoidance models based on image processing algorithms have been implemented using analogue very-large-scale integration (VLSI) designs, but none are as efficient as this neuron in terms of energy consumption or size. Here, insect-inspired collision detection algorithms are implemented in conjunction with in-sensor processing. This is enabled by innovative optoelectronic integrated circuits based on atomically-thin and photosensitive memtransistor technology and can greatly simplify collision detection at night. The proposed collision detector eliminates the need for image capture and image processing, yet demonstrates timely escape responses for cars on collision courses under various real-life scenarios at night. In-sensor processing, which can reduce energy and hardware burden for many machine vision applications, is currently lacking in the state-of-the-art APS technology. Photosensitive and semiconducting 2D materials can bridge such a technology gap by integrating image capture (sense) and image processing (compute) capabilities in a single device. Here, a 2D APS technology based on a 900-pixel monolayer MoS2 phototransistor array is introduced, where each pixel uses a single programmable phototransistor (1T cell), leading to a significant reduction in footprint and energy consumption. Further, a near-ideal yield and uniformity in photoresponse across the 2D APS array is demonstrated. The proposed low-power 2D APS technology can be transformative for many computer-vision applications.