Development of an AI-based real-time monitoring scheme of micro-meso scale combustors using multi-location chemiluminescence spectra with limited available data
Abstract
The emergence of ferroelectricity in various materials owing to their non-centrosymmetric atomic arrangements and ability to control ferroelectric polarization presents many opportunities for technological advancements such as neuromorphic computing, sensors and detectors, energy harvesting and optically controlled FE-FETs and memristors. The recent discovery of 2D ferroelectric materials introduces a new avenue in this domain. Combining 2D ferroelectric materials with other 2D materials offers the potential to realize electrically and optically controlled neuromorphic devices and detectors. However, these advancements come with energy consumption concerns and limited operational speeds, typically in the nanosecond range. Addressing the growing demand for fast computing and data processing with minimal power consumption is crucial. By manipulating the arrangement of atoms in two single layers of 2D semiconductors, inversion symmetry is disrupted, leading to the emergence of ferroelectricity. The direction of ferroelectric polarization alignment depends on the sliding direction, facilitated by charge transfer between the layers. Electrical manipulation of this polarization offers the potential for precise control in neuromorphic computing applications. Integrating sliding ferroelectric materials with graphene and other 2D materials enables the observation of ferroelectric behavior through modulation of the Fermi level due to charge transfer phenomena. Optical control of ferroelectric polarization, characterized by its low power requirements, could significantly enhance its utility. By leveraging this understanding, we aim to demonstrate applications such as neuromorphic computing, reconfigurable logics, and high-speed photodetectors through heterostructure integration with graphene. vdW heterostructures with 2D ferroelectric integration can be designed for various functionalities such as FE-memristors, FE-FETs with gate tunable polarization switching and multi-level polarization-resistance states. These can be used to develop capabilities for in-memory computing (Fig. 1a) which can break the bottlenecks of the current Von Neuman architecture-based computing leading to fast, efficient and low power computation technology.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Feb 05, 2025
- Source ID
- FA23862414094
Entities
People
- Moon Soo Bak
Organizations
- Air Force Office of Scientific Research
- Sungkyunkwan University
- United States Air Force