A novel transistor structure has been theoretically analyzed that has the potential to amplify signals at frequencies up to and beyond 10 Tera-Hertz (THz), which is more than an order of magnitude higher frequency than possible with existing silicon transistors. The goal of the proposed research project is to carry out experimental work to fabricate proof-of-concept prototype transistors and to model and characterize the performance of these transistors in order to verify the potential of the proposed novel transistor structure to achieve current gain greater than unity at extremely high frequencies. The proposed novel transistor has the potential to enable integrated circuit oscillators, amplifiers and mixers at frequencies up to and above 3 THz. They have the potential to enable THz spectrographic analysis of materials using small (hand-held) room-temperature power-efficient units for security (hidden weapon detection) and medical imaging (dental) applications. These ultra-fast transistors also offer the potential to enable the construction of digital circuits operating at clock frequencies that are an order of magnitude faster than those achieved by fully scaled CMOS technologies. And, these ultra-fast transistors could also enable an order of magnitude increase in the maximum operating frequencies of data converters, optical networking components, high-speed communications circuits and high-speed radar systems.
Theoretical calculations indicate that All-Metal-Terminal Hot-Electron Transistors (AMTHETs) have the potential to achieve a current gain of greater than 10 and to operate at frequencies up to or beyond 10 THz. The goal of the proposed research project is to verify and refine these theoretical calculations and to determine if further research work on the AMTHET is warranted. The proposed AMTHET is a metal-semiconductor-metal-semiconductor-metal sandwich in which the mean-free-path of the dominant carrier in the semiconductor layers is longer than the thickness of the semiconductor layers; and, the mean free path of electrons in the middle metal layer is longer than the thickness of the middle metal layer. This structure forms a bipolar junction transistor (BJT) in which the emitter, base and collector are metal layers separated from each other by EB and BC semiconductor layers. We propose to fabricate proof-of-concept prototype AMTHETs using two silicon-on-insulator (SOI) wafers bonded together to form the silicon-metal-silicon layer stack. This will be followed by removal of the bulk of the SOI wafers and deposition of metal to complete the AMTHET structure. Low-frequency and high-frequency measurements of the performance of these prototype devices as the materials and geometries are changed will be used to inform the modeling and analysis of AMTHETs and to guide the design of future iterations. A physics-based device model for the AMTHET, suitable for use in the design of future AMTHETs with particular performance targets, will be developed and fit to the experimental measurements. This work has the potential to advance understanding of bipolar transistors in which a majority of the charge carriers transit directly from the emitter to the collector without scattering.
