Theory and Design of Smith-Purcell Semiconductor Terahertz Sources

Abstract

A wealth of unique physics makes the frequency band from 0.3 – 3.0THz technologically relevant and compact room-temperature semiconductor sources would en- able many new spectroscopic applications. All semiconductor sources currently have serious difficulties reaching the THz, and no THz device of any type operates on the full range at room temperature. This dissertation proposes a novel semiconductor source which utilizes the transferred-electron (Gunn) effect and the Smith-Purcell effect to operate over the majority of the 0.3 – 3.0THz band at room temperature. Mathematical derivations provide a theoretical analysis of the device and computer simulations explore its performance as a function of design.

The dissertation begins with a description of the device and an overview of the field of terahertz science. A literature review establishes the relevance and uniqueness of the work and highlights the physical principles required to model the operation of the device. Next, the mathematical “machinery” required model the device is built, starting with a derivation of the Butcher-Fawcett “equal-areas” method used to calculate the Gunn effect. A description of the computer code written to implement the equal-areas method and examples validating its correct operation follow. A derivation of the Smith-Purcell effect provides a closed-form solution for the electric field, which is then combined with the Gunn effect results.

With the theoretical methods thus established, a detailed explanation of the simulation methods used to model the device is provided, followed by a detailed comparison between theory and simulation – in which it is shown that the theoretical methods display a high degree of agreement with computer simulations of the device. The results of an extensive design study are then presented that map the range of predicted results for this terahertz device. The Smith-Purcell device is found to perform better than or comparable to any state-of-the-art semiconductor THz source. At room temperature, a single 100µm-wide device is predicted to generate 365µW of power at 0.28THz and 1µW at 2.5THz. The dissertation concludes with suggestions for subsequent research – most urgently for an attempt to experimentally verify the predictions of this work.