NSF/ENG/ECCS-BSF: Semiconductor Polytype Heterostructures: A Pathway to Superior Power Electronics

Regents Of The University Of Michigan - Ann Arbor, Ann Arbor MI

Investigators

Abstract

Non-technical Description: This project aims to develop a new building block for electronics, semiconductor polytype heterostructures, which consist of adjacent layers of lattice-matched materials differing only in their atomic stacking sequences. Semiconductor polytype heterostructures are expected to result in the formation of a polarization-doped two-dimensional electron gas, with both high carrier concentration and high carrier mobility, resulting in ultra-high conductivity; thus, they are expected to offer a timely solution to a near-decade plateau in transistor speed. The project provides training to graduate, undergraduate, and high school students, engaging them in a collaboration between the University of Michigan and Ben-Gurion University. The collaboration integrates the expertise of the U.S. investigators (molecular-beam epitaxy and crystallographic characterization of semiconductor polytype films and heterostructures) with that of the Israeli investigators (spectroscopic characterization of polytype heterostructures and fabrication/characterization of high-electron mobility transistors). The new knowledge gained will be broadly disseminated through publications and presentations, and graduate and undergraduate curriculum development. Outreach activities emphasize the mentoring of a broad range of students. Technical Description: The project seeks new understanding ZB vs. WZ polytype selection and the electronic states/transport properties of ZB/WZ polytype heterostructures, thereby informing strategies for fabrication of polytype heterostructures. The interplay between surface reconstruction, polytype selection, and local electronic states will be monitored in real-time during epitaxy using in-situ reflection high-energy electron diffraction (RHEED), multi-beam optical stress sensing, and scanning-tunneling microscopy. In addition to examining growth kinetics, the influence of electrostatic phenomena, including thermal and electron-induced charging, on WZ vs. ZB polytype selection in both ZB-preferring (GaAs) and WZ-preferring (GaN) materials will be explored. A machine-learning approach using convolutional neural networks will be used to quantify and classify RHEED patterns, thereby accelerating the process of identifying appropriate growth kinetics and induced surface charging to select WZ vs. ZB polytypes. Following epitaxy, the interface structure and polarity will be examined using high-resolution and scanning transmission electron microscopy, selected-area and convergent-beam electron diffraction, and x-ray diffraction. The electronic states will be examined using scanning tunneling spectroscopy and optical spectroscopic tools based upon the Franz-Keldysh effect. Upon identification of the key growth kinetics and/or electrostatic phenomena to tailor polytype selection, ZB/WZ polytype hetero-structures for HEMTs will be fabricated. Expected outcomes of this work include the identification of strategies for polytype selection during epitaxy of thin films that prefer the ZB or WZ polytype, as well as the design and fabrication of ZB/WZ polytype HEMT structures that will facilitate the discovery of new strategies for transistors, with the potential for integration of logic and memory beyond Moore's Law. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.