Molecular-Beam Epitaxy and Device Applications of III-V Semiconductor Nanowires

Publisher Website
Google Scholar
Abstract
A scaling-down of feature sizes into the nanometer range is a common trend in silicon and compound semiconductor advanced devices. That this trend will continue is clearly evidenced by the fact that the “roadmap” for the Si ultralarge-scale-integration circuit (USLI) industry targets production-level realization of a 70-nm minimum feature size for the year 2010. GaAs- and InP-based heterostructure devices such as high-electron-mobility transistors (HEMTs) and heterojunction bipolar transistors (HBTs) have made remarkable progress by miniaturization, realizing ultrahigh speeds approaching the THz range with ultralow power consumption. Due to progress in nanofabrication technology, feature sizes of scaled-down transistors are rapidly approaching the Fermi wavelength of electrons in semiconductors, even at the production level. This fact may raise some concerns about the operation of present-day devices based on semiclassical principles. However, the progress of nanofabrication technology has opened up the exciting possibility of constructing novel quantum devices, based directly on quantum mechanics, by utilizing artificial structures such as quantum wells, wires, and dots. In these structures, new physical effects appear, such as the formation of new quantum states in single and coupled quantum structures, artificial miniband formation in superlattices, tunneling and resonant tunneling in single and multiple barriers, propagation of phase-coherent guided electron waves in quantum wires, conductance oscillations in small tunnel junctions due to single-electron tunneling, and so on. We expect that these effects will offer rich functionality in next-generation semiconductor quantum ULSIs based on artificial quantum structures, with feature sizes in the range of one to a few tens of nanometers. Beyond this, molecular-level ULSIs using exotic materials and various chemical and electrochemical processes other than the standard semiconductor ones may appear, butat present, they still seem to be too far in the future for realistic consideration for industrial applications.