University PhD Dissertation Defense
Energy Conversion and Transport in Silicon Nanostructures
Jeremy A. Rowlette
Research Advisor: Professor Ken Goodson
Thursday August 20, 2009 @ 1:00 p.m. (Refreshments at 12:45 p.m.)
Paul Allen Building (formerly CIS-X), Auditorium
Abstract
We examine fundamental energy conversion and transport processes in silicon nanostructures which are relevant to the operation of emerging silicon-based nanoelectronic and nanophotonic devices. The theoretical and experimental work presented here will facilitate the design of improved nanodevices ranging from compact transistors, to quantum dot-based optical sources, detectors, and modulators.
In the first half of the talk, we discuss electron-phonon and phonon-phonon energy conversion within nanoscale silicon transistors under conditions of strong departure from thermal equilibrium. As "hot" electron relaxation tends to favor optical phonon (OP) emission, the conversion from OPs to the acoustic phonons (APs) responsible for heat conduction can yield an energy conversion bottleneck in the drain leading to reduced drive currents and even negative differential conductance effects [1]. We determine the conditions necessary for reaching this critical state in silicon-based nanodevices by developing self-consistent Monte Carlo device simulations which fully couple the electron and phonon systems while accurately accounting for electron and phonon energy dispersion [2]. To assist these calculations, we use anharmonic perturbation theory to compute the two-phonon final-states spectrum and lifetime for selected longitudinal OPs, which have high occupancies during transistor switching.
In the second half of the talk, we discuss photon-electron and electron-electron energy conversion within dense (~10^18 cm^-3) systems of small (~5 nm) luminescent silicon nanocrystals (NCs) embedded in amorphous dielectric films, a family of materials which includes annealed silicon rich oxides and nitrides (SRO, SRN), as well as porous silicon (PS). These materials exhibit stable, room-temperature visible and near-infrared luminescence as well as the ability to sensitize codoped Er3+ ions and are therefore promising for the development of inexpensive, CMOS-compatible short-range optical sources. Despite their promise, it has been shown that fast (< ns) nonradiative carrier recombination (NRCR), coupled with long (> µs) photoluminescence lifetimes, severely limit the prospects for achieving practical levels of optical gain. We characterize these NRCR and associated energy conversion processes by measuring the excited carrier dynamics of optically pumped NCs using a custom-built, two-color picosecond pump-probe measurement system. The unique dependence of the excited carrier losses vs. pump-probe delay and vs. pump intensity reveals enhanced NRCR at high NC occupancies, which we determine to be caused by long–range Coulombic dipole-dipole (d-d) interaction and energy transfer between excited NCs [3]. Finally, we derive the scaling of the effective d-d interaction strength of the NC ensemble in low-dimensional systems and present recent experimental results on quasi-2D SRO films which further supports the interacting d-d model. Monte Carlo analysis is used to provide additional insight into the spatially distributed d-d energy conversion in selected low-dimensional systems.
[1] E. Pop, D. Mann, J. Cao, Q. Wang, K. Goodson, H. Dai. Phys. Rev. Lett. 95 155505 (2005)
[2] J. Rowlette and K. Goodson. IEEE Trans. Elect. Dev. 55 220 (2008)
[3] J. Rowlette, R. Kekatpure, M. Panzer, M. Brongersma, K. Goodson. Phys. Rev. B 80 045314 (2009)
No comments:
Post a Comment