Ph.D. Oral Examination

Angie McConnell angela.mcconnell at stanford.edu
Tue Apr 26 13:57:00 PDT 2005


Stanford University Ph.D. Oral Examination

"Thermal Properties of Micro- and Nanoscale Materials"

Angela D. McConnell
Mechanical Engineering Department
Thermosciences Division

Thornton 110
Friday, April 29, 2005
8:00 AM - 9:00 AM
(Refreshments at 7:45 AM)


In the drive to improve the performance of integrated circuits and 
microelectromechanical systems (MEMS), device designers have continued to 
decrease component dimensions to micrometer and nanometer 
lengthscales.  However, as devices become smaller, they tend to experience 
increased thermal loads due to higher current densities generating heat in 
smaller volumes.  This results in elevated temperatures that are sufficient 
to significantly influence device performance and also adversely affect 
long-term reliability and stability.  In this way, heat transfer has become 
a critical design constraint that limits the performance of modern 
electronic devices.  Heat transport models with accurate thermal property 
data are required to enable designers of integrated circuits and MEMS to 
effectively manage heat transfer.  This thesis details research into the 
thermal properties of two materials at the micro- and nanoscale:  thin 
polycrystalline silicon films and carbon nanotubes.

Thin doped polycrystalline silicon films are common in MEMS and integrated 
circuit applications.  Previous studies show that film processing 
conditions strongly influence the thermal conductivity by altering the 
polysilicon microstructure, but existing thermal conductivity data for 
doped polysilicon films in the literature is difficult to quantitatively 
compare because relevant microstructural details such as impurity 
concentrations and grain sizes and shapes are not reported.  The current 
study quantifies the relative effects of impurities and grain boundaries on 
the thermal conductivity of LPCVD polysilicon films of thickness near 1 
micron doped with boron and phosphorus at varying concentrations for a 
range of temperatures.  Thermal conductivity of the polysilicon films is 
measured using Joule heating and electrical resistance thermometry in a 
suspended membrane structure.  The data shows strongly reduced thermal 
conductivity values at all temperatures compared to similarly doped 
single-crystal silicon films, indicating that phonon scattering on grain 
boundaries is a dominant factor in the polysilicon thermal 
resistance.  Further analysis using a thermal conductivity model based on 
the Boltzmann transport equation reveals that phonon transmission through 
the grains is high at low temperatures, leading to large phonon mean free 
paths and higher than expected thermal conductivity.

Carbon nanotubes represent a promising area for nanotechnology research and 
development due to their high electrical conductivity, tensile strength, 
and thermal conductivity.  However, it is difficult to extract accurate 
properties of an individual single-walled carbon nanotube (SWNT) from the 
existing experimental data since most of the measurements are for bulk 
specimens comprising multiple nanotubes.  In the current study, thermal 
conductance of an individual SWNT is measured using AC electrical 
resistance thermometry in a MEMS-based device, which comprises a nanotube 
bridging two heavily doped, free-standing polysilicon legs.  Individual 
nanotubes are grown directly on the device using chemical vapor deposition 
with iron particles as the growth catalyst.  Electrical measurements from 
the setup are processed using a steady-periodic model of heat generation 
and heat transfer in the structure to compute temperatures, heat fluxes and 
thermal properties for the system.  The measurement approach is validated 
using thin silicon nitride filaments of known thermal conductivity.  The 
resulting data for an individual SWNT represents the intrinsic conductance 
of an isolated nanotube.  Comparing this with data for SWNT bundles from 
the literature indicates that intertube coupling can decrease the intrinsic 
conductance of the constituent nanotubes in a bundle by more than two 
orders of magnitude.




More information about the labmembers mailing list