Environmental Engineering Reference
In-Depth Information
5.1 Introduction
Low-dimensional nanostructures have attracted considerable inter-
estinthescientificcommunityduetotheirdimension-tunableprop-
erties, as well as for their potential applications in nanotechnology.
Amongvariouslow-dimensionalstructures,one-dimensionalSiNWs
are especially attractive [1-3] because of their good compatibility
with conventional microelectronic technology and their promising
candidates as the components of nanoelectronic devices. Although
bulkSiisapoorthermoelectric(TE)material,arecentbreakthrough
madeinindependentexperimentsbyHochbaum
et al.
[4]andBukai
et al.
[5] revealed significantly improved ZT values in SiNWs, which
have inspired the research community to reconsider the Si as a TE
material.
The performance of TE materials is characterized by the figure
of merit, ZT
=
σ
T
/κ
,where
α
is the Seebeck coe
cient,
σ
is the electrical conductivity,
T
is the absolute temperature, and
κ
is the total thermal conductivity (
κ
=
κ
p
+
κ
e
,where
κ
p
and
κ
e
are the phonon and electronic contributions, respectively. In
semiconductors, the phonon's contribution is dominant at room
temperature.) The larger the value of ZT, the better the e
ciency
of the TE cooler or power generator. The thermal conductivity
of semiconductor nanostructures must be minimized to improve
the e
ciency of TE devices [6]. Therefore, searching for methods
that reduce the thermal conductivity of Si nanostructures has
become a crucial issue in current Si-based TE research. However,
the measurement of the thermal conductivity is very common for
bulk materials, but so less for nanostructures that it has been
rarely reported in nanostructured materials (e.g., SiNWs [7]). Thus,
theoretical modeling and simulation tools are of necessity [8, 9].
Particularly, theoretical model and numerical simulations not only
predict the thermal conductivity of nanostructures that have not
already been fabricated or that are di
cult to characterize in
experiment, but also provide information of thermal transport at
the nanoscale, which is helpful in designing nanomaterials with
tailored thermal conductivity. So far, the thermal conductivity of
SiNWs has been extensively studied theoretically by using different
methods, including molecular dynamics (MD) simulations [9-15],
α
2
