Chemistry Reference
In-Depth Information
Fig. 10.14
(
a
) Cross-sectional TEM image, (
b
) high-resolution TEM image of GaSb nanodots,
(
c
) RHEED pattern, and (
d
) STM image of 65ML nanodots formed by codeposition at 450
◦
Cand
subsequently keeping at 450
◦
C for 30min in Sb flux
when compared with the nanodots that had not undergone this process (Fig.
10.12
d).
We changed the GaSb deposition amount from 1 to 200ML and some results are
shown as STM images in Fig.
10.15
a, b. As a result,
L
changed from 17 to
∼
100 nm
10
9
cm
−
2
) was unchanged. The PL spectra for these
nanodots of various sizes were measured, as shown in Fig.
10.15
c. The relation
between the peak position and
L
in the inset of Fig.
10.15
c shows the PL peak
blue-shifted with the decrease of the nanodot size.
In our experiment, the nanodots were sandwiched by a vacuum and the SiO
2
films forming the type I structure. The void size in the SiO
2
films beneath the
nanodots is so small that the SiO
2
films can work as a confinement barrier for the
carriers in the nanodots [
16
,
37
]. The hard wall square potential model is considered
to be reasonable because these two barriers are high enough (a few electron volts).
For an isotropic system, such as spherical nanodots with a radius
r
, therefore, PL
energy
E
confined
is easily described as
and the nanodot density (5
×
h
2
2
π
E
confined
(
)
=
E
bulk
+
L
(10.2)
r
2
2
μ
where
E
bulk
is PL energy for bulk material and
is the reduced mass of electron
and hole. Equation (
10.2
) can approximately be applied to domed nanodots with an
aspect ratio between 1 and 0.3 by replacing
r
with
L
μ
2[
37
]. In fact, the present
nanodots have various anisotropic shapes, which make the formulation of the quan-
tum confinement effect more complicated. The simple model in (
10.2
), however,
represents the essential quantum confinement effect and serves as a good reference.
We approximated the present nanodot shape to a dome with an aspect ratio of
/
5,
namely, a hemisphere and calculated the theoretical curve of
E
confined
in (
10.2
)
∼
0
.
