COL 10(12), 121901(2012) CHINESE OPTICS LETTERS December 10, 2012
Exact investigation of the electronic structure and the linear
and nonlinear optical properties of conical quantum dots
M. Dezhkam and A. Zakery
∗
Department of Physics, College of Sciences, Shiraz University, Shiraz 71454, Iran
∗
Corresponding author: zakeri@susc.ac.ir
Received May 2, 2012; accepted June 20, 2012; posted online October 24, 2012
Intersubband linear and third-order nonlinear optical properties of conical quantum dots with infinite
barrier potential are studied. The electronic structure of conical quantum dots through effective mass
approximation is determined analytically. Linear, nonlinear, and total absorption coefficients, as well as
the refractive indices of GaAs conical d ots, are calculated. The effects of the size of the dots and of the
incident electromagnetic field are investigated. Results show that the total absorption coefficient and t he
refractive index of the dots largely depend on the size of the dots and on the intensity and polarization of
the incident electromagnetic field.
OCIS codes: 190.4720, 160.4760.
doi: 10.3788/COL201210.121901.
Quantum dots (QDs) are quasi-zero-dimension systems,
the carriers of which are confined in all the three spa-
tial dimensions. These quantum systems, which were
first studied by Esaki in 1970
[1]
, are described as “ar-
tificial atoms” because of their δ-function-like density of
states
[2]
.
Unlike bulk crystals with band energies, QDs have dis-
crete subbands because of their three-dimensional (3D)
confinements. Intersubband transitions result in physical
and optical properties and make QDs useful for infrare d
(IR) optoelectronic devices. The abs orption coefficient
(AC) and refractive index (RI) of the host material
change because of the considerably large dipole matrix
element and the low energy of intersubband transitions.
In particular, the enhancement of nonlinear optical prop-
erties is important in QDs. Nonlinear properties depend
on incident optical intensity; thus, a t high incident inten-
sities, the nonlinear properties should be co ns idered.
Researchers have recently investigated the electronic
structure and the linear and nonlinear optical prope rties
of different shapes of QDs, such as the box-shaped
[1]
,
parabolic cy linder
[3]
, lens-shaped
[4]
, spherical
[5]
, and
disc-like
[6]
QDs with finite, infinite, or Gaussian
confining potential.
Self-assembled InAs/GaAs QDs formed through
Stranski–Krastanow growth can be pyramid-shaped
[2]
.
These systems are important for laser applications
[7]
.
Thus, researchers have approximated the pyramids by
using cones to obtain the energy levels of conical quan-
tum dots (CQDs)
[8−10]
.
In this letter, a CQD with an infinite barrier potential
is considered. The Schr¨odinger equation in the e ffective
mass approximation is solved a nalytically to determine
the electronic structure of CQDs. Subsequently, the lin-
ear, third-order nonlinear, and tota l AC and RI changes
are inves tigated. Moreover, the dependencies of these op-
tical properties on QD size and on the intensity and po-
larization of the incident electromagnetic field are stud-
ied.
We consider that an electron confined in a CQD has
an infinite barrier potential, i.e., V (r) = 0 inside and
V (r) = ∞ outside the CQD. To solve this problem
analytically, we consider several assumptions, i.e., we
deem the potential outside of the QD to be infinite, al-
though in reality, the wavefunctions penetra te into the
host material
[10]
. In the effective mass approximation,
the Schr¨odinger equation for an electron inside the cone
is
−
~
2
2m
∗
∇
2
ψ(r) = Eψ(r), (1)
where ~ is the Planck’s constant divided by 2π, and, m
∗
is the electro n effective mass, E is the energ y eigenva lue,
and ψ(r) is the eigenfunction of the system. The geome-
try of the cone is shown in Fig. 1. R
′
, h, and α are the
basis radius, height, and apex angle of the cone, respec-
tively. The azimuthal symmetry facilitates the approxi-
mation of the cone by means of spherical coor dinates. To
obtain the approximation, we consider the origin of the
frame at the cone apex. The polar axis is the co ne axis.
R = (R
′2
+ h
2
)
1/2
represents the radius of the sphere, a
part of which is the cone. The area inside the CQD is
sp e cified by r < R, 0 6 ϕ 6 2π, and θ < α.
Equation (1) is rewritten by us ing spherical coordi-
nates, and is se parable due to the infinite barr ie r po-
tential. Thus, the eigenfunction ψ(r) can be written as
ψ(r, θ, ϕ) = Nf (r)g(θ)e
imϕ
, (2)
where N is the normalization consta nt. To create a
single-valued azimuthal part, m must be an integer if the
full az imuthal rang e is allowed (|m| = 0, 1, 2, · · · ).
Substituting Eq. (2) into Eq. (1) gives radia l and θ
equations. The solutions of the θ equation are associated
Legendre functions of the first and second kinds, i.e.,
P
m
ν
(cos θ) and Q
m
ν
(cos θ). As mentioned previously, we
derive cos α 6 cos θ 6 1. Q
m
ν
(cos θ) diverges at c os θ = 1
and is thus omitted
[11]
. The associated Legendr e func-
tion P
m
ν
(cos θ) is regular at cos θ = 1 and fo r |cos θ| < 1,
but singular at cos θ = −1 unless ν is an integer. In
this instance, cos θ = −1 is excluded; thus, ν is real
but nonintegral
[12]
. The spherical Bessel functions of
the first and second kinds are solutions to the radial
equation. The spher ical B e ssel functions of the second
kind, which are called “spherical Neumann functions,”
1671-7694/2012/121901(4) 121901-1
c
2012 Chinese Optics Lett ers