Beating Pattern in Quantum Magnetotransport Coefficients of Spin-Orbit Coupled Dirac Fermions in Gated Silicene
Beating Pattern in Quantum Magnetotransport Coefficients of Spin-Orbit Coupled Dirac Fermions in Gated Silicene
Beating Pattern in Quantum Magnetotransport Coefficients of Spin-Orbit Coupled Dirac Fermions in Gated Silicene
r
X
i
v
:
1
4
0
4
.
6
1
8
8
v
1
[
c
o
n
d
-
m
a
t
.
m
e
s
-
h
a
l
l
]
2
4
A
p
r
2
0
1
4
Beating pattern in quantum magnetotransport coecients of spin-orbit coupled Dirac
fermions in gated silicene
SK Firoz Islam and Tarun Kanti Ghosh
Department of Physics, Indian Institute of Technology-Kanpur, Kanpur-208 016, India
(Dated: April 25, 2014)
We report theoretical study of magnetotransport coecients of spin-orbit coupled gated silicene
in presence and absence of spatial periodic modulation. The combined eect of spin-orbit coupling
and perpendicular electric eld manifests through formation of regular beating pattern in Weiss and
SdH oscillations. Analytical results, in addition to the numerical results, of the beating pattern
formation are provided. The analytical results yield a beating condition which will be useful to
determine the spin-orbit coupling constant by simply counting the number of oscillation between
any two successive nodes. Moreover, the numerical results of modulation eect on collisional and
Hall conductivities are presented.
PACS numbers: 72.10.-d,72.25.-b,73.43.Qt,73.50.-h
I. INTRODUCTION
Silicene, a new class of two-dimensional (2D) electron
system, possess graphene-like hexagonal lattice struc-
ture except the atoms are silicon instead of carbon
1,2
.
Several experiments of synthesizing monolayer silicene
have been performed successfully
38
. Relatively larger
atomic size of silicon atom causes the 2D lattice to
be buckled, in which two planes of sublattice A and
sublattice B are separated by d 0.46
A
9,10
. Theo-
retical investigations
9,11,12
show that silicene possess
an intrinsic spin-orbit coupling which is very strong
in comparison to graphene. This is due to the fact
that Si atoms have large intrinsic spin-orbit coupling
strength than C atoms. The application of electric
eld (E
z
) perpendicular to the buckled silicene sheet
generates a staggered sublattice potential dierence
(
z
= E
z
d) between the two atomic planes and opens
an electric eld dependent band gap between conduction
and valence bands
9,13,14
. The charge carriers in silicene
also obey Dirac-like Hamiltonian
6
around the corners of
its hexagonal Brillouin zone with additional properties
of having intrinsic spin-orbit coupling and electrically
tunable band gap
9
. Moreover, the low-energy dispersion
around K and K
so
and
z
can be determined experimentally by ana-
lyzing beating pattern in magnetotransport coecients
in spin-orbit coupled gated silicene in presence and ab-
sence of the spatial periodic modulation.
This paper is arranged in following order. In section II,
2
we present energy eigenvalues, the corresponding eigen-
states and density of states of the charge carriers in sil-
icene sheet subjected to transverse magnetic and electric
elds. In section III, we present analytical and numeri-
cal results of diusive, collisional and Hall conductivities
in presence and absence of the periodic modulation. In
section IV we summarize our results.
II. ENERGY EIGENVALUE AND
EIGENFUNCTION
We are considering a buckled 2D silicene sheet in which
Dirac electrons obey a nite gapped graphene-like Hamil-
tonian. The Hamiltonian of an electron with charge e
in presence of both eld, the perpendicular magnetic eld
B and electric eld E = E
z
z, is
12,15
H = v
F
(
x
y
) s
so
z
+
z
z
, (1)
where v
F
is the Fermi velocity, = p + eA is the 2D
momentum operator with vector potential A, = +()
denotes K(K
0
=
(s
so
z
). For n 1, energy spectrum for the
electron band is
E
=
_
n
2
+ (
so
s
z
)
2
, (2)
where {n, s, }, =
c
and
c
=
2v
F
/l is the
cyclotron frequency with l =
_
/eB is the magnetic
length scale. The Landau levels around K and K
points
split into two branches due to presence of both
so
and
z
. The splitting vanishes when either
so
or
z
is zero.
The normalized eigenstates are (for n 1)
=+
n,s
(r) =
e
ikyy
_
L
y
N
n,s
_
+
n,s
n1
(x + x
0
)
n
(x + x
0
)
_
(3)
and
=
n,s
(r) =
e
ikyy
_
L
y
N
n,s
_
n,s
n
(x + x
0
)
n1
(x + x
0
)
_
, (4)
where
n
(x) = (1/
_
2
n
n!l)e
x
2
/2l
2
H
n
(x/l) is the nor-
malized harmonic oscillator wave function centred at x =
x
0
with x
0
= k
y
l
2
. Here, the coecients are N
n,s
=|
n,s
|
2
+ |
n
|
2
with
n,s
=
,s
+
_
2
,s
+ n(
c
)
2
and
n
= i
c
n. Here,
,s
= (
so
s
z
).
The analytical form of density of states
38
is given by
D
,s
(E) D
0
(E)
_
1 + 2
k=1
exp
_
k
_
2
0
E
2
_
2
_
cos
_
s
_
E
2
2
,s
_
/
2
__
, (5)
where D
0
(E) = E/(2
2
v
2
F
) and
0
is the impurity in-
duced Landau level broadening.
III. ELECTRICAL MAGNETOTRANSPORT
In this section we shall study magnetotransport coef-
cients such as diusive, collisional and Hall conductivi-
ties. The diusive conductivity in absence of any modu-
lation exactly vanishes because of the zero group velocity
due to k
y
degeneracy in the energy spectrum. The pres-
ence of modulation imparts drift velocity to the charge
carriers along the free direction of its motion and gives
rise to diusive conductivity. The oscillations in the dif-
fusive conductivity known as Weiss oscillation which is
dominant at low magnetic eld regime. On the other
hand, the collisional conductivity arises due to scattering
of the charge carriers with localized charged impurities
present in the system. The quantum oscillation in colli-
sional conductivity is known as SdH oscillation. The Hall
conductivity due to the Lorentz force is independent of
any collisional mechanisms. However, the external spa-
tial periodic modulation induces periodic oscillation on
collisional as well as Hall conductivities. We shall use
formalism of calculating dierent magnetotransport co-
ecients developed in Ref.
39
.
A. Diusive conductivity
To study diusive conductivity, a spatial weak electri-
cal modulation V = V
0
cos(Kx) with K = 2/a is ap-
plied along the x direction of the silicene sheet. Here, a
is the modulation period. We can treat this modulation
as a weak perturbation as long as V
0
. The energy
correction due to the modulation is calculated approxi-
mately by using rst-order perturbation theory. Then to-
tal energy is given by E
= E
+E
, where {, k
y
}
and E
= G
(u) cos(Kx
0
) with
G
(u) =
V
0
e
u/2
N
n,s
_
|
n,s
|
2
L
n1
(u)+ |
n
|
2
L
n
(u)
_
.
(6)
Here, u = (Kl)
2
/2 and L
n
(u) is the Laguerre polynomial
of degree n. The energy correction E
transforms the
degenerate Landau levels into bands due to k
y
depen-
dency, which leads to non-zero drift velocity.
The diusive conductivity is calculated by using the
standard semiclassical expression as
dif
yy
=
e
2
(1 f
)(v
y
)
2
, (7)
where = L
x
L
y
is the area of the system, = (E
F
)
is the electron relaxation time at the Fermi energy E
F
which is calculated in the next paragraph, f
is the Fermi-
Dirac distribution function at E = E
and = (k
B
T)
1
with k
B
is the Boltzmann constant. Also, v
y
= | v
y
|
3
0
0.05
0.1
0.15
0.2
0.25
0.3
y
y
(
e
2
/
h
)
0.001 0.5 1.0 1.5
0
0.05
0.1
0.15
0.2
0.25
0.3
B
a
/B
y
y
(
e
2
/
h
)
T= 2K
T= 4K
FIG. 1. Plots of the exact results of the diusive conductivity
versus dimensionless inverse magnetic eld .
is the average value of the velocity operator v
y
and it
does not vanish due to the k
y
dependency of the energy
levels. It is given by
v
y
=
1
k
y
=
Kl
2
sin(Kx
0
)G
(u). (8)
After inserting drift velocity given by Eq. (8) into Eq. (7)
and integrating over k
y
variable, we get
dif
yy
= (e
2
/h)
with
=
u
f(E
n,s
)[1 f(E
n,s
)][G
(u)]
2
(9)
is the dimensionless total diusive conductivity. Note
that a given Landau level splits into two branches due
to the simultaneous presence of
so
and
z
. The
rst branch is E
n
=
_
n
2
+ (
so
z
)
2
for {s, } =
{+, +} and {, }. The second branch is E
+
n
=
_
n
2
+ (
so
+
z
)
2
for {s, } = {+, } and {, +}. So,
for = +1 (K-valley) there are two energy branches due
to spin splitting and same goes for = 1 (K
-valley)
also. The total conductivity is written as =
+
+
.
Here,
+
is the contribution from the second branch E
+
n
and
n
.
Before simplifying further, we shall derive the Fermi
energy. At Fermi level E
F
=
_
(v
F
k
F
)
2
+
2
with
= (
so
z
), then one can write
(k
+
F
)
2
(k
F
)
2
= 4
so
z
(v
F
)
2
. (10)
On the other hand, carrier density is given by
n
e
=
2
(2)
2
_
2
0
_
kF
0
kdkd =
1
2
_
(k
+
F
)
2
+(k
F
)
2
_
. (11)
Solving the above two equations for Fermi energy, we
get E
F
=
_
(E
0
F
)
2
+
2
so
+
2
z
with E
0
F
= v
F
k
0
F
and
k
0
F
=
n
e
.
Before presenting exact analytical results, we would
like to get approximated analytical results. To do so we
consider the system at very low temperature in which
Landau levels close to the Fermi energy contribute to
transport properties. Therefore, we can use some approx-
imations which are valid for higher values of n. Around
Fermi level, for higher value of n, we have
e
u/2
L
n
(u)
1
_
nu
cos
_
2
nu /4
_
. (12)
We can also use n n 1 for higher values of Landau
levels, which give G
(u) V
0
exp(u/2)L
n
(u). To ob-
tain analytical expression we convert the summation into
integration by using the relation
_
0
dn
2
2
_
0
EdE. (13)
By using the above two approximations, the exact ex-
pression of the dimensionless diusive conductivity given
by Eq. (9) reduces to
A
0
u
c
cos
2
(c
t + d
)
cosh
2
(t/2)
dt,
(14)
where A
0
= V
2
0
0
/(
2
), t = (E E
F
), and d
=
c
E
F
with
c
=
2
_
(E
0
F
)
2
2
so
z
E
F
. (15)
Using the standard integral
40
, Eq. (14) reduces to
= A
0
u
c
_
1 + H
_
T/T
a
_
cos
_
2f
__
. (16)
Here, = B
a
/B with B
a
= /ea
2
and
f
=
2a
v
F
_
(E
0
F
)
2
2
so
z
(17)
are two closely spaced frequencies of Weiss oscillations
of the two energy branches induced by the simultaneous
presence of spin-orbit interaction and gate induced elec-
tric eld. The temperature dependent damping factor is
given by
H
_
T/T
a
_
=
T/T
a
sinh(T/T
a
)
,
where the characteristic temperature (T
a
) is dened as
T
a
= v
F
E
F
B/
_
4
2
k
B
aB
a
_
(E
0
F
)
2
2
so
z
_
. (18)
Typically the dierence between T
+
a
and T
a
is very
small. T
a
= 0.76 K when B = 0.6 T. Another point
4
0.001 0.1 0.2 0.3 0.4 0.5 0.6 0.7
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
B
a
/B
y
y
(
e
2
/
h
)
FIG. 2. Plots of the exact and analytical results of diusive
conductivity versus dimensionless inverse magnetic eld. The
dark and red lines stand for analytic and numerical results.
is that T
+
a
is increasing with electric eld (
z
) or spin-
orbit interaction (
so
) while T
a
is decreasing.
As (E
0
F
)
2
2
so
z
, the total diusive conductivity
is given by
2A
0
u
c
_
1 + H
_
T/T
a
_
cos(2f
av
) cos(2f
d
)
(19)
with c = 2
uE
F0
/(E
F
), T
a
=
v
F
E
F
B/(4
2
k
B
aB
a
E
0
F
), f
av
= (f
+
+ f
)/2 and
f
d
= (f
+
f
z
must be satised to see the
beating pattern. The location of the beating node can
be obtained from the condition cos(2f
d
) |
B=B
j
= 0,
which gives
2f
d
B
a
B
j
=
_
j +
1
2
_
, (20)
where j = 0, 1, 2.... Another periodic term, cos(2f
av
),
gives number of oscillation between two successive beat-
ing nodes as
N
osc
= f
av
_
B
a
B
j+1
B
a
B
j
_
=
1
2
f
av
f
d
. (21)
In explicit form, it is given by
2N
osc
=
_
(E
0
F
)
2
+ 2
so
z
+
_
(E
0
F
)
2
2
so
z
_
(E
0
F
)
2
+ 2
so
_
(E
0
F
)
2
2
so
z
.
(22)
Here we make couple of important remarks on the above
equation: 1) number of oscillation between any two suc-
cessive beat nodes is independent of the modulation pe-
riod a and magnetic eld and 2) the expression of N
osc
can be used to estimate the spin-orbit coupling constant
by simply counting the number of oscillations between
any two successive beat nodes.
We use the following parameters for various plots: elec-
tron density n
e
= 4 10
15
m
2
, spin-orbit coupling con-
stant
so
= 4 meV, perpendicular electric eld induced
energy
z
= 12 meV, Fermi velocity v
F
= 2 10
5
ms
1
,
modulation period a = 150 nm, and modulation strength
V
0
= 0.1 meV. Figure 1 shows exact numerical results of
the diusive conductivity given by Eqs. (7) and (8) for
two dierent temperature. It shows beating pattern in
both the cases, and oscillation gets damped with increas-
ing temperature. In Fig. 2 we compare the approximate
analytical result given by Eq. (19) with the exact numer-
ical result obtained from Eq. (7). The analytical result
for diusive conductivity matches very well with the ex-
act result. The appearance of beating pattern is due to
the superposition of Weiss oscillation coming from two
oscillatory drift velocity with dierent frequencies (f
)
corresponding to two energy branches. For the numerical
parameters used here, the number of oscillations calcu-
lated from Eq. (22) is 14, same as counted from Fig.
1.
B. Collisional Conductivity
First we shall study collisional conductivity in absence
of modulation. The eect of modulation on SdH oscilla-
tion will be discussed in the later half of this section. The
standard expression for collisional conductivity is given
by
39
col
=
e
2
2
(1 f
)W
,
(
)
2
. (23)
Here, f
= f
.
Also,
= |r
q
|U(q)|
2
||e
iq(rR)
|
|
2
(E
), (24)
where q = q
x
x + q
y
y is the two-dimensional wave-
vector and U(q) = 2e
2
/(
_
q
2
x
+ q
2
y
+ k
2
s
) is the Fourier
transform of the screened impurity potential U(r) =
(e
2
/4)(e
ksr
/r), where k
s
is the inverse screening
length and is the dielectric constant of the material.
Equation (23) can be re-written for higher values of Lan-
dau level (n n 1) as given by
col
xx
=
e
2
h
N
I
U
2
0
l
2
(2n + 1)
_
f
E
_
E=E
. (25)
A closed-form analytical expression of the above equation
is obtained by replacing the summation over quantum
number as
n
2l
2
_
0
D(E)dE with D(E) is the
5
1 2 3 4 5 6
0
0.5
1
1.5
2
2.5
3
3.5
4
1/B (1/T)
x
x
c
o
ll
(
e
2
/
h
)
n
e
= 8 10
15
/m
2
0
= 0.01 meV
T=1.5 K
FIG. 3. Plots of the exact and analytical results of the colli-
sional conductivity versus inverse magnetic eld in absence of
modulation. Here, red and dark lines stand for numeric and
analytic results, respectively.
density of states (see Eq. (5)) and it is given by
col()
xx
0
(
c
0
)
2
E
2
F
2
_
1 + 2
D
H
_
T/T
c
_
cos
_
2
/B
__
. (26)
Here,
0
= e
2
0
E
F
/
2
is Drude-like conductivity and
= [(E
0
F
)
2
2
z
so
]/(2ev
2
F
) are the SdH oscilla-
tion frequencies for the two energy branches E
+
n
and E
n
.
Also, impurity induced damping factor is
D
= exp
_
_
2
0
E
F
2
_
2
_
(27)
and the temperature dependent damping factor is
H(x) = x/ sinh(x). Here, x = T/T
c
with T
c
=
2
/(2
2
k
B
E
F
) is the critical temperature.
Following the same procedure as in the diusive con-
ductivity case, here we get the location of a beating
node (B
j
) as B
j
(j + 1/2) = 2
so
z
/(ev
2
F
) and num-
ber of oscillations between any two successive nodes as
N
s
osc
= (E
0
F
)
2
/(4
so
z
). Unlike Rashba spin-orbit cou-
pled two-dimensional electron gas, N
s
osc
in silicene is same
for a set of given parameters and does not depend on spe-
cic choices of the successive nodes. In Fig. 3, we show
the beating pattern in the collisional conductivity vs in-
verse magnetic eld. For the parameters used here, we
have N
s
osc
= 14 which is same as shown in Fig. 3.
Now we will describe eect of of weak modulation on
the collisional conductivity. The modulation eect enters
mainly through the total energy in Fermi distribution
function. The collisional conductivity in presence of weak
modulation is given by
col
xx
e
2
h
N
I
U
2
0
a
0
(2n + 1)M
, (28)
where M
is given by
M
=
_
a/l
2
0
_
f
E
_
E=E
dk
y
. (29)
0.025 0.035 0.045 0.055
3
2
1
0
1
2
3
x 10
3
B (T)
x
x
c
o
ll
(
e
2
/
h
)
n
e
= 4 10
15
/m
2
T=2 K
FIG. 4. Plots of change in the collisional conductivity due to
modulation versus magnetic eld.
It is dicult to get a closed-form analytical expression
of the collisional conductivity in presence of the modu-
lation. The change in collisional conductivity due to the
modulation [
coll
xx
=
coll
xx
(V
0
)
coll
xx
(0)] is calculated
from Eq. (28) numerically and shown in Fig. 4. Figures
(3) and (4) clearly shows that eect of the modulation on
coll
xx
is very small and vanishes with increasing B. The
location of the beating node and number of oscillations
between any two successive nodes are determined by Eqs.
(20) and (21), respectively. To understand why the beat-
ing pattern appears in
coll
xx
follows the same condition
as in the diusive conductivity, we expand M
as given
by
M
=
_
a/l
2
0
__
f
E
_
+ E
,ky
_
E
_
+
(E
,ky
)
2
2!
_
E
_
+ .....
_
E=E
,ky
dk
y
=
a
l
2
__
f
E
_
+
{G
(u)}
2
4
_
E
_
...
_
E=E
. (30)
We can see the modulation dependent dominant term
is of the order of (G
(u))
2
V
2
0
, which is same as in
the diusive conductivity (see Eq. 9). The modulation
induced Weiss oscillation in collisional conductivity, as
shown in Fig. 4., also follows the same beating condition
as in the diusive conductivity.
From the numerical result we can see that modulation
eect is dominant at low range of magnetic eld i.e; when
the energy scale of Landau level is not much higher than
the energy correction due to modulation. As magnetic
eld increases, SdH oscillation starts to dominate over
Weiss oscillation.
6
0.025 0.050 0.075 0.100
0
0.2
0.4
0.6
0.8
1
B (T)
x
y
(
e
2
/
h
)
n
e
= 4 10
15
/m
2
T= 2 K
FIG. 5. Plots of the modulation eect on Hall conductivity
versus magnetic eld.
C. The Hall conductivity
In this sub section, we will see modulation eect on the
Hall conductivity. The Hall conductivity is given by
39
yx
=
ie
2
(E
)
2
| v
y
|
| v
x
|. (31)
Using unperturbed eigenstates velocity matrix ele-
ments are given by
< n, s, + | v
x
| n + 1, s, + >=
+
n+1,s
n
N
+
n+1,s
N
+
n,s
v
F
(32)
and
< n + 1, s, + | v
y
| n, s, + >= i
+
n+1,s
n
N
+
n+1,s
N
+
n,s
v
F
. (33)
Substituting Eqs. (32) and (33) into Eq. (31), we get
yx
= 2
e
2
h
l
2
a
n,s
_
a/l
2
0
|
+
n+1,s
|
2
|
n
|
2
| N
+
n+1,s
|| N
+
n,s
|
f(E
+
n,s,ky
) f(E
+
n+1,s,ky
)
[
n + z
n + 1 + z
n,ky
]
2
dk
y
. (34)
Here z = [(
so
s
z
)/]
2
and
n,ky
= E
n+1,s,ky
E
n,s,ky
V
0
u
n
e
u/2
L
1
n1
cos (Kx
0
). (35)
Here, a factor of 2 is multiplied because of the iden-
tical states between two valleys but with opposite spin.
The modulation induced change in the Hall conductiv-
ity [
yx
=
yx
(V
0
)
yx
(0)] is plotted in Fig. 5. It
shows beating pattern in the Weiss oscillation of the Hall
conductivity. The node position B
j
and number of oscil-
lations between any two successive nodes are determined
by Eqs. (20) and (21), respectively. The increase of
magnetic eld diminishes the modulation eect on Hall
conductivity as expected.
IV. SUMMARY
We have shown the appearance of beating pattern in
quantum oscillations of magnetotransport coecients of
spin-orbit coupled gated silicene with and without spa-
tial periodic modulation. There is a spin-splitting of the
Landau energy levels due to the presence of both the
spin-orbit coupling and electric eld perpendicular to the
silicene sheet. The formation of beating pattern is due to
the superposition of oscillations from two dierent energy
branches but with slightly dierent frequencies depend-
ing on the strength of spin-orbit coupling constant and
perpendicular electric eld. In addition to the numerical
results we also provide analytical results of the beating
pattern in Weiss and SdH oscillations. The approximated
analytical results are in excellent agreement with the ex-
act numerical results. The analytical results yields a sim-
ple equation which can be used to determine the strength
of spin-orbit coupling constant by simply counting the
number of oscillations between any two successive beat
nodes. There have been few theoretical calculation, al-
ready mentioned earlier, which estimated the spin-orbit
coupling constant. Here we have proposed a way to de-
termine the spin-orbit coupling constant experimentally.
Finally for the sake of completeness, modulation eect on
collisional and Hall conductivities are also studied numer-
ically. Moreover, the analytical results of the Weiss and
SdH oscillations frequencies reduce to graphene mono-
layer case
36,37
by setting
so
= 0 or
z
= 0.
V. ACKNOWLEDGEMENT
This work is nancially supported by the CSIR,
Govt. of India under the grant CSIR-JRF-09/092(0687)
2009/EMR F-O746.
1
K. Takeda and K. Shiraishi, Phys. Rev. B 50, 14916 (1994)
2
G. G. Guzman-Verri and L. C. L. Yan, Phys. Rev. B 76,
075131 (2007)
3
B. Lalmi, H. Oughaddou, H. Enriquez, A. Kara, S. Vizzini,
B. Ealet, and B. Aufray, Appl. Phys. Lett. 97, 223109
(2010)
7
4
P. De Padova, C. Quaresima, C. Ottaviani, P. M.
Sheverdyaeva, P. Moras, C. Carbone, D. Topwal, B.
Olivieri, A. Kara, H. Oughaddou, B. Aufray, and G. L.
Lay, Appl. Phys. Lett. 96, 261905 (2010)
5
P. De Padova, C. Quaresima, B. Olivieri, P. Perfetti, and
G. Le Lay, Appl. Phys. Lett. 98, 081909 (2011)
6
P. Vogt, P. De Padova, C. Quaresima, J. Avila, E.
Frantzeskakis, M. C. Asensio, A. Resta, B. Ealet, and G.
Le Lay, Phys. Rev. Lett. 108, 155501 (2012)
7
C. L. Lin, R. Arafune, K. Kawahara, N. Tsukahara, E. Mi-
namitani, Y. Kim, N. Takagi, and M. Kawai, Appl. Phys.
Express 5, 045802 (2012)
8
A. Fleurence, R. Friedlein, T. Ozaki, H. Kawai, Y. Wang,
and Y. Yamada-Takamura, Phys. Rev. Lett. 108, 245501
(2012)
9
N. D. Drummond, V. Zolyomi, and I. V. Falko, Phys. Rev.
B 85, 075423 (2012)
10
Z. Ni, Q. Liu, K. Tang, J. Zheng, J. Zhou, R. Qin, Z. Gao,
D. Yu, and J. Lu, Nano Lett. 12, 113 (2012)
11
C. C. Liu, W. Feng, and Y. Yao, Phys. Rev. Lett. 107,
076802 (2011)
12
C. C. Liu, H. Jiang, and Y. Yao, Phys. Rev. B 84, 195430
(2011)
13
Z. Ni, Q. Liu, K. Tang, J. Zheng, J. Zhou, R. Qin, Z. Gao,
D. Yu, and J, Lu, Nano Lett. 12 113 (2012)
14
M. Ezawa, New J. Phys. 14, 033003 (2012)
15
M. Ezawa, J. Phys. Soc. Jpn. 81, 064705 (2012)
16
M. Tahir and U. Schwingenschlogl, Scientic reports 3,
1075 (2013)
17
M. Tahir, A. Manchon, K. Sabeeh, and U. Schwingen-
schlogl, Appl. Phys. Lett. 102, 162412 (2013)
18
C. J. Tabert and E. J. Nicole, Phys. Rev. Lett. 110, 197402
(2013)
19
C. J. Tabert and E. J. Nicole, Phys. Rev. B 87, 235426
(2013)
20
D. Weiss, K. von Klitzing, K. Ploog, and G. Weimann,
Euro. Phys. Lett. 8, 179 (1989)
21
R. R. Gerhardts, D. Weiss, and K. von Klitzing, Phy. Rev.
Lett. 62, 1173 (1989)
22
R. W. Winkler, J. P. Kotthaus, and K. Ploog, Phys. Rev.
Lett. 62, 1177 (1989)
23
F. M. Peeters and P. Vasilopoulos, Phys. Rev. Lett. 63,
2120 (1989)
24
C. Zhang and R. R. Gerhardts, Phys. Rev. B 41, 12850
(1990)
25
F. M. Peeters and P. Vasilopoulos, Phys. Rev. B 46, 4667
(1992)
26
C. W. J. Beenakker, Phy. Rev. Lett. 62, 2020 (1989)
27
F. M. Peeters and P. Vasilopoulos, Phys. Rev. B 47, 1466
(1993)
28
P. Vasilopoulos and F. M. Peeters, Superlatt. Microstruct.
7, 393 (190)
29
B. Das, D. C. Miller, S. Datta, R. Reifenberger, W. P.
Hong, P. K. Bhattachariya, J. Sing, and M. Jae, Phys.
Rev. B 39, 1411 (1989)
30
J. Nitta, T. Akazaki, H. Takayanagi, and T. Enoki, Phys.
Rev. Lett. 78, 1335 (1997)
31
X. F. Wang and P. Vasilopoulos, Phys. Rev. B 67, 085313
(2003)
32
S. F. Islam and T. K. Ghosh, J. Phys.: Condens. Matter,
24, 035302 (2012)
33
F. M. Peeters and P. Vasilopoulos, Phys. Rev. B 71, 125301
(2005)
34
SK F. Islam and T. K. Ghosh, J. Phys.: Condens. Matter
24, 185303 (2012)
35
Z. Tan, C. Tan, L. Ma, G. T. Liu, L. Lu, and C. L. Yang,
Phys. Rev. Lett. 84, 115429 (2011)
36
A. Matulis and F. M. Peeters, Phys. Rev. B 75, 125429
(2007)
37
R. Nasir, K. Sabeeh, and M. Tahir, Phys. Rev. B 81,
085402 (2010)
38
SK F. Islam ad T. K. Ghosh, J. Phys.: Condens. Matter
26 165303 (2014)
39
M. Charbonneau, K. M. Van Vilet, and P. Vasilopoulos, J.
Math. Phys. 23, 318 (1982)
40
I. S. Gradshtyen, I. M. Ryzhik, Table of Integrals, Series
and Products