Superlattices and Microstructures 100 (2016) 867e875

Contents lists available at ScienceDirect

Superlattices and Microstructures

journal homepage: www.elsevier.com/locate/superlattices

Electron transport in AlxGa1
xAs d-MIGFETs: Conductivity

enhancement induced by magnetic field effects
O. Oubram a, I. Rodríguez-Vargas b, c, *, L.M. Gaggero-Sager d, L. Cisneros-

squez-Aguilar a, M. Limo
Villalobos a, A. Bassam e, J.G. Vela
n-Mendoza a
a
Facultad de Ciencias Químicas e Ingeniería, Universidad Auto
noma del Estado de Morelos, Av.Universidad 1001, Col. Chamilpa, 62209
Cuernavaca, Morelos, Mexico
b
Unidad Acad

noma de Zacatecas, Calzada Solidaridad Esquina con Paseo la Bufa S/N, 98060 Zacatecas,
emica de Física, Universidad Auto
Zac., Mexico
c
n en Ciencias, IICBA, Universidad Auto
Centro de Investigacio
noma del Estado de Morelos, Av. Universidad 1001, Col. Chamilpa, 62209
Cuernavaca, Morelos, Mexico
d
CIICAp, IICBA, Universidad Auto
noma del Estado de Morelos, Av. Universidad 1001, Col. Chamilpa, 62209 Cuernavaca, Morelos, Mexico
e
Facultad de Ingeniería, Universidad Auto
noma de Yucata
n, Av. Industrias no Contaminantes por Perif

erico Norte, Apdo. Postal 150
Cordemex, M

n, Mexico
erida, Yucata

a r t i c l e i n f o a b s t r a c t

Article history: The electronic structure and the transport phenomena of d-MIGFETs have been studied in
Received 13 June 2016 an AlxGa1
xAs host matrix. The subband structure and mobility calculations were per-
Received in revised form 12 September 2016 formed within the effective mass approximation and relative mobility formula, respec-
Accepted 19 October 2016
tively. Both the electronic structure and the transport properties are calculated as
Available online 21 October 2016
dependent on the applied magnetic field (B), the aluminum molar fraction (x) and the
contact voltage in one of the gates (VC1). It was found that the mobility and conductivity
Keywords:
are enhanced by increasing the magnetic field for appropriate aluminum molar fraction
Electron transport
Mobility
and contact voltage. In particular, the mobility (conductivity) is improved almost 26% (32%)
Conductivity for VC1 ¼ 900 meV (850 meV), x ¼ 0.2, and B ¼ 20 T.
AlxGa1
xAs © 2016 Elsevier Ltd. All rights reserved.
d-MIGFET

1. Introduction

Low Dimensional Electronic Devices (LDEDs) have always been a technological challenge for industry and being an interest
subject for the scientific community due their tremendous properties and applications [1e24].
Thanks to the rapid development of the growth techniques, such as Molecular Beam Epitaxy (MBE) and Metal-Organic
Chemical Vapor Deposition (MOCVD), the possibilities for design devices with variety of potential profiles have increased
[1e15,21,25e27]. Undoubtedly, these new structures open an unlimited field in electrical and optical applications. Specially,
they offer excellent transport properties. For instance, LDEDs based on d-doped semiconductor structures provide high
mobility and conductivity as compared with conventional devices [1e15,21]. A key point of these delta-doped semiconductor
structures is the incorporation of an atomic layer of impurities. The so called delta-doping provides high concentration of

* Corresponding author. Permanent address: Unidad Acade
mica de Física, Universidad Auto

noma de Zacatecas, Calzada Solidaridad Esquina con Paseo a
la Bufa S/N, 98060 Zacatecas, Zac., Mexico.
E-mail address: isaac@fisica.uaz.edu.mx (I. Rodríguez-Vargas).

http://dx.doi.org/10.1016/j.spmi.2016.10.056
0749-6036/© 2016 Elsevier Ltd. All rights reserved.
868 O. Oubram et al. / Superlattices and Microstructures 100 (2016) 867e875

charge carriers as well as efficient quantum confinement. By tailoring the design and characteristics of the delta-doped
structures the transport properties can be improved substantially. For instance, Ploog et al. [5] report that a delta-doped
Field Effect Transistor or d-FET in GaAs attains high electron concentration, large breakdown voltage, high intrinsic trans-
conductance and high mobility. Likewise, Chakhanakia and collaborators [6] report high transconductance and large linearity
in d-FETs. Kao et al. [28] also investigate the transport properties in Simple, Double and Triple-d-doped GaAs FETs. They
demonstrated that the Triple-d-doped GaAs structure achieved an extremely high carrier density along with an enhanced
mobility.
Recently, new LDEDs, called multi-gate transistors, have earned great popularity due to its outstanding transport prop-
erties. Double Gate (DG) MOSFET transistor is one of them. These DG-MOSFETs offer: better control of short channel effects,
low subthreshold leakage and higher On-Current, high scalability and an excellent level of transconductance [21e24,30e36].
In short, this type of transistors reveals better transport properties with respect to conventional ones. To this respect,
Chaisantikulwat et al. [21] have studied by magnetoresistance measurements the electron mobility in DG-MOSFETs. They
showed that the mobility in DG-MOSFETs is significantly higher than in Simple Gated MOSFETs.
Other possibility to improve the transport properties of LDEDs is through the incorporation of external effects such as
electric and magnetic fields, electromagnetic radiation and hydrostatic pressure to name a few [1,37e45]. In the case of the
magnetic field, it has been reported that it can be a flexible tool to control the electronic structure and adjust the mobility and
conductivity in d-doped systems [1,41e45]. Within this context, Dong-Sheng Luo et al. [1] report transport measurements on
a d-doped quantum well system with extra modulation doping and with tilted-magnetic field. They show that the resistivity is
very sensitive to magnetic field. Similarly, Kulbachinskii et al. [41] have realized a theoretical and experimental study about
magnetoresistance in Sn d-doped GaAs with different orientations of the magnetic field. They report that the magnetore-
sistance decreases as a perpendicular magnetic field. The magnetoresistance also shows oscillations at strong magnetic fields.
This oscillatory behavior of the magnetoresistance in d-doped GaAs was reported by Kul'bachinskii et al. [42] as well. They
attributed these properties to the emptying and filling of subbands as the magnetic field increases. Recently, it has been
reported that the confinement, population and localization in p-d-doped quantum wells are quite sensitive to applied
magnetic fields [2]. Likewise, significant enhancement of the mobility with the magnetic field in double Si d-doped GaAs is
documented [43]. E. Daw et al. [44] have experimentally studied the effect of high magnetic fields on the transport properties
and noise temperature of a Heterostructure Field Effect Transistor (HFET). They demonstrate that the transconductance
decreases and the noise temperature increases as a function of the magnetic field. They explained such behavior as a
consequence of the alteration of electron paths, between the source and drain of the HFET, caused by the Lorentz force
induced by the applied field [44].
On the other hand, ternary alloys are other possibility to enhance the transport properties of delta-doped-based LDEDs
[13,46e49]. In particular, the molar fraction of the aluminum concentration in AlxGa1
xAs delta-doped systems can be
adjusted to modulate the degree of confinement and localization of the charge carriers, and hence to change (improve) the
transport properties in these systems. For instance, Lamas et al. had grown AlxGa1
xAs parabolic quantum wells by MBE on
top of GaAs (311) substrates. They found that the mobility was more than twice higher than samples grown on GaAs (100)
substrate [9]. Recently, Sahoo and Sahu [13] analyze the influence of an applied electric field on the multisubband electron
mobility in a barrier delta-doped AlxGa1
xAs parabolic quantum well structure. They found a sudden enhancement of the
mobility at an applied electric field. They attributed this behavior to the transition from double subband occupancy to single
subband one.
In this work, we study the electronic structure, mobility (mrel) and conductivity (srel) in a d-MIGFET with a ternary alloy
(AlxGa1
xAs) as the semiconductor host matrix. In particular, we take advantage of the high electron density provided by the
d-doping, the modulation of the confinement via DG electrodes and the aluminum concentration to show that an applied
magnetic field can improve significantly the transport properties. In fact, by appropriately choosing the relevant parameters
of the d-doped MIGFET the mobility and conductivity can be improved more than 23% and 30% respectively, by adjusting the
magnetic field to 20 T.

2. Theoretical background

The system that we are interested in this work is the d-MIGFET (d-Multiple Independent Gate Field Effect Transistor) in an
AlxGa1
xAs host matrix (see Fig. 1). The transistor is formed by a double Metal-Semiconductor contact (MSC) also known as
Schottky barrier, and a two-dimensional electron gas (2DEG) as conduction channel between the double MSC. Wherein, the
double MS controls the conduction channel, which is formed by a delta-doping layer. The degree of confinement can be
controlled through the contact voltages applied to the MSCs, the aluminum concentration in the semiconductor host matrix,
and the magnetic filed strength. So, we will pay special attention to how these effect are incorporated in our theoretical
description.
The first thing to do in order to provide an adequate description of a d-MIGFET in AlxGa1
xAs under the influence of a
magnetic field is to find the energy spectrum of the quantum structure using the Schro€dinger equation. In order to do that, we
need the confinement potential, which will be a combination of the potentials due to MSCs, the delta-doped quantum well
and the applied magnetic field. In the case of the MSCs and the delta-doped quantum well the potential comes as [8]:
 O. Oubram et al. / Superlattices and Microstructures 100 (2016) 867e875 869

Fig. 1. Schematic representation of a d-MIGFET in AlxGa1
xAs. The conduction channel (d-doped quantum well), drain, source and gates are shown. The main
advantage of this device is the higher confinement of charge carriers achieved by the double gate and d-doped quantum well. In addition, the aluminum
concentration and the magnetic field further improve the transport properties.

2pe2 a2n
 0  2pe2

2
VðzÞ ¼ Nd ðz þ d
lÞ2 qð
d þ l
zÞ
q z
lp q l
z þ N ðz
d þ l0 Þ qðz
d þ l0 Þ; (1)
εr ðxÞ ðan jzj þ z0n Þ4
p
εr ðxÞ d

where εr(x) is the electric permittivity constant of AlxGa1
xAs, which is dependent on the aluminum concentration. Here, we
focus our attention in the direct gap regime of the ternary alloy, that is, when the aluminium concentration is x < 0.45. Hence,
the fundamental physical parameters with aluminum concentration are given as [50]: εr(x) ¼ 13.18
3.12x and
m*(x) ¼ 0.067 þ 0.083x, where m*(x) is the electron effective mass in AlxGa1
xAs.
In this model, we locate the first (second) MSC at z ¼
d (z ¼ d), with z the growth direction. The delta-doped quantum
0
well is located at z ¼ 0 l and l are the screening distances for the top and bottom MSCs (see Fig. 1), which are given as:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
εr ðxÞVC1
l¼ ; (2)
2pe2 Nd

and
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0 εr ðxÞVC2
l ¼ ; (3)
2pe2 Nd

being e the elementary charge, Nd ¼ 1  1018 cm
3 the background impurity density, VC1 the first contact voltage, VC2 the
second contact voltage and q the Heavside step function. The other potential parameters are an ¼ 2/(15p) and
z0n ¼ ða3n =pN2d Þ1=5 . N2d ¼ 7.5  1012 cm
2 is the two-dimensional impurity density of the n-type d-doped quantum well. In
0
our model l and l also represent the width of the depletion regions, so, V(lp) ¼ 0 and Vðl0p Þ ¼ 0. Then, under these conditions,
distances lp and l0p delimit the depletion regions.
Once, we have completed the model for d-MIGFET, we solve the following Schro €dinguer equation, in which we are
including the contribution of the applied magnetic field:
 2 2 
Z2 d2 jðzÞ e B

 þ þ VðzÞ jðzÞ ¼ EjðzÞ; (4)
2m ðxÞ dz2 2m ðxÞc2

where c is the speed of light in free space and B is the magnetic field strength applied perpendicular to the growth direction.
After obtaining the electronic structure, we can compute the relative mobility as [8,51]:
Z Z
mVC1 ;x;B reVC1 ¼500meV;x¼0;B¼0T ðz0 Þrimp ðzÞ z
z0 dzdz0
mrel ¼ ¼ Z Z ; (5)
mVC1 ¼500meV;x¼0;B¼0T reVC1 ;x;B ðz0 Þrimp ðzÞ z
z0 dzdz0

where: rVe C1 ;x;B ðz0 ÞðrVe C1 ¼500meV;x¼0;B¼0 ðz0 ÞÞ is the density of electrons when the first contact potential is VC1 (VC1 ¼ 500 meV), the
magnetic field is B (B ¼ 0) and the Al concentration is x (x ¼ 0) respectively; rimp(z) is the density of impurities, with
rimp ¼ N2d  d(z) and d(z) the Dirac delta function. It is worth mention that in this work, we only consider the ionized donor
scattering mechanism, because it is the most important at low temperature. By substituting the electron and impurity
densities and integrating over z we can write the relative mobility as,
870 O. Oubram et al. / Superlattices and Microstructures 100 (2016) 867e875

Z
P Ref 2 h

i

ðm ðx ¼ 0Þ=εr ðx ¼ 0ÞÞ i Fi ðz0 Þ ln 1 þ eðEF
Ei Þ=kB T z0 dz0
Ref Ref

mrel ¼ Z ;
P 2 h i (6)

ðm ðxÞ=εr ðxÞÞ i FiVar ðz0 Þ ln 1 þ eðEF
Ei Þ=kB T z0 dz0
Var Var

0
where EF, Ei and Fi(z ) are the Fermi energy, the i-th energy level and the i-th envelope function. Ref stands for the system with
VC1 ¼ 500 meV, x ¼ 0 and B ¼ 0 T, while Var stands for a system with arbitrary VC1, x and B. In addition, kB and T represent the
Boltzmann constant and the system temperature. In the limit of zero temperature the latter expression can be written as,
Z
P 2 



ðm ðx ¼ 0Þ=εr ðx ¼ 0ÞÞ i FiRef ðz0 Þ EFRef
EiRef z0 dz0
mrel ¼ Z :
P 2   (7)

ðm ðxÞ=εr ðxÞÞ i FiVar ðz0 Þ EFVar
EiVar z0 dz0

The relative density can be defined in the same sense as the mobility,

nVC1 ;x;B
nrel ¼ : (8)
nVC1 ¼500meV;x¼0;B¼0T

And the conductivity as the product of the relative density and the relative mobility

srel ¼ nrel  mrel : (9)

At this point, it is important to mention that the transport properties, mobility and conductivity, that we will show in the
next section correspond to the zero temperature limit.

3. Results and discussion

As mentioned above, the properties that we are interested in, are the mobility and conductivity of a d-MIGFET in Alx-
Ga1
xAs. Especially, we focus our attention to the effects of the magnetic field, Al concentration and contact voltages on the
transport properties. Therefore, we start to analyze their effects on the electronic structure of the device.
Fig. 2 (a) shows the confining potential profile and the subband energies with their envelope wavefunctions for x ¼ 0.2,
VC1 ¼ 500 meV and VC2 ¼ 300 meV. The solid and dashed lines correspond to magnetic fields of B ¼ 0 T and B ¼ 10 T,
respectively. As we can see the main effect of magnetic field for B ¼ 10 T is the rise of Schottky barrier height. Besides, the
delta-doped quantum well potential is practically unaffected by the magnetic field, as a consequence the electron subbands
are slightly modify.

600
B= 0T 0 Ef
500 B=10T
(a) (b)
400 -25
E2
Energy (meV)

300 -50 E1
V (meV)

200
-75
100
-100
0
-100 -125
E0
-200 -150
-300-200-100 0 100 200 300 0 10 20 30 40
z (Angstrom) B (T)
Fig. 2. (a) Confinement potential profile, subband energy levels and their corresponding wave functions as a function of the growth direction z. The solid and
dashed curves correspond to magnetic fields of 0 and 10 T, respectively. The contact voltages considered were VC1 ¼ 500 meV and VC2 ¼ 300 meV, while the
aluminum composition was x ¼ 0.2. (b) Electron energy levels as a function of magnetic field for the same contact voltages and aluminum concentration as in (a).
 O. Oubram et al. / Superlattices and Microstructures 100 (2016) 867e875 871

Additionally, in Fig. 2 (b) we show the electron subbands as a function of the magnetic field strength. The contact voltages
and the aluminum concentration are the same as in Fig. 2 (a). As we can see the degree of confinement of the electron
subbands and the Fermi level decreases as the magnetic field strength increases. This trend is compatible with the one re-
ported in different quantum structures [2,18,43,52]. It is worth to mention that the number of filled states changes with the
magnetic field, since for B < 18 T there are three filled states, while for B > 18 T there are only two. This change in the number
of filled subbands is in agreement with depopulation reported in d-doped GaAs quantum wells under magnetic field [2,41].
We can also note that the upper levels are more susceptible to the magnetic field than the ground state ones.
In Fig. 3 we show the electronic structure of a d-MIGFET in AlxGa1
xAs as a function of the aluminum concentration. We
have fixed the contact voltages at VC1 ¼ 500 meV and VC2 ¼ 300 meV and the magnetic field at B ¼ 20 T. As we can see the
electron levels and the Fermi energy decrease as the aluminum concentration increases, in other words, the degree of
confinement increases as a function of the aluminum concentration. Even when the effect of the aluminum concentration on
the electronic structure is opposite to the one induced by the magnetic field, the depopulation of the electron levels is also
presented. In particular, the total number of filled states for x < 0.14 is three, and two for x > 0.14. This is due basically to the
change of the physical parameters, the effective mass m*(x) and the dielectric constant ε(x), with the aluminum concentration
[50]. Therefore if the aluminum concentration can modify the electronic structure, then it is also expected that the transport
properties be affected.
Let us turn our attention to the transport properties. In Fig. 4 we present the relative mobility and relative conductivity of
the d-MIGFET in AlxGa1
xAs as a function of the first contact voltage. We have considered five different magnetic fields B ¼ 0 T,
B ¼ 5 T, B ¼ 10 T, B ¼ 15 T and B ¼ 20 T, which correspond to the solid, dashed, dotted-dashed, double-dotted-dashed and
dotted curves, respectively. We set the second contact voltage to VC2 ¼ 300 meV and the aluminum concentration to x ¼ 0. We
can notice that as the magnetic field increases, both the mobility and conductivity, increase as well. The enhancement of these
quantities spans practically all contact voltages. Indeed, our results follow the same trend as in simple Sn and double Si d-
doped GaAs quantum wells under the influence of a magnetic field [41,43]. We also note that as the contact voltage increases,
the mobility and conductivity increase in a linear fashion, until reach an optimal value. Thence, the mobility and conductivity
tend to decrease. This drop of the transport properties can be understood if we consider that for large contact voltages the
conduction channel tends to close. The improvement of the mobility and conductivity can reach 20% and 30% by choosing the
magnetic field and the contact voltage at B ¼ 20 T and VC1 ¼ 800 meV, respectivey. At this point, it is also important to mention
that the upward linear region and the downward one could be of interest technologically speaking, since the former could
provide stable electronic amplification, while the latter could help to design oscillatory circuits.
In Fig. 5 we show the same as in Fig. 4, but now the aluminum concentration is x ¼ 0.2. As we can see the transport
properties sustain the same trend as in the case of x ¼ 0, that is, the mobility and the conductivity improve as the magnetic
field increases, the transport properties show a linear upward region and a descending one, and the enhancement of the
transport properties is presented irrespective of the contact voltage considered. The main effect of the change in the
aluminum concentration is a sudden rise in the transport properties, more pronounced in the conductivity. In fact, the
mentioned rise is red-shifted as the magnetic field is increased.
We can also see that the mobility is less sensitive to the magnetic field, and as the conductivity is the product of the
mobility and the charge density, we can expect that the pronounced rise in the conductivity comes from the number of
subbands involved in the transport as well as how the charge carriers are distributed on these subbands. To this respect, we
can notice that for the contact voltages at which the abrupt changes in the conductivity take place correspond to the same
voltages at which the number of subbands involved in the transport changes. In particular, the number of subbands

0
Energy (meV)

-50

E0
-100
E1
E2
Ef
-150

0 0.1 0.2 0.3 0.4

Al concentration, x
Fig. 3. Energy levels as a function of the aluminium composition. The contact voltages were fixed at VC1 ¼ 500 meV and VC2 ¼ 300 meV, and the magnetic field at
B ¼ 20 T.
872 O. Oubram et al. / Superlattices and Microstructures 100 (2016) 867e875

1.2
1.1
(a)
1 B=0T

μrel
0.9 B=5T
0.8 B=10T
0.7 B=15T
0.6 B=20T
100 200 300 400 500 600 700 800 900 1000
1.2 (b)
σrel 1.1
1
0.9
0.8
100 200 300 400 500 600 700 800 900 1000
VC1(meV)

Fig. 4. (a) Relative mobility and (b) relative conductivity as a function of the first contact voltage VC1. The solid, dashed, dotted-dashed, double-dotted-dashed and
dotted curves correspond to magnetic fields of 0, 5, 10, 15 and 20 T, respectively. The second contact voltage is fixed to VC2 ¼ 300 meV, and the aluminum
concentration to x ¼ 0.

1.2 (a)
1.1 B=0T
B=5T
μrel

1
0.9 B=10T
0.8 B=15T
B=20T
0.7
100 200 300 400 500 600 700 800 900 1000
1.3 (b)
1.2
σrel

1.1
1
0.9
100 200 300 400 500 600 700 800 900 1000
VC1 (meV)

Fig. 5. The same as in Fig. 4, but now x ¼ 0.2.

diminishes at those mentioned voltages, see Fig. 6. This kind of behavior has been also reported in AlxGa1
xAs quantum wells
[46e49]. For instance, Trinath et al. [49] have studied the transport properties in delta-doped AlxGa1
xAs parabolic quantum
wells. They found that the mobility is improved as the aluminum concentration increases. They also attribute this
improvement to the change in the number of subbands involved in the transport. In our case the overall enhancement for the
mobility is of 26% for B ¼ 20 T, x ¼ 0.2 and VC1 ¼ 900 meV. In the case of the conductivity the enhancement is of 32% for
VC1 ¼ 850 meV with the same magnetic field and aluminum concentration as for the mobility. At this point, it is important to
mention that for the mobility and conductivity is quite relevant not only the number of subbands, but also: 1) the carriers
density in each subband, 2) how the carriers are distributed and 3) the interaction between carriers and impurities. All these
details are included in the expressions for the transport properties, see Eq. (7). In particular, the denominator of this equation
is fundamental. For example, if we have three subbands, the denominator will have three terms. However, if one of these
terms is not significant, we will not see a sudden change in the mobility, even when the subband closes. This is precisely what
happen in Fig. 4, and in Figs. 5 and 7 when the magnetic field increases. To put this in numbers, we have computed the
mentioned weights for the mobility in Figs. 4 and 5 prior to the closing of the subband. For x ¼ 0.2 and B ¼ 0 the weights for
the first, second and third subband are 46%, 44% and 10%. In this case the contribution of the third subband is significant, then
when this subband closes the numbers in the first and second subband changes, causing a sudden change in the mobility. If
we increase the magnetic field up to 20 T the numbers become 50%, 46% and 4%. So, the weight of the third subband is less
important and the sudden change in the mobility is barely perceptible. For x ¼ 0 and B ¼ 20 T the weights are 50%, 48% and 2%.
In this case, the weight of the third subband is practically insignificant, and consequently we cannot see the sudden rise of the
 O. Oubram et al. / Superlattices and Microstructures 100 (2016) 867e875 873

x=0 x=0.2
200
(a) (b)

EF-EI (meV)
150 150
100 B=0T 100 B=0T
50 50
0 0
x=0 x=0.2
200 200
(d)
EF-EI (meV)
150
(c) 150
100 B=20T 100 B=20T
50 50
0 0
200 400 600 800 1000 200 400 600 800 1000
VC1(meV) VC1(meV)

Fig. 6. The subband occupancy is proportional to the difference between the Fermi energy and the electron energy levels. Hence, in this figure we are showing
the mentioned occupancy as a function the first contact voltage. (a) and (c) correspond to B ¼ 0 T and B ¼ 20 T with x ¼ 0, while (b) and (d) to B ¼ 0 T and B ¼ 20 T
with x ¼ 0.2. It is also important to remark that the magnetic field and the aluminum concentration clearly change the number of subbands involved in the
transport. And even more, this change (decrease) of the subbands coincides with the sudden rise of the transport properties as a function of the contact voltage.

mobility. Other important number to appreciate the magnitud of the change after the closing of the subband is the relative
change of the denominator. For x ¼ 0.2 and B ¼ 0 T there is a reduction of 11% in the denominator, while for B ¼ 20 T the
reduction is only 5%. For x ¼ 0 and B ¼ 20 T the reduction is of 3%. Then, as far as we can see it is quite important how relevant
the subband prior to the closing is.
In Fig. 7 we show our results for the transport properties, mobility and conductivity, but in this case x ¼ 0.3. If we compare
Figs. 4, 5 and 7, we can notice that by increasing the aluminum concentration the mobility and conductivity improve in
general. However, this improvement is not monotonic with the aluminum concentration. As far as we can see the optimum
values for the transport properties are obtained when x ¼ 0.2. This comes from the fact that the mobility and conductivity are
quite sensitive to the number of energy levels (transport channels) in the system, the degree of confinement (spatial dis-
tribution of the wave functions) of the charge carriers and the density of carriers in the energy levels. Then, when we change
the aluminum concentration we are modifying the electron effective mass and the electric permittivity, and hence, all
mentioned characteristics and consequently the transport properties. It is important to mention that we also explore the case
x ¼ 0.4, results not shown. We find that the mobility and conductivity present similar characteristics that for x ¼ 0.2 and
x ¼ 0.3. However, the improvement is not as good as in the case of x ¼ 0.2.
To end this section, we consider that it is important to comment two points about temperature:

1 Our low-temperature assumption, as far as we know, it is quite practical. In fact, the design and modelling of practically all
devices, from the historical pn junction to current cutting-edge devices, start with this consideration. This is under-
standable if we consider that the calculations of electronic structure at non zero temperature are quite complicated. In our
system this assumption is even more relevant if we take into account that temperature barely affects the electronic
structure [53]. Furthermore, the improvement of mobility at low temperature is quite relevant because new quantum
phenomena arises under such conditions. For instance, mobility has been enhanced by a factor of 7000 in 30 years [54].
From the technological standpoint the low-temperature assumption is also important, especially in space applications, in
which temperature is not a problem.
2 In our system it is known that the electronic structure is practically unaffected by the temperature in the range of 0e60 K
[53]. Then, we can in principle extend our calculations for the transport properties to this temperature range. Specifically,
we can use the electronic structure that correspond to zero temperature and then incorporate temperature effects via the
electron charge density re(z). Even, we could consider temperatures above 80 K, since the changes in the carrier con-
centration are not significant [53].

4. Conclusions

In summary, we have studied the electronic structure and transport properties of a d-MIGFET in AlxGa1
xAs under the
influence of an external magnetic field. The natural design of the d-MIGFET gives the possibility to enhance the transport
properties by adjusting the aluminum concentration and the contact voltages in the MSCs. Here, we have shown that the
mobility and conductivity can be further improved by increasing the magnetic field strength. As far as we could corroborate
874 O. Oubram et al. / Superlattices and Microstructures 100 (2016) 867e875

1.2 (a)
1.1
B=0T

μrel
1 B=5T
0.9 B=10T
B=15T
0.8 B=20T
100 200 300 400 500 600 700 800 900 1000
1.3
(b)
1.2
σrel 1.1
1
0.9
100 200 300 400 500 600 700 800 900 1000
VC1(meV)

Fig. 7. The same as in Fig. 5, but now x ¼ 0.3.

this improvement obeys the change in the degree of confinement in the system. In particular, the magnetic field greatly affects
the occupancy of the electron subbands, changing the landscape of the transport properties.

References

[1] D.-S. Luo, L.-H. Lin, Y.-C. Su, Y.-T. Wang, Z.F. Peng, S.-T. Lo, K.Y. Chen, Y.-H. Chang, J.-Y. Wu, Y. Lin, S.-D. Lin, J.-C. Chen, C.-F. Huang, C.-T. Liang, Nanoscale
Res. Lett. 6 (2011) 139.
[2] E. Ozturk, Chin. Phys. Lett. 27 (2010) 077302.
[3] L.M. Gaggero-Sager, R. Pe


rez-Alvarez, J. Appl. Phys. 78 (1995) 4566.
[4] E.F. Schubert, A. Fischer, K. Ploog, IEEE Trans. Electron. Dev. 33 (1986) 625.
[5] K. Ploog, M. Hauser, A. Fisher, Appl. Phys. A 45 (1988) 233.
[6] Z.D. Chakhnakia, L.V. Khvedelidze, N.P. Khuchua, R.G. Melkadze, G. Peradze, T.B. Sakharova, Proc. SPIE 5401 (2004) 354.
[7] A. Baraldi, C. Ghezzi, A. Parisini, A. Bosacchi, S. Franchi, Appl. Surf. Sci. 50 (1991) 405.
[8] O. Oubram, L.M. Gaggero-Sager, A. Bassam, G.A. Luna-Acosta, PIER 110 (2010) 59.
[9] T.E. Lamas, A.A. Quivy, C.S. Sergio, G.M. Gusev, J.C. Portal, J. Appl. Phys. 97 (2005) 076107.
[10] G. Yu, S.A. Studenikin, A.J. SpringThorpe, G.C. Aers, D.G. Austing, J. Appl. Phys. 97 (2005) 103703.
[11] K.H. Gao, G. Yu, Y.M. Zhou, W.Z. Zhou, T. Lin, J.H. Chu, N. Dai, A.J. Spring Thorpe, D.G. Austing, J. Appl. Phys. 105 (2009) 013712.
[12] S. Palo, N. Sahoo, T. Sahu, Phys. E 64 (2014) 33.
[13] N. Sahoo, T. Sahu, J. Semicond. 35 (2014) 012001.
[14] O. Oubram, L. Cisneros-Villalobos, L.M. Gaggero-Sager, M. Abatal, Rev. Mex. Fis. 60 (2014) 22.
[15] R.S. Balmer, I. Friel, S. Hepplestone, J. Isberg, M.J. Uren, M.L. Markham, N.L. Palmer, J. Pilkington, P. Huggett, S. Majdi, R. Lang, J. Appl. Phys. 113 (2013)
033702.
[16] V. Grimalsky, O. Oubram, S. Koshevaya, C. Castrejon-M, in: Proc. 29th International Conference on Microelectronics (MIEL 2014), Belgrade, Serbia, 2014,
pp. 241e244.
[17] V. Grimalsky, L.M. Gaggero-Sager, S. Koshevaya, Phys. B 406 (2011) 2218.
[18] E.C. Niculescu, M. Cristea, A. Radu, Phys. E 57 (2014) 138.
[19] E. Ozturk, I. So €kmen, Superlattices Microstruct. 48 (2010) 312.
[20] M.E. Mora-Ramos, L.M. Gaggero-Sager, C.A. Duque, Phys. E 44 (2012) 1335.
[21] W. Chaisantikulwat, M. Mouis, G. Ghibaudo, S. Cristoloveanu, J. Widiez, M. Vinet, S. Deleonibus, Solid State Electron 51 (2007) 1494.
[22] T. Poiroux, M. Vinet, O. Faynot, J. Widiez, J. Lolivier, T. Ernst, B. Previtali, S. Deleonibus, Microelectron. Eng. 80 (2005) 378.
[23] L. Mathew, et al., in: Digest of Technical Papers 2005 Symposium on VLSI Technology, 2005, pp. 200e201.
[24] J.W. Yang, J.G. Fossum, IEEE Trans. Electron Dev. 52 (2005) 1159.
[25] O. Oubram, L.M. Gaggero-Sager, O. Navarro, M. Ouadou, PIER 118 (2011) 37.
[26] N. Li, X. Li, S. Quan, PIER 122 (2012) 47.
[27] J.C. Martínez-Orozco, K.A. Rodríguez-Magdalenoa, J.R. Su
arez-Lpez, C.A. Duque, R.L. Restrepo, Superlattices Microstruct. 92 (2016) 166.
[28] M.J. Kao, W.C. Hsu, R.T. Hsu, Y.H. Wu, T.Y. Lin, C.Y. Chang, Appl. Phys. Lett. 66 (1995) 2505.
[30] J.F. Gong, P.C.H. Chan, M. Chan, Solid-State Electron. 52 (2008) 259.
[31] J. Kretz, L. Dreeskornfeld, J. Hartwich, W. Ro € sner, Microelectron. Eng. 67e68 (2003) 763.
[32] V.M. Srivastava, K.S. Yadav, G. Singh, Microelectron. J. 42 (2011) 1124.
[33] P. Baviskar, S. Jain, P. Vinchurkar, Int. J. Comput. Appl. 1 (2010) 42.
[34] M.J. Kumar, G.V. Reddy, Jpn. J. Appl. Phys. 44 (2005) 6508.
[35] G.-X. Hu, R. Liu, Z.-J. Qiu, L.-L. Wang, T.-A. Tang, Jpn. J. Appl. Phys. 49 (2010) 034001.
[36] V.M. Srivastava, K.S. Yadav, G. Singh, J. Comput. Electron 10 (2011) 229.
[37] E. Kasapoglu, H. Sari, I. So €kmen, Superlattices Microstruct. 29 (2001) 25.
[38] I. Karabulut, M.E. Mora-Ramos, C.A. Duque, J. Lumin. 131 (2011) 1502.
[39] H. Sghaier, L. Sfaxi, L. Bouzaïene, H. Maaref, Sens. Actuator A Phys. 113 (2004) 147.
[40] M.T. Pe
rez-Maldonado, L.M. Gaggero-Sager, G. Monsivais, C. Rodríguez-Castellanos, Rev. Mex. Fis. S 53 (2007) 135.
[41] V.A. Kulbachinskii, V.G. Kytin, R.A. Lunin, A.V. Golikov, V.G. Mokerov, A.S. Bugaev, A.P. Senichkin, R.T.F. van Schaijk, A. de Visser, P.M. Koenraad,
Semicond. Sci. Technol. 14 (1999) 1034.
[42] V.A. Kul’bachinskii, V.G. Kytin, R.A. Lunin, V.G. Mokerov, A.P. Senichkin, A.S. Bugaev, A.L. Karuzskii, A.V. Perestoronin, R.T.F. van Schaijk, A. de Visser,
Semiconductors 33 (1999) 771.
[43] E. Ozturk, Superlattices Microstruct. 46 (2009) 752.
 O. Oubram et al. / Superlattices and Microstructures 100 (2016) 867e875 875

[44] E. Daw, R.F. Bradley, J. Appl. Phys. 82 (1997) 1925.

[45] S.-T. Lo, K.Y. Chen, Y.-C. Su, C.-T. Liang, Y.H. Chang, G.-H. Kim, J.-Y. Wu, S.-D. Lin, Solid State Commun. 150 (2010) 1104.
[46] N. Sahoo, T. Sahu, J. Appl. Phys. 116 (2014) 043703.
[47] N. Sahoo, T. Sahu, AIP Adv. 4 (2014) 127106.
[48] T. Sahu, A.K. Panda, S. Palo, in: Proc. 2012 International Conference on Emerging Electronics (ICEE), Mumbai, India, 2012.
[49] T. Sahu, S. Palo, A.K. Panda, J. Appl. Phys. 113 (2013) 083704.
[50] S. Adachi, J. Appl. Phys. 58 (1985) R1.
[51] I. Rodríguez-Vargas, L.M. Gaggero-Sager, V.R. Velasco, Surf. Sci. 537 (2003) 75.
[52] E. Ozturk, I. So€kmen, J. Lumin. 145 (2014) 387.
[53] L.M. Gaggero-Sager, R. Pe


rez-Alvarez, Appl. Phys. Lett. 70 (1997) 212.
[54] E.H. Hwang, S. Das Sarma, Phys. Rev. B 77 (2008) 235437.