MBE growth and magnetic properties of GaSb/MnSb superlattices
ABSTRACT We have grown a 37-period GaSb(2 A)/MnSb(A) superlattice on GaAs(001) substrate with 100 A GaSb buffer layer by solid-source molecular beam epitaxy. We have observed that the streaky RHEED patterns of GaSb and MnSb were maintained even when the growth was finished in 37 periods, indicating 2-dimensional layer-by-layer growth of MnSb on GaSb or vice versa. In-terestingly, a GaSb(2 A)/MnSb(A) superlattice showed ferromagnetic ordering up to above 400 K, with a coercive field of 380 Oe at 10 K.
Journal of the Korean Physical Society, Vol. 47, November 2005, pp. S497∼S499
MBE Growth and Magnetic Properties of GaSb/MnSb Superlattices
Jiyoun Choi, Jeongyong Choi, Sungyoul Choi, Soon Cheol Hong and Sunglae Cho∗
Department of Physics, University of Ulsan, Ulsan 680-749
M. H. Sohn
Korea Electrotechnology Research Institute, Changwon 641-120
Yongsup Park, Kyu-Won Lee and Hyun-Min Park
Korea Research Institute of Standards and Science, Daejon 305-600
Jong Hyun Song and J. B. Ketterson
Department of Physics and Astronomy, Northwestern University, Evanston, IL 60208, U.S.A.
We have grown a 37-period GaSb(25˚ A)/MnSb(2˚ A) superlattice on GaAs(001) substrate with
1000˚ A GaSb buffer layer by solid-source molecular beam epitaxy. We have observed that the
streaky RHEED patterns of GaSb and MnSb were maintained even when the growth was finished
in 37 periods, indicating 2-dimensional layer-by-layer growth of MnSb on GaSb or vice versa. In-
terestingly, a GaSb(25˚ A)/MnSb(2˚ A) superlattice showed ferromagnetic ordering up to above 400
K, with a coercive field of 380 Oe at 10 K.
PACS numbers: 73.21.Cd, 75.70.Cn
Keywords: MBE, Superlattices
Ferromagnetic semiconductors have attracted great in-
terest because of their potential spintronic device appli-
cations. By using ferromagnetic metal as a spin injec-
tor, the spin-polarization of electrons is easily caused
to disappear in the interface between metal and semi-
conductor, due to the scattering effect caused by the
differences in crystal structure, electronic structure and
chemical bonding and conductivity mismatch. Diluted
magnetic semiconductors (DMSs), in which transition
metals were substituted into semiconductors, have been
widely studied as a possible spin injector and/or spin
detector in spin-FET (field-effect transistor), spin-LED
(light-emitting diode) and spin-RTD (resistance temper-
ature diode), etc. . Ferromagnetism was observed in
various DMSs such as group IV, II-V, and III-V, etc. [2–
4]. However, the conventional DMS, in which magnetic
ions are randomly substituted into the semiconductor,
had low solubility of magnetic ions in the host semicon-
An approach to enhancing the solubility is to make a
monolayer superlattice, in which a magnetic monolayer is
periodically substituted between semiconducting layers.
Here, we report on the growth and magnetic properties
of a GaSb/MnSb superlattice: a monolayer of MnSb be-
tween GaSb semiconducting layers.
II. SUPERLATTICE GROWTH
We have grown a 37-period GaSb(25 ˚ A)/MnSb(2
˚ A) superlattice on GaAs(100) substrates with 1000
˚ A GaSb buffer layer by solid-source MBE (molecular
beam epitaxy, VGsemicon Inc.). The base pressure of
the growth chamber was on the order of 10−10Torr.
The growth temperature of the GaSb buffer layer and
GaSb/MnSb superlattice was 400◦C. The evaporation
ratio of Sb/(Ga, Mn) was about 2.
energy electron diffraction (RHEED) was used to exam-
ine the specific surface reconstruction, the growth mode,
and growth orientation of the deposited layers.
III. RESULTS AND DUSCUSSION
In growing a superlattice with sharp interfaces, 2-
dimensional (D) nucleation is preferred over 3D nucle-
ation, since the latter may introduce height variations
and other defects where the 3D islands coalesce. The
present in-situ RHEED experiments showed streaky pat-
terns for the MnSb layer grown on GaSb(100), as shown
-S498- Journal of the Korean Physical Society, Vol. 47, November 2005
Fig. 1. RHEED patterns of (a) 1000˚ A thick GaSb buffer
layer on GaAs(100), and (b) 37-period GaSb(25˚ A)/MnSb(2
˚ A) superlattice on GaSb buffer layer.
Fig. 2. θ-2θ XRD pattern of MnSb/GaSb superlattice on
in Figure 1 and also for the growth of GaSb on MnSb,
regardless of the GaSb layer thickness. These observa-
tions confirm 2D layer-by-layer growth for both MnSb
on GaSb and GaSb on MnSb.
In order to characterize the crystal structure of a
GaSb/MnSb superlattice, we performed θ-2θ X-ray
diffraction (XRD) studies as shown in Figure 2. Only
GaSb peaks are seen in the figure, except for the sub-
strate peaks, implying that the film is well aligned with
the GaSb(100) axis normal to the substrate. The lattice
constant of zinc-blende GaSb is a = 6.09˚ A. However,
it is known that metallic MnSb has a NiAs-type hexag-
onal crystal structure. We have searched for secondary
phases such as MnSb and MnSb2; none was observable.
In this XRD measurement, we could not observe super-
lattice satellites. One possible explanation on this issue
is that the 400◦C growth temperature may be enough
Fig.3.Temperature dependent magnetizationof
MnSb/GaSb superlattice. M-H curve at 10 K is shown in
Temperature dependent electrical resistivity of
to diffuse Mn into GaSb, resulting in the rough interface
between GaSb/MnSb in the superlattice. We may con-
clude the possible zinc-blende crystal structure growth
of a monolayer of MnSb between GaSb layers. Similarly,
zinc-blende growths of monolayers of MnAs and CrSb
were reported [5–9].
We investigated the magnetic and transport properties
of a 37-period GaSb(25˚ A)/MnSb(2˚ A) superlattice by
using magnetic and physical property measurement sys-
tems (MPMS and PPMS, Quantum Design, Inc.). The
temperature-dependent magnetization (M-T) data un-
der 200 Oe magnetic field and the M-H (magnetic field)
curve at 10 K are shown in Figure 3. M-T and M-H data
indicate that the superlattice has ferromagnetic (FM)
ordering above room temperature. At 10 K, they show
ferromagnetic hysteresis with coercive field of Hc= 380
Oe as shown in the inset of Figure 3.
Figure 4 shows the temperature-dependent electrical
resistivity of the GaSb/MnSb superlattice between 5 and
390 K. With increasing temperature, the electrical resis-
tivity slightly decreases.We obtained p-type carriers
with a carrier density of 1.34 × 1019cm−3at 300 K by
Hall measurement. The magnetoresistance (MR) data
at 10, 200, and 300 K are shown in Figure 5. At 200 and
MBE Growth and Magnetic Properties of GaSb/MnSb Superlattices – Jiyoun Choi et al.-S499-
Fig. 5. Magnetoresistance of MnSb/GaSb superlattice at
10, 200, and 300 K.
300 K, the MR change with magnetic field is positive,
while at 10 K it is negative. The negative MR at 10 K
may be due to dominant spin related scattering at low
temperature, which is typically observed in ferromagnets
including metal and ferromagnetic semiconductor. In the
conventional DMSs, Mn ions are distributed uniformly.
Thus, the conduction is governed by spin dependent scat-
tering with negative MR below Tc. However, in a rel-
atively ordered GaSb(25˚ A)/MnSb(2˚ A) superlattice we
expect two different types of dominant conduction: spin-
dependent scattering and phonon-related scattering, in
MnSb/neighboring GaSb layer and GaSb layer, respec-
tively. At low temperature with weaker phonon scatter-
ing, we may observe the negative MR only at low tem-
We have successfully grown a 37-period GaSb(25
˚ A)/MnSb(2 ˚ A) superlattice on GaAs(100) by using
MBE. We have observed that the streaky RHEED pat-
terns of GaSb and MnSb were maintained, even when
the growth was finished in 37 periods, indicating 2-
dimensional layer-by-layer growth of MnSb on GaSb or
vice versa. The RHEED pattern of the MnSb layer
was consistent with that of GaSb, indicating the possi-
ble growth of a monolayer of zinc-blende MnSb between
GaSb layers. A GaSb(25˚ A)/MnSb(2˚ A) superlattice
showed ferromagnetic ordering up to above 400 K, with
coercive field of 380 Oe at 10 K.
This work was supported by KOSEF via the Electron
Spin Science Center at POSTECH, KIST under Vision
21, KOSEF Grant # R08-2003-000-10353-0, the Brain
Korea 21 Project in 2003, and KISTEP Grant # M1-
 Scott A. Chambers, Materials Today 5, 34 (2002).
 N. Kim, S. J. Lee, T. W. Kang, K. S. Yi and G. Ihm, J.
Korean Phys. Soc. 39, 1050 (2002).
 S. J. Lee, S. Souma, N. Kim, J. W. Kim and T. W. Kang,
J. Korean Phys. Soc. 45, S647 (2004).
 H. J. Kim and K. S. Yi, J. Korean Phys. Soc. 42, S263
 J. H. Zhao, F. Matsukura, K. Takamura, E. Abe, D. Chiba
and H. Ohno, Appl. Phys. Lett. 79, 2776 (2001).
 Kanta Ono, Jun Okabayashi, Masaki Mizuguchi, Masa-
haru Oshima, Atsushi Fujimori and Hiro Akinaga, J.
Appl. Phys. 91, 8088 (2002).
 Ya-Qiong Xu, Bang-Gui Liu and D. G. Pettifor, Phys.
Rev. B 66, 184435 (2002).
 M. Shirai, J. Appl. Phys. 93, 6844 (2003).
 Bang-Gui Liu, Phys. Rev. B 67, 172411 (2003).