Magnetic out-of-plane component in MnAs/GaAs(001)

Article abstract of DOI:10.1063/1.1615682

The magnetic out-of-plane component in MnAs/GaAs film was discussed. Its temperature dependence was substantially different from the dominating in-plane magnetization. The analysis showed that the out-of-plane component was due to small isolated magnetic

Full text of DOI:10.1063/1.1615682

Magnetic out-of-plane component in MnAs/GaAs(001)
A. Ney, T. Hesjedal, C. Pampuch, J. Mohanty, A. K. Das, L. Däweritz, R. Koch, and K. H. Ploog
Citation: Applied Physics Letters 83, 2850 (2003); doi: 10.1063/1.1615682
View online: http://dx.doi.org/10.1063/1.1615682
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/83/14?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Thickness-dependent magnetic domain change in epitaxial MnAs films on GaAs(001)
Appl. Phys. Lett. 89, 232506 (2006); 10.1063/1.2402235
Effect of strain on the local phase transition temperature of MnAs/GaAs(001)
Appl. Phys. Lett. 83, 2829 (2003); 10.1063/1.1615836
Magnetic structure of epitaxially grown MnAs on GaAs(001)
J. Appl. Phys. 88, 2766 (2000); 10.1063/1.1287607
Magnetic moment of Mn in the ferromagnetic semiconductor (Ga 0.98 Mn 0.02 )As
Appl. Phys. Lett. 76, 2928 (2000); 10.1063/1.126519
Uniaxial magnetic anisotropy of submicron MnAs ferromagnets in GaAs semiconductors
Appl. Phys. Lett. 73, 387 (1998); 10.1063/1.121843
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.105.215.146 On: Thu, 18 Dec 2014 09:45:10

APPLIED PHYSICS LETTERS
VOLUME 83, NUMBER 14
6 OCTOBER 2003
Magnetic out-of-plane component in MnAsÕGaAs„001…
A. Ney,a) T. Hesjedal, C. Pampuch, J. Mohanty, A. K. Das, L. Daweritz, R. Koch,
¨
and K. H. Ploog
¨
¨
Paul-Drude-Institut fur Festkorperelektronik, Hausvogteiplatz 5-7, D-10117 Berlin, Germany
͑Received 23 April 2003; accepted 14 August 2003͒
From highly sensitive superconducting quantum interference device magnetometry and magnetic
force microscopy, we deduce a small out-of-plane magnetization component of MnAs/GaAs͑001͒
films. Its temperature dependence is substantially different from the dominating in-plane
magnetization, particularly as it is still detectable above the phase transition temperature of MnAs
films. Our measurements indicate that the out-of-plane component is due to small isolated magnetic
‘‘grains’’ within the film. © 2003 American Institute of Physics. ͓DOI: 10.1063/1.1615682͔
Ferromagnetic manganese arsenide films on GaAs are
promising candidates for future spin-injection devices.1
Nowadays high-quality epitaxial MnAs(1100) films are rou-
rate of 20 nm/h. The epitaxial orientation with respect
¯
to the substrate is MnAs(1100)
ʈ
GaAs(001) and
¯
¯
MnAs 0001 GaAs 110 , see Refs. 4 and 13. The easy axis
of magnetization is MnAs 1120 , i.e., the in-plane a axis,
ʈ
͓
͔
͓
͔
¯
tinely grown by molecular beam epitaxy ͑MBE͒ on the two
technologically leading semiconductor substrates, Si͑001͒
and GaAs͑001͒.2–5 In the bulk, MnAs loses its long-range
magnetic order abruptly at about 40 °C, when the ferromag-
netic ␣-phase transforms into the paramagnetic orthorhombic
␤-phase by a first-order transition.6,7 In thin films of
͓
͔
while the magnetically hardest direction is along the in-plane
MnAs͓0001͔ (c-axis͒.5,14 This leads to the unusual situation
that the out-of-plane anisotropy field (ϳ1.0 T) is smaller
than the in-plane one (ϳ2.0 T).5 The magnetic measure-
ments were carried out using a commercial SQUID magne-
tometer with the external field applied in the film plane and
out of plane. The temperature was varied from 0 °C to 60 °C
and backward to exclude thermal hysteresis effects. Each
hysteresis loop was measured at stabilized temperatures
(ϽϮ0.05 °C).
Figure 2 shows hysteresis loops of a 60 nm MnAs/
GaAs͑001͒ film at 0 °C and 50 °C recorded in plane along
the a axis ͓Fig. 2͑a͔͒ and out of plane ͓Fig. 2͑b͔͒. The in-
plane measurements reveal an almost perfectly squarelike
loop indicating that this direction is indeed the easy axis.
Since the remanence is not reduced, the formation of do-
mains can be excluded and the saturation value of the mag-
netization of 0.65Ϯ0.02 MA/m is in good agreement with
the magnetization of 0.67 MA/m reported for bulk ␣-MnAs
at 10 °C.15 Compared to 10 °C, the hysteresis at 50 °C is
strongly reduced; note, that the data are enlarged by a factor
of 50. Obviously, the system loses its long-range order and
appears to be completely transformed into the paramagnetic
¯
MnAs(1100) on GaAs͑001͒, on the other hand, the transition
from ␣- to ␤-phase does not proceed abruptly. Instead, there
is a broad temperature region from 10–40 °C, where both
phases coexist8 forming a self-organized stripe pattern of the
two phases.9,10 Figure 1͑a͒ illustrates the striped pattern at
15 °C using magnetic force microscopy ͑MFM͒. The ferro-
magnetic ␣-phase can be identified by the zigzag-shaped
magnetic contrast due to its domain pattern in the demagne-
tized state. At 41 °C, the striped pattern has disappeared
͓Fig. 1͑b͔͒, however, some residual magnetic contrast is lo-
cally detectable in the ␤-phase ͑marked by arrows͒. A de-
tailed temperature dependent MFM analysis revealed that
these dots serve as pinning centers for the formation of the
ferromagnetic ␣-phase upon cooling.11
To study the origin of the magnetic contrast in more
detail, we performed highly sensitive measurements by a su-
perconducting quantum interference device ͑SQUID͒ magne-
tometer of the magnetic properties of MnAs/GaAs͑001͒ in-
plane along the a axis and out of plane. We recorded
magnetic hysteresis loops evolving as a function of the tem-
perature across the phase coexistence range. From the in-
plane measurements, we can confirm the loss of long-range
order at about 50 °C whereas the out-of-plane measurements
reveal a small hysteresis loop which is present in the tem-
perature range of 0–55 °C, thus indicating that a small frac-
tion of the MnAs film exhibits a finite out-of-plane compo-
nent of the magnetization.
As substrates, commercial epiready GaAs͑001͒ wafers
were coated by standard solid-source MBE ͑Ref. 12͒ with a
100-nm-thick GaAs͑001͒ buffer layer after oxide removal.
MnAs was deposited at a substrate temperature of 250 °C, at
a As4 /Mn beam equivalent pressure ratio of 250, and at a
FIG. 1. MFM images of a 180 nm MnAs/GaAs͑001͒ film ͑see Ref. 13͒ at ͑a͒
15 °C and ͑b͒ 41 °C. ͑a͒ shows the coexistence of the ferromagnetic ␣- and
the paramagnetic ␤-phase; the ␣-phase is demagnetized thus showing a
zigzag-shaped domain pattern of the ␣-stripes. In image ͑b͒, some residual
contrast of small isolated dots is visible ͑arrows͒.
a͒
Electronic mail: ney@pdi-berlin.de
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
0003-6951/2003/83(14)/2850/3/$20.00 2850 © 2003 American Institute of Physics
129.105.215.146 On: Thu, 18 Dec 2014 09:45:10

Appl. Phys. Lett., Vol. 83, No. 14, 6 October 2003
Ney et al.
2851
accordance with Ref. 5, we find hard-axis behavior with an
anisotropy field of 1 T, containing both the shape and the
crystalline contribution. The demagnetizing field ͑shape an-
isotropy͒ can be evaluated from the saturation magnetization
yielding ϳ0.5 T, the out-of-plane crystalline anisotropy field
is, therefore, ϳ0.5 T. The upper inset of Fig. 2 shows a
magnification of the low-field region at 0 °C exhibiting both
clear remanence and coercivity, thus revealing a small out-
of-plane component of the magnetization. We want to point
out that the loop is not squarelike indicating either that the
out-of-plane direction is not the easy axis of the detected
component or it originates from weakly coupled ‘‘grains’’
similar to the behavior of polycrystalline materials. Assum-
ing that the out-of-plane contribution is due to regions which
possess the same magnetization as the MnAs film, we derive
a volume fraction of 2%–3% from the SQUID measurement
͑which essentially measures the product of magnetization
and volume͒. We remark, that a small hysteresis can be rec-
ognized
in
previous
out-of-plane
magnetic
measurements,3,5,16 however, it was not addressed in these
studies. To corrobate our findings, we performed analogous
measurements at higher temperatures, e.g., at 50 °C shown in
the lower inset of Fig. 2͑b͒. Surprisingly, nearly all of the
same hysteretic behavior as at 0 °C with essentially the same
remanence and coercivity is found. This observation—
contrary to older work, where no temperature dependent
measurements have been performed out of plane—excludes
that the signal at 0 °C is due to the misalignment of the
sample with respect to the external field. In that case, the
out-of-plane signal would exhibit the same temperature de-
pendence as the in-plane component and should be drasti-
cally reduced compared to 0 °C. Therefore, we can also con-
clude that the out-of-plane signal is due to a magnetization
aligned along or close to the surface normal. For similar
reasons, the out-of-plane component can not be attributed to
domain-wall formation or tilted magnetization contributions
at furrows as discussed in Ref. 5.
FIG. 2. Magnetic hysteresis measurements of a 60 nm MnAs/GaAs͑001͒
film at 0 °C and 50 °C. The external field is applied ͑a͒ in the film plane
along the a axis and ͑b͒ out-of-plane. Note, that the in-plane hysteresis at
50 °C is normalized to the saturation magnetization at Tϭ0 °C (MS
ϭ0.65 MA/m) and magnified by a factor of 50. The two insets of ͑b͒ en-
large the low-field range revealing clear out-of-plane hysteretic behavior.
␤-phase. The whole temperature dependence of the in-plane
remanent magnetization throughout the phase coexistence
range from 0 °C to 60 °C is summarized in Fig. 3 ͑squares͒.
Figure 2͑b͒ shows a hysteresis loop of the same sample
traced along the out-of-plane direction from Ϫ4 to ϩ4 T.
The diamagnetic contribution of the substrate and the sample
holder derived from the high-field data is subtracted, the
sample holder alone showed pure diamagnetic behavior. In
Figure 3 summarizes our SQUID measurements showing
the in-plane ͑squares͒ and out-of-plane ͑circles͒ component
of the remanent magnetization of 60 nm MnAs/GaAs͑001͒.
Whereas the in-plane component—which by far dominates at
lower temperatures—gradually decreases until it vanishes
around 50 °C, the out-of-plane component shows only a
weak temperature dependence and is present even when the
in-plane magnetization has already disappeared. The exis-
tence of an out-of-plane component of the magnetization has
several implications. ͑i͒ The out-of-plane component does
not follow the temperature dependence of the in-plane mag-
netization, therefore, it is decoupled from the structural and
magnetic phase transistion. This points toward isolated re-
gions within the film which are consistent with the MFM
findings in Fig. 1͑b͒. ͑ii͒ For being magnetized out-of-plane,
these regions must have considerable crystalline anisotropy,
otherwise, the demagnetizing field and the coupling to the
remainder of the film would push their magnetization back to
the film plane at lower temperatures. ͑iii͒ Since we detect an
out-of-plane ferromagnetic signal above 55 °C ͑see inset of
Fig. 3͒, which is significantly above the ordering temperature
of bulk MnAs (40 °C), a structural configuration with a
FIG. 3. Remanent magnetization, in and out of plane, derived from hyster-
esis measurements in the temperature range of 0–60 °C. The inset shows
that the out-of-plane component is nonzero throughout the whole tempera-
ture range and it decreases only slightly while the in-plane component van-
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
ishes.
higher transition temperature compared to the bulk has to be
129.105.215.146 On: Thu, 18 Dec 2014 09:45:10

2852
Appl. Phys. Lett., Vol. 83, No. 14, 6 October 2003
Ney et al.
¨
present. This finding may turn out to be important for tech-
nical applications because, for room-temperature devices, the
bulk transition temperature is quite low. ͑iv͒ Careful hard-
axis magnetic measurements can serve as a highly sensitive
probe for testing the structural homogeneity of MnAs films.
To unambiguously clarify the structural origin of the out-
of-plane magnetiztion component further, careful investiga-
tions are necessary. From our magnetic measurements, we
can only derive indirect conclusions. The finite crystalline
anisotropy of the observed component requires crystallinity
of these regions. Their easy axis should be close to out of
plane and the rounded shape of the out-of-plane loop sug-
gests that these regions are well separated and weakly
coupled. The out-of-plane magnetization seems to be linked
to the granular magnetic contrast recorded by MFM above
the transition temperature. Recently, Iikawa et al.17 observed
a MnAs component in MnAs/GaAs͑001͒ which is rotated by
30° along the c axis, thus exhibiting an a axis perpendicular
to the substrate plane. It is not clear, however, why this struc-
turally equivalent MnAs fraction should have a higher tran-
sition temperature. We therefore actually favor Mn-rich in-
clusions with differing magnetic properties as a possible
explanation, which were found recently at the MnAs/GaAs
interface.18
The authors thank C. Herrmann and M. Kastner for the
sample preparation, and J. Herfort for technical assistance.
1
¨
M. Ramsteiner, H. Y. Hao, A. Kawaharazuka, H. J. Zhu, M. Kastner, R.
Hey, L. Daweritz, H. T. Grahn, and K. H. Ploog, Phys. Rev. B 66,
¨
081304͑R͒ ͑2002͒.
2 M. Tanaka, J. P. Harbison, M. C. Park, Y. S. Park, T. Shin, and G. M.
Rothberg, J. Appl. Phys. 76, 6278 ͑1994͒.
3 K. Akeura, M. Tanaka, M. Ueki, and T. Nishinaga, Appl. Phys. Lett. 67,
3349 ͑1995͒.
4 M. Tanaka, Physica E ͑Amsterdam͒ 2, 372 ͑1998͒.
5
¨
F. Schippan, G. Behme, L. Daweritz, K. H. Ploog, B. Dennis, K.-U. Neu-
mann, and K. R. A. Ziebeck, J. Appl. Phys. 88, 2766 ͑2000͒.
6 C. P. Bean and D. S. Rodbell, Phys. Rev. 126, 104 ͑1962͒.
7 R. H. Wilson and J. S. Kasper, Acta Crystallogr. 17, 95 ͑1964͒.
8
¨
V. M. Kaganer, B. Jenichen, F. Schippan, W. Braun, L. Daweritz, and K.
H. Ploog, Phys. Rev. Lett. 85, 341 ͑2000͒.
9
10
11
12
13
¨
T. Plake, M. Ramsteiner, V. M. Kaganer, B. Jenichen, M. Kastner, L.
Daweritz, and K. H. Ploog, Appl. Phys. Lett. 80, 2523 ͑2002͒.
V. M. Kaganer, B. Jenichen, F. Schippan, W. Braun, L. Daweritz, and K.
¨
¨
H. Ploog, Phys. Rev. B 66, 045305 ͑2002͒.
¨
¨
T. Plake, T. Hesjedal, J. Mohanty, M. Kastner, L. Daweritz, and K. H.
Ploog, Appl. Phys. Lett. 82, 2308 ͑2003͒.
¨
¨
M. Kastner, C. Herrmann, L. Daweritz, and K. H. Ploog, J. Appl. Phys.
92, 5711 ͑2002͒.
¨
F. Schippan, A. Trampert, L. Daweritz, and K. H. Ploog, J. Vac. Sci.
In conclusion, we presented experimental evidence for
an out-of-plane component of the magnetization of MnAs
films on GaAs͑001͒ by temperature dependent SQUID mea-
surements. This component seems to be linked with small
dots of residual magnetic contrast in MFM images at el-
evated temperatures consistent with the observation of a
rounded hysteresis loop. Magnetic separation of these grains
is supported by their different temperature dependent mag-
netic behavior with respect to the dominating in-plane mag-
netization of MnAs/GaAs͑001͒.
Technol. B 17, 1716 ͑1999͒.
14 M. Tanaka, J. P. Harbison, M. C. Park, Y. S. Park, T. Shin, and G. M.
Rothberg, Appl. Phys. Lett. 65, 1964 ͑1994͒.
15 N. Menyuk, J. A. Kafalas, K. Dwight, and J. B. Goodenough, Phys. Rev.
177, 942 ͑1969͒.
16 A. M. Grishin, S. I. Khartsev, and K. V. Rao, Appl. Phys. Lett. 68, 2008
͑1996͒.
17 F. Iikawa, M. J. S. Brasil, O. D. O. Couto, C. Adriano, C. Giles, R.
¨
Magalhas-Paniago, P. V. Santos, and L. Daweritz ͑unpublished͒.
18
¨
¨
L. Daweritz, M. Kastner, T. Hesjedal, T. Plake, B. Jenichen, and K. H.
Ploog, International Conference on Molecular Beam Epitaxy, San Fran-
cisco ͑IEEE, New York, 2002͒, p. 413.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
129.105.215.146 On: Thu, 18 Dec 2014 09:45:10