Nanotechnology
Nanotechnology 25 (2014) 505702 (7pp)
doi:10.1088/0957-4484/25/50/505702
Superlattice of FexGe1−x nanodots and
nanolayers for spintronics application
Tianxiao Nie1, Xufeng Kou1, Jianshi Tang1, Yabin Fan1, Murong Lang1,
Li-Te Chang1, Chia-Pu Chu1, Liang He1, Sheng-Wei Lee2, Faxian Xiu3,
Jin Zou4 and Kang L Wang1
1
Device Research Laboratory, Department of Electrical Engineering, University of California, Los
Angeles, California, USA, 90095
2
Institute of Materials Science and Engineering, National Central University, Taiwan
3
State Key Laboratory of Surface Physics, Fudan University, Shanghai 200433, People’s Republic of
China
4
Materials Engineering and Centre for Microscopy and Microanalysis, The University of Queensland,
Brisbane QLD 4072, Australia
E-mail: nietianxiao@gmail.com and wang@seas.ucla.edu
Received 14 June 2014, revised 28 September 2014
Accepted for publication 27 October 2014
Published 25 November 2014
Abstract
FexGe1−x superlattices with two types of nanostructures, i.e. nanodots and nanolayers, were
successfully fabricated using low-temperature molecular beam epitaxy. Transmission electron
microscopy (TEM) characterization clearly shows that both the FexGe1−x nanodots and
nanolayers exhibit a lattice-coherent structure with the surrounding Ge matrix without any
metallic precipitations or secondary phases. The magnetic measurement reveals the nature of
superparamagnetism in FexGe1−x nanodots, while showing the absence of superparamagnetism
in FexGe1−x nanolayers. Magnetotransport measurements show distinct magnetoresistance (MR)
behavior, i.e. a negative to positive MR transition in FexGe1−x nanodots and only positive MR in
nanolayers, which could be due to a competition between the orbital MR and spin-dependent
scatterings. Our results open a new growth strategy for engineering FexGe1−x nanostructures to
facilitate the development of Ge-based spintronics and magnetoelectronics devices.
S Online supplementary data available from stacks.iop.org/NANO/25/505702/mmedia
Keywords: FexGe1−x superlattice, nanodot, nanolay, molecular beam epitaxy, electron
microscopy, magnetoresistance
(Some figures may appear in colour only in the online journal)
devices to resolve the power consumption and variability
issues in today’s microelectronics industry [4, 5]. To date,
attempts within various semiconductor materials, such as Si
[6], Ge [7, 8], GaAs [9] and ZnO [10], have been widely
undertaken to produce DMSs with a high Curie temperature
and good crystallinity. Among them, Si-/Ge-based DMS
systems are highly favored for integrated spintronics because
of their high compatibility with the present and mature Si
technology.
Since the first report of MnxGe1−x thin film with electricfield controlled ferromagnetism, great efforts have been
devoted to such material systems [11]. One direct and
1. Introduction
Recently, diluted magnetic semiconductors (DMSs) have
attracted a great deal of attention, since they could greatly
expand the functionalities of ordinary semiconductors by
combining not only the electrical charge for information
processing but also the spin of electrons for information
storage [1–3]. In addition, compared with ordinary ferromagnetic metal-based spintronics devices, the advantage of
electric-field controlled ferromagnetism in DMSs can be
taken to produce more functional devices, and it has the
potential for developing a new generation of electronic
0957-4484/14/505702+07$33.00
1
© 2014 IOP Publishing Ltd Printed in the UK
Nanotechnology 25 (2014) 505702
T Nie et al
effective way to realize/implement a room-temperature
operation on a DMS system with a higher Curie temperature
is to increase the magnetic dopant concentration in the Ge
matrix. Unfortunately, Mn has a strong tendency to aggregate
in Ge to form metallic precipitates and secondary phases [12],
which prohibited the improvement of the Curie temperature
and limited its comprehensive application in spintronics. In
contrast, another transition metal Fe is a much more promising dopant because of its weaker tendency to form clusters, and it has much stronger ferromagnetism compared to
that of Mn [13]. Despite its pronounced advantages, the
nature of low solubility of Fe in Ge still makes it challenging
to form pure DMS materials without iron germanium
compounds.
While continuous efforts to address the fundamental
issue of the Fe incorporation limitation has been made, realizing the functionality of spin injection into traditional
semiconductors also requires a lattice-coherent Fe-doped
DMS for less interface scattering and for a small conductivity
mismatch [14]. To this end, the hybrid structure with a layer
of a nonmagnetic semiconductor and a ferromagnetic semiconductor is needed to realize spin injection. Furthermore, the
precise engineering of a lattice-coherent FexGe1−x structure
provides a novel way to design new nanodevices such as
nanodot memories and nanochannels for spin injection
[15, 16]. However, it is still a great challenge to create and
control such a hybrid system, particularly considering the
random nucleation of the FexGe1−x nanostructures.
In this paper, a superlattice with alternative FexGe1−x and
Ge layers was successfully grown by employing a low-temperature molecular beam epitaxy (MBE). A FexGe1−x nanodot
and nanolayer structure could be deliberately designed with
excellent reproducibility by tuning the Fe concentration. The
magnetic measurement excluded the presence of FexGe1−x
intermetallic compounds in the formed superlattice structure.
An apparent distinction in magnetotransport properties can be
obtained by engineering the nanostructures, providing an
extraordinary material candidate for building future spintronics devices.
FexGe1−x and Ge layer, ten periods of the Ge/FexGe1−x
superlattice were grown on the Ge substrate. By tuning the Fe
concentration, FexGe1−x nanodots and a FexGe1−x nanolayer
superlattice were successfully achieved with good
reproducibility.
The microstructure and composition of the grown
FexGe1−x/Ge superlattice were investigated by TEM (Philips
Tecnai F20) equipped with energy-dispersive spectroscopy
(EDS). Their magnetic property was measured by the superconducting quantum interference device (SQUID). Furthermore, the temperature-dependent magnetotransport properties
were studied by the physical property measurement system (PPMS).
3. Results and discussions
Figure 1(a) is a schematic illustration of the designed FexGe1−x
superlattice structure in which ten periods of FexGe1−x
layers are embedded in the Ge matrix. Figures 1(b) and (c) are
typical cross-section TEM images with low (∼1%) and high
(∼3%) Fe doping, respectively, in which ten periods of dark
layers can be seen. From their insets, the FexGe1−x layers can
be found in the form of nanodots (for the case of low Fe
doping) and in the form of nanolayers (for the case of high Fe
doping). To confirm their composition, EDS was employed,
and an example is shown in figure 1(d) in which the Fe and
Ge peaks are clearly seen. To reveal the detailed lattice
structure of the nanodots and nanolayers, high-resolution
TEM (HRTEM) was employed. Figures 1(e) and (f) are typical
HRTEM images taken from both samples from which a
coherent lattice of the FexGe1−x nanodots and nanolayers with
their Ge matrix can be seen. In the case of low Fe doping, the
size of the nanodots shows a certain diameter distribution in the
range of 2.5−4 nm with the dominant size of ∼3 nm, according
to our extensive investigation. When the Fe doping concentration is higher, the nanodots are changed to nanolayers
with a thickness of ∼3 nm, as marked by the dotted red lines in
figure 1(f). To understand the Fe concentration-dependent
structure evolution, we note that Fe in Ge has a certain tendency to agglomerate to form nanoparticles [17]. In our sample
with a low Fe doping concentration, Fe and Ge co-evaporated
on the Ge surface at the initial stage. The incoming Fe atoms
preferred to nucleate on the Ge surface, and they acted as the
energy-preferred sites to attract the new coming Fe atoms.
Consequently, a structure of Fe-rich nanodots with a Fe-poor
adjacent Ge matrix would be formed. As to the sample with the
higher Fe doping concentration, the density of the nucleation
would become larger, and Fe-rich nanodots would become
bigger and bigger because much more Fe dopants were supplied. Considering the distance between the Fe nanodots from
the TEM characterization, ∼3 times the Fe doping concentration should be large enough to connect all of the Fe nanodots
together, which consequently leads to a relatively uniform Fe
doping concentration.
Magnetic properties of the two types of FexGe1−x
nanostructures were measured by a SQUID magnetometer.
Temperature-dependent M-H hysteresis loops of the FexGe1−x
2. Experimental details
The growth of the FexGe1−x superlattice was carried out in an
ultrahigh vacuum PerkinElmer solid-source MBE system.
High-purity Ge and Fe sources were evaporated from the
Knudsen effusion cells. Before loading into the growth
chamber, the Ge (100) wafer was carefully cleaned by rinsing
in acetone, in isopropyl alcohol with ultrasonic agitation and
finally in diluted hydrofluoric acid (HF). After degassing the
substrate at 600 °C for 30 min, the substrate was cooled to
250 °C for Ge buffer layer growth with a thickness of about
50 nm. Subsequently, the growth temperature was lowered to
70 °C for the superlattice growth. The first FexGe1−x layer was
grown with a Ge growth rate of 0.2 Å s−1 and an adjustable Fe
flux as the dopant source. The nominal deposition thickness
of this layer was ∼3 nm. Subsequently, a ∼10 nm thick Ge
space layer was deposited. By alternating the growth of the
2
Nanotechnology 25 (2014) 505702
T Nie et al
Figure 1. (a) Schematic illustration of the ten periods of the FexGe1−x layer. (b)–(c) Low-resolution TEM images of the FexGe1−x nanodots
and nanolayers, respectively, clearly showing their ten layers of nanostructures. (d) EDS spectrum of the FexGe1−x sample, confirming
the existence of Fe and Ge. (e)–(f) HRTEM images of the FexGe1−x nanodots and nanolayers, respectively, showing the lattice-coherent
FexGe1−x layer.
nanodots sample are shown in figure 2(a), where the in-plane
magnetic field is applied parallel to the superlattice film. At
10 K, it shows a very small coercivity field of 10 Oe, as
displayed in the inset of figure 2(a). As the temperature
increases, the coercivity field decreases. The linear M-H curve
at 200 K implies that the FexGe1−x nanodots have become
paramagnetic [18], indicating a Curie temperature (Tc) below
200 K. Meanwhile, it could be estimated that the saturated
magnetic moment per Fe atom was 0.35 μB at 10 K. At 200 K,
the near-zero magnetic moment indicates that the paramagnetic FexGe1−x nanodots have a small susceptibility. In
comparison, FexGe1−x nanolayers show a similar magnetic
behavior as that of the FexGe1−x nanodots, but the saturated
magnetic moment per atom is a little larger, which suggests
that the exchange coupling between the Fe ions is enhanced.
It is well known that Fe in Ge can generate deep impurity
acceptor levels, which will increase the hole density [19].
Meanwhile, the increased Fe doping concentration will
decrease the distance between the Fe ions. Following the
Ruderman–Kittel–Kasuya–Yoshida (RKKY) mechanism
[20], the increased Fe doping concentration induced the
increase of the hole density, and the decreasing Fe ion distance should be responsible for the magnetic moment
enhancement.
To quantitatively understand the prevailing magnetism of
the FexGe1−x superlattice samples, detailed zero-field-cooled
(ZFC) and field-cooled (FC) magnetization measurements
were performed. The ZFC curve was obtained by cooling the
sample under a zero magnetic field and subsequently measuring the magnetic moment while warming up the sample
3
Nanotechnology 25 (2014) 505702
T Nie et al
Figure 2. (a) Hysteresis loops measured on the FexGe1−x nanodot sample at different temperatures from 10–200 K. The inset is the magnified
M-H loops. (b) The hysteresis loops measured from the FexGe1−x nanolayer sample. The corresponding magnified M-H loops are shown in
the inset. (c) ZFC and FC curves of the FexGe1−x nanodot, showing a Tb at about 12 K. (d) ZFC and FC curves from the FexGe1−x nanolayer,
indicating no existence of superparamagnetic particles.
under a field of 50 Oe. For the FC curve, however, the sample
was cooled in the presence of a magnetic field (50 Oe). The
difference between the two processes gives insight into
detecting superparamagnetic nanostructures, estimating the
blocking temperature (Tb) and probing the anisotropy barrier
distribution [21]. As shown in figure 2(c), the FexGe1−x
nanodots display a bifurcation between the ZFC and FC
curves with a Tb at ∼12 K, which means that there exists one
type of superparamagnetic nanostructure attributed exclusively to FexGe1−x nanodots. Below 12 K, the anisotropy
energy barrier would prefer the magnetic moments to be
aligned along the easy axis. The relatively sharp peak of the
Tb indicates that the FexGe1−x nanodots have a narrow size
distribution, which agrees well with our TEM observation
showing a narrow size distribution of 2.5–4 nm. Above 12 K,
FexGe1−x nanodots became superparamagnetic. When we
conducted the experiment, the thermal energy became larger
than the anisotropy energy barrier of the FexGe1−x nanodots,
of which the magnetization direction can randomly flip during
the measurement time. In comparison, the ZFC and FC curves
in FexGe1−x nanolayers are well superimposed in the absence
of Tb, as shown in figure 2(d). To understand the substantial
differences in the ZFC and FC curves of these two nanostructures, the equation to describe the Tb as a function of the
particle volume is given by [22]
Tb =
KV
ln ( τSQUID τ0 ) k B
=
KV
,
25k B
where K is the total magnetic-anisotropy energy per unit
volume, τSQUID is the characteristic measure time of the
SQUID (∼10 s) and τ0 is the natural period of the gyromagnetic precession (∼10−9–10−13 s) [18]. Obviously, the Tb is
proportional to the particle volume. In our FexGe1−x nanodots
sample, only one Tb exists in the ZFC and FC curve, indicating no other metallic precipitants or secondary phases are
formed except the TEM-observed FexGe1−x coherent nanodots. In comparison, the superimposed ZFC and FC curve
rules out the presence of superparamagnetic nanostructures in
the FexGe1−x nanolayer sample, which further confirms that
Fe ions uniformly distribute in the FexGe1−x layer without the
formation of any FeGe alloy precipitates.
Since two types of FexGe1−x nanostructures with distinct
magnetic properties are formed by simply engineering the Fe
concentrations, it is of great interest to investigate their
4
Nanotechnology 25 (2014) 505702
T Nie et al
Figure 3. (a) Magnetotransport measurement of the FexGe1−x nanodot sample at different temperatures from 2–300 K, showing a transition
from negative MR to positive MR. (b) Magnetotransport measurement of the FexGe1−x nanolayer sample only showing a positive MR. The
inset is the curves of MR versus (μ0H)2.
magnetotransport properties. The samples were fabricated
into standard Hall bars with a typical channel width of
500 μm. For transport measurements, the magnetic field was
applied perpendicular to the surface of the samples with
current flowing in the plane. The measurement was performed
at temperatures from 2 K to 300 K with an external magnetic
field of up to 5 T. Figure 3(a) shows the temperature-dependent magnetoresistance (MR) of the FexGe1−x nanodots,
where an obvious transition from negative to positive MR
could be observed when the temperature increases from 10 K
to 15 K. Taking the ZFC and FC curve into account, the MR
transition in the FexGe1−x nanodots is very plausibly related to
the Tb. Previous theoretical studies [23–25] have indicated
that the negative MR in granular films comes from the spindependent scattering of the conduction electrons transported
in magnetic clusters and at the interfaces between the magnetic grains and the nonmagnetic matrix. Below 12 K,
because of the weakly localized magnetic interaction between
the nanodots, all of these FexGe1−x nanodots will reach a state
called the spin-glass state [18]. In this state, the magnetic
moments between different nanodots are analogous with a
real glass or amorphous solid, where atoms are randomly
distributed without any ordering. Both the spin-up and spindown electrons will suffer from the spin-dependent scattering
from the randomly distributed magnetic moments between
different nanodots and thus give rise to a much larger scattering to the conduction electrons in the absence of a magnetic
field. Therefore, an applied magnetic field will well align the
magnetic moments of FexGe1−x nanodots to reduce the spindependent scattering of the electrons, resulting in a negative
MR Meanwhile, it is of great interest to point out that the
negative MR will change to a positive value at a high magnetic field. To understand this phenomenon, we should
mention that another competitive mechanism, namely orbital
MR, always coexists in the magnetotransport process; such a
type of MR is approximately proportional to (μB)2, hence
inducing a positive MR in which μ is the carrier mobility [26].
At such a high magnetic field, FexGe1−x nanodots are already
magnetically polarized and therefore reduce the sensitivity of
the spin-dependent scattering. Under such a circumstance, the
orbital MR component is expected to dominate the conduction behavior, and MR curves display the nearly parabolic
behavior, as highlighted in the inset of figure 3(a).
As the temperature increases above 12 K, the magnetic
moments in each ferromagnetic FexGe1−x nanodot will overcome the anisotropy energy barrier and flip randomly in two
energy-favored directions as a result of thermal fluctuations.
Correspondingly, the FexGe1−x nanodots will change from
ferromagnetic to superparamagnetic. The effect of the magnetic field on the ordering of FexGe1−x nanodots in superparamagnetism is not as strong as that in ferromagnetism [27].
The weakened negative MR will lose the dominant role in the
magnetotransport property. Therefore, only a positive MR
behavior is observed, and the carriers mainly undergo a
scattering characteristic of orbital MR.
In contrast, the FexGe1−x nanolayer only shows a positive
MR in the whole temperature range, suggesting that orbital
MR is the dominant mechanism in this sample, as shown in
figure 3(b). We should point out that the different magnetotransport behaviors unveiled in figures 3(a) and (b) mainly
reflect their nanostructures deviation. Fe dopants distributed
uniformly in the FexGe1−x nanolayer and the relatively wellaligned moments of the Fe atoms due to hole-mediated
RKKY interaction give rise to the smaller spin-dependent
scattering of the carrier in the absence of a magnetic field. A
weaker effect could be expected on the spin-dependent scattering after the magnetic field applied, therefore generating
negligible negative MR. Only the MR contribution from the
orbital MR is dominant, and a positive MR is demonstrated.
Additionally, we noticed that at a low temperature, the MR
deviated from the simple parabolic dependence on the applied
magnetic field. The MR response shows a marked decrease in
field sensitivity at a field comparable to the saturation magnetization of the sample. This effect might be attributed to
ferromagnetic ordering of Fe ions in the Ge matrix, where the
effect on the orbital MR comes from not only the external
magnetic field but also from Fe ion magnetic moments, which
have a strong response to the external magnetic field until
5
Nanotechnology 25 (2014) 505702
T Nie et al
saturated. In the high-temperature region, the magnetic
ordering of the matrix is randomized, so we expect and indeed
observe an enhancement of MR to dissipate following a true
parabolic dependence on the external applied field. To further
clearly understand this phenomenon, the curves of (μ0H)2
versus MR are plotted in the insets of figure 3(b). At a low
temperature, the dependence of MR on (μ0H)2 exhibits two
separate contributions: a linear dependence at a high field
corresponding to the orbital MR and a nonlinear dependence
at a low field corresponding to magnetization-enhanced MR,
which disappears at a high temperature as well. All of the
results further confirm our speculation about the transport
mechanism.
[5] Tang J et al 2013 Electrical spin injection and detection in
Mn5Ge3/Ge/Mn5Ge3 nanowire transistors Nano Lett. 13
4036–43
[6] Dietl T, Ohno H, Matsukura F, Cibert J and Ferrand D 2000
Zener model description of ferromagnetism in zinc-blende
magnetic semiconductors Science 287 1019–22
[7] Wang K L, Zhao Z and Khitun A 2008 Spintronics for
nanoelectronics and nanosystems Thin Solid Films 517
184–90
[8] Wang Y, Xiu F, Wang Y, Kou X, Jacob A P, Wang K L and
Zou J 2010 Mn-rich clusters in GeMn magnetic
semiconductors: structural evolution and magnetic property
J. Alloy. Compd. 508 273–7
[9] Ohno H 1998 Making nonmagnetic semiconductors
ferromagnetic Science 281 951–6
[10] Ueda K, Tabata H and Kawai T 2001 Magnetic and electric
properties of transition-metal-doped ZnO films Appl. Phys.
Lett. 79 988–90
[11] Park Y D, Hanbicki A T, Erwin S C, Hellberg C S,
Sullivan J M, Mattson J E, Ambrose T F, Wilson A,
Spanos G and Jonker B T 2002 A group-IV ferromagnetic
semiconductor: MnxGe1−x Science 295 651–4
[12] Cho S, Choi S, Hong S C, Kim Y, Ketterson J B, Kim B-J,
Kim Y C and Jung J-H 2002 Ferromagnetism in Mn-doped
Ge Phys. Rev. B 66 033303
[13] Shuto Y, Tanaka M and Sugahara S 2008 Epitaxial growth and
magnetic properties of ferromagnetic semiconductor Ge1
−xFex thin films epitaxially grown on Si (001) substrates Jpn.
J. Appl. Phys. 47 7108–12
[14] Kioseoglou G, Hanbicki A T, Sullivan J M, van’t Erve O M,
Li C H, Erwin S C, Mallory R, Yasar M, Petrou A and
Jonker B T 2004 Electrical spin injection from an n-type
ferromagnetic semiconductor into a III–V device
heterostructure Nat. Mater. 3 799–803
[15] Tang J et al 2010 Single-crystalline Ni 2Ge/Ge/Ni 2Ge
nanowire heterostructure transistors Nanotechnology 21
505704
[16] Tang J et al 2012 Ferromagnetic germanide in Ge nanowire
transistors for spintronics application Acs Nano 6 5710–7
[17] Goswami R, Kioseoglou G, Hanbicki A, van t Erve O,
Jonker B and Spanos G 2005 Growth of ferromagnetic
nanoparticles in Ge: Fe thin films Appl. Phys. Lett. 86
032509–0325011
[18] Venugopal R, Sundaravel B, Cheung W Y, Wilson I H,
Wang F W and Zhang X X 2001 Magnetic properties of
nanoclusters formed by implantation of Fe into Ge using a
metal-vapor vacuum arc ion source Phys. Rev. B 65
014418
[19] Sze S M and Irvin J C 1968 Resistivity, mobility and impurity
levels in GaAs, Ge, and Si at 300 °K Solid-State Electron.
11 599–602
[20] Gareev R R, Bugoslavsky Y V, Schreiber R, Paul A,
Sperl M and Döppe M 2006 Carrier-induced ferromagnetism
in Ge(Mn,Fe) magnetic semiconductor thin-film structures
Appl. Phys. Lett. 88 222508
[21] Devillers T, Jamet M, Barski A, Poydenot V,
Bayle-Guillemaud P, Bellet-Amalric E, Cherifi S and
Cibert J 2007 Structure and magnetism of self-organized
Ge_{1 − x}Mn_{x} nanocolumns on Ge(001) Phys. Rev. B
76 205306
[22] Babonneau D, Briatico J, Petroff F, Cabioc’h T and Naudon A
2000 Structural and magnetic properties of Fex–C1−x
nanocomposite thin films J. Appl. Phys. 87 3432–43
[23] Yang R, Zhang W and Song W J 1998 Theory of giant
magnetoresistance of magnetic granular films J. Appl. Phys.
84 2044–7
[24] Ferrari E F, da Silva F C S and Knobel M 1999 Theory of giant
magnetoresistance in granular alloys Phys. Rev. B 59
8412–5
4. Conclusions
FexGe1−x nanodots and nanolayers were successfully fabricated by engineering the Fe doping concentration in
low-temperature MBE. A TEM observation finds that both
FexGe1−x nanodots and nanolayers have a lattice-coherent
structure with the surrounding Ge matrix. The magnetic
property measurements reveal FexGe1−x nanodots have a Tb at
about 12 K and a Curie temperature below 200 K, while
FexGe1−x nanolayers have no superparamagnetic particles.
The MR measurement shows that the magnetotransport
property could be engineered through manipulating the
FexGe1−x nanostructures. The understanding and engineering
of FexGe1−x nanostructures via the superlattice approach
provide an important platform for the design of future spintronics and magnetoelectronics devices.
Acknowledgments
This work was supported in part by the FAME Center, one of
six centers of STARnet, a Semiconductor Research Corporation program sponsored by MARCO and DARPA. We
also gratefully acknowledge the financial support from the
National Science Foundation through grant ECCS 1308358.
References
[1] Xiu F, Wang Y, Kim J, Hong A, Tang J, Jacob A P, Zou J and
Wang K L 2010 Electric-field-controlled ferromagnetism in
high-Curie-temperature Mn 0.05Ge 0.95 quantum dots Nat.
Mater. 9 337–44
[2] Jamet M et al 2006 High-Curie-temperature ferromagnetism in
self-organized Ge1−xMnx nanocolumns Nat. Mater. 5 653–9
[3] Xiu F, Wang Y, Kou X, Upadhyaya P, Zhou Y, Zou J and
Wang K L 2010 Synthesis of high-Curie-temperature
Fe0.02Ge0.98 quantum dots J. Am. Chem. Soc. 132
11425–7
[4] Wolf S A, Awschalom D D, Buhrman R A, Daughton J M,
von Molnár S, Roukes M L, Chtchelkanova A Y and
Treger D M 2001 Spintronics: a spin-based electronics
vision for the future Science 294 1488–95
6
Nanotechnology 25 (2014) 505702
T Nie et al
ferromagnetic nanoclusters in a diluted magnetic
semiconductor matrix Appl. Phys. Lett. 78 2739–41
[27] Hickey B, Howson M, Musa S and Wiser N 1995 Giant
magnetoresistance for superparamagnetic particles: meltspun granular CuCo Phys. Rev. B 51 667–9
[25] Yu R H, Zhang X X, Tejada J and Zhu J 1995 Asymmetric
giant magnetoresistance in Co_{10}Cu_{90} magnetic
granular alloys Phys. Rev. B 52 R6987–90
[26] Park Y D, Wilson A, Hanbicki A T, Mattson J E, Ambrose T,
Spanos G and Jonker B T 2001 Magnetoresistance of Mn:Ge
7