- JSAP Journals

Proc. Int. Conf. and Summer School on Advanced Silicide Technology 2014
JJAP Conf. Proc. 3 (2015) 011504
©2015 The Japan Society of Applied Physics
Spin Valve behavior in Current-Perpendicular-to-Plane
Crossover Structural Fe3Si/FeSi2/Fe3Si Trilayered junctions
Yuki Asai1, Ken-ichiro Sakai1,2*, Kazuya Ishibashi1, Kaoru Takeda3, and
Tsuyoshi Yoshitake1**
1Department
of Applied Science for Electronics and Materials, Kyushu University, Kasuga,
Fukuoka 816-8580, Japan
2Department of Control and Information Systems Engineering, Kurume National College
of Technology, Kurume, Fukuoka 830-8555, Japan
3Department of Electrical Engineering, Fukuoka Institute of Technology, Fukuoka 8110295, Japan
E-mail: *[email protected]; **[email protected]
(Received July 31, 2014)
Current-perpendicular-to-plane structural Fe3Si/FeSi2/Fe3Si trilayered junctions were prepared on
Si(111) by facing targets direct-current sputtering combined with a mask method. The shape of
magnetization curves evidently exhibited that an antiparallel alignment is realized owing to a
difference in the coercive force between the top and bottom Fe3Si layers. In addition, it was
demonstrated that the antiparallel alignment in the wide range of applied magnetic field can be
realized by forming a crossover structure, which is owing to an enhanced difference in the effective
magnetic field between the top and bottom Fe3Si layers aligned perpendicularly to each other.
1.
Introduction
Spintronics devices that utilize not only the charge of carriers but also their spin have considerable
attention from physical and practical viewpoints. Giant magnetoresistance (GMR) [1,2] and
tunneling magnetoresistance (TMR) [3-9] junctions are the representatives of spintronics devices.
The core structures of GMR and TMR junctions are ferromagnetic metal/nonmagnetic metal
multilayers and ferromagnetic metal/insulator/ferromagnetic metal trilayers, respectively.
We have studied Fe3Si/FeSi2 artificial lattices, in which the ferromagnetic Fe3Si and
semiconducting FeSi2 layers were alternatively accumulated in nano scale by facing targets direct
current sputtering (FTDCS) and their spin-dependent electrical properties accompanied by a change
in the interlayer coupling induced between the Fe3Si layers across the FeSi2 layer. The combination
of Fe3Si and FeSi2 has the following merits [10-16]: (i) magnetoresistance effects in the currentperpendicular-to-plane (CPP) structures is easily detectable since the electrical resistivity of FeSi2
spacer layers is distinctively larger than that of Fe3Si layers; (ii) a spin injection efficiency might be
higher than that of TMR films; (iii) the epitaxial growth of Fe3Si layers on Si(111) substrates is
successively kept up to the top Fe3Si layer across FeSi2 spacer layers, which is beneficial to the
coherent transportation of spin-polarized electrons; (iv) Fe3Si is feasible for a practical use since it
has a high Curie temperature of 840 K and a large saturation magnetization, which is half of that of
Fe.
Spin-dependent scattering of polarized carriers have been studied through GMR and TMR
effects thus far. A switch between the parallel and antiparallel magnetization alignments of
ferromagnetic layers in multilayered films, so-called spin valve, is a key operation for modulating
spin-polarized current [17-20]. Spin valves are classified into two types. One is a multilayered film
wherein antiferromagnetic interlayer coupling is induced between ferromagnetic layers at a zero
magnetic field. The antiparallel alignment is switched to a parallel alignment by applying magnetic
011504-1
011504-2
JJAP Conf. Proc. 3 (2015) 011504
fields. The other is multilayer films comprising ferromagnetic layers with different coercive forces.
The antiparallel/parallel alignments, which depends on the applied magnetic field, are induced owing
to the different coercive forces. Practically, a difference in the coercive force is produced as follows:
(i) the combination of different materials for ferromagnetic layers; (ii) the employment of pining
layers; and (iii) purposely changing the crystalline quality (polycrystalline or epitaxial) and film
thickness between ferromagnetic layers, for instance between top and bottom layers for trilayered
films.
In this work, the (iii) method was employed. In addition, in order to produce antiparallel
alignments in the wide range of applied magnetic field, Fe3Si/FeSi2/Fe3Si trilayered films were
prepared in crossover structure by employing a mask method. It was experimentally demonstrated
that the magnetic field range for producing antiparallel alignments can be extended owing to a
magnetic shape anisotropy.
2.
Experimental procedure
Fe3Si (700 nm)/FeSi2 (0.75 nm)/Fe3Si (100 nm) trilayer films were deposited by FTDCS, combined
with a mask method in a procedure schematically shown in Fig. 1(a). First, a p-type Si(111) substrate
with a specific resistance range of 1000-3000 Ω cm, which was produced by a floating zone (FZ)
method, was cleaned with 1% hydrofluoric acid and rinsed in deionized water before it was set into
a FTDCS apparatus together with a mask. An Fe3Si bottom layer (100 nm) was deposited on the
Si(111) substrate, using the 1st mask with line widths of X (X = 0.4, 0.6, and 0.8) mm. After the
deposition, the sample was temporarily took out from the FTDCS apparatus to replace the mask.
Line width = X mm
Junction area = X2 mm2
(a)
(b)
Fe3Si (700 nm)
FeSi2 (0.75 nm)
Fe3Si (100 nm)
Si(111)
(c)
Fig. 1. (a) Schematic of preparation procedure of CPP trilayered junction with masks,
(b) top view of CPP trilayered junction, and (c) cross-sectional schematic of
intersection part of CPP trilayered junction.
011504-3
JJAP Conf. Proc. 3 (2015) 011504
After replacing the mask, FeSi2 (0.75nm) and Fe3Si (700 nm) layers were successively deposited. All
the depositions were carried out at a substrate temperature of 300 C. The base pressure was lower
than 3×105 Pa and the film deposition was carried out at 1.33×101 Pa. The crystalline structure of
the films was characterized by X-ray diffraction (XRD) using Cu K radiation. The magnetization
curves were measured at room temperature using a vibration sample magnetometer (VSM). The
external magnetic field was applied parallel to the line of bottom Fe3Si layer with the thickness of
100 nm. It corresponds to the horizontal direction in Fig. 1(b).
3.
Results and discussion
20
40
2[degree]
60
80
Fe 3Si 222
Si 222
Fe 3Si 220
Si 111
Intensity [arb. unit]
Si 222
Intensity [arb. unit]
Si 111
The 2θ-θ XRD patterns of a Si(111) substrate as a background and CPP film deposited on the Si(111)
20
40
2[degree]
60
(a)
(b)
Fig. 2. 2θ-θ X-ray diffraction patterns of (a) Si(111) substrate (background) and
(b) CPP trilayered film deposited on the Si(111) substrate.
90
β
α
180
20 40 60 80
0
270
Fig. 3. Pole-figure concerning Fe3Si-422 planes.
80
011504-4
JJAP Conf. Proc. 3 (2015) 011504
Line width = 0.8 mm
Junction area = 21.16 mm2
(a) Square shape.
Line width = 0.6 mm
Junction area = 0.64 mm2
(b) Line width X = 0.8 mm.
Line width = 0.4 mm
Junction area = 0.36 mm2
Junction area = 0.16 mm2
(c) Line width X = 0.6 mm.
(d) Line width X = 0.4 mm.
Fig. 4. Top views of (a) non-crossover structural trilayered film and (b) 0.8,
(c) 0.6, and (d) 0.4 mm line width crossover structural trilayered films.
substrate are shown in Fig. 2(a) and 2(b), respectively. The diffraction peaks of Fe3Si-220 and 222
are observed. The Fe3Si-220 peak is attributable to non-oriented crystallites. Figure 3 shows a polefigure concerning the Fe3Si-422 plane with a rotation axis of Fe3Si [222]. It was confirmed that
oriented crystallites are also in-plane ordered. Totally considering the results including those of our
previous research, wherein Fe3Si thin films are epitaxially grown on Si(111) substrate even at room
temperature, the bottom Fe3Si layer should epitaxially be grown on the Si(111) substrate. On the
other hand, although the top Fe3Si thick layer deposited on FeSi2 layer might partially be oriented
with the same orientation relationship as the bottom layer, it might be predominantly polycrystalline
due to the temporal exposure to air for the replacement of the masks.
Figure 4 shows the top view of trilayered films prepared with no masks and three types of masks.
It is confirmed that the trilayered films prepared with the masks have crossover structures, as
expected.
Figure 5 shows the hysteresis loops of the trilayered films. As shown in Fig. 5(a), the trilayered
film with no crossover structure exhibits the hysteresis loop with a weak step, which indicated that a
difference in the coercive force between the top and bottom Fe3Si layers is not so large.
The hysteresis loops of the crossover structural films exhibit steps clearly as shown in Fig. 5(b),
5(c), and 5(d), as compared with that of non-crossover structural film. The clearness of the step is
evidently enhanced with decreasing line width, which implies that the effective magnetic field is
strongly affected by the geometry, concretely line width. The effective magnetic field is defined as
follows:
𝐻𝑒𝑓𝑓 = 𝐻𝑒𝑥 + 𝐻𝑑
𝐻𝑒𝑓𝑓 : effective magnetic field
𝐻𝑒𝑥 : external magnetic field
𝐻𝑑 : demagnetizing field
011504-5
JJAP Conf. Proc. 3 (2015) 011504
1
M/Ms
M/Ms
1
0
–1
0
–1
–20
0
–20
20
Magnetic field [Oe]
(a) Square shape.
20
(b) Line width X= 0.8 mm.
1
M/Ms
1
M/Ms
0
Magnetic field [Oe]
0
–1
0
–1
–20
0
20
Magnetic field [Oe]
(c) Line width X = 0.6 mm.
–20
0
20
Magnetic field [Oe]
(d) Line width X = 0.4 mm.
Fig. 5. Hysteresis loops of trilayered films with (a) no crossover structure,
(b) 0.8, (c) 0.6, and (d) 0.4 mm line width crossover structures.
The demagnetization field is strongly dependent on the shape of ferromagnetic materials. With
decreasing line width, the difference in the demagnetizing field between the top and bottom Fe3Si
layers is increased, which results in the effective magnetic field being further differently applied to
the top and bottom Fe3Si layers. Since the magnetic field is applied parallel to the line of the bottom
Fe3Si layer, the demagnetization field in the bottom layer is weak. On the other hand, for the line of
the top Fe3Si layer, the magnetic field is applied perpendicular to the line. Thus, the demagnetizing
field become so large. With decreasing line width, the difference in the demagnetizing field between
the top and bottom Fe3Si layers become extreme, in other words, the effective magnetic field is
remarkably reduced for the top Fe3Si layer and is hardly changed for the bottom Fe3Si layer.
Considering the above-mentioned magnetic shape anisotropy, the change in the magnetization
alignment can qualitatively considered as follows. The magnetization of the bottom Fe3Si layer
follows the reverse applied magnetic field at first, from the parallel alignment. As a result, the
magnetization of the bottom Fe3Si layer become parallel with the reverse magnetic field while the
magnetization of the top Fe3Si layer keeps antiparallel with the reverse magnetic field. In the reverse
magnetic field range, antiparallel alignments are produced between the top and bottom Fe3Si layers
magnetizations. With increasing reverse magnetic field, the magnetization of the top Fe3Si layer is
reversed, and both magnetizations become parallel. The reverse of the top layer magnetization from
the antiparallel to parallel alignment is further delayed by the enhanced demagnetizing field. Since
the demagnetizing field is enhanced with decreasing line width, the crossover structural films with
the small line widths exhibits the clear steps in the hysteresis loops.
JJAP Conf. Proc. 3 (2015) 011504
4.
011504-6
Conclusion
CPP-structural Fe3Si/FeSi2/Fe3Si trilayered junctions with crossover structures were prepared on
Si(111) by FTDCS combined with a mask method. The hysteresis loops evidently exhibited steps
that originates from the antiparallel alignment between the top and bottom Fe3Si layers
magnetizations owing to a difference in the coercive force of the Fe3Si layers. The difference in the
coercive force is caused by differences in the layer thickness and crystalline quality (epitaxial or
polycrystalline). In addition, it was experimentally demonstrated that the magnetic field range for
producing the antiparallel alignment can be extended by forming crossover structures. It should be
attributable to a magnetic shape anisotropy, and the line width dependent change in the hysteresis
loop profile with the applied magnetic field was qualitatively explained well.
Acknowledgment
The measurement of magnetization curves was performed at Fukuoka Institute of Technology.
References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
G. Binasch, P. Grünberg, F. Saurenbach and W. Zinn: Phys. Rev. B 39 (1989) 4828.
M. N. Baibich, J. M. Broto, A. Fert, F. Nguyen Van Dau and F. Petroff: Phys. Rev. Lett. 61 (1988) 2472.
M. Julliere: Phys. Lett. 54A (1975) 225.
T. Miyazaki and N. Tezuka: J. Magn. Magn. Mater. 139 (1995) L231.
J. S. Moodera, L. R. Kinder, T. M. Wong and R. Meservey: Phys. Rev. Lett. 74 (1995) 3273.
T. Miyazaki and N. Tezuka: J. Magn. Magn. Mater. 151 (1995) 403.
K. Inomata, S. Okamura, R. Goto and N. Tezuka: Jpn. J. Appl. Phys. 42 (2003) L419.
S. Yuasa, T. Nagahama, A. Fukushima, Y. Suzuki and K. Ando: Nat. Mater. 3 (2004) 868.
R. Moriya, K. Hamaya, A. Oiwa and H. Munekata: Jpn. J. Appl. Phys. 43 (2004) L825.
T. Sadoh, T. Takeuchi, K. Ueda, A. Kenjo and M. Miyao: Jpn J. Appl. Phys., 45 (2006) 3598.
Y. Jing, Y. Xu and J.-P. Wang: J. Appl. Phys. 105 (2009) 07B520.
M. Shaban, H. Kondo, K. Nakashima and T. Yoshitake: Jpn. J. Appl. Phys. 47 (2008) 5420.
E. E. Fullerton, J. E. Mattson, S. R. Lee, C. H. Sowers, Y. Y. Huang, G. Felcher, S. D. Bader and F. T.
Parker: J. Appl. Phys. 73 (1993) 6335.
K. Inomata, K. Yusu and Y. Saito: Phys. Rev. Lett. 74 (1995) 1863.
A. Chaiken, R. P. Michel and M. A. Wall: Phys. Rev. B 53 (1996) 5518.
Y. Endo, O. Kitakami, and Y. Shimada: Phys. Rev. B 59 (1999) 4279.
W. H. Meiklejohn, and C. P. Bean: Phys. Rev. 102 (1956) 1413.
V. Jelinek: Tectonophysics, 79 (1981) T63.
A. E. Berkowitz, and K. Takano: J. Magn. Magn. Mater. 200 (1999) 552.
Q. Song, and Z. J. Zhang: J. Am. Chem. Soc. 126 (2004) 6164.