Download Switching magnetic vortex core by a single nanosecond current

1
Switching magnetic vortex core by a single nanosecond
current pulse
Keisuke Yamada1, Shinya Kasai1, Yoshinobu Nakatani2, Kensuke
Kobayashi1 & Teruo Ono1
1
Institute for Chemical Research, Kyoto Univeristy, Uji 611-0011, Japan
2
University of Electro-communications, Chofu 182-8585, Japan
Abstract
In a ferromagnetic nanodisk, the magnetization tends to swirl around in the
plane of the disk and can point either up or down at the center of this
‘magnetic vortex’. This binary state can be useful for information storage.
It is demonstrated that a single nanosecond current pulse can switch the
core polarity. This method also provides the precise control of the core
direction, which constitutes fundamental technology for realizing a vortex
core memory.
1
2
Since the experimental confirmation of the existence of a nanometerscale core with out-of-plane magnetization in a magnetic vortex,1, 2 the
dynamics of a vortex has been investigated intensively,3-11 and the magnetic
vortex core has been expecting to be a candidate for future nonvolatile data
storage devices because of its small size and its good thermal stability.
However, flipping the vortex core direction from up to down requires a
fairly large magnetic field.12 Recently, this technological hurdle has been
reduced by utilizing the resonance effect of the vortex core motion induced
either by the application of an ac magnetic field13, 14 or by injecting an ac
electric current into a magnetic disk.9, 15 We have shown that the core
switches when its velocity reaches a certain value regardless of the value of
the excitation current density, indicating that the core switching is triggered
by a strong dynamic field which is produced locally by a rotational core
motion at a high speed of several hundred m/s in either case of the
magnetic field or electric current excitation.15 Although the final core
direction can not be controlled in this method, this understanding leads to
the idea that the core can be switched if the core is accelerated to high
enough velocity without any help of the resonance.16-18 Here, we report that
2
3
a single short current pulse injecting into a ferromagnetic nanodisk can
determine the final direction of the core.
Figure 1 shows an atomic force microscope image of the sample and
the schematic configuration used for the measurements. The samples were
fabricated on sapphire (Al2O3) substrates by the lift-off method in
combination with e-beam lithography. The sample consists of a Permalloy
(Fe19Ni81) disk and two wide electrodes (Au(70 nm)/Cr(10 nm)), through
which an electric current pulse is supplied. The thickness of the disk is 55
nm, and the diameter is 1.55 µm.
Experimental procedure is the following. First, the direction of a core
magnetization is determined by magnetic force microscope (MFM)
observation. Then, an electric current pulse is injected through the
magnetic disk, and the core direction is again checked by MFM. Figure 2
shows the successive MFM observations with one pulse current applied
between each consecutive image. The current density and the pulse
duration were 1.3 × 1012 A/m2 and 2.5 ns, respectively. A dark spot at the
center of the disk in Fig. 2(b) indicates that the core magnetization directs
upward with respect to the paper plane. As shown in Fig. 2(c), the dark spot
3
4
at the center of the disk changed into bright after the injection of the pulse
current, indicating that the direction of the core magnetization has been
switched. The switching occurred after every injection of the current pulse
as shown in Fig. 2(b)-2(l). Thus, we can precisely control the core direction
by the application of the nanosecond current pulse.
Figures 3(a)–3(e) are successive snapshots of the calculated results
for the magnetization distribution during the process of core motion and
switching induced by injection of the current pulse. The current-induced
dynamics of a vortex was calculated by micromagnetic simulations in the
framework of the Landau-Lifshitz-Gilbert (LLG) equation with a spintransfer term.9, 15, 19, 20 We performed the two-dimensional calculation by
dividing the disk into rectangular prisms of 4 × 4 × 50 nm3; the
magnetization was assumed to be constant in each prism. The typical
material parameters for Permalloy were used: saturation magnetization Ms
= 1 T, exchange stiffness constant A = 1.0 × 10–11 J/m, and damping
constant α = 0.01. The spin polarization of the current was taken as P =
0.7.21 Noteworthy is the development of an out-of-plane magnetization
(dip) opposite to the core magnetization, which is also observed in the case
4
5
of the core switching by the resonant excitation.13, 15 The creation of a pair
of the vortex and anti-vortex just before the core switching was also
confirmed in the simulation. Thus, the mechanism of the switching by the
pulse current can be considered to be the same as in the case of the
switching by the resonant excitation.
Figure 4 shows the switching probabilities as a function of pulse
duration. The switching probability is defined as the ratio of the number of
the core switching events to the total trial number. The current density is
1.3 × 1012 A/m2. The high switching probabilities are obtained for the pulse
duration from 2.3 to 3.0 ns, while no switching is observed for the pulse
duration below 1.8 ns. Results for the pulse duration above 3 ns could not
be obtained because the sample was damaged possibly due to the severe
Joule heating for the longer pulse. We could not access to the higher
current density above 1.3 × 1012 A/m2 due to the same reason. We
confirmed that the core switching does not occur for the current density of
1.1 × 1012 A/m2 regardless of the pulse duration, indicating that the current
density of 1.3 × 1012 A/m2 is just above the threshold current density.
5
6
The results of the micromagnetic simulation are superimposed on the
experimental results in Fig. 4. While the simulation reproduces the
existence of the threshold pulse duration below which the core is not
switched, two discrepancies are found between the experimental results and
the simulation; (i) The critical current density for the switching in the
simulation (2.8 × 1012 A/m2) is higher than the experimental value (1.3 ×
1012 A/m2). (ii) The critical pulse duration is longer in the experiments.
Although the reason for these discrepancies is not clear at this stage, most
probable are the sample heating and/or the magnetic field which are
induced by the application of the current pulse.22 The heating and the
magnetic field can contribute to the reduction of the current density for the
core switching in the experiments.
We have shown that a single short current pulse injecting into a
ferromagnetic nanodisk can switch the direction of the core. This enables
us the fast and exact control of the core direction, which is indispensable
for realizing a memory cell where the bit data is stored as the direction of
the nanometre-scale core magnetization. The experimentally obtained
critical current density for the switching is smaller than the value in the
6
7
simulation, indicating the importance of the sample heating and/or the
magnetic field induced by the current pulse. Designing the cell structure to
utilize the field effect in a cooperative manner to reduce the critical current
is of importance from the technological point of view.
The present work was partly supported by MEXT Grants-in-Aid
for Scientific Research in Priority Areas and JSPS Grants-in-Aid for
Scientific Research.
7
8
References
1. T. Shinjo, T. Okuno, R. Hassdorf, K. Shigeto, and T. Ono, Science 289,
930-932 (2000).
2. A. Wachowiak, J. Wiebe, M. Bode, O. Pietzsch, M. Morgenstern, and R.
Wiesendanger, Science 298, 577-580 (2002).
3. K. Y. Guslienko, B. A. Ivanov, V. Novosad, Y. Otani, H. Shima, and K.
Fukamichi, J. Appl. Phys. 91, 8037-8039 (2002).
4. J. P. Park, P. Eames, D. M. Engebretson, J. Berezovsky, and P. A.
Crowell, Phys. Rev. B 67, 020403 (2003).
5. S. B. Choe, Y. Acremann, A. Scholl, A. Bauer, A. Doran, J. Stöhr, and H.
A. PadmoreChoe, Science 304, 420-422 (2004).
6. H. Stoll, A. Puzic, B. V. Waeyenberge, P. Fischer, J. Raabe, M. Buess,
T. Haug, R. Höllinger, C. Back, D. Weiss, G. Denbeaux, Appl. Phys. Lett.
84, 3328-3330 (2004).
7. V. Novosad, F. Y. Fradin, P. E. Roy, K. S. Buchanan, K. Y. Guslienko,
and S. D. Bader, Phys. Rev. B 72, 024455 (2005).
8. K. Y. Guslienko, X. F. Han, D. J. Keavney, R. Divan, and S. D. Bader,
Phys. Rev. Lett. 96, 067205 (2006).
8
9
9. S. Kasai, Y. Nakatani, K. Kobayashi, H. Kohno, and T. Ono, Phys. Rev.
Lett. 97, 107204 (2006).
10. J. Shibata, Y. Nakatani, G. Tatara, H. Kohno, and Y. Otani, Phys. Rev.
B 73, 020403 (2006).
11. T. Ishida, T. Kimura, and Y. Otani, Phys. Rev. B 74, 014424 (2006).
12. T. Okuno, K. Shigeto, T. Ono, K. Mibu, and T. Shinjo, J. Magn. Magn.
Mater. 240, 1 (2002).
13. B. V. Waeyenberge, A. Puzic, H. Stoll, K. W. Chou, T. Tyliszczak, R.
Hertel, M. Fähnle, H. Brückl, K. Rott, G. Reiss, I. Neudecker, D. Weiss, C.
H. Back and G. Schütz, +ature 444, 462 (2006).
14. S. K. Kim, K. S. Lee, Y. S. Yu, and Y. S. Choi, Appl. Phys. Lett. 92,
022509 (2008).
15. K. Yamada, S. Kasai, Y. Nakatani, K. Kobayashi, H. Kohno, A.
Thiaville, and T. Ono, +ature Materials 6, 269 (2007).
16. R. Hertel, and C. M. Schneider, Phys. Rev. Lett. 97, 177202 (2006).
17. Q. F. Xiao, J. Rudge, B. C. Choi, Y. K. Hong, G. Donohoe, Appl. Phys.
Lett. 89, 262507 (2006).
9
10
18. Y. Liu, S. Gliga, R. Hertel, and C. M. Schneider, Appl. Phys. Lett. 91,
112501 (2007).
19. A. Thiaville, Y. Nakatani, J. Miltat, and N. Vernier, J. Appl. Phys. 95,
7049-7051 (2004).
20. A. Thiaville, Y. Nakatani, J. Miltat, and Y. Suzuki, Europhys. Lett. 69,
990-996 (2005).
21. J. Bass, and W. P. PrattJr, J. Magn. Magn. Mater. 200, 274-289 (1999).
22. M. Bolte, G. Meier, B. Krüger, A. Drews, R. Eiselt, L. Bocklage, S.
Bohlens, T. Tyliszczak, A. Vansteenkiste, B. V. Waeyenberge, K. W. Chou,
A. Puzic, and H. Stoll, Phys. Rev. Lett. 100, 176601 (2008).
10
11
Figure captions
Figure 1 Atomic force microscope image of the sample and
schematic illustration of the experimental setup. The thickness of the
disk is 55 nm, and the diameter is 1.55 µm. Two wide electrodes,
through which an electric current pulse is supplied, are also seen.
Figure 2 (a) AFM image of the sample. (b) MFM image before the
application of the current pulse. A dark spot at the center of the disk
indicates that the core magnetization directs upward with respect to
the paper plane. (c) MFM image after injecting the current pulse of
1.3 × 1012A/m2 with the pulse duration of 2.5 ns. The dark spot at the
center of the disk in Fig. 2(b) changed into the bright spot, indicating
the switching of the core magnetization from up to down. (d-l)
Successive MFM images with a current pulse applied similarly
between consecutive images. The direction of the core
magnetization is switched after every injection of the current pulse.
Figure 3 Perspective view of the magnetization with a moving vortex
structure. The height is proportional to the out-of-plane (z)
11
12
magnetization component. Rainbow colour indicates the in-plane
component as exemplified by the white arrows in (a). (a) Initially, a
vortex core magnetized upward is at rest at the disk center. (b) On
application of the current pulse, the core starts to move. (c) There
appears a region with downward magnetization (called ‘dip’ here) on
the inner side of the core. (d) The dip grows as the core is
accelerated. (e) After the completion of the reversal, the energy is
released to a substantial amount of spin waves. Calculation with the
same geometry as the experimental sample. 3D movie of
micromagnetic simulation is available as “Supplementary
Information”.
Figure 4 Switching probabilities as a function of pulse duration. The
switching probability is defined as the ratio of the number of the core
switching events to the total trial number, which ranges between 20
and 80 times for each duration. The big symbols represent the
experimental data, where the blue and red ones correspond to the
case for the current densities of 1.1× 1012 A/m2 and 1.3 × 1012 A/m2,
12
13
respectively. The green dotted line is the simulation results with the
current density of 2.8 × 1012A/m2.
13
14
Fig. 1
14
15
Fig. 2
15
16
Fig. 3
16
17
Fig. 4
17