Content uploaded by Young Sun
Author content
All content in this area was uploaded by Young Sun
Content may be subject to copyright.
Structure and magnetic properties of the self-assembled Co52Pt48
nanowire arrays
Jian-Hua Gao, Da-Li Sun, Xiang-Qun Zhang, Qing-Feng Zhan, Wei He,
Young Sun, and Zhao-Hua Chenga兲
State Key Laboratory of Magnetism and Beijing National Laboratory for Condensed Matter Physics,
Institute of Physics, Chinese Academy of Sciences, Beijing 100080, People’s Republic of China
共Received 13 January 2008; accepted 18 February 2008; published online 10 March 2008兲
Co52Pt48 nanowire arrays with diameter of about 10 nm have been fabricated by electrodeposition
into the anodic aluminum oxide templates. The as-deposited nanowire arrays exhibit fcc structure
with 关110兴-preferred texture along the wire axes. The nanowire arrays present both large
magnetocrystalline anisotropy along the 关111兴direction and strong shape anisotropy along the wire
axis, resulting in high coercive fields parallel and perpendicular to the wire axes, respectively. From
experimental results and micromagnetic simulation, the coercive field variation as a function of the
angles evidenced that the 关111兴magnetocrystalline anisotropy plays important role besides shape
magnetic anisotropy. © 2008 American Institute of Physics.关DOI: 10.1063/1.2894199兴
In the recent years, magnetic nanowire arrays have re-
ceived growing interest due to not only strong potential can-
didates for high density magnetic recording media but also
the important role in the fundamental research. Among them,
transition metal alloys such as Co–Pt and Fe–Pt alloys, due
to high magnetic perpendicular anisotropy 关⬃2⫻106J/m3
for Co–Pt 共Ref. 1兲and ⬃6⫻106J/m3for Fe–Pt 共Ref. 2兲兴,
are considered promising candidates for potential application
in magnetic storage media. It is only very recently that Co–Pt
and Fe–Pt nanowire arrays have been fabricated.3–13 Huang
et al.3firstly fabricated Co–Pt and Fe–Pt nanowires by elec-
trodeposition. Yasui et al.5have also electrodeposited Co–Pt
nanowires and try to adjust perpendicular magnetic aniso-
tropy by textured underlayer. Mallet et al.6and Dahmane
et al.14 and Min et al.15 have reported that the as-deposited
nanowires exhibit fcc structure with soft magnetic properties,
an annealing treatment is necessary to obtain L10phase with
large coercive field higher than 10 kOe. Unfortunately, the
nanowires usually lose uniaxial magnetic anisotropy and ex-
hibit isotropic anisotropy after the postannealing. Another
disadvantage of annealed nanowires is that it is difficult to
obtain excellent single phase.12,14 In addition, most reports
are concentrated on the fabrication, and little investigation
focused on the magnetization reversal mechanism, in which
magnetic anisotropy is the heart of understanding the mag-
netization reversal process. In the present work, magnetic
properties of the electrodeposited Co52Pt48 nanowire arrays
have been investigated. The contribution of shape magnetic
anisotropy and magnetocrystalline anisotropy to magnetiza-
tion reversal has been simulated by micromagnetic simula-
tion.
The fabrication of nanowire arrays began with nano-
porous anodic aluminum oxide 共AAO兲templates prepared by
two-step anodic anodization process.16 Co52Pt48 nanowires
are electrodeposited into the AAO templates by ac deposition
from the electrolytes consisted of 0.01MCoSO4and 0.01M
Pt共NO2兲2共NH3兲2. Using inductively coupled plasma analysis,
the composition of the nanowires is determined with atomic
percentage Co52Pt48. The crystal structure of the nanowire
arrays is characterized by x-ray diffraction 共XRD兲. The mag-
netic hysteresis loops have been measured by a supercon-
ducting quantum interference device 共SQUID兲magnetometer
and vibrating sample magnetometer 共VSM兲. The micromag-
netic simulation has been performed using the object ori-
ented micromagnetic computing framework 共OOMMF兲
code.17
Figure 1shows the XRD pattern of the Co52Pt48 nano-
wire arrays with the Cu K
␣
radiation. From the diffraction
peaks, the constant lattice can be determined and the value is
a=3.806 Å, corresponding to fcc structure of the Co–Pt al-
loy. Furthermore, fcc structure is the only phase and prefer-
ential growth of the 关110兴orientation along wire axes is ob-
served for the as-deposited nanowires. Therefore, the
Co52Pt48 as-deposited nanowire arrays have fcc structure
with strong 关110兴-preferred texture along the wire axes.
Magnetic hysteresis loops measured by SQUID magne-
tometer at 5 K and calculated by micromagnetic simulation,
is shown in Fig. 2. Figure 2共a兲is the measured loops with
magnetic field parallel and perpendicular to the wire axis.
For the applied field parallel to the nanowire arrays, the hys-
teresis loop is relatively square with coercive field HC
储
=4.52 kOe and remanence squareness S
储
=0.76. While when
the applied field is perpendicular to the wire axes, the smaller
coercive field HC⬜=2.98 kOe and the remanence squareness
a兲Author to whom correspondence should be addressed. Electronic mail:
zhcheng@aphy.iphy.ac.cn.
FIG. 1. X-ray diffraction spectrum of the as-deposited Co52Pt48 nanowire
arrays with the wire diameter of 10 nm.
APPLIED PHYSICS LETTERS 92, 102501 共2008兲
0003-6951/2008/92共10兲/102501/3/$23.00 © 2008 American Institute of Physics92, 102501-1
Downloaded 10 Mar 2008 to 159.226.36.156. Redistribution subject to AIP license or copyright; see http://apl.aip.org/apl/copyright.jsp
S⬜=0.28 are obtained. Therefore, we can conclude that the
as-deposited Co52Pt48 nanowire arrays exhibit a hard ferro-
magnetic behavior, with easy magnetization axes along the
wire axes. Similar results were reported by Mallet et al. for
the CoxPt1−xnanowire arrays.4
According to the chain spheres of symmetric fanning
mechanism,18 the coercive field parallel to the wire axes
should be HC
储
共sy兲=共
MS/6兲共6Kn−4Ln兲=1.49 kOe, where
MS⬇920 emu/cc is the saturation magnetization for bulk
Co–Pt alloy within the error range.19 The coercive field per-
pendicular to the wire axes should be zero. However, the
experimental coercive fields in the two directions are HC
储
=4.52 kOe and HC⬜=2.98 kOe, respectively, which are
much larger than the chain sphere prediction. This large dis-
crepancy is considered to be the magnetocrystallline aniso-
tropy contribution of the Co52Pt48 nanowire arrays. It has
been predicted by calculation that Co50Pt50 films with fcc
structure exhibit large intrinsic magnetocrystallline aniso-
tropy along the 关111兴direction.20,21 Since the nanowire ar-
rays are textured along the wire axes, i.e., 关110兴direction,
therefore, the shape anisotropy should be along the 关110兴
direction. The easy magnetocrystalline axes is along the
关111兴direction, which makes angles of about 35° with the
wire axis. According to the calculation prediction, the mag-
nitudes of the two anisotropies are comparable, so this mag-
netocrystalline anisotropy should be taken into account.
In order to further confirm the magnetocrystalline aniso-
tropy contribution, micromagnetic simulation is used to cal-
culate the hysteresis loops. The simulated hysteresis loops
with different magnetocrystalline anisotropy configurations
are shown in Fig. 2共b兲, for comparison, the measured loop
is also shown. As dissussed in details for Fe3Pt nanowires,22
the dipolar interaction between nanowires is very important
for nanowire arrays. In fact, the dipolar interaction is taken
into account in the OOMMF code during the simulation when
the wire diameter, length, and distance of nanowire array are
given. In the simulation, a hexagonal cell of 4⫻4 array of
nanowires is chosen, with the unit cell size 2.5⫻2.5
⫻5nm
3. The geometry parameters come from the experi-
ment, with the wire diameter d=10 nm and length L
=400 nm. Since the wire diameter is up to 10 nm, the satu-
ration magnetization MSand exchange stiffness Aare close
to corresponding bulk values, with MS=9.2⫻105and A
=2.65⫻10−12 A/m and magnetocrystalline anisotropy con-
stant Ku=2.0⫻105J/m3. If we assume that the magneto-
crystalline anisotropy axes are fixed in a certain direction,
the simulated coercive field of 5.5 kOe is much higher than
the experimental results. Although the nanowires have strong
关110兴-preferred texture along the wire axes, the 关111兴direc-
tion of crystallites distributes randomly with the azimuthal
angle in the cross section perpendicular to the wire axis ow-
ing to the polycrystalline nanowires 关inset of Fig. 2共b兲兴.
Therefore, we simulated the hysteresis loops with the 关111兴
magnetocrystalline axis randomly distributed in the cross
section. The simulated coercive field of 4.59 kOe is in good
agreement with measured coercivity of HC
储
=4.52 kOe. The
relatively large deviation in remanence squareness S
储
is re-
lated to the surface magnetic moment deviating from the
axes of the nanowires.
Figure 3shows the angle dependence of the coercive
fields 关HC共
兲兴 for Co52Pt48 nanowire arrays obtained by ex-
periment and micromagnetic simulation. The solid squares
are the experimental points measured by VSM magnetometer
at room temperature. There is a platform in the HC共
兲curve
for the angles below 20° and the coercive fields decrease
rapidly with further increasing angles. The HC共
兲variation
exhibits a bell-type shape as reported for Ni nanowires.23
This HC共
兲variation shape is determined by rotation reversal
mechanism instead of curling mechanism,24 which is not sur-
prising since the diameter of the Co52Pt48 nanowire arrays is
smaller than the critical diameter about 25 nm 共dcoh
=冑24A/
0MS
2兲.25 Also, we should keep in mind that the
HC共
兲variation is different from the infinite cylinders with
uniaxial anisotropy. As we discussed above, due to the 关111兴
magnetocrystalline easy axes, making an angle of 35° with
FIG. 2. 共Color online兲共a兲Magnetic hysteresis loops of the Co52Pt48 nano-
wire arrays measured by SQUID magnetometer with magnetic field parallel
and perpendicular to wire axis at 5 K. 共b兲The simulated loops by micro-
magnetic simulation with magnetic field parallel to the wire axes for differ-
ent magnetocrytalline axes distribution. The solid triangles and open circles
correspond to magnetocrytalline axes in fixed direction and distributes ran-
domly in the cross section.
FIG. 3. 共Color online兲Angle dependence of the coercive fields for the
Co52Pt48 nanowire arrays. The solid squares indicate experimental points
obtained by VSM magnetometer at 300 K. The solid triangles and open
circles denote the simulation results with magnetocrytalline axis in fixed
direction and distributed in the cross section with the randomly azimuthal
angles, respectively.
102501-2 Gao et al. Appl. Phys. Lett. 92, 102501 共2008兲
Downloaded 10 Mar 2008 to 159.226.36.156. Redistribution subject to AIP license or copyright; see http://apl.aip.org/apl/copyright.jsp
the wire axis, the Co52Pt48 nanowire arrays are not ideal sys-
tem with uniaxial anisotropy. Using micromagnetic simula-
tion, we have calculated the coercive fields at different
angles, taking magnetocrytalline anisotropy into account
with different configurations, as shown in Fig. 3. The open
circles are the simulation results with magnetocrytalline axis
fixed in a certain direction. There appears a peak around 15°
in the curve which exihibits large deviation with the experi-
ment results. We assume that the 关111兴magnetocrytalline
axes distribute randomly in the cross section perpendicular to
the wire axis and the simulated curve is shown by the solid
triangles. For this magnetocrytalline axes configuration, the
HC共
兲variation in coercive fields has improved better at
small angles, even there exists relative large deviation at the
large angles. We can conclude that besides shape anisotropy
along the wire axis, magnetocrytalline anisotropy also plays
an important role for fcc Co52Pt48 nanowire arrays.
In summary, the structure and magnetic properties of
Co52Pt48 nanowire arrays with the diameter 10 nm have been
investigated. The XRD pattern shows that as-deposited nano-
wire arrays exhibit fcc structure with the 关110兴orientation
along the wire axis. Due to shape anisotropy along the wire
axes and magnetocrystalline anisotropy along the 关111兴di-
rection, the nanowire arrays exhibit large coercive fields both
parallel and perpendicular to the wire axis. Magnetocrystal-
line anisotropy plays important role is evidenced by the ex-
periment and micromagnetic simulation.
This work was supported by the State Key Project of
Fundamental Research and the National Natural Science
Foundation of China.
1R. A. M. Currie and P. Gaunt, Philos. Mag. 13,567共1966兲.
2O. A. Ivanov, L. V. Solina, and V. A. Demshina, Phys. Met. Metallogr. 35,
81 共1973兲.
3Y. H. Huang, H. Okumura, G. C. Hadjipanayis, and D. Weller, J. Appl.
Phys. 91, 6869 共2002兲.
4J. Mallet, K. Y. Zhang, C. L. Chien, T. S. Eagleton, and P. C. Searson,
Appl. Phys. Lett. 84, 3900 共2004兲.
5N. Yasui, A. Imada, and T. Den, Appl. Phys. Lett. 83,3347共2003兲.
6J. Mallet, K. Y. Zhang, S. M. Tempfli, and L. Piraux, J. Phys. D 38,909
共2005兲.
7F. M. F. Rhen, E. Backen, and J. M. D. Coey, J. Appl. Phys. 97, 113908
共2005兲.
8A. I. Gapin, X. R. Ye, J. F. Aubuchon, L. H. Chen, Y. J. Tang, and S. Jin,
J. Appl. Phys. 99, 08G902 共2006兲.
9Y. Peng, T. Cullis, G. Möbus, X. J. Xu, and B. Inkson, Nanotechnology
18, 485704 共2007兲.
10J. Zhang, W. X. Li, G. A. Jones, and T. H. Shena, J. Appl. Phys. 99,
08Q502 共2006兲.
11J. H. Gao, D. L. Sun, Q. F. Zhan, W. He, and Z. H. Cheng, Phys. Rev. B
75, 064421 共2007兲.
12W. X. Li, J. Zhang, J. Thompson, T. H. Shen, E. Arenholz, and S. A.
Morton, M. Ellerby, Appl. Phys. Lett. 91, 133111 共2007兲.
13T. S. Eagleton, J. Mallet, X. M. Cheng, J. Wang, C. L. Chien, and P. C.
Searson, J. Electrochem. Soc. 152,C27共2005兲.
14Y. Dahmane, L. Cagnon, J. Voiron, S. Pairis, M. Bacia, L. Ortega, N.
Benbrahim, and A. Kadri, J. Phys. D 39, 4523 共2006兲.
15J. H. Min, B. H. An, J. U. Cho, H. M. Ji, S. J. Noh, Y. K. Kim, H. L. Liu,
J. H. Wu, Y. D. Ko, and J. S. Chung, J. Appl. Phys. 101, 09K513 共2007兲.
16J. H. Gao, Q. F. Zhan, W. He, D. L. Sun, and Z. H. Cheng, Appl. Phys.
Lett. 86, 232506 共2005兲.
17The simulation code from object oriented micromagnetic framework
共OOMMF兲by M. Donahue and D. Porter is used, 共http://math.nist.gov/
oommf兲.
18S. Jacobs and C. P. Bean, Phys. Rev. 100,1060共1955兲.
19H. P. J. Wijn, Data in Science and Technology, Magnetic Properties of
Metals 共Springer, Berlin, 1991兲.
20S. S. A. Razee, J. B. Staunton, B. Ginatempo, F. J. Pinski, and E. Bruno,
Phys. Rev. Lett. 82, 5369 共1999兲.
21S. S. A. Razee, J. B. Staunton, B. Ginatempo, F. J. Pinski, and E. Bruno,
Phys. Rev. B 64, 014411 共2001兲.
22D. L. Sun, J. H. Gao, X. Q. Zhang, Q. F. Zhan, W. He, Y. Sun, and Z. H.
Cheng 共unpublished兲.
23G. C. Han, B. Y. Zong, P. Luo, and Y. H. Wu, J. Appl. Phys. 93, 9202
共2003兲.
24W. F. Brown, Phys. Rev. 105, 1479 共1957兲.
25R. Skomski and J. M. D. Coey, Permanent Magnetism 共Institute of Phys-
ics, Bristol, 1999兲.
102501-3 Gao et al. Appl. Phys. Lett. 92, 102501 共2008兲
Downloaded 10 Mar 2008 to 159.226.36.156. Redistribution subject to AIP license or copyright; see http://apl.aip.org/apl/copyright.jsp