Magnetoelectrolysis of Co nanowire arrays grown in a track-etched polycarbonate membrane

Article abstract of DOI:10.1016/j.jmmm.2006.09.020

Arrays of Cobalt nanowires with a controlled length of 6 μ m have been fabricated by electrochemical deposition into the pores of track-etched polycarbonate membranes with a nominal pore diameter of 30 nm. The magnetic properties of Co-deposited nanowires and the effects of a magnetic field applied during electrodeposition of the arrays have been studied. An enhancement of the mass deposition rate due to the presence of a 50 Oe magnetic field along the nanowire axis has been observed by measuring the experimental development of the current in the electrochemical cell during the fabrication process. X-ray diffraction measurements reveal a different polycrystalline degree for each deposition configuration, indicating that the crystalline structure of the deposited material has been substantially modified. Magnetic measurements show a clear dependence of the anisotropy directions on the orientation of the magnetic field applied during the electrodeposition.

Full text of DOI:10.1016/j.jmmm.2006.09.020

ARTICLE IN PRESS
Journal of Magnetism and Magnetic Materials 312 (2007) 99–106
www.elsevier.com/locate/jmmm
Magnetoelectrolysis of Co nanowire arrays grown in a
track-etched polycarbonate membrane
Ã
J. Sa
´
nchez-Barrigaa, , M. Lucasb, G. Riveroa, P. Marina, A. Hernandoa
aInstituto de Magnetismo Aplicado (UCM-RENFE-CSIC), P.O. Box 155, 28230, Las Rozas, Madrid, Spain
bTechnische Universita¨t Berlin, Institut fu¨r Theoretische Physik, Hardenbergstr. 36, D-10623 Berlin, Germany
Received 12 October 2005; received in revised form 9 September 2006
Available online 16 October 2006
Abstract
Arrays of Cobalt nanowires with a controlled length of 6 mm have been fabricated by electrochemical deposition into the pores of
track-etched polycarbonate membranes with a nominal pore diameter of 30 nm. The magnetic properties of Co-deposited nanowires and
the effects of a magnetic field applied during electrodeposition of the arrays have been studied. An enhancement of the mass deposition
rate due to the presence of a 50 Oe magnetic field along the nanowire axis has been observed by measuring the experimental development
of the current in the electrochemical cell during the fabrication process. X-ray diffraction measurements reveal a different polycrystalline
degree for each deposition configuration, indicating that the crystalline structure of the deposited material has been substantially
modified. Magnetic measurements show a clear dependence of the anisotropy directions on the orientation of the magnetic field applied
during the electrodeposition.
r 2006 Elsevier B.V. All rights reserved.
Keywords: Nanowires; Magnetoelectrolysis; Electrodeposition; Magnetic properties; Polycarbonate membranes
1. Introduction
advantage of increasing the storage capacity in magnetic
recording media and a variety of novel magnetic proper-
ties, such as perpendicular magnetic anisotropy (PMA) [6]
or giant magnetoresistance (GMR) [7]. Some alternatives
to expensive techniques to fabricate templates for arrays of
nanostructures, such as electron beam or imprint litho-
graphy [8,9], include irradiation with heavy ions [10,11] and
chemical etching of polymers (i.e. polycarbonate) or mica
[12], and self-assembling methods for the fabrication of
templates in membranes [13]. Among them, a two-step
anodization process for the fabrication of Al2O3 mem-
branes with hexagonally ordered pores became certainly
interesting [14]. On the other hand, the nuclear track-
etched technology is based on the bombardment of a
polymer film (typically 5–50 mm thick) with heavy ions of
high energy accelerated in a cyclotron. The irradiated film
is then etched in a caustic solution where tracks are
revealed, leading to the formation of pores depending on
the etching time. Membranes with pore diameters as
small as 10 nm are fabricated by this method and are
commercially available with nanopore densities close to
109 pores=cm2.
Fabrication of low-dimensional structures has attracted
a great interest due to their important applications in
ultrahigh-density magnetic recording, biological assays,
chemical sensors and other electronic and optical devices
[1–4]. Many different techniques have been used to produce
nanoscaled magnetic structures. Among them, electrode-
position [5] is a simple and inexpensive bottom–up
technique that involves the fabrication of magnetic
nanowires and templates with a given order. Over the past
few years, electrodeposition has been focused on the
fabrication of self-assembled nanowire arrays and on the
use of different industrial templates as nanoelectrodes,
which are easily filled with magnetic materials. In
technological applications, this offers the interesting
Ã
Corresponding author. Present address: BESSY GmbH, Albert-
Einstein-Strasse 14, 12489 Berlin, Germany. Tel.: +49 1734221844;
fax: +49 63924709.
E-mail address: sbarriga@bessy.de (J. Sanchez-Barriga).
´
0304-8853/$ - see front matter r 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmmm.2006.09.020

ARTICLE IN PRESS
´
J. Sanchez-Barriga et al. / Journal of Magnetism and Magnetic Materials 312 (2007) 99–106
100
If a magnetic field is present in the electrochemical cell
were observed with a metallographic microscope to
confirm the presence of the electrodeposited material. This
method can be used when the nanopores are completely
filled with cobalt and a three-dimensional layer starts to
cover the surface, but not for the observation of the
nanowires. The high resolution SEM characterization of
the membrane surface and nanowires morphology was
performed by a JEOL scanning microscope JSM-6400
working at 24 kV. In order to observe the nanowires, the
deposition was interrupted before the complete filling of
the nanopores. The preparation of the samples for
standard observation involved the partial dissolution of
the polycarbonate membranes in a droplet treatment of
Cl2CH2 at 40 ꢁC. Before membrane dissolution, when the
nanowires are parallel to each other, the crystallographic
structure of Co inside the nanowires was investigated by X-
ray diffraction (XRD) technique in a SIEMENS D-5000
during the fabrication of the nanowires, all the charged
species moving in the electrolyte will experience the effect
of a Lorentz force. In such a scenario, convective currents
are created in the solution that influence the mass transport
rate and affect the diffusion controlled processes. This
effect on the electrodeposition process is known as the
magnetohydrodynamic (MHD) effect [15–18,22–24],
whereas the effect on the crystal structure is known as
the magnetocrystalline effect (ME) [19–21]. In general, the
magnetic field effects are investigated analyzing the
structure of the deposit when the field is applied parallel
and perpendicular to the electrode surface [25,26].
In this work, polycarbonate membranes with nominal
pore diameters of 30 nm are used for the fabrication of
cobalt nanowires with 6 mm length. After the growth
process, the structural and magnetic behavior of the
electrodeposited samples is analyzed by means of scanning
electron microscopy (SEM), X-ray diffraction (XRD) and
vibrating sample magnetometer (VSM). As it has been
demonstrated in previous works on magnetoelectrodepos-
ited nanowires [27,28], superimposing a large magnetic field
during electrochemical deposition influences the growth
mechanism of the nanowires, depending on its orientation,
strength and homogeneity. However, the use of high
intensity fields makes difficult to achieve a proper control
of the hysteresis loops of the samples. In this report we
show how it is possible to control the magnetic properties
of the nanowires fabricated in a 50 Oe magnetic field,
indicating that magnetoelectrolytic systems can exhibit a
different behavior in this regime.
˚
diffractometer with Cu–Ka radiation (l ¼ 1:5418 A),under
working conditions of 40 kV and 30 mA. The field
dependent magnetization hysteresis was measured in a
VSM magnetometer for Co nanowire arrays at a fixed
temperature of 300 K, with a saturation magnetic field of
1 T applied parallel (Hk) and perpendicular (H?) to the
nanowires symmetry axis. For the magnetic characteriza-
tion, a magnetic field of 50 Oe inside kthe electrochemical
cell was previously applied parallel (HCELL) and perpendi-
cular (H?CELL) to the nanowire axis during electrodeposi-
tion by placing the electrolytic system between a pair of
Helmholtz coils supplied with a current of 1 A.
3. Results and discussion
2. Experimental procedure
3.1. Membrane characterization
Cobalt nanowires were electrodeposited inside an array
of empty nanopores of commercially available ion track-
etched polycarbonate membranes. The pores have a
nominal diameter of 30 nm and a fixed length of 6 mm,
which are the nominal values of the whole membrane given
by the membrane manufacturer. All the membranes were
covered with a 250 nm Cu film by sputtering technique,
using a deposition current of 40 mA at 0.05 mbar work
pressure. Before mounting the membrane into the electro-
chemical cell, it was cleaned in deionized water under
ultrasonic agitation for 1 min. The membrane was then
mounted on an electrode leaving a 64 mm2 membrane area
exposed to the electrolyte. Electroplating was done using
an agitation system to prevent hydrogen evolution over the
sample, which is important in order to obtain reproducible
results and a homogeneous growth inside the nanopores.
Cobalt nanowires with 6 mm length were grown in
The properties of the Co nanowire arrays are directly
related to the properties of the nanoporous template, such
as the pore size distribution, surface roughness of the
nanopores and relative pore orientations in the assembly.
The morphological characterization of the polycarbonate
template prior to the electrodeposition process, performed
by SEM technique, is shown in Fig. 1A. Top view images
of the membrane surface with 30 nm nominal pore
diameter were recorded with different magnifications in
order to measure the pore diameter and pore density
distribution. Some defects are observed, probably due to
contamination and stressing effects, as well as a low
ordering degree of pores. The pore density and pore
diameter distributions, shown in Figs. 1B and C, respec-
tively, are calculated after averaging over more than 20
SEM membrane images. The obtained mean pore density is
(6:1 ꢃ 2:4Þ ꢄ 108 nanopores cmꢂ2, which is in good agree-
ment with the value of 6 ꢄ 108 nanopores cmꢂ2 given by
the manufacturer. Due to the reduced lateral resolution, it
is difficult to estimate the apparent nanopore diameter
from the images, and a slightly dispersive mean value of
ð39 ꢃ 21Þ nm is obtained.
polycarbonate membranes in
a mixed solution [6]
(pH ¼ 4:5) of CoSO4 ꢀ 7H2O (252 g/L), H3BO3 (50 g/L)
and NaCl (7 g/L) at 25 ꢁC with a constant applied voltage
of ꢂ0:95 V. A saturated calomel electrode (SCE) was used
as reference in potentiostatic mode, together with a Co film
as counter electrode. After electrodeposition, the samples

ARTICLE IN PRESS
J. Sanchez-Barriga et al. / Journal of Magnetism and Magnetic Materials 312 (2007) 99–106
´
101
(A)
Fig. 2. Experimental development of the current in the electrochemical
cell during the reduction of Co2þ ions in polycarbonate membranes with
30 nm pore diameter, starting with and without an applied external
magnetic field.
(B)
(C)
nanopores up to the membrane surface. The transition
charges are 285, 316 and 254 mC, respectively, with a mean
value of (285 ꢃ 30Þ mC. Taking into account that the
transferred charge is Q ¼ zFNpVp=Vm, where z ¼ 2 for the
divalent Co2þ, F is the Faraday constant, Vp the nanopore
mean volume, Vm the molar volume of Co and Np the total
number of pores, we obtain an averaged mass of ð87 ꢃ
9Þ mg and an electrochemical nanowire diameter of
(74 ꢃ 24Þ nm for all samples. As a second stage, a strong
increase in the electrochemical current is observed at longer
times, indicating a three-dimensional growth on top of the
nanowires: hemispherical caps are formed, which grow in
size until coalescence on the membrane surface occurs. At
this point, the effective electrode area and the electro-
chemical current increase rapidly. If the magnetic field is
absent, the electrodeposition mechanisms in the solution
involve metallic atoms which are mainly bounded to H2O
molecules or form complexes with other ions from the
solution. Applying a potential between the electrodes, the
electric field generated in the solution drives the system out
of electrostatic equilibrium, and the reduction process
takes place upon the displacement of the Co2þ ions in the
direction of the Cu cathode. During their approach to the
electrode area, a convective–diffusion boundary layer of
typically 1 mm thickness is formed near the membrane
surface. In this region, transport phenomena supply the
electrode with ions coming from the solution. Due to the
action of the electric field, the H2O dipoles are oriented
backwards, pointing in the direction of the counter
electrode. At a distance of a few nanometers from the
cathode, the ions are released from the H2O molecules,
leading to the formation of a double Helmholtz layer which
is covered on one side by oriented H2O molecules and on
the other side by Co2þ ions moving over the surface. The
deposition process finishes when the atoms find their
positions of minimum energy on the charged surface inside
the nanopores. In general, the diffusion layer reduces the
effectiveness of the entire process due to the cloud of
Fig. 1. (A) Top view SEM image of an amorphous track-etched
polycarbonate membrane with 30 nm nominal diameter recorded at
24 kV and 23 000ꢄ magnification. (B) and (C): Normalized diameter
and pore density distributions obtained after averaging over more than 20
SEM membrane images for different magnifications. The calculated mean
pore density is (6:1 ꢃ 2:4Þ ꢄ 108 nanoporescmꢂ2 and the measured
apparent pore diameter (39 ꢃ 21Þ nm.
3.2. Chronoamperometric measurements
After membrane characterization, the nanowire length is
controlled by potentiostatic electrodeposition at room
temperature. Fig. 2 shows the experimental development
of the current in the electrochemical cell during the
reduction of Co2þ ions in polycarbonate membranes with
30 nm average pore diameter, starting with and withokut an
external magnetic field of 50 Oe applied parallel (HCELL
)
and perpendicular (H?CELL) to the nanowire axis. Interest-
ingly, it is observed that the presence of a magnetic field in
the electrochemical cell enhances the mass transport rate,
leading to a more pronounced effect when it is applied
along the nanowire axis. In this measurement, two different
stages of the growth process are observed. At t ¼ 0, an
increased electrochemical current of 4.7 mA is measured
for both orientations of the applied magnetic field. After an
initial transient, the electric current reaches a constant
value and the electrodeposition process is balanced. At
t1 ¼ 171 s, t2 ¼ 212 s and t3 ¼ 274 s for HkCELL, HC?ELL and
no applied field, respectively, the nanopores are completely
filled with magnetic material. For all samples fabricated in
this work, nanowires with a mean length close to 6 mm were
grown by stopping the deposition process exactly at these
times, as indicated in Fig. 2. By stopping the growth
process at shorter times, one can fabricate nanowires with
different lengths. The integration of the curves from t ¼ 0
to t1, t2 and t3 gives the total charge needed to fill all the

ARTICLE IN PRESS
´
J. Sanchez-Barriga et al. / Journal of Magnetism and Magnetic Materials 312 (2007) 99–106
102
oriented H2O molecules formed in front of the cathode.
Thermal or mechanical agitation is usually utilized to
liberate the sample from the barrier of molecules through
hydrogen evaporation out of the electrolytic system.
In order to microscopically clarify the enhancement of
the mass transport rate and its dependence on the
orientation of the magnetic field, we will restrict ourselves
to a qualitative interpretation of the magnetoelectrodepo-
sition process in the basics of a magnetically induced
vorticity in the Co2þ ions, caused by the effect of the
Lorentz force on the fluid dynamics (MHD effect).
Although magnetic forces are small compared to other
forces present in an electrochemical reaction, their effect is
non-negligible if they are combined with natural convective
movements in the solution [29,30]. The trajectories of the
ions in the solution are governed by the Navier–Stokes
equation (NSE) [31–33] and some analytical and numerical
approaches for simple geometries and boundary conditions
have been reported [32,34,35]. Detailed studies of the
vortex and turbulence evolution for different boundary
conditions reveal that the vorticity effects are larger at the
electrode–electrolyte interfaces [31]. Therefore, the vorti-
city-based approach can be used to interpret the measured
enhancement of the mass transport rate: the presence of the
magnetic field (for both parallel and perpendicular
orientations) favors ionic collisions, which are responsible
for an increased hydrogen evaporation near the sample.
The additional thermal agitation induced by the vortex
formation causes a reduction of the convective–diffusion
boundary layer due to an energy transfer from the vortexes
to the neighboring molecules. Consequently, the cloud of
H2O molecules is drained out of the nanopores in a more
effective way, allowing the Co ions to reach the Helmholtz
layer.
ions occurs in a plane perpendicular to the membrane
surface, which results in an additional back and forward
movement of the ions that only favors hydrogen evapora-
tion but not the formation of vortexes with axes oriented
along the nanopores. Therefore, we conclude that the
enhancement of the mass transport rate in the presence of a
magnetic field depends on how the additional rotation of
the charged species due to the Lorentz force contributes to
the natural rotation of the fluid near and inside the
nanopores, being more pronounced when the rotation
planes are oriented parallel to the deposition surface.
3.3. Structural characterization
An example of the observed shape of the Co-deposited
nanowires is given in Fig. 3a, in which SEM images at
24 kV and 6000ꢄ of Co nanowires grown without field
(a)
Since we cannot neglect non-zero values of the ion
velocity components in arbitrary directions due to the
complex fluid motion near the membrane surface and
inside the nanopores, the Lorentz force has an effect for
both orientations of the magnetic field, as it is observed in
Fig. 2. Actually, our interpretation is that, depending on
the field direction, the different orientation of the rotation
planes of the ions by means of an induced vorticity in the
system is a crucial parameter that strongly influences the
Co2þ mass deposition rate. When the magnetic field is
applied parallel to the nanowires axis (HkCELL), a rotation
of the charged species in a plane oriented parallel to the
membrane surface will occur, leading to the formation of a
sea of eddies with axes preferentially oriented perpendi-
cular to the membrane plane. Assuming that the nanopores
act as centers for the formation of vortexes, the trajectories
of the ions over the membrane surface will be modified
following the vortex rotation and therefore accelerating
part of the Co2þ ions towards these centers. Inside the
nanopores, these helicoidal currents will favor the trans-
port phenomena and hydrogen evaporation. On the other
hand, when the magnetic field is applied perpendicular to
the nanowires axis (H?CELL), the preferred rotation of the
(c)
(b)
(d)
Fig. 3. SEM images at 24 kV and 6000ꢄ of Co nanowires grown without
field after the partial dissolution of the polycarbonate membrane in
CH2Cl2: (a) while inside the membranes all nanowires are parallel to each
other, after its dissolution they reorganize in different positions over all
the sample area, (b) zoom into image (a): as an example, an isolated
nanowire with 6:78 mm length. (c) and (d): nanowire length and diameter
distributions calculated with SEM.

ARTICLE IN PRESS
J. Sanchez-Barriga et al. / Journal of Magnetism and Magnetic Materials 312 (2007) 99–106
´
103
were recorded after the partial dissolution of the poly-
carbonate membrane in CH2Cl2. The electrodeposition
process was stopped at t3 ¼ 274 s. Whereas all nanowires
are parallel to each other inside the membrane, after its
dissolution they reorganize in different positions all over
the sample area. As an example, an isolated nanowire with
6:78 mm length is shown in Fig. 3b. Analyzing different
SEM images, the mean length obtained from the nanowire
length distribution shown in Fig. 3c is (6:7 ꢃ 2:2Þ mm, which
is in good agreement with the specification of 6 mm given by
the membrane manufacturer. Despite the limited SEM
resolution, it is observed that the nanowires are not
perfectly cylindrical. Fig. 3d shows the nanowire diameter
distribution calculated by SEM. The mean half-length
nanowire diameter is (105 ꢃ 31Þ nm, which differs approxi-
mately in a factor 3 from the membrane manufacturer
specifications. Although SEM images are systematically
overestimating the diameter of the wires, the pore widening
is also confirmed by an apparent pore diameter of
(74 ꢃ 20Þ nm calculated from the electrochemical charge
transference in the system during the fabrication process.
The origin of a widened pore diameter is usually explained
as a consequence of the ion impact during the exposure of
the polycarbonate layer to the ion beam [36–38]. Besides
the primary damage caused by the ion impact, a large
number of secondary ions could affect the neighboring
nanopores, producing a progressive widening of the pore
diameters by a proximity effect, similar to the well-known
proximity effect in e-beam lithography.
(a)
In order to investigate the influence of the magnetic field
on the crystal structure, we have examined the XRD
diffraction patterns of the fabricated samples. Fig. 4a
shows the X-ray diffraction pattern of Co nanowires grown
witkh and without a magnetic field of 50 Oe applied parallel
(HCELL) and perpendicular (H?CELL) to the membrane
plane. During characterization, all nanowires are parallel
to each other and the sputtered Cu layer on one side of the
membrane is still present. It is possible to observe how the
applied magnetic field influences the crystalline structure of
the fabricated nanowires. When the external field is absent,
a Co-FCC peak oriented along the [1 1 1] direction at
44:375ꢁ is dominating the diffraction pattern. A coexistence
of Co-FCC and Co-HCP phases oriented along the [1 1 1]
and [1 0 0] directions respectively is found when the
magnetic field is applied perpendicular to the nanowires
axis. Moreover, a double peak of Co-HCP (0 0 2) and Co-
FCC (1 1 1) phases is found coexisting with a Co-HCP
(1 0 0) phase when the magnetic field is applied parallel to
the nanowire axis (perpendicular to the membrane plane).
Although the Co-HCP (1 0 0) texture appears approxi-
mately at 44:561ꢁ for both orientations of the magnetic
field, the Co-HCP (0 0 2) and Co-FCC (1 1 1) for HCk ELL
appear to be shifted respect to the standard values of
44:473ꢁ and 44:370ꢁ, respectively. For clarification, a zoom
of the double peak is shown in Fig. 4b. The double peak
shows two maxima indicating a mixture of Co-HCP (0 0 2)
and Co-FCC (1 1 1) phases, which appear at approximately
(b)
Fig. 4. (a) X-ray diffraction pattern of Co nanowires grown with and
without an applied magnetic field of 50 Oe parallel and perpendicular to
the membrane plane, (b) Co double peak indicating a mixture of FCC
(1 1 1) and HCP (0 0 2) phases when the field is applied along the
nanowires during growth.
44:406ꢁ and 44:246ꢁ, respectively. The relative distance
between peaks is 0:16ꢁ, according with the standard value
of 0:10ꢁ. The shift and broadening of these Co peaks is
commonly attributed to structural defects or stacking
faults being present in the crystal [39], which are expected
when different phases coexist. No other features are
distinguishable from the background at other angles.
Analyzing the XRD patterns, it is possible to know
the orientations of each crystalline phase respect to the
nanowire axis. Since y–2y geometry has been used, the
X-ray measurements are preferentially sensitive to
the symmetry planes of the structure oriented parallel to
the membrane plane. However, when Co-HCP and Co-
FCC phases coexist, it is not possible to quantify by means
of the peak intensity the amount of cobalt of each phase
present in the nanowires. At zero field, the observed Co-
FCC phase indicates that the [1 1 1] direction is mainly

ARTICLE IN PRESS
´
J. Sanchez-Barriga et al. / Journal of Magnetism and Magnetic Materials 312 (2007) 99–106
104
oriented along the axis of the nanowires, and therefore the
basal (1 1 1) planes are parallel to the membrane plane. On
the other hand, for H?CELL, an in-plane Co-HCP (1 0 0)
texture with (1 0 0) symmetry planes oriented perpendicular
to the nanowire axis is induced. In this case, the Co-HCP c-
axis tends to lie in the membrane plane, reinforcing the in-
plane shape anisotropy. Certainly, other in-plane textures
following the direction of the field would be expected, but
they are not distinguishable in the diffraction patterns.
Finally, for HkCELL, the presence in the double peak of a
Co-HCP (0 0 2) texture (with basal (0 0 2) planes oriented
parallel to the membrane surface) and a Co-FCC (1 1 1)
texture contributes to a reorientation of the c-axis along the
nanowires. The simultaneous appearance of a Co-HCP
(1 0 0) peak in this pattern can be related to inhomogene-
ities in the applied magnetic field caused by the back
reaction of the fluid movement on the field lines that would
be responsible for a misalignment of the dipoles, producing
phases with different orientations to the expected ones. A
further clarification of the observed textures in the crystal-
line structure is only possible in combination with magnetic
measurements. According to this idea, Paulus et al. [40,41]
found that, with no field applied during the growth process,
the direction of the hexagonal c-axis is nearly independent
of the nanowire diameter and the observed change of the
easy axis of magnetization is mainly due to the contribu-
tion of a Co-FCC phase, which takes place when the size of
the nanopores is reduced. Fert et al. [42] argued that this
change is due to a reorientation of the hexagonal c-axis,
and other authors [43,44] showed that there is a coexistence
of two different Co phases with different orientations for a
constant wire diameter. Since the final texture of the
nanowires also depends on the plating procedure, it is still
unclear why the amount of FCC-Co increases with a
decreasing nanowire diameter. Other studies with Co
nanoparticles [45] show that the presence of a HCP phase
or a FCC phase is related to the particle diameter; in that
case, the transition from HCP to FCC was ascribed only to
a size effect due to the lower surface energy of the FCC
phase. Following this idea, we argue that the action of a
weak magnetic field during electrodeposition can partially
influence the surface energy of the crystalline structure, and
this would also explain why a Co-FCC (1 1 1) peak is
always present in the measured spectra. However, other
works in magnetoelectrodeposited nanowires under high
fields up to 5 T [28] show contradictory results, in which
the X-ray diffraction patterns do not seem to agree with the
magnetization measurements. The results shown in the
present work also confirm that the magnetic field has an
influence on the crystallographic structure (ME) and
therefore on the magnetocrystalline anisotropy. We con-
clude that the application of a low intensity magnetic field
is favorable to produce nanowires with an oriented texture
in the direction of the field. This field is responsible for the
alignment of the magnetic dipoles of the deposited ions
along a preferred direction, reducing the surface energy
that is necessary to achieve a particular Co phase.
3.4. Magnetic characterization
The effect of the magnetic field applied during electro-
deposition on the magnetic properties of the fabricated
samples has been investigated by measuring the hysteresis
loops as a function of the orientation of the field during
characterization. This type of measurement allow us to
understand the magnetic behavior of the nanowire array as
a whole and determine the easy axis of magnetization of
individual wires. Firstly, to check if an effective PMA
(parallel to the nanowire axis) is present, we measured the
longitudinal and transverse (in-plane of the membrane)
hysteresis loops for each deposition configuration. The
applied magnetic fields for VSM characterization were
Hkꢅ1 T (out-of-plane) and H?ꢅ1 T (in-plane) for the
parallel and perpendicular directions to the nanowires axis,
respectively. Fig. 5 shows the hysteresis loops measured at
room temperature of Co nanowire arrays fabricated with
and without a magnetic field applied during electrodeposi-
tion parallel and perpendicular to the nanowire symmetry
(a)
(b)
(c)
Fig. 5. In-plane (H?) and out-of-plane (Hk) room temperature hysteresis
loops of Co nanowires grown with and without an applied magnetic field:
(a) Without magnetic field, (b) with a field of 50 Oe applied parallel to the
nanowires axis (HkCELL), and (c) perpendicular to the nanowires axis
(H?CELL). A significant variation of the anisotropy directions is observed as
a function of the orientation of the applied field during the electrodeposi-
tion process.

ARTICLE IN PRESS
J. Sanchez-Barriga et al. / Journal of Magnetism and Magnetic Materials 312 (2007) 99–106
´
105
axis (HkCELL and H?CELL, respectively). To understand these
results, we must take into account that the hysteresis loops
are directly related to the magnetization reversal mechan-
isms and to the different anisotropies dominating the
magnetic behavior of the samples. The total magnetic
anisotropy of the system is given by the interplay between
the shape anisotropy, magnetocrystalline anisotropy and
dipolar interactions among nanowires [46,47]. The magne-
tostatic interactions give rise to an effective stray field
originated between neighboring nanowires, depending on
the mean interwire distance, that adds to the applied field.
The strength of the stray field depends on the geometrical
characteristics of the array [48,49] and has a minor effect
due to the large aspect ratio of the nanowires compared to
the inter-wire spacings characteristic of polycarbonate
membranes. As observed in Fig. 5a, the magnetic hysteresis
loops show a clear difference between the axial and in-
plane loops with larger remanence and coercitivity when
the field is applied along the nanowire axis (Hk). This allow
us to conclude that the magnetic anisotropy is mainly
oriented along the nanowire axis. At zero field, for a Co-
FCC(1 1 1) phase oriented along the nanowire axis, the
magnetocrystalline anisotropy energy density K1 ¼ 6:3 ꢄ
105 erg=cm3 is almost one order of magnitude smaller than
the shape anisotropy energy density p ꢀ M2S ¼ 6:0 ꢄ
106 erg=cm3 [50]. Hence, the effective magnetic anisotropy
of Co-FCC nanowires fabricated without field is domi-
nated by their shape anisotropy, so that the easy axis of
magnetization lies along the nanowire axis for sufficiently
long wires, in agreement with the measured hysteresis loops
in Fig. 5a. On the other hand, when the nanowires were
magnetoelectrodeposited at HkCELL ¼ 50 Oe (Fig. 5b), the
magnetic hysteresis loops show again a clear difference
between the axial and in-plane loops with larger remanence
for the axial loop. In the same way, we can conclude that
the magnetic anisotropy is oriented parallel to the
nanowire axis. Moreover, the remanence and coercitivity
values are even larger than the case in which no field was
applied during electrodeposition. According to the results
of the XRD patterns, the magnetocrystalline anisotropy of
the Co-HCP (0 0 2) and Co-HCP (1 0 0) textures tends to
align the magnetic moments along and perpendicular to the
nanowire axis, respectively. For Co-HCP, the magneto-
crystalline anisotropy energy density K1 ¼ 5 ꢄ 106 erg=cm3
is of the same order of magnitude as the shape anisotropy
energy density p ꢀ M2S ¼ 6:0 ꢄ 106 erg=cm3 [50]. Therefore,
the magnetic anisotropy is determined in this case by the
interplay of both magnetocrystalline anisotropy and shape
anisotropy. Fig. 5b shows how the competition between
both anisotropies results in a stronger effective anisotropy
along the nanowire axis, which leads to the larger
coercitivity and squareness of the nanowire array, as well
as an enhancement of the effective PMA. Finally, the
most interesting situation is shown in Fig. ?5c, when the
nanowires are magnetoelectrodeposited at HCELL ¼ 50 Oe.
The hysteresis loops show an induced radial anisotropy
perpendicular to the nanowire axis due to the effect of the
applied magnetic field. The c-axis of the Co-HCP (1 0 0)
phase is preferentially oriented perpendicular to the
nanowire axis, implying that the magnetocrystalline
anisotropy favorable for the c-axis alignment dominates
over the shape anisotropy and, in consequence, the
effective magnetic anisotropy of the whole sample is
oriented perpendicular to the nanowire axis. Notice the
dramatic decrease of the coercitivity for both in-plane and
out-of-plane hysterkesis loops in comparison with the arrays
deposited with HCELL and without field. The obvious
decrease of the magnetic anisotropy along the nanowire
axis results in a lower coercitivity and larger remanence
along the membrane plane. Although the quantitative
relation among these magnitudes is still an open question,
the observed decrease can be due to a more effective
alignment of the c-axis perpendicular to the nanowires.
However, other effects as a larger degree of polycrystalli-
nity, crystalline defects and impurities, or a c-axis not
strictly oriented perpendicular to the nanowire axis may
exist [50]. Nevertheless, this result demonstrates that it is
possible to switch the easy axis of the whole nanowire array
over to the perpendicular direction by simply applying a
low intensity magnetic field along that direction during
electrodeposition.
4. Conclusions
Arrays of Co nanowires have been fabricated by
magnetoelectrodeposition inside the nanopores of track-
etched polycarbonate membranes with a nominal diameter
of 30 nm and 6 mm length. An enhancement of the
mass deposition rate due to the action of the applied
magnetic field has been observed by measuring
the experimental development of the electrochemical
current in the electrochemical cell during the fabrication
process. While Co nanowires deposited without magnetic
field present a Co-FCC structure oriented along the
nanowire symmetry axis, nanowires electrodeposited in
an external field of 50 Oe applied parallel and perpendi-
cular to the membrane surface exhibit induced Co-HCP
phases with different orientations depending on the
direction of the external field. The magnetic characteriza-
tion shows that Co nanowires fabricated without field have
an easy axis oriented along the nanowires due to their
shape anisotropy. The hysteresis loops are in good
agreement with the X-ray diffraction measurements, and
reveal how perpendicular or radial anisotropies have been
switched depending on the orientation of the applied
magnetic field during electrodeposition. These results can
be used for tailoring the magnetic properties of Co
magnetic nanowires electrodeposited in nanoporous tem-
plates. The observed effects can also be extended to other
type of nanostructures or other materials, combining
electrodeposition with advanced techniques as electron
beam lithography.

ARTICLE IN PRESS
´
J. Sanchez-Barriga et al. / Journal of Magnetism and Magnetic Materials 312 (2007) 99–106
106
References
[25] O. Lioubashevski, E. Katz, I. Willner, J. Phys. Chem. 100 (1996)
5913.
[26] Y. Ganesh, D. Vijayaraghavan, V. Lakshminarayanan, Appl. Surf.
Sci. 240 (2005) 286.
[1] L.W. Wang, Y. Liu, Z. Zhang, Handbook of Nanophase and
Nanostructured Materials, Kluwers Academic Publishers, Dordrecht,
2002.
[27] H.W. Kwon, S.K. Kim, Y. Jeong, J. Appl. Phys. 87 (9) (2000) 6185.
[28] N.B. Chaure, P. Stamenov, F.M.F. Rhen, J.M.D. Coey, J. Magn.
Magn. Mater. 290–291 (2005) 1210.
[2] N. Sulitanu, Electrochemical deposition of novel nanostructured
magnetic thin films for advanced applications, 2002.
[3] T. Osaka, Electrochem. Acta 45 (2000) 3311.
[4] Q.A. Pankhurst, J. Conolly, S.K. Jones, J. Dobson, J. Phys. D 36
(2003) R167.
[29] J.M.D. Coey, G. Hinds, J. Alloy Compd. 326 (2001) 238.
[30] J.E. Wegrowe, D. Kelly, A. Franck, S.E. Gilbert, J.Ph. Ansermet,
Phys. Rev. Lett. 82 (1999) 3681.
[31] T.Z. Fahidy, J. Appl. Electrochem. 32 (2002) 551.
[32] T.Z. Fahidy, Progr. Surf. Sci. 68 (2001) 155.
[33] L.D. Landau, E.M. Lifschitz, Course of Theoretical Physics, vol. 6:
Fluid Mechanics, Butterworth-Heinemann, London, 1995.
[34] O. Lioubashevski, E. Katz, I. Willner, J. Phys. Chem. 100 (1996)
5778.
[5] D. Pletcher, F.C. Walsh, Industrial Electrochemistry, Chapman and
Hall, Boca Raton, FL, 1990.
[6] T.M. Whitney, J.S. Jiang, P.C. Searson, C.L. Chien, Science 261
(1993) 1316.
[7] M.N. Baibich, J.M. Broto, A. Fert, F. Nguyen Van Dau, E. Petroff,
P. Etienne, G. Creuzet, A. Friederich, J. Chazelas, Phys. Rev. Lett. 61
(1988) 2474.
[35] N. Leventis, M. Chen, X. Gao, M. Canalas, P. Zhang, J. Phys. Chem.
B 102 (1998).
[36] I. Chlebny, et al., Nano-Struct. Mater. 2 (1993) 637.
[8] J.F. Smyth, S. Schultz, D. Kern, H. Schmid, Appl. Phys. 63 (1988)
4237.
[37] Schonenberger, et al., J. Phys. Chem. B 101 (28) (1997).
¨
[38] M. Hatzakis, IBM J. Res. Dev. 32 (1988) 441.
[9] S.Y. Chou, P.R. Krauss, L. Kong, J. Appl. Phys. 79 (1996) 6101.
[10] T. Devolder, C. Chappert, Y. Chen, E. Cambril, H. Bernas,
[39] R.L. Snyder, J. Fiala, H.J. Bunge, Defect and Microstructure
Analysis by Diffraction, Oxford Science Publications, Oxford, 1999.
[40] P.M. Paulus, F. Luis, M. Kroll, G. Schmid, L.J. De Jongh, J. Magn.
¨
J.P. Jamet, J. Ferre
[11] T. Devolder, C. Chappert, Y. Chen, E. Cambril, H. Launois, H.
Bernas, J. Ferre, J.P. Jamet, J. Vac. Sci. Technol. B 17 (1999) 3177.
´
, Appl. Phys. Lett. 74 (1999) 3383.
´
Magn. Mater. 224 (2001) 180.
[41] M. Kroll, W.J. Blau, D. Grandjean, R.E. Benfield, F. Luis,
¨
[12] L. Sun, P.C. Searson, C.L. Chien, Appl. Phys. Lett. 74 (1999) 2803.
[13] E. Huang, L. Rockford, T.P. Russell, C.J. Hawker, Nature 395
(1998) 757.
P.M. Paulus, L.J. de Jongh, J. Magn. Magn. Mater. 249 (2002) 241.
[42] A. Fert, L. Piraux, J. Magn. Magn. Mater. 200 (1999) 338.
[43] G.J. Strijfers, J.H.J. Dalderop, M.A.A. Broeksteeg, H.J.M. Swagten,
W.J.M. de Jonge, J. Appl. Phys. 86 (9) (1999) 5141.
[14] H. Masuda, K. Fukuda, Science 268 (1995) 1466.
[15] J.C. Shannon, Z.H. Gu, T.Z. Fahidy, J. Electrochem. Soc. 144 (1997)
L314.
[44] R.E. Benfield, D. Grandjean, M. Kroll, R. Pugin, T. Sawitoski,
¨
G. Schmid, J. Phys. Chem. B 105 (10) (2001) 1961.
[16] T.Z. Fahidy, Prog. Surf. Sci. 68 (2001) 155.
[17] T.Z. Fahidy, J. Appl. Electrochem. 13 (1983) 553.
[18] O. Devos, A. Olivier, J.P. Chopart, O. Aaboubi, G. Maurin,
J. Electrochem. Soc. 145 (1998) 401.
[45] O. Kitakami, H. Sato, Y. Shimada, F. Sato, M. Tanaka, Phys. Rev. B
56 (1997) 13849.
[46] L. Clime, P. Ciureanu, A. Yelon, J. Magn. Magn. Mater. 297 (2006)
60.
[19] S. Bodea, L. Vignon, R. Ballov, P. Molho, Phys. Rev. Lett. 83 (1999)
2612.
[47] M. Va
J. Magn. Magn. Mater. 294 (2005) 174.
[48] T.G. Sorop, K. Nielsch, P. Goring, M. Kroll, W. Blau,
zquez, K. Pirota, J. Torrejon, D. Navas, M. Hernandez-Velez,
´ ´ ´ ´
[20] D.Y. Li, J.A. Szpunar, Electrochem. Acta 42 (1997) 37.
[21] H. Matsushima, T. Noria, I. Mogi, Y. Ito, Surf. Coatings Technol.
179 (2004) 245.
¨
¨
R.B. Wehrspohn, U. Gosele, L.J. de Jongh, J. Magn. Magn. Mater.
¨
272–276 (2004) 1656.
[22] T.R. Nı Mhıochain, J.M.D. Coey, J. Magn. Magn. Mater. 226–230
´ ´ ´
(2001) 1281.
[49] J.U. Cho, J.-H. Wu, J.H. Min, S.P. Ko, J.Y. Soh, Q.X. Liu,
Y.K. Kim, J. Magn. Magn. Mater. 303 (2006) e281.
[50] F. Li, T. Wang, L. Ren, J. Sun, J. Phys. Condens. Matter 16 (2004)
8056.
[23] M. Uhlemann, A. Krause, A. Gebert, J. Electroanal. Chem. 577
(2005) 19.
[24] A. Krause, C. Hamann, M. Uhlemann, A. Gebert, L. Schultz,
J. Magn. Magn. Mater. 290–291 (2005) 261.