Electrochemical control and selection of the structural and magnetic properties of cobalt nanowires

Article abstract of DOI:10.1063/1.1866636

In this letter we present a convenient way of controlling the direction of the uniaxial magnetocrystalline anisotropy in arrays of electrodeposited hcp Co nanowires. Combining electron microscopy and ferromagnetic resonance measurements, it is shown that using an appropriate pH of the electrolytic solution, the hcp c axis can be oriented parallel or perpendicular to the wires axes simply by changing the deposition current density or deposition rate. This reorientation of the c axis leads to a drastic change in overall magnetic anisotropy as the crystal anisotropy either competes for perpendicular oriented c axis or adds to the shape anisotropy for parallel oriented c axis.

Full text of DOI:10.1063/1.1866636

Electrochemical control and selection of the structural and magnetic
properties of cobalt nanowires
M. Darques, L. Piraux, A. Encinas, P. Bayle-Guillemaud, A. Popa et al.
Citation: Appl. Phys. Lett. 86, 072508 (2005); doi: 10.1063/1.1866636
View online: http://dx.doi.org/10.1063/1.1866636
View Table of Contents: http://apl.aip.org/resource/1/APPLAB/v86/i7
Published by the American Institute of Physics.
Additional information on Appl. Phys. Lett.
Journal Homepage: http://apl.aip.org/
Journal Information: http://apl.aip.org/about/about_the_journal
Top downloads: http://apl.aip.org/features/most_downloaded
Information for Authors: http://apl.aip.org/authors
Downloaded 12 Oct 2012 to 136.159.235.223. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions

APPLIED PHYSICS LETTERS 86, 072508 ͑2005͒
Electrochemical control and selection of the structural
and magnetic properties of cobalt nanowires
M. Darquesa͒ and L. Piraux
Unité PCPM, Université Catholique de Louvain, Place Croix du Sud 1, B-1348, Louvain-la-Neuve, Belgium
A. Encinas
Instituto de Física, Universidad Autónoma de San Luis Potosí, Av. Manuel Nava 6, Zona Universitaria,
78290 San Luis Potosí, S.L.P., Mexico
P. Bayle-Guillemaud, A. Popa, and U. Ebels
CEA-DRFMC, 17 rue des Martyrs, 38054 Grenoble, Cedex 9, France
͑Received 21 June 2004; accepted 13 January 2005; published online 10 February 2005͒
In this letter we present a convenient way of controlling the direction of the uniaxial
magnetocrystalline anisotropy in arrays of electrodeposited hcp Co nanowires. Combining electron
microscopy and ferromagnetic resonance measurements, it is shown that using an appropriate pH of
the electrolytic solution, the hcp c axis can be oriented parallel or perpendicular to the wires axes
simply by changing the deposition current density or deposition rate. This reorientation of the c axis
leads to a drastic change in overall magnetic anisotropy as the crystal anisotropy either competes for
perpendicular oriented c axis or adds to the shape anisotropy for parallel oriented c axis. © 2005
American Institute of Physics. ͓DOI: 10.1063/1.1866636͔
Among the different magnetic nanowires produced until
now, cobalt has been the focus of much attention motivated
by its large crystal anisotropy when growth occurs in the hcp
structure. Indeed, numerous reports have shown that elec-
trodeposition of Co nanowires leads to a “polycrystalline”
structure in which hexagonal close packing ͑hcp͒ is the usual
structure for most grains. Under usual deposition conditions,
these grains exhibit preferred orientation of the hexagonal
axis within a few degrees normal to the wire axis and are
randomly distributed.1,2 In this case, the large magnetocrys-
talline anisotropy competes directly with shape anisotropy
which tends to align the magnetization along the wire axis,
giving rise to a lowering of the effective anisotropy. Further-
more, the coexistence of varying amounts of the two hcp and
fcc cobalt phases was found by several groups.3–5 These ob-
servations constitute a major drawback in the use of Co
nanowire arrays in applications such as perpendicular record-
ing media, where an alignment of the c axis along the wire
axis is desired. This underlines the importance of controlling
the microstructure and in particular the magnetic easy axis of
the electrodeposited nanowires.
Though a thorough understanding of the mechanism par-
ticipating in the formation of Co films by electrodeposition is
still lacking, the phase selection as related to deposition con-
ditions or substrate influence has been studied in detail.6–8
From these studies, the electrolytic bath acidity, or pH, is
known to be the most important parameter in determining the
structure of electrodeposited cobalt. Other relevant param-
eters of the operative conditions in cobalt electrocrystalliza-
tion, though minor ones compared to the pH, are the current
density and temperature of deposition.9
hcp c axis along the wires axes could be obtained for pH
above 6.0.10 In contrast to this previous study, we show here
that besides the pH, the deposition rate, or current density,
strongly affects the growth process of the cobalt nanowires.
In particular, it leads to a significant shift of the critical pH
range where the reorientation of the c axis from a perpen-
dicular to a longitudinal direction occurs. As a consequence,
for a limited pH range ͑5–6͒, it is possible to selectively
obtain high quality Co nanowires with either a parallel or a
perpendicular orientation of the c axis only by changing the
deposition current. This makes it possible to control in a very
simple way the overall magnetic anisotropy.
Arrays of Co nanowires are grown by electrodeposition
into the cylindrical pores of lab-made track-etched polycar-
bonate membranes using a standard three-probe technique.
The diameter of the pores is 30 nm with membrane porosity,
P, of 3.5% ͑5ϫ109 pores per cm2͒. Prior to deposition a
Cr ͑20 nm͒/Au ͑300 nm͒ layer is evaporated on one face of
the membrane to serve as a cathode. The electrolyte is
250 g/l CoSO4, 7H2O and 30 g/l H3BO3, the pH of the as-
prepared solution is 4.0 and adjusted with diluted NaOH.
Galvanostatic room temperature deposition is done with cur-
rent densities set from 5 to 50 mA/cm2, over a constant
membrane surface of 1.2 cm2. Structural characterization has
been done by transmission electron microscopy ͑TEM͒ after
dissolving the polycarbonate membrane and spreading of the
wires onto a grid.
As shown in a previous report,10 the average, or domi-
nant orientation, of the c axis induces significant changes in
the total effective anisotropy of the system. Therefore, the
effects of the deposition current on the electrocrystallization
characteristics have been determined by measuring the
uniaxial crystal anisotropy field HU in arrays of nanowires as
a function of both the deposition current and the pH of the
electrolyte. For this purpose, ferromagnetic resonance
͑FMR͒ measurements were carried out using a microstrip
transmission line.10
In a previous study our group has shown the role of the
pH for the control of the orientation of the hcp c axis in
arrays of Co nanowires. In particular a reorientation of the
a͒
Author to whom correspondence should be addressed; electronic mail:
darques@pcpm.ucl.ac.be
0003-6951/2005/86͑7͒/072508/3/$22.50
86, 072508-1
© 2005 American Institute of Physics
Downloaded 12 Oct 2012 to 136.159.235.223. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions

072508-2
Darques et al.
Appl. Phys. Lett. 86, 072508 ͑2005͒
FIG. 1. The value of magnetocrystalline anisotropy field as a function of the
pH for 30 nm Co nanowires electrodeposited at current densities of 5 and
50 mA/cm2.
At a given constant frequency f, the ferromagnetic reso-
nance conditions when the magnetic field is swept parallel to
the wires are:
͑i͒ c axis and, hence, magnetocrystalline easy axis parallel to
the wire axis ͑HUϾ0͒:
f
= HR + 2␲MS͑1 − 3P͒ + HU,
͑1͒
␥
͑ii͒ c axis and, hence, magnetocrystalline easy axis perpen-
dicular to the wire axis ͑HUϽ0͒:
FIG. 2. FMR spectra with a magnetic field applied parallel to the wires for
30 nm Co nanowires grown ͑a͒ at pH 5.2 and current densities of 5, 30, and
50 mA/cm2 recorded at 43 GHz and, ͑b͒ at pH 5.0 and 6.4 with current
densities of 5 and 50 mA/cm2 recorded at 35 GHz.
2
f
= ͓H + 2␲M ͑1 − 3P͒ + H ͔
ͩ ͪ
R
S
U
␥
ϫ͓HR + 2␲MS͑1 − 3P͔͒.
͑2͒
than at high currents. At pH 5.2, HU reaches a maximum
value of +4.6 kOe, and then decreases slowly as the pH in-
creases until it reaches a value of 2 kOe at a pH of 6.6.
As suggested by these results ͑Fig. 1͒, the effect of the
current is to shift the pH value at which the structural and
magnetic changes take place. Specifically, the lowering of
the deposition current shifts this transition to lower pH val-
ues.
In Eqs. ͑1͒ and ͑2͒, ␥͑Co͒=3.05 GHz/kOe, HR is the reso-
nance field, 2␲MSϳ8.8 kOe is the shape anisotropy field
͓MS͑Co͒ϳ1370 emu/cm3͔ and 6␲MSP is the dipolar cou-
pling field between the wires ͑ϳ900 Oe for P=3.5%͒.10
Therefore, FMR studies provide a direct quantification of the
anisotropy crystal field.
Measurements were performed at room temperature by
sweeping down the magnetic field applied parallel to the
wires from 10 kOe to zero, at different constant excitation
frequencies up to 50 GHz, see Ref. 10 for further details. The
crystal anisotropy field is obtained by fitting expressions ͑1͒
and ͑2͒ once the resonance field has been determined at dif-
ferent frequencies.11
Another remarkable and important property derived
from Fig. 1 is that, in a limited pH range, both structural
phases ͑c axis parallel and perpendicular to the wires͒ can be
obtained from the same electrolyte simply by changing the
deposition current. This is illustrated in Fig. 2͑a͒ which
shows the resonance spectra measured in two Co nanowire
arrays deposited from a pH 5.2 electrolyte using high and
low currents ͑indicated in the figure͒. As can be observed, by
changing the current density from 50 down to 5 mA cm−2,
the resonance field can change as much as 8 kOe, which is
consistent with a dominant fraction of hcp Co grains having
their c axis perpendicular ͑high current͒ and parallel ͑low
current͒ to the wires axes. Figure 2͑a͒ also shows an absorp-
tion curve measured on a sample grown at an intermediate
deposition current ͑30 mA cm−2͒. In this case, a much
broader peak appears as a clear signature of the inhomoge-
neous crystallographic structure of the so-prepared Co
nanowires. In these same lines, and in agreement with the
results in Fig. 1, below and above a pH of 5.0 and 6.0,
respectively, the orientation of the c axis is independent of
the deposition current. This is shown in Fig. 2͑b͒ for high pH
͑6.4͒, and low pH ͑5.0͒. To be noted is that for the pH 5.0
samples, the resonance fields for the two deposition currents
Figure 1 shows the variation of the crystal anisotropy
field as a function of the pH for samples deposited using a
high current density of 50 mA cm−2 ͑circles͒ and a low cur-
rent density of 5 mA cm−2 ͑squares͒.
For both deposition currents, the overall dependence of
the crystal anisotropy field is similar. In particular, HU is
negative at low pH and positive at high pH, which is con-
sistent with a dominant fraction of grains having ͑a͒ the c
axis perpendicular ͑HUϽ0͒ and, ͑b͒ the c axis parallel ͑HU
Ͼ0͒ to the wires axes. However, the change in sign for HU
takes place around different pH values depending on the
deposition current density. At higher deposition currents, the
crystal anisotropy fields are in the range of −4.0 to
−5.5 kOe for pH values between 4.0 and 5.5. Above this pH
value, HU rapidly increases and reaches values around
+2 to +3 kOe for a pH above 6.0. For low deposition cur-
rents, the increase in HU, takes place at considerable lower
pH values, around 5.0, and the transition is also more abrupt
Downloaded 12 Oct 2012 to 136.159.235.223. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions

072508-3
Darques et al.
Appl. Phys. Lett. 86, 072508 ͑2005͒
responds closely to the earlier mentioned upper bound. Such
large values are consistent with the very good structural
quality demonstrated by our TEM studies. In contrast, values
for HU are well below the maximum expected value when
the c axis is parallel to the wires. Furthermore, a gradual
decrease of HU with increasing pH clearly appears. This be-
havior can be associated to the influence on crystalline
growth by adsorbed hydrolysis products ͑such as Co hydrox-
ides͒ at the cathode when the pH is sufficiently high. Another
possible reason for the reduction of HU are stacking faults,
which might increase in number in the high pH samples.14
However, further structural characterization is required to
provide a better understanding on the relation between the
growth, crystalline structure and the value of the magneto-
crystalline anisotropy field.
In conclusion, the crucial role of the deposition current
on the preferential orientation of the hcp c axis in Co nanow-
ires with respect to the wire axis has been shown. In particu-
lar, both structural phases, hcp Co with the c axis parallel
and the c axis perpendicular to the wires, can be obtained
from a single electrolyte by changing the deposition current
in a limited pH range ͑5.0–5.5͒ by one order of magnitude.
This reorientation of the c axis leads to changes in sign of the
crystalline anisotropy field. This allows not only to chose the
desired orientation of the c axis, but also to simply grow
wires in which both phases are alternated, providing a simple
and convenient way to control the magnetic properties of
electrodeposited cobalt nanowire arrays to a large extent.
The authors would like to thank R. Legras and E. Ferain
for providing the polycarbonate membrane samples used in
this study, Th. Fournier for fabrication of Si/SiN membranes
used in the TEM experiments, and Dr. L. Vila for his valu-
able help in the experiment. This work was supported in part
by the European Program Contract No.: HPRN CT1999
00150 as well as by the Belgian Science Policy through the
Interuniversity Attraction Pole Program PAI ͑5/1/1͒. One of
the authors ͑A.E.͒ thanks CONACyT for financial support
through Grant No. J-40426.
FIG. 3. TEM bright field images and corresponding diffraction patterns of
two 30 nm Co nanowires electrodeposited at pH=5.2 and current densities
of ͑a͒ 5 and ͑b͒ 50 mA/cm2.
differ, but these values are consistent with a c axis oriented
perpendicular to the wire axis.
1J. L. Maurice, D. Imhoff, P. Etienne, O. Durand, S. Dubois, L. Piraux,
J.-M. George, P. Galtier, and A. Fert, J. Magn. Magn. Mater. 184, 1
͑1998͒.
The dependence of the hcp c-axis orientation on the
deposition current is corroborated by TEM investigations as
shown in Figs. 3͑a͒ and 3͑b͒ for Co wires deposited at a pH
of 5.2. In the case of low current density ͓Fig. 3͑a͔͒, the TEM
bright field and diffraction images clearly show an orienta-
tion of the c axis oriented parallel to the wire axis ͑+/−5°͒.
Similarly, in the case of high current density ͓Fig. 3͑b͔͒, the
TEM bright field and diffraction images clearly show an ori-
entation of the c axis oriented perpendicular to the wire axis
͑+/−5°͒. Although only short sections are shown in the im-
ages, no grain boundaries were observed along longer sec-
tions of the ten wires that have been examined in detail.
As pointed out in previous works involving electroplated
Co films, these structural changes can be attributed to the
influence of absorbed hydrolysis products at the cathode as
well as to evolution of hydrogen, which are known to depend
on the pH and the plating current.9,12 For bulk hcp Co, the
magnetocrystalline anisotropy has an upper bound of
7.8 kOe13 ͑K=5.3ϫ106 erg/cm3 and HU=2K/MS͒. It is
noted that for perpendicularly oriented c axis, the maximum
crystal anisotropy field found in this study ͑ϳ7.8 kOe͒ cor-
2R. M. Metzger, V. V. Konovalov, M. Sun, T. Xu, G. Zangari, B. Xu, M.
Benakli, and W. D. Doyle, IEEE Trans. Magn. 36, 30 ͑2000͒.
3V. Scarani, B. Doudin, and J.-P. Ansermet, J. Magn. Magn. Mater. 205,
241 ͑1999͒.
4G. J. Strijkers, J. H. J. Dalderop, M. A. A. Broeksteed, H. J. M. Swagten,
and W. J. M. de Jonge, J. Appl. Phys. 86, 5141 ͑1999͒.
5M. Kröll, W. J. Blau, D. Grandjean, R. E. Benfield, F. Luis, P. M. Paulus,
and L. J. de Jongh, J. Magn. Magn. Mater. 249, 241 ͑2002͒.
6S. Nakahara and S. Mahajan, J. Electrochem. Soc. 127, 283 ͑1980͒.
7T. A. Tochitskii, A. V. Boltushkin, and V. G. Shadrov, Russ. J. Electro-
chem. 31, 178 ͑1995͒.
8I. M. Croll, IEEE Trans. Magn. 23, 59 ͑1987͒.
9T. Chen and P. Cavallotti, Appl. Phys. Lett. 41, 205 ͑1982͒.
10M. Darques, A. Encinas, L. Vila, and L. Piraux, J. Phys. D 37, 1411
͑2004͒.
11In a first fit, Hu was obtained using Eq. ͑1͒. When Hu turned out to be
negative, the fit was repeated using Eq. ͑2͒.
12T. Cohen-Hyams, W. D. Kaplan, and J. Yahalom, Electrochem. Solid-State
Lett. 5, C75 ͑2002͒.
13E. Du Tremolet de Lacheisserie, in Magnétisme, edited by J. Bornarel
͑Presses Universitaires de Grenoble, Grenoble, 1999͒, p. 168.
14B. Bian, W. Yang, D. Laughlin, and D. Lambeth, IEEE Trans. Magn. 37,
1456 ͑2001͒.
Downloaded 12 Oct 2012 to 136.159.235.223. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions