Investigation of thick cobalt films electrodeposited on gold substrates

Article abstract of DOI:10.1016/j.cplett.2012.06.008

We have investigated the morphological and magnetic structures of thick cobalt films obtained by electrodeposition on polycrystalline gold substrates. The deposits became more and more inhomogeneous with increasing deposit thickness. The morphological str

Full text of DOI:10.1016/j.cplett.2012.06.008

Chemical Physics Letters 542 (2012) 117–122
Contents lists available at SciVerse ScienceDirect
Chemical Physics Letters
journal homepage: www.elsevier.com/locate/cplett
Investigation of thick cobalt films electrodeposited on gold substrates
Witold Szmaja a, , Witold Kozłowski , Krzysztof Pola n´ ski , Józef Balcerski , Michał Cichomski ,
⇑
a
a
a
b
b
c
c
Jarosław Grobelny , Marek Zieli n´ ski , Ewa Mi e˛ ko s´
a
Department of Solid State Physics, University of Łód ´z , Pomorska 149/153, 90-236 Łód ´z , Poland
Department of Materials Technology and Chemistry, University of Łód ´z , Pomorska 163, 90-236 Łód ´z , Poland
Department of Inorganic and Analytical Chemistry, University of Łód ´z , Tamka 12, 91-403 Łód ´z , Poland
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
We have investigated the morphological and magnetic structures of thick cobalt films obtained by elec-
trodeposition on polycrystalline gold substrates. The deposits became more and more inhomogeneous
with increasing deposit thickness. The morphological structure of the films consisted of nanocrystalline
grains roundish in shape and having sizes of the order of 100 nm. The magnetic domains were of the
order of 500 nm in size and exhibited sufficiently high perpendicular anisotropy. The deposits were found
to possess mainly the hexagonal close-packed crystalline structure with sufficiently strong [0001] pre-
ferred orientation. The average domain size slightly increased with decreasing average grain size.
Ó 2012 Elsevier B.V. All rights reserved.
Received 6 March 2012
In final form 5 June 2012
Available online 15 June 2012
1
. Introduction
Magnetic films attract much attention from both fundamental and
evaporation on glass [14], silicon [15] and NaCl [14] substrates,
by sputtering on glass [16] and silicon [16,17] substrates, by chem-
ical vapor deposition on glass [18] and silicon [19] substrates, by
electrodeposition on platinum [20], copper [20,21], gold [22], sili-
con [23] and graphite [21] substrates.
technological points of view. On the fundamental side, they exhibit dif-
ferent magnetic properties, such as magnetic anisotropy, magnetic
microstructure, coercivity and magnetoresistance, depending on their
thickness, chemical composition, morphological structure, crystallo-
graphic structure and preparation conditions. From the technological
point of view, magnetic films find a wide range of applications in the
areas of magnetic and magneto-optic recording and magnetic sensors.
In particular, cobalt and cobalt-based films, deposited by vari-
ous techniques on different substrates, have been intensively
investigated in recent years. This appears to be related to the fact
that depending on the film thickness as well as the preparation
method and conditions used, these films exhibit a very wide range
of magnetic properties, and consequently a very wide range of
applications. For example, cobalt and cobalt-based films are used
in spintronic devices such as giant magnetoresistance multilayers,
spin valves, magnetic tunnel junctions and magnetic random ac-
cess memories [1,2], in micro/nanoelectromechanical systems
Electrodeposition is a film deposition technique which offers a
number of advantages, such as precisely controlled near room tem-
perature operation, low-energy requirements, rapid deposition
rates, capability to handle complex geometries, low cost and sim-
ple scale-up with easily maintained equipment. In addition, the
properties of materials can be tailored by controlling solution com-
positions and deposition parameters. Because of these advantages,
electroplated materials are widely used for the fabrication of mag-
netic recording heads and MEMS/NEMS [3,4].
This Letter presents a study of thick cobalt films electrodepos-
ited on gold substrates. The morphological structure of the films
was observed using scanning electron microscopy (SEM) and
atomic force microscopy (AFM), while the magnetic structure
was imaged by magnetic force microscopy (MFM). The morpholog-
ical structure of electrodeposited cobalt films has been extensively
investigated in the past (mainly by SEM), but observations of their
magnetic structure are in fact very rare (see a review article [24]).
To the authors’ knowledge, only recently has the magnetic struc-
ture of electrodeposited cobalt been studied in Refs. [23,25–27]
by a high-resolution technique such as MFM.
(
MEMS/NEMS) such as actuators, sensors, motors, generators and
frictionless microgears [3,4], and cobalt-based films are used as
high-density magnetic recording materials [5,6].
Recently, studies have been performed of cobalt films obtained
by pulsed laser deposition on sapphire [7], silicon [8], carbon [8]
and glass [9] substrates, by electron beam evaporation on GaAs
[
10], sapphire [11] and silicon [12] substrates, by electron-beam
glancing angle deposition on silicon substrates [13], by thermal
2. Experimental
Thick cobalt films were prepared by electrodeposition using a
three-electrode system. The system consisted of a polycrystalline
gold plate as the working electrode with the surface area of
⇑
Corresponding author. Fax: +48 42 6655137.
E-mail address: witoldszmaja@uni.lodz.pl (W. Szmaja).
0
009-2614/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cplett.2012.06.008

1
18
W. Szmaja et al. / Chemical Physics Letters 542 (2012) 117–122
2
0
.1 cm , an auxiliary platinum mesh electrode of much larger sur-
face area and a saturated calomel reference electrode. More details
on the system used are given in Ref. [28]. The working electrode
was successively prepared mechanically (with the use of sandpa-
pers, down to 2000 grade), chemically (using a chromic acid clean-
ing mixture) and electrochemically (polarization in a stock solution
of 0.1 M H
sweep rate of 50 mV/s). The electrolyte solution contained 0.2 M
CoSO O, 0.6 M Na O and 0.1 M H SO . The solution
2
SO
4
, within the potential range of ꢀ1 to +2 V, with a
4
ꢁ7H
2
3
C
6
H
5
O
7
ꢁ2H
2
2
4
pH was 4.5. The cobalt films were electrodeposited at room tem-
2
perature and at constant current density of 20 mA/cm . The films
of different thickness were obtained by varying the deposition
time.
The investigation of the morphological structure of the films
was carried out by SEM using a Tescan Vega 5135 MM instrument
and by AFM using an Omicron instrument. The scanning electron
microscope was equipped with conventional tungsten filament
and images were recorded in the backscattered electron mode at
3
0 keV primary beam energy. AFM images were taken in the con-
tact mode using Omicron silicon cantilevers. The MFM study of
the magnetic structure of the films was made with a NT-MDT
instrument operated in the dynamic mode using MikroMasch sili-
con cantilevers with tips magnetized along the tip axis, which was
perpendicular to the specimen surface. In this case, MFM senses
the vertical component of the derivative of the force between the
specimen and the tip [29,30]. The image signal was detected as
the phase or amplitude shift of an oscillating cantilever. The tips
used were coated with a FeCoNi film of about 60 nm in thickness.
The coercivity of the tips was approximately 8 kA/m.
The MFM measurements were carried out by the two-pass
method. For each raster line, the first pass was made very close
to the specimen surface and yielded knowledge of the surface
topography (AFM). The second pass then followed the recorded
topography but at an increased scan height. Of course, the tip-spec-
imen distance must be large enough to eliminate (or minimize) the
short-range forces that provided the topographic contrast. Then in
the second pass the tip is affected only (mainly) by long-range
magnetic forces and the corresponding MFM image of the speci-
men surface is obtained. All the MFM images shown in this Letter
were collected with a tip-specimen distance of 100 nm.
3
. Results and discussion
SEM was used to determine the thicknesses of the studied co-
balt films. On the basis of SEM images recorded at various places
of the cross-sectional area of the specimens, the thicknesses of
the cobalt deposits under investigation were determined to be
(
14 ± 1) lm, (28 ± 2) lm and (55 ± 5) lm. The first two films were
relatively uniform in thickness, in contrast to the third film. None
of the films detached from the substrate despite the fact that they
were thick (up to 55 lm in thickness), thus the polycrystalline gold
substrates may be considered suitable.
Figure 1a–c show low magnification SEM images of the surface of
the cobalt deposits 14, 28 and 55 lm in thickness, respectively. At
the presented scale, an inhomogeneous character of the films sur-
face is clearly revealed, especially in the case of the thickest film
shown in Figure 1c. Elevations and cavities in the form of stripes
Figure 1. Low magnification SEM images of the surface of cobalt films 14
lm (a),
28 m (b) and 55 m (c) thick. The double headed arrow in each image is parallel to
l
l
the direction of the current flow during the electrodeposition process. The insets
present the two-dimensional Fourier transform of the images.
(
of the order of 100 lm in width), as well as smaller scale cavities
and some cracks, are present on the surface of the films. The depos-
its become more and more inhomogeneous when the deposit thick-
ness is increased, as proved also by the corresponding Fourier
transform patterns presented in the insets of Figure 1. It was verified
that the topographic stripes run in the direction parallel to the direc-
tion of the current flow during the electrodeposition process, but
the exact reason for their occurrence is not known. The cracks were
present in the surface region of the deposits, and they did not extend
through the whole thickness of the deposits, as evidenced in SEM
cross-sectional and surface images of the deposits. This is in contrast
to electrodeposited chromium films of similar thickness, where
cracks are reported to be commonly observed in the volume of the
films [31].

W. Szmaja et al. / Chemical Physics Letters 542 (2012) 117–122
119
The presence of cracks in electrodeposited cobalt and cobalt-
based films is reported in many papers (e.g. Refs. [20,28,31]). The
most likely reason for this undesired effect is the presence of ten-
sile stress in the film [3,28,32]. In general, there are several possi-
ble sources for generation of tensile stress. These include (i)
interfacial stress between the film and the substrate, (ii) crystallo-
graphic texture and grain size, (iii) coalescence and stress evolu-
tion during the film growth, and (iv) hydrogen adsorption/
desorption. The latter mechanism, related to the incorporation of
hydrogen into the deposit (adsorption) during electrodeposition
and subsequent diffusion out of the deposit (desorption) after the
interruption of electrodeposition, is found to be the most likely
dominant stress mechanism [32].
The tensile stress in the investigated cobalt deposits is found
not to be very large because the cracks occurred only in the surface
region of the deposits and the deposits exhibited good adhesion to
the substrate. There are few ways to reduce the stress (and conse-
quently to reduce or eliminate the cracks) in electrodeposited film.
These ways include, for example, the use of pulse electrodeposition
[
32], adding a stress reducer (such as saccharine, thiourea, benzene
sulfamide, benzene sulfonic acid, etc.) to the electrolytic solution
[
[
3,32], increasing the temperature of the electrolytic solution
32] or the application of magnetic field during the electrodeposit-
ion process [28].
AFM was used to study the morphological structure of the films
with high resolution. Figure 2a–c show original AFM images of
the surface of the cobalt films 14, 28 and 55 lm thick, respectively;
the insets present the two-dimensional Fourier transform of
the images. The morphological structure is composed of grains.
The grains are packed very densely, they are roundish in shape
and possess sizes at the nanometer range. Note, however, that in
the case of the 28
sense, the morphological structure of the film 28
ent from that of the films 14 and 55 m thick.
AFM images taken from sufficiently small film surface area,
m, usually showed some directionality,
lm thick film the grains form clusters. In this
l
m thick is differ-
l
smaller than about 2 ꢂ 2
l
i.e. the grains exhibited preferred orientations (which in general
varied when moving from one film surface area to another). This
is the case for the images of Figure 2a and c, as proved by the cor-
responding Fourier transform patterns in the form of an ellipse
(
presented in the insets). Nevertheless, images recorded from suf-
m, exhib-
ficiently large film surface area, larger than about 5 ꢂ 5
l
ited in fact no directionality, i.e. the grains ran in random
directions in the image plane. In general, no preferred elongation
of the grains was observed.
The original AFM images usually did not possess uniform back-
ground intensity, i.e. some regions were displayed as considerably
darker or brighter (cf. Figure 2). As a result, the visibility of grains
in such regions was unsatisfactory. To obtain better quality images,
appropriate for quantitative analysis of grain sizes, the original
images were subjected to the procedure of image differentiation.
After differentiating of the images from Figure 2, the corresponding
images presented in Figure 3 were obtained. On the basis of such
images as those in Figure 3, the grain size distributions and average
grain sizes in the plane of the investigated films were derived using
the standard linear intercept method [33–35]. In this method, a
large number of test straight lines running in random directions
is used; the method is illustrated in the image of Figure 3a, where
five test lines are drawn. The object size (the grain size in the con-
sidered case) S is determined from the formula [33]:
Figure 2. AFM images of the surface of cobalt films 14 lm (a), 28 lm (b) and 55 lm
(c) thick. The double headed arrow in each image is parallel to the direction of the
current flow during the electrodeposition process. The insets present the
two-dimensional Fourier transform of the images.
!
,
!
X
X
We used the standard linear intercept method with 1000 inter-
cepts counted for each AFM image. The grain sizes were in a wide
range; they varied from about 30 to 230 nm, from about 40 to
250 nm and from about 40 to 220 nm for the films of 14, 28 and
S ¼ ð2=
where l
p
Þ
l
i
n
i
ð1Þ
i
i
i
is the length of the ith test line and n
i
is the number of
intersections of the ith test line with object boundaries.
55 lm in thickness, respectively. The average grain sizes were

1
20
W. Szmaja et al. / Chemical Physics Letters 542 (2012) 117–122
for the films of 14, 28 and 55
lm in thickness, respectively. These
results show that the 55 m thick film had the roughest surface (as
l
expected from SEM study, cf. Figure 1) and the smallest grains.
Thanks to the application of SEM and AFM, useful complemen-
tary information on the morphological structure of the specimen
could be obtained. Although both techniques probe the surface of
the specimen, SEM is certainly better suited for investigating large
areas of the surface, whereas AFM allows for the study of small
areas with high sensitivity and high spatial resolution. Moreover,
in contrast to SEM, AFM provided additional information in the
direction perpendicular to the specimen surface, i.e. information
on the surface roughness. Note also that in the case of a specimen
composed of very small grains, SEM fails due to its insufficient spa-
tial resolution, as reported for example in Ref. [36].
Figure 4a–c present MFM images of the cobalt films 14, 28 and
55 lm in thickness, respectively. The magnetic structure consists
of maze stripe domains. In some regions the magnetic domains
were more maze in character (like those in Figure 4b), and in other
regions they were elongated, more like stripes than maze in char-
acter (like those in Figure 4a and c). In the case of the elongated do-
mains, they were typically from about 8 to 14 lm in length. The
reason for the undulation of the domains (the maze structure) is
the reduction in the magnetostatic energy at the cost of a larger to-
tal domain wall area. In general, the domains ran in random direc-
tions in the film plane, i.e. no directionality in the orientation of the
domains in the film plane was observed. The domains are dis-
played as bright and dark areas in the MFM images of Figure 4,
which proves unambiguously that they have opposite magnetiza-
tion components perpendicular to the film surface. This means also
that the magnetostatic interaction between the domains and the
tip was substantially non-perturbing. And more precisely, during
scanning process the tip was not remagnetized by the stray fields
of these domains each time when passing from one domain to a
neighboring domain (i.e. at the domain walls), providing informa-
tion on the direction of their magnetization component perpendic-
ular to the film surface [37].
The observed domain structure is characteristic of materials with
sufficiently high perpendicular magnetic anisotropy [7,11,35,38–
4
1]. Due to the lack of geometric alignment of the grains (the anisot-
ropy of the demagnetizing field), the magnetic anisotropy of the
studied films is obviously of crystalline origin (the magnetocrystal-
line anisotropy). In general, cobalt films can crystallize in the hexag-
onal close-packed (HCP) or face-centered cubic (FCC) phase. These
two phases have very different magnetic properties. The HCP phase
possesses only one magnetic easy axis [0001] (i.e. along the hexag-
onal axis), while the FCC phase has four easy axes [111] (i.e. along
the cube diagonals). Moreover, the FCC phase of cobalt has magneto-
crystallineanisotropymuchlower(about20–30times)thantheHCP
phase[42,43]. Whatismostimportanthere, theformerphasecannot
be the source of perpendicular magnetic anisotropy. As a conse-
quence, the investigated cobalt films have mainly the HCP crystal-
line structure. As proved by the MFM study, the films possess
perpendicular magnetic anisotropy high enough to overcome the
shape anisotropy of the film (the anisotropy of the demagnetizing
field) which tends to align the magnetization parallel to the film sur-
face. Because of the polycrystalline character of the films (Figure 2),
this in turn means that the films exhibit sufficiently strong crystallo-
graphic alignment (texture) of the cobalt grains with the hexagonal
axis (being the only easy axis of magnetization) perpendicular to the
film surface, i.e. the films exhibit sufficiently strong [0001] pre-
ferredorientation. To determinequantitatively thedegreeofcrystal-
lographic texture, investigations by the X-ray Schulz pole figures [7]
or the electron backscattered diffraction (EBSD) method [34] are
required.
Figure 3. Images (a–c) were obtained by applying a differentiation procedure to the
AFM images from Figure 2a–c, respectively. On the image (a), five test straight lines
are additionally superimposed to illustrate the standard linear intercept method
used for determining the grain sizes.
(
2
123 ± 9) nm, (130 ± 14) nm and (111 ± 6) nm for the films of 14,
8 and 55 m in thickness, respectively. Moreover, the root mean
l
square (rms) values of the surface roughness, obtained directly by
AFM measurements using images taken at different places on the
specimen surface, were (20 ± 3) nm, (30 ± 3) nm and (80 ± 20) nm
For obtaining quantitative data about the sizes of magnetic do-
mains in the studied films, we used the standard linear intercept

W. Szmaja et al. / Chemical Physics Letters 542 (2012) 117–122
121
known that the magnetic domain size becomes larger as the mag-
netic anisotropy of the specimen decreases [44,45]. In the case of
the studied films, the magnetic anisotropy is apparently higher
when the crystallographic alignment of the grains is better. As a
consequence, the crystallographic texture is found to increase with
increasing grain size. The occurrence of such dependence for elec-
trodeposited thick cobalt films was indicated in Ref. [24]. It is also
worth noting that reported recently in Ref. [46] electrodeposited
cobalt films with very high [0001] preferred orientation possessed
similar morphology (with grains roundish in shape and having
sizes in the range of about 50–100 nm) to the films studied by us.
In the case of the films investigated in this work, the average
domain size was few times larger than the average grain size,
which means that each magnetic domain usually consisted of
many crystalline grains. As a general rule, when the average grain
size is about 40 nm or less, a strong exchange coupling among the
grains occurs [47,48]. This is not the case for the studied films, in
which the interactions between the grains are expected to be pre-
dominantly magnetostatic in nature. It is worth noting that the
presence of a strong exchange interaction in cobalt films composed
of grains smaller than about 30 nm in size leads to wide magnetic
domains of a few micrometers in size [49].
In contrast to previous studies, we performed a detailed inves-
tigation of the grain and magnetic domain structures of the thick
cobalt films electrodeposited on polycrystalline gold substrates,
by high-resolution techniques of AFM and MFM, respectively.
The average grain and domain sizes for the films were determined
using the standard linear intercept method. The magnetic domain
structure determines the hysteresis loop of the material, and the
domain configuration is affected by the grain structure. The ob-
tained results show the correlation between the domain size and
the grain size. A detailed knowledge and understanding of the
grain structure, magnetic domain structure and the correlation be-
tween these structures are not only of basic research interest, but
also of practical significance. From the practical point of view, they
are important for tailoring of the materials with the required prop-
erties. The knowledge of the magnetic domain behavior in relation
to the grain structure of the specimens is also significant for theo-
retical modeling of magnetic properties.
4
. Conclusions
A study has been made of the morphological and magnetic
structures of thick cobalt films obtained by electrodeposition on
polycrystalline gold substrates. The films were 14, 28 and 55
lm
in thickness, as revealed by SEM. They were electrodeposited from
a cobalt sulfate bath with pH value of 4.5 at room temperature and
2
at constant current density of 20 mA/cm . The morphological
structure of the deposits was imaged by SEM and AFM, whereas
their magnetic structure was made visible with MFM.
Despite their large thickness, the films did not detach from the
substrate, thus in this regard the used polycrystalline gold sub-
strate can be recognized as suitable. Nevertheless, at the microm-
eter scale the morphology of the deposits was inhomogeneous in
nature. Elevations and cavities in the form of stripes (of the order
Figure 4. MFM images of cobalt films 14 lm (a), 28 lm (b) and 55 lm (c) thick. The
double headed arrow in each image is parallel to the direction of the current flow
during the electrodeposition process.
of 100 lm in width), as well as smaller scale cavities and some
cracks, were present on the surface of the films. The deposits be-
came more and more inhomogeneous as the deposit thickness
was increased. But it is necessary to note that the deposits were
found not to be highly stressed because they showed good
adhesion to the substrate and the cracks occurred only in the sur-
face region of the deposits.
The smaller-scale morphological structure of the deposits was
composed of nanocrystalline grains which were roundish in shape,
possessed sizes of the order of 100 nm, and generally exhibited no
method with 1000 intercepts counted for each MFM image. The
average domain sizes were determined to be (560 ± 20) nm,
(
530 ± 20) nm and (610 ± 30) nm for the films 14, 28 and 55 lm
in thickness, respectively. Taking into account the average grain
sizes of the studied films, one can notice a correlation between
the domain size and the grain size. Namely, the domain size in-
creases with decreasing grain size. On the other hand, it is widely

1
22
W. Szmaja et al. / Chemical Physics Letters 542 (2012) 117–122
[15] M. Cichomski et al., J. Alloys Compd. 507 (2010) 273.
preferred elongation and no geometric alignment. The recorded
images of the grain structure were subjected to digital processing
and analysis for obtaining quantitative information on grain sizes.
The average grain sizes for the investigated films were determined
using the standard linear intercept method. Thanks to applying the
SEM and AFM techniques, useful complementary information on
the morphological structure of the films could be obtained.
The magnetic structure of the films was composed of domains.
The domains formed a maze pattern and possessed sizes of the or-
der of 500 nm. The domain structure was characteristic of materi-
als with sufficiently high perpendicular magnetic anisotropy. As a
consequence, the deposits were found to crystallize mainly in the
HCP structure with sufficiently strong [0001] preferred orienta-
tion. The average domain sizes for the studied films were deter-
mined using the standard linear intercept method. The obtained
results show that the average domain size slightly increased with
decreasing average grain size.
[
[
16] W. Gil, D. Görlitz, M. Horisberger, J. Kötzler, Phys. Rev. B 72 (2005) 134401.
17] C.-P. Liu, J.-J. Chang, S.-W. Chen, H.-C. Chung, Y.-L. Wang, Appl. Phys. A 80
(
2005) 1601.
[18] M.F. Chioncel, H.S. Nagaraja, F. Rossignol, P.W. Haycock, J. Magn. Magn. Mater.
13 (2007) 135.
19] N. Deo, M.F. Bain, J.H. Montgomery, H.S. Gamble, J. Mater. Sci.: Mater. Electron.
6 (2005) 387.
[20] R. Fukui, Y. Katayama, T. Miura, Electrochim. Acta 56 (2011) 1190.
3
[
1
[
21] H. Harti, J.-L. Bubendorff, A. Florentin, C. Pirri, J. Ebothe, J. Cryst. Growth 319
2011) 79.
22] I. Flis-Kabulska, J. Appl. Electrochem. 36 (2006) 131.
(
[
[23] M.V. Rastei, S. Colis, J.P. Bucher, Chem. Phys. Lett. 417 (2006) 217.
[
[
[
24] S. Armyanov, Electrochim. Acta 45 (2000) 3323.
25] C.A. Moina, L. de Oliveira-Versic, M. Vazdar, Mater. Lett. 58 (2004) 3518.
26] M. Rivera, C.H. Rios-Reyes, L.H. Mendoza-Huizar, Appl. Surf. Sci. 255 (2008)
1754.
[
27] M. Rivera, C.H. Rios-Reyes, L.H. Mendoza-Huizar, J. Magn. Magn. Mater. 323
(
2011) 997.
[
28] M. Zieli n´ ski, E. Mi e˛ ko s´ , J. Appl. Electrochem. 38 (2008) 1771.
[29] P. Grütter, H.J. Mamin, D. Rugar, in: R. Wiesendanger, H.-J. Güntherodt (Eds.),
Scanning Tunneling Microscopy II, Springer, Berlin, 1992, p. 151.
[
30] W. Szmaja, in: P.W. Hawkes (Ed.), Advances in Imaging and Electron Physics,
vol. 141, Elsevier, Amsterdam, 2006, p. 175.
Acknowledgements
[
[
31] D.P. Weston, P.H. Shipway, S.J. Harris, M.K. Cheng, Wear 267 (2009) 934.
32] I. Tabakovic, J. Gong, S. Riemer, V. Venkatasamy, M. Kief, Electrochim. Acta 55
(
2010) 9035.
This work was supported by the Łód z´ University. Witold Szmaja
dedicates the Letter to the memory of his mother.
[
[
33] A. Hubert, R. Schäfer, Magnetic Domains: The Analysis of Magnetic
Microstructures, Springer, Berlin, 1998. p. 329.
34] T.G. Woodcock, K. Khlopkov, A. Walther, N.M. Dempsey, D. Givord, L. Schultz,
O. Gutfleisch, Scripta Mater. 60 (2009) 826.
References
[
[
[
[
[
[
35] W. Szmaja, J. Magn. Magn. Mater. 301 (2006) 546.
36] J. Garcia-Torres, E. Gómez, E. Vallés, Mater. Chem. Phys. 122 (2010) 463.
37] W. Szmaja, J. Grobelny, M. Cichomski, Appl. Phys. Lett. 85 (2004) 2878.
38] H.S. Park, D. Shindo, S. Mitani, K. Takanashi, IEEE Trans. Magn. 41 (2005) 3724.
39] F. Stobiecki et al., Appl. Phys. Lett. 92 (2008) 012511.
[
[
1] H. Zabel, Superlatt. Microstruct. 46 (2009) 541.
2] P.P. Freitas, R. Ferreira, S. Cardoso, F. Cardoso, J. Phys.: Condens. Matter 19
(
2007) 165221.
3] N.V. Myung, D.-Y. Park, B.-Y. Yoo, P.T.A. Sumodjo, J. Magn. Magn. Mater. 265
2003) 189.
4] N.M. Dempsey, in: J.P. Liu, E. Fullerton, O. Gutfleisch, D.J. Sellmyer (Eds.),
Nanoscale Magnetic Materials and Applications, Springer, Dordrecht, 2009, p.
[
[
40] X.M. Li, Y.K. Fang, Z.H. Guo, T. Liu, Y.Q. Guo, W. Li, B.S. Han, Chin. Phys. B 17
(
(
2008) 2281.
[
[
[
41] D. Nunes, R. Colaço, J.Th.M. de Hosson, A.P. Gonçalves, L.C.J. Pereira, P.A.
Carvalho, J. Alloys Compd. 487 (2009) 11.
42] Y.T. Chen, S.U. Jen, Y.D. Yao, J.M. Wu, C.C. Lee, A.C. Sun, IEEE Trans. Magn. 42
6
61.
[
[
[
[
5] G. Srajer et al., J. Magn. Magn. Mater. 307 (2006) 1.
6] H.J. Richter, J. Phys. D: Appl. Phys. 40 (2007) R149.
7] J. Brandenburg, R. Hühne, L. Schultz, V. Neu, Phys. Rev. B 79 (2009) 054429.
8] F. Dumas-Bouchiat, H.S. Nagaraja, F. Rossignol, C. Champeaux, G. Trolliard, A.
Catherinot, D. Givord, J. Appl. Phys. 100 (2006) 064304.
(
2006) 278.
43] J. Lindner, M. Farle, in: H. Zabel, S.D. Bader (Eds.), Springer Tracts in Modern
Physics, vol. 227, Springer, Berlin, 2007, p. 45.
44] Y. Gao, J. Zhu, Y. Weng, B. Han, Appl. Phys. Lett. 74 (1999) 1749.
45] W. Szmaja, K. Pola n´ ski, K. Dolecki, J. Magn. Magn. Mater. 130 (1994) 147.
46] A. Vicenzo, P.L. Cavallotti, Electrochim. Acta 49 (2004) 4079.
47] G.C. Hadjipanayis, J. Magn. Magn. Mater. 200 (1999) 373.
[
[
[
[
[
[
9] H.Y. Kwong, Y.W. Wong, J. Alloys Compd. 497 (2010) 267.
[
10] A. Sharma, S. Tripathi, N. Lakshmi, P. Sachdev, T. Shripathi, Solid State
Commun. 149 (2009) 1033.
48] W. Szmaja, J. Grobelny, M. Cichomski, S. Hirosawa, Y. Shigemoto, Acta Mater.
[
[
[
11] Y. Kageyama, T. Suzuki, J. Magn. Magn. Mater. 310 (2007) e789.
12] M.T. Umlor, Appl. Phys. Lett. 87 (2005) 082505.
13] D. Schmidt, A.C. Kjerstad, T. Hofmann, R. Skomski, E. Schubert, M. Schubert, J.
Appl. Phys. 105 (2009) 113508.
5
9 (2011) 531.
[
49] R. Schäfer, in: J.P. Liu, E. Fullerton, O. Gutfleisch, D.J. Sellmyer (Eds.), Nanoscale
Magnetic Materials and Applications, Springer, Dordrecht, 2009, p. 275.
[
14] W. Szmaja, W. Kozłowski, J. Balcerski, P.J. Kowalczyk, J. Grobelny, M.
Cichomski, J. Alloys Compd. 506 (2010) 526.