Magnetic properties of nanocrystalline iron group thin film alloys electrodeposited from sulfate and chloride baths

Article abstract of DOI:10.1016/S0013-4686(02)00773-9

Systematic studies of iron group binary (NiCo and CoFe) and ternary (CoNiFe) thin film alloys relating their magnetic properties with film composition, grain size and the corresponding crystal structure were investigated. Anions influence current efficiencies, magnetic properties, surface morphology and phases of electrodeposited films. Higher current efficiencies in chloride baths compared to sulfate baths were observed for CoFe, NiCo and CoNiFe alloys. The higher deposition current efficiencies in chloride baths were attributed to a catalytic effect. Anion types in CoFe and CoNiFe thin film alloys influenced the microstructures and the resulting magnetic properties (coercivity and squareness). The microstructures of NiCo alloys depend on the deposit Co contents rather than anion types. The surface morphologies of CoFe, NiCo and CoNiFe thin films were independent of anion types. CoFe deposits exhibited relatively smooth surface morphology and turned into fine crystallites with increasing solution Fe⁺² concentration. NiCo deposits showed very smooth surface morphology. CoNiFe deposits had the surface morphology of polyhedral crystallites. The deposit Fe content in CoFe electrodeposits linearly increased with increasing solution Fe⁺² concentration for both chloride and sulfate baths. Similar linear behavior of deposit Co contents was observed in NiCo electrodeposits.

Full text of DOI:10.1016/S0013-4686(02)00773-9

Electrochimica Acta 48 (2003) 819Á830
/
www.elsevier.com/locate/electacta
Magnetic properties of nanocrystalline iron group thin film alloys
electrodeposited from sulfate and chloride baths
Daheum Kim 1, D.-Y. Park, B.Y. Yoo, P.T.A. Sumodjo 2,3, N.V. Myung2 *
Department of Chemical Engineering, University of California, Los Angeles, CA 90095-1592, USA
Received 31 July 2002; received in revised form 23 October 2002
Abstract
Systematic studies of iron group binary (NiCo and CoFe) and ternary (CoNiFe) thin film alloys relating their magnetic properties
with film composition, grain size and the corresponding crystal structure were investigated. Anions influence current efficiencies,
magnetic properties, surface morphology and phases of electrodeposited films. Higher current efficiencies in chloride baths
compared to sulfate baths were observed for CoFe, NiCo and CoNiFe alloys. The higher deposition current efficiencies in chloride
baths were attributed to a catalytic effect. Anion types in CoFe and CoNiFe thin film alloys influenced the microstructures and the
resulting magnetic properties (coercivity and squareness). The microstructures of NiCo alloys depend on the deposit Co contents
rather than anion types. The surface morphologies of CoFe, NiCo and CoNiFe thin films were independent of anion types. CoFe
deposits exhibited relatively smooth surface morphology and turned into fine crystallites with increasing solution Feꢀ2
concentration. NiCo deposits showed very smooth surface morphology. CoNiFe deposits had the surface morphology of
polyhedral crystallites. The deposit Fe content in CoFe electrodeposits linearly increased with increasing solution Feꢀ2
concentration for both chloride and sulfate baths. Similar linear behavior of deposit Co contents was observed in NiCo
electrodeposits.
# 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Electrodeposition; Magnetic films; NiCo alloys; CoFe alloys; CoNiFe alloys; Magnetic properties
1. Introduction
is the best known iron group thin film alloy in magnetic
thin film recording heads [1,2] and MEMS applications
Electrodeposited magnetic thin films of the iron-
group metals (Fe, Co and Ni) have been developed
because of potential applications in computer read/write
heads [1,2] and microelectromechanical systems
[3Á5] with magnetic saturation of 0.97 T, low coercivity,
/
and low magnetostriction. However, new soft magnetic
materials with higher performance are needed because
of the dramatic increase of the areal density (60% per
year) in computer drives [2,6] and further miniaturiza-
tion and/or better performance of electromagnetic
devices in MEMS [5,6]. Various CoFe- and CoNi-based
ternary and quarternary alloys including CoFeB [7],
(MEMS) [3Á5]. Current computer read/write heads are
/
separately fabricated using physical deposition methods
for GMR spin valve read heads, and electrodeposition
for write heads. Electrodeposited permalloy (80Ni20Fe)
CoFeCu [2], CoNiFe [5,6,8Á11], CoNiFeS [12], CoFeP
/
[13] and CoFeSnP [13] have been considered as possible
candidates. Among those alloy systems, the ternary
CoNiFe alloys have been demonstrated as one of the
promising systems due to their high magnetic saturation
* Corresponding author. Present address: MEMS Technology
Group, Jet Propulsion Laboratory, 4800 Oak Glove Blvd.,
Pasadena, CA 91109, USA. Tel.: ꢀ
/
1-818-393-3239; fax: ꢀ1-818-393-
/
4540.
E-mail address: nosang.v.myung@jpl.nasa.gov (N.V. Myung2).
flux density (BSꢀ
low coercivity field strength (B
Liao [7] suggested the fabrication of read/write heads
using CoFeB alloys with 0.1Á2 wt.% B and 7Á12 wt.%
Fe contents. He used a sulfate bath with sodium citrate
/
2.0Á
/
2.1 T) combined with reasonably
1
Department of Chemical Engineering, Kwangwoon University,
Wolgye-dong, Nowon-gu, Seoul, Republic of Korea.
/2 Oe) [6,14,15].
2
ISE member.
Visiting scholar from Institute of Chemistry, University of Sao
Paulo, Sao Paulo, Brazil.
˜
3
/
/
˜
0013-4686/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 1 3 - 4 6 8 6 ( 0 2 ) 0 0 7 7 3 - 9

820
D. Kim et al. / Electrochimica Acta 48 (2003) 819Á830
/
and dimethyl amine borane (DMAB) as reducing agent
for boron at pH 3.5. He reported a decrease of coercivity
to less than 1 Oe due to boron addition into CoFe alloy,
while other properties remained the same as the proper-
ties of the binary CoFe alloy. Andricacos and
Robertson [2] reviewed the requirements, including: (i)
ing crystal structures. This paper is the extended study
based on our previous works [30]. In this study, we
discuss the influence of solution compositions and
electrodeposition parameters on the film compositions
using chloride and sulfate baths. The resulting micro-
structures and magnetic properties of Co, Ni, Fe, CoFe,
NiCo and CoNiFe alloys were also studied.
high magnetic saturation (MSꢁ
/
1 T); (ii) low coercivity
HCB1 Oe; (iii) optimal anisotropy field (HK) for high
/
permeability; (iv) close to zero saturation magnetostric-
tion (l); (v) high electrical resistivity (r); (vi) good
corrosion resistance (Rp), for improved thin-film record-
ing heads. They studied the dependence of magnetic
properties of CoFeCu films on composition and sug-
2. Experimental
Co, Ni, Fe, CoFe, NiCo and CoNiFe alloys were
electrodeposited from chloride and sulfate baths. Table
1 gives the plating solution compositions investigated.
Two different anions (chloride and sulfate electrolytes)
gested that addition of Cu causes a decrease (B1.5 Oe)
/
in the coercivity of the film. Chessnutt [10] studied
electrodeposited CoNiFeB alloys for recording head. He
obtained a magnetic saturation of 1.5 T in Co80Ni10Fe10
alloy. It was observed that addition of boron in CoNiFe
alloy decreased the coercivity from 1.5 to 0.6 Oe. Ohashi
et al. [11] suggested electrodeposited Co65Ni12Fe23 film
from sulfate bath as one of the promising materials for
high density recording MR heads. They reported that
Co65Ni12Fe23 films with mixed bcc and fcc phases
Table 1
Bath compositions for Co, Ni, Fe, CoFe, CoNi and CoNiFe
electrodeposits
Metal/ Anion
alloy
Composition
type
exhibit higher magnetic saturation (MSꢀ
/
2.0Á2.1 T),
/
improved corrosion resistance and good thermal stabi-
lity than other Co65Ni12Fe23 films with bcc phases. In a
subsequent study [14], Osaka et al. obtained electro-
Co
Chloride 0.2 M CoCl2ꢀ
0.0075 M saccharin
0.2 M CoSO4ꢀ0.7 M Na2SO4ꢀ
0.0075 M saccharin
Chloride 0.2 M CoCl2ꢀ0.7 M NaClꢀ
0.0075 M saccharinꢀ0.05 M L’ascorbic acid
0.2 M CoSO4ꢀ0.7 M Na2SO4ꢀ0.4 M H3BO3ꢀ
0.0075 M saccharinꢀ0.05 M L’ascorbic acid
/
0.7 M NaClꢀ
/
0.4 M H3BO3ꢀ
/
Sulfate
/
/0.4 M H3BO3ꢀ
/
deposited Co65Ni12Fe23 alloys (fine grain sizesꢀ
nm) with high magnetic saturation (MSꢀ2.1 T) and low
coercivity (B1.2 Oe) using sulfate baths without sulfur-
/
10Á20
/
/
/0.4 M H3BO3ꢀ
/
/
/
/
Sulfate
/
/
/
containing additives such as saccharin and thiourea.
Takai et al. [12] studied soft magnetic CoNiFe films
electrodeposited from sulfate bath with thiourea as
additive. They obtained (Co73Ni12Fe15)99.1S0.9 with
/
Ni
Chloride 0.2 M NiCl2ꢀ0.7 M NaClꢀ0.4 M H3BO3ꢀ
0.0075 M saccharin
0.2 M NiSO4ꢀ0.7 M Na2SO4ꢀ
0.0075 M saccharin
Chloride 0.2 M FeCl2ꢀ0.7 M NaClꢀ
0.0075 M saccharinꢀ0.05 M L’ascorbic acid
0.2 M FeSO4ꢀ0.7 M Na2SO4ꢀ0.4 M H3BO3ꢀ
0.0075 M saccharinꢀ0.05 M L’ascorbic acid
Chloride 0.2 M CoCl2ꢀx M FeCl2ꢀ0.7 M NaClꢀ0.4 M
H3BO3ꢀ0.0075 M saccharinꢀ0.05 M L’ascor-
bic acid
0.2 M CoSO4ꢀ
0.4 M H3BO3ꢀ0.0075 M saccharinꢀ
L’ascorbic acid
Chloride 0.2 M NiCl2ꢀx M CoCl2ꢀ
H3BO3ꢀ0.0075 M saccharin
0.2 M NiSO4ꢀx M CoSO4ꢀ
0.4 M H3BO3ꢀ0.0075 M saccharin
CoNiFe Chloride 0.2 M NiCl2ꢀ0.15 M CoCl2ꢀy M FeCl2ꢀ
M NaClꢀ0.4 M H3BO3ꢀ0.0075 M saccharinꢀ
0.05 M L’ascorbic acid
0.2 M NiSO4ꢀ0.15 M CoSO4ꢀ
0.7 M Na2SO4ꢀ0.4 M H3BO3ꢀ
saccharinꢀ0.05 M L’ascorbic acid
x: Feꢀ2 concentration in CoFe and Coꢀ2 concentration in NiCo,
0.015x 5 y 5
0.16 M; y: Feꢀ2 concentration in CoNiFe, 0.0055
0.16 M.
/
/
/
Sulfate
/
/
0.4 M H3BO3ꢀ
/
small grain size (5Á
(MSꢀ1.7 T), high resistivity up to 51 mVcm and low
saturation magnetostriction (lSꢀ4.4ꢁ
10ꢂ6). They
suggested that low coercivity and high resistivity re-
sulted from the formation of nano grains (5Á10 nm).
/10 nm), high magnetic saturation
Fe
/
/
0.4 M H3BO3ꢀ
/
/
/
/
/
Sulfate
/
/
/
/
/
CoFe
/
/
/
Hironaka and Uedaira [13] investigated amorphous
CoFeP and CoFeSnP films electrodeposited from sulfa-
mate bath. CoFeP and CoFeSnP films exhibit a
/
/
Sulfate
/
x M FeSO4ꢀ
/
0.7 M Na2SO4ꢀ
/
/
/
0.05 M
magnetic saturation of 1Á1.35 T. They observed that
/
the addition of Sn into CoFeP increases the corrosion
resistance without decreasing magnetic saturation.
Although numerous studies had been carried out to
investigate the binary (CoFe, NiCo, NiFe) and ternary
(CoNiFe) iron group magnetic thin films, they mostly
focused on the mechanism of anomalous codeposition
NiCo
/
/
0.7 M NaClꢀ
/
0.4 M
/
Sulfate
/
/
0.7 M Na2SO4ꢀ
/
/
/
/
/0.7
/
/
/
[16Á24], the effects of various additives [15,25,26], the
/
effects of pulse plating [27,28], corrosion properties
[29,30], and performance studies for recording head
[10,31]. There is a lack of systematic studies relating
magnetic properties of electrodeposited iron group thin
films including CoFe, NiCo, NiFe and CoNiFe alloys
with deposit composition, grain size and the correspond-
Sulfate
/
/
y M FeSO4ꢀ
/
/
/0.0075 M
/
/
/
/
/

D. Kim et al. / Electrochimica Acta 48 (2003) 819Á
/830
821
Fig. 1. Dependence of deposit Fe content and current efficiency with Feꢀ2 solution concentration in binary CoFe electrodeposits (a) chloride bath
and (b) sulfate bath ([Coꢀ2]ꢃ
0.2 M).
/
were used for comparison. NaCl and Na2SO4 were used
as supporting electrolytes in chloride and sulfate baths,
respectively. Boric acid was added as pH buffer.
Saccharin was used to reduce deposit stress and
L’ascorbic acid to minimize Feꢀ2 oxidation in CoFe
and CoNiFe solutions. Solutions were exposed to air
and solution pH was adjusted to 3 by adding HCl
(chloride baths), H2SO4 (sulfate baths) or NaOH
(chloride and sulfate baths); experiments were con-
ducted at 10 mA/cm2, pH 3, 10 coulombs/cm2, and
room temperature without stirring.
The effect of the solution [Coꢀ2]/[Feꢀ2] ratio on the
deposit composition of binary CoFe alloy films was
conducted by varying the Feꢀ2 concentration from 0.01
to 0.16 M, with Coꢀ2 concentration set at 0.2 M. The
effect of the solution [Niꢀ2]/[Coꢀ2] ratio on the deposit
Fig. 2. Dependence of magnetic saturation (MS), coercivity (HC) and squareness (S) of binary CoFe electrodeposits with deposit Fe content from
chloride and sulfate baths (a) magnetic saturation, (b) coercivity, and (c) squareness ([Coꢀ2]ꢃ
0.2 M).
/

822
D. Kim et al. / Electrochimica Acta 48 (2003) 819Á830
/
Fig. 4. Comparison of deposit Fe content for chloride and sulfate
baths as a function of solution Feꢀ2 concentration in binary CoFe
electrodeposits ([Coꢀ2]ꢃ
0.2 M).
/
deposits were examined with a scanning electron micro-
scope (Stereoscan 250, Cambridge, Scientific
Instruments Ltd.). Magnetic properties such as magnetic
saturation (MS), coercivity (HC) and squareness (Sꢃ
/
Mr/Ms) of electrodeposited films were determined using
a vibrating sample magnetometer (Model 880, ADE
Technologies Inc.).
3. Results and discussion
3.1. CoFe alloys
To study anion effects on current efficiencies and
magnetic properties, binary CoFe thin films were
electrodeposited from chloride and sulfate baths. Fig.
1 shows dependence of deposit Fe contents and current
efficiencies with solution Feꢀ2 concentration. Deposit
Fe contents practically increased linearly with increasing
solution Feꢀ2 concentration for chloride and sulfate
Fig. 3. XRD patterns of Fe, Co and CoFe electrodeposits from: (a)
chloride and (b) sulfate baths.
composition of binary NiCo films was investigated by
varying the Coꢀ2 concentration from 0.01 to 0.16 M,
with Niꢀ2 concentration set at 0.2 M. The electrodepo-
sition of ternary CoFeNi alloys was carried out with
fixed Niꢀ2 concentration of 0.2 M, Coꢀ2 concentration
of 0.15 M, and variable concentrations of Feꢀ2 from
0.005 to 0.16 M. Brass was used as substrate; Fe, Co or
Ni was used as soluble anode.
baths. Current efficiencies (approximately ꢀ
independent of solution Feꢀ2 concentrations in chloride
baths, but decreased to level at ꢀ18% in sulfate baths.
Current efficiency (ꢀ52%) in chloride baths were higher
than that (ꢀ18%) in sulfate baths. Hokans [17,32]
/52%) was
/
/
/
Deposit Co, Ni and Fe contents were analyzed using
atomic absorption spectroscopy (AA) (Model 280,
investigated hydrogen evolution using Na2SO4, NaCl
and NaClO4 solutions during the electrodeposition of
NiFe alloys. She obtained higher limiting currents of
Hꢀ reduction in Na2SO4 than in NaCl and NaClO4.
She observed a significant decrease in the reduction
potential of Ni, Fe and NiFe in chloride baths and
suggested that chloride catalyses deposition of Ni, Fe,
and NiFe through the formation of an ion bridge
between the electrode and the metal ion being dis-
charged. Myung and Nobe [30] also observed that
PerkinÁElmer). Alloy compositions are given in weight
/
percent. X-ray diffractometer (XRD) (Model 42202,
Norelco, North American Philips Company Inc.) with
CuKa radiation was used for the identification of the
phases and the measurement of grain size in the
electrodeposits. The conditions of XRD were a scanning
range (40Á
/
808) with 0.038 increments and 1 s dwell time.
Microstructures of CoFe, NiCo and CoNiFe electro-
Fig. 5. Surface morphology of Co, Fe and CoFe electrodeposits from chloride [(a)Á
/
(d)] and sulfate baths [(e)Á(f)] containing L’ascorbic acid (a) pure
/
Co, (b) 74Co24Fe, (c) 52Co48Fe, (d) pure Fe, (e) pure Co, (f) 78Co22Fe, (g) 52Co48Fe, and (h) pure Fe.

D. Kim et al. / Electrochimica Acta 48 (2003) 819Á
/830
823
Fig. 5

824
D. Kim et al. / Electrochimica Acta 48 (2003) 819Á830
/
current efficiencies in chloride baths were higher than in
sulfate baths during the electrodeposition of NiFe thin
film alloys. Therefore, higher current efficiency for
CoFe electrodeposition in chloride baths compared to
sulfate baths in our experiments may result from a
similar catalytic mechanism.
Fig. 2 shows the effect of deposit Fe content from
chloride and sulfate baths on magnetic saturation (MS),
coercivity (HC) and squareness (S). Magnetic saturation
of CoFe electrodeposits was independent of anion types
and increased with increasing deposit Fe content. The
observed MS variation is reasonable from the fact that
MS is an intrinsic magnetic property, which is dependent
only on the film composition. In CoFe alloys, when the
solution Feꢀ2 concentrations in chloride and sulfate
baths are equal, nearly same amounts of Fe were
electrodeposited as shown in Fig. 1. Coercivity of
CoFe electrodeposits from chloride baths increased
ties. There exist difficulties to exactly index the peak at
45.18 [either hcp (002) or bcc (110)] for CoFe alloys
ꢀ
/
with Fe contents (22 wt.%) from chloride and (32 wt.%)
sulfate baths (Fig. 3). The difficulties of indexing result
from two reasons. One comes from that JCPDS files for
the CoFe alloys are not available and other results from
a shift of peak positions due to the slight difference of
atomic radius between Co (0.125 nm) and Fe (0.124
nm). From the phase diagram of CoFe alloys [34], Co
and Fe form a complete substitutional solid solution at
room temperature. As the deposit Fe amount in CoFe
electrodeposits increase, the peak position will shift to
higher angle according to the following Eqs. (1) and (2)
[33]; 2u of hcp (002) is 44.7628 from JCPDS file #5-727,
l2
sin2 uꢃ
sin2 uꢃ
(h2 ꢀk2 ꢀl2) for cubic
(1)
4a2
ꢀ
ꢁ
l2 h2 ꢀ hk ꢀ l2
l2 × l2
4 × c2
from 17 (at 0 wt.% Fe) to 40 Oe (ꢂ
/
ꢀ/18 wt.% Fe),
ꢀ
for hexagonal (2)
3
a2
and then reached a plateau. Coercivity of CoFe electro-
deposits from sulfate baths increased from ꢀ
Oe with increasing deposit Fe content with B
wt.%, and then decreased to ꢀ
content of 22Á50 wt.%. Squareness of CoFe alloys from
chloride baths increased from ꢀ0.4 to ꢀ0.6 with
increasing deposit Fe contents of B 20 wt.% Fe,
and then reached a plateau with deposit Fe contents of
20 wt.%. Squareness from sulfate baths decreased
from 0.8 to 0.3 at B 20 wt.% Fe, and then increased
to be 0.5 at ꢂ50 wt.% Fe.
/
17 to ꢀ
/
60
where, l is the wavelength of radiation, a is the size of
unit cell, and h, k, l are the Miller indices of reflecting
/
ꢀ
/
22
/
18 Oe at deposit Fe
plane. The peaks at ꢀ
/
45.18 for CoFe electrodeposits
/
containing 22Á32 wt.% Fe from chloride and sulfate
/
/
/
baths can be attributed to the hcp (002) planes rather
than bcc (110) planes, because the deposit Co contents
were higher than the deposit Fe contents. The deposit Fe
/
ꢀ
/
ꢂ
/
ꢀ
/
contents was analyzed to be 22Á
obscure to index the peaks at ꢀ
/
32 wt.%. However, it is
45.18 for 0.14 M Feꢀ2
/
ꢀ
/
/
/
concentration for chloride and sulfate baths due to the
nearly same amount of deposit Co and Fe (52 and 48) as
shown in Fig. 4.
Pure Co electrodeposits from chloride baths contain-
ing L’ascorbic acid showed acicular morphology as
shown in Fig. 5a. As the deposit Fe content in CoFe
To study the influence of anion type on the micro-
structure of CoFe electrodeposits, XRD measurements
were carried out as shown in Fig. 3 (XRD measurements
of pure Co and Fe deposits were conducted for
comparison). The grain size was calculated from the
peak broadening using the Scherrer formula [33]. In
chloride baths, CoFe electrodeposits containing deposit
32 wt.% Fe (0.08 M Feꢀ2 concentration in the bath)
deposits increased from 24 to 48 wt.% (0.06Á0.14 M
/
Feꢀ2 concentration), crystalline size decreased (Fig. 5b
and c). Pure Fe electrodeposits were not compact and
exhibited a coercivity of ꢀ41 Oe. From the XRD
/
patterns (Fig. 3a), this film consisted of hcp (002) [or bcc
(110)] and bcc (200) phases with an average grain size of
ꢀ63 nm. For deposits with higher Fe content, the
/
microstructures did not changed, but the intensities of
hcp (002) [or bcc (110)] phase decreased (as shown in
Fig. 3a for the alloy containing 48 wt.% Fe). However,
fairly constant coercivities (ꢀ40 Oe) are observed. Co-
/
22 wt.% Fe films plated from sulfate baths (0.06 M Feꢀ2
concentration), which had a maximum coercivity of
ꢀ
/
60 Oe, consisted of hcp (002) [or bcc (110)] planes
with the average grain size of ꢀ40 nm. The coercivity of
CoFe electrodeposits containing 48 wt.% Fe (0.14 M
Feꢀ2 concentration) was B
30 Oe. The microstruc-
ture of this film showed hcp (002) [or bcc (110)] and bcc
(200) planes with an average grain size of ꢀ33 nm.
/
/
ꢀ
/
/
Fig. 6. Dependence of deposit Co content and current efficiency with
Coꢀ2 solution concentration in binary CoNi electrodeposits from
Therefore, anion types did not affect on the phases of
CoFe deposits, but influenced on the magnetic proper-
sulfate bath ([Niꢀ2]ꢃ
0.2 M).
/

D. Kim et al. / Electrochimica Acta 48 (2003) 819Á
/830
825
exhibited polyhedral crystallites. On the other hand,
pure Co deposits (Fig. 5e) from sulfate baths with
L’ascorbic acid had very smooth surface. As the deposit
Fe content increased, similar feature (Fig. 5f and g) of
the surface morphology compared to those from chlor-
ide baths was observed. The needle shape morphology
at the right and upper corner part (marked with an
arrow in Fig. 5g) resulted from a fast rusting of CoFe
electrodeposits. Pure Fe deposits from sulfate bath (Fig.
5h) showed more compact and dense polyhedral crystal-
lites compared to that from chloride bath (Fig. 5d).
3.2. CoNi alloys
Fig. 6 shows the dependence of deposit Co content
and current efficiency with Coꢀ2 solution concentration
during the electrodeposition of CoNi thin films from
sulfate baths. Deposit Co content increased linearly with
increasing solution Coꢀ2 concentration. Unlike NiCo
electrodeposits from chloride bath reported previously
[30], current efficiency in sulfate baths showed a max-
imum (ꢀ
45%) for Coꢀ2 concentrations around 0.9 M.
/
The magnetic saturation, coercivity, and squareness of
electrodeposited NiCo films from chloride and sulfate
baths are shown in Fig. 7. The magnetic saturation (MS)
increased linearly with increasing deposit Co content
(Fig. 7a). The coercivity of Ni electrodeposit from
chloride bath was ꢀ60 Oe. NiCo films electrodeposited
/
from chloride bath exhibited a decrease in coercivity
with increasing Co content up to approximately 55 wt.%
Fe. Higher Co content resulted in an increase in
coercivity. The same trend was also observed when
sulfate bath was used. A minimum coercivity of ꢀ20 Oe
/
was observed for the Ni and Co ratio (50:50). Fig. 7c
Fig. 7. Effect of binary CoNi electrodeposits with deposit Co contents
on magnetic saturation (MS), coercivity (HC) and squareness (S) from
chloride and sulfate baths (a) magnetic saturation, (b) coercivity, and
Fig. 8. XRD patterns of Ni, Co and CoNi electrodeposits: (a) chloride
and (b) sulfate baths.
(c) squareness ([Niꢀ2]ꢃ
0.2 M).
/

826
D. Kim et al. / Electrochimica Acta 48 (2003) 819Á830
/
Fig. 9

D. Kim et al. / Electrochimica Acta 48 (2003) 819Á
/830
827
Fig. 10. Dependence of deposit Co, Fe and Ni contents and current
efficiencies with Coꢀ2 solution concentration in ternary CoNiFe
electrodeposits (a) chloride and (b) sulfate baths ([Coꢀ2]ꢃ
0.2 M).
/
shows the squareness of NiCo thin film alloys from
sulfate and chloride baths. Squareness increased from
ꢀ
/
0.1 to ꢀ
70 wt.% for sulfate and chloride baths.
The NiCo films electrodeposited from chloride and
sulfate baths with 49Á53 wt.% Co (0.08 M Co concen-
tration) exhibited coercivities of ꢀ16 Oe for sulfate and
22 Oe for chloride baths. This film consisted of hcp
(002) [or fcc (111)] and fcc (200) phases with the average
grain size of ꢀ18 nm, as shown in Fig. 8. However, the
NiCo electrodeposit with higher Co content, which have
different coercivities (ꢀ60 Oe for sulfate and ꢀ121 Oe
/
0.7 with increasing deposit Co content for
B
/
ꢀ
/
/
/
ꢀ
/
Fig. 11. Dependence of magnetic saturation (MS), coercivity (HC) and
squareness (S) of ternary CoNiFe electrodeposits with deposit Fe
content from chloride and sulfate baths (a) magnetic saturation, (b)
/
coercivity, and (c) squareness ([Coꢀ2]ꢃ
0.2 M).
/
/
/
chloride bath exhibits higher coercivity (120 Oe) than
the coercivity of 60 Oe from sulfate baths.
for chloride baths), consisted of hcp (100), hcp (002) [or
fcc (111)] and hcp (101) phases with the average grain
Pure Co and NiCo thin films (Fig. 9aÁ
/
c and eÁg)
/
size of ꢀ13 nm. Hence, it is clear that the microstruc-
/
electrodeposited from chloride and sulfate baths exhib-
ited very smooth surface morphologies. Relatively
smooth surface morphologies were observed on the
surfaces of pure Ni electrodeposits (Fig. 9d and h)
from chloride and sulfate baths.
tures of NiCo alloy are independent of anion types.
However, the microstructures changed with increasing
deposit Co content for both chloride and sulfate baths.
The NiCo electrodeposits (containing 0.15 M Coꢀ2
concentration; deposit Co content of ꢀ80 wt.%) from
/
Fig. 9. Surface morphology of Co, Ni and NiCo electrodeposits from chloride [(a)Á
/
(d)] and sulfate baths [(e)Á(f)]; (a) pure Co, (b) 51Ni49Co, (c)
/
22Ni78Co, (d) pure Ni, (e) pure Co, (f) 47Ni53Co, (g) 19Ni81Co, and (h) pure Ni.

828
D. Kim et al. / Electrochimica Acta 48 (2003) 819Á830
/
3.3. CoNiFe alloys
chloride and sulfate baths. Binary NiCo alloys exhibited
high coercivity of ꢀ70 Oe for sulfate and chloride
/
Fig. 10 shows the dependence of deposit Co, Fe and
Ni contents and current efficiency with solution Feꢀ2
concentration during the electrodeposition of CoNiFe
thin films from chloride and sulfate baths. In chloride
baths, with increasing solution Feꢀ2 concentration, and
baths. When chloride bath is used to deposit the
CoNiFe alloy, the presence of iron for deposit Fe
content less than 10 wt.%, caused a sharp decrease in
the coercivity down to 3 Oe. For deposit Fe content in
the range 10Á
15wt.%Fe alloy exhibited the maximum attained coer-
civity of ꢀ26 Oe. Coercivity from sulfate baths was
independent of deposit Fe contents, ꢀ3 Oe. Squareness
/
40 wt.%, the 80wt.%Co-5wt.%Ni-
thus with increasing deposit Fe content up to ꢀ
wt.%, Co content decreased from ꢀ88 to ꢀ50 wt.%
and deposit Ni content remained practically constant.
75% and independent of
/
40
/
/
/
/
from sulfate and chloride baths decreased from 0.6 to
0.2 with increasing deposit Fe content.
80Co5Ni15Fe alloys from chloride bath, which has a
maximum coercivity of 26 Oe, consisted of hcp (100), fcc
(111) and hcp (002) [or bcc (110)] with the average grain
Current efficiency was ꢀ
/
solution Feꢀ2 concentration. In sulfate baths, deposit
Co, Ni and Fe contents have the similar behavior of
observed deposit Co, Ni and Fe contents compared to
those from chloride baths. Current efficiency of binary
CoNi alloys without solution Feꢀ2 concentration was
size of ꢀ22 nm as shown in Fig. 12. The microstruc-
/
tures changed to hcp (002) [or bcc (110)] and bcc (200)
with increasing the solution Feꢀ2 concentration from 15
(0.04 M Feꢀ2 concentration) to 42 wt.% Fe (0.14 M
Feꢀ2 concentration). 76wt.%Co-7wt.%Ni-17wt.%Fe
from sulfate bath, which has a low coercivity of 9 Oe,
consisted of hcp (002) [or bcc (110)] and bcc (200) with
measured to be ꢀ
practically constant, ꢀ
Feꢀ2 concentration up to B
concentration, it decreased to ꢀ
/
52%. However, current efficiency was
75%, with increasing solution
0.06 M. For higher Feꢀ2
52%.
/
/
/
Fig. 11 shows the effect of deposit Fe contents on
magnetic saturation, coercivity and squareness from
the average grain size of ꢀ42 nm. CoNiFe deposits with
/
46 wt.% Fe (0.1 M Fe concentration) exhibited hcp (002)
[or bcc (110)] phases. Therefore, anion type in CoNiFe
alloys influenced on the microstructures and the result-
ing coercivities. The addition of Fe into electrodeposited
CoNi alloys for chloride and sulfate baths caused the
decrease of the coercivity. Binary CoNi from chloride
and sulfate baths showed very smooth surfaces as shown
in Fig. 13. As solution Feꢀ2 concentration in CoNiFe
electrodeposits increased, the decrease of the size of
polyhedral crystallites were observed.
4. Conclusion
4.1. CoFe alloys
Current efficiencies of the electrodeposited CoFe thin
film alloys from chloride baths (ꢀ
dent of the Feꢀ2 solution concentrations. The current
efficiencies from sulfate baths decreased to be B 20%
/
52%) were indepen-
/
ꢀ
/
with increasing the solution Feꢀ2 concentrations. Cur-
rent efficiencies of the electrodeposited CoFe thin film
alloys from chloride baths were significantly higher than
that from sulfate baths. The higher current efficiencies in
chloride baths may be the results of a low H2 limiting
current compared to those from sulfate baths. Magnetic
saturation (MS) of electrodeposited CoFe films both in
chloride and sulfate baths increased from ꢀ
T as deposit Fe content increased. Coercivities of the
electrodeposited CoFe thin films increased from ꢀ20 to
60 Oe in sulfate baths and from ꢀ20 to ꢀ40 Oe in
/
1.5 to ꢀ
/
2.3
/
Fig. 12. XRD patterns of CoNi and CoNiFe electrodeposits (a)
chloride and (b) sulfate baths.
ꢀ
/
/
/

D. Kim et al. / Electrochimica Acta 48 (2003) 819Á
/830
829
Fig. 13. Surface morphology of CoNi and CoNiFe electrodeposits from chloride [(a)Á/(c)] and sulfate baths [(d)Á/(f)]: (a) 80Co20Ni, (b)
80Co5Ni15Fe, (c) 49Co5Ni46Fe, (d) 80Co20Ni, (e) 76Co7Ni17Fe, (f) 49Co5Ni46Fe.
chloride baths as the deposit Co content increased up to
20 wt.%. A minimum squareness of 0.2 was observed in
4.2. NiCo alloys
Current efficiencies (ꢀ
/
75%) from chloride baths were
sulfate baths at ꢀ
squareness of 0.7 was observed in chloride solution at
18 wt.% Co. Anion types affected the magnetic
/
18 wt.% Co, while a maximum
higher than that (25Á45%) from sulfate baths. Magnetic
/
saturation of the alloys with increasing deposit Co
contents exhibited similar linear behavior for chloride
and sulfate baths. The electrodeposits of NiCo thin film
alloys from chloride baths showed higher coercivities
ꢀ
/
properties. However, no effects of anion types on the
phases and surface morphology in CoFe electrodeposits
were observed. Surface morphology were affected by
solution Feꢀ2 concentration.
(120 Oe) than that from sulfate baths (ꢀ60 Oe) at the
/

830
D. Kim et al. / Electrochimica Acta 48 (2003) 819Á830
/
[4] C. Liu, T. Tsao, G.-B. Lee, J.T.S. Leu, Y.W. Yi, Y.-C. Tai, C.-M.
Ho, Sens. Actuators 78 (1999) 190.
deposit Co contents of ꢀ80 wt.%. Electrodeposited
NiCo thin film alloys from both chloride and sulfate
baths exhibited the similar linear behavior of coercivities
/
[5] F.E. Rasmussen, J.T. Ravnkilde, P.T. Tang, O. Hansen, S.
Bouwstra, Sens. Actuators A 92 (2001) 242.
up to ꢀ50 wt.% Co. The microstructure of NiCo
/
[6] T. Osaka, M. Takai, K. Hayashi, K. Ohashi, M. Saito, K.
Yamada, Nature 392 (1998) 796.
[7] S.H. Liao, U.S. Pat. 5,168,410 (1992).
deposits did not depend on anion types, but on deposit
Co contents. NiCo electrodeposits from chloride and
sulfate baths exhibited very smooth surface morphol-
ogy.
[8] X. Liu, G. Zangari, L. Shen, J. Appl. Phys. 87 (2000) 5410.
[9] I. Tabakovic, V. Inturi, S. Riemer, J. Electrochem. Soc. 149
(2002) C18.
[10] R. Chesnutt, J. Appl. Phys. 73 (1993) 6223.
[11] K. Ohashi, Y. Yasue, M. Saito, K. Yamada, T. Osaka, M. Takai,
K. Hayashi, IEEE Trans. Magn. 34 (1998) 1462.
[12] M. Takai, K. Hayashi, M. Aoyaki, T. Osaka, J. Electrochem. Soc.
144 (1997) L203.
4.3. NiCoFe alloys
Current efficiencies from chloride solutions (ꢀ75%)
/
were independent of solution Feꢀ2 concentrations,
whereas those from sulfate solutions gradually de-
[13] K. Hironaka, S. Uedaira, IEEE Trans. Magn. 26 (1990)
2421.
creased from ꢀ
/
75 to ꢀ
50% as solution Feꢀ2 concen-
trations increased. Coercivity of electrodeposited CoNi
thin films from chloride and sulfate baths measured to
/
[14] T. Osaka, M. Takai, Y. Sogawa, T. Momma, K. Ohashi, M.
Saito, K. Yamada, J. Electrochem. Soc. 146 (1999) 2092.
[15] I. Tabakovic, S. Riemer, V. Inturi, P. Jallen, A. Thayer, J.
Electrochem. Soc. 147 (2000) 219.
be ꢀ70 Oe. Coercivities of CoNiFe deposits from
/
[16] H. Dahms, I.M. Croll, J. Electrochem. Soc. 112 (1965) 771.
[17] J. Horkans, J. Electrochem. Soc. 128 (1981) 45.
[18] S. Hessami, C.W. Tobias, J. Electrochem. Soc. 136 (1989) 3611.
[19] W.C. Grande, J.B. Talbot, J. Electrochem. Soc. 140 (1993)
669.
sulfate baths sharply decreased to be 3 Oe with
increasing the deposit Fe contents. The squarenesses in
chloride and sulfate baths showed gradual decreasing
tendency from ꢀ
/
0.6 to ꢀ0.2 in sulfate and 0.1 in
/
[20] W.C. Grande, J.B. Talbot, J. Electrochem. Soc. 140 (1993) 675.
[21] K.Y. Sasaki, J.B. Talbot, J. Electrochem. Soc. 142 (1995) 775.
[22] N. Zech, E.J. Podlaha, D. Landolt, J. Appl. Electrochem. 28
(1998) 1251.
chloride baths. Microstructures depend on anion types
and the addition of Fe into CoNi deposits caused the
decrease of the coercivity. CoNeFe electrodeposits had
the surface morphology of polyhedral crystallites.
[23] N. Zech, E.J. Podlaha, D. Landolt, J. Electrochem. Soc. 146
(1999) 2886.
[24] N. Zech, E.J. Podlaha, D. Landolt, J. Electrochem. Soc. 146
(1999) 2892.
Acknowledgements
[25] T. Osaka, T. Sawaguchi, F. Mizutani, T. Yokoshima, M. Takai,
Y. Okinaka, J. Electrochem. Soc. 146 (1999) 3295.
[26] L. Ricq, F. Lallemand, M.P. Gigandet, J. Pagetti, Surf. Coat.
Tech. 138 (2001) 278.
This work was supported by the DARPA MEMS
program DAB63-99-1-0020 and the Research Grant of
Kwangwoon University in 2001.
[27] X. Liu, G. Zangari, IEEE Trans. Magn. 37 (2001) 1764.
[28] T. Nakanishi, M. Ozaki, H.-S. Nam, T. Yokoshima, T. Osaka, J.
Electrochem. Soc. 148 (2001) C627.
[29] M. Saito, K. Yamada, K. Ohashi, Y. Yasue, Y. Sogawa, T.
Osaka, J. Electrochem. Soc. 146 (1999) 2845.
References
[30] N.V. Myung, K. Nobe, J. Electrochem. Soc. 148 (2001) C136.
[31] S.H. Liao, IEEE Trans. Magn. 26 (1990) 328.
[32] J. Hokans, J. Electrochem. Soc. 126 (1979) 1861.
[33] B.D. Cullity, Elements of X-ray Diffraction, 2nd Ed., Addison-
Wesley Publishing Company Inc, Reading, MA, 1978, p. 91, 501.
[34] M. Hansen, K. Anderk (Eds.), Constitution of Binary Alloys, 2nd
ed., Genium Publishing Corporation, Schenectady, NY, 1989, pp.
[1] A. Chiu, I. Croll, D.E. Heim, R.E. Jones, Jr., P. Kasiraj, K.B.
Klassen, C.D.R.G. Simmons, IBM J. Res. Develop 40 (1996) 283.
[2] P.C. Andricacos, N. Robertson, IBM J. Res. Develop 42 (1998)
671.
[3] E.J. O’Sullivan, E.I. Cooper, L.T. Romankiw, K.T. Kwietniak,
P.L. Trouilloud, J. Horkans, C.V. Jahnes, I.V. Babich, S.
Krongelb, S.C. Hegde, J.A. Tornello, N.C. LaBianca, J.M. Cotte,
T.J. Chainer, IBM J. Res. Develop 42 (1998) 681.
471Á474.
/