C136 Journal of The Electrochemical Society, 148 共3兲 C136-C144 共2001兲

0013-4651/2001/148共3兲/C136/9/$7.00 © The Electrochemical Society, Inc.

Electrodeposited Iron Group Thin-Film Alloys

Structure-Property Relationships
N. V. Myung*,z and K. Nobe*
Department of Chemical Engineering, University of California, Los Angeles, California 90095-1592, USA

Various iron group alloys have been electrodeposited and evaluated for properties including corrosion resistance, microstructure,
electrical resistivity, magnetoresistance and other magnetic properties. Corrosion resistance depends on deposit composition and
microstructure, which are controlled by solution composition and deposition variables. Maximum corrosion resistance was ob-
served for 50Ni50Fe and 70Co30Ni binary alloys. The corrosion resistance of electrodeposited CoFe films was an order of
magnitude lower than CoNi and NiFe films. Substantially increased corrosion resistance was obtained by adding Ni to CoFe
alloys. However, addition of B only slightly increased corrosion resistance. For electrodeposited NiFe and CoNi thin film alloys,
three distinct structural regions were observed: for NiFe alloys, fcc, mixed fcc and bcc, and bcc phases as deposit Fe content
increased; for CoNi alloys, fcc, mixed fcc and hcp, and hcp as deposit Co content increased. The smallest grain size was obtained
in the mixed phase region of both NiFe and CoNi alloys. For CoFe alloys, a mixed phase region was not observed; only fcc and
bcc phases were obtained. Electrodeposited NiFe films from sulfate baths show superior soft magnetic properties including both
lower coercivity and higher squareness than from chloride baths. Magnetic saturation (M S) of electrodeposited NiFe films follow
bulk alloys where magnetic saturation increased with increased deposit Fe content.
© 2001 The Electrochemical Society. 关DOI: 10.1149/1.1345875兴 All rights reserved.

Manuscript submitted December 7, 1999; revised manuscript received October 31, 2000.

Electrodeposited magnetic thin films are important in computer rates between 60 and 80% Ni.6 Marcus et al. showed that the anodic
read/write heads and microelectromechanical systems 共MEMS兲 due dissolution rate is affected by the presence of sulfur and depended
to their flexibility, capability, quality, and low cost. Electrodeposited on the crystal orientation.7
permalloy 共80Ni20Fe兲 has been the alloy of choice in thin-film re- There have been several corrosion studies of electrodeposited
cording heads. However, as the areal density of computer drives has nickel and nickel-based alloys such as Permalloy and CoNiFe.
increased dramatically 共60% per year兲, new higher performance soft Frankel et al. compared the corrosion behavior of electrodeposited
magnetic materials are needed. Andricacos and Robertson reviewed and sputter-deposited Permalloy 共80Ni20Fe兲 and found that elec-
requirements for improved thin-film recording heads.1 These in- trodeposited Permalloy was more susceptible to corrosion than bulk
alloys and sputter-deposited film due to inclusion of sulfur from
clude: 共i兲 high magnetic saturation (B s Ⰷ 1 T), 共ii兲 low coercivity
organic additives 共saccharin or thiourea兲, which promoted anodic
(H c ⬍ 1 Oe), 共iii兲 optimal anisotropy field (H k) for high permeabil- dissolution and prevented passive film formation.8 Similar results
ity 共␮兲, 共iv兲 close to zero magnetostriction 共␭兲, 共v 兲 high electrical have been reported for pure nickel electrodeposits.9,10 Saito et al.
resistivity 共␳兲, and 共vi兲 good corrosion resistance. Various CoFe and investigated electrodeposited CoNiFe and found that pitting poten-
CoNi ternary and quaternary alloys have been considered to meet tials shifted to more negative potentials with increasing saccharin
the challenges of improved corrosion resistances with superior soft concentration.11 The corrosion behavior was also affected by depo-
magnetic properties; these included CoFeB, CoFeCr, CoFeP, sition parameters and can be improved by reducing grain size.11
CoFeCu, CoNiFe and CoNiFeS, CoFeNiCr, CoFeSnP, and Co Even though there are many studies on electrodeposited iron-
NiFeB alloys. For example, ternary 80Co10Ni10Fe electrodeposits group alloys, there is a lack of systematic studies relating properties
provided zero magnetostriction with M S of 1.6 T. Magnetic satura- of electrodeposited iron group thin films including NiFe, CoNi,
tion decreased with increasing in Ni content, while corrosion resis- CoFe, and their ternary alloys with deposit composition, grain size,
tance increased.1 Liao and Tolman reported that electrodeposited and crystal structure. In this paper, we investigated and compared
CoFe alloys, especially 90Co10Fe, have promising magnetic prop- the effects of solution compositions and electrodeposition param-
erties including high magnetic saturation 共1.9 T兲, low coercivity 共1 eters on the film compositions and resulting structures and proper-
Oe兲, and near-zero magnetostriction.2 However, binary CoFe alloys ties of iron group alloys such as microstructure, electrical properties,
exhibited poor corrosion resistances. magnetoresistance and other magnetic properties, as well as corro-
sion resistance. Particular emphasis was placed on corrosion resis-
Considerable improvement in the corrosion resistance of bulk
tance, in light of the diminishing length scale of next-generation
NiFe alloys with increased Ni content has been observed. Uhlig
recording materials.
explained this behavior by his electron configuration theory of
passivity.3 Based on this theory, the corrosion rate should decrease
substantially with increased Ni content up to a critical composition
of 34% Ni. For greater than 34% Ni content, the corrosion behavior Experimental
should approximate that of pure bulk nickel. Nickel, iron, cobalt, NiFe, CoFe, CoNi, CoFeNi, and CoFeB
Corrosion studies of bulk NiFe alloys in various electrolytes in- alloys were electrodeposited from various plating solutions. Table I
dicated that corrosion rate is not only dependent on the film compo- gives the plating solution compositions investigated. All solutions
sition but also on the electrolyte. March4 and Landau and Oldach5 contained supporting electrolytes, saccharin to reduce deposit stress,
reported that the corrosion rate of bulk NiFe alloys in sulfuric acid and L’ascorbic acid to minimize Fe⫹2 oxidation. Solution pH was
decreased substantially as the nickel content increased from 0 to adjusted to 3 with either HCl, H2SO4, or NaOH; experiments were
40% and was relatively constant at higher nickel content. Sayano conducted at room temperature with no stirring. In the case of NiFe
and Nobe measured the dependence of corrosion rate of NiFe alloys alloys, two different solutions 共chloride or sulfate electrolytes兲 were
on Ni content in 0.5 M H2SO4 and obtained minimum corrosion used. The effect of the solution Ni⫹2/Fe⫹2 ratio on the deposit com-
position was investigated by varying the Fe⫹2 concentrations from
0.005 to 0.06 M, with Ni⫹2 concentrations set at 0.2 M.
* Electrochemical Society Active Member. The effect of the solution Ni⫹2/Co⫹2 ratio on the deposit com-
z
E-mail: [email protected] position of CoNi alloy films was investigated by varying the Co⫹2
Downloaded on 2018-03-16 to IP 130.88.75.55 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 Journal of The Electrochemical Society, 148 共3兲 C136-C144 共2001兲 C137

Table I. Solution compositions.

A. NiFe 共chloride兲 0.2 M NiCl2 ⫹ X M FeCl2 ⫹ 0.7 M NaCl ⫹ 0.4 MH3BO3 ⫹ 0.0075 M saccharin
⫹ 0.05 M L’ascorbic acid

B. NiFe 共sulfate兲 0.2 M NiSO4 ⫹ X M FeSO4 ⫹ 0.7 M Na2SO4 ⫹ 0.4M H3BO3

⫹ 0.0075 M saccharin ⫹ 0.05 M L’ascorbic acid.

C. CoNi 0.2 M NiCl2 ⫹ X M CoCl2 ⫹ 0.7 M NaCl ⫹ 0.4 MH3BO3 ⫹ 0.0075 M saccharin

D. CoFe 0.3 M CoSO4 ⫹ X M FeCl2 ⫹ 1.0 M NH4Cl ⫹ 0.0075 M saccharin ⫹ 0.05 M L’ascorbic acid

E. CoNiFe 0.2 M NiCl2 ⫹ 0.005 M FeCl2 ⫹ 0.15 M CoCl2 ⫹ 0.7 M NaCl ⫹ 0.4 M H3BO3
80Co10Ni10Fe ⫹ 0.0075 M saccharin ⫹ 0.05 M L’ascorbic acid

45Fe20Co35Ni 0.2 M NiCl2 ⫹ 0.025 M FeCl2 ⫹ 0.02 M CoCl2 ⫹ 0.7 M NaCl ⫹ 0.4 M H3BO3
⫹ 0.0075 M saccharin ⫹ 0.05 M L’ascorbic acid

F. CoFeB 共electrodeposited 0.3 M CoSO4 ⫹ 0.015 M FeCl2 ⫹ 1.0 M NH4Cl ⫹ 0.0075 M saccharin
and electroless兲 ⫹ 0.05 M L’ascorbic acid ⫹ 0.1 to 0.4 M DMAB

concentrations from 0.005 to 0.15 M, with Ni⫹2 concentrations set at In chloride-containing baths, the anode is active, but electrode-
0.2 M. posits have higher internal stress than sulfate or sulfamate bath. In
CoFe plating baths were mixed electrolytes of chloride and sul- sulfate baths, electrodeposits have larger grains and less internal
fate, varying the Fe⫹2 concentrations containing NH4Cl. In the case stress than chloride bath; however, the anode passivates and causes
of electrodeposited CoFeB films, 0.1-0.5 M dimethylamineborane the metal ion concentration and solution pH to change.
共DMAB兲 was added to the CoFe plating solutions. To investigate anion effects on the resulting electrodeposits,
Brass or stainless steel were used as substrates; Fe, Co, or Ni separate chloride and sulfate baths were studied. Figure 1 shows the
were used as soluble anodes. Current density and total charge were dependence of deposit Fe content and current efficiency on Fe⫹2 ion
set at 10 mA cm⫺2 and 10 C cm⫺2, respectively, except for NiFe concentrations. In chloride baths, the deposit Fe content increased
deposition where the current density was reduced to 5 mA cm⫺2. linearly with increased Fe⫹2 ion concentration, whereas deposit Fe
Boron content in CoFeB electrodeposits was determined colori- content increased nonlinearly with Fe⫹2 ion concentration in sulfate
metrically at ␭ ⫽ 575 nm using 1% Alizarin Red S in concentrated baths. In both chloride and sulfate baths, current efficiencies were
sulfuric acid, based on the procedure described in Ref. 12. independent of Fe⫹2 concentrations, although current efficiencies in
Corrosion resistances of the electrodeposited thin-film alloys in chloride baths were higher by about 20%. Horkans studied hydrogen
0.5 M NaCl solutions were determined by impedance spectroscopy gas evolution in Na2SO4, NaCl, and NaClO4 and observed higher
after immersion for 1 h. An EG&G PAR potentiostat/galvanostat limiting currents of H⫹ reduction in Na2SO4 than in either NaCl or
and a Solartron impedance/gain-phase analyzer were used. Imped- NaClO4. The difference in limiting currents was attributed to differ-
ance spectra were collected from 100 kHz to 0.05 Hz at rest poten- ent diffusion coefficients.16,17 She observed a significant decrease in
tials. Prior to corrosion testing, the crystal structure and grain size of the overpotential of Ni, Fe, and NiFe in chloride baths and sug-
the films were determined by X-ray diffraction 共XRD兲. gested that chloride catalyzes deposition of Ni, Fe, and NiFe through
The electrical resistivity and magnetoresistance of the electrode- the formation of an ion bridge between the electrode and the metal
posited films were determined by the standard four-point probe dc
current technique in the presence and absence of an external mag-
netic field of 1500 Oe. Magnetic properties of electrodeposited films
were measured with a vibrating sample magnetometer 共model 1660
ADE Technologies兲.

Results and Discussion

Electrodeposition characteristics, electrical properties, and
anisotropic magnetoresistances.—Electrodeposition of NiFe, CoNi,
and CoFe alloys has been characterized as ‘‘anomalous codeposi-
tion.’’ That is, electrodeposition of pure Ni⫹2 is faster than pure
Fe⫹2, yet during codeposition the rates are reversed.13 The degree of
‘‘anomalous codeposition’’ decreased in the order
CoNi⬎NiFe⬎CoFe. 14,15
NiFe alloys.—NiFe alloys have important magnetic properties. Per-
malloy 共80Ni20Fe兲 is the most common soft magnetic film, which
has zero magnetostriction (␭ ⫽ 0), a large magnetoresistive effect
(⌬␳/␳ AMR ⫽ 5%), relative high magnetic saturation (M S
⫽ 1.0 T) with low coercivity, high magnetic permeability, and low
hysteresis loss. Bulk 50Ni50Fe has a higher magnetic saturation
(M S ⫽ 1.6 T) and electrical resistivity (␳ ⫽ 32 ␮⍀ cm) than Per-
malloy. However, Ni50Fe50 possesses a positive magnetostriction.
NiFe thin films have been electrodeposited from different plating
solutions. The most commonly used electrolytes are chloride, sul- Figure 1. Dependence of Fe deposit content and current efficiency on Fe⫹2
fate, mixed sulfate and chloride 共Watts type兲, and sulfamate. solution concentration: 5 mA cm⫺2, unstirred, 关 Ni⫹2 兴 ⫽ 0.2 M.

Downloaded on 2018-03-16 to IP 130.88.75.55 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 C138 Journal of The Electrochemical Society, 148 共3兲 C136-C144 共2001兲

Figure 2. Dependence of grain size of electrodeposited NiFe on Fe content:

5 mA cm⫺2.

ion being discharged.16,17 Higher current efficiency in chloride baths

compared to sulfate baths in our experiments may be due to a lower
H2 limiting current in chloride baths.
The crystal structure and grain size of electrodeposited NiFe al-
loys were determined by XRD. As the Fe content increased, the
intensity of 共111兲 diffraction peaks 共44.46°兲 decreased while the
peak angle shifted to lower values with peak broadening. When the
deposit Fe content exceeded 50%, a second peak appeared around
44.67° while the intensity of the first peak decreased. When the
deposit Fe content reached 68%, the first peak essentially disap-
peared, while the intensity of the second peak increased. Based on
these observations, electrodeposited NiFe thin-film alloys have three
distinct crystalline regions: fcc, mixed fcc and bcc, and bcc at ⬍50,
60-68, ⬎68% deposit Fe content, respectively, independent of the
type of anions present in solution.
The grain size was calculated from the peak broadening for the
共111兲 plane of fcc and the 共200兲 plane of bcc orientation using the
Scherrer formula.18 Figure 2 shows the dependence of grain size on
deposit composition. In the Ni-rich fcc phase, the grain size de-
creased from 210 to 70 nm as the Fe content increased from 0 to
50%. In the mixed phase region, the grain size was around 70 nm. In
the Fe-rich bcc alloys, the grain size increased from 70 to 150 nm as
the Fe content increased from 74 to 100%. Mixed phase alloy de-
posits from both chloride and sulfate baths had the smallest grain Figure 3. Dependence of 共a兲 electrical resistivities and 共b兲 AMR of elec-
size. Grimmett et al. also observed the smallest grain size 共⬃20 nm兲 todeposited NiFe on Fe content: 共䊉兲 chloride bath, 共䊊兲 sulfate bath, and 共—兲
in the mixed phase region 共62-68% Fe兲 of NiFe deposits from bulk alloy.
sulfamate-chloride baths.19 Iwase and Nasu20 and Aotani21 observed
mixed phase regions at 46-86% and 38-92% Fe, respectively, from
sulfate baths. Ueda and Takahashi deposited NiFe from sulfate baths
containing 3 g/L NaCl at pH 3.4 and found that the mixed phase alloys. Electrical resistivities are indicated as ␳⬜ and ␳ 储 when cur-
region occurred between 51 and 53% Fe and was dependent on rent and magnetic field are perpendicular and parallel to each other,
thickness, shifting to higher iron content with increasing thickness.22 respectively, and ␳ AV ⬅ 1/3␳ 储 ⫹ 2/3␳⬜ . In AMR materials includ-
The electrical resistivity and magnetoresistance of electrodepos- ing Ni, Co, Fe, and their alloys, parallel resistivity (␳ 储 ) and trans-
ited NiFe alloys were measured in an external magnetic field of verse resistivity (␳⬜ ) increase and decrease with increase in mag-
1500 Oe. Figure 3a 共top兲 shows that the electrical resistivity in- netic field, respectively. AMR was strongly dependent on the film
creased monotonically from 5.5 to 32 ␮⍀ cm as Fe content in- composition but not affected by the anions in the bath. A maximum
creased from 0 to 40%. The electrical resistivity of NiFe films with AMR of 9.8% was observed for electrodeposited 85Ni15Fe,
high Fe content 共⬎40%兲 could not be measured due to the stressed, whereas the bulk alloy exhibits a low AMR magnitude of 5%.
cracked deposits. Deposit stress increased with increasing Fe con- McGuire and Potter observed that AMR changed significantly de-
tent. Figure 3b 共bottom兲 compares the anisotropic magnetoresistivity pending on thickness, grain size, and sputtering deposition
共AMR兲 ratio (␳⬜ ⫺ ␳ 储 /␳ AV) at various film compositions with bulk parameters.23
Downloaded on 2018-03-16 to IP 130.88.75.55 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 Journal of The Electrochemical Society, 148 共3兲 C136-C144 共2001兲 C139

Figure 4. Dependence of magnetic properties of electrodeposited NiFe on Fe content: 共䊉兲 chloride bath, 共䊊兲 sulfate bath, and 共 兲 bulk alloy: 共a兲 magnetic
saturation, 共b兲 coercivity, and 共c兲 squareness.

Figure 4a 共top兲 compares the magnetic saturation (M S) of elec- from sulfate baths exhibited softer magnetic properties. Coercivity
trodeposited films and bulk alloys. Magnetic saturation of electrode- was substantially lower (H C ⫽ 1-10 Oe) with much higher square-
posits was independent of anions and follows bulk data as expected, ness (0.1 ⬍ S ⬍ 0.65). Superior soft magnetic films from sulfate
because M S is an intrinsic magnetic property which is only depen- baths may be attributed to lower stress in the deposits.
dent on the film composition. Coercivity (H C) and squareness (S
⫽ M R /M S) are extrinsic magnetic properties which are dependent CoNi alloys.—Figure 5 shows the dependence of deposit Co content
on the film microstructure including grain size, preferred orientation, and current efficiency on Co⫹2 concentration. Deposit Co content
and stress. Figure 4b and c 共top and bottom, respectively兲 shows the increased with increased Co⫹2 concentration. Current efficiency 共ap-
coercivity and squareness of electrodeposited NiFe from chloride proximately 75%兲 was independent of Co⫹2 concentration. For bath
and sulfate baths, respectively. C 共Table I兲, three different crystal regions, fcc, mixed fcc and hcp,
Coercivity and squareness of electrodeposited NiFe from chlo- and hcp were observed at ⬍72, 72-78, and ⬎78% Co, respectively.
ride baths decreased from 56 to 8 Oe and 0.3 to 0.05 as deposit Fe Aotani found that the mixed phase region is dependent on the solu-
content increased to 60 wt %, respectively. Electrodeposited NiFe tion pH; for pH 6.4, the mixed fcc and hcp region was between 70
Downloaded on 2018-03-16 to IP 130.88.75.55 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 C140 Journal of The Electrochemical Society, 148 共3兲 C136-C144 共2001兲

Figure 5. Dependence of deposit Co content and current efficiency on Co⫹2

solution concentration: 10 mA cm⫺2, unstirred, 关 Ni⫹2 兴 ⫽ 0.2 M.

and 100% Co.21 However, at pH 1.2, deposits were fcc throughout

the entire composition range.
Figure 6 shows the dependence of grain size on deposit compo-
sition. In the nickel-rich fcc phase, the grain size of about 140 nm
was fairly constant as the Co content increased from 0 to 50% Co.
However, the grain size dramatically decreased to 50 nm in the
mixed phase region 共72-78% Co兲. Pure cobalt deposits 共hcp兲 had a
grain size of 120 nm.
The electrical resistivity and AMR of electrodeposited CoNi al-
loys are shown in Fig. 7. The electrical resistivity of electrodepos-
ited CoNi exhibited an apparent maximum at around 20 wt % de-
posit Co content. At ⭓50 wt %, Co content the resistivity was
equivalent to bulk alloys. A maximum AMR of 8% was observed
for electrodeposited CoNi of 25% Co, whereas a maximum AMR of
6% had been reported for the bulk CoNi alloy of 22% Co.24

Figure 7. Dependence of 共a兲 electrical resistivities and 共b兲 AMR of elec-

trodeposited CoNi 共Cl⫺ bath兲 on Co content: 共⽧兲 electrodeposit and 共 兲
bulk alloy.

CoFe alloys.—The CoFe films were deposited from mixed sulfate-

chloride solutions 共solution D, Table I兲. Figure 8 shows the deposit
Fe content and current efficiency as a function of Fe⫹2 concentra-
tion. Deposit Fe content increased with increasing Fe⫹2 concentra-
tion. Current efficiency 共⬃82%兲 was independent of Fe⫹2 concen-
tration. A mixed phase region was not obtained as with NiFe and
CoNi deposits. Figure 9 shows distinct fcc and bcc phases at ⬍16
and ⬎16% Fe, respectively. The presence of Fe promoted the fcc
structure, whereas the presence of Ni promoted hcp structures for
Co-rich alloys at the same pH. Figure 9 shows the dependence of
grain size on deposit Fe content. For Co-rich fcc deposits, the grain
size was around 250 nm. For the bcc phase, grain size gradually
Figure 6. Dependence of grain size of electrodeposited CoNi thin films on increased from 50 to 120 nm as deposit Fe content increased from
Co content: 10 mA cm⫺2, unstirred. 10 to 40% Fe. Minimum deposit grain size of ⬃50 nm was obtained
Downloaded on 2018-03-16 to IP 130.88.75.55 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 Journal of The Electrochemical Society, 148 共3兲 C136-C144 共2001兲 C141

Figure 8. Dependence of Fe content and current efficiency 共CoFe, Cl⫺ and

SO4⫺2 bath兲 on Fe⫹2 solution concentration: 10 mA cm⫺2, unstirred,
关 Co ⫹ 2 兴 ⫽ 0.3 M.

at 16-23% Fe. The electrical resistivity and magnetoresistance of

CoFe could not be measured due to the highly stressed and cracked
deposits.
Corrosion resistance.—The corrosion resistances of electrode-
posited iron group alloys in 0.5 M NaCl were measured by imped-
ance spectroscopy. The topography of electrodeposited NiFe after
testing in the corrosive environment was also examined by SEM.
NiFe alloys.—The corrosion resistance increased as the deposit Fe
content increased slightly in the Ni-rich fcc phase, reaching a maxi-
mum at 50% Fe, and then sharply decreased in the Fe-rich bcc phase
共Fig. 10a top兲. In Fig. 11, impedance plots indicate that only one
time constant can be observed for pure Ni 共100 Hz兲 and Fe 共0.1 Hz兲
deposits; however, relaxation processes are apparent at ca. 100 and

Figure 10. Corrosion resistance of 共a, top兲 NiFe, 共b, middle兲 CoNi, and 共c,
bottom兲 CoFe alloys in 0.5 M NaCl after 1 h immersion.

0.1 Hz for the alloys. For the Ni-rich fcc phase, the high-frequency
time constant 共100 Hz兲 is relatively well-defined while the second
time constant becomes more distinct as the Fe content increased. For
the 50Ni50Fe alloys, the distinct relaxation processes were clearly
evident. For the Fe-rich bcc phase, the high-frequency process be-
come less apparent as the Fe content approached 100%.
The corrosion resistance of the Ni-rich fcc phase is an order of
magnitude higher than the Fe-rich bcc phase. The corrosion resis-
tance of electrodeposited alloys were comparable to bulk alloys pre-
Figure 9. Dependence of grain size of electrodeposited CoFe thin films on pared by solidification; the latter also have the highest corrosion
Fe content. resistance at 50% Fe.3
Downloaded on 2018-03-16 to IP 130.88.75.55 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 C142 Journal of The Electrochemical Society, 148 共3兲 C136-C144 共2001兲

ucts were observed; pit size was greater, but pit density was much
smaller than electrodeposited nickel.
CoNi.—As shown in Fig. 10b 共middle兲, the corrosion resistance
increased as the deposit Co content increased, reaching a maximum
at 70% Co and then sharply decreasing in the Co-rich hcp phase.
The Ni-rich fcc phase had an order of magnitude higher corrosion
resistance than Co-rich hcp alloys. The corrosion potentials de-
creased monotonically from ⫺180 to ⫺400 mV with increasing Co
content.
CoFe, CoFeB, and CoNiFe.—As shown in Fig. 10c 共bottom兲, the
corrosion resistance reached a minimum at 16-23% Fe content and
increased as the Fe content increased to 40%. CoFe fcc structures
were more susceptible to corrosion than NiFe fcc structures. The
corrosion potentials of electrodeposited CoFe decreased monotoni-
cally from ⫺405 to ⫺800 mV with increasing Fe content.
Based on the CoFe and CoNi studies, the solution compositions
and deposition conditions for the electrodeposition of ternary
80Co10Ni10Fe and 45Fe20Co35Ni were determined. The addition
of Ni to 90Co10Fe 共80Co10Ni10Fe兲 substantially increased the cor-
rosion resistance; there was only a small reduction in magnetic satu-
ration. Magnetic saturation increased 55Fe45Ni 共45Fe20Co35Ni兲
when Co was added. Liao and Tolman incorporated B into binary
CoFe alloys and obtained either nanocrystalline or amorphous de-
posits depending on the B content,2 possibly increasing the corro-
sion resistance without decreasing magnetic saturation. DMAB 共0.1-
0.4 M兲 was added to our CoFe solutions to electrodeposit CoFeB
ternary alloys. In order to deposit significant amounts of B into
CoFe, the solution pH was increased from 3 to 6. Figure 13 shows
the dependence of deposit B content and corrosion resistance of
electrodeposited CoFeB on DMAB concentration. Both the deposit
boron content and the corrosion resistance increased with increased
DMAB concentration. Figure 14 compares corrosion resistances of
various iron group alloys.

Conclusion
NiFe alloys.—Current efficiencies from chloride baths were
higher than sulfate baths.
Three different regions, fcc, mixed fcc and bcc, and bcc, were
obtained for NiFe alloys as the deposit Fe content was increased.
The mixed phase region was around 60-68% Fe independent of the
type of anions present in solution.
The 50Ni50Fe deposits exhibited the greatest corrosion resis-
tance. The Ni-rich fcc phase had an order of magnitude greater
corrosion resistance than the Fe-rich bcc phase.
The grain size was smallest for the mixed phase region 共70 nm兲.
The electrical resistivity monotonically increased from 5.5 to 32
␮⍀ cm when deposit Fe content increased from 0 to 40%. AMR
was largest for electrodeposited 85Ni15Fe film 共9.8%兲.
Soft magnetic properties of NiFe electrodeposited thin films from
sulfate baths were superior to those from chloride baths.
Figure 11. Dependence of impedance spectroscopy of NiFe deposits on Fe
content: complex plane and Bode plots 共immersion time 5 h, electrolyte 0.5 CoNi alloys.—Current efficiency 共72%兲 was independent of
M NaCl兲: 共a兲 complex plane plots, 共b兲 phase angle Bode plots, and 共c兲 im- Co⫹2 concentration.
pedance plots. 共䊉兲 100 Ni, 共䊊兲 85Ni15Fe, 共䉱兲 70Ni30Fe, 共䉭兲 58Ni42Fe, Three different phase regions, fcc, mixed fcc and hcp, and hcp,
共⽧兲 50Ni50 Fe, 共〫兲 32Ni68Fe, 共䊐兲 28Ni74Fe, and 共䊏兲 100 Fe. were obtained as the deposit Co content was increased. The transi-
tion from fcc to mixed fcc-hcp occurred with deposit Ni content
between 22 and 28% Ni.
The 70Co30Ni deposit exhibited the greatest corrosion resis-
Corrosion potentials of electrodeposited NiFe decreased mono- tance. The Co-rich hcp phase had an order of magnitude lower cor-
tonically from ⫺180 to ⫺800 mV vs. SCE with increasing Fe con- rosion resistance than the fcc phase.
tent. Similar results have been reported by others for NiFe alloys in The smallest grain size was in the mixed phase region 共50 nm兲.
sulfuric acid.22,25,26 AMR was largest for 75Ni25Co deposits 共8%兲.
Pits were observed on nickel with an average size of about 1 ␮m.
SEM micrographs show that pit density decreased and pit size in- CoFe, CoFeNi, CoFeB alloys.—Current efficiency of about 82%
creased with increasing Fe content in the Ni-rich phase 共Fig. 12兲. was independent of Fe⫹2 concentration.
For 50Ni50Fe, pits and corrosion products were not observed. In Corrosion resistance of CoFe alloys were an order of magnitude
electrodeposited Fe thin films, a few pits containing corrosion prod- lower than NiFe and CoNi alloys.
Downloaded on 2018-03-16 to IP 130.88.75.55 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 Journal of The Electrochemical Society, 148 共3兲 C136-C144 共2001兲 C143

Figure 12. SEM micrographs of eletrodeposited Ni, NiFe, and Fe thin films in 0.5 M NaCl after 5 h immersion.

Figure 13. Dependence of B content and corrosion resistances of electrode- Figure 14. Comparison of corrosion resistances of iron group alloys in 0.5
posited CoFeB on DMAB concentration. M NaCl after 1 h immersion.

Downloaded on 2018-03-16 to IP 130.88.75.55 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 C144 Journal of The Electrochemical Society, 148 共3兲 C136-C144 共2001兲

Substantial increase in corrosion resistance was obtained by ad- 10. M. Kouncheva, G. Raichevski, St. Vitkova, and M. Prazak, Surf. Coat. Technol.,
dition of Ni to CoFe deposits. However, the addition of B had only 31, 137 共1987兲.
11. M. Saito, K. Yamada, K. Ohashi, Y. Yasue, Y. Sogawa, and T. Osaka, J. Electro-
a slight effect on the corrosion resistance.
chem. Soc., 146, 2845 共1999兲.
Acknowledgment 12. N. V. Myung, M. Schwartz, and K. Nobe, in Fundamentals of Electrochemical
Deposition and Dissolution, D. Landolt, M. Matlosz, and Y. Sato, Editors, PV
The latter part of this work was supported by the DARPA 99-33, pp. 264-278, The Electrochemical Society Proceedings Series, Pennington,
MEMS Program DABT63-99-1-0020. DMAB was supplied by J. NJ 共1999兲.
Morrison of Bayer. Morton Schwartz read the manuscript and pro- 13. A. Brenner, Electrodeposition of Alloys, Principle and Practice, Vol. II, Chap. 31,
p. 239, Academic Press, New York 共1963兲.
vided useful comments.
14. D. L. Grimmett, M. Schwartz, and K. Nobe, Plat. Surf. Finish, 75, 94 共1988兲.
The University of California, Los Angeles, assisted in meeting the publi- 15. N. H. Phan, M. Schwartz, and K. Nobe, Plat. Surf. Finish, 75, 46 共1988兲.
cation costs of this article. 16. J. Horkans, J. Electrochem. Soc., 126, 1861 共1979兲.
17. J. Horkans, J. Electrochem. Soc., 128, 45 共1981兲.
References 18. B. D. Cullity, Elements of X-ray Diffraction, Chap. 3, Addison-Wesley Publishing
Company, Inc., Reading, MA 共1956兲.
1. P. C. Andricacos and N. Robertson, IBM J. Res. Dev., 42, 671 共1998兲, and refer- 19. D. L. Grimmett, M. Schwartz, and K. Nobe, in Magnetic Materials, Processes, and
ences therein. Devices, L. T. Romankiw and D. A. Harman, Editors, PV 88-23, p. 561, The
2. S. H. Liao and C. H. Tolman, U.S. Pat. 4,756816 共1988兲. Electrochemical Society Proceeding Series, Pennington, NJ 共1988兲.
3. H. H. Uhlig, Corrosion Handbook, p. 194, John Wiley and Sons, Inc., New York
20. K. Iwase and N. Nasu, Bull. Chem. Soc. Jpn., 7, 305 共1932兲.
共1948兲.
21. K. Aotani, Nippon Kessho Gakkaishi, B14, 55 共1950兲.
4. J. S. March, The Alloys of Iron and Nickel, Vol. 1, p. 503, McGraw-Hill Book Co.,
Inc., New York 共1938兲. 22. Y. Ueda and M. Takahashi, J. Phys. Soc. Jpn., 49, 477 共1980兲.
5. R. Landau and C. S. Oldach, Trans. Electrochem. Soc., 81, 521 共1942兲. 23. T. R. McGuire and R. I. Potter, IEEE Trans. Magn., MAG-11, 1018 共1975兲.
6. R. R. Sayano and K. Nobe, Corrosion (Houston), 25, 260 共1969兲. 24. R. M. Bozorth, Ferromagnetism, Chap. 5, D. Van Nostrand Co. New York 共1963兲.
7. P. Marcus, A. Teissier, and J. Oudar, Corros. Sci., 24, 259 共1984兲. 25. G. Economy, R. Speiser, F. M. Beck, and M. G. Fontana, J. Electrochem. Soc.,
8. G. S. Frankel, V. Brusic, R. G. Schad, and J.-W. Chang, Corros. Sci., 35, 63 108, 337 共1961兲.
共1993兲. 26. D. O. Condit, Corrosion and Polarization Studies of Fe-Ni Alloys in H 2 SO 4 at
9. J. Bielinski and J. Przyluski, Surf. Technol., 9, 65 共1979兲. 25°C, Aerospace Research Laboratories, Report No. 70-0082 共1970兲.

Downloaded on 2018-03-16 to IP 130.88.75.55 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).