The relationship between nano crystallite structure and internal stress in Ni coatings electrodeposited by watts bath electrolyte mixed with supercritical CO2

Article abstract of DOI:10.1149/2.061206jes

In this study, the internal stress of Ni coating electroplated in the supercritical-CO₂ (Sc-CO₂) mixed Watts bath was investigated with a focus on the effects of plating parameters and additives. Both the internal stresses of the Ni coatings and their corresponding microstructures were examined. A relationship is found that the internal stress of the Ni coating was inversely proportional to its grain size and the Ni{200}Ni{111} peak fraction in the measured X-ray diffraction pattern. The smaller grain size, accompanied with higher Ni{111} intensity, resulted in the higher internal stress of the coating, and vice versa. A proposed theoretical model, which took both the grain size and preferred crystalline orientation into account, correlated closely with experimental results in the internal stress. As a result, the presence of surfactant in the Sc-CO₂ electrolyte increased the internal stress of Ni coating. However, a reduction of internal stress is made possible by varying four factors: the raising of current density, the increasing of plating temperature, the lowering of plating pressure, and the addition of saccharin into electrolyte. A comparison to the internal stress of Ni coatings plated by the conventional method was also discussed in this study.

Full text of DOI:10.1149/2.061206jes

Journal of The Electrochemical Society, 159 (6) D393-D399 (2012)
D393
0
013-4651/2012/159(6)/D393/7/$28.00 © The Electrochemical Society
The Relationship between Nano Crystallite Structure and Internal
Stress in Ni Coatings Electrodeposited by Watts Bath Electrolyte
Mixed with Supercritical CO
2
V. C. Nguyen, C. Y. Lee, L. Chang, F. J. Chen,c,d and C. S. Linc
a
a,z
b,∗
a
Department of Mechanical Engineering, National Taipei University of Technology, Taipei 106, Taiwan
Department of Materials Science and Opto-electronic Engineering, National Sun Yat-Sen University,
b
Kaohsiung 804, Taiwan
c
Department of Materials Science and Engineering and dDepartment of Mechanical Engineering, National Taiwan
University, Taipei 106, Taiwan
In this study, the internal stress of Ni coating electroplated in the supercritical-CO2 (Sc-CO2) mixed Watts bath was investigated
with a focus on the effects of plating parameters and additives. Both the internal stresses of the Ni coatings and their corresponding
microstructures were examined. A relationship is found that the internal stress of the Ni coating was inversely proportional to its
grain size and the Ni{200}/Ni{111} peak fraction in the measured X-ray diffraction pattern. The smaller grain size, accompanied
with higher Ni{111} intensity, resulted in the higher internal stress of the coating, and vice versa. A proposed theoretical model,
which took both the grain size and preferred crystalline orientation into account, correlated closely with experimental results in the
internal stress. As a result, the presence of surfactant in the Sc-CO2 electrolyte increased the internal stress of Ni coating. However,
a reduction of internal stress is made possible by varying four factors: the raising of current density, the increasing of plating
temperature, the lowering of plating pressure, and the addition of saccharin into electrolyte. A comparison to the internal stress of
Ni coatings plated by the conventional method was also discussed in this study.
©
2012 The Electrochemical Society. [DOI: 10.1149/2.061206jes] All rights reserved.
Manuscript submitted January 12, 2012; revised manuscript received March 5, 2012. Published May 3, 2012.
Recently, the electrodeposition of nickel through the use of su-
percritical carbon dioxide (Sc-CO ) electrolyte has been stated as a
novel method for fabricating the nano-material products. The spe-
cial characteristics of the thus prepared nickel coatings include uni-
formity, high hardness, nano-sized grain, bright appearance, smooth
surface, and high corrosion resistance. Thereby, the well penetration
crystallographic orientation and stress have already been observed
16–18
2
individually in the past.
yet been fully investigated, especially with Sc-CO
ings. Therein, not only hydrogen, the CO at supercritical state has
also been incorporated/occluded into the Ni electrodeposits prepared
However, the combined effects have not
1–7
2
electroplated coat-
2
9
2
by Sc-CO method. Thus, the objective of this work is to study
capability of the Sc-CO
and narrow surface profile of plating template presents an advanta-
geous strength in the fabrication of nano-sized electronic devices,
2
mixed electrolyte/solution into complicated
the probable correlation between the nano-crystalline Ni deposit and
its generated internal stress. Subsequently, the interactive mechanism
within the relationships would be addressed.
8
which would be difficult to achieve with conventional methods. How-
9
ever, in our previous work, the internal stress of Sc-CO
2
Ni coating
Experimental
was found significantly higher than that of conventional one. In the
coating applications, keeping the internal stress of a deposition under
control is crucial for its structural integrity. Hence, there have been
researches devoted to studying the underlined mechanisms as well
as controlling the coating’s internal stress. Some have reduced the
internal stress of a coating by using additives such as saccharin in the
Materials.— The compositions of Watts bath electrolyte and ad-
ditives used for the nickel electrodeposition are listed in Table I. The
non-ionic surfactant used in this work was polyoxyethylene sorbi-
114
tan monolaurate (C58H O26). All the experimental chemicals were
purchased from First Chemical Co. Ltd, Taiwan and employed in
the electrodeposition without further refinement. Carbon dioxide with
minimum purity of 99.9% was acquired from CC Gaseous Corp., Tai-
wan. The copper plate with dimensions of 7 mm wide, 30 mm long,
and 0.1 mm thick, was employed as the working substrate. This long
strip configuration was to facilitate the measurement of curvature in
longitudinal direction while the effect of the curvature in the lateral
direction could be minimized.
10,11
electrolyte and/or controlling plating parameters.
There have been hypotheses proposed to explain the particular ex-
perimental observations in the internal stress accordingly. In the ma-
jority of these hypotheses, the following factors causing the change of
deposit volume are considered: coalescence of crystallites, incorpora-
tion of hydrogen, occlusion of other foreign species, and generation of
structural defects. The theory based on the crystallite joining has been
12
13
14
postulated by Hoffman, Nix and Clemens and Thompson. Basi-
cally, the theory suggested that the deposit forms first in a dispersed
state and then recrystallizes. They assumed that when the individual
crystallites grow, the attractive forces attempt to reduce the gap be-
tween them. However, because the constraint of the substrate does not
allow any free displacement of the crystallites, a stress in the coat-
ing emerges. Moreover, microscopic observations established that this
was also essentially true after the deposit became continuous. Con-
sequently, the continuous coalescence of columnar grains with gaps
between them is necessary for deposit thickening. Since the volume
decrease is resulted from the joining of individual crystallites, one
might expect a strong relationship between the internal stress and
the parameters characterizing the deposit microstructure. In fact, the
tie between grain size and stress, and the close relationship between
Experimental apparatus and method.— The schematic diagram
of the Sc-CO electroplating system used in this work is depicted
2
in Fig. 1. The reaction chamber, which had a volume of ca.180 cc,
15
Table I. Compositions of the electrolyte used for nickel
electrodeposition.
Chemicals
Concentrations
NiSO4.6H2O, g/L
NiCl2.6H2O, g/L
H3BO3, g/L
300
50
30
Surfactant, vol.%
Saccharin, g/L
0, 0.1, 0.5, 1.0, 1.5, 2.0
0, 0.1, 0.2, 0.5, 1.0, 2.0
∗
Electrochemical Society Active Member.
E-mail: leech@ntut.edu.tw
z
Downloaded on 2015-04-12 to IP 129.100.58.76 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D394
Journal of The Electrochemical Society, 159 (6) D393-D399 (2012)
Figure 1. The experimental apparatus used for electroplating with an Sc-CO2 mixed electrolyte. It is labeled as follows: 1. CO2 cylinder; 2-7-10.Valve; 4.Cooler;
.High pressure pump; 3-6-9.Pressure indicator; 8.Power supply; 11. Outlet electrolyte contain; 12.Stirring bar; 13.Agitator; 14.Water pump; 15. Hot water
5
reservoir; 16.Thermometer; 17.Anode; 18.Cathode
was made of stainless steel with inner Teflon lining for electrical
insulation and chemical inertness. The underneath magnetic agitator
and a surrounding temperature controlled water jacket were used for
mixing and maintaining the constant temperature of reaction chamber,
respectively. Inside the reaction chamber, the cathode and anode were
configured in horizontal orientation with a working distance of 20 mm
for allowing adequate fluid sweeping on both electrode surfaces. The
cathode was a copper substrate and the anode was a nickel plate of the
same cross section area. Both the cathode and anode were connected
to the bipolar DC power supply, BP4610 NF Corp., Japan, through
titanium rods to the lead wires outside the chamber.
function. Furthermore, the average thickness of coating was calculated
by the weight gain of the specimen after electroplating measured by
an electronic balance, Mettler Toledo AB204, Switzerland, with a res-
olution of 0.1 mg. Accordingly, the internal stress of nickel coating
21
was calculated by the Stoney equation, as shown in the following:
2
s
1
E
s
t
σ
f
= 6
[1]
R (1 − υ
are the Young’s modulus and Poisson’s ratio of the
substrate material, respectively. In Eq. 1, R is the measured radius of
curvature, while t and t are the substrate and coating thicknesses,
s f
)t
s s
where E and υ
s
f
Prior to deposition, the polished copper substrate was treated se-
quentially in 10 wt% NaOH solution for degreasing, 10 wt% HCl
solution for surface activating and then rinsed in de-ionized water.
The cathode was masked and attached to a Teflon fixture with ex-
respectively. Additionally, in order to reduce the effect of exhausted
22
hydrogen diffusion, the internal stress measurement was taken two
hours after the specimen has been removed from the chamber.
2
posed electrode area about 200 mm . The electroplating solution was
Microstructural examination.— The average grain size of the
deposit was measured by the X-ray Diffraction (XRD), Philips
first put into the reaction chamber. With the chamber cap closed, liq-
uid CO was introduced into the chamber using a high pressure air
2
1
830/Mac, Netherlands, at a sweeping rate of 2 deg/min and cal-
culated by the Scherrer equation with taking the broadening effect of
the XRD machine into account. The X-ray was generated by a Cu-K
driven liquid pump, Haskel DSF-60, U.S., and the chamber was pres-
surized to a predetermined pressure. The solution was then agitated
with a magnetic stirrer at a speed of 400 rpm for at least 30 min. be-
α
target operated at 30 kV and 20 mA with a wavelength of 0.15418 nm.
In order to focus on the effect of crystalline structure on the internal
stress of deposited coating, the curvature of deformed Ni coating and
its corresponding grain size were both measured and calculated on the
same specimen.
19
fore the electroplating process, for solution stabilizing and cathode
20
cleaning.
Each experiment was carried out with fresh solution in the planned
plating parameters shown in Table II. After plating, the copper cathode
with nickel coating was rinsed in deionized water then dried with warm
stream of air. Before every examination and measurement, all samples
Results and Discussion
2
were dry cleaned by Sc-CO .
Fig. 2 presents the internal stresses of the nickel coatings prepared
Internal stress measurement.— After electrodeposition, the spec-
imen geometrical configuration along the central line in the longi-
tudinal direction was measured at seven equi-spaced locations using
a laser displacement probe, KEYENCE LC-2440, Japan. The cur-
vature of the plated specimen in the longitudinal direction was then
determined by curve-fitting the configuration points with a circular arc
2
with Sc-CO and conventional electroplating methods at different de-
position thicknesses. All specimens were plated with the additive-free
Watts bath solution under the condition of 14 MPa pressure, 30 vol.%
◦
2
CO
imens with different coating thicknesses were obtained by controlling
the plating time. Because the cathodic efficiency of the Sc-CO spec-
2
, temperature of 60 C, and a current density of 0.02 A/cm . Spec-
2
imen was lower than the conventional one, slight difference in the
coating thicknesses was observed for each plating time. As a result,
the internal stresses of coatings were all in tensile mode and their
magnitudes decreased as the thickness increased. This should be in-
Table II. Plating parameters used for the Ni electrodeposition.
23
debted to the diminishing structural mismatch and grain coarsening.
Clearly shown in Fig. 2 is the significantly higher internal stress for
Sc-CO electroplating than its conventional counterpart. The plausi-
Parameters
Values
2
CO2 fraction, vol.%
Pressure, MPa
Temperature, C
Agitation, rpm
Current density, A/cm
30
8, 10, 12, 14, 16, 18
40, 45, 50, 55, 60, 65
400
9
ble mechanism has been reported and discussed in our previous work
already. The effects of grain refinement, change in crystallographic
◦
2 2
orientation and the H /CO charging will be discussed later.
Under the special characteristics of supercritical state, one can
imagine that adjusting every plating parameter and electrolyte
2
0.005, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5
Downloaded on 2015-04-12 to IP 129.100.58.76 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, 159 (6) D393-D399 (2012)
D395
4
3
3
2
2
1
1
00
50
00
50
00
50
00
600
14
Sc-CO plating
2
550
Conventional plating
1
3
500
450
400
350
300
250
200
12
1
1
10
9
Internal Stress
Grain size
5
0
0
40
45
50
55
60
65
o
0
1
2
3
4
5
6
7
8
9
10
Temperature, C
Thickness, μm
Figure 4. Effect of plating temperature on internal stress and grain size of Ni
coating prepared by Sc-CO2 electroplating. Plating condition: P = 14 MPa,
Figure 2. Internal stress of Ni coating plated in different thicknesses from
additive-free Watt bath electrolyte by Sc-CO2 and conventional methods, re-
spectively.
2
CD = 0.02 A/cm .
were examined on the coating with approximate thickness of 1 μm.
The results show that the higher the plating pressure, the finer the
grains and the higher the internal stress. On the contrary, Fig. 4 shows
the different effects of plating temperature on the internal stress of Ni
coating. Herein, raising the plating temperature effectively decreased
the internal stress while slightly increased the grain size of coating.
Under constant bath temperature, this pressure effect on the inter-
composition probably alters the characteristics of the obtained elec-
trodeposits. Therein, as mentioned in our previous work, an increase
of CO volume fraction in the reaction chamber introduces more CO
2 2
bubbles into the electrolyte and subsequently changes the size, con-
centration and/or its distribution in the aqueous plating solution. In
fact, for a constant volume of reaction bath, the more fraction of
9
nal stress shows similar trend as that of CO
The result denotes that the higher plating pressure, the more CO
reaction chamber and more dispersed CO is introduced into elec-
trolyte. Thus, the relevant effects of CO in aqueous electrolyte are
enhanced accordingly. More specifically, the higher plating pressure
not only increases the CO concentration but also reduces the size of
CO bubbles in the electrolyte. That explains the increase of elec-
trical conductivity of the Sc-CO solution when plating with higher
Thereby, both the increase of plating pressure
volume fraction results in grains refinement. Accord-
2
volume fraction’s role.
CO
With the same volume of solution and under the constant stirring
speed, the more CO bubbles inside electrolyte results in more ex-
changed covering rate on cathode surface. Hence, the addition of CO
2 2
is equivalent to more soluble CO bubbles in the electrolyte.
2
in
2
2
2
2
bubbles in electrolyte apparently increases the periodic frequency.
Once the on-time is initiated, more hydrogen is evolved when the pH
has to be raised sufficiently so that the nickel electrodeposition can
2
3
2
24
2
recommence. Hence, the higher the periodic frequency, the more
on-time beginnings appear per unit time. Therein, the ions have re-
plenished and the nuclei re-formed at every short on-time interval. On
the other hand, the shorter on-time interval has limited the growing
duration of metal nucleus. Thus, the grains are refined and become
1,25
pressure conditions.
as well as CO
2
ingly, an increasing trend of internal stress with respect to the pressure
was observed.
It is known that the energy of grain nucleus formation depends
one of the reasons for raising internal stress of Sc-CO
2
plating films.
26
on the cathodic overpotential. A large cathodic overpotential re-
duces the energy of nucleus formation, and therefore increases the
The effect of plating pressure and temperature.— In Sc-CO
troplating, the effects of pressure on grain size and internal stress of
Ni coating are shown in Fig. 3. Note that, all the following results
2
elec-
10
nucleus densities. In Sc-CO
2
plating, the electrical conductivity of
the electrolyte improves with the higher temperature. Thereby, the
grain nucleation rate is reduced as a result of decreasing cathodic
23,27
overpotential with increasing the bath temperature.
On the other
hand, at higher temperatures a faster surface diffusion of adatoms
550
500
450
400
350
300
250
200
20
28,29
and nuclei causes an enhanced grain growth.
Thus, raising plating
Internal stress
Grain size
temperature not only increases grain growth rate but also decreases
nucleation rate. Both of them result in coarser grain and subsequently
contribute to the lowering in the internal stress of the Ni coating, as
depicted in Fig. 4.
1
1
1
1
8
6
4
2
The effect of plating current density.— Fig. 5 depicts the rela-
tionship between plating current density and internal stress of the
Ni coatings. The result shows an increase in current density from
2
2
0
.005 A/cm to 0.2 A/cm drastically decreased the internal stress
from 500 MPa to 260 MPa. Further increase of current density just
slightly reduced the internal stress of coating. The increase of current
2
density in Sc-CO electroplating has also been found to coarsen the
3
grain size of Ni deposit in the work of Chang and Sone, and their
results are included in Fig. 5 for comparison.
10
8
10
12
14
16
18
The XRD spectra of the coatings prepared with different current
densities are also presented in Fig. 6. The results illustrate that the
current density played an important role in controlling the Ni {200}
peak intensity of the XRD pattern. Applying higher current density
resulted in stronger Ni {200} peak intensity. On the other hand, the
Pressure, MPa
Figure 3. Effects of plating pressure on internal stress and grain size of Ni
coating prepared by Sc-CO2 electroplating. Plating condition: T = 60 C, CD
◦
2
=
0.02 A/cm .
Downloaded on 2015-04-12 to IP 129.100.58.76 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D396
Journal of The Electrochemical Society, 159 (6) D393-D399 (2012)
5
5
4
4
3
3
2
2
50
24
400
350
26
24
22
20
Internal stress
Grain size from Chang & Sone
Grain size (this work)
00
50
00
50
00
50
00
2
1
1
1
9
6
1
8
5
2
3
2
00
50
200
1
1
50
00
50
0
12
8
Internal stress
Grain size from Chang & Sone
Grain size
4
0
0
.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.00
0
.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50
2
Current density, A/cm
Surfactant, vol.%
Figure 5. Effects of plating current density on internal stress and grain size of
Figure 7. Effects of surfactant concentration on the internal stress and grain
size of Ni coating prepared by Sc-CO2 electroplating. Plating condition: T =
◦
Ni coating prepared by Sc-CO2 electroplating. Plating condition: T = 60 C, P
◦
2
=
14 MPa.
60 C, P = 14 MPa, CD = 0.5 A/cm .
intensity of the Ni {111} peak showed little change. In this work, the
average grain size of the coating was calculated through the diffraction
peak corresponding to Ni {111}. To reduce the noise influence caused
by relative thin coating, the peak parameters were determined after
applying the nonlinear regression of Gaussian function and neglecting
the influence of signal from copper substrate.
increase of surfactant concentration seemed only to raise the internal
stress insignificantly. In contrast, the addition of surfactant in Sc-CO2
electrolyte resulted in finer grains. Similar effect of grain refinement
3
from Chang & Sone was included in Fig. 7 for comparison. The
difference between the grain sizes of this work and Chang & Sone’s
could be caused by the different surfactants used and the corrected
grain size calculation by considering the broadening effect of the XRD
measurement. Clearly shown in Fig. 7, again, is the inversely propor-
tional relationship between the internal stress and the corresponding
grain size of the Ni coating.
Clearly seen in Fig. 6 is that when increasing the plating current
density, the intensity of Ni {200} crystallographic orientation was
developed accordingly. In other words, the increase of current density
resulted in a smaller fraction of Ni{111}/Ni{200}. This typical transi-
tion should be attributed to the fact that higher current density reduces
In Sc-CO plating, the presence of surfactant in the electrolyte
2
2
+
24
concentration of Ni at the electrode surface. The lower ion density
near cathode surface, again, reduces the nucleation rate and results in
coarser grains. Hence, when the current density is increased, the more
Ni {200} accompanying with coarser grain size subsequently lowers
the internal stress of Ni coatings.
further improves the dispersion of CO into aqueous electrolyte with
2
smaller bubble size. Subsequently, the better homogenized electrolyte
with smaller CO bubbles shortens the on-time interval then limits the
2
growing duration of metal nucleus. Therein, the ions have replenished
and the nuclei re-formed at every short on-time interval. Thus, the Ni
grains, again, are refined and the Ni coating possesses higher hardness
7
as well as less pinhole on coating surface. The consistent trends of
The effect of additives in the electrolyte.— The presence of sur-
factant in Sc-CO mixed electrolyte increased the internal stress but
2
the addition of saccharin in solution sharply decreased it, as shown in
Fig. 7 and Fig. 8, respectively. Fig. 7 shows an addition of surfactant
to 0.5 vol.% notably increased the internal stress. However, further
grain refinement in this work and Chang and Sone’s result are depicted
in Fig. 7. The result also shows the inversely proportional trend of the
internal stress with respect to the grain size of coatings.
Meanwhile, as presented in Fig. 8, the addition of saccharin up to
g/L in the electrolyte continued to decrease both the internal stress
2
of Ni deposit and its grain size.
Ni-111
Ni-200
Cu
300
24
Cu
g)
2
2
1
1
2
0
8
6
2
2
1
50
00
50
f)
e)
d)
c)
b)
a)
100
Internal stress
Grain size
14
50
12
42
43
44
45
46
47
48
49
50
51
52
53
54
0.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.00
2
theta, deg.
Saccharin, g/L
Figure 6. XRD patterns of Sc-CO2 Ni coatings plated with different current
Figure 8. Effects of saccharin concentration on the internal stress and grain
2
densities: a) 0.005; b) 0.05; c) 0.1; d) 0.2; e) 0.3; f) 0.4; and g) 0.5 A/cm .
size of Ni coating prepared by Sc-CO electroplating. Plating condition: T =
2
◦
◦
2
Plating condition: T = 60 C, P = 14 MPa.
60 C, P = 14 MPa, CD = 0.5 A/cm .
Downloaded on 2015-04-12 to IP 129.100.58.76 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, 159 (6) D393-D399 (2012)
D397
separated grains together, two times the surface energy is traded for
the grain boundary energy. This effect should be operative from the
beginning of coalescence to the formation of a continuous film. Once
a continuous film has been formed through nucleation, growth, and
coalescence process, thickening is normally expected to occur via
epitaxial growth on pre-existing grains. Practically, Steel et al. had
experimented and obtained a good correlation between the buildup of
the tensile stress and the coalescence of the film. Thereby, the surface
energy of isolated crystallites before coalescence can be estimated by:
Δ
d=2a
15
2
hγsv
E
1
= E
0
+
[3]
a
0
where E is the free energy per unit film area associated with both the
top surface of the film and the film/substrate interface, and the second
term represents the free energy of the side faces of the crystallites,
per unit film area. The term γsv represents for the surface free energy
of the crystallites. By applying Eq. 3 with a consideration of varying
grain diameter within a specific unit film area, the distribution of
estimated surface area associated with grain size is depicted in Fig. 9.
Herein, by taking the gap between Ni adjacent crystals into account, ꢀ
2
.5
5
10
25
50
100
Grain size, nm
Figure 9. Relationship between surface area and grain size per unit film area,
after Nix and Clemens’ criteria.
33
=
0.249 nm, only the results corresponding to grains larger than
nm were plotted. As a result, in a specific unit film area, the smaller
3
There appears to be a general agreement, with both experimental
and theoretical models, that saccharin is the organic additive play-
ing the role of the internal stress reducer as well as grain refiner for
grains that possess more surface area would be able to generate higher
surface energy accordingly. After coalescence, the free energy per unit
13
1
0,30–32
film area becomes
the Ni electrodeposit.
Actually, once saccharin presents in the
plating electrolyte, its molecules are adsorbed on the cathode surface
during electroplating. Consequently, by blocking the surface with the
formation of complex compounds, which increases the frequency of
nucleation and decreases the surface diffusion of nickel ions adsorbed
on cathode surface, the saccharin retards the crystalline growth and
finally results in grain refinement. Furthermore, the molecules are
often incorporated into the metal deposit and the tensile stress is di-
minished accordingly. Thus, the reduction of internal stress and the
refinement in grain size of Ni coating associated with the addition of
ꢀ
ꢁ
2
hγgb
E
ꢀ
E
2
= E
0
+
+
h
[4]
a
1 − υ
2a
where the second term is the grain boundary free energy and the third
term is the strain energy per unit film area. The crystallites are assumed
to be elastic and isotropic with Young’s modulus E and Poisson’s ratio
υ. By setting E
2
−E = 0, Nix and Clemens had derived the following
1
saccharin in electrolyte herein again confirms its function in Sc-CO
metal electroplating. It is shown that the role of saccharin in Sc-CO
2
equation for estimating the stress in the film:
2
electroplating remains unchanged as in its conventional counterpart.
The results shown in this work indicated the sensitive influences on the
internal stress of Ni deposits from the Sc-CO electroplating parame-
2
ters as well as the addition of surfactant and saccharin into electrolyte.
Herein, from the previous results presented in the aforementioned fig-
ures, there is evidence that the grain size and crystallite orientation
have strong correlations with the internal stress of Ni coatings.
ꢂ
ꢅ
1
2
ꢃ
ꢄ
2γsv − γgb
E
σ
max
=
[5]
a
1 − υ
Eq. 4 denotes that the smaller grain possesses higher surface free
energy. Thus, crystallite coalescence can lead to the higher tensile
stresses in the film, Eq. 5. For grains in the order of 10 nm, crystallite
coalescence can be sufficiently high for driving the development of
Relationship between crystallite structure and internal stress.— In
12
13
order to determine the internal stress of thin film, Hoffman started
with the idea that crystallites snap together because the surface free
energies are greater than the free energy of the grain boundary created
by the coalescence. At coalescence of the individual grains, the overall
energy of the system is lowered by reducing surface area, hence surface
energy. Because of the elastic energy of the film substrate system, the
interaction of the substrate does not allow any free displacement;
hence, a tensile stress emerges in the film. The estimated stress of thin
film is inversely proportional to its grain size value as clearly seen in
Eq. 2:
stresses in the grains on the order of GPas. This relationship has
been also observed in the electrodeposited iron-nickel alloy film.17
On the other hand, the film coarsening, coalescence, and grain
growth characteristics depend on its size as well as on the crystallo-
graphic orientation of a grain relative to that of its neighbors.23 When
two adjacent grains possess different crystallographic orientations, the
boundary that connects the two adjacent crystalline grains has energy
of its own. That energy depends on the orientation differences of the
two adjacent grains. The greater mismatch between adjacent grains,
the higher grain boundary energy at the surface of the boundaries.
Moreover, for monocrystalline metal, the Young’s moduli in dif-
ferent lattice directions are different. Young’s modulus of Ni{111}
is 303 GPa, the highest value comparing to 137 GPa and 233 GPa
E
ꢀ
σ = 1
[2]
− υ d
3
3
where E/(1−υ) is the biaxial modulus of the film and d is the grain
size, ꢀ is the gap between adjacent crystallites, as illustrated in
Fig. 9.
for Ni{100} and Ni{110}, respectively. Additionally, the strain en-
23
ergy density of {111} texture film is higher than that of the {100}.
34
Through experiments, Rahman et al. proved that the Young’s mod-
ulus of Ni film plated through Sc-CO method is higher than that of
the conventional one. These likely relate to the higher tensile internal
By considering this criterion, Nix and Clemens13 have figured
out the distribution of crystals in a specific film area with hexagonal
crystallites growing on a substrate. Generally, polycrystalline film
is formed through the nucleation of isolated crystals on a substrate
surface. Coalescence of the islands is one of the mechanisms employed
to explain the occurrence of tensile stress. By joining two formerly
2
18,35
stress in the {111} texture film previously observed.
In this work,
the transition of crystallite orientation associated with varying inter-
nal stress is clearly observed, as depicted in Fig. 6, when Ni{200}
intensity developed with increasing current density.
Downloaded on 2015-04-12 to IP 129.100.58.76 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D398
Journal of The Electrochemical Society, 159 (6) D393-D399 (2012)
6
5
4
3
2
1
00
one becomes reasonable. However, it should be noted that the effect
of the grain refinement of Sc-CO nickel coatings over conventional
Sc-CO2
2
Sc-CO with Saccharin
ones still prevails and causes the higher internal stress in the Sc-CO
coatings presented in Fig. 2.
2
2
00
00
00
00
00
0
Conventional
2
Furthermore, the internal stress of Sc-CO nickel coatings prepared
with the presence of saccharin in plating electrolyte should be eluci-
dated as well. As mentioned previously, the saccharin in the electrolyte
is usually co-deposited into coating. Once Saccharin is deposited into
coating, the molecules that locate around the grain boundaries play
the role as the foreign substance incorporation for lowering tensile
internal stress. The mechanism discussed here is similar to that of
compressive stress generation suggested by Hearne and Floro.39 On
the other hand, impurity-induced suppression of grain growth during
deposition can lead to an increase in the grain boundary energy per
23
volume. Hence, the grain boundary energy might be increased with
the incorporation of saccharin molecules in Ni coating. This can fur-
0
.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
2
ther decrease the parameter K comparing with that of Sc-CO plating
S
without saccharin. Thus, the more saccharin concentration in elec-
trolyte, the higher incorporated density in the coating. Consequently,
the measured internal stress fits better with lower K, as clearly ob-
served by the three isolated triangle points in Fig. 10, in which the
electrolyte were prepared with high saccharin concentration.
Figure 10. Crystal structure and Internal stress relationship of Ni electrode-
posits: K = 2.5 for conventional results, K = 1.0 for the Sc-CO2 results, and K
=
0.5 for the results of Saccharin effects. Films’ thicknesses are approximate
to 1μm.
Conclusions
By taking the effect of crystallographic orientation into account,
Eq. 5 can be modified as:
The experimental results in this work indicated that the internal
stress of the Ni deposit was sensitive to the parameters employed
in the Sc-CO electroplating process. The relationship between the
⎡
⎤
ꢃ
ꢄ
1
2
ꢆ
ꢇ
1
2
2
γ
sv − γgb
E
I111
⎣
⎦
2
σ
max
=
1
2
1
− υ I111 + I200 + I220 + I311
microstructure characteristics, mainly grain size and crystallographic
a
orientation, and the internal stress of nickel electrodeposits, estab-
lished with the proposed theoretical model, has correlated well to the
measurements. Furthermore, the higher plating pressure and the pres-
ence of surfactant in electrolyte increased the internal stress of the Ni
coating, respectively. Nevertheless, the internal stress of the coating
can be reduced by raising plating temperature, and increasing the cur-
rent density. Finally, the effects of saccharin in Sc-CO electroplating
2
on refining grains and reducing tensile stress remain unchanged over
in its conventional counterpart.
[
6]
where Ihkl is intergrated intensity of the {hkl} diffraction peak. The
first bracket term denotes for the effect of surface and grain boundary
energies associated with its grain size and the second one considers
for the effect of anisotropic Young’s modulus of the coating associated
with crystallographic orientation of grains. In order to correlate this
criterion with the experimental results, a parameter S is distilled from
Eq. 6:
ꢆ
ꢇ ꢆ
ꢇ
1
2
K
I
111
S =
[7]
1
2
I
111 + I200
Acknowledgments
a
In the above equation, S is defined as an internal stress parameter,
which accounts for both effects of grain size and its crystallographic
orientation. In the meantime, K is the constant accommodates for the
difference between surface energy and grain boundary energy per unit
film area. By using the parameter S in Eq. 7, the measurements of the
internal stress obtained from different plating methods are presented
in Fig. 10. Because the coatings prepared from different methods
may have different surface energy and grain boundary energy due to
different defects and impurity solution elements, different Ks need to
be employed for different plating methods and material systems. In
this work, K = 2.5 was applied for conventional method; K = 1.0 and
The partial financial support from National Science Council, Tai-
wan under the grant No. NSC 98-2221-E-027 -083 -MY3 is gratefully
acknowledged.
References
1. H. Yoshida, M. Sone, A. Mizushima, H. Yan, H. Wakabayashi, K. Abe, X. T. Tao,
S. Ichihara, and S. Miyata, Surf. Coat. Technol., 173, 285 (2003).
. H. Wakabayashi, N. Sato, M. Sone, U. Takada, H. Yan, K. Abe, K. Mizumoto,
S. Ichihara, and S. Miyata, Surf. Coat. Technol., 190, 200 (2005).
. T. F. M. Chang and M. Sone, Surf. Coat. Technol., 205, 3890 (2011).
4. T. F. M. Chang, M. Sone, A. Shibata, C. Ishiyama, and Y. Higo, Electrochimica Acta,
2
3
5
5, 6469 (2010).
K = 0.5 were implemented for Sc-CO
and without the addition of saccharin into Watts bath electrolyte. With
the chosen parameter S and K , the results in Fig. 10 clearly show the
2
electroplating methods with
5
. H. Yan, M. Sone, A. Mizushima, T. Nagai, K. Abe, S. Ichihara, and S. Miyata, Surf.
Coat. Technol., 187, 86 (2004).
s
6
. S. T. Chung, H. C. Huang, S. J. Pan, W. T. Tsai, P. Y. Lee, C. H. Yang, and M. B. Wu,
Corros. Sci., 50, 2614 (2008).
. S. T. Chung and W. T. Tsai, J. Electrochem. Soc., 156, D457 (2009).
. A. H. Romang and J. J. Watkins, Chem. Rev., 110, 459 (2010).
. V. C. Nguyen, C. Y. Lee, F. J. Chen, C. S. Lin, and T. Y. Liu, Surf. Coat. Technol.,
internal stress of the coating is closely related to the parameter S.
In the conventional nickel electroplating results, the incorporation
of hydrogen around grain boundaries (1–2 nm) had been observed
7
8
9
36
through the TEM micrograph. Once the occluded hydrogen diffused
and escaped from the coating, the tensile internal stress was increased
2
06–14, 3201 (2012).
10. A. M. Rashidi and A. Amadeh, Surf. Coat. Technol., 204, 353 (2009).
1. Y. Xuetao, W. Yu, S. Dongbai, and Y. Hongying, Surf. Coat. Technol., 202, 1895
1
6,32
1
accordingly.
ously, H /CO
inclusion of more H
the grain boundary energy of coating. Moreover, the surface energy
In the Sc-CO
was occluded into electrodeposits during plating. The
/CO bubbles around grain boundaries increases
2
electroplating, as mentioned previ-
(2008).
2
2
1
1
2. R. W. Hoffman, Thin Solid Films, 34, 185 (1976).
3. W. D. Nix and B. M. Clemens, J. Mater. Res., 14–8, 3467 (1999).
14. C. V. Thompson and R. Carel, J. Mech. Phys. Solids, 44–5, 657 (1996).
5. S. C. Steel, C. V. Thompson, S. J. Hearne, and J. A. Floro, J. Appl. Phys., 88, 7079
2000).
6. C. S. Lin, C. Y. Lee, F. J. Chen, and W. C. Li, J. Electrochem. Soc., 152–6, 370
2005).
17. F. Czerwinski, J. Electrochem. Soc., 143–10, 3327 (1996).
8. L. Chang, C. H. Chen, and H. Fang, J. Electrochem. Soc., 155–1, D57 (2008).
2
2
37
38
1
of Ni {111} is smaller than those of Ni {100} and Ni {110}. Thus,
the Sc-CO
nickel coatings that possess less Ni {200} than the con-
ventional ones
the definition of K, i.e. the difference between surface energy and grain
boundary energy, a smaller K for Sc-CO plating than conventional
(
2
1
6,7,9
should have lower surface energy. By referring to
(
1
2
Downloaded on 2015-04-12 to IP 129.100.58.76 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, 159 (6) D393-D399 (2012)
D399
1
2
2
9. S. R. P. d. Rocha, P. A. Psathas, E. Klein, and K. P. Johnston, J. Colloid Interface
Sci., 239, 241 (2001).
0. H. Yoshida, M. Sone, A. Mizushima, K. Abe, X. T. Tao, S. Ichihara, and S. Miyata,
Chem. Lett., 020579, 1086 (2002).
1. G. C. A. M. Janssen, M. M. Abdalla, F. van Keulen, B. R. Pujada, and B. van Venrooy,
30. R. Weil, Plating, 57, 1231 (1970) and 58, 50–56, 137–146 (1971).
31. S. H. Kim, H. J. Sohn, Y. C. Joo, Y. W. Kim, T. H. Yim, H. Y. Lee, and T. Kang, Surf.
Coat. Technol., 199, 43 (2005).
32. S. Armyanov and G. S. Chakarova, J. Electrochem. Soc., 139, 3454 (1992).
33. M. A. Meyers and K. K. Chawla, Mechanical Behavior of Materials P. 275, Prentice
Hall, New Jersey, US, 1999.
Thin Solid Films, 507, 1858 (2009).
2
2
2
2
2
2
2
2
2. S. J. Hearne and J. A. Floro, J. Appl. Phys., 97, 014901.1 (2005).
3. C. V. Thompson, Annu. Rev. Mater. Sci., 30, 159 (2000).
34. Md. Z. Rahman, M. Sone, M. Eguchi, K. Ikeda, S. Miyata, and T. Yamamoto,
International Conference on Mechanical Engineering, pp. 28–30 (Dec, 2005).
35. S. P. Kim, H. M. Choi, and S. K. Choi, Thin Solid Films, 322, 298 (1998).
36. F. D. Torre, H. V. Swygenhoven, and M. Victoria, Acta Materialia, 50, 3957
(2002).
37. Y. Y. Huang, Y. C. Zhou, and Y. Pan, Physica. B, 405, 1335 (2010).
38. L. Vitos, A. V. Ruban, H. L. Skriver, and J. Koll a´ r, Surface Science, 411, 186 (1998).
39. S. J. Hearne and J. A. Floro, J. Appl. Phys., 97, 014901.1 (2005).
4. A. M. El-Sherik, U. Erb, and J. Page, Surf. Coat. Technol., 88, 70 (1996).
5. M. S. Kim, J. Y. Kim, C. K. Kim, and N. K. Kim, Chemosphere, 58, 459 (2005).
6. E. Budevski, G. Staikov, and W. J. Lorenz, Electrochimica Acta, 45, 2559 (2000).
7. H. Natter and R. Hempelmann, Z. Phys. Chem., 222, 319 (2008).
8. H. Natter, M. Schmelzer, and R. Hempelmann, J. Mater. Res., 13, 1186 (1998).
9. J. X. Kang, W. Z. Zhao, and G. F. Zhang, Surf. Coat. Technol., 203, 1815 (2009).
Downloaded on 2015-04-12 to IP 129.100.58.76 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).