Effect of additives on microstructure and mechanical properties of nickel plate/mask fabricated by electroforming process

Article abstract of DOI:10.1149/1.3060625

We fabricated a Ni plate by the electroforming process and evaluated the effects of the addition of brighteners to the electrolyte on its microstructure and mechanical properties. It was found that the microstructure and mechanical properties varied consi

Full text of DOI:10.1149/1.3060625

D108
Journal of The Electrochemical Society, 156 ͑3͒ D108-D112 ͑2009͒
0
013-4651/2009/156͑3͒/D108/5/$23.00 © The Electrochemical Society
Effect of Additives on Microstructure and Mechanical
Properties of Nickel Plate/Mask Fabricated by Electroforming
Process
z
Jun Hyung Lim, Eui Cheol Park, Jinho Joo, and Seung-Boo Jung
School of Advanced Materials Science and Engineering, Sungkyunkwan University, Suwon,
Gyeonggi, Korea
We fabricated a Ni plate by the electroforming process and evaluated the effects of the addition of brighteners to the electrolyte
on its microstructure and mechanical properties. It was found that the microstructure and mechanical properties varied consider-
ably with the different combinations of brighteners added ͑i.e., the addition of class-I, class-II, or both class-I and class-II
brighteners͒. The grain size and degree of crystallization of the Ni plate decreased when the class-I or class-II brightener was
added, while the degree of crystallization decreased even further and the degree of preferred ͑002͒ orientation decreased when both
the class-I and class-II brighteners were added. These results are thought to be due to the “inhibited process,” which was
accelerated by the presence of both additives. In addition, the hardness, yield strength, and wear properties increased when the
brighteners were added. The highest mechanical properties ͑a hardness of 638.2 MPa, yield strength of 1200 MPa, and friction
coefficient of 0.66͒ were obtained when both the class-I and class-II brighteners were added, which is probably due to the fine
grain, poor degree of crystallization, and smooth surface.
©
2009 The Electrochemical Society. ͓DOI: 10.1149/1.3060625͔ All rights reserved.
Manuscript submitted July 2, 2008; revised manuscript received December 9, 2008. Published January 21, 2009.
Electroplating has been widely applied in industry to decorate
material components and improve the corrosion and mechanical
properties of electronic devices. In contrast, the electroforming pro-
cess, which is similar to electroplating, can be applied to the manu-
facture of three-dimensional structures or thicker layers than those
that can be achieved by electroplating. In recent years, electroform-
troforming process and evaluated the effects of the brighteners on its
microstructure and mechanical properties, such as its hardness,
strength, and wear behavior. In addition, in order to explore the
possibility of employing this process for MEMS applications, we
fabricated a Ni metal mask with fine holes and pitch using the elec-
troforming process in combination with photolithography.
1
ing by the LIGA method has been considered to be a key process to
Experimental
fabricate complex structures, such as the devices adopted in micro-
electromechanical system ͑MEMS͒ structures, microprobes, micro-
nozzles, nanocantilevers, and metal masks with a fine pitch, etc.,
because it provides various structures or shapes with a high aspect
ratio and submicrometer precision.
A Ni plate was fabricated by the electroforming process. The
cathode was made from a stainless steel ͑SUS 304͒ plate with di-
mensions of 400 ϫ 400 ϫ 0.1 mm, and the anode consisted of a Ni
ball in a titanium basket. Before the electroforming process, the SUS
substrate was immersed into
a
solution of NaOH
The increasing technological demand for the development of me-
tallic components for these technological areas necessitates Ni elec-
troforming with micro- or nanostructures/plates and well-defined
mechanical properties. It should be noted that the mechanical prop-
erties of electroformed plates are affected by the processing param-
+
Na PO ·12H O + Na CO at 60°C for 5 min, in order to remove
2
3
2
2
3
the surface contaminants.
The electrolyte solutions consisted of Ni-sulfamate
Ni͑NH SO ͒ ·H O, 500 mL/L͔, boric acid ͑H BO , 45 g/L͒, and a
͓
2
3 2
2
3
3
2
3
brightener in distilled water. Two different types of brightener, viz. a
class-I ͑C H NaO S + C H O ͒ and class-II ͑C H NaO S
eters, such as the current density, pH, and electrolyte
4
-8
composition. Among them, the addition of a brightener to the
electrolyte is effective in controlling the crystal nucleation and grain
growth and, hence, results in changes in the microstructure and me-
3
3
3
4
6
2
3
5
3
+ NiCl2·6H2O͒, were selected, and we determined the effect of
these brighteners on the microstructure, hardness, strength, and fric-
tion property of the Ni plate. Table I1 shows the chemical compo-
sitions of the four types of electrolytes used. The electrolyte was
kept at 50°C, and its pH value was set to 4.0. The electrolyte was
purified by filtering the solution through carbon. The current density
8
chanical properties.
In general, the commercial brighteners used in electroplating/
electroforming can be classified into class-I and class-II brighteners,
and C H NaO S and 2-butyne 1,4-diol ͑C H O , BD͒ are com-
3
3
3
4
6
2
7
-9
2
and plating time were 1 A/dm and 4 h, respectively, and the Ni
monly used as class-I and class-II brighteners, respectively. It is
known that the addition of a class-I brightener leads to significant
grain refining in the deposited metal, while a class-II brightener is
usually employed in combination with a class-I brightener because
deposit ͑thickness of 30 ␮m͒ was separated from the substrate after
the electroforming process. The metal plate fabricated in the solu-
tion without a brightener, with the class-II brightener, with the
class-I brightener, and with both the class-I and class-II brighteners
are hereafter denoted as type-0, type-A, type-B, and type-C, respec-
tively.
9
of the compatibility and synergy of the two species.
Substantial research has been conducted to understand the role of
brighteners in the electroplating/electroforming process and the ef-
8
,10
In order to fabricate the metal mask with a fine pitch, additive
patterns were introduced onto the SUS plate through a photolithog-
fects of these two types of brighteners on the microstructure,
surface morphology, and grain orientation
10
11,12
of Ni plates have
been reported. Most of the published studies evaluated the grain
morphology by indirect methods, such as X-ray diffraction ͑XRD͒
analysis or scanning electron microscope ͑SEM͒ surface morphol-
ogy analysis, and only a few by the direct method of transmission
electron microscopy ͑TEM͒ image. In addition, there have been in-
sufficient systematic studies conducted to characterize the effect of
the brighteners on the microstructure and resultant mechanical prop-
erties.
Table I. Chemical composition and condition of the Ni
electrolyte.
Sample
Type 0 Type A Type B Type C
Ni·͑NH SO ͒ ·4H O ͑mL/L͒
500
45
500
45
500
45
20
500
45
20
2
50
4.0
2
3
2
2
H BO ͑g/L͒
3
3
In this study, therefore, we fabricated a Ni plate using an elec-
Class-I ͑mL/L͒
Class-II ͑mL/L͒
Temperature ͑°C͒
pH
2
50
4.0
50
4.0
50
4.0
z
E-mail: jinho@skku.edu

Journal of The Electrochemical Society, 156 ͑3͒ D108-D112 ͑2009͒
D109
Figure 1. Top-view SEM microstructures
of ͑a͒ type-0, ͑b͒ type-A, ͑c͒ type-B, and
d͒ type-C samples.
͑
raphy process. Laminating dry photoresist film, photomasking, ex-
posure, and development procedures were used in this process. The
minimum width and pitch of the additive pattern were 50 and
the surfaces were denser and smoother than those for type-0. AFM
analysis showed that the root-mean-square roughness values were
estimated to be 79.2, 42.1, 9.1, and 7.2 nm, respectively, for types 0,
A, B, and C, indicating that the surface roughness was enhanced
when the class-I or class-II brightener was added and it was further
enhanced when both the class-I and class-II brighteners were added.
TEM images along with the selected area diffraction patterns
1
00 ␮m, respectively, and the remaining photoresist film was
stripped with NaOH solution. A Ni mask with fine holes and pitch
was then made using the same electroforming process as that used
for the Ni plate.
The phase identification of the Ni metal mask was performed by
XRD analysis ͑Bruker-AXS͒. The microstructures and surface
roughness were observed by SEM ͑Hitachi-1000͒, TEM ͑JEM
͑
SADP͒ were obtained to observe the grain morphology, as shown
in Fig. 2. For type 0, the microstructure was found to be crystalline
structure, corresponding to the spot patterns in the SADP, and the
grain size was in the range of 7.2–19.0 nm. For the type-A speci-
men, we observed that the microstructure mainly consisted of a
crystalline structure and that the grain size was slightly decreased to
the range of 4.4–18.2 nm. However, the grains observed in the TEM
images had a broader size distribution. The SADP shows some large
diffraction spots encased in the diffraction ring, which is in agree-
ment with the observed broad grain size distribution. The type-B
sample had a much smaller grain size than the type-A sample: its
grain size was measured to be in the range of 0.8–8 nm. In addition,
the SADP was diffuse in the center and the ring pattern was ob-
served to be less scattered. This suggests that the structure of the
type-B sample had a more nearly amorphous structure and less grain
orientation than the type-A one. The crystalline grain size of the
type-C sample was similar to that of type-B; however, the ratio of
the amorphous structure increased further as shown in Fig. 2d.
It is noted that the grain size all the samples is very small. This
fine structure is reasonable because the grain fabricated by electro-
plating is generally much smaller than that obtained by other pro-
3010͒, and atomic force microscopy ͓͑AFM͒, APEM-1000͔ in con-
tact mode with a detection area of 20 ϫ 20 ␮m. The grain size of Ni
was evaluated by measuring diameters of ϳ10 grains from the TEM
image. The microhardness of the specimen was determined using a
Vicker’s microhardness indenter with a load of 2 kg. The final hard-
ness quoted was the average value of 10 measurements. The yield
strength was measured by the tension test conducted on a universal
testing system ͑Instron 4485͒. The 0.2% offset criterion was used,
and approximately four to six specimens were tested for each mea-
surement. The friction and wear behavior was tested by a ball-on-
disk method under dry sliding conditions. An AISI-52100 steel ball
͑
7 mm diam͒ was used as the counter body. All the tests were per-
formed under a load of 200 g and a sliding speed of 0.1 m/s, and the
friction coefficient and sliding time were recorded automatically.
Results and Discussion
Microstructural evolution of nickel plates with addition of
brighteners.— Figure 1 shows the SEM images of the surface of the
electroformed Ni specimens. A distinct difference was observed in
the microstructures obtained in the different electrolytes. For the
type-0 specimen ͑Fig. 1a͒, a uniform granular-type microstructure
formed without any microcracks/cracks. This morphology is typical
9
cesses, such as casting or powder metallurgy. For type 0, Ni and H
ions combine on the surface of the cathode during the reaction in the
electrolyte and, subsequently, Ni and H2 are absorbed or adhered by
8
for a layer made using the electroplating process. It is likely that,
for the types- A, B, and C specimens, the grains were smaller and
Reactions 1 and 2. H , a by-product, can act as a barrier to grain
2

D110
Journal of The Electrochemical Society, 156 ͑3͒ D108-D112 ͑2009͒
700
600
500
400
300
200
(
a)
Type 0
Type A
Type B
Type C
1600
1400
1200
1000
800
600
400
200
Figure 2. TEM microstructures of ͑a͒ type-0, ͑b͒ type-A, ͑c͒ type-B, and ͑d͒
type-C samples.
(
b)
Type 0
Type A
Type B
Type C
1
1
0
0
0
0
0
.2
.0
.8
.6
.4
.2
.0
Figure 4. ͑a͒ The Vicker’s hardness and ͑b͒ the yield strength of the
samples.
growth and thus inhibit the crystal growth during Ni plating, result-
ing in fine grains in the range of several tens of nanometers for the
type-0 sample
+
−
2
Ni + 2H + 2e → 2Ni–H → 2Ni + H2
͓1͔
ads
+
−
Ni–Hads + H + e → Ni + H
͓2͔
2
(
a)
In addition, further decrease in the grain size resulting from the
addition of the brighteners can be explained as follows. In general,
the purpose of a brightener is to improve the surface activity ͑i.e., it
tends to adhere to or be adsorbed on the cathode surface during
electroplating͒. It is known that the adsorbed agents can also inhibit
grain growth, causing a decrease in the grain size according to the
type-0
type-A
type-B
type-C
0
.6
9
,12
inhibited process.
In the process of adsorption, the agent needs to
0
0
0
0
0
0
.5
.4
.3
.2
.1
be ionized or decomposed into an adsorbent/adherent by-product. In
the case of the class-I brightener, 2-butyne-1,4 diol ͑C H O ͒, can
4
6
2
be transformed into 2-butene-1,4 diol ͑C H O ͒ by the hydrogena-
4
8
2
tion of the unsaturated bonds in the electrolyte by the following
reactions
+
2
−
Ni + 2H O + 2e → Ni͑OH͒ + H
͓3͔
2
2
2
+
HOCH C ϵ CCH OH + 2H → HOCH CH = CHCH OH
2
2
2
2
͓
4͔
It is suggested that the formation of 2-butene-1,4 diol ͑C H O ͒
4
8
2
.0 (b)
activates the ionization of C H NaO S, which adsorbs on the cath-
3
3
3
+
ads
type-0
type-A
type-B
type-C
ode. For the class-II agent, NiCl can be consumed from NiCl and
2
Cl gas during the electroplating, which play the roles of an adsor-
2
Figure 3. ͑a͒ The fwhm of the ͑002͒ plane and ͑b͒ the relative intensity ratio,
I͑111͒/͕I͑111͒ + I͑002͖͒, of the samples.
bent and adherent, respectively, indicating the following possible
reaction

Journal of The Electrochemical Society, 156 ͑3͒ D108-D112 ͑2009͒
D111
Figure 5. The morphologies of the worn
surface of the ͑a͒ type-0, ͑b͒ type-A, ͑c͒
type-B, and ͑d͒ type-C samples.
2
+
−
+
−
+
the interface prefers to grow toward the direction of the high energy
plane ͑i.e., the slowest-growing plane͒. In the case of the type-C
2
NiCl → 2Ni + 4Cl → 2NiCl + 2Cl → 2NiCl + Cl2
2
ads
ads
9
−
+
2e
͓5͔
sample, the C H O solution used for the class-I brightener can
4
6
2
Thus, it is considered that the C H NaO S and NiCl+ together
+
3
3
3
3
5
3
react with both C H NaO S and C H NaO S used as the class-I and
3
3
3
ads
ads
class-II brighteners, respectively, and ionize them, thereby further
increasing the number of adsorbents on the cathode. Addition of
with Cl can act as growth barrier for the class-I and class-II bright-
2
eners, respectively.
both agents can also accelerate the formation of H and Ni͑OH͒
The crystallization and degree of texturing were evaluated using
the full width at half-maximum ͑fwhm͒ of the ͑002͒ plane and the
ratio of the peak intensities of the ͑002͒ plane to the ͑111͒ plane,
respectively, as shown in Fig. 3. We observed that the Ni ͑111͒ and
2
2
from the electrolyte, which tend to adhere to the surface. Thus, it is
expected that more adsorbent was formed on the cathode and their
molecular size had a broader distribution, which interfered with the
growth of the Ni atoms in the crystalline mode by disturbing the
atomic movement/arrangement and the growth along the “free
growth mode,” resulting in the degraded crystallinity and improved
͑111͒ orientation in type C.
͑
002͒ planes formed parallel to the surface of the specimens without
any other planes in all specimens, the relative intensity and the
fwhm of the major peaks differed from one another. The fwhms of
the ͑002͒ peak for the types-0, A, B, and C samples were measured
to be 0.28, 0.30, 0.44, and 0.96°, respectively. This result is consis-
tent with the TEM observations; the degree of crystallization was
degraded in types A and B and degraded further in type C. It is
believed that the degradation of the crystallinity in types- A, B, and
C specimens, as noted earlier, is related partly to the formation of
amorphous structures and the fine grain size in the crystalline struc-
ture.
The relative intensity ratios, I͑111͒/͕I͑111͒ + I͑002͖͒, were found to
be similar; the ratios were 0.13, 0.06, and 0.10, for types-0, A, and B
samples, respectively. The preferred orientation was shown to be the
002͒ plane, which formed parallel to the deposited surface. The
002͒ orientation is known to be the “free growth mode” when Ni is
electroplated. The relative intensity ratio of type C was 0.43, sug-
gesting that the ͑111͒ plane was well developed compared to the
other samples.
Mechanical properties of nickel plates and its application in
metal mask.— In order to evaluate the mechanical properties of the
nickel plates, their microhardness, yield strength, and friction coef-
ficient were measured. Figure 4 shows the Vicker’s hardness and
yield strength of the samples. The hardness values of types 0, A, B,
and C were measured to be 227.7, 294.2, 404.0, and 638.2 HV,
respectively, and the corresponding yield stresses were 449.9, 500.5,
627.5, and 1200.6 MPa. These mechanical properties increased in
the sequence of types 0, A, B, and C, and these improvements are
related to their reduced grain size and degraded crystallization as
shown in Fig. 2 and 3. In addition, it is known that the Ni͑002͒ plane
presents the higher ductility and lower hardness than those of the
͑
͑
1
2
1
3
other preferred orientation planes. The presence of the mixed ori-
entation, ͑002͒ and ͑111͒, is another reason for type C having the
1
3
It is to be noted that the crystallinity and degree of texture of
type C is quite different from those of both type-A and type-B
samples. The significant degradation of the crystallinity and the de-
velopment of the ͑111͒ orientation in type C is probably due to the
synergistic effects of the two brighteners. In general, the crystal
growth occurs by the atomic movement either across the substrate
surface or along the surface ͑lateral diffusion or ledge growth͒ and
highest hardness and strength values. It is also noted that the
strength values had a wider distribution for the samples to which a
wetting agent was added than for type 0. These distributions seem to
be related to the broad grain size distribution or the presence of the
mixed orientation, as discussed in the TEM and XRD analyses.
To evaluate the friction and wear properties, the ball-on-disk
method under dry sliding conditions was used. The friction coeffi-

D112
Journal of The Electrochemical Society, 156 ͑3͒ D108-D112 ͑2009͒
oped, which plastically deformed along the sliding direction. This
type of surface is observed in the case of adhesive wear, which is
produced by the formation and subsequent shearing of the welded
junctions between the two sliding surfaces. For type C, the surface is
relatively flat and the depth and width of the wear grooves are de-
creased. Furthermore, the wear debris was smaller and fewer with
less plastic deformation, compared to the other samples, and this
surface morphology is consistent with its lowest friction coefficient.
The improved wear behavior of this specimen is due to its improved
mechanical properties, which resulted from its microstructure such
as its fine grain, low crystallinity, and smooth surface, as noted
earlier. Therefore, it was thought that the wear resistance of type C
is better than that of the other samples, which would provide it with
a longer service life.
In order to apply this electroforming process to MEMS applica-
tions, we fabricated a Ni metal mask with fine and small holes using
the type-C electrolyte after optimizing the photolithography. Figure
6
shows the SEM image of the Ni mask with holes of 50 ␮m and a
pitch of 150 ␮m. The thickness of the structural layer is currently
limited to 30 ␮m by the thickness of the photoresist film. The image
shows that the surface is smooth and strain-free, and the holes are
round with uniform sidewall geometry. The vertical surface of the
inner hole was also smooth and dense without any microcracks. The
Ni metal mask having the desired 50 ␮m diam geometry was suc-
cessfully fabricated, and the shape and dimensions of the remaining
positive photoresist film were very close to the shape and size of the
holes in the Ni mask ͑i.e., the size of the holes could be controlled to
be Ͻ1 ␮m tolerance limit͒.
Conclusions
A Ni plate was fabricated by the electroforming process, and its
microstructure and mechanical properties were found to vary signifi-
cantly, depending on the different combination of brighteners that
was used. The grain size and degree of crystallization of the Ni plate
decreased when the class-I or class-II brightener was added. When
both the class-I and class-II brighteners were added ͑type C͒, the
degree of crystallization decreased further and the degree of ͑002͒
orientation decreased, while the ͑111͒ orientation was increased. It is
believed that the addition of the brightener forms an adsorbent or
adherent on the cathode surface, which can act as a barrier to grain
growth in crystalline mode, resulting in a decrease in the grain size
and degree of crystallization. The lowest crystallization and the for-
mation of a nonpreferred orientation for type-C are probably due to
the synergistic effect of both brighteners. In addition, the hardness,
yield strength, and wear properties increased when the brightener
was added and were improved further for type-C, which is probably
due to its fine grain, poor crystallization, and smooth surface.
Figure 6. SEM microstructures of the ͑a͒ metal mask with 50 ␮m holes and
100 ␮m pitch and ͑b͒ magnified hole image.
cients of types-0, A, B, and C were estimated to be 0.77, 0.70, 0.72,
and 0.66, respectively. In general, it is known that the wear volume
14
͑
V͒ can be expressed by Archard’s law as follows
Sungkyunkwan University assisted in meeting the publication costs of
this article.
A␮2–4WS
V =
͓6͔
References
H
1
. R. Klein and A. Neyer, Electron. Lett., 30, 1672 ͑1994͒.
where ␮, W, S, and H are the friction coefficient, applied load,
sliding distance, and hardness, respectively. In order to increase the
wear resistance of a substance, the friction coefficient needs to be
decreased and the hardness should be increased.
2. K. P. Wong, K. C. Chan, and T. M. Yue, Surf. Coat. Technol., 115, 132 ͑1999͒.
3
4
. C.-M. Cheng and R.-H. Chen, Microelectron. Eng., 75, 423 ͑2004͒.
. C. S. Lin, P. C. Hsu, L. Chang, and C. H. Chen, J. Appl. Electrochem., 31, 925
͑2001͒.
5
. V. Darrort, M. Troyon, J. Ebothe, C. Bissieux, and C. Nicollin, Thin Solid Films,
The morphology of the worn surface of the samples is shown in
Fig. 5. Type-0 sample shows a typical worn surface resulting from
abrasive wear, and this type of wear generally occurs when a hard
surface slides against and cuts grooves into a softer surface. It was
thought that the harder ANSI 52100 steel ball ͑hardness of
ϳ865 HV͒ contacts the sample at some protruding points and, con-
sequently, scuff and adhesion occur, which hinder the movement
between the Ni sample and the ball. It is believed that this wear
behavior is due to the lower hardness and coarser grain of the type-0
sample, which is the reason for its having the highest friction coef-
ficient.
265, 52 ͑1995͒.
6
7
8
. C. Gu, J. Lian, and Z. Jiang, Scr. Mater., 57, 233 ͑2007͒.
. D. Mockute and G. Bernotiene, Surf. Coat. Technol., 135, 42 ͑2000͒.
. E. A. Pavlatou, M. Raptakis, and N. Spyrellis, Surf. Coat. Technol., 201, 4571
͑2007͒.
9. J. O. Bockris and G. A. Razumney, Fundamental Aspects of Electrocrystallisation,
p. 115, Plenum, New York ͑1967͒.
0. A. A. Rasmussen, R. Møller, and M. A. J. Somers, Surf. Coat. Technol., 200, 6037
1
͑
2006͒.
1. V. M. Kozlov and L. P. Bizelli, Mater. Chem. Phys., 77, 289 ͑2002͒.
12. J. Amblard, I. Epelboin, M. Froment, and G. Maurin, J. Appl. Electrochem., 31,
25 ͑2001͒.
1
9
1
3. J. D. Verhoeven, Fundamentals of Physical Metallurgy, p. 176, John Wiley & Sons,
New York ͑1975͒.
4. E. Rabinowicz, Friction and Wear of Materials, 2nd ed., John Wiley & Sons,
Hoboken, NJ ͑1995͒.
For types A and B, it was shown that microcracks formed on the
wear surface and remarkably more and larger wear debris devel-
1