Electroless nickel plating for nanofabrication in optics

Article abstract of DOI:10.1149/1.1393311

The electroless plating method has played an important role as an indispensable metallization technology for miniaturization of electronic components. This investigation discusses selective metallization on the fine area (nanometer size) by electroless plating. The experimental purpose is to fabricate a probe which is used for scanning near-field optical microscopy. The probe (cone angle 20°, probe size 4 μm) is comprised of an optical fiber covered with a metal film except for an aperture at the apex of the fiber. Nickel-plated probes with an aperture of 100 nm were fabricated by optimization of the plating conditions (dissolved oxygen concentration, bath temperature, and bath pH) and the addition of a catalytic poison into the plating bath.

Full text of DOI:10.1149/1.1393311

1046
Journal of The Electrochemical Society, 147 (3) 1046-1049 (2000)
S0013-4651(98)10-118-0 CCC: $7.00 © The Electrochemical Society, Inc.
Electroless Nickel Plating for Nanofabrication in Optics
T. Kobayashi,a J. Ishibashi,a S. Mononobe,c M. Ohtsu,c,d and H. Honmab,*,z
aGraduate School and bFaculty of Engineering, Kanto Gakuin University, Kanazawa-ku, Yokohama-shi,
Kanagawa 236-0032, Japan
cKanagawa Academy of Science and Technology, Takatsu-ku, Kawasaki-shi, Kanagawa 213-0012, Japan
dInterdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology, Midori-ku, Yokohama-shi,
Kanagawa 226-0027, Japan
The electroless plating method has played an important role as an indispensable metallization technology for miniaturization of
electronic components. This investigation discusses selective metallization on the fine area (nanometer size) by electroless plating.
The experimental purpose is to fabricate a probe which is used for scanning near-field optical microscopy. The probe (cone angle
20Њ, probe size 4 ␮m) is comprised of an optical fiber covered with a metal film except for an aperture at the apex of the fiber.
Nickel-plated probes with an aperture of 100 nm were fabricated by optimization of the plating conditions (dissolved oxygen con-
centration, bath temperature, and bath pH) and the addition of a catalytic poison into the plating bath.
© 2000 The Electrochemical Society. S0013-4651(98)10-118-0. All rights reserved.
Manuscript submitted October 29, 1998; revised manuscript received September 8, 1999.
Miniaturization and high efficiency of electronic devices depend
solution with a composition ratio of 10:1:1 for 90 min. Clad diame-
ter could be reduced to 100, 50, and 25 ␮m by an etching solution
with a composition ratio of 1.7:1:1. Figure 2 shows SEM micro-
graphs of the sharpened fiber probe with a magnified view of the
apex region. The radius of curvature is less than 5 nm and the cone
angle is 20Њ. In this way, controlling the shape of the apex of the
sharpened core can be achieved with high reproducibility. All treat-
ments were performed at 25ЊC.
heavily on advances in semiconductor-processing technologies. Cur-
rently, the semiconductor devices are scaled down to 0.25/0.18 ␮m
as a design rule. Consequently, fine processing technology will enter
into the nanosize fabrication stage in the near future. At present,
thin-film formation by dry processes and pattern formation by litho-
graphic techniques are applied widely. Application of electroless
plating for fine wiring formation on semiconductor devices has been
studied.1-3 In this paper, we discuss the preparation of the fiber probe
for a near-field optical microscope as a nanofabrication technique
using electroless plating.
Resolution of an optical microscope is restricted by the wavelength
of the incident light. According to the theory of the diffraction limit of
Abbe, the limit of resolution is about 60% of the incident wavelength.
However, the resolution of the optical microscope can be improved by
detecting the evanescent field occurring at the vicinity of the surface
of the object. Scanning near-field optical microscopy (SNOM) is real-
ized based on this fact.4,5 The resolution of SNOM depends on the size
of the opening at the probe tip, which detects the evanescent field.
When an optical fiber with a nanosized tip is used as the probe, the
main factor determining the resolution of the SNOM imaging is the
apex size of the probe. The sharp edges of the fiber must be coated
with metal except for the apex region to suppress scattering and gen-
eration of low-spatial-frequency components of the near field. Vacu-
um deposition techniques, such as evaporation or sputtering, have
been reported as methods for coating the fiber probe.6,7 After the fiber
is vacuum-coated with chromium and gold, the coating must be re-
moved from the apex region. Conventionally, the fiber is dipped in an
acrylic resin solution. When the fiber is withdrawn from the resin solu-
tion, a submicrometer apex region of the sharpened core is exposed,
which is then etched by KI-I2 solution. These procedures for probe-tip
preparation are very complicated, however, and reproducibility is low.
As a more efficient and simpler alternative, the applicability of elec-
troless nickel plating for preparation of fine aperture on the protruded
fiber is examined in this report.
Electroless nickel plating process.—The sharpened fiber probes
with flattened apex were coated with nickel by electroless plating.
Pretreatment processes are shown in Table I. The basic bath compo-
sition and operating conditions of the electroless nickel plating bath
are shown in Table II. The impurity in the electroless plating solu-
Experimental
Protruded probe preparation method8.—Optical fibers with
125 ␮m diameter cladding and 2 ␮m diam core, doped with GeO2,
were used. These fibers were sharpened by immersing them in an
etching solution containing NH4F (40 wt %), HF (50 wt %), and
H2O. We denote the volume ratio of etching solution as X, 1:1,
which corresponds to the order of NH4F, HF, and H2O, respectively.
In this study, two types of etching processes were performed as
shown in Fig. 1. For sharpening the core, the fibers were etched in a
* Electrochemical Society Active Member.
z
E-mail: honma@kanto-gakuin.ac.jp
Figure 1. Schematic diagram of sharpening process for fiber probe.
Downloaded on 2015-04-13 to IP 157.182.150.22 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, 147 (3) 1046-1049 (2000)
1047
S0013-4651(98)10-118-0 CCC: $7.00 © The Electrochemical Society, Inc.
Figure 2. SEM micrographs of fine probe for SNOM.
Figure 3. Effect of dissolved oxygen concentration on deposition config-
uration.
tion could have influenced the deposition at the probe surface. To
avoid this, the plating solution was filtrated through the membrane
filter with a pore size of 0.2 ␮m.
The deposition morphology on the fine apex areas were observed
by scanning electron microscopy (SEM). Dissolved oxygen concen-
tration in the plating bath was measured under the plating condition
using a dissolved oxygen (DO) meter (UD-1 type, Central Science
Co., Ltd., resolution 0.01 ppm).
These are many differences between our observations and previ-
ously reported results,10 which revealed that oxygen reduction did
not inhibit metal deposition from an electroless nickel bath without
any stabilizers. It is supposed that the oxygen is reduced only at the
apex area of fiber below the order of submicron for electroless nick-
el plating. Moreover, the condition of nonlinear diffusion for DO is
accelerated in this condition, since only three fiber probes are im-
mersed into the 100 mL of plating bath. Thus, a lot of factors are in-
volved to interpret these differences. Therefore, analysis of the elec-
trochemical activity in this extremely small area is necessary to clar-
ify these contradicting phenomena.
Results and Discussion
Influence of the DO concentration in plating solution.—It is a
well-known fact that the deposition behavior on a small substrate
area is influenced by the DO concentration in the plating bath.9-11
The DO effect has been explained with a mixed potential theory.12-14
Accordingly, we have examined the effect of DO in our plating solu-
tion. We have seen that when DO exceeds over 3 ppm, the entire area
of protruded probe was not plated at all. This is because DO in the
plating solution concentrates at the probe by nonlinear diffusion.
Therefore, reduction of oxygen occurs preferentially and the nickel
deposition reaction is suppressed. Nickel was deposited on the probe
tip by introducing nitrogen gas in the bath before plating (DO
0.7 ppm) or by continuously purging with nitrogen during plating
(DO 0.1 ppm) as shown in Fig. 3. The deposition configurations
might have been delicately influenced by the several deposition-
related factors at the protruded points of the probe, such as the dif-
ferences of DO level, dynamics at the reaction interface, concentra-
tion of reaction species, etc.
From these observations, DO was adjusted at 0.7 ppm by purging
with nitrogen gas before plating to avoid the stirring effect and plat-
ing was performed for 15 min under the static condition in subse-
quent experiments.
Influence of the clad diameter.—We have studied the behavior of
nickel deposition by using the fibers with different sizes of clad
Table I. Experimental conditions.
Optical fiber (␾ 125 ␮m)
d
Preparation for protruded probe
d
Rinsing (25ЊC, 1 min)
d
0.1 g/L SnCl2 (25ЊC, 3 min)
d
Rinsing (25ЊC, 1 min)
Repeated
two times
d
0.05 g/L PdCl2 (25ЊC, 3 min)
d
Rinsing (25ЊC, 1 min)
d
Electroless Ni plating
Table II. Bath composition and plating conditions.
NiSO4и6H2O
CH3COONH4
NaH2PO2иH2O
0.10 M
0.40 M
0.20 M
Temperature
pH
50ЊC
05.0
Figure 4. Influence of clad diameter on deposition configuration.
Downloaded on 2015-04-13 to IP 157.182.150.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

1048
Journal of The Electrochemical Society, 147 (3) 1046-1049 (2000)
S0013-4651(98)10-118-0 CCC: $7.00 © The Electrochemical Society, Inc.
Figure 5. Relationship between plating time and deposition configuration.
Figure 6. Relationship between bath pH and deposition configuration.
diameters, since reaction species and oxygen are concentrated by
nonlinear diffusion at the protruded area. Figure 4 shows the depen-
dence of plating on the clad diameter. Nickel coating was complete
on the protruded area of the fiber probe with a clad diameter of
100 ␮m. However, no nickel deposition was observed up to a length
of 1 ␮m from the apex at the protruded area for a 50 ␮m diam fiber.
Nickel deposition at the protruded area did not occur at all with a
clad diameter of 25 ␮m. These differences in deposition can be ex-
plained by the mixed-potential theory.12-14 From these results we can
conclude that when the clad diameter of the probe is greater than
100 ␮m, DO does not influence nickel deposition. However, the par-
tial anodic current needed for oxygen reduction increases with the
decrease in clad diameter. Therefore, the reduction of oxygen and
the oxidation of reducing agent occur preferentially and nickel re-
duction is terminated.
deposition morphology was obtained between the pH range 4.5-6
and temperature range 45-55Њ. Nickel was not deposited at the apex
area below the bold line, nickel completely covered the apex area
above the line.
Thus, the process window for our aimed shape is very narrow and
reproducibility is low. These deposition differences can be attributed
to various factors like chemical adsorption of reaction species, charge
transfer, electrical double layer, adsorption and desorption of the
From these results, we adjusted the DO to 0.1 ppm and 100 ␮m
diameter of fiber was selected. Figure 5 shows the relationship be-
tween the plating time and deposition configuration. Ideal deposition
configuration was obtained corresponding to 15 min. However, nick-
el gradually reached the apex area and all apex area was completely
covered within 30 min.
Influence of pH and temperature of the plating solution.—The
deposition morphology at the protruded area is greatly influenced by
the condition of the plating solution. Accordingly, we studied the
bath pH dependence of the deposition morphology at the probe tip at
50ЊC for 30 min. As shown in Fig. 6, nickel was not deposited at the
apex of the protruded area below pH 4.5. On the other hand, thick-
ness of deposition gradually increased with increasing pH. Ideal
deposition for our purpose was achieved at the pH range 5-5.5.
Plating bath temperature is also a key factor for the deposition
morphology at the protrusion area. Therefore, the bath temperature
and pH dependence were investigated. As shown in Fig. 7, the ideal
Figure 7. Dependence of pH and temperature on deposition configuration.
Downloaded on 2015-04-13 to IP 157.182.150.22 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, 147 (3) 1046-1049 (2000)
1049
S0013-4651(98)10-118-0 CCC: $7.00 © The Electrochemical Society, Inc.
0.5 ppm in the plating bath. Nickel was deposited on the fiber except
the apex area by the addition of 1 ppm of 2 MBT. This configuration
is ideal for our proposed purpose.
Thus, we confirmed that the addition of inhibitors in the plating
bath which was reported elsewhere is an effective method for con-
trolling the deposition condition even at the extremely fine area such
as the apex of the fiber probe.
Conclusion
Preparation of the optical fiber probe for SNOM was studied as
an application of electroless plating to coat a minute area. The fol-
lowing conclusions were reached
1. Nickel deposition occurred at the protruded area of the probe
when the dissolved oxygen in the plating solution was lowered to
0.1 ppm.
2. Nickel deposition at the protruded area depends on the plating
reaction area. The deposition morphology on the protruded point
was greatly affected by the clad diameter of the fiber probe.
3. Shielding the fiber with electroless nickel, except the apex
area, can be accomplished by the addition of a suitable catalytic poi-
son into the plating solution.
Kanto Gakuin University assisted in meeting the publication costs of this
article.
References
Figure 8. Effect of addition of catalytic poison on deposition configuration.
1. V. M. Dubin, Y. Shacham-Diamand, B. Zhao, P. K. Vasudev, and C. H. Ting,
J. Electrochem. Soc., 144, 898 (1997).
2. H. Watanabe and H. Honma, J. Electrochem. Soc., 144, 471 (1997).
3. S. Abe, T. Fujinami, T. Aono, and H. Honma, Hyomen Gijutsu, 48, 433 (1997).
4. S. Jiang, N. Tomita, H. Ohsawa, and M. Ohtsu, Jpn. J. Appl. Phys., 30, 2107
(1991).
reducing agent, etc. These deposition-related factors at the vicinity of
the protruded reaction site are greatly different from the bulk.
5. S. Jiang, H. Ohsawa, K. Yamada, T. Pangaribuan, M. Ohtsu, K. Imai, and A. Ikai,
Jpn. J. Appl. Phys., 31, 2282 (1992).
Effect of addition of inhibitors.—Our target is to shield the fiber by
electroless nickel deposition except the apex region. To ensure con-
trolability and reproducibility of the process, the effects of different
catalytic poisons were evaluated. Lead (Pb) and 2-mercaptobenzo-
thiazole (2-MBT) were selected as catalytic poisons. Effects of addi-
tion of lead are shown in Fig. 8. Appearance of the nickel deposit on
the probe was not influenced by the addition of Pb below 0.01 ppm.
Nickel deposition was inhibited only at the apex area when the
amount of Pb was raised to 0.1 ppm. The plating reaction is com-
pletely inhibited at the protruded area by the addition of Pb over
6. R. Uma Maheswari, S. Mononobe, and M. Ohtsu, Appl. Opt., 35, 6740 (1996).
7. S. Mononobe, M. Naya, T. Saiki, and M. Ohtsu, Appl. Opt., 36, 1496 (1997).
8. S. Mononobe and M. Ohtsu, J. Lightwave Technol., 14, 2231 (1996).
9. J. W. M. Jacobs and J. M. G. Rikken, J. Electrochem. Soc., 135, 2822 (1988).
10. A. M. T. van der Putten and J. W. G. de Bakker, J. Electrochem. Soc., 140, 2221
(1993).
11. A. M. T. van der Putten and J. W. G. de Bakker, J. Electrochem. Soc., 140, 2229
(1993).
12. M. Paunovic, Plating, 55, 1161 (1968).
13. P. Bindra and J. Tweedie, J. Electrochem. Soc., 130, 1112 (1983).
14. P. Bindra and J. Roldan, J. Electrochem. Soc., 132, 2581 (1985).
Downloaded on 2015-04-13 to IP 157.182.150.22 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).