Influence of the complexing agents on the thickness uniformity of electroless nickel plating in deep-recessed trenches

Article abstract of DOI:10.1149/1.1784211

Electroless plating has been extensively applied to electronic devices because it provides superior throwing power and uniform film formation for complicated geometries. In this study, the formation of the conductive layer in the deep-recessed trenches using electroless Ni-B plating was investigated. Uniformity of plating films was greatly influenced by the selection of the complexing agent. Superior uniformity of Ni films in deep-recessed trenches was obtained by the electroless plating bath containing DL-malic acid. On the other hand, Ni films from a bath containing glycine were not uniformly deposited and no deposition was obtained at the bottom of deep-recessed trenches. Dependence on a selected complexing agent to the initial plating reaction was investigated by the electrochemical analysis. It is confirmed that the induction time before initiation of deposition was influenced by the complexing agent, and adsorbed Pd catalyst in the trenches was dissolved during the induction time and Ni films are not uniformly deposited by the dissolution of adsorbed Pd in the trenches if the plating bath shows the long induction time.

Full text of DOI:10.1149/1.1784211

C610
Journal of The Electrochemical Society, 151 ͑10͒ C610-C613 ͑2004͒
0013-4651/2004/151͑10͒/C610/4/$7.00 © The Electrochemical Society, Inc.
Influence of the Complexing Agents on the Thickness
Uniformity of Electroless Nickel Plating in Deep-Recessed
Trenches
Takanori Tsunoda,a Toshiyuki Okabe,a and Hideo Honmab,
*
b
aGraduate School and Faculty of Engineering, Kanto Gakuin University, 1-50-1 Mutsuurahigashi,
Kanazawa-ku, Yokohama-shi, Kanagawa 236-8501 Japan
Electroless plating has been extensively applied to electronic devices because it provides superior throwing power and uniform
film formation for complicated geometries. In this study, the formation of the conductive layer in the deep-recessed trenches using
electroless Ni-B plating was investigated. Uniformity of plating films was greatly influenced by the selection of the complexing
agent. Superior uniformity of Ni films in deep-recessed trenches was obtained by the electroless plating bath containing DL-malic
acid. On the other hand, Ni films from a bath containing glycine were not uniformly deposited and no deposition was obtained at
the bottom of deep-recessed trenches. Dependence on a selected complexing agent to the initial plating reaction was investigated
by the electrochemical analysis. It is confirmed that the induction time before initiation of deposition was influenced by the
complexing agent, and adsorbed Pd catalyst in the trenches was dissolved during the induction time and Ni films are not uniformly
deposited by the dissolution of adsorbed Pd in the trenches if the plating bath shows the long induction time.
© 2004 The Electrochemical Society. ͓DOI: 10.1149/1.1784211͔ All rights reserved.
Manuscript submitted November 3, 2003; revised manuscript received March 3, 2004. Available electronically September 1, 2004.
Electroplating and electroless plating are actively applied to the
troless Ni-B plating was evaluated from partial polarization curves.
Electrochemical measurement was performed with potentiostat
͑Hokuto Denko, HZ3000͒. The partial anodic polarization behavior
was measured using the nickel bath without nickel sulfate. The par-
tial cathodic polarization behavior was measured using the nickel
bath without DMAB.
formation of metal films for high-density electronic components. In
particular, electroless plating is applied to the fabrication of micro-
components on nonconductors such as silicon, ceramics, glass, and
resin substrates because the metals can be uniformly formed on the
nonconductive substrates.1-6 Among electroless plating methods,
Ni-B plating features good solderability and electric conductivity.7
In this study, applicability of electroless Ni-B plating using
dimethylamine-borane ͑DMAB͒ as a reducing agent to achieve uni-
form nickel deposition on the ceramics comprising deep trenches
was investigated.
Experimental
Electroless nickel plating.—Ceramics (1 ϫ 2 cm͒ with trenches
of aspect ratios of 1, 3, 5, and 7 were used as samples. Figure 1
shows a cross-sectional view of the sample ceramics.
The electroless plating bath was composed of nickel sulfate as a
metal source and DMAB as a reducing agent. As the pattern is
formed on the actual devices, an organic resist is necessary. There-
fore, the electroless nickel plating bath should be operated at mod-
erate conditions to avoid damage to the organic resist. The pH of the
plating bath was adjusted to 6.0 using H2SO4 and NaOH. The tem-
perature was maintained at 60°C. The basic plating bath composi-
tion and operating conditions are shown in Table I. Deposition thick-
ness uniformity was evaluated using electroless nickel plating baths
containing glycine, DL-malic acid, lactic acid, ethylenediamine, or
acetic acid as a complexing agent.
After degreasing and acid treatment, Pd catalysts were deposited
on the sample with sensitizing and activation, and then followed by
the deposition of nickel. The pretreatment process is shown in Table
II.
Evaluation of deposition thickness uniformity.—After the elec-
troless nickel plating, the samples were embeded in epoxy resin and
cross-sectional views of trenches were observed by an optical mi-
croscope.
Figure 2 shows the cross-sectional image of deposited Ni-B films
in the trenches. Thickness of Ni-B films was measured at each few
points around the aperture and the bottom area. The uniformity of
Ni-B film thickness in the trenches was calculated from (B
͕
1
ϩ B ϩ B )/3 / (A ϩ A ϩ A ϩ A )/4 ϫ 100.
͖ ͕
͖
2
3
1
2
3
4
Reactivity of electroless nickel plating.—The reactivity of elec-
Figure 1. Trench patterned ceramics and the cross-sectional view of the
* Electrochemical Society Active Member.
sample ceramics.
Downloaded on 2015-06-08 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, 151 ͑10͒ C610-C613 ͑2004͒
C611
Table I. The basic plating bath composition and operating con-
ditions.
Component
Concentration ͑mol/dm3͒
0.1
0.3-1.0
0.05
NiSO4•6H2O
Complexing agent
DMAB
Adjust pH to 6.0
H2SO4 and NaOH
A standard three-electrode configuration was employed with one
side of the nickel rod as a working electrode, an Ag/AgCl/3M KCl
electrode (E ϭ 0.222 V vs. NHE at 25°C͒ as a reference electrode,
and a Pt plate as a counter electrode. The linear sweep voltammo-
grams were measured at a sweep rate of 10 mV/s, and were scanned
first to the negative and then to the positive direction. The pH of the
plating bath was adjusted to 6.0 with dilute H2SO4 and NaOH. The
pH buffer solution was not used to confirm the clear influence of a
complexing agent.
The nickel disk electrode was used for the measurement of the
electrode potential around the aperture of the trenches. A nickel rod
was buried in PTFE, and the exposed top surface was used as a disk
electrode. The potential of the bottom area was measured using a
trench-bottom electrode with trenches having an aspect ratio of 10.
The trench-bottom electrode was designed for these experiments.
Figure 3 shows the pattern diagrams of these electrodes.
Immersion potential and mixed potential8.—Immersion potential
and mixed potential at the aperture part and the bottom area were
measured to elucidate the initial plating reaction using a disk elec-
trode and the trench-bottom electrode. To avoid the influence of
evolved hydrogen gas due to the electroless plating reaction, the
working electrode was placed upward under the counter electrode.
Figure 2. Uniformity calculation of the Ni deposit in the trench.
bottom region of the trenches above aspect ratio 3. On the other
hand, as shown in Fig. 5, Ni-B films were deposited even at the
bottom of aspect ratio of 7 using DL-malic acid as a complexing
agent.
Comparison of deposition uniformity in the trenches was mea-
sured using glycine, DL-malic acid, acetic acid, lactic acid, or eth-
ylenediamin as a complexing agent in the electroless Ni solution.
Deposition uniformity in the trenches at aspect ratio of 7 was
about 27% from the acetic acid bath, and about 75% from the lactic
acid bath, and about 33% from the ethylenediamine bath. The DL-
malic acid bath showed the highest thickness uniformity ͑82%͒
among these complexing agents.
Analysis of the electrode reaction during the induction
period.—The quartz crystal microbalance ͑QCM͒ method was used
to analyze a reaction mechanism during the initial plating reaction.
QCM were measured with an electrochemical quartz crystal mi-
crobalance by an ALS/CHI model 400 ͑ALS/CH Instruments͒.
Results and Discussion
Influence of various complexing agents to the thickness
uniformity.—Concentration of the chemical ion species in the plat-
ing solution, pH, and temperature are the important factors govern-
ing the thickness uniformity of the deposition. It is believed that the
thickness uniformity is greatly affected by the complexing agent in
the plating solution.
In this way, remarkable differences of thickness uniformity were
indicated among of the complexing agents in the plating bath.
Uniformity of Ni-B plating film thickness in the trenches was
evaluated using various kinds of complexing agents. No dependence
on the complexing agent was noticed and uniform deposits were
formed on the trenches of a aspect ratio of 1. However, significant
differences of the thickness uniformity, in the trenches using aspect
ratios above 3 were confirmed between the complexing agents. Fig-
ure 4 shows the cross-sectional views of deposited nickel films in
the case of glycine bath, no Ni-B film deposits were observed in the
Table II. Pretreatment process.
Cleaning
᭞
Sensitizing (SnCl2 : 0.1 g/dm3͒ 25°C, pH 2.0, 2 min
᭞
Activating (PdCl2 : 0.05 g/dm3͒ 25°C, pH 3.0, 2 min
᭞
Sensitizing (SnCl2 : 0.1 g/dm3͒ 25°C, pH 2.0, 2 min
᭞
Activating (PdCl2 : 0.05 g/dm3͒ 25°C, pH 3.0, 2 min
᭞
Electroless NiB plating 60°C ᭞ Rinse
Figure 3. Schematic diagrams of the electrodes used.
Downloaded on 2015-06-08 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).

C612
Journal of The Electrochemical Society, 151 ͑10͒ C610-C613 ͑2004͒
Figure 4. Ni-B films in the trenches of aspect ratio 1 to 7 by the glycine
bath.
Partial anodic polarization curve9.—To elucidate the reactivity
of the electroless Ni-B plating in the trenches, the electrochemical
technique was employed.
The partial anodic polarization curve was measured using gly-
cine and DL-malic acid bath which showed the most typical differ-
ences in the thickness uniformity.
Figure 6. Partial anodic polarization curves with DL-malic acid and glycine
bath.
As shown in Fig. 6, when the disk electrode which assumed the
aperture part of the trench was used for the measurement, the oxi-
dation peak based on the DMAB of the glycine bath was smaller
than that of the DL-malic acid bath. Note, however, that the oxida-
tion peaks of DMAB decreased using the trench-bottom electrode.
This may be due to the retardation of the ion diffusion at the bottom
of the trenches. The peak potential of the DL-malic acid bath was
not significantly changed, while the oxidation curve of DMAB of
the glycine bath was greatly shifted in the noble direction. There-
fore, it was confirmed that plating rate of glycine bath was greatly
suppressed at the bottom of the trenches.
Electrochemical analysis by the mixed potential8.—The mixed
potential was calculated by the partial anodic polarization and the
partial cathodic polarization. Figure 7 and 8 show the results. As
shown in Fig. 7, the mixed potential and the exchange current den-
sities of the aperture and the bottom area of the trenches were almost
the same using DL-malic acid bath. Therefore, plating progressed at
approximately the same rate on the aperture part and the bottom of
the trenches.
On the other hand, as shown in Fig. 8, mixed potential at the
bottom of the glycine bath shifted to the noble direction, and the
exchange current value was greatly decreased.
Figure 7. Mixed potential curves with DL-malic acid bath.
Thus, the plating rate of the glycine bath at the bottom decreased
remarkably. These electrochemical measurements indicated that the
deposition behavior in the trenches corresponded to each of the elec-
troless Ni-B plating conditions.
Figure 5. Ni-B films in the trenches of aspect ratio 1 to 7 by the DL-malic
acid bath.
Figure 8. Mixed potential curves with glycine bath.
Downloaded on 2015-06-08 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, 151 ͑10͒ C610-C613 ͑2004͒
C613
Figure 9. Immersion potential curves with DL-malic acid bath.
Immersion potential and induction time.—The time dependence
of deposition potential was measured using nickel disk and trench-
bottom electrodes as a working electrode.
Figure 9 and 10 show the immersion potential of nickel elec-
trodes under the each condition. As shown in Fig. 9, the induction
time was about the same on the disk and the trench-bottom elec-
trodes from the DL-malic acid bath. Consequently, the plating reac-
tion initiates simultaneously on the aperture and the bottom parts of
trench by the DL-malic acid bath.
Figure 11. Change of resonance frequency.
On the other hand, the induction time was extended at the trench-
bottom electrode from the glycine bath, as shown in Fig. 10.
It is assumed that the adsorbed Pd catalyst on the bottom area
dissolves into the glycine bath during the induction time. The den-
sity of the catalyst decreases and catalytic reaction will be sup-
pressed. Consequently, plating rate at the bottom is retarded by the
glycine bath.
the plating reaction, Pd may have dissolved from the bottom area of
the trenches during the induction time, terminating the deposition
reaction.
Conclusion
Thickness uniformity of deposits by the electroless nickel plating
was examined using the trench wall construction of high aspect
ratios and the following conclusions were obtained.
Uniformity of the plated films thickness was greatly influenced
by the complexing agents.
Ni film thickness obtained from a glycine bath was not uniform.
No deposition was observed at the bottom of the deep-recessed
trenches.
Superior uniformity of Ni film thickness was obtained by the
plating bath containing DL-malic acid as a complexing agent.
The induction time at the bottom of the trench was influenced by
the complexing agents.
Glycine bath showed longer induction time than DL-malic acid
bath. Pd catalyst might be dissolved during the induction time, caus-
ing less deposition at the bottom of the trench in the glycine bath.
Analysis of the electrode reaction in the induction time.—The
quartz crystal which deposited Pd nuclei by sensitizing and activa-
tion was immersed in the solution containing complexing agents,
and the change of resonance frequency was measured. The result is
shown in Fig. 11 by DL-malic acid and glycine baths.
In the solution containing DL-malic acid, the frequency changed
gradually. On the other hand, in the solution containing glycine, the
frequency rapidly changed in the dissolving direction, compared
with the case of the solution containing DL-malic acid. This phe-
nomenon is due to the dissolution of Pd catalysts from the quartz
crystal.
From the result of Fig. 10, the induction time of the glycine bath
was about 7 min at the trench-bottom electrode. Therefore, before
Kanto Gakuin University assisted in meeting the publication costs of this
article.
References
1. H. Honma, Electrochim. Acta, 47, 75 ͑2001͒.
2. H. Honma and S. Mizushima, J. Met. Finish. Soc. Jpn., 33, 380 ͑1982͒.
3. S. Yamamoto, K. Tashiro, and H. Honma, J. Surf. Finish. Soc. Jpn., 53, 256 ͑2002͒.
4. T. Kobayashi, J. Ishibashi, S. Mononobe, M. Ohtsu, and H. Honma, J. Electro-
chem. Soc., 147, 1046 ͑2000͒.
5. T. Fujinami and H. Honma, Plat. Surf. Finish., 84, 22 ͑1997͒.
6. J. Fils and D. J. Duquette, J. Electrochem. Soc., 131, 254 ͑1984͒.
7. K. Tashiro, K. Ohtaka, S. Yamamoto, S. Kamashima, and H. Honma, J. Jpn. Inst.
Electron. Pack., 5, 591 ͑2002͒.
8. A. M. T. van der Putten and J. W. G. de Bakker, J. Electrochem. Soc., 140, 2221
͑1993͒.
9. I. Ohno, O. Wakabayashi, and S. Haruyama, J. Electrochem. Soc., 132, 2323
͑1985͒.
Figure 10. Immersion potential curves with glycine bath.
Downloaded on 2015-06-08 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).