Effect of process variables on electrotinning in a methanesulfonic acid bath

Article abstract of DOI:10.1149/1.1393315

The electrodeposition of tin in a methanesulfonic acid-based plating solution as studied. The effects of agitation, current density, and temperature on the plating efficiency and coating morphology were examined using a rotating cylinder cathode and a rotating disk electrode. A good correlation was found between the plating efficiency and coating morphology; moreover, both of them were affected by the diffusion-limited current density (i_L) of the stannous ion. A compact crystalline deposit was obtained with >95% plating efficiency at current densities up to i_L in this plating bath. When the current density exceeds i_L, not only does the plating efficiency begin to decrease but also the coating changes to micronodular and then to dendritic. Through the determination of the temperature dependence of the viscosity of the bath and the diffusion coefficient of the stannous ion, the synergistic effects of temperature and agitation on i_L were quantified.

Full text of DOI:10.1149/1.1393315

Journal of The Electrochemical Society, 147 (3) 1071-1076 (2000)
1071
S0013-4651(99)08-106-9 CCC: $7.00 © The Electrochemical Society, Inc.
The Effect of Process Variables on Electrotinning in a
Methanesulfonic Acid Bath
Yung-Herng Yau
Bethlehem Steel Corporation, Bethlehem, Pennsylvania 18016, USA
The electrodeposition of tin in a methanesulfonic acid-based plating solution was studied. The effects of agitation, current densi-
ty, and temperature on the plating efficiency and coating morphology were examined using a rotating cylinder cathode and a rotat-
ing disk electrode. A good correlation was found between the plating efficiency and coating morphology; moreover, both of them
were affected by the diffusion-limited current density (i ) of the stannous ion. A compact crystalline deposit was obtained with
L
>
95% plating efficiency at current densities up to i in this plating bath. When the current density exceeds i , not only does the
L
L
plating efficiency begin to decrease but also the coating changes to micronodular and then to dendritic. Through the determination
of the temperature dependence of the viscosity of the bath and the diffusion coefficient of the stannous ion, the synergistic effects
of temperature and agitation on i were quantified.
L
©
2000 The Electrochemical Society. S0013-4651(99)08-106-9. All rights reserved.
Manuscript submitted August 26, 1999; revised manuscript received November 9, 1999.
The electrotinning process based on methanesulfonic acid
MSA) has gained acceptance in continuous steel strip plating
lines after years of applications in the electronics industry. In the
reach 3000 revolutions per minute (rpm) without causing a vortex.
The shaft of the RCC was driven by a rotator, model AFASR from
Pine Instrument Company. The plating current was supplied by a
10 A dc power supply, model MSK 10-10M from Kepco Inc.
The steel coupon was cleaned, weighed, then mounted on the
rotating cylinder cathode. The assembled electrode was electrolyti-
cally cleaned in an alkaline cleaner (Elf Atochem SM 92L, 25 g/L,
82ЊC), rinsed in hot tap water, activated in a 5% sulfuric acid solu-
tion at ambient temperature, and rinsed in hot tap water again fol-
lowed by distilled water. Three rotation speeds of 500, 1000, and
2000 rpm were used because they provided a reasonable simulation
(
1
-3
reel-to-reel plating of electronics applications, the line speed is about
2
0
.1 m/s, the current density is about 2000 A/m , and the stannous ion
3
4
concentration is in the range of 420 to 840 mol/m (50 to 100 g/L).
In contrast, in a continuous steel strip plating line, the steel strip trav-
els at speeds from 2 to 10 m/s while the current density varies from
2
1
000 to 6500 A/m to encompass the entire product mix of various
steel grades and coating weights. Furthermore, the stannous ion con-
3
centration is much lower, often in the range of 84 to 168 mol/m to
7
minimize the loss of costly chemicals through leaks and drag-out in
a continuous steel strip plating line. Because of the differences in
operating conditions, the production experiences from the reel-to-
reel plating need to be expanded to be used for the continuous strip
plating line. Although a guideline on the interactive effects between
to production conditions. The current density was varied from 60 to
2
6000 A/m . The total charge transferred was kept constant at
2
18200 C/m , which, at 100% plating efficiency, would produce a
2
coating weight of 11.2 g/m , typical of the heaviest coating com-
monly used for commercial tinplate. After plating, the assembly was
rinsed in hot water; the steel coupon was then dismounted, quenched
in acetone, dried in air, and weighed again. The plating efficiency
was calculated from the weight gain. The coating morphology was
examined by scanning electron microscopy (SEM) using an Amray
3200-C microscope.
5
processing parameters had been established, the basic information
on the transport properties of the MSA bath was still lacking. The
purpose of this study was to examine how the deposition process is
affected by mass transfer under the hydrodynamic conditions creat-
ed by the fast moving strip. Such an understanding is critical for
proper process control.
A demountable rotating disk electrode (RDE) from Lucent Tech-
nologies was used for the electrochemical measurements so that the
resulting coating morphology was readily available for examination.
The working electrode on the RDE was made of copper with a sur-
Experimental
A patented MSA process, Ronastan-TP-HCD from Shipley
6
2
Ronal, was used in this study. The bath consisted of 50 mL/L stan-
face area of 0.5 cm . A platinum wire and a saturated calomel elec-
nous methanesulfonate (Ronastan TP Tin Concentrate 300, with 300
g/L stannous ions), 30 mL/L 70% methanesulfonic acid (Ronastan
TP acid 70), pH 0.4, 50 mL/L grain refiner (Ronastan TP-HCD pri-
mary additive), and 20 mL/L antioxidant (Ronastan TP antioxidant).
The grain refiner was nonionic and had a cloud point of 70ЊC. In
trode (SCE) were used as the counter and reference electrode, re-
spectively. All potentials, reported with respect to the SCE, were cor-
rected for solution resistance (i.e., IR drop). The RDE was driven by
the Pine rotator. The electrochemical measurements were conducted
using a CMS100D Corrosion Measurement System from Gamry
Instruments, Inc.
3
addition, iron (179 mol/m , or 10 g/L) was added to simulate its
buildup due to corrosion of the steel strip by the electrolyte, that is
pertinent to a horizontal plating cell where only the bottom side of
For each electrochemical measurement, a Cu electrode was
freshly polished to a 1 ␮m finish, cleaned in distilled water and ace-
tone. The electrolyte was deaerated by purging with argon gas for
30 min before the test, but not during the test to avoid trapping gas
7
,8
3
the steel strip is plated. For the 179 mol/m reagent-grade iron
powder dissolved, 40 mL/L of methanesulfonic acid was added to
maintain the free acid concentration. The ionic composition of the
MSA bath tested is listed in Table I. The bath was evaluated at four
temperatures: 27, 38, 52, and 60ЊC.
A rotating cylinder cathode (RCC) from Lucent Technologies
was used because the hydrodynamic conditions and electrical field
are well defined. The plating cell consisted of a 2 L Nalgene gradu-
ate cylinder cut to 920 mL, 8.1 cm diam ϫ 20 cm tall. The cathode
was a steel coupon, 2.86 cm wide and 5.70 cm long, wrapped around
the shaft, 1.93 cm in diam, of the RCC. The anode was a cylinder
made of platinum mesh, 4.95 cm in diam and 3.02 cm tall. A baffle
was placed inside the plating cell so that the rotation speed could
Table I. Composition of the MSA plating bath.
Ion
Concentration (mol/m3)
ϩ2
Sn
Fe
127
179
424
ϩ2
Hϩ
CH SO
Ϫ
3
1036
3
Downloaded on 2015-03-14 to IP 128.110.29.83 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

1
072
Journal of The Electrochemical Society, 147 (3) 1071-1076 (2000)
S0013-4651(99)08-106-9 CCC: $7.00 © The Electrochemical Society, Inc.
Figure 2. Effect of deoxygenation on the plating efficiency at the low current
density of 60 A/m at 52ЊC.
Figure 1. Effect of current density and rotation speed on plating efficiency at
2ЊC, determined by RCC.
2
5
stannic ions. To verify this hypothesis, the presence of oxygen in the
electrolyte was minimized either by sparging with argon or by using
a soluble tin anode. As shown in Fig. 2, the plating efficiency stays
at 90% as the rotation speed increases from 500 to 2000 rpm by
either one of these two approaches of deoxygenating. Apparently,
the current from oxygen reduction increases as its limiting current
density increases with increasing rotation speed. However, with lim-
ited solubility, oxygen reduction becomes insignificant as the ap-
plied current density increases.
Figure 3 exemplifies the effect of rotation speed and current den-
sity on the coating morphology at 52ЊC. The coating morphology is
divided diagonally into three groups: (I) compact crystalline at the
lower right region of high rotation speeds and low current densities,
bubbles on the surface of the RDE. For the potentiodynamic test,
used to determine the cathodic polarization curve, the potential was
swept cathodically at a constant scan rate of 5 mV/s from the open-
circuit potential to Ϫ1.2 V or 6000 A/m2 (the capacity of the
CMS100D instrument). For the galvanostatic plating, the current
density was set at various ratios with respect to the diffusion-limited
current density that was determined by the potentiodynamic test at
the same rotation speed. After the electrochemical tests, the Cu elec-
trode was dismounted, rinsed in distilled water and acetone, and
dried for examination.
Results and Discussion
Figure 1 shows the effect of the current density and the rotation
speed on the plating efficiency at 52ЊC, while the trend is similar at
other temperatures. At a given rotation speed, the plating efficiency
(II) dendritic at the upper left region of low rotation speeds and high
increases sharply as the current density increases from 60 to
2
6
00 A/m , reaches a maximum above 95%, and then gradually de-
creases. The current density at which the plating efficiency begins to
2
decrease increases from 1200 to 3000 A/m as the rotation speed in-
creases from 500 to 2000 rpm. As is shown later, the decrease of
plating efficiency is a result of reaching the diffusion-limited current
density of stannous ions and the beginning of hydrogen reduction. At
2
current densities above 3000 A/m the plating efficiency increases as
the rotation speed increases. On the other hand, at current densities
2
below 1500 A/m , the plating efficiency decreases as the rotation
2
speed increases. At the extremely low current density of 60 A/m , the
plating efficiency decreases from 78 to 41% as the rotation speed in-
creases from 500 to 2000 rpm. This unexpected decrease in plating
efficiency at low current densities was further investigated.
As shown in Table II, there are seven possible cathodic reactions
in the MSA bath. Since this low efficiency was also observed in a
7
bath free of iron, this decrease in plating efficiency is most likely to
be caused by the cathodic reduction of dissolved oxygen or residual
Table II. Possible reduction reactions in the MSA bath.
Reaction
Potential (V vs. SCE)
O ϩ 4H ϩ 4eϪ o 2H O
ϩ
Eo ϭ Ϫ0.987
Eo ϭ Ϫ0.528
Eo ϭ Ϫ0.092
Eo ϭ Ϫ0.241
Eo ϭ Ϫ0.378
2
2
ϩ3
Ϫ
ϩ2
Fe ϩ e o Fe
ϩ4
Ϫ
ϩ2
Sn ϩ 2e o Sn
ϩ
Ϫ
2
H
ϩ 2e o H2
ϩ2
Ϫ
Sn ϩ 2e o Sn
ϩ2
Ϫ
Fe ϩ 2e o Fe
Eo ϭ Ϫ0.651
Eo ϭ Ϫ1.070
Figure 3. Effect of current density and rotation speed on coating morpholo-
gy at 52ЊC, determined by RCC.
Ϫ
Ϫ
H O ϩ 2e o H ϩ 2 OH
2
2
Downloaded on 2015-03-14 to IP 128.110.29.83 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) 1071-1076 (2000)
1073
S0013-4651(99)08-106-9 CCC: $7.00 © The Electrochemical Society, Inc.
current densities, and (III) micronodular between the compact crys-
talline and dendritic regions. This behavior falls in the general
9
scheme proposed by Winand. Furthermore, the coating morphology
was found to be well correlated with the plating efficiency. As shown
in Fig. 4, the plating efficiency, excluding those at the extremely low
2
current density of 60 A/m , is above 90% in the compact crystalline
region, between 80 and 92% in the micronodular region, and much
lower on average in the dendritic region.
Under a given hydrodynamic condition, i.e., rotation speed and
temperature, the current density above which the morphology
changes from compact crystalline to micronodular is referred to as
the critical current density, i . As shown in Fig. 5, i is more than dou-
C
C
bled either as the rotation speed increases from 500 to 2000 rpm at a
given temperature, or as the temperature increases from 27 to 60ЊC at
a given rotation speed. It is commonly believed that dendrites start to
form once the current density reaches a certain fraction of the diffu-
sion-limited current density, i , and 0.4 was cited for tin in a phenol-
L
10,11
sulfonic bath.
In general, i can be calculated by the following equation
L
iL ϭ nFkCB
[1]
Figure 5. Effect of rotation speed and temperature on the critical current den-
sity, iC.
where n is the number of electron transferred, F the Faraday con-
stant, k the mass-transfer coefficient, and C the bulk concentration.
b
For a rotating cylinder, the mass-transfer coefficient can be estimat-
The kinematic viscosity was measured using a glass capillary
kinematic viscometer according to ASTM Standard D455-88. As
shown in Fig. 6, the kinematic viscosity of the electrolyte decreases
ed by the Eisenberg correlation12
k ϭ 0.0791 V ReϪ0.3 ScϪ0.644
[2]
Ϫ6
2
from 0.98 to 0.51 centistoke (10 m /s) as the temperature increas-
es from 24 to 60ЊC. This temperature dependence is similar to that
of water and correlates well with the Guzman-Andrade equation,
where V is the peripheral speed of the cylinder, Re the Reynolds
number based on the diameter of the cylinder d and the peripheral
speed of the cylinder V, and Sc the Schmidt number of the electro-
lyte. Introducing k from Eq. 2 into Eq. 1 yields
1
3
which was used to interpolate the kinematic viscosity within the
temperature range. Furthermore, it has been found that the kinemat-
ic viscosity increases as the concentration increases.
To determine the diffusion coefficient of the stannous ion, poten-
tiodynamic tests were conducted using the RDE in an iron-free elec-
trolyte. As shown by the cathodic polarization curves obtained at
Ϫ0.3
ScϪ0.644Cb
iL ϭ 0.0791 nFV Re
[3]
Introducing the definitions of the Reynolds number (i.e., Re ϭ dV/␯)
and the Schmidt number (i.e., Sc ϭ ␯/D), Eq. 3 can be rewritten as
2
7ЊC in Fig. 7, the cathodic current density is very low at potentials
iL ϭ 0.0791 nFdϪ0.3V0.7
␯
Ϫ0.344D0.644Cb
[4]
above Ϫ0.3 V. At Ϫ0.4 V, the current density reaches a plateau of 5
2
to 7 A/m , likely because of the residual oxygen or stannic ions.
where ␯ is the kinematic viscosity and D the diffusion coefficient.
The effects of speed (V) and concentration (C ) on i are explicitly
Once the potential reaches Ϫ0.48 V, the current density increases
quickly as nucleation of tin crystal starts. At about Ϫ0.65 V, the cur-
rent density reaches another plateau as a result of reaching the diffu-
sion-limited current density of stannous ions. This diffusion-limited
current density increases as the rotation speed increases. At a more
negative potential, depending on the rotation speed, the current den-
sity increases again as the proton reduction starts to take place. As is
shown later, the reduction of hydrogen ions is negligible at potentials
b
L
shown in Eq. 4, whereas the effect of temperature is not as obvious.
In order to examine the effect of temperature and to calculate i by
L
using Eq. 4, the kinematic viscosity (␯) of the electrolyte and the dif-
fusion coefficient (D) of the stannous ion were determined as func-
tions of temperature.
Figure 4. Correlation between plating efficiency and coating morphology for
the four temperatures and current densities above 600 A/m2 in the RCC
experiment.
Figure 6. Effect of temperature on the kinematic viscosity of the plating bath.
Downloaded on 2015-03-14 to IP 128.110.29.83 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

1
074
Journal of The Electrochemical Society, 147 (3) 1071-1076 (2000)
S0013-4651(99)08-106-9 CCC: $7.00 © The Electrochemical Society, Inc.
Figure 7. Effect of rotation speed on the cathodic polarization curve in the
iron-free bath at 27ЊC, determined by the RDE.
Figure 9. Effect of temperature on the diffusion coefficient of stannous ions
in the MSA plating bath.
after deposition is greater than the geometric area; it is also possible
above Ϫ0.8 V, therefore, the current density obtained at around
Ϫ0.8 V is primarily from the reduction of stannous ions.
that i for the MSA bath is greater than 0.4 for the phenolsulfonic acid
C
11
bath. The latter was further investigated in order to elucidate the
effects of current density and rotation speed on the plating efficiency.
Since the reaction rates depend on the substrate, for example, the
exchange current density for hydrogen reduction on tin is two orders
The diffusion-limited current density (i ) obtained from the RDE
L
can be related to the rotation speed ␻ (rad/s) by the well-known
1
4
Levich equation
16
iL ϭ 0.622nFD2/3
␻
1/2
␯
1/6Cb
[5]
of magnitude lower than that on iron. Of course, the vast majority of
deposition reaction takes place on tin as soon as a thin layer of tin cov-
ers the substrate, either the copper electrode used in the study or the
steel for tin plate. It is more pertinent to study the reaction on tin, thus,
the copper electrode was flash-coated with tin by a potentiodynamic
scan from the open-circuit potential to Ϫ0.7 V at 2000 rpm, immedi-
ately prior to the second potentiodynamic scan in the test solution.
As shown in Fig. 11, the open-circuit potential of the flash-coat-
ed tin is Ϫ0.45 V, much lower than the Ϫ0.02 V found on copper in
the plating bath, Fig. 7. The cathodic polarization curves exhibit a
distinctive region where the current is limited by diffusion of stan-
nous ions and increases with increasing rotation speed. However, as
the rotation speed increases, the plateau of limiting current becomes
narrower and less well defined. This is primarily due to the increase
in surface area, as the coating becomes thicker and dendrites start
to form.
1
/2
As shown in Fig. 8, i is linear against ␻ for the four temperatures
L
tested and the diffusion coefficient of stannous ions can be obtained
from the slope. As shown in Fig. 9, the diffusion coefficient increas-
Ϫ9
2
es from 0.83 to 1.58 ϫ 10 m /s as the temperature increases from
7 to 60ЊC. The value of the diffusion coefficient at 27ЊC is higher
2
Ϫ9
2
than 0.66 ϫ 10 m /s obtained by Chen et al. from a MSA bath
containing only 2.5 mol/m stannous ions.
3
15
With the determination of the kinematic viscosity of the electro-
lyte and the diffusion coefficient of the stannous ion as functions of
temperature, i of stannous ions was calculated by using Eq. 4. As
L
shown in Fig. 10, while i increases as the Reynolds number increas-
L
es with increasing temperature at each rotation speed, iC is always
greater than iL and the difference increases as the temperature in-
creases at each rotation speed. Although it is possible that the Eisen-
berg correlation underestimates iL because the actual surface area
Figure 10. Comparison of the critical current density (i ) and the diffusion-
C
Figure 8. Effect of rotation speed on the diffusion-limited current density, iL.
limited current density (i ) calculated by Eq. 4 for RCC.
L
Downloaded on 2015-03-14 to IP 128.110.29.83 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) 1071-1076 (2000)
1075
S0013-4651(99)08-106-9 CCC: $7.00 © The Electrochemical Society, Inc.
Figure 11. Cathodic polarization curves on the tin flash-coated RDE in the
plating bath and the tin-free solution at 27ЊC.
Figure 13. Effect of current density on potential and coating morphology in
2
a MSA bath without additive. (RDE, 27ЊC, 500 rpm, 18200 C/m , i
ϭ
L
2
1
000 A/m .)
In the tin-free solution, the polarization curves are independent of
the rotation speed, suggesting that the hydrogen reduction is under
kinetic control and is not affected by mass transfer. The current den-
sity from the tin-free solution is two orders of magnitude lower than
that from the plating bath until the potential reaches below Ϫ0.9 V.
tration. Although the hydrodynamic conditions in the RCC and RDE
are different, the results of RDE support the findings from RCC.
Furthermore, i has been found to be dependent upon the grain re-
C
finers used in commercial plating baths. As shown in Fig. 13, in the
absence of the grain refiner, a thin layer of discrete nuclei was formed,
Thus, when the current density is less than i (or above Ϫ0.8 V), the
L
2
regardless of the current density between 600 and 1400 A/m . How-
current from hydrogen reduction is negligible and the plating effi-
ever, these nuclei do not coalesce, resulting in large crystals of tin. As
the current density increases, the number of crystals increases but the
size of individual crystals decreases. It is also noticed that the poten-
tial stays at Ϫ0.45 V regardless of the current density, suggesting that
either the actual current densities never exceed the limiting current
density due to the increase in surface area or the dendrites protrude the
diffusion boundary layer. Based on this technique, 20 mL/L was
determined as the minimum amount of grain refiner required for a
compact crystalline deposit.
ciency is near 100%. When the current density exceeds i , the plat-
L
ing efficiency begins to decrease, as the proton reduction becomes
increasingly significant.
2
With i determined from the polarization curve, e.g., 1000 A/cm
L
at 500 rpm and 27ЊC, galvanostatic plating was conducted at current
densities of various ratios with respect to i , but with a constant total
L
2
charge of 18200 C/m . As shown in Fig. 12, as the current density
increases from 600 to 1000 A/m , the coating remains compact crys-
2
talline while the potential decreases from Ϫ0.65 to Ϫ0.75 V. As the
2
current density increases to 1200 A/m , the coating changes to
Conclusions
micronodular and the potential suddenly drops to Ϫ0.92 V. As the
1
. The plating efficiency correlates well with the coating mor-
2
current density further increases to 2000 A/m and beyond, the mor-
phology in this plating bath. Both of them depend on the applied cur-
phology changes to dendritic but the potential stays at Ϫ0.95 V. This
phenomenon has been observed at limiting current densities of other
combinations of temperature, rotation speed, and stannous concen-
rent density and the diffusion-limited current density (i ) of stannous
L
ions.
2
. A compact crystalline deposit is obtained with >95% plating
efficiency at current densities up to i in this plating bath. When the
L
current density exceeds i , not only does the plating efficiency begin
L
to decrease but also the coating changes to micronodular and then to
dendritic. Thus, the critical current density (i ) is as high as i in this
C
L
bath.
. The reduction of hydrogen ions is negligible at low overpoten-
3
tials (above Ϫ0.8 V), resulting in the high plating efficiency obtained
at current densities below iL.
4
. The synergistic effects of temperature and agitation on i were
L
quantified through the determination of the temperature dependence
of the viscosity of the bath and the diffusion coefficient of the stan-
nous ion.
5
. The critical current density (i ) also depends on additives. The
C
lower threshold of concentration of the addition agent can be deter-
mined by examining the coating morphology at various current den-
sities relative to iL.
Acknowledgment
The author thanks T. L. Holzinger and E. S. Fodor for the plating
and electrochemical experimentation and L. L. Hahn for the SEM
examination, and thanks Shipley Ronal for providing the Ronastan-
TP additives.
Figure 12. Effect of current density on potential and coating morphology in
Bethlehem Steel Corporation assisted in meeting the publication costs of
this article.
2
2
the plating bath. (RDE, 27ЊC, 500 rpm, 18200 C/m , i ϭ 1000 A/m .)
L
Downloaded on 2015-03-14 to IP 128.110.29.83 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

1
076
Journal of The Electrochemical Society, 147 (3) 1071-1076 (2000)
S0013-4651(99)08-106-9 CCC: $7.00 © The Electrochemical Society, Inc.
References
8. J. L. Swanson, G. A. Federman, D. Hahn, U. Dreissigacker, and N. Brown, in AESF
Proceedings of the 9th Continuous Steel Strip Symposium, p. 111 (1999).
1
.
.
.
D. R. Hultberg and G. A. Federman, in Proceedings of the 3rd North American Tin-
plate Conference, Rosemont, IL, Oct 1994.
F. J. Bottos, B. Alavanjo, G. A. Federman, and J. Swanson, in Proceedings of the
th International Tinplate Conference, p. 38, London, Sept 1996.
H. Espenhahm, in AESF Proceedings of the 8th Continuous Steel Strip Plating
Symposium, p. 263 (1996).
L. Dercsh, Met. Finish., 88, 23 (1990).
9
. R. Winand, Hydrometallurgy, 29, 567 (1992).
1
1
1
0. D. R. Gabe, Plat. Surf. Finish., 82, 69 (1995).
1. D. R. Gabe, Private communication.
2. M. Eisenberg, C. W. Tobias, and C. R. Wilke, Chem. Eng. Prog. Symp. Series, 21,
No. 16, p. 1 (1955).
13. E. N. da C. Andrade, Nature, 125, 309 (1930).
2
3
6
4
5
.
.
G. A. Federman, N. Brown, J. Swanson, and A. Chirafisi, in AESF Proceedings of
the 8th Continuous Steel Strip Plating Symposium, p. 243 (1996).
G. A. Federman, D. W. Thomson, M. P. Toben, and N. D. Brown, U.S. Pat.
14. A. J. Bard and L. R. Faulkner, Electrochemical Methods-Fundamentals and Appli-
cations, p. 288, John Wiley & Sons, New York (1980).
15. C. S. Chen, C. C. Wan, and Y. Y. Wang, Trans. Inst. Met. Finish., 76, 54 (1998).
16. H. H. Uhlig, Corrosion and Corrosion Control, 2nd ed., p 49, John Wiley & Sons,
New York (1971).
6
7
.
.
5
,174,887 (1992).
Y-H. Yau, Plating Surf. Finish., 86, 48 (1999).
Downloaded on 2015-03-14 to IP 128.110.29.83 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).