Nucleation and growth of tin on low carbon steel

Article abstract of DOI:10.1016/j.electacta.2004.10.053

The nucleation and growth of tin from stannous sulfate and sulfuric acid with an organic additive of gelatin was studied on low carbon steel. Investigations were conducted by linear sweep voltammetry and current step methods. The agitation effect on the morphology of tin deposits was also studied. From the results, a mechanism of tin deposition on a steel substrate is suggested. Initially, 3D tin crystallites are formed preferentially on step edges, and this is followed by fast covering of neighboring sites with much smaller and densely packed crystallites of about 150-180 nm in size. Without gelatin, the coverage was poor and tri-modal structures were dominating. It is discovered that the synergy between hydrogen co-evolution and gelatin contributes to fast and complete coverage of the steel substrate and the bi-modal grain size distribution is observed. The diffusion coefficient of tin ion was determined according to the Sand equation.

Full text of DOI:10.1016/j.electacta.2004.10.053

Electrochimica Acta 50 (2005) 2393–2399
Nucleation and growth of tin on low carbon steel
Shixue Wen∗, Jerzy A. Szpunar
Department of Mining, Metals and Materials Engineering, McGill University, Montreal, Que., Canada H3A 2B2
Received 19 July 2004; received in revised form 9 October 2004; accepted 10 October 2004
Available online 23 November 2004
Abstract
The nucleation and growth of tin from stannous sulfate and sulfuric acid with an organic additive of gelatin was studied on low carbon steel.
Investigations were conducted by linear sweep voltammetry and current step methods. The agitation effect on the morphology of tin deposits
was also studied. From the results, a mechanism of tin deposition on a steel substrate is suggested. Initially, 3D tin crystallites are formed
preferentially on step edges, and this is followed by fast covering of neighboring sites with much smaller and densely packed crystallites
of about 150–180 nm in size. Without gelatin, the coverage was poor and tri-modal structures were dominating. It is discovered that the
synergy between hydrogen co-evolution and gelatin contributes to fast and complete coverage of the steel substrate and the bi-modal grain
size distribution is observed. The diffusion coefficient of tin ion was determined according to the Sand equation.
© 2004 Elsevier Ltd. All rights reserved.
Keywords: Tin; Electrodeposition; Chronopotentiometry; Diffusion; Nucleation; Morphology
1. Introduction
anisotropy [3]. On the other hand, hydrogen co-evolution in
deposits decreases the current efficiency of the plating oper-
ation and may result in hydrogen embrittlement.
Nucleation and growth of metal deposits depend on the
nature of plating bath, hydrodynamic conditions and the sub-
strate onto which the electrochemical reduction takes place.
The nucleation process is closely related to the active sites
existing on the substrate [1]. Defects on a substrate surface,
such as edges, corners, boundaries, and kink sites, can act as
active sites for nucleation. However, the number of actual nu-
cleation sites is small at low overpotentials when compared
to the available active sites, instant nucleation on all active
sites takes place at high potentials. Cathodic hydrogen evo-
lution, either on the substrate and/or on the deposit, is also
an important factor which needs to be considered. Hydrogen
evolution is of scientific and technological importance. The
technological importance stems from the fact that electrode-
position of some metals, such as nickel and chromium, is
accompanied by simultaneous hydrogen evolution [2]. Hy-
drogen co-deposition is also responsible for texture variation
in the deposits due to its influences on the surface-energy
One can carry out electrodeposition by applying either
potential or current. For kinetic studies, the potential step
method is frequently used; while for industrial coatings
preparation, the current step method (galvanostatic electrol-
ysis) is mostly applied. The most important benefit of using
the galvanostatic method is that the thickness or weight of de-
posits can be readily controlled according to Faraday’s law
of electrolysis. There is little research, however, on the nu-
cleation and growth of electrodeposits using potential–time
method (chronopotentiometry) at constant currents. No such
studies on tin electrodeposition have been reported although
tin has been extensively used as a coating for a wide range
of metals. Tin coatings on steel (known as tinplates) are ex-
tensively used in the fabrication of containers for the food
processing industry. Some researchers [4–6] have reported
the nucleation and growth of tin on glassy carbon, how-
ever, there is no published work on the initial stage of tin
nucleation and growth on steel. In a previous study [4], the
nucleation and growth mechanism of tin, and the transforma-
tion in the morphology of initial deposits on glassy carbon
∗
Corresponding author. Tel.: +1 514 3984755x09516;
fax: +1 514 3984492.
0013-4686/$ – see front matter © 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2004.10.053

2394
S. Wen, J.A. Szpunar / Electrochimica Acta 50 (2005) 2393–2399
were investigated. It has found that the nucleation and growth
mechanism of tin depends on the applied cathodic potential.
A transition in the deposition mechanism from 3D progres-
sive to instantaneous nucleation and growth was observed
with increasing cathodic potential. The morphology of tin
nuclei also varies with applied potentials. At more positive
potentials, tetragonal nuclei and crystallites are dominant,
and as the potential become more negative, the fraction of
clusters of tin needles increases.
The current study focuses on the nucleation and growth of
tin on low carbon steel substrates. The potential–time behav-
iors at constant current densities were presented along with
the morphology of tin deposits. The diffusion coefficient of
tin ions was also determined.
Fig. 1. Voltammetric response of hydrogen evolution on Fe in sulfuric acid
solution. Scan rate: 50 mV/s; (a) 0.3 M H2SO4 + 2 g/l gelatin; (b) 0.3 M
H2SO4.
2. Experimental details
The electrochemical experiments were done in a conven-
tional three-electrode cell, where a low carbon steel substrate
embedded in epoxy resin was used as the working electrode
with a surface area of 1 cm2 exposed to the tin electrolyte, a
tin rod (99.98%) with 12.7 mm in diameter was used as the
anode, and a saturated calomel electrode (SCE) as a refer-
ence electrode to which all potentials in this paper are re-
ferred. The electrolyte used was composed of 0.3 M H2SO4
and 0.2 M SnSO4, in some experiments, gelatin was added
as an organic additive. All solutions were prepared with dis-
tilled water treated with a Millipore system. The experiments
were done in an oxygen-free electrolyte, which was obtained
by purging the electrolyte with nitrogen gas for at least half-
an-hour. After oxygen was removed from the solution, the
nitrogen bubbler was pulled above the electrolyte surface,
and the inert atmosphere was maintained by saturating the
chamber above the electrolyte with nitrogen. Hence, all the
electrochemical measurements were done in a quiescent so-
lution except for the agitation investigation. Before each mea-
surement, the working electrode (the substrate) was mechan-
ically polished using 600-grit sandpaper, and was then thor-
oughly rinsed with distilled water. All the electrochemical
measurements were done using a computer-controlled Au-
tolab PGSTAT30 potentiostat. For the imaging of samples,
a field-emission-gun scanning electron microscope (Philips
XL30 FEG-SEM) was used.
added, hydrogen evolution occurred at a more negative po-
tential (∼−0.82 V). This increase in hydrogen overpotential
was due to the inhibitory effect of gelatin on hydrogen evolu-
tion. Fig. 2 presents the voltammetric response of hydrogen
evolution on a tin coating in the supporting electrolyte. Com-
paring to steel, hydrogen evolution started at more negative
potentials on tin, −0.8 V, for the sulfuric acid solution with-
out gelatin and −0.95 V while gelatin added. The voltam-
metric response for tin reduction in the 0.3 M H2SO4 and
0.2 M SnSO4 electrolyte is shown in Fig. 3. As can be seen,
in the absence of gelatin, the abrupt increase of current den-
sity in magnitude was observed at the negative potential of
−0.44 V, which was due to the deposition of tin (II) ions. At
around −0.50 mV, the current peaks were observed in both
situations, indicating that there is a nucleation and growth
mechanism controlled by diffusion [4,5]. As potentials be-
came more negative, limiting current density regions were
evident. The magnitude of the limiting current density for
the tin electrolyte without gelatin was 236 A/m2. The limit-
3. Results and discussion
3.1. Linear sweep voltammetry
Fig. 1 shows the voltammetric response of hydrogen evo-
lution on a low carbon steel substrate in the supporting elec-
trolyte (i.e., 0.3 M H2SO4 solution). Since there were no
tin ions in the solution, only hydrogen evolution was ob-
served. When there was no gelatin (Fig. 1b), hydrogen evo-
lution took place around −0.62 V versus SCE; with gelatin
Fig. 2. Voltammetric response of hydrogen evolution on tin in sulfuric acid
solution. Scan rate: 50 mV/s; (a) 0.3 M H2SO4 + 2 g/l gelatin; (b) 0.3 M
H2SO4.

S. Wen, J.A. Szpunar / Electrochimica Acta 50 (2005) 2393–2399
2395
4.8 s, which was due to the concentration depletion of tin
ions on the surface of the electrode. The time taken from the
bulk concentration of tin to zero at the electrode surface is
defined as the transition time τ or the Sand time, which is
given by the Sand equation (Eq. (1)) [7]
nF(πD)1/2c0
|Iτ1/2| =
(1)
2
where I is the current density, n the number of electrons, here
equals 2, F the Faraday’s constant, D the diffusion coefficient
and c0 the tin ion concentration.
The diffusion of tin ions to the substrate by natural con-
vection was then no longer sufficient to maintain the applied
current density, and the potential dropped rapidly to −0.88 V
until hydrogen evolution took place. Through visional obser-
vation, lots of small bubbles approximately 1 mm in diameter
were seen emerging from the sample. It is interesting to have
observed the peaks of the potential in Fig. 4c and the bubbling
on the substrate decreased progressively, and finally the po-
tential increased slowly and the bubbling halted. Fig. 4 also
shows the potential–time transients obtained from sulfuric
acid solution without tin ions on steel, and on steel completely
coated with tin. The potential arrest at −0.88 V was related to
hydrogen evolution on steel rather that tin, which indicated
the steel sample was poorly covered by tin, as seen further
through morphological examination. The potential perturba-
tion between 4 and 11 s (Fig. 4c) can be explained as follows:
the first sharp and biggest peak was due to depletion of tin
ions on the sample and hydrogen evolution on steel. The gen-
eration of hydrogen bubbles at the electrode surface stirred
the solution in the immediate vicinity of the growing tin parti-
cles. As the depletion layer around growing tin particles was
eliminated, tin deposition occurred more favorably and po-
tential restored to a higher value of −0.57 V determined by
the Nernst equation. It is obvious that the restored concentra-
tion of tin ions on the electrode surface is still smaller than
that of the bulk concentration. As deposition progressed, the
potential increased due to the growth of tin dendrites (as can
Fig. 3. Voltammetric responses from tin sulfate solutions on low carbon steel
at a scan rate of 10 mV/s. (a) 0.3 M H2SO4 + 0.2 M SnSO4 + 2 g/l gelatin;
(b) 0.3 M H2SO4 + 0.2 M SnSO4.
ing current was reduced to 185 A/m2 when gelatin was added,
indicating an inhibitory effect of gelatin on the reduction of
tin ions.
3.2. Current step method
The galvanostatic experiments, also known as chronopo-
tentiometry, were used to study the potential behaviors with
time. Fig. 4 shows E–t transients on a steel substrate for
acid stannous sulfate electrolytes without gelatin at dif-
ferent current densities. At different current densities, the
potential–time curves were different. At a low current den-
sity of 100 A/m2, the potential remained constant with time.
At an increased current density of 200 A/m2 (Fig. 4b), there
was a fairly slow decrease in potential up to 35 s and then a
gradual increase in potential. The initial decrease in potential
is determined by the Nernst equation; the gradual increase
resulted from a dendrite tin growth, which increased the sur-
face area with time and the real current density decreased.
Curve (c) for 400 A/m2 shows an abrupt potential drop at
Fig. 4. Potential–time transients for the tin sulfate electrolyte (0.3 M
H2SO4 + 0.2 M SnSO4) on steel at different current densities. (a) 100 A/m2;
(b) 200 A/m2; (c) 400 A/m2; A: 3 s; B: 6 s; C: 8 s; D: 15 s; E: 30 s.
Fig. 5. Potential–time transients for tin sulfate on steel with gelatin at dif-
ferent current densities. (a) 100 A/m2; (b) 200 A/m2; (c) 400 A/m2.

2396
S. Wen, J.A. Szpunar / Electrochimica Acta 50 (2005) 2393–2399
Table 1
Product of I × τ1/2 and calculated diffusion coefficient in the electrolyte
composed of 0.3 M H2SO4 + 0.2 M SnSO4 + 2 g/l gelatin on steel
I/Am−2
τ/s
I × τ1/2
Calculated D/cm2 s−1
200
270
300
400
810
12.063
6.783
5.322
2.739
0.636
69.46
70.31
69.20
66.20
64.60
4.1 × 10−6
4.2 × 10−6
4.1 × 10−6
3.7 × 10−6
3.6 × 10−6
Without organic additives, smooth deposits cannot be ob-
tained and tin dendrites form. In order to get smooth coat-
ings, a variety of organic additives can be used. Fig. 5 shows
the potential–time curves for the tin electrolyte with gelatin
(0.3 M H2SO4 + 0.2 M SnSO4 + 2 g/l gelatin). At 100 A/m2,
the potential remained stable with time. The potential oscil-
lated at 200 A/m2 for the recording time of 3 min, here pre-
sented for only 2 min. The frequency of oscillation remained
constant after the second potential drop. Such oscillations
have also been observed by other authors [8]. This oscillat-
ingfeature isdue to repeatedtinionsdepletionandrestoration
by hydrogen evolution. At an even higher current density of
Fig. 6. Potential–time transient for tin sulfate on steel with gelatin at dif-
ferent current densities. (a)200 A/m2 with agitation; (b) 400 A/m2 without
agitation; (c) 400 A/m2 with agitation.
be seen in the SEM images of Fig. 7e and f). This process
increased the actual electrode surface area, and therefore de-
creased the actual current density below its limiting value;
thus no more hydrogen evolved.
Fig. 7. Morphology of tin deposits at 400 A/m2 for different times corresponding to arrows A (3 s); B (6 s); C (8 s); D (15 s); E (30 s) in Fig. 4. (a) 400 A/m2,
3 s; (b) 400 A/m2, 6 s; (c) 400 A/m2, 8 s; (d) 400 A/m2, 15 s; (e) 400 A/m2, 15 s; (f) 400 A/m2, 30 s.

S. Wen, J.A. Szpunar / Electrochimica Acta 50 (2005) 2393–2399
2397
Fig. 8. Morphology of tin deposits at 200 A/m2 for times corresponding to arrows A (10 s), B (16.3 s), C (25 s) in Fig. 5b, and at 400 A/m2 for times corresponding
to arrows A (2 s), B (4.5 s), C (8 s) in Fig. 6b. (a) 200 A/m2, 10 s; (b) 400 A/m2, 2.5 s; (c) 200 A/m2, 16.3 s; (d) 400 A/m2, 4.5 s; (e) 200 A/m2, 25 s; (f) 400 A/m2,
8 s.
400 A/m2, however, a damping oscillation was observed, and
the potential increased slowly and tended to remain constant
at longer time, which was attributed to hydrogen evolution
on tin coatings. The Sand time can be easily determined. At
different current densities, the product of I × τ1/2 should be
constant as expected for a mass-transfer-controlled process.
Knowing the bulk concentration of tin c0, the diffusion co-
efficient can be calculated from Eq. (1). Table 1 presents the
product of I × τ1/2 and calculated diffusion coefficients in a
tin sulfate electrolyte with gelatin. As expected, the results
reflect a mass-transfer-controlled process with a diffusion co-
efficient of about 3.9 × 10−6 cm2 s−1 (averaged value).
As was mentioned earlier, at low and moderate current
densities, potentials were relatively stable and the deposition
of tin was sufficiently supplied by diffusion. At high current
densities, however, the supply of tin ions by natural convec-
tion was not sufficient. In order to maintain the concentration
of tin ions on the electrode surface, forced convection was
introduced. Fig. 6 presents the potential–time curves for tin
sulfate electrolyte with gelatin on steel in the presence or ab-
sence of agitation at 200 and 400 A/m2. As can be seen in both
cases, the potential remained constant with time when agita-
tion was applied, which means that at this level of agitation,
tin concentration on the sample surface could be maintained.
The small perturbation of potential for 400 A/m2 indicated
there was a difficulty in supplying tin ions at high current
densities even with agitation. It should be noted that the agi-
tation should not exceed a certain level where turbulent flow
occurs. A detailed explanation is beyond the scope of this
paper. By rule of thumb, there should have been no visible
vortices in the solution.
3.3. Scanning electron microscopy
Morphological examinations of tin deposits by the SEM
were carried out to better understand the electrochemical re-
sults. The morphology and coverage of tin coatings are di-
rectly related to the properties of tin coatings and are of great
importance in industrial applications. Fig. 7 shows the mor-
phology of tin deposits at 400 A/m2 for different times cor-
responding to arrows A (3 s), B (6 s), C (8 s), D (15 s), E
(30 s) in Fig. 4. As can be seen in the deposits obtained for

2398
S. Wen, J.A. Szpunar / Electrochimica Acta 50 (2005) 2393–2399
Fig. 9. Morphology of tin deposits (obtained at different current densities) on steel in tin sulfate electrolyte with gelatin (0.3 M H2SO4 + 0.2 M SnSO4 + 2 g/l
gelatin) with and without agitation. (a) 200 A/m2, 60 s, without agitation; (b) 200 A/m2, 60 s, with agitation; (c) 400 A/m2, 30 s, without agitation; (d) 400 A/m2,
30 s, with agitation.
3 s (Fig. 7a), 3D tin crystallites of similar size preferentially
formedalongpolishinglinesorstep-edges. Thisphenomenon
is also called step-edge decoration, which was used to pre-
pare molybdenum nanowires [9]. At a longer deposition time
of 6 s (Fig. 7b), tin crystallites were still of similar size but
bigger than those for 3 s, and the number of tin crystallites
was roughly the same, which indicated instantaneous nucle-
ation. In Fig. 7c, a sub-layer of small grains in the range of
100–200 nm can be seen. Those grains tended to be intercon-
nected. At a much longer deposition time (15 s in Fig. 7d and
e), those grains became densely packed. When comparing
the bi-modal structure of tin deposits with the potential–time
result (Fig. 4c), it is clear that the growth of a sub-layer of
small grains was credited to hydrogen evolution. As hydro-
gen evolution starts, the depletion layer around tin crystallites
was eliminated by the stirring effect of hydrogen evolution,
which made tin reduction happen again. At such negative
potentials (around −0.6 to −0.7 V), instant nucleation took
place on the whole surface, resulting in a rapid coverage of
tin on steel. When there was no additive in the electrolyte,
dendritic growth above early deposits (as shown in Fig. 7e
and f) occurred at certain preferred sites. Fast growth of den-
drites on the sample edges was readily visible with naked
eyes after 8 s at 400 A/m2. It is also evident that without or-
ganic additives, the coverage of steel with tin was poor, and
a tri-modal structure was observed.
deposition times corresponding to arrows A (2s), B (4.5s),
C (8s) in Fig. 6b. At both current densities, tin initially nu-
cleated at polishing lines before hydrogen evolution started.
When hydrogen evolution started at very negative potentials,
there were numerous small crystallites around 150–180 nm
covering steel substrates completely. The thickness of the fine
grain layer was estimated to be around 100 nm according to
the Faraday’s law of electrolysis.
The sudden growth of numerous grains can be explained
as follows. As the onset of hydrogen evolution, the stirring
effect of hydrogen bubbles enhances the supply of tin ions to
thesubstrateforthedepositionoftin. Suchnegativepotentials
allow tin deposition occurring on the whole substrate surface,
resulting in a full coverage of tin. It must be noted, however,
that the full and rapid coverage of tin is not solely due to
hydrogen co-evolution. When comparing images in Fig. 8f
with Fig. 7c and d, it is clear that for the same current density
of 400 A/m2, a full coverage of steel with numerous small tin
grains could not be achieved without gelatin.
Based on the above results, there are two possible factors
that can contribute to the rapid growth of grains with 150–
180 nm in diameter. In order to isolate these factors, agitation
was introduced since the sufficient supply of tin ion for elec-
trochemicalreductionwasfeasiblethroughforcedconvection
(see Fig. 6b and c). Fig. 9 shows the morphology of tin de-
posits at different current densities with or without agitation.
In Fig. 9, the deposition time was chosen to achieve 1 ␮m
thick tin coatings. For both processing conditions, the sub-
strates were well covered with tin and the coatings were com-
pact if no agitation was involved. However, when the elec-
trolyte was stirred (and no hydrogen evolution took place),
In order to achieve a smooth coating and well covered steel
substrate, gelatinwasusedasanorganicadditiveinthetinsul-
fate electrolyte. Fig. 8 shows the morphology of tin deposits
at 200 A/m2 for different deposition times corresponding to
arrows A, B and C in Fig. 5b, and at 400 A/m2 for different

S. Wen, J.A. Szpunar / Electrochimica Acta 50 (2005) 2393–2399
2399
steelcouldnotbewellcovered. Itwasevidentthatwithgelatin
alone, fully covered tin coatings could not be obtained. It is
therefore, discovered that hydrogen co-evolution and gelatin
had a synergistic effect that resulted in a rapid full coverage
of tin on steel. It is also clear that the grain sizes of the tin
deposits obtained at 400 A/m2 were smaller than those ob-
tained at 200 A/m2. The grain size of deposits decreased with
an increase in current density was as expected.
deposition times depending on the current densities. It is also
demonstrated that the chronopotentiometry is a very useful
and efficient tool to study the deposition process when com-
bined with the SEM.
References
[1] L. Bonou, M. Eyraud, J. Crousier, J. Appl. Electrochem. 24 (1994)
906.
[2] M. Paunovic, M. Schlesinger, Fundamental of Electrochemical Depo-
sition, Wiley, New York, USA, 1998, p. 96.
4. Conclusions
[3] D.Y. Li, J.A. Szpunar, Electrochim. Acta 42 (1) (1997) 47.
[4] S. Wen, S. Omanovic, J.A. Szpunar, Proceedings of EPD Congress,
Symposia sponsored by the Extraction and Processing Division (EPD)
of TMS (The Minerals Metals and Materials Society), Charlotte, North
Carolina, 14–18 March, 2004, p. 5.
[5] J. Torrent-Burgues, E. Guaus, F. Sanz, J. Appl. Electrochem. 32 (2002)
225.
[6] E. Gomez, E. Guaus, F. SanZ, E. Valles, J. Electroanal. Chem. 465
(1999) 63.
[7] Southampton Electrochemistry Group, Instrumental Methods in Elec-
trochemistry, Wiley, New York, 1985, p. 73.
[8] A. Aragon, M.G. Figueroa, R.E. Gana, J. Appl. Electrochem. 22
(1992) 558.
For tin electrocrystallization on steel substrates, 3D tin
crystallites are initially formed preferentially on step edges,
followed by fast deposition of other sites with much smaller
and densely packed crystallites of about 150–180 nm in diam-
eter. It is discovered that hydrogen co-evolution and gelatin
have a synergy effect contributing to the fast and complete
coverage of steel substrate with tin and the bi-modal grain
size distribution at very short deposition times. In order to
achieve compact uniform tin deposits, the following condi-
tions should be met: (a) the addition of an organic additive (at
least); (b) hydrogen co-evolution; and (c) at relatively longer
[9] M.P. Zach, K.H. Ng, R.M. Penner, Science 290 (2000) 2120.