Study of the electrochemical deposition of Sn-Ag-Cu alloy by cyclic voltammetry and chronoamperometry

Article abstract of DOI:10.1016/j.electacta.2008.11.015

The electrochemical deposition of Sn-Ag-Cu alloy from weakly acidic baths onto glassy carbon electrodes (GCE) was studied by cyclic voltammetry (CV) and chronoamperometry (CA). The properties of the electrodeposits were characterized by scanning electron microscopy (SEM), energy-dispersive spectrometery (EDS) and X-ray diffraction (XRD). Test results indicate that the two cathodic peaks in the CV curves, at -0.6 V and -0.85 V during the forward scan towards the negative potentials, correspond to the irreversible deposition of a solid solution of tin, silver and copper. The underpotential deposition (UPD) of Sn occurs at -0.6 V during the cathodic period and the amount of Ag and Cu in the Sn-Ag-Cu alloy decreases with increasingly negative cathodic potentials. During the forward scan, towards the positive potentials used in CV testing, cathodic peaks at -0.85 V appear in the CV curves for baths containing mixtures of tin salts and triethanolamine (TEA). This corresponds to a reduction of transient complex ions [Sn(TEA)_x]²⁺ on the surface of the cathode. Furthermore, the formation and reduction of [Sn(TEA)_x]²⁺ is a diffusion controlled process. On the surface of the GCE, the actual nucleus growth mechanism of the Sn-Ag-Cu alloy is represented by the progressive nucleation model.

Full text of DOI:10.1016/j.electacta.2008.11.015

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Electrochimica Acta 54 (2009) 2883–2889
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Study of the electrochemical deposition of Sn–Ag–Cu alloy by cyclic
voltammetry and chronoamperometry
Jinqiu Zhang, Maozhong An∗, Limin Chang
School of Chemical Engineering and Technology, Harbin Institute of Technology, Harbin, China
a r t i c l e i n f o
a b s t r a c t
Article history:
The electrochemical deposition of Sn–Ag–Cu alloy from weakly acidic baths onto glassy carbon elec-
trodes (GCE) was studied by cyclic voltammetry (CV) and chronoamperometry (CA). The properties of
the electrodeposits were characterized by scanning electron microscopy (SEM), energy-dispersive spec-
trometery (EDS) and X-ray diffraction (XRD). Test results indicate that the two cathodic peaks in the CV
curves, at −0.6 V and −0.85 V during the forward scan towards the negative potentials, correspond to the
irreversible deposition of a solid solution of tin, silver and copper. The underpotential deposition (UPD)
of Sn occurs at −0.6 V during the cathodic period and the amount of Ag and Cu in the Sn–Ag–Cu alloy
decreases with increasingly negative cathodic potentials. During the forward scan, towards the positive
potentials used in CV testing, cathodic peaks at −0.85 V appear in the CV curves for baths containing
mixtures of tin salts and triethanolamine (TEA). This corresponds to a reduction of transient complex
ions [Sn(TEA)x]2+ on the surface of the cathode. Furthermore, the formation and reduction of [Sn(TEA)x]2+
is a diffusion controlled process. On the surface of the GCE, the actual nucleus growth mechanism of the
Sn–Ag–Cu alloy is represented by the progressive nucleation model.
Received 11 October 2008
Received in revised form
10 November 2008
Accepted 10 November 2008
Available online 19 November 2008
Keywords:
Electrodeposit
Cyclic voltammetry
Chronoamperometry
Sn–Ag–Cu alloy
Nucleus growth mechanism
© 2008 Elsevier Ltd. All rights reserved.
1. Introduction
trodeposition of a Sn–Ag–Cu alloy coating from an alkaline bath
represents a normal co-deposition [3]. The cyclic voltammograms
Tin–lead alloys are the most widely used solder coatings in
traditional electronic packaging. However, the acute toxicity of
lead has spurred legislation banning its use in electronic prod-
ucts. Near eutectic Sn–Ag–Cu alloys [1–8] are now being developed
as promising lead-free solder coatings. These alloys demonstrate
superior mechanical properties and solderability [9,10]. We elec-
trodeposited bright, compact and smooth Sn–Ag–Cu alloy coatings.
These coatings contain 88–95 wt.% Sn, 5–10 wt.% Ag and 0.5–2 wt.%
Cu. They were produced from weakly acidic baths that contained
three main salts, three complexing agents, one main brightener
and one antioxidant [11]. In this kind of complicated electrolyte,
the properties and quality of the alloy coatings are directly deter-
mined by mechanism of electrocrystallization. Therefore, it is
of great significance to study the electrochemical deposition of
Sn–Ag–Cu alloys to improve the properties and quality of these
coatings.
of a Sn–Ag–Cu electroplating bath that contained methanesul-
phonic acid (MSA) were measured to analyze the effect of iso-octyl
phenoxy polyethoxy ethanol (OPPE) and thiourea on the co-
deposition [5].
CV and CA techniques were used to investigate the electrode-
position behavior, rate controlling step, and the nucleus growth
mechanism of Sn–Ag–Cu alloy electrodeposited from a weakly
acidic bath.
2. Experimental
2.1. Compositions of bath
The electroplating bath for bright Sn–Ag–Cu alloy co−n1-
tained 0.20 mol L−1 Sn(CH3SO3)2, 4.5 mmol L−1 AgI, 1.5 mmol L
Cu(CH3SO3)2, 0.60 mol L−1 K4P2O7, 1.35 mol L−1 KI, 0.225 mol L−1
TEA, 6.4 mmol L−1 heliotropin (HT) and 9 mmol L−1 hydroquinone.
Among these compositions, Sn(CH3SO3)2, AgI and Cu(CH3SO3)2 are
the main salts, K4P2O7, KI and TEA are complexing agents, HT is
a main brightener and hydroquinone is an antioxidant [11]. The
pH of the bath was adjusted to 5.5 by MSA. The bath undergoing
investigation was modified according to the different experimental
requirements. The bath components were modified, while pH was
kept constant.
To the best of our knowledge, not much work has been done
on the electroplating mechanism of Sn–Ag–Cu alloy coatings. The
polarization curves and surface morphology show that the elec-
∗
Corresponding author. Tel.: +86 45186413721; fax: +86 45186415527.
E-mail address: mzan@hit.edu.cn (M. An).
0013-4686/$ – see front matter © 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2008.11.015

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
2884
J. Zhang et al. / Electrochimica Acta 54 (2009) 2883–2889
2.2. Electrochemical evaluation
CV and CA experiments were carried out in a three-electrode
glass cell on a CHI630B electrochemical workstation. A saturated
calomel electrode (SCE) was used as the reference electrode and
platinum foil (99.99%, 2 cm2) was used as the counter-electrode.
The working electrode was a GCE with a working surface of 7.1 mm2
(Ø = 3 mm). A Luggin capillary tip was set 2 mm from the surface
of the working electrode. The bath temperature was maintained
at 20 1 ◦C by thermostat control. CV curves were recorded at a
scanning rate of 100 mV s−1
.
2.3. Properties of Sn–Ag–Cu alloy electrodeposits
The deposition of Sn–Ag–Cu alloy was carried out potentiostat-
ically on a HDV-07 instrument. The working electrode was a Fe/Ni
foil with a working surface area of 750 mm2 (25 mm × 30 mm). The
electrodeposition time varied from 30 min to 120 min. The depo-
sition charge that was calculated from the current–time curves
was 2200 C dm−2. The surface morphologies of the Sn–Ag–Cu alloy
electrodeposits were analyzed using a S-570 SEM. The chemical
composition of the electrodeposits was determined by EDS in the
S-570 SEM. The crystal structure of the deposit was examined using
a D/max-3C XRD with Cu K␣ radiation.
Fig. 2. CV curves of Cu, Ag and Ag–Cu electrodeposited from weakly acidic baths
containing 0.60 mol L−1 K4P2O7, 1.35 mol L−1 KI, 0.225 mol L−1 TEA, 6.4 mmol L−1 HT,
9 mmol L−1 hydroquinone and (a) 1.5 mmol L−1 Cu(CH3SO3)2, (b) 4.5 mmol L−1 AgI,
(c) 4.5 mmol L−1 AgI and 1.5 mmol L−1 Cu(CH3SO3)2 at a scan rate of 100 mV s−1
.
a nucleation and growth process [12]. An anodic peak ‘R’ appears
at −0.62 V during the anodic period. This peak corresponds to the
dissolution of tin. The area of the cathodic peak is larger than that
of the anodic peak because most of the cathodic current generates
hydrogen. The exclusive electrodeposition of Sn is difficult in the
bath without other metal ions.
As shown in Fig. 2a and b, the anodic peaks at −0.335 V and
−0.395 V correspond to the dissolution of copper and silver, respec-
tively. In a simple salt bath, the anodic peak position of Ag should
be more positive than that of Cu. However, the anodic peak position
of Cu in Fig. 2a is more positive than that of Ag in Fig. 2b. This can
be attributed to the large quantity of KI existing in the bath. KI is
used as a complexing agent for Ag+. The anodic peak area in Fig. 2a
is the smallest because the concentration of copper ions is the low-
est among metal ions in the bath. As shown in Fig. 2c, an anodic
peak appears at −0.382 V during the anodic period. The area of the
anodic peak is the largest. This means that the anodic peak is the
dissolution peak of a solid solution of silver and copper.
3. Results and discussion
3.1. Analysis of cyclic voltammograms
Transient cyclic voltammograms are an efficacious method used
to study the reactions on the working electrode surface. The cathode
electrode reactions in the electrodeposition process of Sn–Ag–Cu
ternary alloy are complicated. Therefore, it is necessary to study
the CV behaviors of single metals and binary alloys before the CV
curves of the Sn–Ag–Cu alloy can be studied.
As shown in Fig. 1, during the forward scan towards the neg-
ative potentials, a very small cathodic current is observed. It is
attributed to the reduction of impurities or oxygen dissolved in
the bath. The onset of Sn deposition is at about −1.26 V. There is
a sharp increase in the cathodic current, followed by the current
looping as the direction of the sweep is reversed. The appearance
of this type of hysteresis loop (ABC) is a characteristic feature of
As shown in Fig. 3a–c, scanning towards the positive potentials
result in anodic peaks at −0.638 V, −0.634 V and −0.627 V, respec-
tively. These anodic peaks have a potential value similar to that of
tin (Fig. 1); they also correspond to the dissolution of tin. By adding
tin salt to the bath, the potentials of the peaks at about −0.4 V in
Fig. 3a–c are different from those of the curves shown in Fig. 2a–c.
The anodic currents of the peaks at −0.4 V in Fig. 3a–c are bigger
than the corresponding curves in Fig. 2a–c. This means that these
anodic peaks in Fig. 3 correspond to the dissolution of a solid solu-
tion containing tin. As shown in Fig. 3c, two cathodic peaks at −0.6 V
and −0.85 V form during the cathodic period. In order to deter-
mine the relationships between the cathodic and anodic peaks in
Fig. 3c, we investigated the effect of the scan reversal potential on
anodic peaks of the CV curves. As shown in Fig. 4, cathodic peak
‘A’ corresponds to the anodic peak ‘S’. The Sn–Ag–Cu alloy coat-
ing, electrodeposited potentiostatically at −0.6 V, is composed of
60.1 at% Sn, 11.2 at% Ag and 28.7 at% Cu. This means that peak ‘A’
is the co-deposit peak of the Sn–Ag–Cu alloy, and peak ‘S’ is the
dissolution peak of the solid solution of tin, silver and copper. The
exclusive electrodeposition of Sn is difficult (Fig. 1), therefore, the
reduction of Sn2+ ions at −0.6 V in the weakly acidic bath can be
modeled as underpotential deposition (UPD) of Sn on Ag and Cu
atoms [13]. The composition of the small round grains in region ‘A’
Fig. 1. CV curves of tin electrodeposited from a weakly acidic bath containing
0.20 mol L−1 Sn(CH3SO3)2, 0.60 mol L−1K4P2O7, 1.35 mol L−1 KI, 0.225 mol L−1 TEA,
6.4 mmol L−1 HT and 9 mmol L−1 hydroquinone at a scan rate of 100 mV s−1
.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
J. Zhang et al. / Electrochimica Acta 54 (2009) 2883–2889
2885
Fig. 3. CV curves of Sn–Cu, Sn–Ag and Sn–Ag–Cu electrodeposited from
weakly acidic baths containing 0.20 mol L−1 Sn(CH3SO3)2, 0.60 mol L−1 K4P2O7,
1.35 mol L−1 KI, 0.225 mol L−1 TEA, 6.4 mmol L−1 HT, 9 mmol L−1 hydroquinone
and (a) 1.5 mmol L−1 Cu(CH3SO3)2, (b) 4.5 mmol L−1 AgI, (c) 4.5 mmol L−1 AgI and
1.5 mmol L−1 Cu(CH3SO3)2 at a scan rate of 100 mV s−1
.
of Fig. 5 is 24.9 at% Sn, 29.3 at% Ag and 45.8 at% Cu; this confirms
the above presumption.
As shown in Fig. 6a, the scan reversal potential is −0.90 V. Peak
‘K’ appears at −0.44 V, which is near peak ‘S’, during the anodic
period. When the scan reversal potential is −0.95 V, there is super-
position between peak ‘S’ and peak ‘K’ (Fig. 6b). When the scan
reversal potential is −1.1 V, peak ‘K’ disappears as shown in Fig. 6c.
In addition, as shown in Fig. 7, the coexistence of peak ‘K’ and peak
‘S’ is observed by testing other baths of electrodeposited Sn–Ag and
Sn–Cu alloys when the scan reversal potential is −0.90 V. It is pre-
sumed that, peak ‘K’ is the dissolution peak of some ambiguous
intermetallic compound (IMC). When the scan reversal potential is
increasingly negative, the dissolution peak of the solid solution cov-
ers up the ambiguous IMC. It can be seen that the anodic current of
peak ‘S’ in Fig. 6a is larger than the same ‘S’ peak in Fig. 4. Therefore,
peak ‘B’ is mainly related to peaks ‘S’ and ‘K’. Therefore, peak ‘B’
Fig. 5. SEM image of a Sn–Ag–Cu alloy coating electrodeposited potentiostatically at
−0.6 V from a weakly acidic bath containing 0.20 mol L−1 Sn(CH3SO3)2, 4.5 mmol L−1
AgI, 1.5 mmol L−1 Cu(CH3SO3)2, 0.60 mol L−1 K4P2O7, 1.35 mol L−1 KI, 0.225 mol L−1
TEA, 6.4 mmol L−1 HT and 9 mmol L−1 hydroquinone.
is the co-deposit peak for large quantities of Sn–Ag–Cu solid solu-
tion and small quantities of ambiguous IMC. The Sn–Ag–Cu alloy
coating, electrodeposited potentiostatically at−0.85 V, is composed
of 78.9 at% Sn, 14.8 at% Ag and 6.3 at% Cu. The stone-like grains in
region ‘A’ of Fig. 8 are composed of 72.1 at% Sn, 17.5 at% Ag and
10.4 at% Cu. The higher contents of Ag and Cu in region ‘A’ can be
attributed to the electrodeposition of IMC.
Fig. 3c shows a straight increase in the cathodic current during
the potential change from −1.25 V to −1.4 V. This corresponds to the
co-deposition of the Sn–Ag–Cu alloy. The Sn–Ag–Cu alloy coating,
electrodeposited potentiostatically at −1.4 V, is bright and smooth,
with a fine morphology (Fig. 9). The Ag and Cu contents in the
coating are 8.9 at% and 4.8 at%, respectively. These values are lower
than those from the coating electrodeposited potentiostatically at
−0.6 V and −0.85 V. The Ag and Cu content within the Sn–Ag–Cu
alloy decreases, as the cathodic potential is increasingly negative.
As shown in Fig. 10, the diffraction peaks in the Sn–Ag–Cu alloy
coating can be attributed to the ␤-Sn, Ag3Sn and Cu6Sn5 phases.
Peak ‘R’ in Fig. 3c is clearly the dissolution peak of Sn, and peak ‘S’
in Fig. 3c is clearly the dissolution peak of a solid solution containing
little IMC. Different compositions of solid solution and IMC, under
various conditions, will shift the potential position of the ‘S’ peak.
As shown in Fig. 6c, a strong cathodic peak ‘Q’ appears at about
−0.85 V during the forward scan towards the positive potentials.
The current of peak ‘Q’ is higher than that of peak ‘B’. However, the
current of peak ‘Q’ decreases as the scan reversal potential becomes
more negative (Fig. 3c). This means that peak ‘Q’ is a reductive peak
of some transient complex ions. As shown in Fig. 11, the current
values associated with peaks ‘Q’ and ‘R’ decrease as the number
of scans increase. This implies some sort of interrelation between
peaks ‘Q’ and ‘R’. As shown in Fig. 2, peak ‘Q’ does not appear in
the CV curves for the baths without Sn2+ ions. As shown in Fig. 12,
peak ‘Q’ also does not appear in the CV curves for the bath with-
out TEA. It can be seen from Figs. 2, 3 and 12, that peak ‘Q’ only
Fig. 4. CV curves of Sn–Ag–Cu alloy electrodeposited in a weakly acidic bath
containing 0.20 mol L−1 Sn(CH3SO3)2, 4.5 mmol L−1 AgI, 1.5 mmol L−1 Cu(CH3SO3)2,
0.60 mol L−1 K4P2O7, 1.35 mol L−1 KI, 0.225 mol L−1 TEA, 6.4 mmol L−1 HT and
9 mmol L−1 hydroquinone in scan potential range of 0 V to −0.8 V at a scan rate of
100 mV s−1
.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
2886
J. Zhang et al. / Electrochimica Acta 54 (2009) 2883–2889
Fig. 7. CV curves of Sn–Cu and Sn–Ag alloys electrodeposited from weakly acidic
baths containing 0.20 mol L−1 Sn(CH3SO3)2, 0.60 mol L−1 K4P2O7, 1.35 mol L−1 KI,
0.225 mol L−1 TEA, 6.4 mmol L−1 HT, 9 mmol L−1 hydroquinone and (a) 1.5 mmol L−1
Cu(CH3SO3)2, (b) 4.5 mmol L−1 AgI at a scan rate of 100 mV s−1
.
appears in the CV curves for the bath containing a mixture of tin
salts and TEA. This confirmed that the transient complex ions cor-
respond to tin and TEA. In the bath, TEA is used as a complexing
agent for Cu2+, and TEA acts as an assistant brightener, and can
decrease cathodic polarization [11]. Considering the complexing
effect from TEA, it is presumed that during the cathodic period,
tin pyrophosphate complex ions decompose. This is particularly
because the potential is more negative and the tin–TEA transient
complex ions, expressed as [Sn(TEA)x]2+, form. Peak ‘Q’ and peak
Fig. 6. CV curves of Sn–Ag–Cu alloy electrodeposited from a weakly acidic bath
containing 0.20 mol L−1 Sn(CH3SO3)2, 4.5 mmol L−1 AgI, 1.5 mmol L−1 Cu(CH3SO3)2,
0.60 mol L−1 K4P2O7, 1.35 mol L−1 KI, 0.225 mol L−1 TEA, 6.4 mmol L−1 HT and
9 mmol L−1 hydroquinone in different scan potential ranges: (a) 0 V to −0.9 V, (b)
0 V to −0.95 V and (c) 0 V to −1.1 V at a scan rate of 100 mV s−1
.
Fig. 8. SEM image of a Sn–Ag–Cu alloy coating electrodeposited potentiostati-
cally at −0.85 V from a weakly acidic bath containing 0.20 mol L−1 Sn(CH3SO3)2,
4.5 mmol L−1 AgI, 1.5 mmol L−1 Cu(CH3SO3)2, 0.60 mol L−1 K4P2O7, 1.35 mol L−1 KI,
0.225 mol L−1 TEA, 6.4 mmol L−1 HT and 9 mmol L−1 hydroquinone.