Corrosion Science 50 (2008) 2437–2443
Contents lists available at ScienceDirect
Corrosion Science journal homepage: www.elsevier.com/locate/corsci
Electrochemical corrosion behaviour of Pb-free Sn–8.5Zn–0.05Al–XGa and Sn–3Ag–0.5Cu alloys in chloride containing aqueous solution Udit Surya Mohanty, Kwang-Lung Lin * Department of Materials Science and Engineering, National Cheng Kung University, Tainan 701, Taiwan, ROC
a r t i c l e
i n f o
Article history: Received 27 March 2007 Accepted 17 June 2008 Available online 10 July 2008 Keywords: Anodic polarization Corrosion Pb-free solder alloys
a b s t r a c t The electrochemical corrosion behaviour of Pb-free Sn–8.5Zn–0.05Al–XGa and Sn–3Ag–0.5Cu alloys was investigated in 3.5% NaCl solution by using potentiodynamic polarization techniques. The results obtained from polarization studies revealed that there was a negative shift in the corrosion potential with increase in Ga content from 0.02 to 0.2 wt% in the Sn–8.5Zn–0.05Al–XGa alloy. These changes were also reﬂected in the corrosion current density (Icorr) value, corrosion rate and linear polarization resistance (LPR) of the four element alloy. However, for Sn–3Ag–0.5Cu alloy a signiﬁcant increase in the corrosion rate and corrosion current density was observed as compared to the four element alloys. SIMS depth proﬁle results established that ZnO present on the outer surface of Sn–8.5Zn–0.05Al–0.05Ga alloy played a major role in the formation of the oxide ﬁlm. Oxides of Sn, Al and Ga contributed a little towards the formation of ﬁlm on the outer surface of the alloy. On the other hand, Ag2O was primarily responsible for the formation of the oxide ﬁlm on the outer surface of Sn–3Ag–0.5Cu alloy. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction Pb–Sn alloys have been the most dominant solders widely used in manufacturing, because of their unique combination of material properties and low cost. In view of the toxicity of Pb to the human body and its danger to the environment, more and more countries have begun to prohibit the use of Pb-containing solder alloy in the packaging industry. These solders are mostly based on Sn-containing binary and ternary alloys. The binary systems include Ag–Sn, Au–Sn, Sb–Sn and Sn–Zn alloys [1–7] and some of the ternary alloys include Bi–Sb–Sn , Sn–Zn–In [9,10], etc. The Sn–Zn system is also a promising alternative for Pb–Sn solder alloys. Unlike other common alloying elements, Zn is active both chemically and metallurgically. These aspects dramatically decrease the wettability and corrosion behaviour when formed as a solder alloy. In order to improve the corrosion resistance, the galvanizing industry has incorporated Al into the galvanized coatings for steel . The eutectic Zn–5% Al coating is found to be superior to pure Zn coating as far as atmosphere corrosion is concerned. Lin et al. [12,13] have investigated the corrosion behaviour of Sn–Zn–Al and Sn–Zn–Al– In alloys in 3.5% NaCl solution. They observed that Sn–Zn–Al alloy  underwent more active corrosion than 63Sn–37Pb alloy. In another study  they noticed that 5In–9(5Al–Zn)–YSn and 10In– 9(5Al–Zn)–Sn alloys exhibited electrochemical passivation behaviour and the polarization behaviour of these two alloys were similar to that of 9(5Al–Zn)–Sn alloy. * Corresponding author. Tel.: +886 6 2762709; fax: +886 6 2759602. E-mail address: [email protected]
(K.-L. Lin). 0010-938X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.corsci.2008.06.042
Among the tin-based solder alloys, the Sn–Ag–Cu alloys, such as Sn–3.5Ag–0.7Cu, Sn–4.7Ag–1.7Cu have been generally recommended as substitutes for Pb–Sn eutectic solder . The main beneﬁts of the Sn–Ag–Cu alloy system are its relatively low melting temperature compared to Sn–Ag binary eutectic alloy, its superior mechanical properties as well as relatively good solderability. Mohanty and Lin  have reported the effect of Ag on the electrochemical corrosion behaviour of Sn–8.5Zn–XAg–0.1Al–0.5Ga alloy in 3.5% NaCl solution. They observed that an increase in the Ag content from 0.1 to 1.5 wt% decreased the corrosion current density and shifted the corrosion potential towards more noble values. The presence of Ag atoms in the oxide layer also improved the passivation behaviour of the alloy to a certain extent. The effect of Al on the electrochemical corrosion behaviour of Sn–8.5Zn–0.5Ag– XAl–0.5Ga alloy in 3.5% NaCl solution was also investigated by the same authors  by using potentiodynamic polarization techniques. The results showed that an increase in the Al content to 1.5 wt% decreased the corrosion current density and corrosion rate of the alloy. Further increase in Al content to 3 wt% enhanced the corrosion rate of the 5 element alloy. However, there is very little information available on the electrochemical corrosion behaviour of four element alloy. Polarization studies  made on Sn–8.5Zn–XAl–0.5Ga alloy in 3.5% NaCl solution revealed that addition of Al from 0.02 to 0.05 wt% decreased the corrosion current density and corrosion rate of the alloy. Higher Al content above 0.05 wt% increased the corrosion rate and corrosion current density to a signiﬁcant extent. The present paper investigates the electrochemical corrosion behaviour of Sn–8.5Zn–0.05Al–XGa alloy in 3.5% NaCl solution by using
U.S. Mohanty, K.-L. Lin / Corrosion Science 50 (2008) 2437–2443
Sn-3Ag-0.5Cu Sn-8.5Zn-0.05Al Sn-8.5Zn-0.05Al-0.02 Ga
Potential (mV vs SSCE)
E D C1
log (I) (A/cm2) Fig. 1. Potentiodynamic polarization curves depicting the corrosion behaviour of Sn–8.5Zn–0.05Al–XGa and Sn–3Ag–0.5Cu alloy in 3.5% NaCl solution.
Sn-8.5 Zn-0.05Al-0.05Ga Sn-8.5Zn-0.05Al-0.2Ga Sn-8.5Zn-0.05Al
by the methods described earlier . The alloy constituents were then weighed and melted in a crucible followed by natural cooling in air. The alloy thus formed after cooling was polished with SiC paper of grit 240, 800, 1200 and 2000, respectively, rinsed with distilled water followed by cleaning in an ultrasonic cleaner (Delta D 2000). The specimen was then dried and placed in the rectangular test cell. A detailed description of the rectangular cell has been reported  in our previous paper. Potentiodynamic polarization measurements were conducted in the cell containing 3.5% NaCl solution. The solution was de-aerated with N2 for 1 h prior to the polarization studies. Similar procedures were also followed for Sn–3Ag–0.5Cu alloy. A Pt wire and Ag/AgClsat electrode were used as the counter and reference electrodes, respectively. Potentiodynamic polarization experiments were carried out at a scan rate of 1 mV/s. The polarization curves were recorded +200 mV above and below the OCP in the potential range of 1200 to 1200 mV. The specimens were investigated for their surface corrosion products formed after the polarization study up to the potentials of interest. The morphology of the corrosion products were determined by using scanning electron microscopy (SEM) and energy dispersive X-ray (EDX) analysis. The elements distributed across the outer surface of the alloy and its underlying layers were analysed by using secondary ion mass spectrometry (SIMS). CAMECA made IMS-6f ion microscope and ion microprobe using the Cs+ microbeam energy of 10 keV was used. Primary ion beam of Cs+ with 50 nA beam current was rastered over an area of 150 lm 150 lm on the specimen surface. The resultant intensity of the secondary ions was plotted against the sputtering time to understand the variations in the composition of the various elements distributed across the modiﬁed layer. A sputtering rate of 1 lm of depth per hour was maintained. The Cameca IMS-6f has a realistic limit of detection for elements in the range of 1–100 ppb. Such determinations have a typical spatial resolution of 10–20 lm.
Potential (mVvs SSCE)
3. Results and discussion 3.1. Corrosion studies
A comparative study on the electrochemical corrosion behaviour of three sets of alloys was investigated in 3.5% NaCl solution by using potentiodynamic polarization techniques. Polarization curves for Sn–8.5Zn–0.05Al, Sn–8.5Zn–0.05Al–XGa and Sn–3Ag– 0.5Cu alloy are plotted in Fig. 1 and different points (A–G) in the polarization curves are analysed. The polarization experiments are actually carried out in the potential range of 2100 to +100 mV at a scan rate of 1 mV/s. Since all the corrosion tests were conducted in de-aerated NaCl solution, the only feasible cathodic reaction is the evolution of hydrogen from water reduction (cathodic region AB) 
log (I) (A/cm ) Fig. 2. Potentiodynamic polarization curves depicting the corrosion behaviour of Sn–8.5Zn–0.05Al alloy and Sn–8.5Zn–0.05Al–XGa in 3.5% NaCl solution.
2H2 O þ 2e ! 2OH þ H2
The current rises sharply from point B where dissolution of Zn occurs according to the following reaction 
Zn þ 2H2 O ! ZnðOHÞ2 ðsÞ þ 2Hþ þ 2e
2Zn þ 2NaCl þ 3H2 O ! ZnOZnCl2 ðsÞ þ 2NaOH þ 4H þ 4e potentiodynamic polarization techniques. The polarization measurement of Sn–3Ag–0.5Cu alloy was also examined and a comparative study was made between the two alloys. 2. Experimental Alloys of Sn, Zn, Al and Ga were prepared from pure elements. The Ga content investigated in the Sn–8.5Zn–0.05Al–XGa alloy was 0.02–0.2 wt%. The Sn-8.5Zn-0.05Al-XGa alloy was prepared
Point B is referred as the corrosion potential (Ecorr) where the extrapolated anodic and cathodic slopes intersect and the current becomes zero. The dissolution of zinc starts at point B and continues up to the region C. As a result of oxidation of zinc from the surface of the alloy, other alloying elements like Zn, Ga and Al gets exposed and could be seen on the surface of the alloy. A small current density peak C1 is formed very close to the point C in Fig. 1. The peak might be due to the dissolution of Al  from the surface of the alloy.
U.S. Mohanty, K.-L. Lin / Corrosion Science 50 (2008) 2437–2443 Table 1 List of corrosion parameters obtained by the potentiodynamic polarization of Sn–8.5Zn–0.05Al–XGa and Sn–3Ag–0.5Cu alloys Composition (mass%)
Icorr (mA cm2)
Corrosion rate (mm yr1)
LPR (X cm2)
ic (mA cm2)
ip (mA cm2)
Sn–8.5Zn–0.05Al 0.02Ga 0.05Ga 0.2Ga Sn–3Ag–0.5Cu
951 1197 1144 1266 874
1.27 1.1 1.13 1.19 3.16
34.0 29.4 30.3 31.3 84.3
77.3 89.4 86.7 83.7 31.4
15.8 3.53 0.71 5.79 26.4
10.3 3.2 0.78 4.34 3.49
153 658 741 688 97.8
348 303 371 250 471
Fig. 3. SE micrograph and EDX analysis of Sn-8.5Zn-0.05Al-0.05Ga alloy polarized to 371 mV (Point F).
The dissolution reaction might proceed through the following steps:
AlðsÞ þ OH ! AlðOHÞads þ e
AlðOHÞads þ OH ! AlðOHÞ2ads þ e AlðOHÞ2ads þ OH ! AlðOHÞ3ads þ e
The active dissolution of aluminium continues with increase in potential upto point D where a broad current density peak is obtained. This broad peak may be due to the oxidation of Ga on the surface of the alloy. Further increase in potential increases the anodic current density, which becomes maximum at point E. The point E is actually referred as the passivation potential (Epp) and the current at this potential is known as the critical current density (ic). The transition from active dissolution occurs at the passivation potential where the solid species i.e. oxides of Zn, Al and Ga become more stable
than the parent metal ion. These oxides accumulate near the surface of the electrode and result in a slight decrease in current density with increase in potential towards F. This indicates a slight passivation behaviour or initiation of corrosion with a constant rate. Remarkably it can be noticed from Fig. 2 that the anodic current density for the pure Sn–8.5Zn–0.05Al in the EF region is higher than the four element Sn–8.5Zn–0.05Al–XGa alloys. These changes are clearly seen in the marked increase in the corrosion current density and corrosion rate of the Sn–8.5Zn–0.05Al alloy as compared to the investigated four element alloys (Table 1). On the other hand, a signiﬁcant decrease in anodic current density is observed in the EF region for the Sn–3Ag–0.5Cu alloy (Fig. 1). The anodic current density value noted at the end of region EF in Fig. 1 is termed as the passivation current density denoted by the symbol ip. Table 1 presents a detailed data on the ip, ic and Epp, values obtained from the polarization curves of Sn–8.5Zn–0.05Al–XGa
U.S. Mohanty, K.-L. Lin / Corrosion Science 50 (2008) 2437–2443
Fig. 4. SE micrograph and EDX analysis of Sn–3Ag–0.5Cu alloy polarized upto +100 mV (end of the electrochemical experiment).
and Sn–3Ag–0.5Cu alloys in Fig. 1. The data shows that the Sn8.5Zn–0.05Al–0.05Ga alloy has the lowest ip and ic values compared to the other alloys. Singh and Gupta  have reported that lower values of ip and ic correspond to the stability and protective nature of the passive ﬁlm. The ic value for Sn–3Ag–0.5Cu alloy differs from ip by a higher margin suggesting an increase in corrosion of the alloy (84.3 mm yr1, Table 1). The slight passivation behaviour of Sn– 8.5Zn–0.05Al–XGa alloy as compared to other alloys might be attributed to the formation of high concentration of ZnO with low content of Sn and Al oxides on the outer surface of the alloy. Several authors [24–26] have reported the role of ZnO in the formation of thin layer of oxide ﬁlm. SEM and EDX analysis in Fig. 3 also reveals high concentration of ZnO on the outer surface of Sn–8.5Zn–0.05Al– 0.5Ga alloy polarized to 371 mV. Oxides of Sn and Al are present in considerably lower amounts than Zn. A sharp increase in current density is observed at point F with further increase in potential in the anodic direction. This potential at point F is referred as the breakdown potential (EBr). Cl ion present on the outer surface of Sn–8.5Zn–0.05Al–XGa alloy is responsible for the breakdown of the oxide ﬁlm at 371 mV. Pistorius and Burstein  have reported that chloride ions penetrate into the oxide ﬁlm and form solid metal chloride which causes the mechanical breakdown of the ﬁlm. The EBr value for Sn–3Ag–0.5Cu alloy (Table 1) differs signiﬁcantly from the EBr values obtained for the Sn–8.5Zn–0.05Al–XGa alloys. Since the stability of the oxide ﬁlm depends on the breakdown potential, hence changes in the EBr value are reﬂected in the linear polarization resistance (LPR) of the alloys (Table 1). The corrosion products formed for Sn–3Ag–0.5Cu
alloy after the end of the polarization experiments were also examined by SEM and EDX analysis. EDX analysis conﬁrmed the presence of high amount of Sn and Cl in the coalesced needle shaped structure shown in Fig. 4. This structure was similar to that obtained by Chang et al.  for Sn–3.5Ag alloy. Similar structures were also observed for four element Sn–8.5Zn–XAl–0.5Al alloy in our previous studies . The corrosion potential (Ecorr) and corrosion current density (Icorr) for pure Sn–8.5Zn–0.05Al alloy without Ga are found to be 951 mV and 1.27 mA cm2, respectively (Table 1). Addition of 0.02 wt% Ga to the above alloy shifts the corrosion potential towards more negative values. Further increase in the Ga content to 0.2 wt% progressively increases the corrosion current density and corrosion rate of Sn–8.5Zn–0.05Al–XGa alloy. These changes are also reﬂected in the linear polarization resistance (LPR) of Sn8.5Zn-0.05Al-XGa alloy. On the other hand, a signiﬁcant increase in the corrosion current density (Icorr) and corrosion rate are noted for Sn–3Ag–0.5Cu alloy in comparison to other alloys (Table 1). This might be ascribed to the poor passivation ability shown by the three element alloy. 3.2. Surface analysis by SIMS SIMS technique was used to measure the surface and depth proﬁles of the corroded specimens formed by the potentiodynamic polarization of Sn–8.5Zn–0.05Al–XGa and Sn–3Ag–0.5 Cu alloys in 3.5% NaCl solution. Fig. 5 displays the SIMS depth proﬁle for Sn–8.5Zn–0.05Al–0.05Ga alloy polarized to 1144 mV (point B)
U.S. Mohanty, K.-L. Lin / Corrosion Science 50 (2008) 2437–2443
Fig. 5. SIMS depth proﬁle for Sn–8.5Zn–0.05Al–0.05Ga alloy potentiodynamically polarized to 1144 mV (Point B). Fig. 6. SIMS depth proﬁle for Sn–8.5Zn–0.05Al–0.05Ga alloy potentiodynamically polarized to 1086 mV (Point C).
in the polarization curve. The depth proﬁle results show that ZnO is present in very high concentration on the surface and its atomic concentration is found to decrease with increase in sputtering time to 3000 s. This suggests that most of the zinc might have oxidized and segregated to the outer surface of the alloy. The major alloying element Sn thus gets exposed as a result of oxidation of zinc from the outer surface and its concentration remains constant throughout with increase in sputtering time to 3500 s suggesting that most of the Sn occupies the bulk of the alloy. SIMS depth proﬁle for Sn–8.5Zn–0.05Al–0.05Ga alloy polarized to 1086 mV (Point C) is shown in Fig. 6. The results show that the concentration of Al is highest on the surface followed by Ga. This suggests that as a result of dissolution of zinc from the surface of the alloy, Al and Ga gets exposed and are prone to further oxidation. Thus the possibility of formation of oxides/hydroxides of Al is higher near the surface as a result of dissolution of aluminium. The formation of Al(OH)3 at point C1 very close to the point C is thus conﬁrmed by SIMS analysis. High concentration of Cl ion is also noticed on the outer surface indicating the possible formation of ZnO ZnCl2 at the corrosion potential B. Furthermore, SIMS depth proﬁle was carried out on the Sn– 8.5Zn–0.05Al–0.05Ga alloy polarized to 1013 mV (Point D) and the results are shown in Fig. 7. The results reveal that Ga concentration is highest on the outer surface and its concentration gradually decreases with increase in sputtering time to 2500 s (Fig. 7). This decrease suggests that all the Ga might have segregated to the outer surface and would have oxidized. These results are consistent with our assumptions described in Section 3.1. Since the oxides of Zn, Al and Sn are present in low concentrations on the surface; their formation on the outer surface of the alloy is virtually ruled out. Fig. 8 demonstrates the SIMS depth proﬁle for Sn–8.5Zn– 0.05Al–0.05Ga alloy polarized to 371 mV (Point F). The results
Fig. 7. SIMS depth proﬁle for Sn–8.5Zn–0.05Al–0.05Ga alloy potentiodynamically polarized to 1013 mV (Point D).
U.S. Mohanty, K.-L. Lin / Corrosion Science 50 (2008) 2437–2443
Fig. 8. SIMS depth proﬁle for Sn–8.5Zn–0.05Al–0.05Ga alloy potentiodynamically polarized to 371 mV (Point F).
Fig. 9. SIMS depth proﬁle for Sn–3Ag–0.5Cu alloy potentiodynamically polarized to 365 mV (Point D).
show that the atomic concentration of oxides on the outer surface of the alloy varies in the following order ZnO > SnO > Al2O3 > Ga2O3. From the SIMS analysis it can be said the ﬁlm formed on Sn–8.5Zn–0.05Al–0.05Ga alloy might be composed of high concentration of ZnO with low concentration of Al2O3 and SnO. The stability of these oxides, i.e. oxides of Zn, Sn, Al and Ga can be ascertained from their corresponding Gibbs free energy values (DGf for ZnO = 320.5 kJ/mol, DGf for SnO = 251.9 kJ/mol DGf for Al2O3 = 1582.3 kJ/mol, DGf for Ga2O3 = 998.3 kJ/mol) . The oxide data mentioned above is per mole of oxide formed at 298 K. The above theoretical data on free energy suggests that ZnO being thermodynamically unstable gets easily oxidized from the bulk of the alloy and forms on the outer surface of the alloy. The concentration of Al2O3 is present in lower amounts on the outer surface than oxide of Zn because of its stability and high Gibbs free energy (DGf ). Ga2O3 however contributes a little to the formation of oxide ﬁlm on the outer surface of the alloy. SIMS depth proﬁle (Fig. 9) performed on Sn–3Ag–0.5Cu alloy polarized to 365 mV (Point D) shows that the alloying element Sn possesses high atomic concentration compared to other elements Ag and Cu. The concentration of oxygen and Cl ion seems to be very high on the surface indicating the possible formation of SnCl2 or SnOCl2 at point D. Fig. 10 shows the SIMS depth proﬁle results for the Sn–3Ag– 0.5Cu alloy polarized to 97.8 mV (Point E, Fig. 1). The results reveal that a mixture of Sn, Ag and Cu oxides accumulate near the passivation potential E before the formation of the passive ﬁlm in the region EF. Fig. 11 displays the SIMS depth proﬁle for Sn– 3Ag–0.5Cu alloy polarized to 471 mV (Point F). The results show that among all the oxides, the concentration of Ag2O is found to be the highest on the outer surface of the alloy. Its concentration is found to decrease with gradual increase in sputtering time to
Fig. 10. SIMS depth proﬁle for Sn–3Ag–0.5Cu alloy potentiodynamically polarized to 97.8 mV (Point E).
U.S. Mohanty, K.-L. Lin / Corrosion Science 50 (2008) 2437–2443
2. The corrosion current density (Icorr) and corrosion rate of the Sn-8.5Zn–0.05Al–XGa alloy shows a progressive decrease with increase in Ga content from 0.02 to 0.2 wt%. These changes are also reﬂected in the linear polarization resistance of the alloy. Nevertheless, a signiﬁcant increase in the corrosion rate and corrosion current density is observed for Sn–3Ag–0.5Cu alloy. 3. SIMS depth proﬁle reveal that the outer surface of the alloy for Sn–8.5Zn–0.05Al–XGa alloy is primarily composed of high concentration of ZnO. Oxides of Sn, Al and Ga contribute a little towards the formation of oxide ﬁlm on the surface. 4. Ag2O was primarily responsible for the formation of the oxide ﬁlm on the outer surface of Sn–3Ag–0.5Cu alloy. SnO and Cu2O make a little contribution towards the formation of oxide ﬁlm on the surface of the alloy. 5. The presence of Cl ion on the outer surface of the alloy is responsible for the breakdown of the oxide ﬁlm. Acknowledgement The authors acknowledge the ﬁnancial support of this study from the National Science Council of ROC under NSC94-2811-E006-021. References
Fig. 11. SIMS depth proﬁle for Sn–3Ag–0.5Cu alloy potentiodynamically polarized to 471 mV (Point F).
3000 s, indicating that all the Ag might have segregated to the outer surface and have been oxidized. The concentrations of Cu and its oxides on the outer surface are found to be lower than Ag2O; however the decrease in its atomic concentration is not progressive with the sputtering time suggesting that some Cu is still present in the sputtered layers. The concentration of SnO on the outer surface is however found to be low than the other two oxides. The stability of these oxides on the outer surface of three element alloy can be explained from the standard Gibbs free energy values (DGf for SnO = 251.9 kJ/mol, DGf for Ag2O = 11.2 kJ/mol, DGf for Cu2O = 146 kJ/mol) . From the DGf values it can be stated that the SnO is the most stable oxide and hence does not get completely oxidized from the bulk of the alloy. Ag2O being the most unstable and thermodynamically unfavourable gets easily oxidized from the bulk and thus forms prominently on the outer surface of the alloy. The passivation behaviour of Sn–3Ag–0.5Cu alloy can thus be attributed to the presence of high concentration of Ag2O on the outer surface of the oxide ﬁlm. 4. Conclusions 1. The corrosion potential (Ecorr) value for Sn–8.5Zn–0.05Al–XGa alloy is found to shift towards more negative values with increase in Ga content from 0.02 to 0.2 wt%. The Ecorr value for Sn–3Ag–0.5Cu alloy however tends to shift towards more noble values.
 P.T. Vianco, D.R. Frear, J. Met. 47 (1993) 14.  J.W. Morris, J.L. Freer Goldstein, Z. Mei, J. Met. 47 (1993) 28.  C.H. Raeder, L.E. Felton, V.A. Tanzi, D.B. Knorr, J. Electron. Mater. 23 (7) (1994) 611.  L.E. Felton, C.H. Raeder, D.B. Knorr, J. Met. 47 (1993) 62.  W.L. Winterbottom, J. Met. 47 (1993) 20.  W.J. Tomlinson, I. Collier, J. Mater. Sci. 22 (1987) 1835.  M. Harada, R. Satoh, IEEE Trans. Compon. Hybr. Manuf. Technol. 13 (1990) 736.  G. Ghosh, M. Loomans, M.E. Fine, J. Electron. Mater. 23 (1994) 619.  M. McCormack, S. Jin, H.S. Chen, D.A. Machusak, J. Electron. Mater. 23 (1994) 687.  M. McCormack, S. Jin, J. Electron. Mater. 23 (1994) 635.  L. Allegra, J.C. Zoccola, Mater. Perform. 22 (1993) 5.  K.L. Lin, T.P. Liu, Mater. Chem. Phys. 56 (1998) 171.  K.L. Lin, F.C. Chung, T.P. Liu, Mater. Chem. Phys. 53 (1998) 55.  K.W. Moon, W.J. Boetinger, U.R. Kattner, F.S. Biancaniello, C.A. Handwerker, J. Electron. Mater. 29 (2000) 1122.  U.S. Mohanty, K.L. Lin, Mater. Sci. Eng. A 406 (2005) 34.  U.S. Mohanty, K.L. Lin, Appl. Surf. Sci. 252 (2006) 5907.  U.S. Mohanty, K.L. Lin, J. Electrochem. Soc. 153 (8) (2006) B319–B324.  K.I. Chen, K.L. Lin, in: International Symposium on Electronic Materials and Packaging, 2002, p. 49.  U.S. Mohanty, K.L. Lin, Corros. Sci. 48 (2006) 662.  H. Oulfajrite, A. Sabbar, M. Boulghallat, A. Jouaiti, R. Lbibb, A. Zrinch, Mater. Lett. 57 (2003) 4368.  X. Zhang, S. Lo Russo, A. Miotello, L. Guzman, E. Cattaruzza, P.L. Bonora, L. Benedetti, Surf. Coat. Technol. 141 (2001) 187.  W.A. Badawy, F.M. Al-Kharaﬁ, A.S. El-Azab, Corros. Sci. 41 (1999) 709.  V.B. Singh, A. Gupta, Mater. Chem. Phys. 85 (2004) 12.  T.H. Muster, I.S. Cole, Corros. Sci. 46 (2004) 2319.  E.E. Abd El Aal, Anti-Corrosion 46 (1999) 349.  S.S. Abd El Rehim, E.E. Fouad, S.M. Abd El Wahab, H.H. Haassan, J. Electroanal. Chem. 401 (1996) 113.  P.C. Pistorius, G.T. Burstein, Corros. Sci. 38 (1994) 1525.  T. Chang, M.H. Hon, M.C. Wang, D.Y. Lin, J. Electrochem. Soc. 151 (7) (2004) C484–C491.  David R. Lide, Handbook of Chemistry and Physics, 73rd ed., CRC Press, Boca Raton, Florida, 1993. pp. 5–75.