Currently Cu–Sn coatings are widely used in numerous commercial productions in replacement of Ni layersespecially for decorative arts. In recent decades, researchers are exploring more environment-friendly baths todeposit Cu–Sn coatings to substitute CN− electrolyte which has been widely used as conventional electroplatingbaths for industrial applications. This paper investigated Cu–Sn deposition from a methanesulfonic acid basedbath studying the optimization of deposition parameters. Film composition was varied by changing currentdensity, which leads to a difference in color and final properties. Usually lifetime of tin containing baths is limitedbecause the composition of deposited films has poor stability. In this work the effect of deposition parametersand anode materials on bath and deposits stability is evaluated. In particular, the application of pulse currentwas considered to improve the quality of coatings at fixed duty cycle while deposition frequency was varied.The coatings deposited under different conditions were compared in term of microstructure and corrosionbehavior.
In decorative applications, electrodeposited nickel is routinely usedas an undercoat for noble metal finishes. However, nickel is a knownskin allergen and should avoid contacting human skin such as jewellery[1]. Bronze coatings, CuSn(Zn), are a known alternative to nickel fortheir good corrosion resistance, ductility and solderability and thereforeare widely used in many different applications [2]. To provide a brightsurface and retain the attractive appearance as well as improve thecoating's adhesion to the surface, a lustrous copper underlayer isconventionally placed underneath bronze layer [3]. As coatings, bronzecan be electrodeposited from various electrolytes: the mostly usedbath is cyanide based [4], which is known for its toxic influence toenvironment as well as human being. Various acid electrolytes havebeen studied to be applied as suitable baths in replacement of cyanide,such as sulfuric acid [1,5–7] and pyrophosphate [8,9]. However, theysuffer from problems of low plating speed and low stability of bath,which make them unavailable as industrial options. As an alternative,methanesulfonic acid (MSA) has been studied as the favorable aqueouselectrolyte to electrodeposit metals or metallic oxides due to the bath'sexcellent metal salt solubility, ease of effluent treatment, low toxicity,low corrosivity and biodegradability [10–22]. Also, unlike sulfuric acid,MSA is a less oxidizing acid and therefore the oxidation of tin (II) totin (IV) has a much lower rate improving the bath stability [11,19–21].MSA bath has been studied by some groups in the last 10 years.The works present in literatures focus on 2 main topics: the influence
39 0461 281977.la).
ghts reserved.
rf. Coat. Technol. (2013), http
of different concentrations and ratios between metallic salts [14,21]and the use of surfactants and oxidation inhibitors used to reduce theformation rate of Sn4+ ions [20,23]. Recently, industries are also pursuingdevelopment on commercial bath based on MSA [24].
To obtain bronze with superior properties, pulse plating is usedto control the structure and properties of electrodeposited materials. Ithas been proved to be a useful tool for improvement of metallic coatingproperties [25]. Landolt and Marlot reviewed the effect of pulse current(PC) onmicrostructure and composition ofmetallic coatings [26]. Higherthrowing power of copper deposition can be realized by applying pulseplating [22], leading also to deposition on irregular shapes and surfacesthat can be effectively leveled up [10].
However, previous investigators primarily concentrated on directcurrent (DC) electrodeposition of Cu–Sn alloys from various acids[1,7,27–39], and the effect of various parameters like deposition currentdensity [37] or potential [38], additive agents [1,7,29], metal concen-tration [28,40], annealing conditions [36], substrates [29], electroplatingtime [34] on copper content of the deposit and predictive modelswere explored [27,30]. A few attempts have been made for the PCelectroplating of Cu–Sn alloys [4,41] though they are depositedfrom pyrophosphate or cyanide based baths. S.D. Beattie et al. [41]showed the changes of average stechiometry and phases presentas a function of the position on the substrate foil in a hull cell withlow frequencies. T. Nickchi et al. [4] prepared composite coatingscontaining self-lubricant particles like graphite in bronzefilm to decreasethe coefficient of friction.
To our knowledge, no report has been recorded on the influence ofpulse plating on the composition andmicrostructure of bronze coatingsdeposited from MSA based bath. In this work, three electrodeposition
Fig. 1. Effect of current density on copper content of films deposited with copper anode(solid line) and graphite anode (dotted line).
2 C. Zanella et al. / Surface & Coatings Technology xxx (2013) xxx–xxx
parameters – choice on anode material, current density under directcondition and frequency values under pulsed condition – were variedto control the physical and chemical characteristics of the depositsobtained from the described bath. First, effect of current density on Cucontent was studied by DC method in order to cover the wide rangeof compositions with lustrous appearance in silver-like color. Then,copper and graphite were compared as anodematerial to achieve longerworking time of the bath. Ultimately, to obtain bettermicrostructure andcorrosion resistance behavior compared to DC, different frequenciesunder PC depositionwere investigated. Satisfying results can be achievedunder optimized conditions.
2. Experimental
Methanesulfonic acid based bronze electroplating solution containingCu2+ 6 g/l, Sn2+ 3 g/l, MSA 300 ml/l (mole ratio Cu:Sn = 3.71) wascommercially available from Enthone Group and operated at roomtemperature (25 °C). It contains 300 ml/l acid concentrate, 200 ml/lmake-up, 3 ml/l brightener, 20 ml/l wetting agent and 10 ml/l additiveHG besides the bulk metal salts.
These additives not only are brighteners but also are needed tostabilize tin ions. Other examples of electrodeposition copper–tin alloysfrom methanesulfonic acid-based bath can be found with differentadditives, e.g. 200 g dm−3MSA based bath with addition of primarygrain refiner, second grain refiner and brightener was used toelectrodeposit 99 wt.% Sn at a 10 A/dm2[42] and 2 mol dm−3 MSAbased bath with 25ppm MPS and 75ppm PEG to deposit 7% Sn [43].
Distilled water was used throughout. The depositionwas carried outin a cell with two electrodes fixed vertically parallel with both directcurrent as well as square wave current pulses at room temperature.Either a graphite sheet or a copper plate was used as anode material.Low alloy steel panels (4 cm × 4 cm) were used as substrates fordeposition and were pretreated by polishing with SiC paper (grit4000) degreasing with acetone, pickled in 0.1M H2SO4 for 15min andrinsed. When indicated, the steel panels were coated by bright Cuunderlayer layer (5 μm). Bronze bath was kept constantly undermechanical stirring at 350 rpm.
For deposition with direct current, different current density valueswere tested in range between 0 and 4 A/dm2. The deposition withpulse current was carried out varying the current density between1.3 A/dm2 (on-time) and zero (off-time) with a square wave pulseand a duty cycle of 50%, while different frequencies were applied from0.01 to 10 Hz. Deposition time was varied according to actual currentdensity in order to obtain constant thickness of 5 μm.
In order to monitor the cathodic deposition potential peaks andoxidation reactions on anodes, the bath was tested by linear scanningvoltammetry in a 3 electrode cell by a PAR potentiostat model 273.Platinum plate was used as working electrode, platinum foil as counterelectrode and the potential was swept from 0 to−0.8V vs. Ag/AgCl/3MKCl at a scan rate of 10mV/s.
Regarding the deposited coatings, the surface morphology andcomposition were characterized by Environmental Scanning ElectronMicroscopy (ESEM, Phips XL30) and energy dispersive X-ray spectros-copy (EDXS). A surface profilometer (MarSurf PS1) was used to measurethe surface roughness of steel substrate and electrodeposited films. Thecrystalline structure was investigated with X-ray diffraction (XRD)analysis. X-ray diffraction spectra were collected using a powderdiffractometer (Rigaku D-max), employing CuKα radiation and agraphite monochromator in the diffracted beam; typical scans wereperformed in the 2θ 10–110° range, with a sampling range of 0.05°and 6 s counting time.
Electrochemical behaviors of the deposits were characterized bypotentiodynamic polarization curves using a Princeton Applied ResearchPARSTAT 2273 potentiostat. A three-electrode cell was used with 1 cm2
deposits as working electrode, Ag/AgCl/3 M KCl as reference electrodeand platinum foil as counter electrode. The anodic polarization was
Please cite this article as: C. Zanella, et al., Surf. Coat. Technol. (2013), http
performed in 0.1 M NaCl solution (pH= 6.5) in the range of −0.03 Vvs. OCP to 1V vs. Ag/AgCl at a scan rate of 0.2mV/s at room temperature.
3. Results and discussion
3.1. Effect of direct current density on the composition of deposited coatings
In order to investigate the range of current that is favorable to silver-white layers, thefilmswere produced at different current densities from0 to 4A/dm2 with Cu plate as anode. Fig. 1 showed that copper contentof the deposit first decrease and then increase with the increaseof current density. The minimum of copper content (60 wt.%) wasobtained at about 3 A/dm2. This result can be explained by the factthat copper deposits at more positive potentials than that of tin[41,44]. The higher the current density, the more negative the potentialat the cathode. At very low current densities, the potentials on thecathode are positive enough to deposit only copper, not tin. At highercurrent densities (more negative potentials), both copper and tin areready to deposit. Given a 3.71:1 difference betweenmolarities of copperand tin ions in bath, the tin deposition is limited by diffusion andthe copper deposition is less limited when sufficiently higher currentdensities (sufficiently negative potentials) are applied. By visualobservations, the deposits show light yellow (I b 2 A/dm2) to silverwhite (2 A/dm2 ≤ I≤ 3 A/dm2) and then to light yellow (I N 3 A/dm2)changing trend corresponding to the composition change. Bright andcoherent deposits could be obtained in the range of 2.0–3.0 A/dm2
with target color.Graphite anode was also studied in narrower current range.
Comparable composition (57 wt.% Cu) could be obtained at muchlower current density, 1.3 A/dm2 since the composition trend seemsto be shifted at lower current deposition. Copper content of the layerdeposited with graphite anode is lower compared to the depositobtained at the same current density value with copper anode whencurrent density is lower than 2 A/dm2. Current efficiency of thedepositionprocesseswith both anodeswas very high, 95%–98%, coveringthe whole current range under investigation.
3.2. Effect of anode on the stability of the bath
A copper plate and a graphite sheet were adopted as anodematerialto check its influence on stability of composition of deposited films andlifetimeof the bath. Continuous depositions on separate substrateswerecarried out in two equally prepared baths. Current density was fixed at1.3 A/dm2 under DC and deposited films were controlled with 5 μmthickness. Each deposit requires approximately 200 C (0.056 Ah) toaccomplish. With the increase of electrical quantity passed, deposited
samples with copper anode experienced color change from silverwhite to light yellow at first three samples, then return to silver whitewhile color of the bath shows slightly turbid after several samples.On the contrary, deposited films with graphite anode keep a ratherstable color, silver white, during the whole process and also bath doesnot show any visual change in color. To verify the result, compositionanalysis was carried out on the samples with different aging times asshown in Fig. 2. It can be seen that copper content reached a maximumaround after 600C (0.167Ah), i.e. the third sample. Even the second andfourth samples show distinct different composition compared to thefirst one. On the contrary, samples resulted from graphite maintainconstant composition with only slight changes.
To find out the reason leading to deposit changes, linear scanningvoltammetry was carried out to show the redox species changesof baths which are new-prepared or have accomplished severalelectroplating processes, as shown in Fig. 2. The reduction processinitiates at a potential of approximately −0.17 V vs. Ag/AgCl whichlocates at the same value for copper deposition as reported in Ref [14].The solid line presents the linear scanning voltammetry of new bath.The two well-defined peaks with a maximum located at −0.4 V and−0.62V in the fresh bath represent the reduction of Cu(II) and Sn(II),separately. The transition of Cu(II) to Cu(0) occurs at the same potentialeven after the bath was aged by deposition with either copper orgraphite anode. The reduction of Sn(II) ions to Sn metal starts to occurat −0.5 V both in fresh and in old baths although the peak shapeis slightly steeper for old ones. With the change of anode material, the
200 400 600 800 100055
60
65
70
75
80
85
90
95
Cu anode graphite anode
Co
pp
er c
on
ten
t (w
t%)
Electric charge (C)
A
-0.8 -0.6 -0.4 -0.2 0.0-0.10
-0.09
-0.08
-0.07
-0.06
-0.05
-0.04
-0.03
-0.02
-0.01
0.00
0.01B
aged bath with copper anode aged bath with graphite anode new bath
Potential vs. Ag/AgCl (V)
Cu
rren
t d
ensi
ty (
A/c
m2 )
Fig. 2. (A) Copper content of continuously deposited films with copper anode (solid line)and graphite anode (dotted line) obtained from EDX analysis. (B) Linear scanning curvesin new bath (solid line), aged bath with graphite anode (dotted line) and aged bath withcopper anode (dot-dashed line).
Please cite this article as: C. Zanella, et al., Surf. Coat. Technol. (2013), http
reduction potential remains at −0.5 V showing that anode materialdid not alter the tin deposition potential. This is in agreement withwork carried out by other researchers [11,14]. As the potential becamemore negative, the current density reached a plateau followed by anincrease. The appearance of plateau can be associatedwith the completeconsumption of metallic ions at the electrode surface due to masstransport control of the reduction. The further increases in currentdensity, at a potential more negative than −0.7 V, are attributable tohydrogen evolution as a secondary reaction [1]. The hydrogen reactionstarts at more negative potential and lower rate in aged bath withgraphite anode compared to both new bath and the old one adoptingcopper anode. However, a new peak emerges at −0.48 V vs. Ag|AgClwhich help prove the occurrence of Sn(IV) during the electroplatingprocess with copper anode after 600 C (0.167 Ah) of aging. This newpeak is assumed to be related to the existence of Sn4+ due to oxidationfromSn2+. The bath aged using graphite anode on the contrary does notshow differences in peaks compared to initial bath, which states thatthe oxidation of Sn2+ can be acceleratedwith copper anodemore easilycompared to graphite. This can be explained by the fact that Sn2+ showsless noble oxidation potential than Cu0. As a result Sn2+ contained inthe bath has the tendency to be oxidized on the copper anode surface.Slight increase of reduction peak of Cu2+ can be attributed to the higherratio of Cu/Sn. In order to understand if oxidation extent of Sn2+ isdifferent on the 2 anodes, the anodic polarization curves (Fig. 3) ofdifferent anodes are carried out in baths containing only tin andwithoutmetallic ions. Copper anode exhibits much greater currents whenpolarized positively than that of graphite anode due to the continuousdissolution of bulk material when oxidized. Therefore it is difficult toextrapolate the current intensity due to Sn2+ oxidation, the differencebetween the current recorded in Sn-contained bath and in blank bathcan give an estimation of the oxidation of stannous ions. The increasingdisparity in current with potential increase is significantly largerthan the one of graphite which shows very limited currents in thesame conditions.
3.3. Effect of pulse frequency on the composition
For the PC deposition the graphite anode was used and the appliedcurrent is cycled between 1.3A/dm2 and zero at fixed duty cycle of 0.5.Copper content, reported in Fig. 4, is slightly higher in the entire rangestudied than that deposited with DC. The higher copper content is likelydue to the copper ion exchanges with the plated tin during the off pulseaccording to the reaction Cu+2(aq) + Sn(s)→ Cu(s) + Sn+2(aq) [45].Despite small increase of copper content, the appearance of filmsremains silver-colored. Such a behavior has been verified experimentallyfor Ni–Cu and Co–Cu alloys [46,47]. The change of alloy composition
0.0 0.2 0.4 0.6 0.8 1.0
0.0
0.3
0.6
0.9
1.2
Cu
rren
t d
ensi
ty (
A/c
m2 )
Potential vs. Ag/AgCl (V)
Cu anode in MSA Cu anode in Sn/MSA bath graphite anode in Sn/MSA bath
Fig. 3. Anodic polarization curves for copper anode in MSA electrolyte (solid line),copper anode in MSA electrolyte containing Sn salts (dotted line) and graphite anode inMSA containing Sn salts (dash-dotted line).
Fig. 4. Effect of frequency on copper content of films deposited with graphite anodeat current density of 1.3 A/dm2 and duty cycle of 0.5.
4 C. Zanella et al. / Surface & Coatings Technology xxx (2013) xxx–xxx
resulting from pulse plating has been numerically modeled by somestudies [48,49].
3.4. Microstructure and electrochemical properties of deposited coatings
SEM micrographs revealed the nodular growth morphology of theas-deposited surface of Cu–Sn layers coating under DC and PC platedconditions. In Fig. 5, a comparison between the DC and representativePC samples is presented. Prior to deposition, the substrate was slightlypolished to a surface roughness Ra of around 1.029 μm. The averageroughness of the layers deposited under PC is 0.416 μm which is muchlower than its DC counterparts (0.797 μm). The increasing frequencyshows no obvious influence on surface roughness of resulting layers.
C
A
Fig. 5. Secondary electron micrographs of Cu–Sn alloys obtained at (A) direct current (Band (C) magnification of sample in B.
Please cite this article as: C. Zanella, et al., Surf. Coat. Technol. (2013), http
Electron microscopy observation of the layer electrodepositedwith pulse current shows a nodular microstructure irrespective of theapplied frequency. Also, significantly peak broadening of the XRDpeaks was observed. The nodules in the SEM images together withthe broadened XRD peaks are evidence for a nanocrystalline coatingmorphology. However, increasing the deposition frequency nodulestructure and size seems to slightly increase and in this case aggregatein bigger granules.
Further investigation on microstructure was carried out by analysison X-ray diffraction patterns of deposits obtained under differentdeposition conditions reported in Fig. 6. The broad peaks may also bedue to the fact that Cu–Sn layers were relatively thin (5 μm) besidestheir nanocrystalline microstructure [21].
Peak intensity analysis was carried out by comparison with XRDpatterns in other investigations related to bronze coatings [7,14,21,44].It shows that the deposit is mostly tin, copper with Cu6Sn5 and Cu3Snintermetallics. The curves suggest that copper and tin are present inthe Cu–Sn deposit as Cu6Sn5. With increasing the applied frequencyfrom 0.01Hz to 10Hz, Cu is achieving higher portion in bi-phase depositwhile Sn is reduced in content by comparing the two significantpeaks arising between 40 and 45°. This result is compatible with thecomposition analysis of Cu(~65 wt.%) and Sn(~35 wt.%) obtained byEDXS, where the total copper content comes from the signal of Cu6Sn5alloy whose composition is ~40wt.% Cu and 60wt.% Sn, plus dissolutionof pure copper [23,50]. In other studies, copper–tin alloys can also beobtained via, e.g. heat treatment for 24 h to produce hexagonal Cu6Sn5and cubic Cu3Sn [51] and chemical reduction to produce 20–40 nmCu6Sn5 [52]. The diffraction line of 1Hz is not easily assigned to specificmetallic phase with shifted peaks. The two peaks appeared around35° can be assumed to the existence of meta-stable phases of theintermetallics, Cu6Sn5 and Cu3Sn, which is similar with the resultsobtained by aging alloys above 300 °C [53].
In termof corrosion protection, coatingswere deposited both on steelsubstrate and copper underlayer in order to evaluate the electrochemical
B
) 1 Hz with graphite anode at current density of 1.3 A/dm2 and duty cycle of 0.5,
Fig. 6.Characteristic XRD patterns obtained for Cu–Sn alloy deposited at current density of1.3 A/dm2 and duty cycle of 0.5 at indicated frequencies or DC.
Table 1Tafel data of different Cu–Sn layers deposited on steel substrate in 0.1M NaCl.
behavior by potentiodynamic polarization measurements in 0.1M NaClaqueous solution. For better clarity, polarization curve of deposits withpulse current was represented with that of 0.1 Hz, and detailed datahas been shown in the Tables. Bronze layers are expected to increasethe corrosion potential of substrate and lower the corrosion currents.
1E-8 1E-7 1E-6 1E-5 1E-4 1E-3
-0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Current Density (A/cm2)
Po
ten
tial
(V
)
steel substrate DC 0.1 Hz
A
1E-9 1E-8 1E-7 1E-6 1E-5 1E-4 1E-3
-0.6
-0.3
0.0
0.3
0.6
0.9
Current Density (A/cm2)
Po
ten
tial
(V
)
Cu underlayer DC 0.1 Hz
B
Fig. 7. Polarization curves of bronze films deposited on (A) steel plate, and (B) Cuunderlayer (5 μm) in 0.1M NaCl solution B.
Please cite this article as: C. Zanella, et al., Surf. Coat. Technol. (2013), http
However, as shown in Fig. 7(A) and Table 1, the DC-deposited onesshow poor improvement indicating the defective films caused by porousstructure. On the contrary, the PC indeed improves the layers showingmore compact surface compared to DC. In addition, passive behaviorcan be hardly observed on the film deposited on steel plate, due torelatively high porosity.
However, in the practical application, copper underlayer isdeposited on the substrate in order to make the substrate bright andmore compatiblewith thefinal deposition. Fig. 7(B) and Table 2 showedthe corrosion behavior of bronze deposited on copper layer. As shown,copper underlayer shows significantly lower corrosion potential andreduced corrosion rate compared to steel substrate proving it is aprotective coating free of pores. Passive behavior can be observed onthe bronze deposits on copper underlayer suggesting that copper–tinco-deposits exhibits better protective function than pure copper.The addition of tin to the co-deposition can decrease the corrosionpotential of copper. The deposition of bronze coating on copper reducesthe cathodic Tafel slope and shifts corrosion current density towardslower values, thereby resulting in a reduced corrosion rate. By com-paring the deposits produced with direct current and pulse current,the film obtained by pulse plating can exhibit lower corrosion ratewithin the passive region. It can be concluded that corrosion potentialis proportional to copper content in bronze layers from the fact thatPC-deposited layers have higher corrosion potentials.
4. Conclusions
This work studied the deposition of Cu–Sn coatings from amethanesulfonic based bath for decorative purposes and theoptimization of process parameter in order to obtain layerswith expected composition and high stability. The change ofanode from copper to graphite has been beneficial to both lifetimeof the bath and appearance of deposits. The employment of pulsedcurrent during electrodeposition process can reduce the grain sizeand introduce new crystal structures. In term of corrosion protectiondirect deposition on steel leads to the formation of defects andaccordingly non protective coatings, while a copper underlayer cansolve this problem leading to a protective coating.
Table 2Tafel data of different Cu–Sn layers deposited on copper underlayer in 0.1M NaCl.
6 C. Zanella et al. / Surface & Coatings Technology xxx (2013) xxx–xxx
References
[1] A. Hovestad, R.A. Tacken, H.H.t. Mannetje, Phys. Stat. Solidi C 5 (2008) 3506.[2] J.S. Park, V. Shinde, J. Archaeol. Sci. 40 (2013) 3811.[3] K. Weitershaus, European Patent Office, Buckley, Guy Julian, European, 2013.[4] T. Nickchi, M. Ghorbani, Surf. Coat. Technol. 203 (2009) 3037.[5] A. Survila, Z. Mockus, S. Kanapeckaitė, V. Jasulaitienė, R. Juškėnas, J. Electroanal.
Chem. 647 (2010) 123.[6] D. Padhi, S. Gandikota, H.B. Nguyen, C. McGuirk, S. Ramanathan, J. Yahalom, G. Dixit,
Electrochim. Acta 48 (2003) 935.[7] R. Juškėnas, Z. Mockus, S. Kanapeckaitė, G. Stalnionis, A. Survila, Electrochim. Acta 52
(2006) 928.[8] A. Correia, M. Facanha, P. Delimaneto, Surf. Coat. Technol. 201 (2007) 7216.[9] G.Meng, F. Sun, S.Wang, Y. Shao, T. Zhang, F.Wang, Electrochim. Acta 55 (2010) 2238.
[10] C.T.J. Low, F. Walsh, Surf. Coat. Technol. 202 (2008) 1339.[11] N.M. Martyak, R. Seefeldt, Electrochim. Acta 49 (2004) 4303.[12] P. Perret, Z. Khani, T. Brousse, D. Bélanger, D. Guay, Electrochim. Acta 56 (2011) 8122.[13] C.T.J. Low, F.C. Walsh, Surf. Coat. Technol. 202 (2008) 3050.[14] C.T.J. Low, F.C. Walsh, Trans. Inst. Met. Finish. 86 (2008) 315.[15] I. Sirés, C.T.J. Low, C. Ponce-de-León, F.C.Walsh, Electrochem. Commun. 12 (2010) 70.[16] A. Oury, A. Kirchev, Y. Bultel, E. Chainet, Electrochim. Acta 71 (2012) 140.[17] C.T.J. Low, D. Pletcher, F.C. Walsh, Electrochem. Commun. 11 (2009) 1301.[18] F.J. Recio, P. Herrasti, I. Sirés, A.N. Kulak, D.V. Bavykin, C. Ponce-de-León, F.C. Walsh,
J. Appl. Electrochem. 30 (2000) 987.[24] U. Manz, S. Berger, K. Bronder, K. Leyendecker, B. Weyhmüller, Surf. Tech. White
Papers 77 (2013) 1.[25] K.M. Yin, S.L. Jan, C.C. Lee, Surf. Coat. Technol. 88 (1996) 219.[26] D. Landolt, A. Marlot, Surf. Coat. Technol. 169–170 (2003) 8.[27] K. Subramanian, V.M. Periasamy, M. Pushpavanam, K. Ramasamy, J. Electroanal.
Chem. 648 (2010) 176.
Please cite this article as: C. Zanella, et al., Surf. Coat. Technol. (2013), http
[28] M.B. Turnhouse, J. Paleontol. 75 (1987) 11.[29] A. Survila, Z. Mockus, S. Kanapeckaite, V. Jasulaitiene, R. Juskenas, Electrochim. Acta
52 (2007) 3067.[30] K. Subramanian, V.M. Periasamy, M. Pushpavanam, K. Ramasamy, J. Electroanal.
Chem. 636 (2009) 30.[31] R. Balaji, M. Pushpavanam, K.Y. Kumar, K. Subramanian, Surf. Coat. Technol. 201
(2006) 3205.[32] E. Huttunen-Saarivirta, T. Tiainen, J. Mater. Process. Technol. 170 (2005) 211.[33] Z. Zainal, A. Kassim, M. Hussein, C. Ching, Mater. Lett. 58 (2004) 2199.[34] T.N. Vorobyova, V.P. Bobrovskaya, V.V. Sviridov, Met. Fin. 11 (1997) 14.[35] W.J. Boettinger, C.E. Johnson, L.A. Bendersky, K.W. Moon, M.E. Williams, G.R.
Stafford, Acta Mater. 53 (2005) 5033.[36] E. Huttunensaarivirta, T. Tiainen, J. Mater. Process. Technol. 260 (2005) 476.[37] Z. Mockus, S. Kanapeckaité, V. Jasulaitiené, A. Survila, Prot. Met. 42 (2006) 485.[38] B. Palys, M.I. Borzenko, G.A. Tsirlina, K. Jackowska, E.V. Timofeeva, O.A. Petrii,
Electrochim. Acta 50 (2005) 1693.[39] K. Subramanian, V.M. Periasamy,M. Pushpavanam,M. Ramasamy, J. Electroanal. Chem.
Soc. China 16 (2006) 223.[41] S.D. Beattie, J.R. Dahn, J. Electrochem. Soc. 150 (2003) C457.[42] C. Doyle, N. Brown, M. Bardizeh, Met. Fin. 100 (2002) 10.[43] M. Datta, R.V. Shenoy, C. Johnes, P.C. Andricacos, J. Horkans, J.O. Dukovic, L.T.
Romankiw, J. Roeder, H. Deligianni, H. Nye, B. Agarwala, H.M. Tong, P. Totta, J.Electrochem. Soc. 142 (1995) 3779.
[44] S.D. Beattie, J.R. Dahn, J. Electrochem. Soc. 150 (2003) A894.[45] D. Landolt, A. Marlot, Surf. Coat. Technol. 8 (2003) 169.[46] P.E. Bradley, D. Landolt, Electrochim. Acta 45 (1999) 1077.[47] P.E. Bradley, D. Landolt, Electrochim. Acta 42 (1997) 993.[48] S. Roy, D. Landolt, J. Appl. Electrochem. 27 (1997) 299.[49] S. Roy, D. Landolt, J. Electrochem. Soc. 142 (1995) 3021.[50] M. Hansen, Constitution of Binary Alloys, first ed. McGraw Hill, New York, 1958.[51] N. Tamura, R. Ohshita, M. Fujimoto, S. Fujitani, M. Kamino, I. Yonezu, J. Power Sources
107 (2002) 48.[52] C.H. Wu, F.Y. Hung, T.S. Lui, L.H. Chen, Mater. Trans. 50 (2009) 381.[53] N.F. Kennon, D.P. Dunne, L. Middleton, Metall. Trans. A 13A (1982) 551.
