ANOVA designs

Temperature, current density and cobalt concentration effects on electrodeposited anticorrosive cobalt-tungsten alloys using factorial experiment design and ANOVA techniques M. B. Porto, D. G. Portela and A. F. de Almeida Neto

Department of Process and Product Design, Faculty of Chemical Engineering, State University of Campinas, Campinas, Brazil

ABSTRACT Corrosion processes can compromise the durability of metallic materials. One way to minimise the effects of corrosion is to coat them with corrosion-resistant alloys. In this study, Co–W alloys were produced by electrodeposition and their anticorrosive performance was evaluated at different levels of current density, Co bath concentration and bath temperature, using factorial experiment design and ANOVA techniques. This evaluation consisted of measuring corrosion current and using electrochemical impedance spectroscopy. Adherent and corrosion resistant alloys were obtained at the lowest temperature of experimental design (25°C) and current density (10 mA cm−2) values. From the corrosion tests, X-ray Diffraction and Scanning Electron Microscopy analysis it was possible to conclude that the increase in the tungsten percentage in the composition of the alloy generated an increase in corrosion resistance. Although the alloys’ chemical composition has influence on the corrosion resistance, their physical aspect, such as the presence of cracks, are more significant for the resistance to corrosion.

ARTICLE HISTORY Received 15 April 2019 Accepted 2 August 2019

KEYWORDS Anticorrosive alloys; coating materials; Co–W alloy; electrodeposition; corrosion current

Introduction

Corrosion can affect several types of materials; this phenom- enon appears frequently and under diverse circumstances. The corrosion process can produce undesirable and detrimen- tal changes in the structures of materials. The financial losses generated by corrosion, especially the need for replacement, are a factor of great importance to be considered. For this reason, the development of advanced materials to mitigate or halt corrosion is a necessity.

The use of protective coatings is a way to prevent metallic corrosion and to considerably improve the physical–chemical properties of surfaces. A simple way of coating metal surfaces is through electrodeposition, a process whose procedural par- ameters can be individually controlled, such as the compo- sition of the electrolytic solution, deposition temperature, applied current density, and mass to be deposited. The exper- imental control of these parameters makes studying the phys- icochemical characteristics of obtained metallic coatings possible.1

The use of cobalt is widespread in magnetic materials, storage devices, microelectromechanical systems, and mag- netic recorders; however, these materials are prone to wear and degradation under normal usage conditions and have a limited useful life.2,3 To avoid this problem, cobalt and tung- sten alloys can be used to substitute cobalt-coated materials in many of these applications, with the advantages of being harder and more resistant to corrosion.

Tungsten has many useful properties, and there is cur- rently extensive research to discover additional applications of its alloys, especially as an anticorrosive coating. This metal is non-toxic to aquatic environments, non-carcinogenic, and has a high melting point (3410°C), which, despite making it impossible to thermally deposit it on the surface of any

other metal, it is able to stay solid even at high temperatures. Moreover, tungsten has excellent mechanical properties such as the highest tensile strength of all metals (410 kg mm−2), and is very resistant to corrosion. Tungsten readily forms alloys with metals from the iron group, such as nickel and cobalt, through induced codeposition, a situation in which a metal is codeposited in the presence of another, composing a metallic alloy.4

In general, cobalt and tungsten alloys have many advan- tages over chromium alloys and even over nickel and tung- sten alloys. This is due to cobalt having a hexagonal crystal structure, making it more wear-resistant than nickel, which comparatively has a cubic unit cell with a centred face.5

Electroplating baths containing chromium ions are known to be environmentally harmful, mutagenic and carcinogenic agents.6 Research for materials with good anti-corrosion prop- erties that generate more environmentally acceptable indus- trial tailings, such as metallic cobalt-based coatings, are important to replace chromium.7

Different microstructures and properties of alloys are obtained depending on their electrodeposition conditions. As such, it is important to investigate the process to deter- mine more efficient parameters and amplify specific, desired characteristics. In this context, this work aimed to evaluate the corrosion resistance of various Co–W alloys obtained by electrodeposition, containing different chemical compo- sitions of cobalt and tungsten.

Materials and methods

The electrolytic baths were composed of cobalt sulphate (CoSO4) with concentrations ranging from 0.1, 0.2 and 0.3 M; sodium tungstate (Na2WO4), the tungsten source, with a

© 2019 Institute of Materials Finishing Published by Taylor & Francis on behalf of the Institute

CONTACT M. B. Porto maribporto@hotmail.com Department of Process and Product Design, Faculty of Chemical Engineering, State University of Campinas, 500, Albert Einstein Av., Campinas, SP, 13083-852, Brazil

TRANSACTIONS OF THE IMF 2019, VOL. 97, NO. 6, 305–311 https://doi.org/10.1080/00202967.2019.1668122

concentration of 0.3 M; borax (Na2B4O7) 3.75 . 10 −2 M to give an

amorphous alloy; sodium dodecyl sulphate (1-dodecylsul- phate-Na) 1.04 . 10−4 M, to release hydrogen with higher speed during the electrodeposition avoiding the formation of bubbles in the adhered alloy; ammonium sulphate ((NH4)2SO4) 1.287 . 10−1 M, in order to obtain greater stability to the bath and ammonium citrate ((NH4)2C6H6O7) 0.3 M as the cobalt com- plexing agent. The electrolytic baths had a pH of approximately 6, as determined by metal speciation diagrams.8,9

According to the speciation diagrams, it is possible to observe that Co2+ chemical species are complexed by ammonium citrate at pHs ranging from 6 to 8.9 The electro- lytic bath’s pH affects the electrodeposition process, as well as determining which of the different metal species may exist in the aqueous solutions used during electrolysis. The complexes of electrolytic solutions depend on the conditions of pH, concentration, and ionic strength to establish them. In this context, the present work was based on chemical metal speciation diagrams of determining an optimal pH for electro- lytic baths to obtain Co–W alloys.

The electrodepositions were performed on the surface of a squared copper plate with 8 cm2 of area using a potentiostat/ galvanostat and a hollow cylindrical platinum mesh as depo- sition counter electrode. During the assays, the copper elec- trode was maintained under rotation at 30 rev min−1 and the duration of each test was 3600 s. The deposition efficiency (ε) was calculated according to Equation (1):

1 = m · F i · t

∑ nj · wj Mj

(1)

The coating mass (m) is in grams, deposition time (t) in seconds and the total current used (i) in amperes. In Equation (1), wj refers to the mass fraction of the metal j in the alloy, obtained by EDX, nj is the number of electrons transferred for each atom of the metal j, Mj is the atomic mass of the metal j in g mol

−1

and F is the Faraday constant, which is 96,485 C mol−1. The Co–W electrodeposition process was performed fol-

lowing a factorial design 23, with 3 experiments at central point.8–10 Repetitions of the tests were performed only at central point, making at least 3 repetitions, so the calculation of error is at least 2 degrees of freedom. Thus, the standard deviation calculated for centre point experiments can be con- sidered equal for all the experimental planning, which is a reasonable consideration in practice. The influence of cobalt initial concentration in the bath (CCo), electrical current density (i) and bath temperature (T) on the deposition efficiency (ε) was analysed, as shown in Table 1.

The experiments were performed in a random order with the aim of avoiding systematic error. The regression data analysis was performed using Statistica 10.0 software.

For the corrosion tests, a VersaSTAT3 potentiostat and an electrochemical cell composed of three electrodes in an aqueous solution of sodium chloride (NaCl) 0.1 M were used at room temperature. The reference electrode was |Ag/ AgCl|. The counter electrode was platinum mesh, and the

working electrode consisted of the copper plates coated with the metal alloys. Corrosion current values were calcu- lated from potentiodynamic polarisation curves and electro- chemical processes were analysed by the Nyquist diagrams obtained from the electrochemical impedance.

The potentiodynamic polarisation curves were obtained in the range of −0.25 to 0.25 V, with a scanning speed of 2 mV s−1, recording the correlation between the applied potential at the electrode and the current measured at the potentiostat. The values of Tafel slopes and the coefficient of Stern-Geary were obtained from the Tafel curve, an impor- tant tool to investigate corrosive processes.11 These par- ameters are related by the Butler–Volmer equation. The following Equation (2) presents the anodic Tafel slopes:

ha = aa + balogia (2) Wherein aa = (−2.3RT / βnF) log icorr and ba = 2.3RT / βnF.

And Equation (3) presents the cathodic Tafel slopes:

hc = ac + bclogic (3) where ac = (−2.3RT / (1 – a)nF) log icorr and bc = 2.3RT / (1 – β)nF.

The parameters a and b are the Tafel constants; R is the ideal gas constant; β is the load transfer coefficient; n is the oxidation number of the electroactive species; F is the Faraday constant; i is the measured current density; icorr is the corrosion current density; and η is the relative potential compared to the corrosion potential (E – Ecorr).

Extrapolating the Tafel lines, the corrosion current density, icorr, is obtained. It is related to the polarisation resistance, Rp, and the Stern-Geary coefficient B, a relation shown by Equation (4):12

icorr = 106B Rp

(4)

The unit of Rp is ohm.cm 2, icorr is in μA cm

−2 and B is in V. The Stern-Geary coefficient is related to the anodic and cathodic Tafel slopes as shown in Equation (5):9

B = babc 2.3(ba + bc)

(5)

When the values of the corrosion potentials were obtained, electrochemical impedance spectroscopy (EIS) measurements were taken. EIS is an important tool in the investigation of electrochemical systems, and is useful to study corrosion, evolution of protective layers, batteries, electrodeposition, and semiconductors.13

This method allows for the identification and determi- nation of the parameters of a model based on the frequency response of the studied electrochemical system. The tech- nique provides a complete and detailed view of the electrical characteristics of the electrode/electrolytic solution interface in an alternating current (AC) circuit.

The measure of the ability of a circuit to withstand the elec- tric current is its impedance (Z), which can be obtained analo- gously to Ohm’s law by Equation (6):

E(t) = Z.I(t) (6) where the impedance is also measured in ohms (Ω).

The analysis of the results is performed using a graphical representation of the data on a Nyquist plot. This diagram is a complex (real-imaginary) plane in cartesian coordinates,

Table 1. Levels and coding of the studied factors in the 23complete factorial design.

Variables Level (−1) Level (0) Level (+1) CCo (M) 0.1 0.2 0.3 i /mA cm−2 10 30 50 T /°C 25 42.5 60

306 M. B. PORTO ET AL.

where the x-axis represents the real part (resistive terms) and the y-axis represents the imaginary part (capacitive or inductive).13 The initial potential of corrosion, observed in the polarisation, was applied with a perturbation of amplitude equal to 10 mV in the frequency range of 10 mHz to 10,000 Hz.

Deposited Co–W alloy samples were characterised by X-ray Diffraction (XRD) and Scanning Electron Microscopy (SEM) with X-ray Energy Dispersion Spectroscopy (EDS).

XRD analysis was used to discern the crystallinity of the alloys. The apparatus used was produced by Philips, model X’PERT, with copper Ka radiation, wavelength 1.52 Å, voltage 40 kV, current 40 mA, step size 0.02, and time per step 1 s.

By using EDS, it was possible to perform chemical mapping of the alloy surfaces by qualitative and semiquantitative microanalyses of the chemical elements that make up the metal alloys, also providing insight into the homogeneity of the sample. The micrographs were obtained using a scanning electron microscope produced by LEO Electron Microscopy, Inc., model LEO 440i. Magnifications of 50, 500, 1500, and 5000 times were performed.

Results and discussion

Efficiency of the Co–W electrodeposition

Table 2 shows the 23 complete factorial design used in the electrodeposition process, listing the deposition efficiencies as well as the alloy percentage of cobalt and tungsten obtained by EDX.

Analysing Table 2, it is possible to observe a higher per- centage of cobalt in deposits compared to tungsten. Exper- iment 2 presented the highest deposition efficiency, which was 86.61% at room temperature and also had the lowest current density. A similar result was found in the study carried out by Rafailovic et al.14 in which lower values of current density increased the current efficiency, indicating that lower energy levels in the electrodeposition process can ensure greater process performance. Rafailović et al.14

also observed enhancement of Co deposit at higher current density, however, the authors did not consider vari- ation of Co concentration on the bath, this being one of the most influential parameters for deposition efficiency in Co– W alloys.

Analysis of variance (ANOVA) was performed with a confidence level of 90% for p < 0.10. The ANOVA results demonstrate that the model is statistically significant and pre- dictive for p < 0.10. The model is represented by Equation (7).

1 = 72.9 + 25.18.CCo − 11.15.i (7) where CCo is the cobalt concentration in M.

The variance and regression analyses demonstrated the statistical significance of the model, justifying the use of a linear model for the statistical analysis. The regression analysis of the experimental data showed that the concen- tration of cobalt sulphate followed by the current density were the most influential variables in the deposition efficiency.

The effect of cobalt sulphate concentration and current density on deposition efficiency were evaluated on the response surface, with temperature set in the lower level. Figure 1 shows the surface response.

It was observed that with increasing cobalt sulphate con- centration and lower values of current density, deposition efficiency was higher.

Corrosion resistance of the Co–W alloys

As shown in Equation (4), lower values of corrosion current represent higher polarisation resistance values. Among the experiments that showed deposition efficiency values greater than 80%, the alloy obtained in Experiment 2 had the lowest corrosion current density (1.96 μA cm−2), followed by Experiment 9 (2.27 µA cm−2). In these two experiments, the cobalt concentration in the bath was 0.3 M. Experiment 2 was performed at ambient temperature with a current density of 10 mA cm−2, demonstrating that no large energy values were needed to produce a corrosion-resistant alloy. Table 3 presents the values of corrosion current density for each experiment from the 23

experimental design.

Table 2. Co–W alloy results with the variables applied and their respective levels.

Exp. CCo /M I /mA cm −2 T /°C mCo /% mW /% ε /%

1 0.1 (−1) 10 (−1) 25 (−1) 69.2 30.8 72.19 2 0.3 (+1) 10 (−1) 25 (−1) 81.3 18.7 86.61 3 0.1 (−1) 50 (+1) 25 (−1) 58.0 42.0 44.47 4 0.3 (+1) 50 (+1) 25 (−1) 82.2 17.8 82.20 5 0.1 (−1) 10 (−1) 60 (+1) 71.0 29.0 62.26 6 0.3 (+1) 10 (−1) 60 (+1) 79.0 21.0 82.15 7 0.1 (−1) 50 (+1) 60 (+1) 57.8 42.2 51.61 8 0.3 (+1) 50 (+1) 60 (+1) 78.3 21.7 80.32 9 0.2 (0) 30 (0) 42.5 (0) 71.6 28.4 75.01 10 0.2 (0) 30 (0) 42.5 (0) 70.0 30.0 83.16 11 0.2 (0) 30 (0) 42.5 (0) 70.0 30.0 82.00

Figure 1. Response surface for Co–W deposition efficiency as a function of current density and cobalt concentration at pH 6.

Table 3. Corrosion current density results for Co–W alloys.

Exp m /g E /mV Icorr /μA cm −2

1 0.0643 −603.304 10.24 2 0.0705 −493.242 1.96 3 0.1988 −645.924 7.00 4 0.3640 −568.159 10.31 5 0.0554 −572.101 6.14 6 0.0713 −510.548 9.53 7 0.2307 −615.012 2.40 8 0.3562 −514.833 2.84 9 0.2001 −577.083 2.27 10 0.2220 −608.073 2.90 11 0.2188 −684.847 2.73

TRANSACTIONS OF THE IMF 307

Impedance analysis

In order to characterise the electrochemical systems and the corrosion process of metallic alloys in an aqueous medium, the coatings were submitted to electrochemical impedance tests. Impedance tests were performed by immersing the samples in a 0.1 M NaCl solution. Electrochemical reactions are then allowed depending on the nature of the electrode- solution interface, thermodynamic and kinetic reactions on the electrode and mass transport effects. The corresponding Randles equivalent circuits were applied in this work for data adjustment. The simulation parameters obtained using the EIS Spectrum Analyzer software are listed in Table 4.

The Rf values resulting from EIS data fitting are well consist- ent with Icorr values obtained from Tafel extrapolation, showing the high corrosion resistance of experiment 2. For experiments 5 and 6, the high error fit value explains the difference between the results of both techniques. Low Re values also emphasise high electrolyte conductivity. The n values were obtained using the EIS Spectrum Analyzer software. The results are between 0.5 < n <1, indicating that the electrode surface is rough. All experiments presented constant phase element (CPE) because they are composed of the same elements (Co,W).

Figure 2(A) shows single process behaviour of experiments 3 and 5. In those experiments, Nyquist plots are associated with a circuit configuration that includes the double layer. The equivalent circuit in Figure 2(A) is composed by a con- stant phase element (CPE), which characterises an imperfect capacitor, electrolyte resistance (Re), and coating resistance (Rf), represented on the Nyquist diagram by a capacitive arc. Thus, comparing the two coatings, there is higher impedance and, consequently, an increase of corrosion resistance in the coating obtained in experiment 5.

Figure 2(B) shows a Nyquist diagram for Co–W alloys from experiments 2, 4, 6 and 9. These alloys showed a possible for- mation of two arcs. The equivalent circuit is represented by CPE, Re, Rf and also the Warburg open impedance element (Wo). The Warburg open impedance is related to the finite- length diffusion with reflective boundary.

The Nyquist diagram of experiments 7 and 8 in Figure 2(C) shows a diffusion control region indicated by a linear trend. Nyquist plots are associated with a circuit configuration including the double layer constant phase element (CPE), electrolyte resistance (Re), coating resistance (Rf), and also the Warburg open impedance element (Wo) as shown in Figure 2(C). If the coating resistance is greater than the load transfer resistance, there is no semicircle formation.15

Morphological analysis of Co–W alloys

Figure 3 shows the micrographs obtained in the experiments 5 and 7. These experiments were conducted using the same

concentration of sodium tungstate (0.1 M) and temperature (60°C) but with different value of current density. Experiment 5 was performed at 10 mA cm−2 and experiment 7 was per- formed with 50 mA cm−2.

Since the SEMs were analysed with the same magnification of 500 times, and presented similar appearance and uniform- ity, it is possible to make a comparison between experiments 5 and 7. The alloy from experiment 5 has 29.0% wt. of W and the corrosion current density value of 7 μA cm−2. Experiment 7 produced an alloy with tungsten content of 42.2% wt. and the lowest corrosion current density value of 2.4 μA cm−2, consequently resulting in a higher value of polarisation resist- ance than experiment 5. Owing to the W atomic radius being larger than the Co atomic radius, an increase in tungsten content in Co–W alloy from Experiment 7 resulted in more expanded grains than those observed in experiment 5. Figure 4 shows the micrographs obtained by SEM for the Co–W coatings of experiments 2 and 4. It is possible to

Table 4. Simulated impedance parameters.

Experiment Re

/ohm.cm² Rf

/ohm.cm² E* /%

n1 /a.u

2 53.24 6590.2 3.09 0.76 3 62.87 1386.0 2.83 0.81 4 55.46 886.26 2.23 0.78 5 53.27 2099.2 5.31 0.81 6 57.55 3047.8 5.46 0.83 7 62.87 4162.4 2.83 0.81 8 51.44 88626 1.31 0.73 9 69.00 6493.2 2.43 0.69

*error of Rf.

Figure 2. Nyquist diagrams for Co–W alloys and circuit equivalents.

308 M. B. PORTO ET AL.

observe the presence of cracks in the coating surface of exper- iment 4. These cracks can be associated with the high value of current density, that is the parameter which differentiates these two experiments. Another possibility is that during the formation of homogeneous substitutional structure some stress may have occurred in deposits with more tung- sten, as it has higher atomic radius, causing the cracks. These deformations can compromise the corrosion resistance performance of the alloy since they produce discontinuities in the surface and allow the electrolyte to permeate the alloy. The coating alloy obtained in Experiment 2 has 81.3% wt of Co mass and 18.7% wt of W. The coating from experiment 4 is composed of 82.2% wt cobalt and 17.8% wt of tungsten. Despite the similarity in composition, there is a significant difference between the corrosion current density values: 1.96 and 10.41 μA cm−2, respectively for experiments 2 and 4. As the corrosion current is inversely proportional to the resist- ance to polarisation, the coating from experiment 4 is less cor- rosion resistant than the one obtained in experiment 2.

From the corrosion tests and the morphological analysis of Co–W alloys, it was possible to conclude that the increase in the percentage of tungsten in the composition of the alloy results in an increase in corrosion resistance. Although the chemical composition of the alloys has influence on the resist- ance to corrosion, their physical aspects, such as the presence

of cracks, are more significant for the resistance to corrosion. It was also possible to observe that the increase in current density can increase the number of cracks formed in the alloy deposit.16

Metallic alloy crystallinity

The crystallinity of the Co–W alloys was obtained by XRD. Experiments 2 and 6 produced crystalline samples with well-defined peaks, as shown in Figure 5; deposits from these experiments have compositions and similar mor- phologies with 81.3% wt. Co 18.7% wt. W and 79.0% wt. Co 21% wt. W respectively. Although the temperature did not influence the current efficiency and the composition of the alloys, the little that the composition varied due to the temp- erature was just enough to result in different diffractograms and the corrosion current density, as presented by Equation (7) and Table 2.

The peaks found in these alloys are characteristic of cobalt, which indicates a complete dissolution of tungsten in cobalt and, therefore, the formation of a solid monophasic solution. Studies with Co–W alloys also observed the presence of this monophase.17,18

One can observe the Co peaks, with 41.73 and 47.51 (2θ degrees) corresponding respectively to crystallography (1 0 0)

Figure 3. SEM Micrograph of the Co–W alloy obtained in experiments 5 and 7.

Figure 4. SEM Micrograph of the Co–W alloy obtained in experiments 2, 4 and 6.

TRANSACTIONS OF THE IMF 309

and (1 0 1) and the peaks of W, with 44.6 and 64.9 (2θ degrees) respectively corresponding to (2 0 0) and (2 2 0) crystallogra- phy.19 This confirms that copper substrate was coated by the Co–W alloys and the deposits are crystalline. The alloys obtained were of the homogeneous substitutional type, thus tungsten was capable of replacing cobalt in the crystalline lattice.

For coatings obtained in experiments 1 and 3, amorphous structures are formed by random atomic arrangements and without symmetry or long-range order. A class of solid materials which has an amorphous structure stands out, since they have widely varying characteristics. Components are bonded together by metallic bonds and therefore have high electrical and thermal conductivity as well as a high degree of ductility. These materials also generally exhibit certain qualities such as being easily magnetisable, having high hardness, toughness, and corrosion resistance, and having low thermal expansion. When constituted of appropri- ate elements, these glassy metals have high resistance to cor- rosion.20 Figure 6 shows experiments 1 and 3.

Experiments 1 and 3 differ in their current density values during deposition. The alloys obtained in experiments 1 and 3 exhibited characteristics of amorphous materials, since they have a broad peak of only around 2θ = 45°. Co–W alloys can form thermodynamically stable intermetallic com- pounds, such as Co3W, which are generally amorphous and hard.21–23

Some studies indicate that the amorphous structure is directly affected by the cobalt concentration, which can

significantly influence the formation of either an amorphous or crystalline alloy.9, 24

In the present study, by the correlation of XRD results and alloy compositions, it seems that increasing the concentration of cobalt in the bath results in a structural transition from amorphous to crystalline alloys. Alloys from experiments 1 and 3 were obtained with lower cobalt concentrations and resulted in amorphous alloys. Alloys 2 and 6 were obtained with higher concentrations of cobalt and resulted in crystal- line alloys. As previously mentioned, an amorphous structure is expected to have useful properties, making such coatings desirable for various industrial applications.25

Conclusions

Within the proposed ranges of variables, all alloys showed a good appearance and presented different electrochemical behaviours between experiments. The highest deposition efficiency and lowest corrosion current density values found were 86.61% wt. and 1.96 µA cm−2, respectively. These optimum values were both obtained in experiment 2, using concentrations of 0.3 M cobalt sulphate, a temperature of 25°C, and a current density of 10 mA cm−2. The study also confirmed previous studies which suggested that low energy levels during the electrodeposition process can guar- antee higher performance with the added benefits of redu- cing the consumption of thermal and electrical energy. Also in this study, an increased percentage of tungsten in the alloy was found to result in a structural transition from a crys- talline to an amorphous alloy. The micrographs showed the presence of nodules with a spherical morphology that increased proportionally with tungsten content in Co–W alloys, also resulting in an increasing grain size. The analysis, based on statistical parameters, showed that the effects of temperature and current density can significantly influence corrosion resistance. Although the alloy deposits’ chemical composition has influence on the corrosion resistance, their physical aspect, such as the presence of cracks, is more signifi- cant for the resistance to corrosion.

Acknowledgements

The authors would like to thank FAEPEX/UNICAMP and FAPESP n° 2013/ 25212-0 for financial support.

Disclosure statement

No potential conflict of interest was reported by the authors.

References

1. A. Brenner, D. E. Couch and E. K. Williams: J. Res. Nat. Bur. Stand., 1950, 44(1), 109.

2. A. M. P. Sakita, E. C. Passamani, H. Kumar, D. R. Cornejo, C. S. Fugivara, R. D. Noce and A. V. Benedetti: Mater. Chem. Phys., 2013, 141(1), 576– 581.

3. S. Mary Delphine, M. Jayachandran and C. Sanjeeviraja, Mater. Chem. Phys., 2003, 81(1), 78–83.

4. F. H. Su, C. S. Liu and P. Huang, Appl. Surf. Sci., 2012, 258(17), 6550– 6557. doi:10.1016/j.apsusc.2012.03.075

5. L. Wang, Y. Gao, T. Xu and Q. Xue, Mater. Chem. Phys., 2006, 99(1), 96– 103.

6. A. D. Q. Ferreira, J. Zukerman-schpector and M. C.M. Alves. Química Nova, 2002, 147(1), 572–578.

7. E. W. Brooman, Met. Fin., 2004, 102(1), 42–54.

Figure 5. XRD for the Co–W alloys for experiments 2 and 6.

Figure 6. XRD for the Co–W alloys for experiments 1 and 3.

310 M. B. PORTO ET AL.

8. M. B. Porto, L. B. Alvim and A. F. de Almeida Neto, J. Clean Prod., 2017, 142, 3293–3299.

9. C. F. Baldessin, T. C. de Morais Nepel and A. F. de Almeida Neto, Can. J. Chem. Eng., 2018, 96(6), 1284–1289.

10. B. Bahadormanesh, M. Ghorbani and N. L. Kordkolaei: Appl. Surf. Sci., 2017, 404, 101–109.

11. S. Wollynec: ‘Técnicas eletroquímicas em corrosão,’ (ed. Paulo U de S). São Paulo; 2003.

12. S. W. Dean. J. ASTM Int., 2007, 4(9), 1–18. 13. A. Altube, H. Takenouti, L. Beaunier, M. Keddam, S. Joiret, S.

Borensztajn, F. Pillier and A.R. Pierna, Corros. Sci., 2003, 45(4), 685– 692.

14. L. D. Rafailović, D. M. Minić, H. P. Karnthaler, J. Wosik, T. Trišović and G. E. Nauer, J. Electrochem. Soc., 2010, 157(5), D295. doi:10.1149/1. 3336957

15. G. A. Ekama and G. V. R. Marais: ‘Zusatzliche biologische phosphore- limination beim belebungsverfahren – erfahrungen in sudafrika’ 1985, GWF, Wasser-Abwasser.

16. L. Y. Qin, J. S. Lian and Q. Jiang, Trans. Nonferrous Met. Soc. China, 2010, 20(1), 82–89. doi:10.1016/S1003-6326(09)60101-1

17. N. Tsyntsaru, A. Dikusar, H. Cesiulis, J. P. Celis, Z. Bobanova, S. Sidel’Nikova, S. Belevskii, Yu. Yapontseva, O. Bersirova and V. Kublanovskii, Powder Metall. Met. Ceram., 2009, 48(7–8), 419–428.

18. F. A. Lowenheim: ‘Modern electroplating’, 3rd edn; 1973, New York, Wiley.

19. Z. Ghaferi, S. Sharafi and M. E. Bahrololoom, Appl. Surf. Sci., 2015, 355, 766–773. Available from: http://linkinghub.elsevier.com/retrieve/pii/ S0169433215016505

20. R. A. Donald and P. P. Phulé Wright: ‘The science and engineering of materials’, 4th edn; 2003, (ed. B-T Learning). CA.

21. N. Tsyntsaru, Russ. J. Electrochem., 2016, 52(11), 1041–1047. Available from: http://link.springer.com/10.1134/S1023193516110136

22. N. Tsyntsaru, H. Cesiulis, E. Pellicer, J. P. Celis and J. Sort, Electrochim Acta, 2013, 104, 94–103. doi:10.1016/j.electacta.2013.04.022

23. N. Tsyntsaru, H. Cesiulis, A. Budreika, X. Ye, R. Juskenas and J. P. Celis, Surf. Coat. Technol., 2012, 206(19–20), 4262–4269.

24. M. H. Qiao, S. H. Xie, W. L. Dai and J. F. Deng, Catal. Lett., 2001, 71(3–4), 187–192.

25. H. Einati, V. Bogush, Y. Sverdlov, Y. Rosenberg and Y. Shacham- Diamand, Microelectron. Eng., 2005, 82(3–4 spec. iss.), 623–628.

TRANSACTIONS OF THE IMF 311

Copyright of Transactions of the Institute of Metal Finishing is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.