Introduction

It is well recognized that composite coatings of a single metal and alloys protect steel against corrosion, by improving its mechanical, physical and/or electrochemical properties. As such, Zn-based alloys have attracted a lot of interest in the last decade ^1-9. Zn-Ni alloy has received more attention, due to its high degree of corrosion resistance, mechanical characteristics and thermal stability, compared to Zn only and other Zn alloys coatings ^10-17. Zn-Ni alloy is more toxic than Zn-Cd alloy ^18),(19, although it has been widely used in automotive, aeronautic, marine, building and electronics industries ^20-22.

In order to improve coating quality, using additives is generally recommended. Generally, a small amount of additives affects the electroplating reaction kinetics, mainly by adsorption or complexation ^23,24. Therefore, metals and metal oxides electroplating is often performed in baths containing organic additives ^25-29.

The present study dealt with Zn-Ni coating electroplating from a sulfate bath. The main goal was to find the best bath composition that would produce coatings with high quality and hardness. For this purpose, a mixture design method was examined, which provided maximum information on its constituents, their individual influences and possible interactions. This assessment allowed reducing the experiments number, facilitating the study planning and procedures ^30,31. The desired response depended on the bath components concentration, which were Na₃C6H₅O₇ ^-, C₇H₅NO₃S and 2-butyne-1,4-diol.

Experimental

Material

Low CS, conforming to SAE 1010, ASTM A-366 and QQS-698 standards, was used in this study. CS substrate chemical composition is given in Table 1.

Electroplating

Zn-Ni alloy coatings electrodeposition was carried out in a sulfate acid bath, of which compositions and operating conditions are shown in Table 2. The solution pH was from 4.3 to 4.5. Each experiment was performed in a fresh solution, to avoid problems such as metal ions depletion from the electrolyte.

Zn-Ni deposits electrodeposition was performed in a conventional electrochemical cell with two electrodes (cathode and anode), to which constant ddp was applied, using a generator. The procedure was carried out by fixing the following parameters: T of 30±1 ºC, stirring speed around 300 rpm, electrolyte volume of 100 mL, I of 1 A/dm² and thickness of 15 µm. However, Na₃C6H₅O₇ ^-, C₇H₅NO₃S and 2- butyne-1,4-diol concentrations were changed, according to a matrix given by the mixture design (MINITAB 18) (Table 3).

Results and discussion

Simplex plot

The simplex plot shows points in the plane space, which are: three at the triangle vertices, for 0.3 M Na₃C₆H₅O₇ ^-, 0.3 g/L C₇H₅NO₃S and 0.3 g/L 2butyne1,4diol pure solutions; three on the triangle sides for the mixtures; and one in the triangle center consisting of the three components in equal proportions. These three complete mixtures had all components, but in different proportions (Fig. 1).

Henry's line of residual values

Generally, Henry's line is useful for checking the model normality. It is seen that the points tend to form a line (Fig. 2).

Cox plot

The Cox plot (Fig. 3) presents the evolution of the estimated response by varying the proportion of a single component from a reference mixture, while keeping constant the other components ratio ³⁰. It consists in representing micro-hardness variations along the Cox axis.

Micro-hardness decreased with increased C₇H₅NO₃S concentrations in the mixture, up to the reference line. On the other hand, beyond this line, an increase in micro-hardness was observed. It was also found that 2-butyne1,4diol presence did not have a remarkable effect on micro-hardness. However, Na₃C6H₅O₇ ^- had a positive effect on the obtained coatings micro-hardness.

Mathematical model

The mathematical model applied to the response (Zn-Ni coating micro-hardness) is a quadratic model for three components, with a total of six coefficients for a single response, according to eq. 1.

Graphics contour and surface response

Simplex contour and response surfaces are shown in Fig. 4. On its left side, Na₃C6H₅O₇ ^-, C₇H₅NO₃S, and 2butyne1,4diol contour area and micro-hardness are shown. The best responses are represented by the red contour (>210 HV), which is located at the high level for 0.3 g/L Na₃C6H₅O₇ ^-, and at the low level for 0.1 g/L C₇H₅NO₃S and 2butyne1,4diol each. The response surface represented in a polyhedron (on Fig. 4 right) is a concave triangle shape.

It is seen that microhardness increased with lower C₇H₅NO₃S and 2butyne1,4diol concentrations, and with higher Na₃C6H₅O₇ ^- content.

Optimization

There are many optimization methods. Most of them have been created to deal with the mathematical problem of finding the multi-variable extreme, whether non-linear functions are subject to constraints or not. After several optimizations, the best results are shown in Table 4.

Theoretical and experimental responses

The following figure represents the plot of experimental response as a function of theoretical response. From Fig. 5, we can see that the experimental and estimated responses are on the regression line, with a very good value of 0.9846.

From Table 2, it was confirmed that there was not a significant difference between experimental and theoretical values, which is represented by the relative and absolute incertitudes calculated according to eqs. 2 and 3, respectively.

Characterization

SEM analysis

SEM images show that the bath composition influenced the coating quality. Indeed, with Na₃C6H₅O₇ ^- only (Fig. 6a), a CS non-homogeneous oxidized surface was obtained.

In contrast, the CS surface morphology was improved by C₇H₅NO₃S addition (Fig. 6b), which turned out to be more homogeneous and less oxidized compared than that of Fig. 6a.

By adding Na₃C6H₅O₇ ^- and 2butyne1,4diol, the obtained deposit formed cracks (Fig. 6c). It is also seen that the grains have a large size and cauliflower shape. The same results were found by ^32,33. It was found that the best surface in terms of quality and brightness was obtained in the presence of all additives (Fig. 6d).

EDS analysis

EDS analysis results showed that Zn content of Fig. 7a was lower than that of Fig. 7d, contrary to O. Ni concentration in the deposit varied between 6 and 21%. From literature, it was reported that Zn-Ni coatings ranging from 10 to 15 wt% Ni have better corrosion resistance ³⁴.

Discussion

These findings are consistent with other previous results, and the change in Zn-Ni alloy coatings quality and micro-hardness is explained by the additive molecules adsorption onto the cathode surface.

It was found, by ¹¹, that Na₃C₆H₅O₇ ^- effectively stabilizes Zn-Ni alloy plating baths. According to ³⁵, it was found that Na₃C₆H₅O₇ ^- addition influenced the grains morphology and size.

C₇H₅NO₃S and 2-butyne 1-4 diol organic additives used in this study influenced the obtained coating roughness, by preventing the formation and growth of nodular structures. This finding was also reported by ^36-38.

It was found that O can disrupt the process, due to the formation of Zn(OH)₂ precipitate, which inhibits Ni deposition. This result agrees with that of ³⁹, who found that O incorporation into the deposits is probably due to the formation of Zn(OH)₂ layer.

Conclusion

This study was focused on the protection of low-CS against corrosion, using electrodeposition as a solution. Zn-Ni coatings were produced from a sulfate bath with 0.1 to 0.3 g/L of both C₇H₅NO₃S and 2-butyne1-4diol (0.1 to 0.3 g/L) as additives, and 0.1 to 0.3 M Na₃C6H₅O₇ ^- as complexing agent.

The electrodeposition performance of Zn-Ni coating can be improved with the three components blend, by applying the mixture design method. This method employed a quadratic model that considered microhardness as a response to various constituents. A set of experiences and proposed formulations to determine the optimum conditions was herein built, leading to a gray, smooth, uniform, hard and bright coating.

The mathematical model was validated by comparing theoretical and experimental microhardness results. Indeed, the fit was almost perfect, as the linear correlation constant was 0.9846.

Optimization gave a mixture of three components, of which concentrations were 0.3 M Na₃C6H₅O₇ ^- and 0.1 g/L each C₇H₅NO₃S and 2-butyne1,4diol. This corresponded to a theoretical microhardness of 215.22 HV. Indeed, to confirm and validate the theoretically obtained results, a check for an additional experiment under optimal conditions was performed, and the experimentally obtained microhardness value was 221.60 HV.

From SEM analysis, it was found that the presence of additives played a significant role on the deposit surface morphology and homogeneity.

Acknowledgments

We thank Prof. Rehamnia Rabah Director of the LNCTS Research Laboratory and Dr. Amirat Samia, Head of Chemistry Department, Badji Mokhtar University of Annaba, for providing us with the necessary material resources. The authors thank also the ENSMM-Annaba for the assistance provided in this research, through SEM-EDS analyzes.

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Authors’ contributions

Hawa Bendebane: designed the research work; performed the experimental work; carried out data analysis and interpretation; acted as corresponding author. Salima Bendebane: contributed to the interpretation of results; made the paper English corrections; improved the quality of the paper. Samia Amirat: supervised the work at certain parts. Rabah Rehamnia: provided methodology and investigation. All authors read and reviewed the manuscript.