Abstract

Extensive efforts devoted in recent years to booming structural applications of lightweight magnesium alloys are usually undermined by their insufficient surface properties. Surface modification is therefore necessarily required in most cases for enhanced surface integrity of the alloys. Here, we report construction of aluminum-silica protective layers by cold spray on AZ31 magnesium alloys, and the effect of the silica additives on microstructure and mechanical properties of the coatings was examined. The ceramic particles were dispersed evenly in the coatings, and increased silica content gives rise to enhanced adhesion, antiwear performances, and microhardness of the coatings. The even distribution of silica in the coatings altered the wear regimes from adhesive to abrasive wear. The cold spray fabrication of the aluminum-silica protective coatings would facilitate structural applications of the magnesium alloys.

1. Introduction

Magnesium alloys are promising lightweight materials with extensive applications in automotive, aerospace, and electronics industries. However, challenges persist due to their relatively poor corrosion and wear resistance, which has been one of the major hurdles affecting development of the alloys [1]. One of the most effective solutions to prevent the corrosion is to coat an anticorrosion layer on magnesium alloys. Many surface techniques, such as thermal spraying, chemical vapor deposition, sol-gel, plating, anodizing oxidation, and microarc oxidation, have therefore been attempted to improve the anticorrosion performances of magnesium alloys [27], in which it was reported that a thin layer is effective in preventing penetration of corrosive substances. In many cases, corrosion is accompanied by wear for magnesium alloy components; it is therefore essential that the protective layers must simultaneously possess anticorrosion/antiwear performances.

Addition of ceramic materials to alloy-based coating can effectively improve its wear resistance, corrosion resistance, or temperature oxidation resistance [8, 9]. Lee et al. [10, 11], Shkodkin et al. [12], and Irissou et al. [8] reported that incorporating ceramic particles into metallic coatings not only improved the quality of the coatings by reducing their porosity but also increased their bonding strength. Porosity reduction of the coatings contributes to further enhanced corrosion resistance [8]. A variety of ceramics such as Al2O3 [10, 13], SiC [1417], and TiN [18, 19] have been investigated as the secondary phase in metallic matrix ceramic composites.

As one of the recently developed surface coating techniques, cold spraying differs from the conventional thermal spray methods. During cold spray processing, micron-sized particles are accelerated by an inert gas stream to a high velocity, and the feedstock remains in the solid state throughout the entire processing. This in turn solves the problems associated with high processing temperature, such as chemical reaction, phase transition, oxidation, or unfavorable structural changes. Apart from the advantages of spraying temperature-sensitive materials, cold spray provides the coatings with very dense microstructure [20], giving rise to better mechanical properties and corrosion resistance. For the temperature-sensitive magnesium alloy substrates, cold spray might be an appropriate technique for fabricating protective coatings. It has been clear that addition of hard ceramic particles into cold-sprayed coatings remarkably improves their hardness and wear resistance [12, 21]. In this study, Al-SiO2 composite coatings were deposited by cold spray on the AZ31 magnesium alloy. The effect of SiO2 content on the microstructure, mechanical properties, and anticorrosion properties of the coatings was examined and elucidated.

2. Materials and Methods

2.1. Deposition of Coatings

Spherical commercial pure aluminum powder (Beijing General Research Institute of Mining and Metallurgy, China) with a mean size of 25 µm and angular SiO2 powder (Beijing General Research Institute of Mining and Metallurgy, China) with a mean size of 16 µm were used in this study. As reported in [22], the ceramic particle can only deposit to form a coating unless it impacts on a metallic surface, and that impacting onto a ceramic surface cannot form a coating. Therefore, the ceramic content in the coating is less than that in the designed original powder. Depending on the nature and the particle size of the ceramic phase, the corresponding relation of the ceramic content in the original power and in the composite coating is different. For the new addition phase SiO2, this corresponding relation is unclear. Therefore, SiO2 particles with different contents from 10 vol.% to 40 vol.% were added into the powder by mechanical blending. The composite coatings with different SiO2 content were fabricated by cold spraying (CGT KINETIKS 4000, Germany). Nitrogen was used as acceleration gas and carrier gas at the temperature of 350°C and the pressure of 3 MPa. The spraying distance was 30 mm, and the traverse speed of the gun was 200 mm/s. The pure Al coating and composite coatings with 10 vol.%, 20 vol.%, and 40 vol.% SiO2 particles were named as coating 0, coating 10, coating 20, and coating 40, respectively. AZ31 magnesium alloy (AZ31) plates with a dimension of 20 mm × 20 mm × 3 mm were used as substrates. Prior to spraying, the substrates were surface grit blasted using 60-mesh black fused alumina sands and subsequently degreased by sonication cleaning in acetone.

2.2. Characterization of the Coatings

Phase composition of the powder and the coatings were characterized by X-ray diffraction (XRD, D8 Advance, Bruker AXS, Germany) at a scanning rate of 0.02°/s over a 2θ range of 10°∼90° using Cu Kα radiation operated at 35 mA and 40 kV. Topography and cross-sectional morphology of the coatings were examined by using field emission scanning electron microscope (FESEM, Hitachi S-550N, Japan). Element analyses were carried out by using energy dispersive X-ray spectra (EDX) equipped with FESEM. The porosity of the coating was measured by image analysis, and at least ten images per coating were taken. The microhardness of the coating was measured using a Vickers hardness tester with 100 g load with dwelling time of 10 s. Five points per sample were acquired randomly to obtain reliable data. Bonding strength of the coating was tested following the ASTM standard C633-01.

The tribological properties of the coatings were evaluated using a reciprocating-type ball-on-disc tribometer (JLTB-02, J&L Tech Co. Ltd., Korea). The tests were performed under a load of 3 N for 30 min at 18.6°C with a relative humidity of 70%. Steel balls (1Cr15) with a diameter of 6 mm were used as the counterparts. The coatings and balls were ultrasonically cleaned in acetone prior to the tests, and a new ball or a new position of the ball was used for each friction test. The friction coefficients and sliding time were automatically recorded during the tests. For each sample, the measurement was repeated three times. Electrochemical impedance spectroscopy (EIS) measurements were performed using a Solartron ModuLab system (2100A, UK). The tests were conducted in 3.5 wt.% NaCl solution at room temperature. AC signal of 10 mV and the frequency ranging from 100 kHz to 0.01 Hz were employed. Before the electrochemical measurement, the coating samples were immersed in an aerobic chamber containing 3.5 wt.% NaCl solution for 30 min. Each measurement was repeated three times.

3. Results and Discussion

Figure 1 shows the topographical morphology and cross section of the cold-sprayed coatings produced in this study. As can be seen from Figures 1(a) and 1(b), the pure Al coating and Al + SiO2 coatings exhibited significantly different surface morphologies. For coating 0 (Figure 1(a)), particles on the top of the coating remained smooth features with only few very small carters that were induced by debonding Al particles as indicated by white arrows. However, for the composite coating as shown in Figures 1(b)1(d), the top surface of the coatings was much rougher. Many angular carters can be observed as indicated by white arrows, which were due to the rebound of SiO2 during the coating formation process. For better understanding the deposition features of the coatings, the cross-sectional views of the coatings are provided in Figures 1(e) and 1(f). Clearly, composite coating was much denser than pure Al coating as the rebound SiO2 particles can tamp on the already deposited coating. For quantitative analysis, the porosity and SiO2 content of different coatings are listed in Table 1. The composite coatings demonstrated much denser structure than the pure Al coating. In addition, from Table 1, it is also found that the SiO2 content in the coating was lower than that in the original feedstock. Such difference became even larger as the SiO2 content in the feedstock increased. This fact suggests that a large amount of SiO2 particles will rebound during the deposition process.

Figure 2 shows the typical XRD spectra of the coatings and the EDS mapping of the composite coating named as coating 40. As can be seen from the XRD spectra, all coatings experienced no oxidation during the deposition process, which clearly indicates the advances of cold spray in producing oxide-free coatings. In addition, as the SiO2 content reduced significantly during the deposition process, no SiO2 peaks were detected in the composite coatings. However, from the EDS mapping as shown in Figures 2(b)2(d), the Si phase was clearly seen in the coating, which indicates the existence of SiO2 in the coating. Also, the SiO2 phase was found to be uniformly distributed in the coating.

As for the coating properties, Figure 3 shows the microhardness and adhesion strength of different coatings. It is clearly seen from Figure 3(a) that coating microhardness increased gradually as the SiO2 content increased. The reason for this phenomenon is the increased hard-phase reinforcement. On the one hand, SiO2 itself is much harder than Al matrix; higher SiO2 content certainly led to higher hardness. On the other hand, increased reinforcement helped to compact the coating and resulted in more work-hardening effect. Moreover, in Figure 3(b), the adhesion strength of the coating was also found to increase with increasing SiO2 content. This may also be attributed to the compacting effect of the hard SiO2 phase. Increased SiO2 content resulted in the reduction of pores and defects, hence increasing the adhesion strength. Another possible explanation contributing to the increase of the adhesion strength could be the anchor effect of the ceramic particles [23]. However, the interface between the coatings and the substrates did not present significant difference as a function of the ceramic content in the coatings in this investigation. Therefore, the increase of the adhesion strength of the coatings is mainly attributed to the compacting effect of the ceramic particles. The surface roughness of the substrate also influences the adhesion strength of the coating [24, 25], which would be investigated in the next study.

The tribological behavior of the coating was also investigated. Figure 4 shows the friction coefficient and wear weight loss of different coatings. It is seen that the friction coefficients of the cold-sprayed coatings were higher as compared with those of the substrate (Figure 4(a)), which may be due to the interior defects and ceramic particles [2628]. However, it seems that the pure coating and composite coatings had no significant difference in friction coefficient. Furthermore, the weight loss of different coatings as shown in Figure 4(b) indicates that the composite coatings had less weight loss than the pure Al coating, which clearly demonstrates the wear resistance capability of the Al + SiO2 composite coatings. For better understanding the wear mechanism, the morphology of the wear surface of the coatings is shown in Figure 5. In Figure 5(a) and 5(b), the severe delamination of the substrate material and pure Al coating was observed, indicating the serious wear. However, for the Al + SiO2 composite coatings, plough and extrusion characterization were found on the wear surface without obvious material delamination phenomenon. This explains why the weight loss of the composite coatings was smaller than that of the pure Al coating and substrate.

The corrosion behavior of the composite coating was also evaluated by potentiodynamic polarization technique. Figure 6 shows the polarization curves, corrosion potential, and corrosion current density of the different coatings. As can be seen, the current density of the composite coating is much lower than that of the Mg alloy substrate. The corrosion potential of the composite coating shifts toward positive. Both results suggest that Al + SiO2 composite coating had better corrosion resistance. Moreover, it is also found that the corrosion current density of the composite coating presents a decreasing trend as the ceramic content of the coating increases. The corrosion potential also shifts to positive with the increase of the ceramic content in the composite coating. The increase of the corrosion resistance of the composite coating with a higher ceramic content could be attributed to the decrease of the coating porosity. As shown in Table 1, the porosity of the coating decreases as the ceramic content in the composite coating increases.

4. Conclusions

For preventing Mg alloy from serious corrosion, Al and Al + SiO2 composite coatings were fabricated on the AZ31 magnesium alloy substrate by cold spraying. The results suggest that the Al-SiO2 composite coatings had higher performances than the pure Al coating. The composite coatings showed higher density, microhardness, bonding strength, wear resistance, and corrosion resistance. The content of SiO2 particles in the coatings had no significant effects on the coating properties. Moreover, SiO2 was found to significantly reduce in the coatings as compared with that in the feedstock due to the rebounding during the deposition. With the addition of SiO2 particles, the wear mechanism transfers from adhesive wear to abrasive wear.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This study was supported by the Natural Science Foundation of China (21705158), Fundamental Research Project of Qinghai Province (2016ZJ757), Project of Scientific Innovation Team of Ningbo (2015B11050), National Key Research and Development Program of Zhejiang Province (2015C01036), International Scientific and Technological Cooperation Project of Ningbo (2016D10012), and 3315 Program of Ningbo.