- Email: [email protected]
Microstructural characteristics of oxide layers formed on Mg–9 wt%Al–1 wt%Zn alloy via two-step plasma electrolytic oxidation
Journal of Alloys and Compounds xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom
Microstructural characteristics of oxide layers formed on Mg–9 wt%Al–1 wt%Zn alloy via two-step plasma electrolytic oxidation Kang Min Lee a, Young Gun Ko b,⇑, Dong Hyuk Shin a,⇑ a b
Department of Metallurgy and Materials Engineering, Hanyang University, Ansan 426-791, South Korea School of Materials Science and Engineering, Yeungnam University, Gyeongsan 712-749, South Korea
a r t i c l e
i n f o
a b s t r a c t
Article history: Available online xxxx
A study investigated the formation and microstructural features of the oxide layer formed on Mg–9 wt%Al–1 wt%Zn alloy coated by two-step plasma electrolytic oxidation (PEO) where an acid electrolyte with K2ZrF6 was used for the second PEO coating after the initial coating was done in an alkaline electrolyte. The microstructure, chemical compositions, and constituent compounds of the oxide layers were observed using scanning electron microscopy, electron probe micro-analyzer, and X-ray diffraction, respectively. The microstructural observations showed that the micropores caused by plasma discharge were formed in the oxide layers which were comprised of three different parts, namely, inner, intermediate, and outer layers from the substrate to the surface of the sample. The outer layer contained the highest concentration of Zr element whose amount decreased toward the substrate whereas the concentration of Mg element increased in the order of outer, intermediate, and inner layers. This finding suggested that the outer oxide layer was mainly comprised of ZrO2 compound while both ZrO2 and Mg2Zr5O12 compounds existed together as the main compounds in the intermediate oxide layer. Ó 2014 Elsevier B.V. All rights reserved.
Keywords: Mg alloy Plasma electrolytic oxidation K2ZrF6 Microstructure Chemical compound
1. Introduction Growing interest has been poured into the development of Mg alloys in the fields of electronics and automobile industry owing to their low density, high specific strength, and good machinability [1–3]. Despite their excellent properties, however, the poor corrosion resistance of Mg alloy samples limited the industrial applications in the numerous fields in which the high corrosion resistance was strongly required [4,5]. In order to protect bare Mg alloy samples from severe corrosion environment, several surface treatments, such as chromating, anodizing, and plasma electrolytic oxidation (PEO), have been used extensively to improve the corrosion properties of Mg alloy samples by creating the oxide layer on the top of the Mg alloy samples [6–8]. To date, PEO has been regarded as an eco-friendly surface reforming method that would generate the oxide layer on the metal surface with the excellent adhesion as intense plasma discharges with high energy state activated the plasma-enhanced electrochemical reactions in the electrolyte [9,10]. The surface morphologies of the oxide layer on Mg alloy sample subjected to PEO coating depended highly on processing variables, such as chemical ⇑ Corresponding authors. Tel.: +82 31 400 5224 (D.H. Shin). E-mail addresses: (D.H. Shin).
[email protected]
(Y.G.
Ko),
[email protected]
compositions of the electrolyte, electrical parameters, and post treatments [11–14]. The chemical composition of the electrolyte was most important in determining the microstructure and chemical compounds of the oxide layer. According to the recent investigations on the effects of chemical additives in the electrolyte [15–19], the oxide layers with various Zr-compounds which could be fabricated by adding K2ZrF6 into the electrolyte during the PEO process were desirable for enhancing the corrosion resistance. Luo et al. [15] reported that the incorporation of ZrO2 compound into the oxide layer formed in zirconate electrolytes was propitious to improve the corrosion resistance of AZ91D Mg alloy at relatively high temperature. Liu et al. [16] demonstrated that the Zr-compounds, such as Mg2Zr5O12 and ZrO2, were beneficial for both passivation effect against the corrosion and compactness of the oxide layer formed on AM50 Mg alloy. Few studies, however, have been made to investigate the formation of the oxide layer containing Zr-compounds, in terms of the microstructural variation, chemical composition, and constituent phases affecting the corrosion resistance. Recently, Wang et al. [17] suggested that when K2ZrF6 was added to the alkaline electrolyte in order to form the Zr-compounds, the plasma discharges on Mg alloy samples were difficult to take place during PEO coating since the formation of the dielectric layer was apparently restricted which was associated with the instability of the electrolyte. On the other hand, the use of K2ZrF6 in the acid
0925-8388/$ - see front matter Ó 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jallcom.2014.01.097
Please cite this article in press as: K.M. Lee et al., J. Alloys Comp. (2014), http://dx.doi.org/10.1016/j.jallcom.2014.01.097
2
K.M. Lee et al. / Journal of Alloys and Compounds xxx (2014) xxx–xxx
electrolyte could cause the noticeable anodic spark, which led to the growth of the oxide layer in Mg samples [16]. To alleviate the instability of the electrolyte as well as the difficulty in forming the oxide layer in the electrolyte with K2ZrF6, thus, we propose two-step PEO coating approach where an acid electrolyte with K2ZrF6 is used for the second PEO coating after the initial coating is made in a phosphate alkaline electrolyte. Then, the microstructure and chemical composition, and constituent phases of the oxide layer fabricated via two-step PEO coating of Mg–9 wt%Al– 1 wt%Zn alloy sample were investigated.
2. Experimental Prior to the PEO process, Mg–9 wt%Al–1 wt%Zn alloy samples with a size of 20 30 2 mm3 were mechanically polished with 1000 grit emery paper and subsequently rinsed with distilled water. The samples were then cleaned ultrasonically in ethanol. Two-step PEO coating was conducted utilizing a 6.5 kW DC power supply (Unicorntech, 65010D) in conjunction with stirring and cooling systems. The electrolytic cell consisted of a 2 l glass-vessel with a sample holder and stainless steel of dimensions 15 25 cm2 was used as cathode. The chemical compositions and electrochemical properties of the electrolytes are listed in Table 1. The alkaline and acid electrolytes are denoted as ‘Bath A’ and ‘Bath B’, respectively. Two-step PEO coating comprised two individual PEO coatings within Baths A and B in a sequent manner. First PEO coating was conducted in Bath A for 60 s because the sample would be covered fully with the thin coating layer which might prevent the dissolution of Mg substrate further during second PEO coating. Then, the sample was pulled out from Bath A and immersed in Bath B which was prepared in a different vessel. Second PEO coating for 600 s was made in Bath B containing K2ZrF6 compound which might be known to trigger effectively the growth kinetics of the newly-formed oxide layer with fairly good compactness. To prevent the effects of PEO electrical parameters on the growth behavior of the oxide layer during twostep PEO coating, the current density, frequency, and duty ratio were fixed as 100 mA/cm2, 500 Hz, and 30%, respectively. Schematic illustration of sequent two-step PEO coating is shown in Fig. 1. The microstructure and chemical compositions of the surface of Mg alloy sample were collected using a field-emission scanning electron microscopy (SEM) equipped with an energy dispersive X-ray spectrometer (EDS). The composition profile of the cross section was examined by an electron probe micro-analyzer
Fig. 2. Coating voltage vs. coating time of Mg alloy sample during two-step PEO coating in Bath B and PEO coating in Bath A. (EPMA). The Cu Ka radiation was used for the present X-ray diffraction analysis operated in Bragg–Brentano mode. The wide scan was performed from 20° to 80° with a step size of 0.05°.
3. Results and discussion The coating voltage vs. coating time curve of the present sample processed by two-step PEO coating in Bath B is shown in Fig. 2, and this curve is also compared to that of the sample processed by PEO coating in Bath A during the same time. As reported earlier, the coating voltage of the sample increased steeply due to the partial dissolution of the substrate as well as the occurrence of the thin passivation film working as an insulator at the beginning of PEO
Table 1 Compositions and electrochemical properties of the electrolytes used.
Bath A Bath B
KOH (M)
Na3PO412H2O (M)
H3PO4 (M)
K2ZrF6 (M)
pH
EC (mS/cm)
0.02
0.03
0.02
0.02
12.2 2.7
10.4 7.4
Fig. 1. Schematic illustration of the sequent two-step PEO coating.
Please cite this article in press as: K.M. Lee et al., J. Alloys Comp. (2014), http://dx.doi.org/10.1016/j.jallcom.2014.01.097
K.M. Lee et al. / Journal of Alloys and Compounds xxx (2014) xxx–xxx
coating [20]. As the coating time increased, the plasma discharges in the vicinity of the sample began to appear noticeably when the coating voltage exceeded the breakdown voltage [9]. The breakdown voltages of the samples via PEO coating in Bath A and two-step PEO coating in Bath B were observed to be 330 and 475 V, respectively. Wang et al. [21] reported that the breakdown voltage implied the degree of difficulty to begin to form the oxide layer on metal substrate. In this study, the passivation film of 5 lm in thickness, generated by the first coating prior to the second coating, might play as an insulator to suppress the conduction of electrons from the electrolyte to the substrate during two-step PEO coating, which differed from PEO coating in Bath A. It is clear that the breakdown voltage of the sample during two-step PEO coating in Bath B was higher than that during PEO coating in Bath A. In addition, the final voltage of the sample was also higher with two-step PEO coating in Bath B than with PEO coating in Bath A. This finding was also explained by the use of different electrolytes for Baths A (alkaline electrolyte) and B (acid electrolyte). The chemical compositions, pH, and electrical conductivity of Baths A and B were different as listed clearly in Table 1. Under that condition, when the high current density was applied in the cell, the different oxide layers formed on the substrate because the different electrochemical reactions took place. This would affect the voltage response. Fig. 3 displays the surface image and EDS point analysis of the oxide layer on Mg alloy sample after two-step PEO coating. The microstructure of the oxide layer exhibited a number of micropores (3 lm in diameter), which were formed by the rapid solidification of molten oxides in the relatively cool electrolyte following plasma discharges [10,22]. Interestingly, the uneven nodular structures in the vicinity of micropores were observed. EDS spectra measured at two different regions A and B showed that the mainly chemical components of the oxide layer were O, Mg, and Zr elements. This result suggested that Zr ions in Bath B were involved effectively in the electrochemical reactions to form the oxide layer. Meanwhile, region B contained a reduced amount of Mg, with typical Mg:Zr atomic ratio from EDS point analysis of about 0.14, compared with about 0.21 in the region A. It is of interest that Zr:O atomic ratio of 0.49 in region B was likely to be ZrO2 via the co-electrodeposition of Zr ions and ionized O in the electrolyte.
3
Fig. 4 shows EPMA line profiles of the cross section of the oxide layer on Mg alloy sample after two-step PEO coating. O, Mg, and Zr were selected as the main chemical elements of interest. As shown in back-scattered electron (BSE) image of Fig. 4, the oxide layer of 35 lm in thickness was formed with good adhesion to the substrate. In general, the oxide layer on Mg alloy treated by PEO coating was comprised of two different layers: inner and outer layers [23,24]. In this study, however, depending on the degree of brightness in the BSE image and the gradient of several elements in EPMA results, the oxide layer on Mg alloy sample after two-step PEO coating could be divided into three different parts: inner, intermediate, and outer layers from the substrate to the surface of the sample. The thin inner layer with fairly compactness formed next to the substrate (region A). The intermediate layer occupying a majority of the oxide layer possessed some micropores owing to the generation of the dissolved oxygen gas trapped in the molten oxide during PEO coating (region B) [13,25]. The brightest outer oxide layer contained some cracks due to the thermal stress during PEO coating (region C) [26]. On the basis of EPMA analysis, the concentration of O was reasonably uniform throughout the oxide layer, while that of Mg decreased markedly from the substrate to the coating surface of the sample. Zr was detected not only in the intermediate but also in the outer layers. It is of interest that the amount of Zr in the outer layer was the highest, implying that most of the outer layer might be Zr-rich compounds, such as Zr–O and Mg–Zr–O. Hence, the gradients of these elements in EPMA spectra revealed that main constituent phases of the three different oxide layers should be varied. In order to analyze quantitatively the constituent phases of the oxide layer, XRD analysis of the oxide layers on Mg alloy samples after first PEO coating in Bath A for 60 s and two-step PEO coating in Bath B for 600 s was performed, and the result is shown in Fig. 5. After PEO coating in Bath A for 60 s the main composition of the passivation film on Mg alloy sample was MgO phase. Meanwhile, the oxide layer after two-step PEO coating consisted of major and minor phases as Mg2Zr5O12, ZrO2, and MgO, MgF2 phases, respectively. According to the study proposed by Tolstoy and Altangerel [27], in terms of fractions of Zr4+ when K2ZrF6 was dissolved in the acid solution at pH 3, ZrF3+ complex was mainly present. On the basis of this simulation, the chemical equations related
Fig. 3. Surface morphology and EDS spot analysis of the oxide layer on Mg alloy sample treated by two-step PEO coating.
Please cite this article in press as: K.M. Lee et al., J. Alloys Comp. (2014), http://dx.doi.org/10.1016/j.jallcom.2014.01.097
4
K.M. Lee et al. / Journal of Alloys and Compounds xxx (2014) xxx–xxx
Fig. 4. EPMA analysis of the oxide layer on Mg alloy sample treated by two-step PEO coating. Regions A, B, and C denote inner, intermediate, outer layers, respectively.
Fig. 5. XRD patterns of the oxide layers on Mg alloy samples after (a) first PEO coating in Bath A for 60 s and (b) two-step PEO coating in Bath B for 600 s.
It was considered that from Fig. 5(a), MgO phase could be generated mainly during first PEO coating in Bath A owing to the oxidation of the substrate. Metastable Mg2Zr5O12 phase, reported to be stable in the range of 2123–2373 K [11], was synthesized by thermo-electrochemical reactions with short-lived plasma with locally high temperature and the quenching effect. It implies that Mg2Zr5O12 phase was located mainly in the intermediate layer, because the oxide layer dominantly grew down to the substrate [28]. In addition, as expected above in EDS results of Fig. 3, ZrO2 phase was formed in the outer oxide layer on the substrate via the anodic deposition. Meanwhile, highly reactive F ions in the electrolyte reacted dominantly with Mg2+ ions to form MgF2 phase [29]. In this study, however, since the dissolution of Mg alloy sample during two-step PEO coating was suppressed owing to the presence of the passivation film on the substrate, the oxide layer did not contain numerous MgF2 phase. It was concluded that the outer and intermediate oxide layer was mainly comprised of ZrO2 and Mg2Zr5O12 phase via the electrochemical reactions during two-step PEO coating. In addition, further experimental data would be required to elucidate the mechanism and quantitative analysis of the chemical compounds formed during PEO coating.
to the electrochemical reactions to form the chemical compounds in the oxide layer are given below:
Mg2þ þ O2 ! MgO 5ZrFþ3 þ 2Mg2þ þ 14O2 ! Mg2 Zr5 O12 þ 15F þ 4e þ O2 " ZrFþ3 þ 2O2 ! ZrO2 þ 3F þ e Mg2þ þ 2F ! MgF2
4. Summary We successfully fabricated the oxide layer on Mg alloy sample treated by two-step PEO coating where an acid electrolyte with K2ZrF6 was used for the second PEO coating after the initial coating was made in a phosphate alkaline electrolyte. The breakdown
Please cite this article in press as: K.M. Lee et al., J. Alloys Comp. (2014), http://dx.doi.org/10.1016/j.jallcom.2014.01.097
K.M. Lee et al. / Journal of Alloys and Compounds xxx (2014) xxx–xxx
voltage of the Mg alloy sample during two-step PEO coating was higher than that during first PEO coating in two-step PEO coating. This phenomenon was influenced by the different electrolytes in Baths A and B. The different oxide layers would appear on the substrate because the different electrochemical reactions took place. In addition to the effect of electrolyte, the passivation film formed by first PEO coating would baffle the conduction of electrons. The microstructural observations showed that the oxide layer containing a number of micropores and uneven nodules was divided into three different parts: inner, intermediate, and outer oxide layers from the substrate to the surface of the sample. Due to the electrochemical reactions assisted by plasma discharges and the anodic oxidation, Mg2Zr5O12 and ZrO2 phases could be synthesized dominantly in the intermediate and outer oxide layers, respectively. Acknowledgment This study was supported by a National Research Foundation (NRF) Grant (2012-0000130) funded by the Korean Government. References [1] [2] [3] [4] [5] [6]
K. Kubota, M. Mabuchi, K. Higashi, J. Mater. Sci. 34 (1999) 2255–2262. G.L. Song, A. Atrens, Adv. Eng. Mater. 1 (1999) 11–33. B.L. Mordike, T. Ebert, Mater. Sci. Eng., A 302 (2001) 37–45. G.L. Song, A. Atrens, Adv. Eng. Mater. 5 (2003) 837–858. J.E. Gray, B. Luan, J. Alloys Comp. 336 (2002) 88–113. C. Zhong, F. Liu, Y. Wu, J. Le, L. Liu, M. He, J. Zhu, W. Hu, J. Alloys Comp. 520 (2012) 11–21.
5
[7] C. Blawert, W. Dietzel, E. Ghali, G. Song, Adv. Eng. Mater. 8 (2006) 511–533. [8] J.I. Yu, J.S. Yoon, J.G. Yoon, J.H. Kim, S.D. Choi, H.K. Jang, J.Y. Yu, I.H. Bae, Kor. J. Met. Mater. 51 (2013) 405–411. [9] K.M. Lee, K.R. Shin, S. Namgung, B. Yoo, D.H. Shin, Surf. Coat. Technol. 205 (2011) 3779–3784. [10] A.L. Yerokhin, X. Nie, A. Leyland, A. Matthews, S.J. Dowey, Surf. Coat. Technol. 122 (1999) 73–93. [11] R. Arrabal, E. Matykina, F. Viejo, P. Skeldon, G.E. Thompson, M.C. Merino, Appl. Surf. Sci. 254 (2008) 6937–6942. [12] Y.G. Ko, K.M. Lee, B.U. Lee, D.H. Shin, J. Alloys Comp. 509 (S1) (2011) S468– S472. [13] I.J. Hwang, D.Y. Hwang, Y.G. Ko, D.H. Shin, Surf. Coat. Technol. 206 (2012) 3360–3365. [14] B.U. Lee, I.J. Hwang, J.S. Lee, Y.G. Ko, D.H. Shin, Kor. J. Met. Mater. 49 (2011) 601–607. [15] H. Luo, Q. Cai, B. Wei, B. Yu, J. He, D. Li, J. Alloys Comp. 474 (2009) 551–556. [16] F. Liu, D. Shan, Y. Song, E.-H. Han, Surf. Coat. Technol. 206 (2011) 455–463. [17] L. Wang, L. Chen, Z. Yan, H. Wang, J. Peng, J. Alloys Comp. 493 (2010) 445–452. [18] F. Liu, D. Shan, Y. Song, E.-H. Han, W. Ke, Corros. Sci. 53 (2011) 3845–3852. [19] J. Liang, P.B. Srinivasan, C. Blawert, W. Dietzel, Corros. Sci. 51 (2009) 2483– 2492. [20] Y.G. Ko, S. Namgung, D.H. Shin, Surf. Coat. Technol. 205 (2010) 2525–2531. [21] Y.Q. Wang, K. Wu, M.Y. Zheng, Surf. Coat. Technol. 201 (2006) 353–360. [22] P. Su, X. Wu, Y. Guo, Z. Jiang, J. Alloys Comp. 475 (2009) 773–777. [23] V. Birss, S. Xia, R. Yue, R.G. Rateick, J. Electrochem. Soc. 151 (2004) B1–B10. [24] O. Khaselev, D. Weiss, J. Yahalom, Corros. Sci. 43 (2001) 1295–1307. [25] K.M. Lee, J.O. Jo, E.S. Lee, B. Yoo, D.H. Shin, J. Electrochem. Soc. 158 (2011) C325–C328. [26] R.O. Hussein, X. Nie, D.O. Northwood, A. Yerokhin, A. Matthews, J. Phys. D: Appl. Phys. 43 (2010) 105203. [27] V.P. Tolstoy, B. Altangerel, Mater. Lett. 61 (2007) 123–125. [28] R. Arrabal, E. Matykina, T. Hashimoto, P. Skeldon, G.E. Thompson, Surf. Coat. Technol. 203 (2009) 2207–2220. [29] D.Y. Hwang, Y.M. Kim, D.H. Shin, Mater. Trans. 50 (2009) 671–678.
Please cite this article in press as: K.M. Lee et al., J. Alloys Comp. (2014), http://dx.doi.org/10.1016/j.jallcom.2014.01.097
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom
Microstructural characteristics of oxide layers formed on Mg–9 wt%Al–1 wt%Zn alloy via two-step plasma electrolytic oxidation Kang Min Lee a, Young Gun Ko b,⇑, Dong Hyuk Shin a,⇑ a b
Department of Metallurgy and Materials Engineering, Hanyang University, Ansan 426-791, South Korea School of Materials Science and Engineering, Yeungnam University, Gyeongsan 712-749, South Korea
a r t i c l e
i n f o
a b s t r a c t
Article history: Available online xxxx
A study investigated the formation and microstructural features of the oxide layer formed on Mg–9 wt%Al–1 wt%Zn alloy coated by two-step plasma electrolytic oxidation (PEO) where an acid electrolyte with K2ZrF6 was used for the second PEO coating after the initial coating was done in an alkaline electrolyte. The microstructure, chemical compositions, and constituent compounds of the oxide layers were observed using scanning electron microscopy, electron probe micro-analyzer, and X-ray diffraction, respectively. The microstructural observations showed that the micropores caused by plasma discharge were formed in the oxide layers which were comprised of three different parts, namely, inner, intermediate, and outer layers from the substrate to the surface of the sample. The outer layer contained the highest concentration of Zr element whose amount decreased toward the substrate whereas the concentration of Mg element increased in the order of outer, intermediate, and inner layers. This finding suggested that the outer oxide layer was mainly comprised of ZrO2 compound while both ZrO2 and Mg2Zr5O12 compounds existed together as the main compounds in the intermediate oxide layer. Ó 2014 Elsevier B.V. All rights reserved.
Keywords: Mg alloy Plasma electrolytic oxidation K2ZrF6 Microstructure Chemical compound
1. Introduction Growing interest has been poured into the development of Mg alloys in the fields of electronics and automobile industry owing to their low density, high specific strength, and good machinability [1–3]. Despite their excellent properties, however, the poor corrosion resistance of Mg alloy samples limited the industrial applications in the numerous fields in which the high corrosion resistance was strongly required [4,5]. In order to protect bare Mg alloy samples from severe corrosion environment, several surface treatments, such as chromating, anodizing, and plasma electrolytic oxidation (PEO), have been used extensively to improve the corrosion properties of Mg alloy samples by creating the oxide layer on the top of the Mg alloy samples [6–8]. To date, PEO has been regarded as an eco-friendly surface reforming method that would generate the oxide layer on the metal surface with the excellent adhesion as intense plasma discharges with high energy state activated the plasma-enhanced electrochemical reactions in the electrolyte [9,10]. The surface morphologies of the oxide layer on Mg alloy sample subjected to PEO coating depended highly on processing variables, such as chemical ⇑ Corresponding authors. Tel.: +82 31 400 5224 (D.H. Shin). E-mail addresses: (D.H. Shin).
[email protected]
(Y.G.
Ko),
[email protected]
compositions of the electrolyte, electrical parameters, and post treatments [11–14]. The chemical composition of the electrolyte was most important in determining the microstructure and chemical compounds of the oxide layer. According to the recent investigations on the effects of chemical additives in the electrolyte [15–19], the oxide layers with various Zr-compounds which could be fabricated by adding K2ZrF6 into the electrolyte during the PEO process were desirable for enhancing the corrosion resistance. Luo et al. [15] reported that the incorporation of ZrO2 compound into the oxide layer formed in zirconate electrolytes was propitious to improve the corrosion resistance of AZ91D Mg alloy at relatively high temperature. Liu et al. [16] demonstrated that the Zr-compounds, such as Mg2Zr5O12 and ZrO2, were beneficial for both passivation effect against the corrosion and compactness of the oxide layer formed on AM50 Mg alloy. Few studies, however, have been made to investigate the formation of the oxide layer containing Zr-compounds, in terms of the microstructural variation, chemical composition, and constituent phases affecting the corrosion resistance. Recently, Wang et al. [17] suggested that when K2ZrF6 was added to the alkaline electrolyte in order to form the Zr-compounds, the plasma discharges on Mg alloy samples were difficult to take place during PEO coating since the formation of the dielectric layer was apparently restricted which was associated with the instability of the electrolyte. On the other hand, the use of K2ZrF6 in the acid
0925-8388/$ - see front matter Ó 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jallcom.2014.01.097
Please cite this article in press as: K.M. Lee et al., J. Alloys Comp. (2014), http://dx.doi.org/10.1016/j.jallcom.2014.01.097
2
K.M. Lee et al. / Journal of Alloys and Compounds xxx (2014) xxx–xxx
electrolyte could cause the noticeable anodic spark, which led to the growth of the oxide layer in Mg samples [16]. To alleviate the instability of the electrolyte as well as the difficulty in forming the oxide layer in the electrolyte with K2ZrF6, thus, we propose two-step PEO coating approach where an acid electrolyte with K2ZrF6 is used for the second PEO coating after the initial coating is made in a phosphate alkaline electrolyte. Then, the microstructure and chemical composition, and constituent phases of the oxide layer fabricated via two-step PEO coating of Mg–9 wt%Al– 1 wt%Zn alloy sample were investigated.
2. Experimental Prior to the PEO process, Mg–9 wt%Al–1 wt%Zn alloy samples with a size of 20 30 2 mm3 were mechanically polished with 1000 grit emery paper and subsequently rinsed with distilled water. The samples were then cleaned ultrasonically in ethanol. Two-step PEO coating was conducted utilizing a 6.5 kW DC power supply (Unicorntech, 65010D) in conjunction with stirring and cooling systems. The electrolytic cell consisted of a 2 l glass-vessel with a sample holder and stainless steel of dimensions 15 25 cm2 was used as cathode. The chemical compositions and electrochemical properties of the electrolytes are listed in Table 1. The alkaline and acid electrolytes are denoted as ‘Bath A’ and ‘Bath B’, respectively. Two-step PEO coating comprised two individual PEO coatings within Baths A and B in a sequent manner. First PEO coating was conducted in Bath A for 60 s because the sample would be covered fully with the thin coating layer which might prevent the dissolution of Mg substrate further during second PEO coating. Then, the sample was pulled out from Bath A and immersed in Bath B which was prepared in a different vessel. Second PEO coating for 600 s was made in Bath B containing K2ZrF6 compound which might be known to trigger effectively the growth kinetics of the newly-formed oxide layer with fairly good compactness. To prevent the effects of PEO electrical parameters on the growth behavior of the oxide layer during twostep PEO coating, the current density, frequency, and duty ratio were fixed as 100 mA/cm2, 500 Hz, and 30%, respectively. Schematic illustration of sequent two-step PEO coating is shown in Fig. 1. The microstructure and chemical compositions of the surface of Mg alloy sample were collected using a field-emission scanning electron microscopy (SEM) equipped with an energy dispersive X-ray spectrometer (EDS). The composition profile of the cross section was examined by an electron probe micro-analyzer
Fig. 2. Coating voltage vs. coating time of Mg alloy sample during two-step PEO coating in Bath B and PEO coating in Bath A. (EPMA). The Cu Ka radiation was used for the present X-ray diffraction analysis operated in Bragg–Brentano mode. The wide scan was performed from 20° to 80° with a step size of 0.05°.
3. Results and discussion The coating voltage vs. coating time curve of the present sample processed by two-step PEO coating in Bath B is shown in Fig. 2, and this curve is also compared to that of the sample processed by PEO coating in Bath A during the same time. As reported earlier, the coating voltage of the sample increased steeply due to the partial dissolution of the substrate as well as the occurrence of the thin passivation film working as an insulator at the beginning of PEO
Table 1 Compositions and electrochemical properties of the electrolytes used.
Bath A Bath B
KOH (M)
Na3PO412H2O (M)
H3PO4 (M)
K2ZrF6 (M)
pH
EC (mS/cm)
0.02
0.03
0.02
0.02
12.2 2.7
10.4 7.4
Fig. 1. Schematic illustration of the sequent two-step PEO coating.
Please cite this article in press as: K.M. Lee et al., J. Alloys Comp. (2014), http://dx.doi.org/10.1016/j.jallcom.2014.01.097
K.M. Lee et al. / Journal of Alloys and Compounds xxx (2014) xxx–xxx
coating [20]. As the coating time increased, the plasma discharges in the vicinity of the sample began to appear noticeably when the coating voltage exceeded the breakdown voltage [9]. The breakdown voltages of the samples via PEO coating in Bath A and two-step PEO coating in Bath B were observed to be 330 and 475 V, respectively. Wang et al. [21] reported that the breakdown voltage implied the degree of difficulty to begin to form the oxide layer on metal substrate. In this study, the passivation film of 5 lm in thickness, generated by the first coating prior to the second coating, might play as an insulator to suppress the conduction of electrons from the electrolyte to the substrate during two-step PEO coating, which differed from PEO coating in Bath A. It is clear that the breakdown voltage of the sample during two-step PEO coating in Bath B was higher than that during PEO coating in Bath A. In addition, the final voltage of the sample was also higher with two-step PEO coating in Bath B than with PEO coating in Bath A. This finding was also explained by the use of different electrolytes for Baths A (alkaline electrolyte) and B (acid electrolyte). The chemical compositions, pH, and electrical conductivity of Baths A and B were different as listed clearly in Table 1. Under that condition, when the high current density was applied in the cell, the different oxide layers formed on the substrate because the different electrochemical reactions took place. This would affect the voltage response. Fig. 3 displays the surface image and EDS point analysis of the oxide layer on Mg alloy sample after two-step PEO coating. The microstructure of the oxide layer exhibited a number of micropores (3 lm in diameter), which were formed by the rapid solidification of molten oxides in the relatively cool electrolyte following plasma discharges [10,22]. Interestingly, the uneven nodular structures in the vicinity of micropores were observed. EDS spectra measured at two different regions A and B showed that the mainly chemical components of the oxide layer were O, Mg, and Zr elements. This result suggested that Zr ions in Bath B were involved effectively in the electrochemical reactions to form the oxide layer. Meanwhile, region B contained a reduced amount of Mg, with typical Mg:Zr atomic ratio from EDS point analysis of about 0.14, compared with about 0.21 in the region A. It is of interest that Zr:O atomic ratio of 0.49 in region B was likely to be ZrO2 via the co-electrodeposition of Zr ions and ionized O in the electrolyte.
3
Fig. 4 shows EPMA line profiles of the cross section of the oxide layer on Mg alloy sample after two-step PEO coating. O, Mg, and Zr were selected as the main chemical elements of interest. As shown in back-scattered electron (BSE) image of Fig. 4, the oxide layer of 35 lm in thickness was formed with good adhesion to the substrate. In general, the oxide layer on Mg alloy treated by PEO coating was comprised of two different layers: inner and outer layers [23,24]. In this study, however, depending on the degree of brightness in the BSE image and the gradient of several elements in EPMA results, the oxide layer on Mg alloy sample after two-step PEO coating could be divided into three different parts: inner, intermediate, and outer layers from the substrate to the surface of the sample. The thin inner layer with fairly compactness formed next to the substrate (region A). The intermediate layer occupying a majority of the oxide layer possessed some micropores owing to the generation of the dissolved oxygen gas trapped in the molten oxide during PEO coating (region B) [13,25]. The brightest outer oxide layer contained some cracks due to the thermal stress during PEO coating (region C) [26]. On the basis of EPMA analysis, the concentration of O was reasonably uniform throughout the oxide layer, while that of Mg decreased markedly from the substrate to the coating surface of the sample. Zr was detected not only in the intermediate but also in the outer layers. It is of interest that the amount of Zr in the outer layer was the highest, implying that most of the outer layer might be Zr-rich compounds, such as Zr–O and Mg–Zr–O. Hence, the gradients of these elements in EPMA spectra revealed that main constituent phases of the three different oxide layers should be varied. In order to analyze quantitatively the constituent phases of the oxide layer, XRD analysis of the oxide layers on Mg alloy samples after first PEO coating in Bath A for 60 s and two-step PEO coating in Bath B for 600 s was performed, and the result is shown in Fig. 5. After PEO coating in Bath A for 60 s the main composition of the passivation film on Mg alloy sample was MgO phase. Meanwhile, the oxide layer after two-step PEO coating consisted of major and minor phases as Mg2Zr5O12, ZrO2, and MgO, MgF2 phases, respectively. According to the study proposed by Tolstoy and Altangerel [27], in terms of fractions of Zr4+ when K2ZrF6 was dissolved in the acid solution at pH 3, ZrF3+ complex was mainly present. On the basis of this simulation, the chemical equations related
Fig. 3. Surface morphology and EDS spot analysis of the oxide layer on Mg alloy sample treated by two-step PEO coating.
Please cite this article in press as: K.M. Lee et al., J. Alloys Comp. (2014), http://dx.doi.org/10.1016/j.jallcom.2014.01.097
4
K.M. Lee et al. / Journal of Alloys and Compounds xxx (2014) xxx–xxx
Fig. 4. EPMA analysis of the oxide layer on Mg alloy sample treated by two-step PEO coating. Regions A, B, and C denote inner, intermediate, outer layers, respectively.
Fig. 5. XRD patterns of the oxide layers on Mg alloy samples after (a) first PEO coating in Bath A for 60 s and (b) two-step PEO coating in Bath B for 600 s.
It was considered that from Fig. 5(a), MgO phase could be generated mainly during first PEO coating in Bath A owing to the oxidation of the substrate. Metastable Mg2Zr5O12 phase, reported to be stable in the range of 2123–2373 K [11], was synthesized by thermo-electrochemical reactions with short-lived plasma with locally high temperature and the quenching effect. It implies that Mg2Zr5O12 phase was located mainly in the intermediate layer, because the oxide layer dominantly grew down to the substrate [28]. In addition, as expected above in EDS results of Fig. 3, ZrO2 phase was formed in the outer oxide layer on the substrate via the anodic deposition. Meanwhile, highly reactive F ions in the electrolyte reacted dominantly with Mg2+ ions to form MgF2 phase [29]. In this study, however, since the dissolution of Mg alloy sample during two-step PEO coating was suppressed owing to the presence of the passivation film on the substrate, the oxide layer did not contain numerous MgF2 phase. It was concluded that the outer and intermediate oxide layer was mainly comprised of ZrO2 and Mg2Zr5O12 phase via the electrochemical reactions during two-step PEO coating. In addition, further experimental data would be required to elucidate the mechanism and quantitative analysis of the chemical compounds formed during PEO coating.
to the electrochemical reactions to form the chemical compounds in the oxide layer are given below:
Mg2þ þ O2 ! MgO 5ZrFþ3 þ 2Mg2þ þ 14O2 ! Mg2 Zr5 O12 þ 15F þ 4e þ O2 " ZrFþ3 þ 2O2 ! ZrO2 þ 3F þ e Mg2þ þ 2F ! MgF2
4. Summary We successfully fabricated the oxide layer on Mg alloy sample treated by two-step PEO coating where an acid electrolyte with K2ZrF6 was used for the second PEO coating after the initial coating was made in a phosphate alkaline electrolyte. The breakdown
Please cite this article in press as: K.M. Lee et al., J. Alloys Comp. (2014), http://dx.doi.org/10.1016/j.jallcom.2014.01.097
K.M. Lee et al. / Journal of Alloys and Compounds xxx (2014) xxx–xxx
voltage of the Mg alloy sample during two-step PEO coating was higher than that during first PEO coating in two-step PEO coating. This phenomenon was influenced by the different electrolytes in Baths A and B. The different oxide layers would appear on the substrate because the different electrochemical reactions took place. In addition to the effect of electrolyte, the passivation film formed by first PEO coating would baffle the conduction of electrons. The microstructural observations showed that the oxide layer containing a number of micropores and uneven nodules was divided into three different parts: inner, intermediate, and outer oxide layers from the substrate to the surface of the sample. Due to the electrochemical reactions assisted by plasma discharges and the anodic oxidation, Mg2Zr5O12 and ZrO2 phases could be synthesized dominantly in the intermediate and outer oxide layers, respectively. Acknowledgment This study was supported by a National Research Foundation (NRF) Grant (2012-0000130) funded by the Korean Government. References [1] [2] [3] [4] [5] [6]
K. Kubota, M. Mabuchi, K. Higashi, J. Mater. Sci. 34 (1999) 2255–2262. G.L. Song, A. Atrens, Adv. Eng. Mater. 1 (1999) 11–33. B.L. Mordike, T. Ebert, Mater. Sci. Eng., A 302 (2001) 37–45. G.L. Song, A. Atrens, Adv. Eng. Mater. 5 (2003) 837–858. J.E. Gray, B. Luan, J. Alloys Comp. 336 (2002) 88–113. C. Zhong, F. Liu, Y. Wu, J. Le, L. Liu, M. He, J. Zhu, W. Hu, J. Alloys Comp. 520 (2012) 11–21.
5
[7] C. Blawert, W. Dietzel, E. Ghali, G. Song, Adv. Eng. Mater. 8 (2006) 511–533. [8] J.I. Yu, J.S. Yoon, J.G. Yoon, J.H. Kim, S.D. Choi, H.K. Jang, J.Y. Yu, I.H. Bae, Kor. J. Met. Mater. 51 (2013) 405–411. [9] K.M. Lee, K.R. Shin, S. Namgung, B. Yoo, D.H. Shin, Surf. Coat. Technol. 205 (2011) 3779–3784. [10] A.L. Yerokhin, X. Nie, A. Leyland, A. Matthews, S.J. Dowey, Surf. Coat. Technol. 122 (1999) 73–93. [11] R. Arrabal, E. Matykina, F. Viejo, P. Skeldon, G.E. Thompson, M.C. Merino, Appl. Surf. Sci. 254 (2008) 6937–6942. [12] Y.G. Ko, K.M. Lee, B.U. Lee, D.H. Shin, J. Alloys Comp. 509 (S1) (2011) S468– S472. [13] I.J. Hwang, D.Y. Hwang, Y.G. Ko, D.H. Shin, Surf. Coat. Technol. 206 (2012) 3360–3365. [14] B.U. Lee, I.J. Hwang, J.S. Lee, Y.G. Ko, D.H. Shin, Kor. J. Met. Mater. 49 (2011) 601–607. [15] H. Luo, Q. Cai, B. Wei, B. Yu, J. He, D. Li, J. Alloys Comp. 474 (2009) 551–556. [16] F. Liu, D. Shan, Y. Song, E.-H. Han, Surf. Coat. Technol. 206 (2011) 455–463. [17] L. Wang, L. Chen, Z. Yan, H. Wang, J. Peng, J. Alloys Comp. 493 (2010) 445–452. [18] F. Liu, D. Shan, Y. Song, E.-H. Han, W. Ke, Corros. Sci. 53 (2011) 3845–3852. [19] J. Liang, P.B. Srinivasan, C. Blawert, W. Dietzel, Corros. Sci. 51 (2009) 2483– 2492. [20] Y.G. Ko, S. Namgung, D.H. Shin, Surf. Coat. Technol. 205 (2010) 2525–2531. [21] Y.Q. Wang, K. Wu, M.Y. Zheng, Surf. Coat. Technol. 201 (2006) 353–360. [22] P. Su, X. Wu, Y. Guo, Z. Jiang, J. Alloys Comp. 475 (2009) 773–777. [23] V. Birss, S. Xia, R. Yue, R.G. Rateick, J. Electrochem. Soc. 151 (2004) B1–B10. [24] O. Khaselev, D. Weiss, J. Yahalom, Corros. Sci. 43 (2001) 1295–1307. [25] K.M. Lee, J.O. Jo, E.S. Lee, B. Yoo, D.H. Shin, J. Electrochem. Soc. 158 (2011) C325–C328. [26] R.O. Hussein, X. Nie, D.O. Northwood, A. Yerokhin, A. Matthews, J. Phys. D: Appl. Phys. 43 (2010) 105203. [27] V.P. Tolstoy, B. Altangerel, Mater. Lett. 61 (2007) 123–125. [28] R. Arrabal, E. Matykina, T. Hashimoto, P. Skeldon, G.E. Thompson, Surf. Coat. Technol. 203 (2009) 2207–2220. [29] D.Y. Hwang, Y.M. Kim, D.H. Shin, Mater. Trans. 50 (2009) 671–678.
Please cite this article in press as: K.M. Lee et al., J. Alloys Comp. (2014), http://dx.doi.org/10.1016/j.jallcom.2014.01.097