Formation of In Situ Composite Layer on Magnesium Alloy Surface by Casting Process

Transcription

1 Materials Transactions, Vol. 49, No. 10 (2008) pp to 2398 #2008 The Japan Institute of Metals EXPRESS REGULAR ARTICLE Foration of In Situ Coposite Layer on Magnesiu Alloy Surface by Casting Process Kazunori Asano and Hiroyuki Yoneda Departent of Mechanical Engineering, School of Science and Engineering, Kinki University, Higashiosaka , Japan In order to develop a process to obtain a functional layer on the surface of a agnesiu alloy, the coposite layer including in situ Mg 2 Si was fored on the surface by a gravity die-casting. The optiu fabrication condition to for the coposite layer and the foration echanis of the layer were clarified by observing the icrostructure, and then the wear properties of the coposite layer were also investigated. By casting the agnesiu alloy elt into the peranent old on which the slurry ainly consisting of Si particles was coated, the agnesiu alloy coposite layer in which the in situ Mg 2 Si particles were dispersed was fored. The elt teperature and old teperature required to for the coposite layer, which perfectly covers the casting, were above 1073 K and above 673 K, respectively. The coposite layer was a agnesiu alloy in which the fine Mg 2 Si particles of approxiately 40 vol% were dispersed. The thickness and hardness of the layer was about 600 and 180 HV, respectively. Under the dry sliding wear, the weight loss of the coposite layer was lower than that of the agnesiu alloy. These results lead to the conclusion that the wear resistant agnesiu alloy coposite layer in which the in situ Mg 2 Si particles were dispersed can be fored in the present process. [doi: /atertrans.mer ] (Received June 23, 2008; Accepted July 18, 2008; Published Septeber 25, 2008) Keywords: agnesiu, Mg 2 Si, coposite layer, in situ foration, casting process, icrostructure, wear 1. Introduction Fro the viewpoint of the global environent and energy savings, reduction in the weight and size of the products has been recently prooted in the industrial autoobile, aerospace and electrical fields. This trend leads to the requireents of lightweight and high strength aterials. Based on these requireents, agnesiu alloys have recently been suggested as an alternative to steel and aluinu alloys in any products because agnesiu is the lightest in the practical alloys. However, the agnesiu alloy easily defors and wears fro ipact and frictional forces due to its low hardness. In addition, a surface treatent for preventing oxidation is generally required because the agnesiu alloy is easily oxidized. These probles would be resolved by foring a coposite layer having a high hardness and corrosion resistance on the alloy surface. To iprove the wear and corrosion resistance of the agnesiu alloy, coatings of ceraics by laser-cladding 1) and spraying 2,3) have been ainly perfored. However, these techniques involve any processes such as a pretreatent, leading to a high cost. Moreover, a recycling proble reains when the coat includes the eleents different fro the ones usually added to coercial agnesiu alloys. Mg 2 Si, an interetallic copound, has a good oxidation resistance, high elting point and high hardness. 4) Therefore, the foration of Mg 2 Si on the agnesiu alloy surface would iprove its corrosion resistance and wear resistance. The agnesiu alloy coated with Mg 2 Si can be reused as the AS alloy (Mg-Al-Si alloy), which is classified by ASTM specifications; therefore, it would be easy to recycle. In addition, Mg 2 Si can be in situ fored by a reaction between agnesiu and silicon. Kondoh et al. 5,6) fabricated a agnesiu alloy coated with Mg 2 Si by sputtering. If the in situ coposite layer including Mg 2 Si can be fored by a casting process, it is industrially beneficial because it leads to a considerable reduction in the fabrication processes and cost. However, there are no reports on the icrostructure and properties of the coposite layer including Mg 2 Si fabricated by the casting process. In the present study, the coposite layer including Mg 2 Si fabricated by the casting process was fored to develop an in situ foration process of the functional layer on the agnesiu alloy surface. The icrostructure, foration echanis and wear properties of the coposite layer were then investigated. 2. Experiental Procedure The AZ91D alloy with its cheical coposition shown in Table 1 was used as the cast alloy. Pure silicon (Si) particles (99.9 ass%si) were used as the starting aterial to for Mg 2 Si by the reaction with agnesiu in the alloy elt. Figure 1 is a SEM icrograph of the Si particles. The average size of the Si particles easured by an iage analyzer using 10 fields in the SEM icrographs was 4. Figure 2 shows a scheatic illustration of the old. The old has 5 cavities in a radial anner and 5 speciens can be obtained per one casting process. To obtain a plate specien Table 1 Cheical coposition of AZ91D alloy (ass%). Al Zn Mn Si Cu Fe Mg Bal. Fig. 1 5µ SEM icrograph of Si particles.

2 Foration of In Situ Coposite Layer on Magnesiu Alloy Surface by Casting Process 2395 Vent hole Sprue( φ 19) Pin ( φ 5 5) Coposite layer Sliding direction Disc (SK4, φ30 5) Coat Fig. 4 Scheatic illustration of wear test. Bolt hole 30 5 Fig. 2 Scheatic illustration of old. (a) T 4 p =643K (b) T p =673K Fig. 5 Macrostructure of specien surface (elt teperature; T ¼ 1073 K). T p represents the old teperature. Fig. 3 10µ SEM icrograph of cross-section of coat. with a coposite layer containing Mg 2 Si, the slurry including Si particles was coated on the upper plane of the cavities, followed by drying. The AZ91D alloy elt is poured fro the sprue, and then the coposite layer containing the Mg 2 Si is obtained by the reaction between the elt and Si in the coat. A vent hole was situated at the end of each cavity to prevent blow holes fro foring in the specien. The slurry was fabricated by ixing distilled water, Si particles and silica (SiO 2 ) sol to iprove the adhesion of the coat, and then brushed on the cavity in the old. In order to deterine the coating conditions to obtain the optiu coat, a preliinary experient was perfored. As a result, the optiu conditions were as follows: Si content in the slurry: 40 ass%; SiO 2 sol content: 7 ass%; old teperature: 323 K; and coat thickness: 0.5. Figure 3 shows a SEM icrograph of the cross-section of the coat. One hundred gras of the AZ91D alloy elt was poured into the coated old, and then pressure of 4 MPa was quickly applied by a plunger and aintained until the solidification was coplete. The elt teperatures ranged fro 1003 to 1123 K, and the old teperature ranged fro 473 to 723 K. The acrostructure of the surface and cross-section of the speciens and icrostructure of the coposite layer were observed. X-ray diffractoetry (XRD) and electron probe icroanalysis (EPMA) were perfored to identify the coposite layer. Hardness of the coposite layer was easured by a Vickers hardness tester under a load of 1.47 N. Dry sliding wear tests were perfored using a pin-on-disc type wear apparatus, as scheatically illustrated in Fig. 4. The pin test piece having a 5 diaeter and 5 length was achined fro the specien. JIS-SK4 disc test piece having a 30 diaeter and 5 thickness was used as the counterface. The disc test pieces were quenched and tepered to obtain the hardness of 700 HV. The wear tests were carried out at a sliding distance of 500, at a load of 19.6 N, and at sliding velocities of 0.1, 0.3 and 0.5 s 1. The weight losses were calculated fro the differences in weight of the test piece easured before and after the test every Results and Discussion 3.1 Macrostructure and icrostructure Figure 5 shows the acrostructure of the specien surface (elt teperature: T ¼ 1073 K). When the old teperature (T p ) was 643 K, a dark substance, probably fored by the reaction between the coat and the alloy elt, was partially observed on the surface, as shown in Fig. 5(a). However, the product coprised only half of the surface under this condition. When the T was 1073 K and T p was 673 K, the product fored over the entire surface, as shown in Fig. 5(b). When the T was below 1073 K and T p was below 643 K, no products were fored on the surface. On the other hand, the product covered the entire surface when T was above 1073 K and T p was above 673 K. Therefore, the speciens fabricated by setting T at 1073 K and T p at 673 K were selected to exaine the icrostructure and the wear properties. Figure 6(a) shows the acrostructure of the vertical crosssection of a specien (T ¼ 1073 K, T p ¼ 673 K), indicating that a layer is fored on the specien s surface by the reaction. Figure 6(b) is an optical icrograph of the vertical cross-section near the layer of the specien, showing that the thickness of the layer is about 600. The enlarged views of the icrostructure enclosed by squares in Fig. 6(b) are shown in Figs. 6(c) and 6(d), indicating that the layer consists of a agnesiu alloy atrix and fine dispersed particles. Figure 6(d) shows the typical icrostructure of the layer. The area fraction of the particles easured by an iage analysis using 10 fields in the icrographs was 40%. Figure 6(c) is the icrostructure near the boundary between the layer and

3 2396 K. Asano and H. Yoneda (a) Layer (b) Intensity(arb. units) Mg Mg 17 Al 12 Mg 2 Si (c) (d) 150µ Fig. 8 Diffraction Angle, 2θ X-ray diffraction pattern of coposite layer. 30µ 20µ Fig. 6 Macrostructure and icrostructure of vertical cross-section of specien (T ¼ 1073 K, T p ¼ 673 K). alloy part, showing that the particles gradually decreased towards the alloy part near the boundary. Figure 7 is SEM and X-ray iages of the layer, showing that the particles dispersed in the layer consists of Mg and Si. Figure 8 is the XRD pattern of the layer, showing that Mg, Mg 17 Al 12, and Mg 2 Si were detected. Based on the results shown in Figs. 7 and 8, it can be concluded that the particles are Mg 2 Si, while no residual Si exists in the layer. Subsequently, the shape and size of the Mg 2 Si were investigated in detail. Figure 9 is a SEM icrograph of the Mg 2 Si particles extracted fro the layer by 30% nitric acid, showing that the average size of the Mg 2 Si particles in the layer is less than 5 and any of the appear to aggloerate with each other. These results leads to the conclusion that the layer obtained in the present process is the agnesiu alloy coposite layer in which fine Mg 2 Si particles are dispersed. Fig. 9 SEM icrograph of Mg 2 Si particles extracted fro coposite layer. 3.2 Foration echanis of coposite layer Based on these results, the foration echanis of the Mg 2 Si in the present study is discussed. Mg 2 Si is fored by the reaction between Mg in the alloy and Si, as designated by eq. (1): 7) 2Mg þ Si! Mg 2 Si ð1þ A binary equilibriu phase diagra of the Mg-Si syste 8) shows that the iniu teperature to for Mg 2 Si is 912 K. As shown in Fig. 5, in the present study, the coposite layer covered the entire surface of the specien when T was above 1073 K and T p was above 673 K. Although a previous report showed that this reaction occurs at approxiately 800 K, at which both the Mg alloy and Si are in the solid state, 7) higher teperatures of the elt and old would be 5µ SEI Mg Si O 3µ Fig. 7 SEM and X-ray iages of coposite layer.

4 Foration of In Situ Coposite Layer on Magnesiu Alloy Surface by Casting Process 2397 Vickers Hardness Coposite layer AZ91D alloy Distance fro Surface, d / Weight Loss, w /g Coposite layer AZ91D Disc(coposite layer) Disc(AZ91D) Fig. 10 Vickers hardness of specien plotted versus distance fro surface. The error bars indicate the scatter in easured values. required in the present study because the solidification rate of the elt by the gravity die-casting is high. Since the coat thickness is approxiately equal to the coposite layer thickness, the coat was infiltrated with the agnesiu alloy elt followed by the reaction between the elt and Si with subsequent foration of the coposite layer. The fine dispersion of the Mg 2 Si particles is probably due to the suppression of the coarsening of the particles based on the rapid cooling in the peranent old. In addition, it is reported that the Mg 2 Si is fored by the reaction between Mg and SiO 2, as designated by eq. (2): 5) 4Mg þ SiO 2! Mg 2 Si þ 2MgO Since the slurry used in the present study includes a SiO 2 sol, MgO ight exist in the coposite layer. However, as shown in Figs. 7 and 8, MgO was not detected in the coposite layer. This indicates that the aount of MgO in the layer is inor even though it exists. 3.3 Hardness and wear properties of coposite layer Figure 10 shows the Vickers hardness plot of a specien versus distance fro the surface. The error bars in the figure indicate the scatter in the easured values. The hardness of the coposite layer was about 180 HV, while that of the AZ91D alloy part was about 80 HV. This indicates that the dispersion of the Mg 2 Si particle having a hardness of 460 HV 4) increased the hardness of the specien surface. As shown in Fig. 6(c), since the particles gradually decrease near the boundary, the hardness siultaneously decreases in the sae area: The hardness at the boundary was about 120 HV. Figure 11 shows the weight loss of the AZ91D alloy, coposite layer and their counterfaces plotted versus the sliding distance ( ¼ 0:3 s 1 ). Although the weight loss of all speciens increased as the sliding distance increased, the weight loss of the coposite layer was lower than that of the AZ91D alloy. This indicates that the dispersion of the Mg 2 Si particles decreased the weight loss of the alloy. The weight loss of the disc counterface (SK4) cobined with the coposite layer was alost the sae as that cobined with the AZ91D alloy; the dispersion of the Mg 2 Si particles only slightly affected the weight loss of the SK4 counterface. The wear rate calculated fro the slope of the linear part of the weight loss vs. sliding distance curves was 11.1 g/ for the coposite layer, while the wear rate was 16.7 g/ for the AZ91D alloy. This result also indicates that the Mg 2 Si particles offer an ð2þ effective iproveent in the wear resistance of the agnesiu alloy. Figure 12 is a plot that shows the weight loss of the AZ91D alloy, coposite layer and their disc counterfaces (SK4) versus the sliding velocity. At all sliding velocities, the weight loss of the coposite layer was lower than that of the AZ91D alloy, and the weight loss of the counterface cobined with the coposite layer was alost the sae as that cobined with the AZ91D alloy. 4. Conclusions Sliding Distance, d s / Fig. 11 Weight loss of AZ91D alloy, coposite layer and disc counterfaces (SK4) plotted versus sliding distance ( ¼ 0:3 s 1 ). Weight Loss, w /g Coposite layer AZ91D Disc(coposite layer) Disc(AZ91D) Sliding Velocity, v /s 1 Fig. 12 Weight loss of AZ91D alloy, coposite layer and disc counterfaces (SK4) plotted versus sliding velocity. The coposite layer containing in situ Mg 2 Si particles was fored on the AZ91D alloy surface by a casting process. The icrostructure, foration echanis and wear properties of the coposite layer were investigated. The following results were obtained. (1) By casting the AZ91D alloy elt into the peranent old on which a slurry ainly consisting of Si particles was coated, the coposite layer in which the in situ Mg 2 Si particles were dispersed was fored. (2) The coposite layer was a agnesiu alloy in which the fine in situ Mg 2 Si particles were dispersed. The hardness of the layer was about 180 HV, which was higher than that of the agnesiu alloy. (3) Microscopically, the Mg 2 Si particles gradually decreased towards the alloy part near the boundary. No residual Si was observed in the coposite layer.

5 2398 K. Asano and H. Yoneda (4) Under the dry sliding wear at a load of 19.6 N and at sliding velocities ranging fro 0.1 to 0.5 s 1, the weight loss of the coposite layer was lower than that of the AZ91D alloy. This result indicates that the foration of the coposite layer iproved the wear resistance of the alloy. REFERENCES 1) Y. Gao, C. Wang, H. Liu and M. Yao: Appl. Surf. Sci. 253 (2007) ) M. Parco, L. Zhao, J. Zwick, K. Bobzin and E. Lugscheider: Surf. Coat Technol. 201 (2007) ) M. Parco, L. Zhao, J. Zwick, K. Bobzin and E. Lugscheider: Surf. Coat Technol. 201 (2006) ) G. Froeyer, S. Beer and K. Oldenburg: Z. Metallk. 85 (1994) ) K. Kondoh, R. Tsuzuki, W. Du and S. Kaado: Materia Japan 43 (2004) ) T. Yaaguchi, T. Serikawa, M. Heni, H. Oginua and K. Kondoh: Mater. Trans. 47 (2006) ) R. Tsuzuki, K. Kondoh, W. Du, T. Aizawa and E. Yuasa: Mater. Sci. Foru (2003) ) T. B. Massalski: Binary Alloy Phase Diagras (ASM International, 1990) p