High Speed High Quality Friction Stir Welding of Austenitic Stainless Steel

Transcription

1 , pp High Speed High Quality Friction Stir Welding of Austenitic Stainless Steel Takeshi ISHIKAWA, 1) Hidetoshi FUJII, 2) Kazuo GENCHI, 1) Shunichi IWAKI, 1) Shigeki MATSUOKA 1) and Kiyoshi NOGI 2) 1) Tokyu Car Corporation, 3-1 Ohkawa, Knazawa-ku, Yokohama Japan. 2) Joining and Welding Research Institute, Osaka University, 11-1 Mihogaoka, Ibaraki, Osaka Japan. (Received on February 3, 2009; accepted on February 26, 2009; originally published in Tetsu-to-Hagané, Vol. 94, 2008, No. 11, pp ) Although several studies on the friction stir welding of high temperature materials have recently been reported, their practical use has not yet occurred due to some problems to be solved such as decreases in the corrosion resistance and joint efficiency during a high-speed joining. In this study, the effect of the welding speed on the joint efficiency was investigated in detail. As a result, it has been clarified that at a rotational speed of 600 rpm, the friction stir welding of 304 austenitic stainless steel is possible up to the joining speed of mm/min, and the tensile strength of the joint exceeds that of the parent material up to the joining speed of mm/min. In addition, the corrosion resistance is significantly improved at the higher welding speeds. No rust was observed during the salt spray testing of the mm/min joint. Thus, an increase in the welding speed can improve productivity and the product quality by decreasing corroded region. These results are expected to extend to an actual application to products by further improving the technology. KEY WORDS: tensile strength; high-speed friction stir welding; joint efficiency; hardness distribution; macro structure; micro structure; corrosion resistance. 1. Introduction Friction stir welding (FSW) is a new welding process that was invented by TWI in 1991, and a kind of solid-state welding using the friction heat generated between the tool and the material. 1) This process features the fact that the highest temperature during the friction stir welding is lower than the melting point. Therefore, this process has some merits such as a low distortion and high strength due to a grain refinement. Especially, practical applications for the light alloys like the aluminum alloys have been developed, and it has already been applied to the bodies of cars, rolling stock, ocean vessels and airplanes. 2 6) On the other hand, for the friction stir welding of high temperature materials, it is necessary that the proper tool is developed for enduring the high temperature during the friction stir welding. 7) Recently, various tool materials such as polycrystalline cubic boron nitride(pcbn), 8) W alloys 9,10) and WC Co 11) have been tested. A number of authors have shown the possibility of friction stir welded high temperature materials of Mo, 12) Ti, 13,14) IF steel 15) and carbon steel, 16 19) etc. There are some reports 20 29) on the friction stir welding of austenitic stainless steel. Though these research studies mainly reported the characteristics of corrosion 20 22) or the microstructures and mechanical properties ) There are only few reports of the welding condition effects on the joint strength. 28,29) In this study, the friction stir welding of 304 austenitic stainless steel is carried out in various welding conditions, by changing the heat input based on the change in the welding speed. This study s objective is to increase the joint strength and corrosion resistance by high speed welding. 2. Experimental Procedures The base material is a 160 mm long, 35 mm wide and 1.5 mm thick 304 austenitic stainless steel plate. Its chemical compositions and mechanical properties are shown in Table 1. The joint butt interface was machined and cleaned with acetone. Table 1. Chemical composition and mechanical properties of 304 stainless steel (mass%) ISIJ

2 Fig. 1. Designated of welding tool. Fig. 2. Appearance of a joint obtained at welding speed of mm/min. They were butt-welded by friction stir welding using a load control machine. The welding tool was tilted 3 degrees from the plate normal direction. A water-cooled holder was installed around the tool and argon gas was used at the flow rate of 30 L/min as the shielding gas to prevent oxidation of the materials. Although a straight probe tool without threads can also produce sufficient joints, 30) in this study, a tool with a tapered probe was used in order to extend the welding tool life. The tool dimensions and shapes are shown in Fig. 1. The welding tool rotation speed was a constant clockwise 600 min 1, and the welding speed was varied between 60 and mm/min in order to control the heat input. After the welding, the joints properties were evaluated on the basis of the surface appearance, the tensile test, the salt spray testing, the hardness test, and macro and microstructure observations. Three tensile test specimens were used prepared by cutting in the direction vertical to the welding. The tensile tests were carried out without machine finishing the joints, namely the surface had ripples and flash. The crosshead speed was 10 mm/min during the tensile tests at room temperature, so that the strain rate was 40% by JIS G The salt spray testing was performed using 5% salt water and was maintained at 95% humidity and 35 C. The vickers hardness distribution of the weld was measured at 0.5 mm from the top surface and bottom surface on the cross section perpendicular to the welding direction, across the stir zone, HAZ and the base metal materials with a 2.94 N load for 10 s. The macro observation was performed on cross sections perpendicular to the welding direction after they were ground with aluminum abrasives and etched with 10% oxalate solution using a microscope. 3. Results and Discussion Fig. 4. Fig. 3. Tensile strength of 304 stainless steel joints. Appearance of mm/min joints after tensile tests Joints Properties The friction stir welding of the 304 stainless steel was successfully performed at speeds of 60 to mm/min. Figure 2 shows the appearance of the 304 stainless steel joint welded at the high speed of mm/min. No defects are observed. This gap is equal to the distance when the tool moves while it rotates at 360 degrees, 31,32) namely, the value of the welding speed divided by the rotational speed of the tool. For example, when the welding speed is mm/min at the rotational speed of 600 min 1 as in Fig. 2, the value is 2 mm. Although a slight oxidation is observed on the surface, the oxidation can be prevented by expanding the shielding range. Such a high speed friction stir welding has been reported for aluminum alloys. 31) On the other hand, this is first time for use with ferrous materials to the best of the author s knowledge. Figure 3 shows the tensile strength of the joints welded at different welding speeds. Stable joints are obtained at all welding speeds. In particular, the samples from 600 to mm/min are fractured in the base material. In other words, the joint efficiency is 100%, and the joint strength exceeds that of the base material. Figure 4 shows the appearance of the joints welded at mm/min after the tensile test. At this welding speed, no defect such as a kissing bond is observed. This high welding speed for the successful joining of austenitic stainless steel has never been reported, and is an exceptional discovery. The fractured position of the tensile test specimen is in the stir zone (Hereafter, the stirred zone is called the SZ.) at 420 mm/min or less and at mm/min. Because the thermal conductivity of aluminum alloys is higher than that of stainless steels, the frictional heating between the tool and the material transmits more widely for 2009 ISIJ 898

3 the aluminum alloys. Also, because the proof strength at a high temperature is low, the plastic flow generally occurs more easily for the aluminum alloys. As a result, the stirred region broadens, and the shape of the stirred zone becomes trapezoidal. On the other hand, stainless steel has a low thermal conductivity. Especially, the SZ narrows when the friction stir welding is performed at a high welding speed, as in this study. Thus, the plastic flow does not easily occur, when the friction stir welding is performed for a high melting point material whose proof strength is high. Accordingly, high quality joints are not easily obtained, if the tool with the same size and shape used for aluminum alloys is used for the stainless steel. In this study, a large probe diameter is used, when compared to that for the light alloy material in order to increase the size of the SZ. As a result, defects form in the advancing side (Hereafter, the advancing side is called AS.) could be completely removed. In this case, the strength of the SZ exceeds that of the base material, though the tensile test specimens are fractured in the SZ because it becomes slightly thinner than the base material due to the inclination of the tool. In conclusion, the conditions of the tool rotation of 600 min 1 and the welding speed of mm/min are the optimal welding conditions at which a defect free joint is obtained with the minimum heat input among the various welding conditions of this study. Fig. 5. Cross sections of 304 stainless steel joints welded at different welding speeds. Fig. 6. Corrosion test results by salt spray test Microstructure of Stainless Steel Joints Figure 5 shows the typical macrostructure obtained at the various welding speeds. As shown in Fig. 5(a), at 60 mm/min, the heat input is too high, and accordingly, the boundary between the SZ region and the base material region are not easily identified, because the SZ grains grow, though the grain sizes are smaller than those of the base material. Moreover, a region, which is easily corroded, is formed in the AS region of the SZ, though no defect is formed. The corrosion resistance of the 304 austenitic stainless steel friction stir welding joints has been described by Park et al. 20,21) The easily corroded region is observed in the SZ, which is the same as that for the PCBN tool. According to Park et al., 22) there are many precipitations at the grain boundary and inside the grains. This region is identified as the sigma phase that includes a lot of stacking faults inside. However, when the thickness of the base material is as thin as 2 mm, the sigma phase is not observed, because the cooling rate is higher. It generally takes a longer time to precipitate the usual sigma phase. They deduced that during the friction stir welding, the diffusion of the chromium at the grain boundary and in the grains is promoted due to the dynamic recrystallization in the AS, and accordingly, the sigma phase is formed in a very short time. 21) The possible reason why the corrosion region is formed in spite of the thin 1.5 mm plate used in this study is that the maximum temperature is higher than that in Park s experiment because the welding speed of 60 mm/min is extremely slow. As shown in Fig. 5(b), at 600 mm/min, a clear boundary region between the SZ and the base material is observed, and a zonal layer is also observed which expands from the flash to the AS of the SZ. As shown in Fig. 5(c), at mm/ min, no defects are observed and an excellent SZ region is obtained. The area of the zonal layer also significantly decreases when compared with Fig. 5(a), the 60 mm/min area. This zonal layer corresponds to the above-mentioned corrosion region. It was found that the range of the zonal layer decreases with the increasing welding speed. Therefore, the corrosion resistance is improved by increasing the welding speed. Figure 6 shows the surface appearance of the joints after the corrosion test. Much rust is observed on the 180 mm/min sample surface. No rust is formed on the mm/min sample. High speed welding is effective for increasing the joint corrosion resistance. On the other hand, in the AS region in Fig. 5(d) at mm/min, a tunnel defect is seen due to heat input shortage. Therefore, as shown in Fig. 3, the tensile strength decreases and it breaks in the SZ. Figure 7 shows the microstructures of the SZ for the various welding speeds. As shown in Fig. 7(b), when the welding speed is extremely slow, for example, at 60 mm/min, the microstructure is not uniform due to the grain growth ISIJ

4 ISIJ International, Vol. 49 (2009), No. 6 Fig. 8. Effect of FSW speed on grain size. caused by an excessive heat input. As shown in Figs. 7(c) and 7(d), at 600 mm/min and mm/min, the grains are refined, and an equiaxial dynamic recrystallization structure can be observed, rather than a non-uniform structure which is often observed during fusion welding. The grain size is smaller than that of the base material, causing the tensile strength to be higher than the base material. It can be judged that the SZ exceeds that of the base material, because the fracture position is the base material in the tensile test. This result is aligned with the microstructures. As shown in Fig. 7(e) at mm/min, the grain becomes further smaller. However, the stirring shortage and the tunnel defect due to the heat input shortage are observed in the SZ32) in Fig. 7(f). Figure 8 shows the effect of the welding speed on the average grain size in the SZ. The average grain size at mm/min is minimal at 12 m m, which is about 1/3 that of the base materials i.e., 36 m m. As the welding speed increases, the grain size becomes smaller. Although the grain size of 12 m m obtained in this study is similar to that reported for austenitic stainless steels in previous studies,20,26) it is much bigger than the values previously reported for Al alloys and carbon steels.9,11) As the stacking fault energy of austenitic stainless steels is low, the recovery does not occur easily. Based on this fact, the grain size should be smaller than that of Al alloys having large stacking fault energies and the bcc carbon steel. A detailed examination is necessary for this result, although it is postulated that the grains grew because the thermal conductivity of the SUS304 is smaller, the cooling rate after the friction stir welding was low and thus the high temperature was maintained for a comparatively long time Hardness Distribution of Stirred Zone Figure 9 shows the hardness distributions for several welding speeds. The hardness distribution of 0.5 mm from the top surface is not significantly dependent on the welding speed. In particular, at 60 mm/min, the hardness of the stir zone is almost same as that of the base material, because grain growth occurs due to the sufficient heat input from the shoulder. On the other hand, as shown in Fig. 9(b), the hardness of 0.5 mm from the back surface is slightly higher than that near the top surface, indicating that the obtained temperature was lower. However at 60 mm/min, there is no significant difference in the hardness, when compared to the top surface and the base metal, because the heat input spreads to the back of the joint, which is known from the results of the section s macro observation results, indicating that the grain growth sufficiently occurs. At 600 mm/min or more, on the other hand, the temperature near the back surface is much lower than that of the top surface. The in- Fig. 7. Microstructures of 304 stainless steel stir zone formed at different welding speeds ISIJ 900

5 Acknowledgements A part of this research is a result of the grant for steel research promotion from the Iron and Steel Institute of Japan. One of the authors (H.F.) is grateful for the financial support of the Global COE program. REFERENCES Fig. 9. Hardness distribution at different FSW speeds. crease in the dislocation density and the grain refinement due to the stirring action by the probe increases the hardness of the SZ. 4. Conclusion Friction stir welding is performed using 304 austenitic stainless steel, a typical austenitic stainless steel, and the influence of the welding speed on the mechanical properties is investigated. As a result, the following conclusions were obtained. (1) The friction stir welding of 304 austenitic stainless steel is possible at the welding speeds of mm/min. This is the first successful trial at the welding speed of mm/min or more for the friction stir welding of ferrous materials. (2) Under the welding conditions of this study, defectfree friction stir welding joints are obtained at a welding speed of mm/min or less, and the joint efficiency is 100%. A zonal layer which is easily corroded is formed in the stir zone at the slow welding speed of 600 mm/min or less. However, the size of this region decreases with the increasing welding speed. Accordingly, the increase in the welding speed is effective for improving the corrosion resistance of the joints. (3) The grain size is smaller and the hardness of the SZ increases as the welding speed increases. This tendency of the hardness distribution is more significant near the bottom of the sample. 1) W. M. Thomas, E. D. Nicholas, J. C. Needham, M. G.. Murch, P. T. Smith and C. J. Dawes (TWI): International Patent Application No. PCT/GB92/02203 and GB Patent Application No , (1991). 2) H. Okamura, K. Aota, Y. Sakamoto, M. Esumi and K. Ikeuti: Q. J. Jpn. Weld. Soc., 19 (2001), No. 3, ) M. Enomoto: J. Light Met. Weld. Constr., 36 (1998), No. 2, 75. 4) M. Kumagai and S. Tanaka: J. Light Met. Weld. Constr., 39 (2001), No. 1, 22. 5) H. Okamura: J. Jpn. Weld. Soc., 69 (2000), No. 7, ) M. Enomoto: Proc. 3rd Int. Symp. Friction Stir Welding, TWI, UK, (2001), ) M. Inagaki, S. Hirano, K. Okamoto, T. Odakura and K. Masutomi: Bull. Iron Steel Inst. Jpn., 7 (2002), No. 10, ) M. Collier, R. Steel, T. W. Nelson, C. Sorensen and S. Packer: Mater. Sci. Forum, 426 (2003), ) A. P. Reynolds, W. Tang, M. Posada and J. DeLoach: Sci. Technol. Weld. Joining, 8 (2003), No. 6, ) T. J. Lienert, W. L. Stellwag, Jr. B. B. Grimmett and R. W. Warke: Weld. J., 82 (2003), 1-S. 11) H. Fujii, L. Cui, N. Tsuji, M. Maeda, K. Nakata and K. Nogi: Mater. Sci. Eng., A429 (2006), ) H. Fujii, H. Kato, K. Nakata and K. Nogi: Proc. 6th Int. Symp. Friction Stir Welding, TWI, UK, (2006), 2A-2-1-2A ) A. P. Reynolds, E. Hood and W. Tang: Scr. Mater., 52 (2005), No. 6, ) H. Fujii, H. Kato, K. Nakata and K. Nogi: Ceram. Trans., 198 (2007), ) H. Fujii, R. Ueji, Y. Takada, N. Tsuji, K. Nakata and K. Nogi: Mater. Trans., 47 (2006), ) W. M. Thomas, P. L. Threadgill and E. D. Nichoras: Sci. Technol. Weld. Joining, 4 (1999), ) R. Ueji, H. Fujii, L. Cui, A. Nishioka, K. Kunishige and K. Nogi: Mater. Sci. Eng., A423 (2006), ) L. Cui, H. Fujii, N. Tsuji, K. Nakata, K. Nogi, R. Ikeda and M. Matsushita: ISIJ Int., 47 (2007), No. 2, ) L. Cui, H. Fujii, N. Tsuji and K. Nogi: Scr. Mater., 56 (2007), ) S. H. C. Park, Y. S. Sato, H. Kokawa, K. Okamoto, S. Hirano and M. Inagaki: Scr. Mater., 49 (2003), ) S. H. C. Park, Y. S. Sato, H. Kokawa, K. Okamoto, S. Hirano and M. Inagaki: Scr. Mater., 51 (2004), ) S. H. C. Park, Y. S. Sato, H. Kokawa, K. Okamoto, S. Hirano and M. Inagaki: Sci. Technol. Weld. Joining, 10 (2005), No. 5, ) Y. S. Sato, T. W. Nelson, C. J. Sterling, R. J. Steel and C.-O. Pettersson: Mater. Sci. Eng., A397 (2005), ) X. K. Zhu and Y. J. Chao: J. Mater. Process. Technol., 146 (2004), ) J.-H. Cho, D. E. Boyce and P. R. Dawson: Mater. Sci. Eng. A, 398 (2005), ) A. P. Reynolds, W. Tang, T. Gnaupel-Herold and H. Prask: Scr. Mater., 48 (2003), ) C. D. Sorensen, T. W. Nelson and S. M. Packer: Proc. 3rd Int. Symp. Friction Stir Welding, TWI, UK, (2001), ) T. Ishikawa, H. Fujii, K. Genchi, C. Ling, S. Matsuoka and K. Nogi: Q. J. Jpn. Weld. Soc., 24 (2006), No. 2, ) T. Ishikawa, H. Fujii, S. Iwaki, S. Matsuoka and K. Nogi: Proc. 6th Int. Symp. Friction Stir Welding, TWI, UK, (2006), 2A-4-1-2A ) H. Fujii, L. Cui, M. Maeda and K. Nogi: Mater. Sci. Eng., A419 (2006), ) H. J. Liu, M. Maeda, H. Fujii and K. Nogi: J. Mater. Sci. Lett., 22 (2003), ) Masatsukakuhansetsugou, Friction Stir Welding, Japan Welding Society, Tokyo, (2006), ISIJ