Friction-Stir-Welded Lap Joint of Aluminum to Zinc-Coated Steel

Transcription

1 [ 23 p (2005)] Friction-Stir-Welded Lap Joint of Aluminum to Zinc-Coated Steel by ELREFAEY Ahmed**, TAKAHASHI Makoto*** and IKEUCHI Kenji*** This paper has dealt with the feasibility of friction-stir welding a zinc-coated steel plate to a pure aluminum plate in a lap joint configuration (the aluminum plate was top and the steel bottom). The friction stir welding was carried out at rotation speeds of s 1 and travel speeds of mm/s. It was found that the performance of the joint depended strongly on the depth of the pin tip of the FSW tool relative to the steel surface; when the pin depth did not reach the steel surface the joint showed quite weak fracture loads, while the penetration of the pin tip to a depth of 0.1 mm under the steel surface significantly increased the joint strength. The joint strength tended to increase with rotation speed and slightly decrease with the increase in the traveling speed. As compared with the similar joints of an aluminum to a steel without Zn-coating, joints of aluminum to the Zn-coated steel exhibited high strength at all bonding parameters employed in this investigation, which suggested the beneficial effect of the Zn coating on the bond strength. Key Words: Friction stir welding, Lap joint, Aluminum, Zinc-coated steel, Intermetallic compound 1. Introduction 2. Experimental Details The use of zinc-coated steels for the automotive body construction has increased significantly in the last 2 decades to enhance the durability of vehicle structures. The automotive industries are currently working to develop lighter and more fuelefficient vehicles. A significant proportion of this effort is currently being directed toward the substitution of aluminum for steel in the body structure. Aluminum is considerably lower in strength and stiffness than steel, and the design of the spaceframe coupled with the use of thicker material sections, successfully compensates for this 1). Nevertheless, the replacement of material presents a vital challenge with respect to the methods of joining to be used for fabrication in volume production. Over the last decade, friction stir welding (FSW) has offered excellent welding quality to the joining of aluminum, magnesium 2 3), titanium 4), copper 5), and Fe alloys 6 8). Recently, some trials have been made to join the dissimilar materials, for examples dissimilar Al alloys 9) and aluminum to steel joints 10 13). However, most of the FSW efforts to date have not involved joining aluminum to zinc-coated steel. From an industrial point of view, there seems to be considerable interest in extending the process to this joint. Therefore, this research has been aimed at investigating the performance of aluminum-to-zinc coated steel (Al/Zn-coated steel) lap joint by friction stir welding and metallurgical factors controlling the performance. The materials used were a plate of commercially pure aluminum A1100H mm thick and a plate of low carbon Zncoated steel 1.0 mm thick. The steel had Zn coating of 10 m thickness. The chemical compositions of base metals are shown in Tables 1 and 2. The microstructures of the base metals are shown in Fig. 1. The aluminum base metal presented grains elongated in the rolling direction, and the steel base metal showed ferritic structure due to its very low carbon content. The base metals were cut into specimens of 250 mm x 100 mm for welding. Lap welding was made by a FSW machine as schematically shown in Fig. 2. Rotation and travel speeds of the tool employed are listed in Table 3. The depth of the pin tip from the upper surface of the aluminum plate was fixed at 2.0 or 2.1 mm (0.0 and 0.1mm from the surface of the zinc-coated steel plate). The tool of steel SKD61 was comprised of a shank, shoulder and pin as shown in Fig. 3. The tool axis was tilted by 3 with respect to the vertical axis of the plate surface. The FSW tool, fixed to the holder, was slowly pushed into the aluminum plate to the specified pin depth and then forcibly traversed along the joint until the end of the weld was reached. The welding tool was then Table 1 Chemical composition of the aluminum base metal 1100H24 (mass %). * ** *** Student Member, Graduate Student, School of Engineering of Osaka University Member, Joining and Welding Research, Institute Osaka University Table 2 Chemical composition of the Zncoated steel base metal (mass %).

2 Table 3 Welding parameters. Fig. 1 Microstructures of the base metals of aluminum (a) and Zn-coated steel (b). Fig. 2 Schematic illustration of the FSW of an aluminum plate to a Zn-coated steel plate. Fig. 4 Schematic view of the specimen used for peel test. Fig. 3 Schematic illustration of the tool used for FSW. retracted while the tool continued to turn. The surface for the observation of microstructure was etched by 3% Nital to reveal the steel microstructure and subsequently by 1% HF aqueous solution to reveal the aluminum microstructure. The microstructure was observed with an optical microscope and SEM (Scanning Electron Microscope) for closer observations. A peel test was employed to estimate the fracture load of the obtained joint. The schematic view of the specimen for the peel test is shown in Fig. 4. Temperature measurements at positions close to the zinccoated steel surface (0.3 mm down from the surface) were carried out by using a thermocouple (K-type) percussion-welded to the bottom of holes drilled from the back surface of the steel specimen. X-ray diffraction analyses were carried out to identify phases present on the fracture surfaces of joint after peel tests.

3 188 ELREFAEY Ahmed et al. Friction-Stir-Welded Lap Joint of Aluminum to Zinc-Coated Steel 3. Experimental Results and Discussion 3.1 Characteristics features of joints The macroscopic views of traverse sections of joints welded at pin depth of 2.1 mm are shown in Fig. 5. It should be noted that a great change in the steel microstructure was observed in the area below the pin tip when the rotation speed was not less than 25.0 s 1. Meanwhile, at a rotation speed of 16.7 s 1 the change in steel microstructure at the interface was quite small, but a remarkable change was observed in the aluminum stir zone to the direction of advancing side. The figure showed no clear onion ring structure or thermo-mechanically affected zone in contrast to the general stir zone reported in many previous papers 14 16) about FSW of aluminum alloys. A few characteristic regions can be identified in both aluminum side and zinc-coated steel side of the joints as shown in Fig. 6 (a). The microstructure of the aluminum corresponding to the stir zone was characterized by equiaxed fine grains as shown in Figs. 6 (b) and 6 (c). The grain size near the bond interface (area II) was slightly coarser than the upper surface of aluminum (area I), suggesting the effect of heat generated by friction between the pin tip and Zn-coated steel on the grain size of the aluminum. Several authors suggested that the equiaxed fine grain in the stirred zone was formed through the dynamic recrystalization followed by the static grain growth for a short period during the cooling process 17 19). Between the stir zone and the base metal, narrow heataffected-zones (areas III and IV) were observed which were characterized by slightly coarse grain size as shown in Figs. 6 (d) and 6 (e). Area V in the zinc-coated steel side was characterized by much finer equiaxed-grains than the base metal as shown in Fig. 6 (f). The average grain size of this zone was approximately 3 m at a rotation speed of 25.0 s 1 at a traveling speed of 3.3 mm/s, while it was 7 m at a rotation speed of 41.7 s 1. The grain size was reduced with an increase in traveling speed from 3.3 mm/s to 5.0 mm/s. In the transition zone between area V and zinc-coated steel substrate, as shown in Fig. 6 (g), a coarser grain structure representing the Zn-coated steel HAZ was observed (area VI). The very fine grain size in the steel close to the Al/Zn-coated steel interface (area V) can be attributed to the recrystalization of the area deformed heavily by the friction with the rotating pin, since the maximum temperature measured with thermocouples, at points 0.3 mm down from the steel surface was about 779 K, much lower than a transformation of the steel. The size of the recrystalized grains depends mainly on the strain which took place in the grains and the temperature relative to the melting point of the material. It can be expected that the deformation of the steel fine grain zone was smaller than that of aluminum, since the tool pin tip does not reach this zone. Therefore, the much smaller grains size of the steel fine grain zone can be attributed to the higher melting point of the steel than the aluminum. In the aluminum stir zone close to the interface, some iron- Fig. 5 Macrostructures of the transverse sections of welds: (a) Weld No. 2, (b) Weld No. 8, (c) Weld No. 14, and (d) Weld No. 20 (welding parameters are listed in Table 3). Fig. 6 Characteristic microstructures of different areas in Al/Zinc-coated steel FSW joint (Weld No. 14): (a) macrostructure of the joint, (b) fine equiaxed grains zone of aluminum (area I), (c) fine equiaxed grains zone of aluminum (area II), (d) aluminum HAZ on the advancing side (area III), (e) aluminum HAZ on the retreating side (area IV), (f) fine equiaxed grains zone of steel (area V), and (g) steel HAZ (area VI).

4 rich particles were observed as shown in Fig. 6 (a). These particles were separated from the steel surface by the stirring Fig. 7 Effect of travel speed on fine grain area of Zn-coated steel: (a) Weld No. 20, (b) Weld No.22, and (c) Weld No. 24. effect of the pin which pulled them from the zinc-coated steel surface and scattered in the aluminum substrate. With an increase in the rotation speed, grains of both aluminum and zinc-coated steel in all characteristic areas were coarsened. This is probably due to the increase in heat input which increases the maximum temperature of the area and decreases the cooling rate. This result agrees with many previous papers ). In contrast to the rotation speed, increasing the travel speed decreased the heat input, which in turn decreased the grain size and decrease the steel fine grain zone area as obviously seen in Fig. 7. Close observations with a SEM of the Al/Zn-coated steel interface revealed that layered structures containing high percentage of aluminum, from 20 to 40 %, formed in the Zncoated steel fine-grain zone adjacent to the weld interface at rotation speeds higher than 16.7 s 1. The layer structure was more developed in size and thickness at higher rotation speeds as shown in Fig. 8. Meanwhile, at a rotation speed of 16.7 s 1, a layer involving Al, Fe, and Zn was formed at the interface as shown in Fig. 9, and it extended to the advancing side. The chemical composition of this Al-Fe-Zn layer was well above the solid solubility of Fe in Al, as shown in Table 4. This suggests that the Al-Fe-Zn layer involved Al-Fe or Al-Fe-Zn intermetallic compounds. The hardness value of the aluminum was almost constant ( 30 Hv) in the equiaxed fine grain zone, HAZ, and base metal. The hardness of the zinc-coated steel showed the highest value in the layer structure, reaching 584 Hv. This could be related to the formation of intermetallic compounds within this structure as Fig. 9 layer of deformed Al/Fe/Zn observed at the Al/Zn-coated steel interface. Weld No. 2. Table 4 Chemical analyses at points 1 to 5 indicated in Fig. 9 (at %). Fig. 8 Layered structure observed at the Zncoated steel interface: (a) weld No. 20 and (b) weld No. 14.

5 190 ELREFAEY Ahmed et al. Friction-Stir-Welded Lap Joint of Aluminum to Zinc-Coated Steel suggested by EDX analyses and X-ray diffraction analyses of the fracture surfaces of the joints (see Fig.16). The very fine grain area in the steel close to interface was higher in hardness than the base metal. Hardness values of different areas at Al/Zn-coated steel interface are shown in Fig. 10. At a rotation speed of 16.7 s 1, as shown in Fig. 11, the Al- Fig. 10 Hardness values of different structures at Al/Zn-coated steel interface. (Weld No. 20). Fe-Zn layer observed at the interfacial region (see Fig.9) showed hardness higher than the aluminum and zinc-coated steel base metals, suggesting the presence of intermetallic compounds. The extension of this layer to the advancing side in the aluminum stir zone showed a similar hardness level to those at the interface. The extension of this layer was surrounded by areas diluted with aluminum showing hardness of 34 Hv. The fracture load of joints on peel test is shown in Fig. 12. It seems that increasing travel speed from 3.3 to 5.0 mm/s slightly decreased the fracture load. Meanwhile, increasing the rotation speed from 16.7 to 25.0 s 1 raised significantly the fracture load, while further increase from 25.0 to 41.7 s 1 exhibited a slight positive effect on the fracture load. For most joints bonded at rotation speeds of 25.0 to 41.7 s 1, the fracture on the peel test occurred along the path as shown in Fig. 13 (a). Iron-rich fragments stuck to the fractured surface of the aluminum side which contained mainly layer structure, similar to those observed in Fig. 8. Results from point analyses of the layered structure are listed in Table 5. The Fe and Al contents of the layered structure suggest the presence of intermetalic Fig. 11 Hardness value of the Al-Fe-Zn Intermetallic compound layer: (a) traverse section, (b) Al/Zncoated steel interface, (c) extension of the Al-Fe-Zn layer to the stir zone of aluminum. Fig. 13 Cross section of fracture surface of aluminum and zinc-coated steel sides (Weld No.22): (a) iron-rich fragment on the fractured surface, and (b) closer view of the rectangular area in (a). Table 5 Chemical analyses at points 1 to 5 indicated in Fig. 13 (b) (at %). Fig. 12 Relation between travel speed and fracture load at various rotation speeds. Fig. 14 Fracture path through aluminum at both sides of fracture surface of the joint. (Weld No.22)

6 Fig. 16 X-ray diffraction patterns from fracture surfaces of the aluminum (a) and Zn-coated steel sides (b) (Weld No.20), where diffraction lines from aluminum, iron, Al 13 Fe 4, and Al 5 Fe 2 were indicated by,,, and respectively. Fig. 15 Fracture surfaces of joint after peel test (Weld No. 8): (a) and (b) ductile morphologies on the steel side and aluminum side respectively. grain zone and layered structure in the zinc-coated steel. 3.2 Comparison with joint of aluminum to steel without zinc coating compounds. It can be considered that the crack propagated either in the aluminum substrate or layer structure. It was frequently noted that the fracture path developed into the aluminum substrate in the areas indicated by arrows in Fig 14. These joints achieved higher fracture loads than others. SEM micrographs of the fracture surfaces corresponding to the fracture path shown in Fig. 13 are shown in Fig. 15. The fracture surfaces were mainly ductile with some brittle areas. The ductile morphology was more prominent in joints that showed higher fracture loads. Meanwhile, ductile fracture morphologies were observed in the area where the crack propagated in the aluminum substrate as shown in Fig. 14. In order to identify the intermetallic compounds formed in the layered structure of the joint bonded at higher rotation speeds ( s 1 ), X-ray diffraction patterns from fractured surfaces of the aluminum and steel sides were analyzed as shown in Figs. 16 (a) and 16 (b). As can be seen from these, diffraction lines that were attributable to intermetallic compounds of Al 13 Fe 4 and Al 5 Fe 2 were detected from both aluminum side and steel side. This suggests that these intermetallic compounds were involved in the layer structure and responsible for the brittle fracture on peel test. The pin depth of 2.0 mm was not deep enough to penetrate the pin tip to the zinc-coated steel side. It only reached the top surface of zinc layer and hence, there was no evidence for fine The fracture loads of Al/Zn-coated steel joint at 2.1 mm pin depth were higher than those of similar joints of aluminum to steel without Zn coating (Al/steel) 13) in spite of similarity in microstructure of joints. It should be also mentioned that while the Al/Zn-coated steel joint exhibited considerable fracture load at pin depth of 2.0 mm, Al/steel joints were so week that they fractured during preparation of the specimen for metallugraphy at the same pin depth. In fact, the zinc-coated layer is the only reason for this improvement in the fracture load of the joint. There were two main factors controlling the performance of the solid-state bonded joint of dissimilar metals 21). One of them is the intimate contact between aluminum and steel, and the other is the microstructure, particularly the formation of brittle intermetallic compounds. In regard to the microstructure, the amounts of the layer structure and intermetallic compounds in Al/Zn-coated steel joints decreased in comparison with those observed in the Al/steel joint. According to Al-Fe-Zn ternary phase diagram 22), Zn increases the solid solubility of Fe in Al, which probably contributed to decrease the intermetallic compounds in the layer structure. Moreover, it is conceivable that the zinc layer acted as lubricant during the FSW process because zinc itself is softer than aluminum and steel. Moreover, aluminum becomes softer by alloying with zinc owning to the decrease in its melting point (the lowest melting point in the Al-Zn system is 654 K much lower than the peek temperature of the fine grain

7 192 ELREFAEY Ahmed et al. Friction-Stir-Welded Lap Joint of Aluminum to Zinc-Coated Steel zone 780 K). It is expected that the softer zone formed in the interfacial region plays a role of lubricant in the friction between the harder aluminum and steel substrates. The area of the steel fine grain zone was significantly smaller when it was bonded to the Zn-coated steel. This support the conclusion that the friction between the tool and steel was reduced by the zinc coating, since the zone was considered to be heavily deformed by the friction between the tool and steel. Practically, the sound and vibration of the tool during FSW were weaker when the zinc coated steel was welded to aluminum. This observation indicated the effect of zinc as lubricant in the joint. The lubricant effect of zinc decreased the shear stress imposed on the steel surface. This effect of the zinc layer probably contributes to the reduction of the formation of layer structure, since the shear stress ca be considered to be the driving force to incorporate the aluminum into the steel to form the layer structure. The intimate contact will be enhanced by increasing the heat input, viz. increasing the rotation speed or decreasing the traveling speed, since it facilitates the materials flow in the stir zone through lowering the flow stress, and increases the period held at high temperatures. On the other hand, it enhances the formation of intermetallic compounds, through the promotion of the mechanical mixing of the two metals and the diffusion of elements forming the intermetallic compound. In the present investigation, as shown in Fig. 12, the bond strength was increased with the heat input. The bond strength of the joint of aluminum to steel without Zn coating was also increased with the heat input. These results suggest that the bond strength of these joints was controlled mainly by the attainment of intimate contact at the interface. Therefore, the effect of Zn coating on the bond strength can be attributed mainly to the enhancement of the intimate contact at the bond interface. As mention above in the explanation of the effect of zinc coating on the layer structure, the zinc coating can be considered to introduce a softened zone in the interfacial region which contributes to the enhancement of the intimate contact similar to the increase in the heat input. However, the formation of the intermetallic compounds in the layer structure limited the rise in the bond strength of the joint. These effects of the Zn coating on the attainment of intimate contact and the formation of layer structure are probably responsible for the improvement of the bond strength observed in the Al/Zn-coated steel joint. This result also suggests that further improvement can be obtained by applying a zinc intermediate layer to the joining of aluminum to steel. 4. Conclusions 1. The feasibility of the FSW lap joint of a commercially pure aluminum plate to a zinc-coated steel plate was exhibited. The Al/zinc-coated steel joint showed higher fracture strength than the Al/steel joint, suggesting that the zinc coating had a beneficial effect on the bond strength. 2. The Al/zinc-coated steel joint welded at 2.1 mm pin depth were much stronger than that welded at 2.0 mm pin depth, and its fracture strength showed a general tendency to decrease with an increase in traveling speed. 3. The aluminum microstructure of the joint consisted of the stir zone and the heat affected zone on the retreating and advancing sides of the stir zone. The stir zone consisted of equiaxed fine-grains, and the heat affected zone was characterized by coarser grains than the base metal and stir zone. Meanwhile, the steel just under the tool pin showed a very fine grain zone that involved a hard layer structure in the area close to the weld interface. 4. Fracture of joints bonded at rotation speeds of s 1 occurred mainly in the aluminum substrate and the layer structure where intermetallic compounds such as Al 13 Fe 4 and Al 5 Fe 2 were formed. The joints having higher fracture strength showed more ductile fracture morphology which corresponds to the fracture in the aluminum substrate. References 1) T. A. Barnes, and I.R. Pashby, Joining Techniques for Aluminum Spaceframes used in Automobiles, Part I - solid and liquid phase welding, Journal of Material Processing Technology, 99-1(2000), ) S. H. C. Park, Y. S. Sato and H. Kokawa, Mechanical Properties and Microstructure in Friction-Stir-Weld of Magnesium alloy AZ61, Proceedings of the 7th International Symposium, JWS, Kobe, Japan, (2001), ) W. B. Lee, Y. M. Yeon, and S. B. Jung, Joint Properties of Friction Stir Welded AZ31B-H24 Magnesium Alloy, Material Science and Technology, 19-1 (2003), ) M. C. Juhas, G. B. Viswanathan and H. L. Fraser, Microstructural Evolution in a Titanium Alloy Friction Stir Weld, Proceedings of the 2 nd Annual Symposium on Friction Stir Welding, Gothenburg, Sweden, June, (2000). 5) Won-Bae Lee, and Seung-Boo Jung, The Joint Properties of Copper by Friction Stir Welding, Material Letters, 58-6 (2004), ) W. M. Thomas, P.L. Threadgill, and E. D. Nicholas, Feasibility of Friction Stir Welding Steel, Science and Technology of Welding and Joining, 4-6 (1999), ) T. J. Lienert, W. L. Stellwag, JR., B. B. Grimmett, and R.W. Warke, Friction Stir Welding Studies on Mild Steel, Welding Journal, 82-1 (2003), ) A.P. Reynolds, Wei Tang, T. Gnaupel-Herold, and H. Prask, Structure, Properties, and Residual Stress of 304L Stainless Steel Friction Stir Welds, Scripta Meterialia, 48-9 (2003), ) Ying Li, L. E. Murr, and J. C. McClure, Flow Visualization and Residual Microstructures Associated with the Friction-Stir Welding of 2024 Aluminum to 6061 Aluminum, Material Science and Engineering A, (1999), ) K. Yoshikawa, and T. Harano, Numerically Controlled Friction Stir Welding in Layered Dissimilar Metal Materials of Aluminum and Steel, Proceedings of the third symposium on friction stir welding,

8 Kobe, Japan, (2001). 11) T. Watanabe, A. Yanagisawa, and H. Takayama, Bonding of Steel and Aluminum Alloy by Interfacial Active Adhesion Bonding Method -Interfacial Active Adhesion Bonding of Dissimilar Materials with a Rotationg Needle-, Preprints of the National Meeting of J.W.S., 71 (2002), (in Japanese) 12) M. Tsubak, M. Fukumoto, and T. Yasui, Evaluation of Joining Property between Steel and Aluminum by Friction Stirring, Preprints of the National Meeting of J.W.S., 72 (2003), (in Japanese) 13) A. Elrefaey, M. Takahashi, and K. Ikeuch, Friction Stir Welding of Aluminum-to-Steel Lap Joint, 175 th Meeting of the Technical Commission on Welding Metallurgy, JWS, Feb., (2004), Doc. WM ,. 14) G. Oertelt, S. S. Babu, S. A. David, and E. A. Kenik, Effect of Thermal Cycling on Friction Stir Welds of 2195 Aluminum Alloy, Welding Research Supplement, 3 (2001), ) W. D. Lockwood, B. Tomaz, and A. P. Reynolds, Mechanical Response of Friction Stir Welded AA 2024: Experiment and Modeling, Material Science and Engineering A, (2002), ) M. Guerra, C. Schmidt, J. C. McClure, L. E. Murr, A. C. Nunes, Flow Patterns during Friction Stir Welding, Material Characterization, 49-2 (2002), ) Y.S. Sato, H. Kokawa, K. Ikeda, M. Enomoto, S. Jogan, and T. Hashimoto, Microtexture in the Friction-Stir Weld of an Aluminum Alloy, Metallurgical and Metals Transactions A, 32-4 (2001), ) K.V. Jata, and S.L. Semiatin, Continuous Dynamic Recrystallization during Friction Stir Welding of High Strength Aluminum Alloys, Scripta Materialia, 43-8 (2000), ) J.Q. Su, T.W. Nelson, R. Mishra, M. Mahoney, Microstructral Investigation of Friction Stir Welded 7050-T651 Aluminum, Acta Materialia, 51-3 (2003), ) C. G. Rhodes, M. W. Mahoney, W. H. Bingel, and M. Calabrese, Fine-Grain Evolution in Friction-Stir Processed 7050 Aluminum, Scripta Materialia, (2003), ) K. Kobayashi, K. Nishimoto, and K. Ikeuchi, Zairyo Setsugo Kogaku No Kiso (Introduction to Joining Engineering of Materials), Sampo, 2002, (in Japanese) 22) P. Villars, A. Prince, and H. Okamoto, Handbook of Ternary Alloy Phase Diagram, ASM International, 1995,