2021 Volume 62 Issue 3 Pages 448-452
Although pure copper has good electrical and thermal conductivity, it is often used in combination with other metals due to its poor strength. An inexpensive austenitic stainless steel that has excellent corrosion resistance is a good joining material for pure copper, and the joining of these dissimilar metals enables the fabrication of parts for a wide range of applications. However, it is difficult to weld pure copper and stainless steel due to the large differences between their thermal properties. Therefore, in this study, the pulsed-electric-current bonding was applied to achieve solid state bonding of oxygen-free copper (OFC) and SUS304 austenitic stainless steel, and the bonding characteristics were investigated. The results show that the joint tensile strength improves with the increase in the bonding temperature and the applied pressure, and the sample bonded at 973 K with an applied pressure of 20 MPa exhibited a high strength of greater than 200 MPa, which caused a fracture in the OFC base material.
This Paper was Originally Published in Japanese in J. Jpn. Inst. Copper 59 (2020) 91–95.
Although pure copper has good electrical and thermal conductivity, it is often used in combination with other metals due to its poor strength. An inexpensive austenitic stainless steel as a material with excellent corrosion resistance has a good bonding property with pure copper, and the satisfactory bonding of these dissimilar metals is harnessed in components fabrication in a wide range of applications. However, it is difficult to weld pure copper and stainless steel due to the differences between their melting points and thermal conductivities as well as the low solid solution limit of copper compared to stainless steel.1,2) Therefore, the solid-phase bonding method without melting is considered to be a suitable technique for joining these metals. Solid-state bonding methods mainly include diffusion bonding, friction welding, hot pressure welding, cold pressure welding, explosion pressure welding and gas pressure welding.3) Precedents for solid-state bonding of copper/stainless steel include friction welding, friction stir welding (FSW), explosive welding, and diffusion bonding, all of which do achieve good bond strength. In the cases of friction welding, FSW, and explosion welding, it has been reported that the hardness of the samples increases due to the effect of plastic deformation.4–6) Diffusion bonding is a method in which a limited pressure is applied such that there is moderate plastic deformation, and the diffusion of atoms occurring at the bonding interface is utilized in bonding.7) Compared with friction welding that requires a rotary drive in the welding process, diffusion bonding is suitable for joining mechanical parts that require high precision due to the degree of freedom of the weldable components and the large shape of components that can be joined. Considering these advantages, we propose that pulsed-electric-current bonding, which is a kind of solid-phase diffusion bonding, is an effective joining technique.
The pulsed-electric-current bonding is called spark plasma sintering bonding (SPS bonding); it is a method in which the bonding material is sandwiched by punches of the same material as the graphite die, and a pulse current is applied directly to the bonding material. Thus, the joint is achieved by utilizing the discharge plasma arising from the minute discharge generated at the initial stage of the energization. The important parameters to be considered in this study include discharge impact pressure, the Joule heat diffusion, and the electric field diffusion effect by the electric field. Since the joining surfaces are is locally heated by directly energizing the components, joining with reduced energy consumption is possible even in a shorter time compared with the hot isostatic pressing (HIP) and hot pressing methods. Also, since a joint surface without inclusions can be achieved by cleaning the adjoining surfaces, extensive researches are ongoing reports on the joining of dissimilar metals and ceramics.8–19) Based this understanding, we attempted direct joining of oxygen-free copper and austenitic stainless steel using the pulsed-electric-current bonding method and evaluated the characteristics of this method.
The samples used were ϕ10 × 45 mm round bars of oxygen-free copper (OFC) and austenitic stainless steel (SUS304). The surfaces of the samples were lathe-processed and ground to #2000 using wet emery paper, then buffed using Al2O3 powder with a particle size of 1 µm for a mirror finish, and finally washed with acetone ultrasonically.
A spark plasma sintering machine (model: SPS-1020) manufactured by Fuji Electronic Industrial Co., Ltd (former Sumitomo Coal Mine Co., Ltd.) was used for the pulsed-electric-current bonding. The OFC and SUS304 were introduced into a graphite die with an internal diameter of 10 mm and pressed vertically. The temperature of the joining surfaces was measured by inserting a sheath of K-type thermocouple into a ϕ4 mm hole at the side surface of the mold. After the bonding, the load was removed, and the assembly was cooled in a chamber under atmospheric pressure. Then, the chamber, dies, and sample components were cooled, and air was charged into the assembly to allow removing the sample from the mold. The joining conditions were vacuum pressure of 10 Pa or less, a heating rate of 50 K/min, holding time of 30 min, applied pressures of 15 and 20 MPa, and bonding temperatures of 773, 873, and 973 K.
Using three samples, the joint tensile strength was analyzed by Shimadzu Autograph (AG-IS). Then, observation of the microstructure of the joint interface and the fractured surface was conducted using a scanning electron microscope (SEM) (model: JSM-6060LV, JEOL), and the elemental analysis was performed using and an energy dispersive X-ray (EDX) attached to the SEM. The SEM was conducted at an acceleration voltage of 15 kV and the observed surfaces were mirror-polished prior to the characterization. Furthermore, to identify the elemental compounds existing on the fractured surface after the tensile test, X-ray diffraction (XRD) analysis was performed using a diffraction X-ray tester (model: RINT-2550V) manufactured by Rigaku Corporation.
From the observation of the cross-sectional microstructure, no crack occurred at an applied pressure of 15 and 20 MPa, and no intermetallic compound was formed according to the XRD analysis. Since no void or crack was observed at the joint interface, it was confirmed that a good joint was achieved. Figure 1 shows the results of a line analysis of the bond interface by SEM-EDX at a pressure of 20 MPa and bonding temperatures of 773 and 973 K. The diffusion of Cu could not be confirmed at 773 and 973 K, but at 973 K there was a gradation in the Cu concentration at the bond interface which suggests the possibility of bonding at the interface.
SEM and EDX-line analysis of joint interface of OFC/SUS304 bonded by SPS with pressure of 20 MPa at 773 and 973 K.
Figure 2 shows the experimental results of the tensile test. The right vertical axis of the figure represents the tensile strength of the OFC base metal at 100% joint efficiency. At a pressure of 15 MPa, the joint tensile strength was 27 MPa at 773 K, 51 MPa at 873 K, and 74 MPa at 973 K. Thus, it can be deduced that the tensile strength improved as the bonding temperature increased. Achieving a bonding strength even at 773 K suggests that the chromium oxide film on the SUS304 surface can could be dissociated also at a relatively low temperature. This is attributed to the fact that the pulsed-electric-current bonding method applies the current directly to the sample. This direct application of current causes local heat generation due to the resistance of the oxide film on the bonding surfaces and the contact resistance at the bonding interface. Furthermore, the tensile strength of the joints bonded under the pressure of 20 MPa was higher than that of joint at 15 MPa; the strengths were 36 MPa at 773 K, 86 MPa at 873 K and 222 MPa at 973 K. Particularly, the joint bonded at 973 K had the highest strength and fractured in the OFC matrix. It was assumed that the surface unevenness deformed as the applied pressure increased, thereby more increasing the contact area and ultimately results in better contact between the joining interface. Specifically, the tensile strength of the joints increased at the bonding temperature of 973 K because the high temperature strength of OFC is extremely low and the unevenness is easily deformed.
Joint tensile strength of OFC/SUS304 joint.
Figure 3 shows the results of the SEM observation of the fractured surface after the tensile test. At a bonding temperature of 773 K, the fractured surface of the specimens was relatively smooth at 15 and 20 MPa pressures. Considering that the tensile strength of the joint is low at this temperature as described in 3.2, suggests that the driving force at the joint was insufficient at the temperature. At the bonding temperatures of 873 and 973 K, different microstructures were observed at the center region and the outer edge of the fracture surface. Specifically, different microstructures were evident at the fracture surface under the pressure of 15 MPa at 873 K and the difference was more pronounced under 20 MPa at 873 K. This is because there is an interfacial resistance between the graphite mold and the sample surface, and the temperature tends to rise locally. In addition, it was considered that Joule’s heat was less likely to be generated due to the high conductivity of OFC, indicating that the thermal diffusion from the graphite mold had a great influence. The reason for the increased difference in the fractured surface morphologies observed at the outer edge and the center region as the pressing force increases is attributed to the fact that the deformation of the surface irregularities at the center region caused the adherence of the bonding surfaces and decrease in the interfacial resistance. Ultimately, there is an increased difficulty in the heating process.20) On the other hand, the outer edge may be in contact with the graphite mold, causing it to be easily affected by heat.
SEI of fractured surface of the OFC/SUS304 joint bonded by SPS.
Figures 4 and 5 show SEM images and EDX elemental maps of the SUS304 fractured surface at bonding temperatures of 873 and 973 K and a pressure of 15 MPa. At both temperatures, the mating material Cu was detected on the outer edge of the SUS304, however, more Cu was detected at 973 K, shown in Fig. 5. It was assumed that fracture occurred in the OFC matrix. However, no Cu was detected in the central region of the SUS304. This suggests that in pulsed-electric-current bonding, interfacial bonding commences interface bonding proceeds from the outer edge.
SEI and EDX analysis of fractured surface of SUS304 side of the OFC/SUS304 joint bonded by SPS with pressure of 15 MPa at 873 K.
SEI and EDX analysis of fractured surface of SUS304 side of the OFC/SUS304 joint bonded by SPS with pressure of 15 MPa at 973 K.
Figure 6 shows the EDX analysis of the fractured surface of the sample bonded under 15 MPa at 973 K. The elemental map of O did not show a corresponding Cr enrichment, thus, fails to provide the information of inclusions such as chromium oxides formed by the aggregation of the chromium oxide film at the fractured surface. In the diffusion bonding of OFC and SUS304, it was reported that the process of the chromium oxide film removal was through condensation, and the inclusion of the chromium oxide film was confirmed on the fractured surface.21) On the other hand, in the pulsed-electric-current bonding, these inclusions did not exist even in the bonding under a low vacuum of 10 Pa or less, indicating that the oxide film was not aggregated at the high temperature and undestroyed by the pressure. It was presumed that the oxide film was destroyed by direct energization on the contact surface, and a good joint with no inclusions was obtained. There are some examples regarding the report of joining dissimilar metal by the pulsed current under low vacuum lesser than below 10 Pa.20,22)
SEI and EDX analysis of fractured surface of the OFC/SUS304 joint bonded by SPS with pressure of 15 MPa at 973 K.
Figure 7 shows the XRD pattern of the OFC and SUS304 fractured surfaces. The diffraction line of Cr23C6 was detected on the SUS304. This suggests that a chromium carbide phase was formed near the bonding interface. This was attributed to the fact that chromium carbide was formed at the grain boundaries by due to its prolonged retainment at 873 to 1073 K where the solid solubility limit of carbon is low.23) This can be avoided by increasing the bonding temperature or using low carbon materials. On the OFC and SUS304, as shown in Fig. 4 and Fig. 5, the elements of the mating material are detected at the outer edges. However, the XRD pattern in Fig. 7 did not show Cu and γ-Fe diffraction lines. This is because both Cu and γ-Fe have the FCC structure and the diffraction lines exist at very close angles, making it difficult to distinguish the fractured surfaces of Cu and SUS304. Also, it was assumed that the chromium oxide film formed on the SUS304 bonding surface after the tensile test was extremely thin and imperceptible.
X-ray diffraction patterns of fractured surface of the OFC/SUS304 joint specimen after tensile test.
The following findings were obtained from the characterization and evaluation of directly welded oxygen-free copper and austenitic stainless steel via pulsed-electric-current bonding.
