- Email: [email protected]
Ti alloy
Optics and Laser Technology 115 (2019) 149–159
Contents lists available at ScienceDirect
Optics and Laser Technology journal homepage: www.elsevier.com/locate/optlastec
Full length article
Effect of copper-nickel interlayer thickness on laser welding-brazing of Mg/ Ti alloy ⁎
T
⁎⁎
S.T. Auwala,b, S. Ramesha, , Zequn Zhangc,d, F. Yusofa, Jinge Liuc, Caiwang Tanc,d, , S.M. Manladane, F. Tarlochanf a
Center for Advanced Manufacturing and Materials Processing, Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia b Department of Mechanical Engineering, Faculty of Engineering, Kano University of Science and Technology, Wudil, 3244 Kano, Nigeria c State Key Laboratory of Advanced Welding and Joining, Harbin Institute of Technology, Harbin 150001, China d Shandong Provincial Key Laboratory of Special Welding Technology, Harbin Institute of Technology at Weihai, Weihai 264209, China e Department of Mechanical Engineering, Faculty of Engineering, Bayero University, Kano, 3011 Kano, Nigeria f Department of Mechanical and Industrial Engineering, College of Engineering, Qatar University, Doha, Qatar
H I GH L IG H T S
Mg alloy to titanium with varied thickness Cu-Ni coating using laser welding-brazing process. • Join interlayer elements content were found to exert varied strengthening influence. • Different • The optimum fracture load was obtained with appropriate thickness of Cu-Ni coating.
A R T I C LE I N FO
A B S T R A C T
Keywords: Mg alloy Laser welding-brazing Mechanical properties Interfacial reaction Microstructure Ti alloy
Dissimilar lap joining of AZ31B Mg alloy to Cu-Ni coated Ti-6Al-4V was carried out by laser welding-brazing method. The effect of Cu and Ni contents on interfacial reaction and joint fracture load were analyzed. For the joint in which the Ni coating (15.36 µm) was thicker than the Cu coating (5.47 µm), thick intermetallic compound (IMC) composed of a mixture of light gray Al-Ni-Ti + Ti3Al + Ti2Ni phases mingled with dark gray Ti3Al was produced at the interface. In comparison, a mixed interfacial reaction layer consisted of Ti2Ni and Ti3Al was formed from the direct irradiation zone to the weld toe zone of the joint with comparable Ni and Cu coating thicknesses (10.78 µm Cu–9.30 µm Ni). In this case, the thickness of the mixed layer was below the critical thickness of 10 µm. For the joint in which the Cu coating is much higher than the Ni coating thickness (17.12 µm Cu–4.23 µm Ni), Ti3Al and Ti2Ni mixed reaction layer was produced at the brazed interface of direct irradiation zone, whereas, only Ti3Al phase was formed at the middle zone. At the weld toe zone, Ti2Cu uneven interfacial reaction layer was evolved. With increasing Cu and decreasing Ni coating thicknesses, the fracture load first increased and then slightly decreased, the maximum tensile-shear fracture load attained 2020 N for joints with comparable Cu and Ni coating thicknesses. This is twofold higher than that of uncoated joint. The tensile-shear investigation showed that the joint would fracture at the fusion zone when the coating thickness of Ni was comparable or higher than Cu. In contrast, interfacial failure was observed when the thickness of Cu was much higher than the Ni. For the joint with interfacial failure mode, tear ridge was observed from the fracture surface, whereas, the fusion zone fracture surfaces was noted to display a typical dimple feature.
⁎ Corresponding author at: Center for Advanced Manufacturing and Materials Processing, Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia. ⁎⁎ Co-corresponding author at: Shandong Provincial Key Laboratory of Special Welding Technology, Harbin Institute of Technology at Weihai, Weihai 264209, China. E-mail addresses: [email protected] (S. Ramesh), [email protected] (C. Tan).
https://doi.org/10.1016/j.optlastec.2019.02.024 Received 12 October 2018; Received in revised form 21 November 2018; Accepted 3 February 2019 0030-3992/ © 2019 Elsevier Ltd. All rights reserved.
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
1. Introduction
the Mg-Ni interface, whereas, solid-state diffusion occurred at the Ni-Ti interface. The analysis of the joint quality revealed that the maximum tensile-shear of only 39 MPa was achieved. The low shear strength observed was attributed to the fact that the foils thickness produced a thick IMCs. These IMCs reduced the direct contact area and hence, affected bond strength of the joint. However, further increased in the joint shear strength was achieved with electroplated coatings [23,24]. The electrodeposition of thin coats was found to be beneficial in improving the flow and uniformity of the liquid. In addition, the authors reported maximum shear strength with an electrodeposited Ni coating containing a dispersion of nanoparticles of pure Cu [25]. The addition of nanoscale copper dispersion in Ni coating facilitates the bonding process through shorter diffusion distances. In our recent work, the benefit of using Ni coating to join Mg/Ti by LWB method was explored. The presence of the addition of Ni coating was found to be beneficial in improving the Mg/Ti joint performance [26,27]. These studies suggested that the addition of Cu and Ni in the form of thin coats proved to be effective in realizing an interfacial reaction between the immiscible Mg/Ti systems with excellent joint quality. Furthermore, the interlayer elements content could affect the dissimilar joints’ microstructural evolution and mechanical properties [24,28]. Thus, different thickness of interlayer is expected to exert varied strengthening influence. In this study, the effect of laser weldedbrazed Mg/Ti lap joints with and without electrodeposited Cu-Ni layer of different thickness was studied and defined. The joint appearance, microstructure development and joint fracture load were examined. The joining mechanism of the Mg/Ti joints with varied Cu and Ni thicknesses was also discussed.
Magnesium (Mg) has been described as a green engineering material and one of the most promising material categories in the 21st century [1]. It has unique properties such as formability, good recyclability and high specific strength [2,3]. Because of these properties, Mg alloys are being used to replace not only heavy metals such as steels, but also light alloys, including aluminum alloys, in a wide range of applications [4,5]. Recently, they attract special attention in automotive and aerospace industries where weight reduction is crucial [6–8]. On the other hand, titanium (Ti) and its alloys have been used extensively in aerospace, medical, chemical, biomaterial, and petrochemical industries for their high creep resistance, excellent corrosion resistance and high strength to weight ratio [9]. The ability to join Mg alloys to Ti effectively has received great interest in high-technology based industries such as aerospace and automotive manufacturing sectors where light-weight components are crucial in order to exploit the advantages of both materials, reduce fuel consumption, greenhouse gases and improve performance of energy converting system [10,11]. However, the significant differences in their physical (e.g. melting point, coefficient of thermal expansion, thermal conductivity) and metallurgical characteristics (nearly zero solubility) make joining them together very difficult [12]. Although, there were reports of direct joining of dissimilar metals such as Ti alloy to steel [13] and Ti alloy to Al [14,15] without the use of interlayer material or filler metal, this would depend on the base metals to be join and their ability to form an intermetallic layer as well as careful manipulation of the welding parameters. In view of the immiscible characteristics of Mg/Ti alloys, thus, a direct and economical approach to joining them would be to use a suitable interlayer material during the welding process. In choosing a suitable interlayer would largely depends on the material composition. Ideally, the interlayer material should be able to react with the base metals to provide excellent wetting and bonding without generating thick layers of hard and brittle intermetallic compounds (IMCs) at the joint interface. Moreover, when choosing the joining process, minimization of the thickness of any brittle IMCs along the interfaces of the Mg alloy-interlayer-Ti joint and minimization of intermixing between the Mg and Ti in the molten-state are the main factors that must be considered [16]. Hence, selecting an appropriate welding technique and interlayer material composition to produce and control interfacial layers became the focus of Mg/Ti joining. Based on the existing literature, intermediate elements such as Al, Ni have been employed to achieve interfacial bonding between Mg/Ti by various joining techniques. During friction stir welding [17], mutual diffusion of aluminum from the Mg-Al-Zn alloy and titanium was achieved as a result of combined action of pressure and stirring. The analysis of the interface characteristics showed that TiAl3 intermetallic compound was produced. With increasing the aluminum content in the Mg-Al-Zn base metal, the thickness of the TiAl3 phase increases, whereas, the tensile strength decrease. Cao et al. [18] reported that for cold metal transfer (CMT) welded Mg to Ti alloy, the Al from the Mg filler metal performed an important role in joining the immiscible couple. Interfacial reaction layer identified as Ti3Al phase was observed. In our previous work, joining Mg/Ti with AZ91 filler metal was achieved by laser welding brazing (LWB) method [19,20]. The reaction between the Al from the Mg-Al-Zn filler metal and titanium resulted to Ti3Al IMC layer formation at the interface. Recently, Zang et al. [21] examined the influence of Al interlayer thickness on laser conduction welded AZ31/Ti-6Al-4V dissimilar joint characteristics. It was found that TiAl3 phase was produced at interface. However, increasing the Al thickness had little effect on the newly formed reaction layer. The results of these studies proved that the Al intermediate element performed an important role in joining the immiscible couple. On the other hand, Atieh et al. [22] investigated the diffusion behavior of Mg alloy to Ti joint with Ni foil. It was found that eutectic reaction was formed at
2. Experimental 2.1. Materials and electrodeposition process The materials used in this study were commercially available 1 mm thick Mg alloy (AZ31B) (95–97 wt% Mg, 2.5–3.5 wt% Al, 0.5–1.5 wt% Zn) and 1.5 mm Ti alloy (Ti-6Al-4V). Both alloys were cut into specimens with length of 100 mm and width of 55 mm. Furthermore, commercially available AZ92 filler wire (89 wt% Mg, 9 wt% Al, 2 wt% Zn) with diameter of 3.5 mm was employed. The electrodeposition of the pure copper and nickel on Ti substrate was carried out based on American Society for Metals (ASM) standard [29]. Prior to electrodeposition, Ti alloy surfaces was acid pickled to remove the oxide film using an acid solution containing 15% hydrochloric acid, 5% hydrofluoric acid and 80% deionized water for 3 min and then cleaned with water. The electroplating of the pure Cu and Ni followed a sequence of Cu and finally Ni (Cu-Ni) on the Ti substrate was performed in a 500 mL beaker containing the plating solution as given in Table 1 and Table 2. During the electro-plating process, the interlayer was used as the anode while the Ti was the cathode. Furthermore, DC power was used during the Cu-Ni electrodeposition process, with pH value of 5 at 35 °C. Furthermore, the magnetic stirrer to stir the bath was kept at 200 rpm. In order to get a 20 µm thick uniform Cu and Ni layer on the Ti substrate, different cathode current densities and electrodeposition times were tried. When the electroplating was carried out at current density of 0.3 A/dm2 for 60 min, an approximate 20 µm Cu and Ni layer on the Ti substrate of different metals layer contents were observed. For each coating condition, the average of at least five coating thicknesses was measured and listed in Table 3. The scanning electron microscope (SEM) cross sectional micrograph of various Table 1 Composition of Cu electro-plating solution.
150
Compositions
K4P2O7.3H20
C6H17N3O7
CuSO4·5H2O
Na2HPO4·12H2O
Amount (g/L)
220
20
48
24
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
was cut into pieces and preset at the join interface before welding. The Mg filler metal was irradiated with the vertical laser beam. To increase the spot size of the laser beam, the beam was defocused. Thus, the laser beam diameter of 2.4 mm was used at the defocused distance. The joint was shielded from oxidation with Ar gas having a purity of 99.99%. The LWB parameters employed in the current work is shown in Table 4.
Table 2 Composition of Ni electro-plating solution. Compositions
NiSO4·6H2O
NaCl
Na2SO4
H3BO3
MgSO4·7H2O
Amount (g/L)
180
10
70
30
60
Table 3 Electroplating time and corresponding Cu and Ni coating thicknesses. Joint
MT-0 MT-1 MT-2 MT-3 MT-4 MT-5
Electroplating time (min)
Coating thickness (µm)
Cu
Ni
Cu
Ni
0 10 20 30 40 50
0 50 40 30 20 10
0.00 5.47 8.41 10.78 14.49 17.12
0.00 15.36 12.07 9.30 6.97 4.23
2.3. Characterization Total
Standard procedures were adopted to prepare the metallographic specimens after welding. OLYMPUS-DSX510 Optical microscopy (OM) and Zeiss Merlin Compact SEM equipped with EDS were adopted to examine the cross sectional microstructures morphologies and fracture surface. The reaction products formed were verified by X-ray diffraction (XRD) (Panalytical X’pert 3 Powder) using a copper target at 35 mA and 40 kV. The 2θ scans were carried out from 20° to 90°, at a step size of 0.04° and 1 s per step. The joint quality is estimated through room temperature tensile-shear testing on 10 mm wide specimens machined from the joints using Instron 5967 tensile tester at a cross heat speed of 1 mm/min. Shims were added to each specimen to minimize induced couples. An average of at least 3 tensile shear testing specimens was used to evaluate the mechanical performance of the joint.
0.00 20.83 20.48 20.08 21.46 21.35
coating layer thicknesses with uniform characteristics is shown in Fig. 1. In addition, Fig. 1(f) demonstrates the relationship between the coating thickness and the samples coating condition. To assess the uniformity of the Cu-Ni coating of the final deposit, energy-dispersive X-ray spectrometer (EDS) mapping was performed as typically presented in Fig. 2. As shown on Fig. 2 relatively uniform coat thickness was achieved. Nevertheless, small discontinuities is inevitable as observed in Fig. 2(b), which could be associated with electrochemical heterogeneity over a work-piece surface under electrodeposition, which resulted in localization of electrodeposition reactions and rates [30].
3. Results and discussion 3.1. Weld appearance and cross sections The joints typical appearance and cross section at various coating conditions is presented in Fig. 3. Under the various coating thicknesses, smooth weld surface were observed, suggesting that the addition of the thin deposited coats enhanced the wettability of the brazing alloy. For comparison, joining of the uncoated joint was also carried out as shown in Fig. 3(a). As illustrated on Fig. 3(a), large contact angle was formed, which was associated with unwetted filler alloy. This behavior is attributed to poor affinity of the Mg/Ti. Furthermore, it can be observed from the corresponding joints cross-sections that between the brazing alloy and Mg base metal (BM), fusion was observed. In comparison,
2.2. Laser welding-brazing process A 6-kW fiber laser (IPG-6000) having a beam parameter product of 8 mm·mrad and 1070 nm wavelength was employed in the current work. The join design adopted was lap configuration with magnesium base metal placed at the top of the titanium. A stainless-steel wire brush was used to clean the magnesium surface oxide layer. The brazing alloy
Fig. 1. SEM cross-sectional micrographs of the Cu-Ni coating layer at different electroplating time: (a) MT-1; (b) MT-2; (c) MT-3; (d) MT-4; (e) MT-5 and (f) Relationship between the coating thickness and the samples coating condition. 151
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
Fig. 2. EDS mapping analysis of Cu-Ni layer for MT-3: (a) SEM image; (b–d) Cu, Ni, Ti; and (e) line scanning result. Table 4 Laser welding-brazing parameters used in the experiment. Laser power (W)
1400
Welding speed (m/min) Defocus distance from Ti substrate (mm) Flow rate of the Ar shielding gas (L/min)
0.3 +30 20
brazing was formed at fusion zone/Ti substrate. Fig. 3(a–d) show the fusion lines differentiating the fusion zone (FZ) from the magnesium base metal. Nevertheless, flat titanium surfaces were produced, confirming that the titanium remained un-melted. This could be associated with positive defocus distance employed in the current work. To further compare the wettability of the brazing alloy in the coated and uncoated joining situations, contact angle and seam width were drawn in Fig. 4. A contact angle of 50.8˚ and a seam width of 5.42 mm were observed for the uncoated joint, whereas, contact angle of less than 35˚ and seam width of up to 6.47 mm for Cu-Ni coated joints were produced. However, slight variation of the seam width and contact angle was observed under different coating conditions. 3.2. Interfacial microstructure The interfacial microstructure of the uncoated (direct) and Cu-Ni coated joints at various coating conditions were studied under similar welding parameters. For the uncoated joint, relatively smooth interface was observed, suggesting no obvious reaction layer. However, ultrathin Ti3Al IMC was formed across the brazed interface. Detail analysis of the microstructural evolution and the related bonding mechanism were reported in our previous work [20,31]. In this study, emphasis was given to Cu-Ni coated joints. Figs. 5, 6 and 8 show the interfacial microstructures under various coating condition. After welding, the original coating layer was not visible, suggesting suitable heat-input adopted in this work. The high temperature gradient during LWB process caused the interfacial microstructure along the interface to vary [32]. Thus, the FZ-Ti brazed interface was divided into 3 parts namely: laser irradiation, middle and weld toe zones. Fig. 5(a–e) show the SEM images of MT-1 joint, i.e. Ni coating
Fig. 3. Mg/Cu-Ni coated Ti laser welded-brazed joints with various Cu and Ni coating thicknesses: (a) MT-0; (b) MT-1; (c) MT-2; (d) MT-3; (e) MT-4; and (f) MT-5.
thicker than that of Cu (5.47 µm Cu-15.36 µm Ni). EDS analyses were performed to characterize phases formed as presented in Table 5. The phases formed at different zones were homogenous. At the FZ, the EDS analysis showed that the dark phase (Point P1 in Fig. 5a) consisted of 90.96 at.% Mg, 9.04 at.% Al, suggesting α-Mg formation. Meanwhile, 152
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
the secondary phases (P2 in Fig. 5a) precipitated in the FZ was identified as Mg17Al12 since it contained 27.30 at.% Al, 72.70 at.% Mg. The formation of Mg17Al12 phase was formed during eutectic reaction [33]. In addition, bulk phases P3 of Fig. 5b aggregated in the fusion zone nearby the interface, which consisted of mainly 39.62 at.% Al, 42.69 at. % Mg, 17.69 at.% Ni, suggesting they were Mg-Al-Ni ternary phases [27]. The formation of these bulk phases could be associated with melting and diffusion of Ni coating into the brazing alloy. At the brazed interface, light gray (P4 in Fig. 5b) and dark gray (P5 in Fig. 5b) mixed reaction layer was observed. The light gray phase consisted of 33.07 at. % Al, 42.31 at.% Ti, 19.84 at.% Ni, 4.39 at.% Cu, which was identified as IMC that composed of a mixture of Al-Ni-Ti + Ti3Al + Ti2Ni phases [34], whereas, the dark gray phase contained 29.66 at.% Al, 61.87 at.% Ti, 6.10 at.% Ni, 2.12 at.% Cu, suggesting it was Ti3Al based on Ti-Al phase diagram [35]. An EDS line scan was carried out across the middle zone (Fig. 5f). The results disclosed that titanium content increased while magnesium decreased with concentration of nickel and aluminum at the interface, signifying the possible formation of Al-Ni-Ti, TiNi and Ti-Al IMCs. Furthermore, no Cu segregation was noted, indicating that the Cu content was low and under the action of flow and vortex, the thin Cu layer could have dissolved into the liquid filler. Fig. 6(a–e) present the morphologies of MT-3 joint interfacial microstructures, i.e. comparable Ni and Cu coating thicknesses (10.78 µm
Fig. 4. Varied contact angles and seam widths under different coating conditions.
Fig. 5. Interfacial microstructure morphologies of the MT-1 joint: (a) direct irradiation zone; (b) higher magnification of region b; (c) middle zone; (d) weld toe zone; (e) higher magnification of region e; (f) line scanning result. 153
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
Fig. 6. Interfacial microstructure morphologies of the MT-3 joint: (a) direct irradiation zone; (b) higher magnification of region b; (c) middle zone; (d) weld toe zone; (e) higher magnification of region e; (f) line scanning result.
to the relatively limited Ni atoms. On the other hand, a mixed layer of light grey (P8 in Fig. 6b) and dark grey (P9 in Fig. 6b) was produced at interface. The light grey phase consisted of 11.30 at.% Al, 56.81 at.% Ti, 26.82 at.% Ni and 4.14 at.% Cu, which was identified as Ti2Ni IMC, whereas, the light grey phase contained 22.72 at.% Al, 68.78 at.% Ti, 5.12 at.% Ni and 3.25 at.% Cu, suggesting the possible formation of Ti3Al IMC. Noteworthy, compared to MT-1 joint the thickness of the interfacial reaction layer decreased. At the weld toe zone, the precipitation of second phase (P10 in Fig. 6d) inside the FZ increased. The second phases consisted of 13.57 at.% Al, 79.64 at.% Mg, 0.14 at.% Ti, 0.30 at.% Ni, 6.35 at.% Cu, which were identified as Mg17(Al,Cu)12 due to the little copper percentage. Study have shown that the Mg17(Al,Cu)12 formation was attributed to the replacement of aluminum atoms in Mg17Al12 by copper because of their comparable atomic radius [36]. Some dendritic structures P11 dispersed about αMg matrix with lamellar morphology were also observed nearby the interface (Fig. 6e). These dendritic phases consisted of 15.87 at.% Al, 67.62 at.% Mg, 4.20 at.% Ni and 12.21 at.% Cu. Coupled with our with our prior work [37], these phases were characterized as Mg-Cu eutectic structure (α-Mg + Mg2Cu). Eutectic reaction occurred at 485 °C: L ↔ Mg + Mg2Cu. The eutectic structure was also observed during welding of dissimilar Mg to steel with Cu interlayer [38]. EDS line scanning results (Fig. 6f) across the weld toe zone of FZ-Ti brazed interface
Table 5 EDS results of various points indicated in Figs. 5, 6 and 8 (at.%). Point
Mg
Al
Ti
Ni
Cu
Possible phases
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15
90.96 72.70 42.60 0.40 0.25 0.03 42.53 0.93 0.13 79.64 67.62 0.72 6.84 1.32 0.30
9.04 27.30 39.62 33.07 29.66 17.99 29.75 11.30 22.72 13.57 15.87 12.69 20.70 27.35 6.69
— — — 42.35 61.87 71.57 0.90 56.81 68.78 0.14 0.10 59.00 60.89 67.19 62.52
— — 17.78 19.39 6.10 8.08 25.32 26.82 5.12 0.30 4.20 22.12 6.45 2.27 4.38
— — — 4.79 2.12 2.33 1.50 4.14 3.25 6.35 12.21 5.47 5.12 1.87 26.11
α-Mg Mg17Al12 + α-Mg Mg-Al-Ni Al-Ni-Ti + Ti3Al + Ti2Ni Ti3Al Ti3Al Mg-Al-Ni Ti2Ni Ti3Al Mg17(Al, Cu)12 + α-Mg α-Mg + Mg2Cu Ti2Ni Ti3Al Ti3Al Ti2Cu
Cu–9.30 µm Ni). The phases formed at different zones were observed to be inhomogeneous. At the laser irradiation and middle zones adjacent to the brazed interface, some bulk phases (P7 in Fig. 6a) identified as Mg-Al-Ni compounds were observed. However, compared with MT-1 joint (Fig. 5), the distribution area of these bulk phases was small, due 154
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
Fig. 7. EDS mapping of MT-3 joint at the weld toe region: (a) corresponding SEM image; (b–f) Al, Ni, Cu, Ti, and Mg.
be inhomogeneous. At the laser irradiation zone, Mg17(Al,Cu)12 secondary particles precipitated in the entire FZ. At the joint brazed interface, an interfacial reaction layer comprising light grey P12 and dark grey P13 phases, which were confirmed as Ti2Ni and Ti3Al, respectively was observed. Owing to the limited Ni contents, only Ti3Al (P14 in Fig. 8b) formed at the middle zone, according to the compositions examination. Furthermore, limited content of Ni content coupled with strong stirring caused the Mg-Al-Ni bulk compounds to be concentrated at the FZ of weld toe zone. In addition, large amount of Mg-Cu dendritic structures with lamellar morphology dispersed around the α-Mg matrix as shown in Fig. 8(d). Owing to the high Cu contents, the distributions
confirmed the Ti-Al and Ti-Ni IMCs formation. In contrast, the Cu concentration observed in Fig. 6(f) could be associated with the presence of Mg-Cu phase stick to the interface. EDS mapping analysis at the weld toe zone (Fig. 7) showed that the Al and Ni contents concentrated along the interface as shown in Fig. 7(b and c), indicating possible formation of Ti3Al and Ti2Ni IMCs with thickness of approximately 5.34 µm. Fig. 7(d) indicates that the copper atoms diffused into the FZ to contribute in the microstructural evolution. Fig. 8(a–d) show the interfacial microstructure morphologies of MT5 joint where the Cu coating thickness higher than that of Ni (17.12 µm Cu–4.23 µm Ni). The phases formed at different zones were observed to
Fig. 8. Interfacial microstructure morphologies of the MT-5 joint: (a) direct irradiation zone; (b) middle zone; (c) weld toe zone; and (d) higher magnification of region d. 155
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
Fig. 9. EDS mapping of MT-5 joint at the weld toe region: (a) corresponding SEM image; (b–f) Al, Ni, Cu, Ti, and Mg.
Fig. 10. XRD analysis of the MT-5 joint.
Fig. 11. Simulation of thermal cycles of MT-5 joint at various zones.
of this denser eutectic structure was more than that of MT-3 joint. An uneven IMC layer P15 was formed at the interface of the weld toe zone. This uneven layer contained 6.69 at.% Al, 62.52 at.% Ti, 4.38 at.% Ni and 26.11 at.% Cu, thus suggesting it was Ti2Cu. IMCs containing Ti-Al was not observed in this region due to insufficient Ti and Al diffusion force. EDS mapping was carried out at the weld toe zone (Fig. 9). The spreading of copper along the FZ-Ti brazed interface was discontinuous (Fig. 9d), suggesting possible formation of uneven Ti2Cu phase. In comparison, there were no aluminum or nickel concentration at the brazed interface as shown on Fig. 9(b and c), which corresponded to the analysis of microstructure morphology in Fig. 8(d). To further confirm the phases observed, XRD analysis was carried out on MT-5 joint and the result is presented in Fig. 10. Diffraction peaks belonging to Ti3Al, Ti2Ni, Ti2Cu and Mg2Cu were detected, thus verifying the result obtained in Fig. 8.
(Fig. 11), the bonding mechanism of the joint is was proposed. Fig. 12 shows the typical schematic illustrations of MT-5 joining mechanism. During heating, the laser beam irradiated the brazing alloy which melted. As shown Fig. 12(a), the brazing alloy then dropped on the deposited coats and spread-out. Owing to the violent stirring at the direct irradiation and middle zones, the copper coating dissolved into the fusion zone and the melted nickel coating diffused into Ti and FZ. At elevated temperature, the titanium interacted with molten filler and part of it became energetic. As shown in Fig. 12(c), the Al, Ni and Cu in the molten pool diffused toward the titanium and interacted with Ti atoms, whereas, part of Cu atoms being in direct contact with Ti adhered to the Ti substrate towards the weld toe zone due to weaker stirring. According to the temperature field simulation analysis shown on Fig. 11, at laser irradiation (P1) and middle (P2) zones, the maximum temperature was higher than the formation temperature of Ti3Al phase (> 1180 °C). Thus, when the temperature dropped to 1180 °C, Ti3Al IMC was the first to be precipitated in these regions (Fig. 12d). Further decreased in temperature (984 °C), the Ni atoms diffused toward the titanium side scattering all over the precipitated Ti3Al phase at the laser
3.3. Bonding mechanism Based on the microstructure development, confirmation of the newly formed phases and the numerical simulation of temperature field 156
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
Fig. 12. Schematic illustration of bonding mechanism of MT-5 joint: (a–b) heating process; (c) atomic diffusion; (d–h) solidification behavior with the decrease in temperature. 2500
Tensile-shear fracture load (N)
irradiation zone, which resulted to Ti2Ni IMC formation (Fig. 12e). When the temperature dropped below 980 °C or below, only Ti2Cu phase was produced at the weld toe zone (Fig. 12f). The maximum temperature at the weld toe region (P3 in Fig. 11) was 1069.39 °C, which was less than the formation temperature of Ti3Al phase. This phenomenon was attributed to the insufficient Al and Ti diffusion force in this zone. Further decreased in temperature (700 °C), resulted to the concentration of Al, Ni and Mg atoms inside the molten pool of weld toe zone and Mg-Al-Ni reaction product was produced (Fig. 12g) [39]. At 487 °C temperature, the diffused copper atoms produced eutectic reaction with magnesium atoms (α-Mg + Mg2Cu) at the weld toe zone. Lastly, with decreased in temperature Mg17(Al,Cu)12 was produced at both laser irradiation and middle zones (Fig. 12h).
2000
1500
1000
500
3.4. Mechanical properties 0
The tensile shear performance of the joints under different coating condition is shown in Fig. 13. Under similar welding parameter, the tensile shear performance of the coated joints was more than the direct joint. For instance, when the nickel coating was thicker than the copper, referring to MT-1 and MT-2, the fracture load reached 1713 N, nearly 74% more than that of the direct joint (MT-0). The formation of Al-NiTi, Ti-Al and Ti-Ni IMCs hindered crack propagation and thus enhanced the interfacial strength, resulting to high fracture load observed. Despite the limited Cu content, the Cu atoms enhanced mutual reaction between Ti and Al as demonstrated in our previous study [37], which could improve metallurgical bonding and increase the joint mechanical property. Nevertheless, the thickness of the IMCs formed was greater than the recommended thickness of 10 µm (Fig. 5f) [40], which could adversely affect the joint strength. On the other hand, MT-3 and MT-4 joints presented the maximum mechanical resistance, which could be attributed to the Ti-Al and Ti-Ni mixed reaction layer formation whose
MT-0
MT-1
MT-2
MT-3
MT-4
MT-5
Sample condition Fig. 13. Tensile-shear fracture loads of laser welded-brazed Mg/Ti joints under different coating conditions.
thickness was less than the critical thickness. The joint mechanical performance attained a maximum value of 2020 N with MT-3 joint, growing approximately 100% more than the direct joint. For MT-5 joint, the thick Cu coating enhanced the Ti-Al reaction at the laser irradiation and middle zones and promoted Ti-Cu IMC formation at the weld toe zone, which improved the joint mechanical resistance. The fracture load reached 1660 N, which is lower than the remaining coated joints. The declined in the joint performance was associated with the more formation of denser Mg-Cu eutectic structure with increasing Cu coating in the fusion zone with network-like morphology being formed 157
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
Fig. 14. Mg/Cu-Ni coated Ti fracture surface under different coating conditions at different fracture modes: (a) fracture location of MT-3 with FZ fracture; (b) Mg side of (a); (c) higher magnification of the square area indicated in (b); (d) fracture location of MT-5 with interfacial failure; (e) laser irradiation zone; (f) higher magnification of square area indicated in (e); (g) middle zone; and (h) higher magnification of square area indicated in (g).
as shown in Fig. 8(d). The denser eutectic structure resulted in the decreased in the tensile-shear fracture load. Furthermore, the huge differences in thermal conductivity and thermal expansion coefficient between the magnesium and titanium caused residual stress formation [41]. Thus, the uneven Ti2Cu layer observed at the weld toe zone (Fig. 8c) resulted to high stress concentration, which could further deteriorate joint performance and caused interfacial failure [37]. With exception of MT-5 joint that fractured along the interface, all the coated joints fractured at fusion zone during the tensile shear testing. Fig. 14 shows the micrograph of the fracture surfaces in the two fracture patterns and the relevant EDS analysis are listed in Table 6. For joints that fractured at the fusion zone, the crack extended along the interface at the weld toe zone and then moved to the fusion zone as
Table 6 EDS analysis of various points indicated in Fig. 14 (at.%). Point
Mg
Al
Ti
Ni
Cu
Possible phases
P1 P2 P3
94.82 94.63 5.96
4.65 4.36 23.23
0.08 0.76 63.84
0.21 0.11 3.63
0.24 0.14 3.34
α-Mg α-Mg Ti3Al
shown in Fig. 14(a), indicating that the joint strength at the interface was more than that of the fusion zone. The fracture surface exhibit a typical dimple feature (Fig. 14c). However, as for the joint with interfacial failure mode, tear ridge was formed from the fracture surface at both direct irradiation and middle zones as shown in Fig. 14(e–g), 158
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
indicating severe deformation occurred during tensile shear testing. A number of particles (P3 in Fig. 14e) mingled with residual Mg (P2 in Fig. 14e). These particles contained 23.23 at.% Al and 63.84 at.% Ti, which were confirmed as Ti3Al [19].
Manuf. Technol. (2018) 1–17. [12] P. Villars, A. Prince, H. Okamoto, Handbook of ternary alloy phase diagrams, ASM Int. (1995). [13] S. Chen, M. Zhang, J. Huang, C. Cui, H. Zhang, X. Zhao, Microstructures and mechanical property of laser butt welding of titanium alloy to stainless steel, Mater. Des. 53 (2014) 504–511. [14] S. Chen, X. Huo, C. Guo, X. Wei, J. Huang, J. Yang, S. Lin, Interfacial characteristics of Ti/Al joint by vaporizing foil actuator welding, J. Mater. Process. Technol. 263 (2019) 73–81. [15] S. Chen, D. Yang, J. Yang, J. Huang, X. Zhao, Nanoscale structures of the interfacial reaction layers between molten aluminium and solid steel based on thermophysical simulations, J. Alloys Compd. 739 (2018) 184–189. [16] A.M. Nasiri, Laser Brazing of Magnesium to Steel Sheet, 2013. [17] M. Aonuma, K. Nakata, Effect of alloying elements on interface microstructure of Mg–Al–Zn magnesium alloys and titanium joint by friction stir welding, Mater. Sci. Eng. B 161 (1) (2009) 46–49. [18] R. Cao, T. Wang, C. Wang, Z. Feng, Q. Lin, J.H. Chen, Cold metal transfer weldingbrazing of pure titanium TA2 to magnesium alloy AZ31B, J. Alloys Compd. 605 (2014) 12–20. [19] C. Tan, X. Song, B. Chen, L. Li, J. Feng, Enhanced interfacial reaction and mechanical properties of laser welded-brazed Mg/Ti joints with Al element from filler, Mater. Lett. 167 (2016) 38–42. [20] C. Tan, B. Chen, S. Meng, K. Zhang, X. Song, L. Zhou, J. Feng, Microstructure and mechanical properties of laser welded-brazed Mg/Ti joints with AZ91 Mg based filler, Mater. Des. 99 (2016) 127–134. [21] C. Zang, J. Liu, C. Tan, K. Zhang, X. Song, B. Chen, L. Li, J. Feng, Laser conduction welding characteristics of dissimilar metals Mg/Ti with Al interlayer, J. Manuf. Process. 32 (2018) 595–605. [22] A.M. Atieh, T.I. Khan, Effect of process parameters on semi-solid TLP bonding of Ti–6Al–4V to Mg–AZ31, J. Mater. Sci. 48 (19) (2013) 6737–6745. [23] A.M. Atieh, T.I. Khan, TLP bonding of Ti-6Al-4V and Mg-AZ31 alloys using pure Ni electro-deposited coats, J. Mater. Process. Technol. 214 (12) (2014) 3158–3168. [24] A.M. Atieh, T.I. Khan, Effect of interlayer thickness on joint formation between Ti6Al-4V and Mg-AZ31 alloys, J. Mater. Eng. Perform. 23 (11) (2014) 4042–4054. [25] A.M. Atieh, T.I. Khan, Application of Ni and Cu nanoparticles in transient liquid phase (TLP) bonding of Ti-6Al-4V and Mg-AZ31 alloys, J. Mater. Sci. 49 (22) (2014) 7648–7658. [26] C. Tan, C. Zang, X. Zhao, H. Xia, Q. Lu, X. Song, B. Chen, G. Wang, Influence of Nicoating thickness on laser lap welding-brazing of Mg/Ti, Opt. Laser Technol. 108 (2018) 378–391. [27] C.W. Tan, Q.S. Lu, B. Chen, X.G. Song, L.Q. Li, J.C. Feng, Y. Wang, Influence of laser power on microstructure and mechanical properties of laser welded-brazed Mg to Ni coated Ti alloys, Opt. Laser Technol. 89 (2017) 156–167. [28] M. Ding, S. Liu, Y. Zheng, Y. Wang, H. Li, W. Xing, X. Yu, P. Dong, TIG–MIG hybrid welding of ferritic stainless steels and magnesium alloys with Cu interlayer of different thickness, Mater. Des. 88 (2015) 375–383. [29] M. Handbook, Vol. 5: Surface Cleaning, Finishing, and Coating, American Soc. for Metals, 1982, 412–416. [30] Y.-J. Tan, K.Y. Lim, Understanding and improving the uniformity of electrodeposition, Surf. Coat. Technol. 167 (2–3) (2003) 255–262. [31] S.T. Auwal, S. Ramesh, F. Yusof, C.W. Tan, Z.Q. Zhang, X.Y. Zhao, S.M. Manladan, Comparative study on characteristics of laser welded-brazed AZ31/Ti-6Al-4V lap joints with and without coatings, Int. J. Adv. Manuf. Technol. (2018). [32] C. Tan, X. Song, S. Meng, B. Chen, L. Li, J. Feng, Laser welding-brazing of Mg to stainless steel: joining characteristics, interfacial microstructure, and mechanical properties, Int. J. Adv. Manuf. Technol. 86 (1–4) (2016) 203–213. [33] C. Xu, G.M. Sheng, H. Wang, K. Feng, X.J. Yuan, Tungsten inert gas welding-brazing of AZ31B magnesium Alloy to TC4 titanium alloy, J. Mater. Sci. Technol. 32 (2) (2016) 167–171. [34] K. Lee, P. Nash, The Al-Ni-Ti system (Aluminum-Nickel-Titanium), J. Phase Equilibria 12 (5) (1991) 551–562. [35] V.T. Witusiewicz, A.A. Bondar, U. Hecht, S. Rex, T.Y. Velikanova, The Al–B–Nb–Ti system: III. Thermodynamic re-evaluation of the constituent binary system Al–Ti, J. Alloys Compd. 465 (1) (2008) 64–77. [36] S. Du, G. Liu, M. Wang, Microstructure and properties of transient liquid-phase diffusion bonded joint of AZ31B/Cu dissimilar metal, Chin. J. Nonferrous (In Chinese) Met. 5 (2013) 1255–1261. [37] Z. Zhang, C. Tan, X. Zhao, B. Chen, X. Song, H. Zhao, Influence of Cu coating thickness on interfacial reactions in laser welding-brazing of Mg to Ti, J. Mater. Process. Technol. 261 (2018) 61–73. [38] X.Y. Wang, D.Q. Sun, Y. Sun, Influence of Cu-interlayer thickness on microstructures and mechanical properties of MIG-welded Mg-steel joints, J. Mater. Eng. Perform. 25 (3) (2016) 910–920. [39] V. Raghavan, Al-Mg-Ni (Aluminum-Magnesium-Nickel), J. Phase Equilibria Diffus. 30 (3) (2009) 274–275. [40] H. Laukant, C. Wallmann, M. Müller, M. Korte, B. Stirn, H.-G. Haldenwanger, U. Glatzel, Fluxless laser beam joining of aluminium with zinc coated steel, Sci. Technol. Weld. Joi. 10 (2) (2005) 219–226. [41] S.T. Auwal, S. Ramesh, F. Yusof, S.M. Manladan, A review on laser beam welding of titanium alloys, Int. J. Adv. Manuf. Technol. 97 (1) (2018) 1071–1098.
4. Conclusions The effect of the electrodeposited Cu and Ni contents on microstructure development and joint fracture load were analyzed. The reaction products were influenced by the interlayer element content. The major conclusions of this study can be summarized as follows: 1. Compared with uncoated joint (MT-0), superior joints appearance and cross-sections were observed with coated joints (MT1-MT5). Thus, the feasibility of this process depends strongly on the preexisting Cu-Ni layer on the Ti sheet that promotes wetting of the AZ92 filler alloy. 2. Depending on the interlayer Cu and Ni contents, different reaction products formed inside the joint region. Nevertheless, at optimum interlayer contents (i.e. joint with comparable Ni and Cu coating thicknesses MT-3), Ti2Ni mingled with Ti3Al IMC layer was produced at the interface and that the thickness of the reaction layer was below the critical thickness of 10 µm. 3. The maximum tensile-shear fracture load attained 2020 N for the joint with comparable Ni and Cu coating thicknesses (MT-3), which was 100% more than that of the direct joint. With exception of MT-5 joint that fractured at the brazed interface, all the coated joints fractured with fusion zone fracture mode. For the joint with interfacial failure mode, tear ridge was observed at the fracture surface, whereas, the fusion zone fracture surfaces displayed a typical dimple feature. Acknowledgement of the funding sources The authors would like to acknowledge the support provided by University of Malaya (Project No. GPF073A-2018), National Natural Science Foundation of China (Project No. 51504074 and 51875129) and Key Research & Development Program in Shandong Province (Grant No. 2017CXGC0811 and 2017GGX30147). References [1] L. Liu, Welding and Joining of Magnesium Alloys, Elsevier, 2010. [2] X. Cao, M. Jahazi, J. Immarigeon, W. Wallace, A review of laser welding techniques for magnesium alloys, J. Mater. Process. Technol. 171 (2) (2006) 188–204. [3] M.K. Kulekci, Magnesium and its alloys applications in automotive industry, Int. J. Adv. Manuf. Technol. 39 (9–10) (2008) 851–865. [4] C. Blawert, N. Hort, K. Kainer, Automotive applications of magnesium and its alloys, Trans. Indian Inst. Met. 57 (4) (2004) 397–408. [5] S. Manladan, F. Yusof, S. Ramesh, M. Fadzil, Z. Luo, S. Ao, A review on resistance spot welding of aluminum alloys, Int. J. Adv. Manuf. Technol. 90 (1–4) (2017) 605–634. [6] L. Li, C. Tan, Y. Chen, W. Guo, C. Mei, CO2 laser welding–brazing characteristics of dissimilar metals AZ31B Mg alloy to Zn coated dual phase steel with Mg based filler, J. Mater. Process. Technol. 213 (3) (2013) 361–375. [7] S.M. Manladan, F. Yusof, S. Ramesh, M. Fadzil, A review on resistance spot welding of magnesium alloys, Int. J. Adv. Manuf. Technol. 86 (5) (2016) 1805–1825. [8] S. Manladan, F. Yusof, S. Ramesh, Y. Zhang, Z. Luo, Z. Ling, Microstructure and mechanical properties of resistance spot welded in welding-brazing mode and resistance element welded magnesium alloy/austenitic stainless steel joints, J. Mater. Process. Technol. 250 (2017) 45–54. [9] E. Akman, A. Demir, T. Canel, T. Sinmazcelik, Laser welding of Ti6Al4V titanium alloys, J. Mater. Process. Technol. 209 (8) (2009) 3705–3713. [10] C. Xu, G. Sheng, Y. Deng, X. Yuan, K. Tang, Microstructure and mechanical properties of tungsten inert gas welded–brazed Mg/Ti lap joints, Sci. Technol. Weld. Joi. 19 (5) (2014) 443–450. [11] Y.M. Baqer, S. Ramesh, F. Yusof, S. Manladan, Challenges and advances in laser welding of dissimilar light alloys: Al/Mg, Al/Ti, and Mg/Ti alloys, Int. J. Adv.
159
Contents lists available at ScienceDirect
Optics and Laser Technology journal homepage: www.elsevier.com/locate/optlastec
Full length article
Effect of copper-nickel interlayer thickness on laser welding-brazing of Mg/ Ti alloy ⁎
T
⁎⁎
S.T. Auwala,b, S. Ramesha, , Zequn Zhangc,d, F. Yusofa, Jinge Liuc, Caiwang Tanc,d, , S.M. Manladane, F. Tarlochanf a
Center for Advanced Manufacturing and Materials Processing, Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia b Department of Mechanical Engineering, Faculty of Engineering, Kano University of Science and Technology, Wudil, 3244 Kano, Nigeria c State Key Laboratory of Advanced Welding and Joining, Harbin Institute of Technology, Harbin 150001, China d Shandong Provincial Key Laboratory of Special Welding Technology, Harbin Institute of Technology at Weihai, Weihai 264209, China e Department of Mechanical Engineering, Faculty of Engineering, Bayero University, Kano, 3011 Kano, Nigeria f Department of Mechanical and Industrial Engineering, College of Engineering, Qatar University, Doha, Qatar
H I GH L IG H T S
Mg alloy to titanium with varied thickness Cu-Ni coating using laser welding-brazing process. • Join interlayer elements content were found to exert varied strengthening influence. • Different • The optimum fracture load was obtained with appropriate thickness of Cu-Ni coating.
A R T I C LE I N FO
A B S T R A C T
Keywords: Mg alloy Laser welding-brazing Mechanical properties Interfacial reaction Microstructure Ti alloy
Dissimilar lap joining of AZ31B Mg alloy to Cu-Ni coated Ti-6Al-4V was carried out by laser welding-brazing method. The effect of Cu and Ni contents on interfacial reaction and joint fracture load were analyzed. For the joint in which the Ni coating (15.36 µm) was thicker than the Cu coating (5.47 µm), thick intermetallic compound (IMC) composed of a mixture of light gray Al-Ni-Ti + Ti3Al + Ti2Ni phases mingled with dark gray Ti3Al was produced at the interface. In comparison, a mixed interfacial reaction layer consisted of Ti2Ni and Ti3Al was formed from the direct irradiation zone to the weld toe zone of the joint with comparable Ni and Cu coating thicknesses (10.78 µm Cu–9.30 µm Ni). In this case, the thickness of the mixed layer was below the critical thickness of 10 µm. For the joint in which the Cu coating is much higher than the Ni coating thickness (17.12 µm Cu–4.23 µm Ni), Ti3Al and Ti2Ni mixed reaction layer was produced at the brazed interface of direct irradiation zone, whereas, only Ti3Al phase was formed at the middle zone. At the weld toe zone, Ti2Cu uneven interfacial reaction layer was evolved. With increasing Cu and decreasing Ni coating thicknesses, the fracture load first increased and then slightly decreased, the maximum tensile-shear fracture load attained 2020 N for joints with comparable Cu and Ni coating thicknesses. This is twofold higher than that of uncoated joint. The tensile-shear investigation showed that the joint would fracture at the fusion zone when the coating thickness of Ni was comparable or higher than Cu. In contrast, interfacial failure was observed when the thickness of Cu was much higher than the Ni. For the joint with interfacial failure mode, tear ridge was observed from the fracture surface, whereas, the fusion zone fracture surfaces was noted to display a typical dimple feature.
⁎ Corresponding author at: Center for Advanced Manufacturing and Materials Processing, Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia. ⁎⁎ Co-corresponding author at: Shandong Provincial Key Laboratory of Special Welding Technology, Harbin Institute of Technology at Weihai, Weihai 264209, China. E-mail addresses: [email protected] (S. Ramesh), [email protected] (C. Tan).
https://doi.org/10.1016/j.optlastec.2019.02.024 Received 12 October 2018; Received in revised form 21 November 2018; Accepted 3 February 2019 0030-3992/ © 2019 Elsevier Ltd. All rights reserved.
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
1. Introduction
the Mg-Ni interface, whereas, solid-state diffusion occurred at the Ni-Ti interface. The analysis of the joint quality revealed that the maximum tensile-shear of only 39 MPa was achieved. The low shear strength observed was attributed to the fact that the foils thickness produced a thick IMCs. These IMCs reduced the direct contact area and hence, affected bond strength of the joint. However, further increased in the joint shear strength was achieved with electroplated coatings [23,24]. The electrodeposition of thin coats was found to be beneficial in improving the flow and uniformity of the liquid. In addition, the authors reported maximum shear strength with an electrodeposited Ni coating containing a dispersion of nanoparticles of pure Cu [25]. The addition of nanoscale copper dispersion in Ni coating facilitates the bonding process through shorter diffusion distances. In our recent work, the benefit of using Ni coating to join Mg/Ti by LWB method was explored. The presence of the addition of Ni coating was found to be beneficial in improving the Mg/Ti joint performance [26,27]. These studies suggested that the addition of Cu and Ni in the form of thin coats proved to be effective in realizing an interfacial reaction between the immiscible Mg/Ti systems with excellent joint quality. Furthermore, the interlayer elements content could affect the dissimilar joints’ microstructural evolution and mechanical properties [24,28]. Thus, different thickness of interlayer is expected to exert varied strengthening influence. In this study, the effect of laser weldedbrazed Mg/Ti lap joints with and without electrodeposited Cu-Ni layer of different thickness was studied and defined. The joint appearance, microstructure development and joint fracture load were examined. The joining mechanism of the Mg/Ti joints with varied Cu and Ni thicknesses was also discussed.
Magnesium (Mg) has been described as a green engineering material and one of the most promising material categories in the 21st century [1]. It has unique properties such as formability, good recyclability and high specific strength [2,3]. Because of these properties, Mg alloys are being used to replace not only heavy metals such as steels, but also light alloys, including aluminum alloys, in a wide range of applications [4,5]. Recently, they attract special attention in automotive and aerospace industries where weight reduction is crucial [6–8]. On the other hand, titanium (Ti) and its alloys have been used extensively in aerospace, medical, chemical, biomaterial, and petrochemical industries for their high creep resistance, excellent corrosion resistance and high strength to weight ratio [9]. The ability to join Mg alloys to Ti effectively has received great interest in high-technology based industries such as aerospace and automotive manufacturing sectors where light-weight components are crucial in order to exploit the advantages of both materials, reduce fuel consumption, greenhouse gases and improve performance of energy converting system [10,11]. However, the significant differences in their physical (e.g. melting point, coefficient of thermal expansion, thermal conductivity) and metallurgical characteristics (nearly zero solubility) make joining them together very difficult [12]. Although, there were reports of direct joining of dissimilar metals such as Ti alloy to steel [13] and Ti alloy to Al [14,15] without the use of interlayer material or filler metal, this would depend on the base metals to be join and their ability to form an intermetallic layer as well as careful manipulation of the welding parameters. In view of the immiscible characteristics of Mg/Ti alloys, thus, a direct and economical approach to joining them would be to use a suitable interlayer material during the welding process. In choosing a suitable interlayer would largely depends on the material composition. Ideally, the interlayer material should be able to react with the base metals to provide excellent wetting and bonding without generating thick layers of hard and brittle intermetallic compounds (IMCs) at the joint interface. Moreover, when choosing the joining process, minimization of the thickness of any brittle IMCs along the interfaces of the Mg alloy-interlayer-Ti joint and minimization of intermixing between the Mg and Ti in the molten-state are the main factors that must be considered [16]. Hence, selecting an appropriate welding technique and interlayer material composition to produce and control interfacial layers became the focus of Mg/Ti joining. Based on the existing literature, intermediate elements such as Al, Ni have been employed to achieve interfacial bonding between Mg/Ti by various joining techniques. During friction stir welding [17], mutual diffusion of aluminum from the Mg-Al-Zn alloy and titanium was achieved as a result of combined action of pressure and stirring. The analysis of the interface characteristics showed that TiAl3 intermetallic compound was produced. With increasing the aluminum content in the Mg-Al-Zn base metal, the thickness of the TiAl3 phase increases, whereas, the tensile strength decrease. Cao et al. [18] reported that for cold metal transfer (CMT) welded Mg to Ti alloy, the Al from the Mg filler metal performed an important role in joining the immiscible couple. Interfacial reaction layer identified as Ti3Al phase was observed. In our previous work, joining Mg/Ti with AZ91 filler metal was achieved by laser welding brazing (LWB) method [19,20]. The reaction between the Al from the Mg-Al-Zn filler metal and titanium resulted to Ti3Al IMC layer formation at the interface. Recently, Zang et al. [21] examined the influence of Al interlayer thickness on laser conduction welded AZ31/Ti-6Al-4V dissimilar joint characteristics. It was found that TiAl3 phase was produced at interface. However, increasing the Al thickness had little effect on the newly formed reaction layer. The results of these studies proved that the Al intermediate element performed an important role in joining the immiscible couple. On the other hand, Atieh et al. [22] investigated the diffusion behavior of Mg alloy to Ti joint with Ni foil. It was found that eutectic reaction was formed at
2. Experimental 2.1. Materials and electrodeposition process The materials used in this study were commercially available 1 mm thick Mg alloy (AZ31B) (95–97 wt% Mg, 2.5–3.5 wt% Al, 0.5–1.5 wt% Zn) and 1.5 mm Ti alloy (Ti-6Al-4V). Both alloys were cut into specimens with length of 100 mm and width of 55 mm. Furthermore, commercially available AZ92 filler wire (89 wt% Mg, 9 wt% Al, 2 wt% Zn) with diameter of 3.5 mm was employed. The electrodeposition of the pure copper and nickel on Ti substrate was carried out based on American Society for Metals (ASM) standard [29]. Prior to electrodeposition, Ti alloy surfaces was acid pickled to remove the oxide film using an acid solution containing 15% hydrochloric acid, 5% hydrofluoric acid and 80% deionized water for 3 min and then cleaned with water. The electroplating of the pure Cu and Ni followed a sequence of Cu and finally Ni (Cu-Ni) on the Ti substrate was performed in a 500 mL beaker containing the plating solution as given in Table 1 and Table 2. During the electro-plating process, the interlayer was used as the anode while the Ti was the cathode. Furthermore, DC power was used during the Cu-Ni electrodeposition process, with pH value of 5 at 35 °C. Furthermore, the magnetic stirrer to stir the bath was kept at 200 rpm. In order to get a 20 µm thick uniform Cu and Ni layer on the Ti substrate, different cathode current densities and electrodeposition times were tried. When the electroplating was carried out at current density of 0.3 A/dm2 for 60 min, an approximate 20 µm Cu and Ni layer on the Ti substrate of different metals layer contents were observed. For each coating condition, the average of at least five coating thicknesses was measured and listed in Table 3. The scanning electron microscope (SEM) cross sectional micrograph of various Table 1 Composition of Cu electro-plating solution.
150
Compositions
K4P2O7.3H20
C6H17N3O7
CuSO4·5H2O
Na2HPO4·12H2O
Amount (g/L)
220
20
48
24
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
was cut into pieces and preset at the join interface before welding. The Mg filler metal was irradiated with the vertical laser beam. To increase the spot size of the laser beam, the beam was defocused. Thus, the laser beam diameter of 2.4 mm was used at the defocused distance. The joint was shielded from oxidation with Ar gas having a purity of 99.99%. The LWB parameters employed in the current work is shown in Table 4.
Table 2 Composition of Ni electro-plating solution. Compositions
NiSO4·6H2O
NaCl
Na2SO4
H3BO3
MgSO4·7H2O
Amount (g/L)
180
10
70
30
60
Table 3 Electroplating time and corresponding Cu and Ni coating thicknesses. Joint
MT-0 MT-1 MT-2 MT-3 MT-4 MT-5
Electroplating time (min)
Coating thickness (µm)
Cu
Ni
Cu
Ni
0 10 20 30 40 50
0 50 40 30 20 10
0.00 5.47 8.41 10.78 14.49 17.12
0.00 15.36 12.07 9.30 6.97 4.23
2.3. Characterization Total
Standard procedures were adopted to prepare the metallographic specimens after welding. OLYMPUS-DSX510 Optical microscopy (OM) and Zeiss Merlin Compact SEM equipped with EDS were adopted to examine the cross sectional microstructures morphologies and fracture surface. The reaction products formed were verified by X-ray diffraction (XRD) (Panalytical X’pert 3 Powder) using a copper target at 35 mA and 40 kV. The 2θ scans were carried out from 20° to 90°, at a step size of 0.04° and 1 s per step. The joint quality is estimated through room temperature tensile-shear testing on 10 mm wide specimens machined from the joints using Instron 5967 tensile tester at a cross heat speed of 1 mm/min. Shims were added to each specimen to minimize induced couples. An average of at least 3 tensile shear testing specimens was used to evaluate the mechanical performance of the joint.
0.00 20.83 20.48 20.08 21.46 21.35
coating layer thicknesses with uniform characteristics is shown in Fig. 1. In addition, Fig. 1(f) demonstrates the relationship between the coating thickness and the samples coating condition. To assess the uniformity of the Cu-Ni coating of the final deposit, energy-dispersive X-ray spectrometer (EDS) mapping was performed as typically presented in Fig. 2. As shown on Fig. 2 relatively uniform coat thickness was achieved. Nevertheless, small discontinuities is inevitable as observed in Fig. 2(b), which could be associated with electrochemical heterogeneity over a work-piece surface under electrodeposition, which resulted in localization of electrodeposition reactions and rates [30].
3. Results and discussion 3.1. Weld appearance and cross sections The joints typical appearance and cross section at various coating conditions is presented in Fig. 3. Under the various coating thicknesses, smooth weld surface were observed, suggesting that the addition of the thin deposited coats enhanced the wettability of the brazing alloy. For comparison, joining of the uncoated joint was also carried out as shown in Fig. 3(a). As illustrated on Fig. 3(a), large contact angle was formed, which was associated with unwetted filler alloy. This behavior is attributed to poor affinity of the Mg/Ti. Furthermore, it can be observed from the corresponding joints cross-sections that between the brazing alloy and Mg base metal (BM), fusion was observed. In comparison,
2.2. Laser welding-brazing process A 6-kW fiber laser (IPG-6000) having a beam parameter product of 8 mm·mrad and 1070 nm wavelength was employed in the current work. The join design adopted was lap configuration with magnesium base metal placed at the top of the titanium. A stainless-steel wire brush was used to clean the magnesium surface oxide layer. The brazing alloy
Fig. 1. SEM cross-sectional micrographs of the Cu-Ni coating layer at different electroplating time: (a) MT-1; (b) MT-2; (c) MT-3; (d) MT-4; (e) MT-5 and (f) Relationship between the coating thickness and the samples coating condition. 151
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
Fig. 2. EDS mapping analysis of Cu-Ni layer for MT-3: (a) SEM image; (b–d) Cu, Ni, Ti; and (e) line scanning result. Table 4 Laser welding-brazing parameters used in the experiment. Laser power (W)
1400
Welding speed (m/min) Defocus distance from Ti substrate (mm) Flow rate of the Ar shielding gas (L/min)
0.3 +30 20
brazing was formed at fusion zone/Ti substrate. Fig. 3(a–d) show the fusion lines differentiating the fusion zone (FZ) from the magnesium base metal. Nevertheless, flat titanium surfaces were produced, confirming that the titanium remained un-melted. This could be associated with positive defocus distance employed in the current work. To further compare the wettability of the brazing alloy in the coated and uncoated joining situations, contact angle and seam width were drawn in Fig. 4. A contact angle of 50.8˚ and a seam width of 5.42 mm were observed for the uncoated joint, whereas, contact angle of less than 35˚ and seam width of up to 6.47 mm for Cu-Ni coated joints were produced. However, slight variation of the seam width and contact angle was observed under different coating conditions. 3.2. Interfacial microstructure The interfacial microstructure of the uncoated (direct) and Cu-Ni coated joints at various coating conditions were studied under similar welding parameters. For the uncoated joint, relatively smooth interface was observed, suggesting no obvious reaction layer. However, ultrathin Ti3Al IMC was formed across the brazed interface. Detail analysis of the microstructural evolution and the related bonding mechanism were reported in our previous work [20,31]. In this study, emphasis was given to Cu-Ni coated joints. Figs. 5, 6 and 8 show the interfacial microstructures under various coating condition. After welding, the original coating layer was not visible, suggesting suitable heat-input adopted in this work. The high temperature gradient during LWB process caused the interfacial microstructure along the interface to vary [32]. Thus, the FZ-Ti brazed interface was divided into 3 parts namely: laser irradiation, middle and weld toe zones. Fig. 5(a–e) show the SEM images of MT-1 joint, i.e. Ni coating
Fig. 3. Mg/Cu-Ni coated Ti laser welded-brazed joints with various Cu and Ni coating thicknesses: (a) MT-0; (b) MT-1; (c) MT-2; (d) MT-3; (e) MT-4; and (f) MT-5.
thicker than that of Cu (5.47 µm Cu-15.36 µm Ni). EDS analyses were performed to characterize phases formed as presented in Table 5. The phases formed at different zones were homogenous. At the FZ, the EDS analysis showed that the dark phase (Point P1 in Fig. 5a) consisted of 90.96 at.% Mg, 9.04 at.% Al, suggesting α-Mg formation. Meanwhile, 152
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
the secondary phases (P2 in Fig. 5a) precipitated in the FZ was identified as Mg17Al12 since it contained 27.30 at.% Al, 72.70 at.% Mg. The formation of Mg17Al12 phase was formed during eutectic reaction [33]. In addition, bulk phases P3 of Fig. 5b aggregated in the fusion zone nearby the interface, which consisted of mainly 39.62 at.% Al, 42.69 at. % Mg, 17.69 at.% Ni, suggesting they were Mg-Al-Ni ternary phases [27]. The formation of these bulk phases could be associated with melting and diffusion of Ni coating into the brazing alloy. At the brazed interface, light gray (P4 in Fig. 5b) and dark gray (P5 in Fig. 5b) mixed reaction layer was observed. The light gray phase consisted of 33.07 at. % Al, 42.31 at.% Ti, 19.84 at.% Ni, 4.39 at.% Cu, which was identified as IMC that composed of a mixture of Al-Ni-Ti + Ti3Al + Ti2Ni phases [34], whereas, the dark gray phase contained 29.66 at.% Al, 61.87 at.% Ti, 6.10 at.% Ni, 2.12 at.% Cu, suggesting it was Ti3Al based on Ti-Al phase diagram [35]. An EDS line scan was carried out across the middle zone (Fig. 5f). The results disclosed that titanium content increased while magnesium decreased with concentration of nickel and aluminum at the interface, signifying the possible formation of Al-Ni-Ti, TiNi and Ti-Al IMCs. Furthermore, no Cu segregation was noted, indicating that the Cu content was low and under the action of flow and vortex, the thin Cu layer could have dissolved into the liquid filler. Fig. 6(a–e) present the morphologies of MT-3 joint interfacial microstructures, i.e. comparable Ni and Cu coating thicknesses (10.78 µm
Fig. 4. Varied contact angles and seam widths under different coating conditions.
Fig. 5. Interfacial microstructure morphologies of the MT-1 joint: (a) direct irradiation zone; (b) higher magnification of region b; (c) middle zone; (d) weld toe zone; (e) higher magnification of region e; (f) line scanning result. 153
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
Fig. 6. Interfacial microstructure morphologies of the MT-3 joint: (a) direct irradiation zone; (b) higher magnification of region b; (c) middle zone; (d) weld toe zone; (e) higher magnification of region e; (f) line scanning result.
to the relatively limited Ni atoms. On the other hand, a mixed layer of light grey (P8 in Fig. 6b) and dark grey (P9 in Fig. 6b) was produced at interface. The light grey phase consisted of 11.30 at.% Al, 56.81 at.% Ti, 26.82 at.% Ni and 4.14 at.% Cu, which was identified as Ti2Ni IMC, whereas, the light grey phase contained 22.72 at.% Al, 68.78 at.% Ti, 5.12 at.% Ni and 3.25 at.% Cu, suggesting the possible formation of Ti3Al IMC. Noteworthy, compared to MT-1 joint the thickness of the interfacial reaction layer decreased. At the weld toe zone, the precipitation of second phase (P10 in Fig. 6d) inside the FZ increased. The second phases consisted of 13.57 at.% Al, 79.64 at.% Mg, 0.14 at.% Ti, 0.30 at.% Ni, 6.35 at.% Cu, which were identified as Mg17(Al,Cu)12 due to the little copper percentage. Study have shown that the Mg17(Al,Cu)12 formation was attributed to the replacement of aluminum atoms in Mg17Al12 by copper because of their comparable atomic radius [36]. Some dendritic structures P11 dispersed about αMg matrix with lamellar morphology were also observed nearby the interface (Fig. 6e). These dendritic phases consisted of 15.87 at.% Al, 67.62 at.% Mg, 4.20 at.% Ni and 12.21 at.% Cu. Coupled with our with our prior work [37], these phases were characterized as Mg-Cu eutectic structure (α-Mg + Mg2Cu). Eutectic reaction occurred at 485 °C: L ↔ Mg + Mg2Cu. The eutectic structure was also observed during welding of dissimilar Mg to steel with Cu interlayer [38]. EDS line scanning results (Fig. 6f) across the weld toe zone of FZ-Ti brazed interface
Table 5 EDS results of various points indicated in Figs. 5, 6 and 8 (at.%). Point
Mg
Al
Ti
Ni
Cu
Possible phases
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15
90.96 72.70 42.60 0.40 0.25 0.03 42.53 0.93 0.13 79.64 67.62 0.72 6.84 1.32 0.30
9.04 27.30 39.62 33.07 29.66 17.99 29.75 11.30 22.72 13.57 15.87 12.69 20.70 27.35 6.69
— — — 42.35 61.87 71.57 0.90 56.81 68.78 0.14 0.10 59.00 60.89 67.19 62.52
— — 17.78 19.39 6.10 8.08 25.32 26.82 5.12 0.30 4.20 22.12 6.45 2.27 4.38
— — — 4.79 2.12 2.33 1.50 4.14 3.25 6.35 12.21 5.47 5.12 1.87 26.11
α-Mg Mg17Al12 + α-Mg Mg-Al-Ni Al-Ni-Ti + Ti3Al + Ti2Ni Ti3Al Ti3Al Mg-Al-Ni Ti2Ni Ti3Al Mg17(Al, Cu)12 + α-Mg α-Mg + Mg2Cu Ti2Ni Ti3Al Ti3Al Ti2Cu
Cu–9.30 µm Ni). The phases formed at different zones were observed to be inhomogeneous. At the laser irradiation and middle zones adjacent to the brazed interface, some bulk phases (P7 in Fig. 6a) identified as Mg-Al-Ni compounds were observed. However, compared with MT-1 joint (Fig. 5), the distribution area of these bulk phases was small, due 154
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
Fig. 7. EDS mapping of MT-3 joint at the weld toe region: (a) corresponding SEM image; (b–f) Al, Ni, Cu, Ti, and Mg.
be inhomogeneous. At the laser irradiation zone, Mg17(Al,Cu)12 secondary particles precipitated in the entire FZ. At the joint brazed interface, an interfacial reaction layer comprising light grey P12 and dark grey P13 phases, which were confirmed as Ti2Ni and Ti3Al, respectively was observed. Owing to the limited Ni contents, only Ti3Al (P14 in Fig. 8b) formed at the middle zone, according to the compositions examination. Furthermore, limited content of Ni content coupled with strong stirring caused the Mg-Al-Ni bulk compounds to be concentrated at the FZ of weld toe zone. In addition, large amount of Mg-Cu dendritic structures with lamellar morphology dispersed around the α-Mg matrix as shown in Fig. 8(d). Owing to the high Cu contents, the distributions
confirmed the Ti-Al and Ti-Ni IMCs formation. In contrast, the Cu concentration observed in Fig. 6(f) could be associated with the presence of Mg-Cu phase stick to the interface. EDS mapping analysis at the weld toe zone (Fig. 7) showed that the Al and Ni contents concentrated along the interface as shown in Fig. 7(b and c), indicating possible formation of Ti3Al and Ti2Ni IMCs with thickness of approximately 5.34 µm. Fig. 7(d) indicates that the copper atoms diffused into the FZ to contribute in the microstructural evolution. Fig. 8(a–d) show the interfacial microstructure morphologies of MT5 joint where the Cu coating thickness higher than that of Ni (17.12 µm Cu–4.23 µm Ni). The phases formed at different zones were observed to
Fig. 8. Interfacial microstructure morphologies of the MT-5 joint: (a) direct irradiation zone; (b) middle zone; (c) weld toe zone; and (d) higher magnification of region d. 155
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
Fig. 9. EDS mapping of MT-5 joint at the weld toe region: (a) corresponding SEM image; (b–f) Al, Ni, Cu, Ti, and Mg.
Fig. 10. XRD analysis of the MT-5 joint.
Fig. 11. Simulation of thermal cycles of MT-5 joint at various zones.
of this denser eutectic structure was more than that of MT-3 joint. An uneven IMC layer P15 was formed at the interface of the weld toe zone. This uneven layer contained 6.69 at.% Al, 62.52 at.% Ti, 4.38 at.% Ni and 26.11 at.% Cu, thus suggesting it was Ti2Cu. IMCs containing Ti-Al was not observed in this region due to insufficient Ti and Al diffusion force. EDS mapping was carried out at the weld toe zone (Fig. 9). The spreading of copper along the FZ-Ti brazed interface was discontinuous (Fig. 9d), suggesting possible formation of uneven Ti2Cu phase. In comparison, there were no aluminum or nickel concentration at the brazed interface as shown on Fig. 9(b and c), which corresponded to the analysis of microstructure morphology in Fig. 8(d). To further confirm the phases observed, XRD analysis was carried out on MT-5 joint and the result is presented in Fig. 10. Diffraction peaks belonging to Ti3Al, Ti2Ni, Ti2Cu and Mg2Cu were detected, thus verifying the result obtained in Fig. 8.
(Fig. 11), the bonding mechanism of the joint is was proposed. Fig. 12 shows the typical schematic illustrations of MT-5 joining mechanism. During heating, the laser beam irradiated the brazing alloy which melted. As shown Fig. 12(a), the brazing alloy then dropped on the deposited coats and spread-out. Owing to the violent stirring at the direct irradiation and middle zones, the copper coating dissolved into the fusion zone and the melted nickel coating diffused into Ti and FZ. At elevated temperature, the titanium interacted with molten filler and part of it became energetic. As shown in Fig. 12(c), the Al, Ni and Cu in the molten pool diffused toward the titanium and interacted with Ti atoms, whereas, part of Cu atoms being in direct contact with Ti adhered to the Ti substrate towards the weld toe zone due to weaker stirring. According to the temperature field simulation analysis shown on Fig. 11, at laser irradiation (P1) and middle (P2) zones, the maximum temperature was higher than the formation temperature of Ti3Al phase (> 1180 °C). Thus, when the temperature dropped to 1180 °C, Ti3Al IMC was the first to be precipitated in these regions (Fig. 12d). Further decreased in temperature (984 °C), the Ni atoms diffused toward the titanium side scattering all over the precipitated Ti3Al phase at the laser
3.3. Bonding mechanism Based on the microstructure development, confirmation of the newly formed phases and the numerical simulation of temperature field 156
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
Fig. 12. Schematic illustration of bonding mechanism of MT-5 joint: (a–b) heating process; (c) atomic diffusion; (d–h) solidification behavior with the decrease in temperature. 2500
Tensile-shear fracture load (N)
irradiation zone, which resulted to Ti2Ni IMC formation (Fig. 12e). When the temperature dropped below 980 °C or below, only Ti2Cu phase was produced at the weld toe zone (Fig. 12f). The maximum temperature at the weld toe region (P3 in Fig. 11) was 1069.39 °C, which was less than the formation temperature of Ti3Al phase. This phenomenon was attributed to the insufficient Al and Ti diffusion force in this zone. Further decreased in temperature (700 °C), resulted to the concentration of Al, Ni and Mg atoms inside the molten pool of weld toe zone and Mg-Al-Ni reaction product was produced (Fig. 12g) [39]. At 487 °C temperature, the diffused copper atoms produced eutectic reaction with magnesium atoms (α-Mg + Mg2Cu) at the weld toe zone. Lastly, with decreased in temperature Mg17(Al,Cu)12 was produced at both laser irradiation and middle zones (Fig. 12h).
2000
1500
1000
500
3.4. Mechanical properties 0
The tensile shear performance of the joints under different coating condition is shown in Fig. 13. Under similar welding parameter, the tensile shear performance of the coated joints was more than the direct joint. For instance, when the nickel coating was thicker than the copper, referring to MT-1 and MT-2, the fracture load reached 1713 N, nearly 74% more than that of the direct joint (MT-0). The formation of Al-NiTi, Ti-Al and Ti-Ni IMCs hindered crack propagation and thus enhanced the interfacial strength, resulting to high fracture load observed. Despite the limited Cu content, the Cu atoms enhanced mutual reaction between Ti and Al as demonstrated in our previous study [37], which could improve metallurgical bonding and increase the joint mechanical property. Nevertheless, the thickness of the IMCs formed was greater than the recommended thickness of 10 µm (Fig. 5f) [40], which could adversely affect the joint strength. On the other hand, MT-3 and MT-4 joints presented the maximum mechanical resistance, which could be attributed to the Ti-Al and Ti-Ni mixed reaction layer formation whose
MT-0
MT-1
MT-2
MT-3
MT-4
MT-5
Sample condition Fig. 13. Tensile-shear fracture loads of laser welded-brazed Mg/Ti joints under different coating conditions.
thickness was less than the critical thickness. The joint mechanical performance attained a maximum value of 2020 N with MT-3 joint, growing approximately 100% more than the direct joint. For MT-5 joint, the thick Cu coating enhanced the Ti-Al reaction at the laser irradiation and middle zones and promoted Ti-Cu IMC formation at the weld toe zone, which improved the joint mechanical resistance. The fracture load reached 1660 N, which is lower than the remaining coated joints. The declined in the joint performance was associated with the more formation of denser Mg-Cu eutectic structure with increasing Cu coating in the fusion zone with network-like morphology being formed 157
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
Fig. 14. Mg/Cu-Ni coated Ti fracture surface under different coating conditions at different fracture modes: (a) fracture location of MT-3 with FZ fracture; (b) Mg side of (a); (c) higher magnification of the square area indicated in (b); (d) fracture location of MT-5 with interfacial failure; (e) laser irradiation zone; (f) higher magnification of square area indicated in (e); (g) middle zone; and (h) higher magnification of square area indicated in (g).
as shown in Fig. 8(d). The denser eutectic structure resulted in the decreased in the tensile-shear fracture load. Furthermore, the huge differences in thermal conductivity and thermal expansion coefficient between the magnesium and titanium caused residual stress formation [41]. Thus, the uneven Ti2Cu layer observed at the weld toe zone (Fig. 8c) resulted to high stress concentration, which could further deteriorate joint performance and caused interfacial failure [37]. With exception of MT-5 joint that fractured along the interface, all the coated joints fractured at fusion zone during the tensile shear testing. Fig. 14 shows the micrograph of the fracture surfaces in the two fracture patterns and the relevant EDS analysis are listed in Table 6. For joints that fractured at the fusion zone, the crack extended along the interface at the weld toe zone and then moved to the fusion zone as
Table 6 EDS analysis of various points indicated in Fig. 14 (at.%). Point
Mg
Al
Ti
Ni
Cu
Possible phases
P1 P2 P3
94.82 94.63 5.96
4.65 4.36 23.23
0.08 0.76 63.84
0.21 0.11 3.63
0.24 0.14 3.34
α-Mg α-Mg Ti3Al
shown in Fig. 14(a), indicating that the joint strength at the interface was more than that of the fusion zone. The fracture surface exhibit a typical dimple feature (Fig. 14c). However, as for the joint with interfacial failure mode, tear ridge was formed from the fracture surface at both direct irradiation and middle zones as shown in Fig. 14(e–g), 158
Optics and Laser Technology 115 (2019) 149–159
S.T. Auwal, et al.
indicating severe deformation occurred during tensile shear testing. A number of particles (P3 in Fig. 14e) mingled with residual Mg (P2 in Fig. 14e). These particles contained 23.23 at.% Al and 63.84 at.% Ti, which were confirmed as Ti3Al [19].
Manuf. Technol. (2018) 1–17. [12] P. Villars, A. Prince, H. Okamoto, Handbook of ternary alloy phase diagrams, ASM Int. (1995). [13] S. Chen, M. Zhang, J. Huang, C. Cui, H. Zhang, X. Zhao, Microstructures and mechanical property of laser butt welding of titanium alloy to stainless steel, Mater. Des. 53 (2014) 504–511. [14] S. Chen, X. Huo, C. Guo, X. Wei, J. Huang, J. Yang, S. Lin, Interfacial characteristics of Ti/Al joint by vaporizing foil actuator welding, J. Mater. Process. Technol. 263 (2019) 73–81. [15] S. Chen, D. Yang, J. Yang, J. Huang, X. Zhao, Nanoscale structures of the interfacial reaction layers between molten aluminium and solid steel based on thermophysical simulations, J. Alloys Compd. 739 (2018) 184–189. [16] A.M. Nasiri, Laser Brazing of Magnesium to Steel Sheet, 2013. [17] M. Aonuma, K. Nakata, Effect of alloying elements on interface microstructure of Mg–Al–Zn magnesium alloys and titanium joint by friction stir welding, Mater. Sci. Eng. B 161 (1) (2009) 46–49. [18] R. Cao, T. Wang, C. Wang, Z. Feng, Q. Lin, J.H. Chen, Cold metal transfer weldingbrazing of pure titanium TA2 to magnesium alloy AZ31B, J. Alloys Compd. 605 (2014) 12–20. [19] C. Tan, X. Song, B. Chen, L. Li, J. Feng, Enhanced interfacial reaction and mechanical properties of laser welded-brazed Mg/Ti joints with Al element from filler, Mater. Lett. 167 (2016) 38–42. [20] C. Tan, B. Chen, S. Meng, K. Zhang, X. Song, L. Zhou, J. Feng, Microstructure and mechanical properties of laser welded-brazed Mg/Ti joints with AZ91 Mg based filler, Mater. Des. 99 (2016) 127–134. [21] C. Zang, J. Liu, C. Tan, K. Zhang, X. Song, B. Chen, L. Li, J. Feng, Laser conduction welding characteristics of dissimilar metals Mg/Ti with Al interlayer, J. Manuf. Process. 32 (2018) 595–605. [22] A.M. Atieh, T.I. Khan, Effect of process parameters on semi-solid TLP bonding of Ti–6Al–4V to Mg–AZ31, J. Mater. Sci. 48 (19) (2013) 6737–6745. [23] A.M. Atieh, T.I. Khan, TLP bonding of Ti-6Al-4V and Mg-AZ31 alloys using pure Ni electro-deposited coats, J. Mater. Process. Technol. 214 (12) (2014) 3158–3168. [24] A.M. Atieh, T.I. Khan, Effect of interlayer thickness on joint formation between Ti6Al-4V and Mg-AZ31 alloys, J. Mater. Eng. Perform. 23 (11) (2014) 4042–4054. [25] A.M. Atieh, T.I. Khan, Application of Ni and Cu nanoparticles in transient liquid phase (TLP) bonding of Ti-6Al-4V and Mg-AZ31 alloys, J. Mater. Sci. 49 (22) (2014) 7648–7658. [26] C. Tan, C. Zang, X. Zhao, H. Xia, Q. Lu, X. Song, B. Chen, G. Wang, Influence of Nicoating thickness on laser lap welding-brazing of Mg/Ti, Opt. Laser Technol. 108 (2018) 378–391. [27] C.W. Tan, Q.S. Lu, B. Chen, X.G. Song, L.Q. Li, J.C. Feng, Y. Wang, Influence of laser power on microstructure and mechanical properties of laser welded-brazed Mg to Ni coated Ti alloys, Opt. Laser Technol. 89 (2017) 156–167. [28] M. Ding, S. Liu, Y. Zheng, Y. Wang, H. Li, W. Xing, X. Yu, P. Dong, TIG–MIG hybrid welding of ferritic stainless steels and magnesium alloys with Cu interlayer of different thickness, Mater. Des. 88 (2015) 375–383. [29] M. Handbook, Vol. 5: Surface Cleaning, Finishing, and Coating, American Soc. for Metals, 1982, 412–416. [30] Y.-J. Tan, K.Y. Lim, Understanding and improving the uniformity of electrodeposition, Surf. Coat. Technol. 167 (2–3) (2003) 255–262. [31] S.T. Auwal, S. Ramesh, F. Yusof, C.W. Tan, Z.Q. Zhang, X.Y. Zhao, S.M. Manladan, Comparative study on characteristics of laser welded-brazed AZ31/Ti-6Al-4V lap joints with and without coatings, Int. J. Adv. Manuf. Technol. (2018). [32] C. Tan, X. Song, S. Meng, B. Chen, L. Li, J. Feng, Laser welding-brazing of Mg to stainless steel: joining characteristics, interfacial microstructure, and mechanical properties, Int. J. Adv. Manuf. Technol. 86 (1–4) (2016) 203–213. [33] C. Xu, G.M. Sheng, H. Wang, K. Feng, X.J. Yuan, Tungsten inert gas welding-brazing of AZ31B magnesium Alloy to TC4 titanium alloy, J. Mater. Sci. Technol. 32 (2) (2016) 167–171. [34] K. Lee, P. Nash, The Al-Ni-Ti system (Aluminum-Nickel-Titanium), J. Phase Equilibria 12 (5) (1991) 551–562. [35] V.T. Witusiewicz, A.A. Bondar, U. Hecht, S. Rex, T.Y. Velikanova, The Al–B–Nb–Ti system: III. Thermodynamic re-evaluation of the constituent binary system Al–Ti, J. Alloys Compd. 465 (1) (2008) 64–77. [36] S. Du, G. Liu, M. Wang, Microstructure and properties of transient liquid-phase diffusion bonded joint of AZ31B/Cu dissimilar metal, Chin. J. Nonferrous (In Chinese) Met. 5 (2013) 1255–1261. [37] Z. Zhang, C. Tan, X. Zhao, B. Chen, X. Song, H. Zhao, Influence of Cu coating thickness on interfacial reactions in laser welding-brazing of Mg to Ti, J. Mater. Process. Technol. 261 (2018) 61–73. [38] X.Y. Wang, D.Q. Sun, Y. Sun, Influence of Cu-interlayer thickness on microstructures and mechanical properties of MIG-welded Mg-steel joints, J. Mater. Eng. Perform. 25 (3) (2016) 910–920. [39] V. Raghavan, Al-Mg-Ni (Aluminum-Magnesium-Nickel), J. Phase Equilibria Diffus. 30 (3) (2009) 274–275. [40] H. Laukant, C. Wallmann, M. Müller, M. Korte, B. Stirn, H.-G. Haldenwanger, U. Glatzel, Fluxless laser beam joining of aluminium with zinc coated steel, Sci. Technol. Weld. Joi. 10 (2) (2005) 219–226. [41] S.T. Auwal, S. Ramesh, F. Yusof, S.M. Manladan, A review on laser beam welding of titanium alloys, Int. J. Adv. Manuf. Technol. 97 (1) (2018) 1071–1098.
4. Conclusions The effect of the electrodeposited Cu and Ni contents on microstructure development and joint fracture load were analyzed. The reaction products were influenced by the interlayer element content. The major conclusions of this study can be summarized as follows: 1. Compared with uncoated joint (MT-0), superior joints appearance and cross-sections were observed with coated joints (MT1-MT5). Thus, the feasibility of this process depends strongly on the preexisting Cu-Ni layer on the Ti sheet that promotes wetting of the AZ92 filler alloy. 2. Depending on the interlayer Cu and Ni contents, different reaction products formed inside the joint region. Nevertheless, at optimum interlayer contents (i.e. joint with comparable Ni and Cu coating thicknesses MT-3), Ti2Ni mingled with Ti3Al IMC layer was produced at the interface and that the thickness of the reaction layer was below the critical thickness of 10 µm. 3. The maximum tensile-shear fracture load attained 2020 N for the joint with comparable Ni and Cu coating thicknesses (MT-3), which was 100% more than that of the direct joint. With exception of MT-5 joint that fractured at the brazed interface, all the coated joints fractured with fusion zone fracture mode. For the joint with interfacial failure mode, tear ridge was observed at the fracture surface, whereas, the fusion zone fracture surfaces displayed a typical dimple feature. Acknowledgement of the funding sources The authors would like to acknowledge the support provided by University of Malaya (Project No. GPF073A-2018), National Natural Science Foundation of China (Project No. 51504074 and 51875129) and Key Research & Development Program in Shandong Province (Grant No. 2017CXGC0811 and 2017GGX30147). References [1] L. Liu, Welding and Joining of Magnesium Alloys, Elsevier, 2010. [2] X. Cao, M. Jahazi, J. Immarigeon, W. Wallace, A review of laser welding techniques for magnesium alloys, J. Mater. Process. Technol. 171 (2) (2006) 188–204. [3] M.K. Kulekci, Magnesium and its alloys applications in automotive industry, Int. J. Adv. Manuf. Technol. 39 (9–10) (2008) 851–865. [4] C. Blawert, N. Hort, K. Kainer, Automotive applications of magnesium and its alloys, Trans. Indian Inst. Met. 57 (4) (2004) 397–408. [5] S. Manladan, F. Yusof, S. Ramesh, M. Fadzil, Z. Luo, S. Ao, A review on resistance spot welding of aluminum alloys, Int. J. Adv. Manuf. Technol. 90 (1–4) (2017) 605–634. [6] L. Li, C. Tan, Y. Chen, W. Guo, C. Mei, CO2 laser welding–brazing characteristics of dissimilar metals AZ31B Mg alloy to Zn coated dual phase steel with Mg based filler, J. Mater. Process. Technol. 213 (3) (2013) 361–375. [7] S.M. Manladan, F. Yusof, S. Ramesh, M. Fadzil, A review on resistance spot welding of magnesium alloys, Int. J. Adv. Manuf. Technol. 86 (5) (2016) 1805–1825. [8] S. Manladan, F. Yusof, S. Ramesh, Y. Zhang, Z. Luo, Z. Ling, Microstructure and mechanical properties of resistance spot welded in welding-brazing mode and resistance element welded magnesium alloy/austenitic stainless steel joints, J. Mater. Process. Technol. 250 (2017) 45–54. [9] E. Akman, A. Demir, T. Canel, T. Sinmazcelik, Laser welding of Ti6Al4V titanium alloys, J. Mater. Process. Technol. 209 (8) (2009) 3705–3713. [10] C. Xu, G. Sheng, Y. Deng, X. Yuan, K. Tang, Microstructure and mechanical properties of tungsten inert gas welded–brazed Mg/Ti lap joints, Sci. Technol. Weld. Joi. 19 (5) (2014) 443–450. [11] Y.M. Baqer, S. Ramesh, F. Yusof, S. Manladan, Challenges and advances in laser welding of dissimilar light alloys: Al/Mg, Al/Ti, and Mg/Ti alloys, Int. J. Adv.
159