Introduction

Owing to their light weight, high strength, excellent conductivity, and outstanding corrosion resistance, titanium and aluminum alloys are in high demand in aerospace, automotive, and smart structures [1]. When welded to titanium grade 2 (G2) tubing, the AA3105-O is a promising candidate for airplane cabin blade cooling fins. Welding these alloys has already been tested by a number of researchers [2,3,4,5,6,7,8,9,10]. The differences in lattice structures and thermophysical properties of Al and Ti alloys have presented a big challenge in assigning a suitable technique for joining them together. From the Ti/Al equilibrium diagram, it is possible to realize that solidified Al does not allow much Ti dissolution, although a Ti-rich solid-solution of up to 12 at.% Al is possible at 327 °C [2]. In conventional fusion welding, negative mixing enthalpies [3] causes an initial dissolution of Ti. After that, different IMCs such as Al3Ti, AlTi, and AlTi3 form.

These days, laser technology is widely used in the welding/brazing of metals. It delivers many benefits over traditional joining methods, such as high precision, high speed, excellent versatility, and low distortion [4]. Chen et al.[5] studied the mechanism for joining Al to Ti utilizing the CO2 laser technique using filler of AlSi12. The brazing region includes α-Ti nano-size particles, Ti7Al5Si12, and TiAl3, and α-Al and ternary formed α-Al, Si, or Mg2Si. They reported that Ti7Al5Si12 formation diminishes Ti dissolution and thereby restricts TiAl3 growth. They [6] also tested the welding of Al to Ti alloys with AlSi12 filler wire using a continuous-wave laser welding/brazing technique. They found that the Ti content had more effect on the diffusion of Si than temperature. They also discovered that when Ti is melted, Si diffuses into the melt and the Ti-filler interface.

Chen et al. [11,12,13] investigated the impact of heat on the microstructure and mechanical properties of the Ti/Al interface during laser welding-brazing with AlSi12 filler. Lower heat inputs are expected to cause Ti dissolution in the filler metal, resulting in thin IMC layers and eventual FZ fracture. Increased heat input causes IMC thickness to increase, which, when combined with porosities, causes the interface to fracture.

Chen et al. [14] observed that the growth of TiAl3 could happen during the interfacial reaction at the solid/liquid Ti/Al laser welding-brazing interface since the formation of Ti7Al5Si12 and segregation of Si atoms hinders the dissolution of Ti alloy. Nevertheless, the process reliability and weld appearance are curtailed by some laser welding-brazing problems such as the spreading and poor wetting operation. In comparison, laser keyhole welding offers excellent versatility and controllability. The high welding speed of laser welding can reduce metallurgical reaction time and consequently suppresses the formation of IMCs.

Tomashchuk et al. [15] observed a transformation interface rich in Al3Ti with a thickness of 90-300 μm as a result of focusing the laser beam on the Ti6Al4V/AA5754 interface. As the laser beam moved toward the AA5754 side, the TiAl precipitates thinned (by 20 μm) at the interface, resulting in maximum output in joint power.

Chen et al. [16] examined the mechanical properties and fracturing activity of the Ti/Al joint and how they were affected by an overlap configuration’s processing parameters in a multimode laser penetration weld. By increasing the laser power and slowing down the welding speed, they were able to strengthen the joint. However, weld spattering and keyhole instability resulted in a subsequent loss of strength. They also discovered that the joint and the Ti-side of the weld at the interface had two fracture modes. When the fracture occurred on the Ti-side of the weld, the joint strength increased to 184 MPa. According to Chelladurai et al. [17], the heat-affected zone (HAZ) joint fracture is caused by the thickness of the IMCs and the microstructure, which was softened by beam energy impact. Rapid cooling of the FZ could result in in-situ reactions and the formation of certain hard-ceramic particles, resulting in increased microscale hardness.

Majumdar et al. [18] reported that during CO2 laser welding of Ti-6Al-4 V to AlMg0.9Si, Al content below 20 at.% the Ti/Al mixing zone and crack-free welding was not achievable at any welding parameters. They discovered that inserting an Nb plate between Ti and Al sheets keeps 11 at.% of the Al in the FZ and results in a crack-free joint.

To eliminate IMC formation in a welded area, Vaidya et al. [19] modified the Ti/Al interface area to reduce the extent of the Al and Ti interfaces and the length of TiAl3. They found that joint modification has doubled the resistance to fatigue crack propagation. However, tensile strength improved slightly. For enhancing the tensile properties, the requirement for new techniques in joining Ti to Al is crucial.

Shehab et al. [20] welded 1 mm thick Ti to 0.5 thick Al strips using a pulsed Nd:YAG laser instrument. The microstructural investigations showed that precipitation of TiAl3 IMCs in the Al FZ near the connection was the reason for most joint failures under shear loading. Joint strength rose to 86% of the base Al strip.

Since all IMCs are brittle and hard, a high-strength/tough Ti/Al joint requires minimum IMC formation [20, 21]. To minimize the formation of IMCs, previous authors [20] used Al-12Si-2.5 Mg filler. By forming a Ti3(Al-Si) eutectic composition, Si reduces Ti dissolution in TiAl3, resulting in hardness relaxation at the Ti/Al interaction zone. The fast dissolution of Si and Mg in the Al re-solidified zone near the Ti/Al interface increases joint strength to 98% of the Al base metal, a 14% improvement over autogenous welding. However, Mg’s existence in the weld microstructure may limit the usage of the joint in a corrosive environment.

In the current study, a tactile seam tracking system using a high-energy laser instrument is implemented for pulsed welding of Ti-G2 to AA3105-O. Utilization of the AlScZr filler with a ring-like spot-weld procedure was the new idea to achieve a high-strength joint. The study continued by looking at the effect of Al-0.2Sc-0.36Zr filler on microstructural improvements as well as the mechanical properties of the new joint obtained.

Materials and Methods

A 1 mm thick Ti-G2 was lap-joined to a 0.5 mm thick AA3105-O using Nd: YAG laser welding utilizing a 0.15 thick AlScZr filler metal. Tables 1 and 2, and 3 list the chemical composition and mechanical and physical properties of base (BMs) and filler metal. The Ti-G2 sheet locates on top of the AA3105-O sheet with a 5 mm wide overlap. Both BM sheets were in dimensions area of 40 × 5 mm2. Before welding, the sheets were cleaned from grease using acetone and then polished with emery papers, followed by further cleaning with 6-10% NaOH alkaline solution for 5 min. Finally, the sheets were rinsed with distilled water and 30% HNO3 + 3% H2SO4 acid solution for 3 min. Ti-G2 sheets were etched with 20% HNO3 + 5% HF acid solution for 5 min, then wiped and rinsed with ethanol and distilled water for 10 min before welding. Figure 1 illustrates a schematic representation of the joint design and welding procedure performed in this research.

Table 1 Chemical composition of base and filler metals used in the study [22]
Full size table
Table 2 Mechanical properties of the base metals [22]
Full size table
Table 3 Physical properties of the base metals [22]
Full size table
Fig. 1
figure 1

Schematic illustration of laser welding Setup geometry of Ti/Al strips

Full size image

Nd: YAG laser equipment (Model IQL-10) welded the samples. The machine’s average power was 400 W, which generated 1-1000 Hz frequency standard square-shape pulses with a period of 2-20 ms and 0-40 J energy. The optical focusing system’s focal length was 75 mm, generating spots of ~250 μm in size. The clamping mechanism was moved under a 0.05 mm laser head by a movable XYZ table, which set precision. During welding, a commercially pure Argon (99.99% purity) was used as a shielding gas to cover the welding area. Gas was blown at a rate of 20 L∙min-1 through a coaxial nozzle.

A Universal Testing Machine type SANTAM model STM-20 with a 5 mm∙min-1 cross-head speed measured the tensile shear results. Three measurements determined each data point. The microstructure and fracture zone of the resulting joints were investigated using an optical microscope and a scanning electron microscope (SEM) model TESCAN MIRA 3 (Czech) supplied with energy dispersive X-ray spectrometry (EDS) to determine the chemical composition of the joint. Microhardness test, using BUEHLER model MMT1, USA, was employed to evaluate the hardness values of the various locations in the joint, including BM, HAZ, and FZ. An applied load of 25 g and a dwell time of 25 s were applied.

Results and Discussion

A high-quality Ti/Al joint was achieved using a seam welding on a ring-like path (that leads to a laser-spot-like shape). Table 4 presented the best welding parameters obtained from the pre-test experiments.

Table 4 Laser welding parameters
Full size table

Weld Appearance

Figure 2 presents the top view of Ti/Al welds with and without the application of filler metal. It can be seen from this figure that the diameter of the weld nugget in Fig. 2a is slightly larger than that of Fig. 2b for the same welding conditions, which indicates that filler metal consumed part of the welding energy during melting. More than two pulses are applied at the same location before the laser power supply is turned off to prevent the creation of weld craters at the final pulse area.

Fig. 2
figure 2

Top view of Ti/Al welds: (a) without filler, (b) with using Al-0.2Sc-0.36Zr filler metal

Full size image

Figure 3 shows a cross-sectional view of the welds which have been sectioned along the indicated lines in Fig. 2. The single-pulse weld boundary is shown in Fig. 2a, while Fig. 2b presents multiple overlapped pulses weld boundaries. Weld penetration is deeper without using filler due to the heat dispersion by filler metal.

Fig. 3
figure 3

 A cross-sectional view of the welds along the indicated lines in Fig. 2 as (a) section "Introduction" and (b) section "Materials and Methods"

Full size image

Weld Microstructure

Figure 4a presents the SEM image of zone A in Fig. 3a. Based on the equilibrium phase diagram of the binary Ti/Al alloy [23] and EDS point scan analysis, the different IMCs formed at the Ti/Al interface were marked in Fig. 4b. The weld zone contains mainly Ti3Al and TiAl at the Ti side, TiAl2, TiAl3 at the Ti/Al interface region, and Al at the FZ at the Ti/Al interface. Cracks are visible near the boundary of the Al FZ, as illustrated in Fig. 4a. The crack could be a result of brittle TiAl3 IMCs formation, where the high thermal gradient between the FZ and solidified alloy can produce high residual stresses. As the brittle TiAl3 cannot withstand this thermal strain, cracks form at the welded samples’ connection region [3].

Fig. 4
figure 4

Microstructures of zone A of Fig. 3a and adjacent regions. A magnified view of Fig. (a) is presented in (b)

Full size image

A view with more magnification of zone A in Fig. 3a is presented in Fig. 5, where the distribution of the elements Al, Ti, Mn and Fe over the specified area are easily recognized with different colors. It can be seen from Fig. 5c and d that the light yellow and light blue colors for respectively Ti and Al distributions at the Ti/Al interface refer to IMC areas. Figure 5d and e refer to the uniform distribution of the elements Mn and Fe respectively.

Fig. 5
figure 5

EPMA mapping analysis of Al, Ti, Mn and Fe elemets of zone A in Fig. 3a

Full size image

Figures 6 and 7 show the equilibrium phase diagram of binary alloys: Al-Sc, Ti-Sc; Al-Zr; Ti-Zr; systems. Based on Figs. 6 and 7, the following points can be inferred:

  1. 1.

    [0 to 0.4 wt% Sc] can dissolve in Al forming Al–Sc solid-solution (α Al), otherwise α Al + Al3Sc are formed above 0.4 Sc wt% as presented in Fig. 6a.

  2. 2.

    [0 to 3 wt% Sc] can dissolve in Ti forming Ti–Sc solid solution (α Ti), Fig. 6b.

  3. 3.

    [0 to 0.14 wt% Zr] from Al-rich side, can dissolve in Al forming Al–Zr solid solution (α Al), otherwise Al3Zr is formed above 0.14 Sc wt%. as shown in Fig. 7a.

  4. 4.

    [0 to 49 wt% Zr] can dissolve in Ti forming Ti–Zr solid solution (α Ti), Fig. 7b.

Fig. 6
figure 6

Binary phase diagrams for (a) Al–Sc alloy, (b) Ti-Sc alloys [23]

Full size image
Fig. 7
figure 7

Binary phase diagrams for (a) Al-Zr alloy, (b) Ti-Zr alloys [23]

Full size image

Based on the analysis results and weight percentages of Al, Ti, Sc, and Zr from the EDS scan of Fig. 8a and b, and c presented in Table 5, the TixAly phases were predicted. One of the modification effects of Sc in Al occurs at hyper eutectic concentrations above 0.4 wt%, where Al3Sc is formed. The addition of 0.15 Zr wt% reducing the Sc concentration to above 0.18 wt% with keeping its modification effects where zirconium addition to binary Al–Sc alloy forms a substitutional solid solution in Al3Sc in which 50 wt% Zr + 10 wt% Ti can be dissolved in Al3(Sc,Zr,Ti) [12].

Fig. 8
figure 8

High magnification SEM images of the zones of Fig. 3b; a: zone B, b: zone C, c: zone D

Full size image

The AlScZr filler is casted with 0.36 wt% Zr, 0.2 wt% Sc in Al base, where the additional amount 0.26 Zr wt% may form Al3Zr in the casted filler according to the Al-Zr phase diagram. During welding, the Zr in the Al3Zr phase is dissolved in Ti/Al molten metal forming Ti-Zr solid solution α Ti, where a high temperature (above 2000 °C) is reached and thus a better connection between Ti and Al may be achieved [24, 25].

Figure 8 shows high magnification weld cross-sections SEM images for the zones of Fig. 3b. It can be seen from these figures that the interface between the mixed molten metal and the melted Al was ‘fluffy’ with some swirls where Al melted but not mixed with melted Ti, and this is due to the high difference in melting temperature between Ti and Al where Ti is solidified while Al is still in the liquid state [11].

Different IMCs [Ti3Al, TiAl, and TiAl3(Sc,Zr,Ti)] were presented in Fig. 8a. It can be seen from this figure that TiAl3(Sc,Zr,Ti) that formed at point G are highly distributed in the Ti/Al weld zone, and this could be related to the high dispersity of Al3(Sc) in the Ti/Al melted volume where the wt% of Sc at point G = (0.17 × 100) / 77 = 0.22 wt% Sc in binary Al-Sc system, which is above the range (0.18 wt.%Sc) for the modification effect of Sc in Al, which occurred in the presence of 50 wt% Zr with respect to Sc (i.e., 0.1wt.% Zr). In the point F, α Al(Sc,Zr,Ti) solid solution + IMP Al3Zr are formed (Table 5), and these phases (Al3Zr) can always exist in the Al FZ in the absence of Ti (Fig. 6). Point A in Fig. 8b has the same phase Ti Al3(Sc, Zr, Ti) of point G Fig. 8a where wt% of Sc = (0.13 × 100) / 66 = 0.2wt.% Sc in binary Al-Sc system at which the modification effect of Sc in Al occurred in the presence of 50 wt% Zr with respect to Sc (i.e., 0.1 wt%). Figure 8c shows how particles of TiAl3(Sc,Zr,Ti) phase (at points C and D, Table 5) are distributed in Al FZ close to Ti/Al interface, which are the zones of weakness, as a result of a high gradient in mechanical and thermophysical properties, as observed in the previous work [19, 20].

Table 5 Weight% of Al, Ti Sc and Zr obtained from EDS of surfaces shown in Fig. 8a, b and c
Full size table

Mechanical Behavior

Shear Strength of the Joint

Table 6 represents shear force at the yielding point, shear strength, and length extension of the samples with and without using filler metal. Figure 9 shows shear force as a function of extension curves with images of fused areas at the bottom of the Ti sheet at the Ti/Al contacted surfaces. Despite that the area of fusion without using filler (3.77 mm2) is larger than that area of fusion with using filler (3.6 mm2), Fig. 9; Table 6 show that the weld with AlScZr filler metal has a higher fracture shear force (300 N) and a higher extension value than the autogenous weld. This result could be related to the modification consequence on the composition of the weld metal by Sc and Zr, which leads to the highest atomic percent strength values when added to Al and to decrease the hardness gradient between the aluminum FZ and Ti/Al zone as illustrated in Fig. 12, which decreases the opportunity of initiation crack under thermal stress [19].

Table 6 Effect of AlScZr filler metal on mechanical properties of the joint
Full size table
Fig. 9
figure 9

Force-extension curves merged with images of fused areas of the joints at bottom of Ti sheet

Full size image

The strengthening is achieved by solid solution and precipitation hardening of Al3(Sc, Zr) particles, which impedes the dislocation cell structure [26, 27]. The modification effect of Sc and Zr for Al was the primary motivation to choose such elements since the cracks in autogenous weld joint located in Al near Ti/Al interface and thus fractured, as observed in the previous works [16, 20, 21, 26].

The effect of filler metal existence on fracture behavior is shown in Fig. 10. When autogenous weld was performed, the fracture occurred at the FZ, as presented in Fig. 10a. Further fracture details were detected by the SEM examination, as shown in Fig. 10c. It was observed that the fractured area occurred between the Al weld zone and the Ti/Al interaction area. The fracture zone featured a cleavage fracture characterized by a river pattern band with some dimples. The cleavage fracture is contributed to the formation of Ti/Al IMCs (like TiAl3 as detected in EDS analysis of point A) at the Ti/Al interaction zone. It is believed that the presence of a river-like pattern is an indication of a crack that moves along the grains through crystallographic planes that form a series of plateaus and joining ledge. The presence of small dimples is an indication of the ductile fracture of Ti in some areas. While the sample’s rupture using filler metal occurred at Al HAZ-base metal (Fig. 10b) with ductile fracture behavior characterized in SEM image by large dimples as shown in Fig. 10d.

Fig. 10
figure 10

Images of the samples fractured under shear load (a,c) without filler, (b,d) with AlScZr filler

Full size image

The strength of the lap joint design is mainly affected by the area of the FZ and the strength of the welded sheets [27, 28]. Considering the shear forces of 264 and 300 N for the joints (without and with filler respectively) and also the fusion surface areas (3.77 mm2 and 3.6 mm2) in Fig. 9, the values of shear strength of 70 MPa and 83.3 MPa were calculated for the joint without and with AlScZr filler, respectively. Figure 11 compares the shear strength values for Ti-Al dissimilar weld joint with and without filler metal and Al base metal. In dissimilar metals Ti/Al joint, it is more logical to compare the value of the shear stress of the joints (with and without AlScZr filler), with the shear stress of the aluminium 82 MPa, (i.e., lower strength base-metal). Due to the addition of AlScZr filler metal, joint shear strength of up to 102% of the Al base metal shear strength was achieved, which was around 16% higher than that obtained by autogenous welding.

Fig. 11
figure 11

Shear strength values of Ti-Al dissimilar weld joint with and without filler metal and Al base metal

Full size image

Microhardness of the Joint

Figure 12 shows Vickers hardness measurements of the Ti/Al welding joint with and without filler metal, covering BM, HAZ, and FZ. Ti and Al BMs had an average hardness value of 120 HV and 43 HV, respectively. There is no discernible difference in HAZ hardness between the two metals, which may be attributable to the LBW process’s low heat input [29], which limits the change in HAZ mechanical properties. At the Ti side, the microhardness increases at the FZ over the BM with an average hardness of 317 HV. This rise in hardness might be due to the existence of acicular alpha-martensite (Fig. 13), produced by thermal-cycling rapid laser welding [13]. A large increase in hardness was observed in the interaction zone, with a greater value than 650 HV. This increase in hardness is an impact of the Ti/Al IMCs developed in this region. While in the FZ at Al side, the hardness was approximately the same degree as the BM.

Fig. 12
figure 12

Vickers hardness measurements of the welding joint

Full size image
Fig. 13
figure 13

Formation of acicular martensite at Ti FZ

Full size image

When filler metal was added, a significant improvement in the microhardness values at the Al FZ was observed compared with Al BM, and this could be assigned to the high cooling rate that modifies the grains size of Al, besides, the strengthening effect of Sc and Zr of filler metal. At the Ti/Al interaction zone, the hardness rising was steady, and the maximum hardness recorded in the interaction zone was lower than that without utilizing the filler metal. The observed fluctuations in the microhardness values in the interaction zone could be attributed to the local composition at each point (IMC formation at Ti/Al interface + strengthening filler effect (Sc and Zr that dissolved in Ti).

Conclusions

In this paper, a pulsed Nd: YAG laser welding was employed for dissimilar joining of Ti-G2 and AA3105-O using AlScZr filler metal. The following conclusions can be drawn from this paper:

  1. 1.

    Most autogenous welding joint failures were caused by the formation of intermetallic compounds (IMCs) such as TiAl3 and TiAl2 at the Al re-solidified zone.

  2. 2.

    With using of AlScZr filler metal, the addition of zirconium to binary Al–Sc alloy formed a substitutional solid solution in Al3Sc in which Zr, Ti dissolved in Al3(Sc,Zr,Ti).

  3. 3.

    Sc has a strengthening effect by solid solution and precipitation hardening by forming Al3(Sc,Zr) particles, which pin the Ti/Al joint’s dislocation cell structure

  4. 4.

    The relaxing effect of Ti(Al,Sc)3 and the strengthening effect of AlScZr filler metal on the Al FZ and Al re-solidified zone (near Ti/Al interface) increase the joint strength from 86% for autogenously laser weld to 102% with filler metal.

  5. 5.

    The high cooling rate of laser welding that decreases Al grain size, and the strengthening effect of Sc and Zr in the filler metal, resulted in a substantial increase in Al FZ hardness.

  6. 6.

    The hardness increased steadily at the Ti/Al interaction zone, and the maximum hardness reported was lower than without the filler metal.