1 Introduction

Molybdenum-based alloys are extensively used in metallurgy, aerospace, nuclear power, and several other industries due to their exceptional stability at high temperatures, high melting points, and excellent thermal conductivity (Ref 1, 2). Moreover, the TZM alloys, enhanced with Zr and Ti elements, are gaining popularity because of their exceptional mechanical qualities and high thermal stability (Ref 3). The Ti-6Al-4V alloys are widely used in aerospace owing to their exceptional strength, excellent corrosion resistance, and low density, making them favored lightweight materials (Ref 4). Implementing the composite structure between the two metals is particularly important in fabricating missile combustion chambers and low-temperature packaging (Ref 5, 6). Hence, evaluating the welding between dissimilar metals of TZM and TC4 and attaining a connection with high strength has significant practical value and promising potential.

Various welding techniques, including brazing (Ref 7), rotary friction welding (Ref 8), laser-GTAW hybrid welding (Ref 9), and laser beam welding (Ref 10), have been used for welding pure Mo and Mo-based alloys. Moreover, heterogeneous welding of TZM has been documented in recent years. Metallurgical regulation and energy control are typically utilized to enhance the mechanical properties of joints in dissimilar metals (Ref 11). Yu Bin et al. found that Fe2Mo intermetallic compounds were produced at the TZM/FZ interface of the joint during the electron beam welding of TZM and 30CrMnSiA, which led to poor joint performance, and the tensile strength is only 162.5 MPa. After adding the V layer, the intermetallic compounds disappeared, and the tensile strength increased to 312.7 MPa (Ref 12). Yin Qianxing et al. stated that the composition of the brittle middle layer changed from continuous σ (FeMo) to intermittent fine σ (FeCr) with the addition of a high-entropy alloy interlayer during electron beam welding of Mo/Kovar, and the tensile strength of the joint increased from 190 to 358 MPa (Ref 13). Gao Xiaolong et al. reported that the Ni interlayer with the laser welding of pure molybdenum and stainless steel compensated for the welding thermal stress and reduced the brittle phases in the welded joints, while the tensile strength was 351 MPa with 0.1 mm interlayer (Ref 14). Zhou Xingwen et al. used the TC4/Ni interlayer to realize the laser brazing of molybdenum and Ta. The research showed that when using TC4 as the interlayer, the solid solution was generated at the interface, while Ni would produce brittle phases, and the tensile strength of joints using TC4 was 19% higher than that of joints using Ni (Ref 15). Mo D et al. studied the effect of the silver interlayer on the electron beam welding of Kovar and titanium alloy and revealed that the silver could suppress the formation of brittle phases when the thickness was 0.6 mm and the maximum tensile strength was 243 MPa (Ref 16). Hongbo Xia (Ref 17) et al. adopted the laser texturing and micro-arc oxidation hybrid process to enhance the tensile-shear strength of CFRTP/TC4 joint. The resulted showed that porous micro-arc oxidation coating enhanced the wetting of molten peek from CFRTP and improved the resistance to the tensile–shear load, and the maximum tensile–shear force was 1587N with the micro-arc oxidation coating thickness of 2.3 μm.

Moreover, many researchers have reported the control of energy in the dissimilar welding of Mo-based alloys. Wang Ting et al. found that when the electron beam current exceeded 24 mA, the tensile strength of the joint of TZM and 30CrMnSiA steel was deficient, while the highest tensile strength was 191 MPa with the beam current of 24 mA, and the fracture occurred in the Fe2Mo intermediate layer on the TZM side (Ref 18). Chen G et al. proved that the offset of the electron beam would increase the tensile strength of Mo and Kovar joints from 190 without deflection to 262 MPa due to a decreased brittle reaction layer at the fusion line, leading to better joint performance (Ref 19). The impact of beam offset on laser welding of pure Mo and stainless steel was investigated by Gao Xiaolong et al., resulting in a decrease of Fe-Mo brittle phase intermediate layer with offset increase, while the highest tensile strength of joints was 290 MPa with 0.3 mm offset (Ref 20). Zhang et al. studied the laser welding of molybdenum and titanium alloy with different offsets. The effective joint could form when the beam focused on the titanium alloy, while the maximum tensile strength of 350 MPa was obtained at 0.3 mm offset (Ref 21). Despite previous research in which the two materials were welded successfully, the base materials were preheated to 450 °C and maintained for several seconds before welding, which may be limited in actual engineering applications. Therefore, further research is needed to improve the mechanical properties of joints and facilitate practical applications.

Based on the Mo-Ti phase diagram, the Mo element exhibits unlimited solubility in β-Ti and limited solubility in α-Ti, leading to a continuous solid solution (Ref 22). While welding molybdenum and titanium do not pose metallurgical challenges, the significant disparity in melting points and physical properties between the two metals suggests that inappropriate energy input and welding techniques may result in weak joint performance and failure to meet the requirements of joint applications. Therefore, electron beam welding, which takes place in a vacuum environment and allows for precise energy control, is well-suited for welding refractory metals (Ref 8).

In this study, the TZM alloy and Ti-6Al-4V alloy were welded by the electron beam with different beam offsets. The influence of the electron beam offset on the morphology, microstructure, element distribution, and mechanical properties of the welded joint was evaluated.

2 Experimental

2.1 Materials and Preparation

TZM molybdenum alloy and Ti-6Al-4V (TC4) titanium alloy were utilized in the experiment and were formed into 60 × 50 × 2 mm plates. The composition and physical characteristics of the base materials are presented in Tables 1 and 2, respectively. Prior to welding, the splicing surfaces underwent a thorough preparation process that involved degreasing with acetone, cleaning with alcohol, and polishing with sandpaper. Figure 1 illustrates the microstructures of the two base materials. A schematic diagram of the welding process can be found in Fig. 2. It is noteworthy that in our experiment, all offset beams were deflected toward the TC4 side, with offsets ranging from 0 to 0.4 mm in increments of 0.1 mm.

Table 1 The chemical composition (wt.%) of the base materials
Full size table
Table 2 The physical properties of the base materials
Full size table
Fig. 1
figure 1

Microstructures of base materials, (a) TZM, (b) Ti-6Al-4V

Full size image
Fig. 2
figure 2

Welding schematic diagram

Full size image

2.2 Analytical Methods

The electron beam welding machine (PTR EVOBUBE SK230L.) was used during the welding experiment. The welding parameters were acceleration voltage 55 KV, welding beam current 14 mA, and welding speed 5 mm/s. The TZM side was etched by a mixture of acids (nitric acid/hydrofluoric acid/water = 5:2:3, by volume) while the TC4 side was etched by another mixture of acid (nitric acid/hydrofluoric acid/water = 1:1:4, by volume). The cross sections and microstructures of the welded joints were observed via optical microscopy, and more detailed analysis was conducted by scanning electron microscope (Zeiss@: SIGMA-300). In addition, the elemental distribution was analyzed by SEM-EDS in spot mode and line scan mode. The phases of the welded joints were confirmed and analyzed by XRD (Rigaku D/max-2200PC). The scanning range was 20°-90°(2θ) at the scanning speed of 1°/min. The microhardness of the welded joints section was tested by the HXD-100TM microhardness tester, and the test conditions involved 100 g continuous compression for 15 s. The mechanical properties of the welded joints were evaluated using the INSTRON 5982 testing machine. Three samples were prepared for each beam offset condition following GB/T 228-2002. The fracture morphology and element distribution were analyzed via SEM after stretching.

3 Results and Discussion

3.1 Weld Appearance and Macrostructure of Joints

The cross sections of the TZM/TC4 welded joints with different beam offsets are presented in Fig. 3. All samples were successfully joined, and the fusion zone (FZ) exhibited a funnel-shaped profile. When the offsets were set to 0 or 0.1 mm, pores were observed near the fusion line due to gas escape during welding, as illustrated in Fig. 3(a) and (b). However, as the offset distance increased, the pores disappeared, indicating that the joints exhibited improved performance through offset shifting. There were pre-existing voids, contamination and inclusions in the TZM base metal associated to the powder metallurgy, which resulted in the formation of pores with more melted TZM base metal. As the beam offset increased, the pores disappeared (Ref 21).

Fig. 3
figure 3

Cross section morphology of the joints with different beam offsets and the area of FZ and HAZ of TC4, (a) 0 mm, (b) 0.1 mm, (c) 0.2 mm, (d) 0.3 mm, (e) 0.4 mm, (f) area of FZ and HAZ of TC4

Full size image

Figure 3(a) clearly demonstrates the accumulation of molten metal on the top side of the TZM base material with the 0 mm offset, due to the recoil force and the high viscosity of the molten TZM (Ref 19). The fusion line on the TZM side had a curved shape in the upper portion, while the lower portion was perpendicular. With a beam offset of 0 mm, significant heat input caused the top TZM to reach melting point and begin to melt. As the heat input decreased in the thickness direction, it was solely conducted to the TZM through the molten pool. Due to the exceptionally high thermal conductivity of TZM, heat was rapidly transferred, resulting in the formation of the fusion line shape. M.J. Torkamany et al. also observed a similar effect in their investigation (Ref 23). As shown in Fig. 3(b), (c), (d) and (e), the curvature of the upper section of the fusion line gradually decreased as the offset increased, eventually becoming entirely straight for offsets of 0.3 and 0.4 mm. With the increasing of beam offset, more heat was input to the TC4 while the melting amount of the TZM decreased, causing the fusion line to be vertical.

The top surfaces of all the welded specimens exhibited depressions measuring 0.3 mm. The presence of weld reinforcement was apparent in the lower portion of the base materials, which is attributed to the excellent wettability of liquid TC4 on TZM. The reinforcement reached the peak at a distance of 0.54 mm with the 0.1 mm offset and gradually decreased as the offset raised. Figure 1(f) illustrates the sizes of the FZ and heat-affected zone (HAZ) of TC4 with varying beam offsets. The results indicate that as the beam offset increased, the dimensions of both the FZ and HAZ in TC4 gradually decreased. Notably, the largest FZ area was observed with a 0.2 mm offset.

3.2 Microstructure and Element Distribution of the Joints

3.2.1 Microstructure of Joints

Figure 4(a) displays the microstructure of the joint with the 0 mm beam offset, while Fig. 4(b), (c) and (d) provides magnified views of the TZM/FZ, FZ, and FZ/TC4 areas, respectively. Figure 4(b) exhibits the consistent distribution of dendritic structures perpendicular to the TZM/FZ fusion line. Conversely, they were randomly distributed within the center of the FZ, as seen in Fig. 4(c). Due to the high viscosity of liquid TZM, the melted material accumulated at the top of the molten pool. The Marangoni effect promoted the formation of a solid solution by mixing liquid TC4 with liquid TZM, and the solid solution subsequently cooled within the molten pool. As a result, the dendrites inside the liquefied pool competed for the most advantageous arrangement of the solid solution, resulting in a chaotic and fragmented appearance. Dendritic structures also accumulated on the periphery of the FZ/TC4 fusion line, proliferating in alignment with the dominant temperature gradient. Additionally, the fusion line was wide and irregular, as shown in Fig. 4(d).

Fig. 4
figure 4

Microstructure of joint with 0 mm offset, (a) cross section morphology, (b) TZM/FZ, (c) FZ, (d) FZ/HAZ of TC4

Full size image

Figure 5 reveals the microstructure morphology of the fusion line in TZM/FZ welded joints with varying offsets. Figure 5(a) illustrates grain boundaries and a limited number of dendrites in the vicinity of the fusion line of the joint with 0.1 mm offset. Figure 5(b) demonstrates that the overall structure of the joint stayed consistent when comparing the 0.2 mm offset to the 0.1 mm offset. As the beam offset continues to increase, the dendritic structure decreases, with only a few discernible grain boundaries observed. Distinct grain boundaries arise near the fusion line, and acicular martensite was dispersed along these grain boundaries in joints with 0.3 and 0.4 mm offset, as shown in Fig. 5(c) and (d).

Fig. 5
figure 5

Microstructure of fusion line on TZM side with different beam offset, (a) 0.1 mm, (b) 0.2 mm, (c) 0.3 mm, (d) 0.4 mm

Full size image

Figure 6 shows the microstructures of the FZ in joints with different offsets. Specifically, Figure 6(a) represents the FZ morphology with the 0.1 mm offset. In comparison to the 0 mm offset, the FZ displayed a consistent arrangement of dendritic structures that extend to the center of the FZ, indicating that the increase in beam offset results in a reduction of TZM base material melting, preventing the formation of solid solution in the FZ and potentially impacting the development of microstructure. The dendrites in the center of the FZ exhibited reduced length due to inadequate supercooling, limiting dendrite expansion. Figure 6(b) shows the FZ with the 0.2 mm displacement. In contrast to Fig. 6(a), the FZ exhibited some grain boundaries and intricate dendritic patterns. As the offset continues to increase, the dendritic structure in the FZ disappeared and underwent a transformation into martensite structure, as seen in Fig. 6(c) and (d).

Fig. 6
figure 6

Microstructure of fusion zone with different beam offset, (a) 0.1 mm, (b) 0.2 mm, (c) 0.3 mm, (d) 0.4 mm

Full size image

The microstructures of the fusion line on the TC4 side were also examined which are shown in Fig. 7. It indicated that dendrites had a tendency for growth in a direction perpendicular to the fusion line, aligning with the normal direction when the offsets were 0.1 and 0.2 mm. This was consistent with the phenomenon observed at 0 mm offset in Fig. 4(d). However, as the offset increased, the dendrites gradually diminished, ultimately leading to a complete transformation into martensitic structure at 0.4 mm offset, as seen in Fig. 7(d). Compared to the TC4 side fusion line with 0 mm offset, the increase in offset distance resulted in the flattening of the fusion line on the TC4 side.

Fig. 7
figure 7

Microstructure of fusion line on TC4 side with different beam offsets, (a) 0.1 mm, (b) 0.2 mm, (c) 0.3 mm, (d) 0.4 mm

Full size image

The difference in beam offsets caused changes in the temperature and heat conduction of the molten pool during TZM/TC4 welding, further affecting the morphology of the weld joints. At the beam offset of 0 mm, a significant amount of the TZM base material melted, forming a junction containing a solid solution of (Mo,Ti). The adjustment of the offset to 0.1 mm resulted in a reduction of the melting quantity of TZM, which was observed through the examination of the bending angle of the fusion line on one side of TZM. An abundance of dendrites has accumulated in the joint. As the offset intensity raised, the electron beam became entirely concentrated on the TC4 side, which surpassed the threshold for the transition of Alpha Ti into Beta Ti (882.5 °C) (Ref 24). The beta Ti experienced martensitic transformation as a result of the rapid cooling rate of the molten pool (Ref 25).

3.2.2 Element Distribution of Joints

The distributions of molybdenum (Mo) and titanium (Ti) elements in welded joints with varying beam offsets are shown in Fig. 8. The Mo content was higher in TZM but significantly lower in the FZ. Conversely, the Ti concentration in TZM decreased significantly. At offsets of 0, 0.1, and 0.2 mm, it was observed that the Mo content at the fusion line on the TC4 side gradually decreased over a distance corresponding to the width of the fusion line, as illustrated in Fig. 8(a), (b) and (c). However, there were no significant changes in the distribution of Mo elements in TC4 and the FZ with 0.3 and 0.4 mm offset due to the decreasing melting of TZM caused by the increasing offset.

Fig. 8
figure 8

Line scanning results of element distribution of TZM/TC4 joints with different beam offset, (a) 0 mm, (b) 0.1 mm, (c) 0.2 mm, (d) 0.3 mm, (e)0.4 mm

Full size image

Scanning electron microscopy was used to conduct an exhaustive investigation of the morphology of the welded joints and analyze the distribution of elements. Figure 9(a) depicts the morphology with the 0 mm offset, whereas Fig. 9(b), (c) and (d) provides high-magnification images of the TZM/FZ, FZ, and FZ/TC4, respectively. Figure 9(a) clearly demonstrates the presence of a uniformly thick interlayer composed of a solid solution (Mo, Ti) (Ref 23, 26). A similar intermediate layer also appeared near the fusion line of the two metals in the study by Hongbo Xia et al (Ref 27). Several fragmented dendritic formations were dispersed along the interlayer. In Fig. 9(d), the fusion line on the TC4 side had an uneven morphology that progressively transitioned from regular blocky structures to a stable state, while the width of the fusion line was 0.08 mm. The EDS analysis was performed on the selected points, as listed in Table 3. The concentration of Mo element was relatively high near the fusion line on the TZM side, amounting to 29.46 at.% while decreasing to 25.87 at.% in the FZ and dropping further to 20.28 at.% near the fusion line of the TC4 side. The concentrations of the Mo element of SPOT E and SPOT F were 9.37 and 2.62 at.%, respectively, indicating a gradual decrease of the Mo element along the fusion line.

Fig. 9
figure 9

Higher magnification of microstructure of joint with 0 mm offset, (a) overall morphology, (b) TZM/FZ, (c) FZ, (d) FZ/HAZ of TC4

Full size image
Table 3 Chemical composition (at.%) of point A-H in joint with 0 mm offset
Full size table

The morphology of the TZM/FZ, with offsets ranging from 0.1 to 0.4 mm, is shown in Fig. 10. Compared with that in Fig. 9(b), the TZM/FZ morphology with 0.1 mm was relatively smooth. When the offset was 0.2 mm, some rounded phases were gathered within the grain boundaries. As the offset increased to 0.3 mm, the short and disordered martensite structure with more obvious grain boundaries existed. Applying the 0.4 mm offset, the structure near the fusion line completely transformed into an ordered martensite structure, as illustrated in Fig. 10(c) and (d). The EDS results indicated that the Mo element content dropped sharply from 80 at to 5% at near the fusion line, as listed in Table 4. Simultaneously, as the beam offset increased, the concentration of the Mo element in the vicinity of TZM/FZ diminished, reducing from 5, 4, and 2 at to 1.55 at.%.

Fig. 10
figure 10

Higher magnification of microstructure of fusion line of TZM side with different offset, (a) 0.1 mm, (b) 0.2 mm, (c) 0.3 mm, (d) 0.4 mm

Full size image
Table 4 Chemical composition (at.%) of point A-H in TZM/FZ area
Full size table

Figure 11 shows the microstructures and element distributions in the FZ with different offsets. Continuous dendritic structures were distributed in the FZ with the offset of 0.1 mm. Figure 11(b) demonstrates the accumulation of round phases and disordered martensite structures in the FZ, with an offset of 0.2 mm. At the offset of 0.3 mm, the circular phase was noticeably diminished and concentrated within the lattice gap, while the martensite structure was dispersed throughout the crystal lattice. The FZ consists of martensite with increased offset, as shown in Fig. 11(d). The contents of elements in the selected spots are listed in Table 5. The results demonstrated a gradual decrease in Mo element content within the FZ with increasing offset. More precisely, the Mo content decreased from 6.49 to 1.23 at.%.

Fig. 11
figure 11

Higher magnification of microstructure of fusion zone with different offset, (a) 0.1 mm, (b) 0.2 mm, (c) 0.3 mm, (d) 0.4 mm

Full size image
Table 5 Chemical composition (at.%) of point A-H in FZ
Full size table

The FZ and heat-affected zone (HAZ) of TC4 with 0.1 and 0.2 mm beam offsets are also analyzed in Fig. 12. At 0.1 mm offset, the fusion line width decreased, and the distribution of the fusion line appeared relatively smooth, resulting in a fusion line width of 0.02 mm, as shown in Fig. 12(a). However, with 0.2 mm offset, the distinct division of the fusion line was no longer evident. It can be observed from Table 6 that the Mo element content in Spot B of the fusion line was 2.10 at.%, while that of Spot C in the HAZ of TC4 was 1.02 at.%, indicating a gradual decrease across the fusion line with the 0.1 mm offset. In contrast, the Mo element contents in Spot E and Spot F were 1.95 and 1.35 at.%, respectively. The decreasing trend of the Mo element slowed down near the fusion line, indicating that as the offset increased, the content in the FZ and HAZ of TC4 tended to be consistent.

Fig. 12
figure 12

Higher magnification of microstructure of fusion line of TC4 side with different offset, (a) 0.1 mm, (b) 0.2 mm

Full size image
Table 6 Chemical composition (at.%) of point A-H in FZ/HAZ of TC4
Full size table

To further clarify the phases in joints, the XRD analysis was carried out which is shown in Fig. 13. When the offset was 0 mm, the Mo content in the FZ was 25.87 at.%, and the FZ was mainly composed of (Mo,Ti) solid solution and α’ phase according to the Mo-Ti phase diagram and XRD result. At the offset of 0.1 mm, the content of Mo element was 6.49 at.%. The formation of a significant quantity of continuous solid solution was impeded by the decreased melting of the TZM. The XRD result showed the main peak of BCC molybdenum, along with the identification of β-Ti and ω-Ti phases. Due to the presence of the Mo element, β-Ti remained stable at ambient temperature. The stability of the β phase affected the generation of the ω phase in β-Ti alloys. The amount of ω phase produced was directly proportional to the stability of the β phase. Zhang et al. reported that when the Mo content was between 4.5 and 8 at.%, β-Ti and ω-Ti were the main components of the Ti-Mo alloy (Ref 28). When the offset increased to 0.2 mm, the Mo content was 4.27 at.%, and the FZ was mainly composed of BCC molybdenum and α’ and α’’ phases. When the offset further increased to 0.3 and 0.4 mm, the Mo contents in the FZ were 2.21 and 1.23 at.%, respectively. There were only single-phase Mo and α’ phases in the FZ, consistent with the research results of W.F. Ho et al. (Ref 29). Therefore, the position of the offset influenced the Mo content in the molten pool, and the varying cooling rates due to different offsets led to distinct phases in the FZ, which subsequently affected the mechanical properties of the joint.

Fig. 13
figure 13

X-ray diffraction patterns of joints with different beam offset

Full size image

3.3 Mechanical Properties of the Welded Joints

3.3.1 Microhardness Analysis of the Welded Joints

Figure 14 displays the joint microhardness with different offsets. It was observed that the microhardness of the FZ in all joints was higher than that of the TZM base material. Furthermore, the microhardness of the HAZ of the TZM was slightly lower than that of the base material, decreasing from 220 HV to 210 HV. This was related to the recrystallization of the TZM. For the rolled TZM alloy, Ping Hu et al. observed recrystallization characteristics near 1300 °C (Ref 30). Due to high heat input of electron beam, recrystallization still occurred in the TZM heat-affected zone, resulting in the decrease in microhardness (Ref 31, 32). Near the fusion line on the TZM side, the joints with 0.3 and 0.4 mm offset exhibited the highest microhardness at 430HV, while the microhardness with other offsets was 350HV. The microhardness of the joints within the FZ exhibited considerable variability. When the offset was 0.3 and 0.4 mm, the microhardness of the FZ was higher than that of the other three joints. The maximum values for the FZ are 425 HV and 418 HV, respectively. Meanwhile, the microhardness of the FZ of 0 and 0.2 mm offset was comparable, measuring about 340 HV, while the FZ of 0.1 mm offset had the lowest microhardness, approximately 310 HV.

Fig. 14
figure 14

The microhardness of joints with different beam offset

Full size image

As the content of stabilizing elements in titanium alloys increased, the average hardness and elastic modulus of the metal would decrease significantly, which was mainly related to the volume fraction of α' and α'' phases (Ref 33). The orthorhombic martensite structure improved the hardness of the joint. According to research by Moiseev et al., the α' and α'' phases had similar mechanical properties, but the α' phase is harder than the α'' phase, while the excessive cooling rate during welding would increase the hardness (Ref 34,35,36). Combining the microstructure of the FZ and XRD results, it can be found that when the offset is 0.3 or 0.4 mm, the FZ was dominated by the α’ phase, resulting in the highest microhardness. Meanwhile, the microhardness of FZ decreased with the offset of 0.2 mm related to the existence α’/α’’ phase in the FZ. When the beam offset was 0 mm, the FZ consisted of the Mo-Ti solid solution, resulting in low microhardness. Additionally, the microhardness of β-Ti was lower than that of α’ and α’’ phases, making the joint with the offset of 0.1 mm have the lowest microhardness (Ref 37). The microhardness of the FZ on the side near the fusion line on the TC4 side increased significantly at the offset of 0, 0.1, and 0.2 mm, which was caused by the rapid cooling of the molten pool, as shown in Fig. 4(d) and 7(a) and (b).

3.3.2 Tensile Strength and Fracture of Welded Joints

Figure 15 presents the tensile strength of welded joints with varying offsets. The tensile strength of the joints initially increased and then subsequently decreased within the beam offset range of 0-0.4 mm. When the offset was 0 mm, the tensile strength and the elongation were 262 MPa and 0.68%, indicating the joint was absolutely brittle. As the offset increased to 0.1 mm, the tensile strength and the elongation increased to 432 MPa and 2.28%. The tensile strength reached a maximum of 482 MPa, which is 56% of the tensile strength of the TZM base material when the offset is increased to 0.2 mm, the fracture strain and elongation were 0.50 and 1.88%, respectively. However, as the offset increased further, the tensile strength progressively diminished to 150 MPa at 0.4 mm. Furthermore, the fracture surface of the TC4 side after fracture was analyzed.

Fig. 15
figure 15

Tensile strength of joints with different beam offsets

Full size image

Figure 16 illustrates the fracture micromorphology of the TC4 side with different beam offsets. It can be observed from Fig. 16(a) and (b) that the fracture surface was undulating, and most of the fracture surfaces exhibited river-like patterns, which indicated that the fracture mode was brittle fracture when the offset was 0 mm. The EDS analysis of the fracture surface listed in Table 7 revealed that the Ti content was 56 at.%, concluding that the fracture took place in the FZ of the welded joints. Although the TZM base metal recrystallized with heated, the strength of the Mo-Ti solid solution with higher Mo content is lower than that of recrystallized TZM. Nthabiseng Moshokoa et al. have shown that as the Mo content increased, the tensile strength decreased significantly (Ref 38). Initially, brittle fracture was extremely prone to occur in Ti-20 at.% Mo solid solution due to its BCC structure.

Fig. 16
figure 16

Fracture morphology of the joints with different beam offsets, (a) 0 mm, (b) details of fracture with 0 mm offset, (c) 0.1 mm, (d) details of fracture with 0.1 mm offset, (e) 0.2 mm, (f) details of fracture with 0.2 mm offset, (g) 0.3 mm, (h) details of fracture with 0.3 mm offset, (i) 0.4 mm, (j) details of fracture with 0.4 mm offset

Full size image
Table 7 Chemical composition (at.%) of point A-J in fracture
Full size table

Figure 16(c) and (d) displays the fracture micromorphology of the joint with a 0.1 mm offset. Compared with that in Fig. 16(a), the fracture in Fig. 16(a) have undergone a significant change, with numerous flat reflective surfaces on the entire fracture surface, indicating intergranular fracture characteristics. The EDS results indicated 3 at.% Ti content on the fracture surface, determining that the fracture occurred in the HAZ of the TZM. The fracture morphologies of 0.2 and 0.3 mm offset were similar to that of 0.1 mm and indicated intergranular fracture status, as displayed in Fig. 16(e), (f), (g) and (h). However, there are notable variations in that the 0.2 mm fracture joint exhibited more prominent undulations, and visible transgranular cracks were observed on the fracture. The EDS results suggested that the Ti content on the fracture surface is at 3 at.%, and the fracture still transpired in the HAZ of TZM.

Although the microstructure and element distribution of the FZ changed a lot with different beam offsets, when the beam offset range was 0.1-0.3 mm, the fractures of the joints all occurred in the HAZ of TZM. Although improving upon the mechanical properties of pure molybdenum, TZM retained the traits of grain boundary segregation and brittle fracture. At the same time, the input of electron beam heat coarsens the HAZ grains on the TZM side, further promoting grain boundary segregation, reducing the grain boundary area, and greatly reducing the bonding strength of the HAZ grain boundary on the TZM side (Ref 39,40,41).

Figure 16(i) and (j) presents the fracture morphology with a 0.4 mm offset. Observations indicate that the entire fracture surface is nearly flat, except for an intergranular fracture at the top. EDS results revealed that the Ti content at the upper was 60 at.%, while it was only 4 at.% in the middle and increased to 60 at.% the bottom, indicating that the fracture transpired along the FZ-base metal-FZ caused by the leakage of heat input with further offset. It can be seen from Fig. 16(c), (d), (e), (f), (g) and (h) that there are fewer neat crystal planes for intergranular fracture, and the tensile test results showed that the elongation increased with beam offset.

The fracture analysis revealed that the FZ becomes mechanically vulnerable due to the significant concentration of solid solution when the deflection beam is 0 mm. As the deflection beam increases to a range of 0.1-0.3 mm, the mechanical vulnerability of the joint transitions into a heat-affected zone (HAZ) of TZM. The occurrence of MoO2 and other phases precipitating and the thermal recrystallization of the heat-affected zone (HAZ) may be the causes of this phenomenon. Further increase in the beam offset would result in inadequate welding and inferior mechanical properties.

4 Conclusion

In this work, the influence of the electron beam offset on the morphology, microstructure, element distribution, and mechanical properties of the TZM/TC4 welded joint was evaluated. The conclusions can be summarized as follows.

  1. 1.

    As the offset was 0 and 0.1 mm, pores appeared near the fusion line of TZM, while the pores vanished as the offset increased. The fusion line of TZM side became straight with the increase of offset. The microstructure in the fusion zone underwent the transition from the fragmented dendritic structure at 0 mm offset to the smooth dendritic structure at 0.1 and 0.2 mm offsets, ultimately resulting in the martensite structure at more significant offsets.

  2. 2.

    The Mo element sharply decreased near the fusion line of TZM. The content of Mo element in the fusion zone was reduced from 20 to 1.55 at.%, which was related to the decrease of molten TZM with offset increased. The phases in the welded joints ranged from (Mo,Ti) solid solution to β-Ti and ω-Ti when the offset was 0.1 mm, and then to the α' and α'' phases with offset of 0.2 mm. When the offsets were 0.3 and 0.4 mm, the α' phase became dominant.

  3. 3.

    The minimum microhardness of the FZ was 310 HV with 0.1 mm offset, while maximum was 380HV with 0.3 and 0.4 mm offset. The tensile strength of the joint initially increased and then decreased with the offset range of 0-0.4 mm. The maximum tensile strength of 480 MPa were obtained at 0.2 mm offset. When the offset was 0 mm, the fracture occurred in the fusion zone due to the (Mo,Ti) solid solution, while the fracture occurred on the HAZ of TZM with 0.1-0.3 mm offset. The fracture path was FZ-base metal-FZ with the offset of 0.4 mm related to the inadequate welding.