Journal of Materials Science & Technology, 2021, 61(0): 16-24 DOI: 10.1016/j.jmst.2020.05.043

Research Article

Effect of interlayer addition on microstructure and mechanical properties of NiTi/stainless steel joint by electron beam welding

H. Niua,b, H.C. Jiang,a,*, M.J. Zhaoa, L.J. Rong,a,*

aCAS Key Laboratory of Nuclear Materials and Safety Assessment, Institute of Metal Research, Chinese Academy of Science, Shenyang, 110016, China

bSchool of Materials Science and Engineering, University of Science and Technology of China, Hefei, 230026, China

Corresponding authors: * E-mail addresses:hcjiang@imr.ac.cn(H.C. Jiang),ljrong@imr.ac.cn(L.J. Rong).

Received: 2020-03-19   Accepted: 2020-05-6   Online: 2021-01-15

Abstract

NiTi/Stainless Steel (SS) sheets have been welded via a vacuum electron beam welding process, with three methods (offsetting electron beam to SS side without interlayer, adding Ni interlayer and adding FeNi interlayer), to promote mechanical properties of the NiTi/SS joints. The joints with different interlayers are all fractured in the weld zone near the NiTi side, which is attributed to the enrichment of intermetallic compounds including Fe2Ti and Ni3Ti. The fracture mechanisms of different joints are strongly dependent on the types of interlayers, and the joints without interlayer, adding Ni interlayer and adding FeNi interlayer exhibit cleavage fracture, intergranular fracture and mixed fracture composed of cleavage and tearing ridge, respectively. Compared with the brittle laves phase Fe2Ti, Ni3Ti phase can exhibit certain plasticity, block the crack propagation and change the direction of crack propagation. The composite structure of Ni3Ti and Fe2Ti will be formed when the FeNi alloy is taken as the interlayer, which provides the joint excellent mechanical properties, with rupture strength of 343 MPa.

Keywords: NiTi ; Stainless steel ; Electron beam welding ; Interlayer ; Mechanical property

PDF (5493KB) Metadata Metrics Related articles Export EndNote| Ris| Bibtex  Favorite

Cite this article

H. Niu, H.C. Jiang, M.J. Zhao, L.J. Rong. Effect of interlayer addition on microstructure and mechanical properties of NiTi/stainless steel joint by electron beam welding. Journal of Materials Science & Technology[J], 2021, 61(0): 16-24 DOI:10.1016/j.jmst.2020.05.043

1. Introduction

As a type of shape memory materials, NiTi alloy with nearly equiatomic compositions continuously has a wide application due to its pseudoelastic, perfect shape memory effect, corrosion resistance and biocompatibility properties [[1], [2], [3], [4], [5]]. It is broadly used in the design of medical devices for orthopedics, dentistry and interventional medicine, such as catheter guide wires, stents or coil anchors [6]. At the same time, stainless steel (SS), as a biomedical material in common use, has excellent mechanical property, machining property, corrosion resistance and a low cost [7]. With the development of surgical instruments and medical treatment, medical devices composed by single material cannot meet the demand in the clinic, instead, medical devices integrated by different materials can provide better and more flexible problem solutions. Therefore, these two kinds of biomedical materials often need to be joined together.

Due to the small size and precision of medical device, fusion welding is usually used in this field. However, the different thermophysical and chemical properties of NiTi and stainless steel often lead to low plastic brittle crack related to the formation of intermetallic compounds such as Fe2Ti, Ti2Ni and Ni3Ti in the weld zone, as well as the existence of significant residual stresses due to physical mismatch of the welded materials [8]. Meanwhile, intermetallic compounds formed in the weld can also cause the brittle fracture of the joint, and result in lower mechanical properties compared with the base materials [[9], [10], [11], [12]]. Therefore, it is difficult to achieve a high quality NiTi/SS welding joint. So far, most of the studies focus on laser welding of NiTi and stainless steel [[13], [14], [15], [16]] owing to its characteristics: low heat input, high energy density and reproducibility. But some problems still remain: the existence of intermetallic compounds in the weld zone and low mechanical properties of the joint, so most of the studies can only conduct on submillimetric wires welding.

To solve these problems, scholars proposed to add interlayer between NiTi and stainless steel during welding and different pure metal (such as Ni, Cu, Ta, Co, etc.) interlayers were added in studies [[17], [18], [19], [20], [21], [22], [23], [24]]. Most of the weld microstructure exhibit dendrite structure, and the strength of dissimilar joints can be improved to 300-500 MPa after adding interlayer. Among them, Ni interlayer was widely adopted, as Ni interlayer with appropriate thickness can improve the mechanical property of the welded samples by suppressing intermetallic compounds, changing their distribution in the weld zone and modifying the joint microstructure [[17], [18], [19], [20]]. Studies show that the strength of joints reached the maximum values (372 MPa) when weld Ni content was 47.25 wt.% in the weld [19]. However, the wide Ni interlayer would degrade the joint properties because of forming more Ni3Ti phase, gas-pores and shrinkage cavities in the weld metals [19]. It was also proved that Cu interlayer with appropriate thickness can improve the mechanical properties of NiTi/SS joints and the maximum strength of joints can reach 521 MPa [21,22]. As the Cu filler metal thickness increased, the tensile strength and elongation of the joints were increased due to the increase of Cu solid solution and the reduction of intermetallic compounds. Further increasing Cu interlayer thickness, the joint properties decreased because of the formation of more Cu-Ti intermetallic compounds at weld/NiTi interface. It was confirmed that other elements such as Co and Ta [23,24] can also improve the properties of the joints by suppressing the harmful intermetallic compounds. The maximum strength of the joints with Co and Ta interlayer can reach 347 MPa and 251 MPa, respectively. The results of these researches indicate that reducing the formation of brittle intermetallic compounds by selecting a kind of appropriate interlayer is an effective method on improving the joint strength of NiTi/SS weldment. What need to be mentioned is that all of the above investigations were conducted on submillimetric wires. Due to the higher welding residual stress, very few investigations on NiTi and SS plates welding are reported.

At present, most of the interlayer material selections about NiTi/SS welding focuses on pure metal, and there are no researchers have been exploited with a type of alloyed interlayer by composition design. It is not known yet whether the intermetallic compounds can be suppressed completely by some kind of alloyed interlayer. FeNi alloy has better mechanical properties and weldability compared to pure Ni and no more elements will be introduced into the weld zone compared with Cu, Co, Ta interlayer. Meanwhile NiTi is the nickel-rich material and SS is iron-rich material, due to the ingredients of FeNi in between of the two base materials, FeNi can improve the uniformity of compositions in the weld zone. Beyond that, the high solubility of Ti element in FeNi alloy can be expected to effectively inhibit the precipitation of titanium-rich compounds (such as Fe2Ti, Ni3Ti, and Ti2Ni) which would induce low mechanical properties of the joint.

So, this study is aiming to investigate the effect of different interlayer addition on microstructure and mechanical properties of NiTi/AISI 304 plate dissimilar joint. Vacuum electron beam welding technology is adopted as it has the same advantages with laser welding as described previously, other than that, its vacuum environment can prevent the introduction of some harmful elements, such as O, N, and H into the weld zone. To the best of authors’ knowledge, no research works have been carried out on NiTi/SS dissimilar electron beam welding by FeNi interlayer. For better explanation of the relationship between microstructure and mechanical properties, samples with Ni interlayer, FeNi interlayer and without interlayer were investigated systematically.

2. Materials and experiments

2.1. Materials and sample preparation

The experimental investigations were carried out using NiTi shape memory alloy (entire B2 phase at room temperature) and AISI 304 stainless steel (γ-Fe). Chemical compositions and mechanical properties are listed in Table 1. The experimental materials were manufactured by means of vacuum smelting, and then rolled into 40 mm thickness plates and the welding samples with the dimension of 30 mm × 15 mm × 1.6 mm were machined from the plates. Before welding, these samples should be operated with surface treatment using the grinding machines until metallic luster. Interlayers of pure Ni (99.95 wt.%) and FeNi (containing 50 wt.% Ni) were made into 0.5 mm thick by cold rolling and then cut into 1.6 mm × 15 mm size. All base materials and interlayers should be ultrasonically cleaned in alcohol bath for 10 min to remove the dirt on the surface before welding.

Table 1   Chemical compositions (wt.%) and mechanical properties of base materials.

MaterialNiTiFeCrMnC + Si + P+SRp0.2(MPa)Rm(MPa)
NiTi55.8644.14----339663
AISI 3048.06-72.2218.011.010.70316779

New window| CSV


2.2. Electron beam welding

For the electron beam welding process, a workable MEDARD46 type vacuum electron beam welding machine made in the French TECHMETA company was employed. The maximum electron gun power and acceleration voltage were 15KW and 60KV, respectively. Base material sheets and interlayers were assembled on a self-made welding fixture to make sure a close contact in vacuum chamber during the welding. The electron beam was concentrated on the center of the interlayer, and moved parallel to the interlayer according to the prior set program. With respect to the sample without interlayer addition, the electron beam should be concentrated on stainless steel base materials, since it can lead to a softer weld zone and a higher tensile strength, due to the formation of less brittle intermetallic compounds compared to that when the laser beam was placed at the NiTi/SS interface [20]. Optimized the offset parameter, the weld has a better welding quality of weld and high performance when electron beam offsets 200 μm to stainless steel side. Besides, the thickness of interlayer has also been optimized, and it was found that the joints with 500 μm thickness interlayer have the better mechanical properties. Fig. 1 shows the schematic diagram of the electron beam welding process. The electron beam welding parameters are optimized to obtain a high and stable strength of joints. Table 2 gives the optimized parameters in the welding. In order to evaluate welding quality, all the samples will be detected by X-ray inspection.

Fig. 1.

Fig. 1.   The schematic diagram of electron beam welding. (a) without interlayer, (b) with interlayer.


Table 2   Welding parameters used in the study.

Welding MethodAcceleration Voltage (KV)Beam Current (mA)Focusing Current (mA)Welding Speed (mm/min)Focal Length (mm)
EBW608-1223251000260

New window| CSV


2.3. Metallography

Samples for metallographic study were polished with sandpapers from 150 to 2000 grits and then polished with 2.5 μm diamond paste. The samples were etched with the solution prepared by HF: HCl: HNO3: H2O = 1:3:4:10 to reveal the microstructure. The microstructure and fracture morphology of the joints were characterized by optical microscopy (OLMPUS GX51) and scanning electron microscopy (SEM, ZEISS MERLIN Compact) equipped with an Oxford Instrument EBSD detector NordlysNano.

Electron backscattered diffraction (EBSD) was performed to reveal the phases distribution in the weld zone. Samples for EBSD were polished in the way as same as the metallographic samples, furthermore, samples should be processed by vibration polishing for 16 h. The polishing solution including the SiO2 particles with 50 nm in diameter was used in this process. On the basis of different magnifications of SEM, 5 μm and 0.1 μm will be chosen as scan step size, respectively. Beyond that, TruPhase calibration model was used during EBSD measurement to distinguish different phases with similar structure, like Ni3Ti and Fe2Ti phase.

2.4. Test of mechanical properties

Static tensile tests were performed by INSTRON 5982 tensile testing machine at room temperature with the cross head speed of 1.6 mm/min. The shape and dimensions of the samples were shown in Fig. 2, which were processed according to ISO 4136-2012. After tensile test, the fracture surfaces and cross sections of the joint were examined to analysis the fracture mechanism. The micro hardness measurement was carried out by microscopic hardness meter 5103, with a load of 1000g-force and a dwell time of 15 s. Each test location was separated 0.25 mm from the adjacent one.

Fig. 2.

Fig. 2.   The schematic diagram of tensile sample.


3. Results

3.1. Microstructures of the joints

Three types of joints were obtained, respectively, with addition of Ni interlayer, FeNi interlayer and without interlayer addition. From the face and back photographs of welded joint shown in Fig. 3(a), it can be observed that the weld zone owns good weld configurations and no obvious weld cracks can be detected from the X-ray inspection in Fig. 3(b). However, some small gas pores exist inside the weld zone with Ni interlayer addition Fig. 3(b2), which are induced by the poor liquidity of pure Ni. The cross section images of welded joint show that the weld zone form “V” shape, wide at the top and narrow at the bottom, with the maximum weld width less than 2.5 mm, as typically shown in Fig. 3(c) for joint with FeNi interlayer.

Fig. 3.

Fig. 3.   (a) Face and back image of NiTi/SS electron beam welded sample. (b) X-ray inspection image of the three samples: 1-without interlayer, 2-with Ni interlayer, 3-with FeNi interlayer.(c) Cross section microstructure of the electron beam welded joint with FeNi interlayer.


The microstructure of weld joints is observed by EBSD. It can be found from the picture of phases distribution in Fig. 4 that there are two main regions in all of the weld zones: austenite region and intermetallic compound region. Austenite(γ-FeNi with solution titanium) shown in yellow is near to SS side, while intermetallic compounds shown in red which are consisted of Laves phase (Fe2Ti) and η phase (Ni3Ti) exist in the weld zone closer to NiTi side in the three joints. The austenite region is in a low resolution in the joint without interlayer, which may be due to the higher residual stress (Fig. 4(a)). However, there is a difference in phase volume fraction in three joint welds, as summarized in Table 3. Compared with the FeNi interlayer weld zone, weld zones without interlayer addition and with Ni interlayer have a higher volume fraction of intermetallic compounds and the volume fraction of intermetallic compounds reaches 47 % and 43 %, respectively, While, there was only 9% in the weld zone with FeNi interlayer addition. Apparently, the volume fraction of intermetallic compounds in the weld zone has been greatly reduced through FeNi interlayer addition.

Fig. 4.

Fig. 4.   EBSD phase distribution of NiTi/SS electron beam welding joints with different interlayers. (a) without interlayer, (b) Ni interlayer, (c) FeNi interlayer. The yellow part represents γ-(Fe,Ni) phase, the red part represents intermetallic compounds contain Fe2Ti phase and Ni3Ti phase.


Table 3   Volume fraction of phases in the weld zone.

No interlayerNi interlayerFeNi interlayer
Austenite53 %57 %91 %
Intermetallic compounds47 %43 %9 %

New window| CSV


Since most of the intermetallic compounds exhibit intrinsic brittleness, too many intermetallic compounds may lead to a brittle weld zone and do harm to the performance of joints. Fig. 5 presents the SEM micrographs of the intermetallic compound region in the three joints marked with white rectangle in Fig. 4, it is clear that they have different microstructure features. The microstructure of intermetallic compound region without interlayer addition and with Ni interlayer addition consisted mainly of fine equiaxed dendrite. However, the microstructure of the intermetallic compound region with FeNi interlayer addition consisted mainly of columnar dendrite with distinct preferred orientation. Generally, the dendrite grows along temperature gradient, from base material to the center of weld zone. A subsequent analysis of EBSD in Fig. 12 manifests that the dark phase is Fe2Ti phase and the bright one is Ni3Ti phase.

Fig. 5.

Fig. 5.   SEM micrographs showing intermetallic compounds region in NiTi/SS electron beam welding joint with different interlayer. (a) without interlayer, (b) Ni interlayer, (c) FeNi interlayer.


3.2. Test of mechanical properties

For the hardness test, the different colors in hardness nephograms represent different hardness ranges, as shown in Fig. 6. The specific values are listed on the right, the red color represents higher hardness, and the blue represents lower hardness. The weld zone is far harder than the two base materials, because of the intermetallic compounds and solution strengthening in the weld zone. There are more high hardness region in the weld zone without interlayer than the other two samples with interlayer addition. Make a comparison between hardness nephograms and phase distribution in Fig. 4, it can be found that the weld zone containing more intermetallic compounds shows higher hardness. For the uneven of the phase distribution, the hardness distribution in the weld zone shows inhomogeneous.

Fig. 6.

Fig. 6.   Vickers microhardness distribution of NiTi/SS electron beam welding joints with different interlayer. (a) without interlayer, (b) Ni interlayer, (c) FeNi interlayer. Weld seams were outlined by white lines from Fig.4.


The stress-strain curves of three joints and two base materials are shown in Fig. 7. From the figure, it can be seen that the joint with FeNi interlayer has the highest rupture strength up to 343 MPa, the joint with Ni interlayer is 264 MPa, while the joint without interlayer has the lowest rupture strength only about 174 MPa. The fracture strength of NiTi and SS base materials are 663 MPa and 779 MPa, respectively. The highest rupture strength of the joints approximately equals to the yield strength of NiTi base material.

Fig. 7.

Fig. 7.   Engineering stress-strain curves of electron beam welding joints with different interlayer addition and two base materials.


3.3. Fracture characteristics analysis

Fig.8(a, b, c) shows the lateral micrographs of fractured samples of three joints. It can be seen that the fracture location is closer to NiTi base material side in all of the three samples, where is the weld zone rather than the interface between weld zone and NiTi base material, as shown in Fig. 8(d, e, f) with higher magnification. The fracture location is corresponding to the intermetallic compound region shown in Fig. 4, which implies that intermetallic compound region in the weld zone determines mechanical property of the joint.

Fig. 8.

Fig. 8.   Lateral micrographs of NiTi/SS electron beam welded joints. (a,d) without interlayer, (b,e) Ni interlayer, (c,f) FeNi interlayer, (d-f) show the higher magnification of red box in (a-c). The interface of the weld zone and NiTi base material was marked by white dotted line.


SEM observation of fracture surfaces shows that three samples exhibit different fracture modes, as shown in Fig. 9. The fracture surface of joint without interlayer addition presents brittle feature, with a lot of cleavage facets on the fracture surface. The fracture surface of joint with Ni interlayer addition shows typical intergranular fracture characteristics and the second cracks can also be found. However, the joint with FeNi interlayer addition exhibits a ductile and brittle mixed fracture mode, with the huge number of facets and tearing ridges.

Fig. 9.

Fig. 9.   Fracture surface morphologies of welded joints. (a) without interlayer, (b) Ni interlayer, (c) FeNi interlayer.


4. Discussion

4.1. Interlayer designing

The multicomponent phase diagram of the various elements present in the base materials can help understanding the phase transition process in the weld zone. The presence of chromium in SS is neglecting, since its low content compared with Fe, Ni and Ti in the weld zone, which are three main elements to be considered in this case. Fig. 10 shows an isothermal section of the Fe-Ni-Ti ternary phase diagram according to J. Vannod et al. [6] and Cacciamani et al. [25]. The black line connects the two-initial phases of the base materials, SS in the low left corner of the isothermal section and NiTi in the middle of the right boundary. In the case of welding without interlayer addition and the electron beam focusing on centerline, the average composition of the weld zone is located in the ternary region composed of Fe2Ti, Ni3Ti and γ-(Fe,Ni), as shown by O point in Fig. 10. As is well-known, intermetallic compounds especially Fe2Ti present the intrinsic brittleness and deteriorate mechanical properties of the joint. So, some methods should be taken to decrease the amount of the intermetallic compounds in the joint and it is feasible to change the compositions of welding pool.

Fig. 10.

Fig. 10.   Fe-Ni-Ti ternary phase diagram, isothermal cut at 1000℃.


In this investigation, three methods have been taken to decrease the amount of the intermetallic compounds. When offsetting electron beam 200 μm to SS side, the melting volume of NiTi base material will be decreased, the average composition of the welding pool (O point) will change along the OA direction to the location nearby A point. The average composition is even close to the γ-(Fe,Ni) single phase region and far away from Fe2Ti and Ni3Ti phase region. In this case, the quantity of γ-(Fe,Ni) phase is increased, while those of the Fe2Ti phase and Ni3Ti phase are decreased in welding pool. Obviously, the quantity of Ni3Ti phase goes down faster than Fe2Ti phase. When 500 μm thickness Ni interlayer is added between the base materials, the average composition will move along the OB direction to the location nearby B point. The average composition is even close to the γ-(Fe,Ni) and Ni3Ti phase region and far away from Fe2Ti single phase region. In this case, the quantity of Fe2Ti phase is decreased, and both of the Ni3Ti and γ-(Fe,Ni) phases are increased. Although the above two ways can play a role of decreasing the number of intermetallic compounds, there are still some limitations. About offsetting electron beam to SS side(OA), the offset distance should be moderate as the binding force of the joint will be decreased with the increase of the distance. Similarly, adding Ni interlayer (OB) may increase Ni3Ti phase but produce redundant gas pores in the welding due to the poor liquidity of Ni. For these reasons, FeNi interlayer is considered to solve these problems. When 500 μm FeNi interlayer is added, the average compositions of welding pool will move to the location nearby C point along the OC direction and the average compositions are located in γ-(Fe,Ni) single phase region. The above composition points (A, B and C) are calculated from the function of weld zone fitted curves.

As a result, both of the Fe2Ti phase and Ni3Ti phase will be reduced in this case. The phase distribution diagram (Fig. 4) verifies these predictions that the joint with addition of FeNi interlayer contains the fewest intermetallic compounds among the three ways (Table 3). In theory, the welding pool should be austenitic completely when adding 500 μm FeNi interlayer according to the calculation result. However, since the inhomogeneous distribution of chemical elements and insufficient cooling speed, there are still some little intermetallic compounds exited in the weld zone (Fig. 4(c)).

4.2. Fracture mechanism

From phase distribution diagram (Fig. 4) and the microstructure on the lateral section (Fig. 8), it can be found that tensile samples are fractured in the intermetallic compound region. However, the three samples exhibited different fracture modes, as shown in Fig. 9. In order to make sure the relation between intermetallic compounds and fracture mechanism, EBSD was used to conduct phase analysis at these regions. Since the similar crystal structures (HCP) and lattice constant between Fe2Ti phase and Ni3Ti phase, traditional calibration mode is hard to distinguish the two phases effectively such as in Fig. 4. Therefore, TruPhase calibration mode was used to identify the Fe2Ti and Ni3Ti phase distribution. Different from the traditional model, TruPhase calibration mode can collect information about structures and elements at the same time. As a result, the two HCP phases can be distinguished more accurately.

Cracks on the lateral section near the fracture surface were analyzed by SEM (Fig. 11(a-c)) and EBSD (Fig. 11(e-f)) respectively. Fe2Ti is dominated phase in intermetallic compounds region, while Ni3Ti phase is in the shape of continuous net inside the Fe2Ti, when offsetting electron beam to SS without interlayer (Fig. 11(a, d)). The cracks can across through Fe2Ti phase, and deflects at the locations where Ni3Ti phase existed. Compared with Fe2Ti phase, Ni3Ti phase blocks the propagation of crack obviously. Also, it can be found that cracks prefer to passes through Ni3Ti phase with thin thickness.

Fig. 11.

Fig. 11.   SEM and EBSD figures show microcracks near the fracture in intermetallic compounds region. (a, d) Without interlayer (b, e) Ni interlayer (c, f) FeNi interlayer. Blue color represents Fe2Ti phase and red color represents Ni3Ti phase. By reason of the internal stress inside the intermetallic compounds region, a few regions cannot be distinguished effectively, these regions show grey color. The cracks were marked by white dotted line.


The microstructure near the fracture section of the sample adding Ni interlayer is shown in Fig. 11(b, e). In this case, Ni3Ti is the majority phase in the intermetallic compound region, while Fe2Ti was distributed in the shape of net. Cracks only form and propagate inside the Fe2Ti phase and are blocked by Ni3Ti phase.

The microstructure near the fracture surface of the joint adding FeNi interlayer is shown in Fig. 11(c, f). Compared with joints without interlayer and with Ni interlayer, the quantity of intermetallic compounds containing Fe2Ti and Ni3Ti phase has a significant reduction and keeps relative balance in the weld zone (OC ways in Fig. 10). Most of Fe2Ti phase precipitates preferentially and presents itself with columnar dendrites, while Ni3Ti phase distributes between the dendrites of Fe2Ti. The columnar dendrite microstructure can be seen clearly in Fig. 5(c). Microcracks occur inside the Fe2Ti dendrites, and then be blocked by Ni3Ti phase in the interdendritic regions. It is observed that Ni3Ti phase can hinder crack propagation in intermetallic compound region, effectively.

Several studies have shown that Fe2Ti (Laves phase) is a brittle phase and Ni3Ti (η phase) phase can present certain plasticity at room temperature [26,27]. For this reason, the main factor that determines the weld zone hardness (Fig. 6) is the quantity of Fe2Ti phase.

From above results, the different crack propagation modes have been drawn schematically in Fig. 12. Red and blue colors represent Ni3Ti phase and Fe2Ti phase, respectively, while black lines represent microcracks. Obviously, three fracture modes have huge difference. The crack propagates through the intermetallic compound region directly in the sample without interlayer addition, because a large number of Fe2Ti phase exists in this region. Slight deflection will occur when the crack meet Ni3Ti phase with enough thickness. Therefore, the joint without interlayer has the characteristic of cleavage fracture (Fig. 12(a)) and a low mechanical property. In Fig. 12(b), Fe2Ti phase has a significant reduction in intermetallic compounds region. Cracks only form and propagate inside the Fe2Ti phase, which result in the fractured joint has the characteristic of intergranular fracture. However, when the FeNi interlayer was added (Fig. 12(c)), the microstructure of the intermetallic compound region forms a composite structure which consists of Fe2Ti and Ni3Ti phase. Under the applied stress, cracks occur preferentially inside Fe2Ti phase and their propagation can be blocked by Ni3Ti phase, since Ni3Ti phase can present certain plasticity at room temperature. If the applied stress continues to increase, a severe plastic deformation will occur in Ni3Ti phase and it will be teared in the end. Therefore, the joint with FeNi interlayer has the characteristic of cleavage and ductile fracture at the same time. This shows that this kind of composite structure can avoid the formation of successive Fe2Ti phase, and the propagation of crack can be blocked, effectively. So, the joint with FeNi interlayer addition owns an excellent mechanical property due to its unique fracture mode.

Fig. 12.

Fig. 12.   Schematic drawing of microcracks propagation in intermetallic compound region: (a) Without interlayer, (b) Ni interlayer, (c) FeNi interlayer.


Through macro observation, the columnar dendritic in intermetallic compound region also changes the main crack propagation path in FeNi adding joint. From the micrograph of lateral section of fractured joint in Fig. 13, we can find that main crack propagation is deflected. Three crack deflection regions have been investigated by means of EBSD, respectively. In these regions (Fig. 14), dendrite growth has a distinct orientation, which has a distinct effect on the crack propagation direction. The direction of the main crack propagation is from top to bottom. In Fig. 14(a, d), the original direction of crack propagation was parallel to the direction of the tensile stress. When meeting the exquiaxed dendrite region (dendrite growth doesn’t have a distinct orientation), the main crack will deflect under the applied tensile stress, changed from a model-II crack (paralleling to the direction of the tensile stress) to a model-I crack (perpendicular to the direction of the tensile stress). This is because the dendrites in this region don’t have a uniform direction. In Fig. 14(b, e), the original direction of crack propagation was perpendicular to the direction of the tensile stress. With the expansion of the crack, it will be blocked by NiTi base material, as was shown in Fig. 8(f). Therefore, the crack propagation path changes to along dendrite growth direction with less resistance, as the Fe2Ti phase is more continuous in this direction. A model-I crack changed to a mode-II crack. For the same reason, in Fig. 14(c, f), the original direction of the crack propagation was perpendicular to the direction of the tensile stress. By reason of the larger resistance in this direction, the crack changed its propagation direction, to paralleling dendrite growth direction. a model-I crack changed to a model-II crack. The main crack propagation process can be represented schematically in Fig. 15. The main crack initiates at the surface of the weld zone in intermetallic compounds region under applied stress. The direction of crack propagation was perpendicular to the direction of the tensile stress until it reaches the NiTi base material. Due to the high crack propagation resistance of NiTi base material, the crack has to choose another path. Obviously, expanding along the brittle columnar dendrites of Fe2Ti is a choice. By the reason of continuous Fe2Ti distribution in this direction, there has been less crack propagation resistance. When meeting the exquiaxed dendrite region, the crack will deflect to the original direction under the applied tensile stress. A mass of columnar and exquiaxed dendrites of Fe2Ti phase deflect the main crack, make a step-shape crack propagation path. On account of the “V” shape weld zone, the deflection will occur multiple times. As a result, the columnar dendrite in FeNi interlayer addition joint can increase the crack propagation resistance, thus improving mechanical performance of the welded joint. However, this phenomenon didn’t occur in the other two samples (without interlayer and with Ni interlayer).

Fig. 13.

Fig. 13.   Fracture section microstructure of the joint with FeNi interlayer addition.


Fig. 14.

Fig. 14.   SEM and EBSD figures of the main crack deflections in FeNi interlayer addition weld. (a, d) present a region in Fig. 13. (b, e) present b region in Fig. 13. (c, f) present c region in Fig. 13.


Fig. 15.

Fig. 15.   Schematic drawing of main crack deflection in the joint with FeNi interlayer. Intermetallic compounds region is on the right side of dotted line, and austenite region with simplified representation is on the left side of dotted line. The columnar and exquiaxed dendrites of Fe2Ti phase are shown in the figure. Ni3Ti phase is distributed around Fe2Ti dendrites. Green lines represent model-I cracks and red lines represent model-II cracks.


5. Conclusions

(1) Electron beam welding has been used to weld of NiTi/stainless sheet with addition of Ni and FeNi interlayers and without interlayer, among them the joint with FeNi interlayer addition has excellent mechanical properties and its fracture strength can reach 343 MPa.

(2) The microstructure of the weld zone has two distinct regions, which are composed of austenite and intermetallic compounds (Fe2Ti, Ni3Ti), separately, but the volume fraction of the intermetallic compound regions is different, with the 47 %, 43 % and 9% for joints without interlayer and with Ni interlayer and FeNi interlayer, respectively.

(3) All tensile samples with welding joints are fractured in the intermetallic compound region closer to NiTi base material, but their fracture modes are different, exhibiting cleavage fracture, intergranular fracture and mixed fracture of cleavage and tearing ridges for joints without interlayer, with Ni interlayer and FeNi interlayer, respectively. This is due to that brittle intermetallic compound Fe2Ti has different volume fraction and distribution in the intermetallic compound region, and the composite structure of Fe2Ti and Ni3Ti phase insure the NiTi/ AISI 304 steel joint with FeNi interlayer excellent mechanical property.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Reference

J.A. Shaw, S. Kyriakides, Acta Mater. 45 (2) (1997) 683-700.

DOI      URL     [Cited within: 1]

Y.T. Hsu, Y.R. Wang, S.K. Wu, C. Chen, Metall. Mater. Trans. 32 (3) (2001) 569-576.

DOI      URL     [Cited within: 1]

K. Otsuka, X. Ren, Prog. Mater. Sci. 50 (5) (2005) 511-678.

DOI      URL     [Cited within: 1]

Y.F. Zheng, B.B. Zhang, B.L. Wang, Y.B. Wang, L. Li, Q.B. Yang, L.S. Cui, Acta Biomater. 7 (6) (2011) 2758-2767.

DOI      URL     PMID      [Cited within: 1]

A new kind of biomedical shape memory TiNiAg alloy with antibacterial function was successfully developed in the present study by the introduction of pure Ag precipitates into the TiNi matrix phase. The microstructure, mechanical property, corrosion resistance, ion release behavior in simulated body fluid, cytotoxicity and antibacterial properties were systematically investigated. The typical microstructural feature of TiNiAg alloy at room temperature was tiny pure Ag particles (at submicrometer or micrometer scales with irregular shape) randomly distributed in the TiNi matrix phase. The presence of Ag precipitates was found to result in a slightly higher tensile strength and larger elongation of TiNiAg alloy in comparison with that of TiNi binary alloy. Furthermore, a maximum shape recovery strain of approximately 6.4% was obtained with a total prestrain of 7% in the TiNiAg alloy. In electrochemical and immersion tests, TiNiAg alloy presented good corrosion resistance in simulated body fluid, comparable with that of CP Ti and TiNi alloy. The cytotoxicity evaluation revealed that TiNiAg alloy extract induced slight toxicity to cells, but the viability of experimental cells was similar to or higher than that of TiNi alloy extract. In vitro bacterial adhesion study indicated a significantly reduced number of bacteria (S. aureus, S. epidermidis and P. gingivalis) on the TiNiAg alloy plate surface when compared with that on TiNi alloy plate surface, and the corresponding antibacterial mechanism for the TiNiAg alloy is discussed.

M.T. Andani, S. Saedi, A.S. Turabi, M.R. Karamooz, C. Haberland, H.E. Karaca, M. Elahinia, J. Mech. Behav. Biomed. Mater. 68 (2017) 224-231.

DOI      URL     PMID      [Cited within: 1]

Near equiatomic NiTi shape memory alloys were fabricated in dense and designed porous forms by Selective Laser Melting (SLM) and their mechanical and shape memory properties were systematically characterized. Particularly, the effects of pore morphology on their mechanical responses were investigated. Dense and porous NiTi alloys exhibited good shape memory effect with a recoverable strain of about 5% and functional stability after eight cycles of compression. The stiffness and residual plastic strain of porous NiTi were found to depend highly on the pore shape and the level of porosity. Since porous NiTi structures have lower elastic modulus and density than dense NiTi with still good shape memory properties, they are promising materials for lightweight structures, energy absorbers, and biomedical implants.

J. Vannod, M. Bornert, J.E. Bidaux, L. Bataillard, A. Karimi, J.M. Drezet, M. Rappaz, A. Hessler-Wyser, Acta Mater. 59 (17) (2011) 6538-6546.

DOI      URL     [Cited within: 2]

The biomedical industry shows increasing interest in the joining of dissimilar metals, especially with the aim of developing devices that combine different mechanical and corrosive properties. As an example, nickel titanium shape memory alloys joined to stainless steel are very promising for new invasive surgery devices, such as guidewires. A fracture mechanics study of such joined wires was carried out using in situ tensile testing and scanning electron microscopy imaging combined with chemical analysis, and revealed an unusual fracture behaviour at superelastic stress. Nanoindentation was performed to determine the mechanical properties of the welded area, which were used as an input for mechanical computation in order to understand this unexpected behaviour. Automated image correlation allowed verification of the mechanical modelling and a reduced stress strain model is proposed to explain the special fracture mechanism. This study reveals the fact that tremendous property changes at the interface between the NiTi base wire and the weld area have more impact on the ultimate tensile strength than the chemical composition variation across the welded area. (C) 2011 Acta Materialia Inc. Published by Elsevier Ltd.

M.F.F.A. Hamidi, W.S.W. Harun, M. Samykano, S.A.C. Ghani, Z. Ghazalli, F. Ahmad, A.B. Sulong, Mater. Sci. Eng. C 78 (2017) 1263-1276.

DOI      URL     [Cited within: 1]

J.P. Oliveira, R.M. Miranda, F.M.B. Fernandes, Prog. Mater. Sci. 88 (2017) 412-466.

DOI      URL     [Cited within: 1]

S. Belyaev, V. Rubanik, N. Resnina, V. Rubanik, O. Rubanik, V. Borisov, I. Lomakin, Phys. Procedia 10 (2010) 52-57.

DOI      URL     [Cited within: 1]

Q. Li, Y. Zhu, J. Mater. Process. Technol. 255 (2018) 434-442.

DOI      URL     [Cited within: 1]

M.J.C Oliveira, R.H.F Melo, T.M. Maciel, C.J. de Araújo, Mater. Chem. Phys. 224 (2019) 137-147.

DOI      URL     [Cited within: 1]

Q. Li, Y. Zhu, J. Guo, J. Mater. Process. Technol. 249 (2017) 538-548.

DOI      URL     [Cited within: 1]

H. Gugel, A. Schuermann, W. Theisen, Mater. Sci. Eng.A 481-482 (2008) 668-671.

URL     [Cited within: 1]

G.R. Mirshekari, A. Saatchi, A. Kermanpur, S.K. Sadrnezhaad, Opt. Laser Technol. 54 (2013) 151-158.

DOI      URL     [Cited within: 1]

The unique properties of NiTi alloy, such as its shape memory effect, super-elasticity and biocompatibility, make it ideal material for various applications such as aerospace, micro-electronics and medical device. In order to meet the requirement of increasing applications, great attention has been given to joining of this material to itself and to other materials during past few years. Laser welding has been known as a suitable joining technique for NiTi shape memory alloy. Hence, in this work, a comparative study on laser welding of NiTi wire to itself and to AISI 304 austenitic stainless steel wire has been made. Microstructures, mechanical properties and fracture morphologies of the laser joints were investigated using optical microscopy, scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS), X-ray diffraction analysis (XRD), Vickers microhardness (HV0.2) and tensile testing techniques. The results showed that the NiTi-NiTi laser joint reached about 63% of the ultimate tensile strength of the as-received NiTi wire (i.e. 835 MPa) with rupture strain of about 16%. This joint also enabled the possibility to benefit from the pseudo-elastic properties of the NiTi component. However, tensile strength and ductility decreased significantly after dissimilar laser welding of NiTi to stainless steel due to the formation of brittle intermetallic compounds in the weld zone during laser welding. Therefore, a suitable modification process is required for improvement of the joint properties of the dissimilar welded wires. (C) 2013 Elsevier Ltd.

G.R. Mirshekari, A. Saatchi, A. Kermanpur, S.K. Sadrnezhaad, J. Mater. Eng. Perform. 25 (6) (2016) 2395-2402.

DOI      URL     [Cited within: 1]

M. Mehrpouya, A. Gisario, M. Elahinia, J. Manuf. Process. 31 (2018) 162-186.

DOI      URL     [Cited within: 1]

S. Fukumoto, T. Inoue, S. Mizuno, K. Okita, T. Tomita, A. Yamamoto, Sci. Technol. Weld. Join. 15 (2) (2010) 124-130.

DOI      URL     [Cited within: 2]

J. Pouquet, R.M. Miranda, L. Quintino, S. Williams, Int. J. Adv. Manuf. Technol. 61 (1-4) (2012) 205-212.

DOI      URL     [Cited within: 2]

H.M. Li, D.Q. Sun, X.L. Cai, P. Dong, W.Q. Wang, Mater. Des. 39 (2012) 285-293.

URL     [Cited within: 4]

A. Shamsolhodaei, J.P. Oliveira, N. Schell, E. Maawad, B. Panton, Y.N. Zhou, Intermetallics 116 (2020), 106656.

URL     [Cited within: 3]

H. Li, D. Sun, X. Gu, P. Dong, Z. Lv, Mater. Des. 50 (2013) 342-350.

URL     [Cited within: 2]

C. Zhang, S. Zhao, X. Sun, D. Sun, X. Sun, Corros. Sci. 82 (2014) 404-409.

URL     [Cited within: 2]

H. Li, D. Sun, X. Cai, P. Dong, X. Gu, Opt. Laser Technol. 45 (2013) 453-460.

URL     [Cited within: 2]

C.H. Ng, E.S.H. Mok, H.C. Man, J. Mater. Process. Technol. 226 (2015) 69-77.

URL     [Cited within: 2]

G. Cacciamani, J. de Keyzer, R. Ferro, U.E. Klotz, J. Lacaze, P. Wollants, Intermetallics 14 (10-11) (2006) 1312-1325.

URL     [Cited within: 1]

K. Hagihara, T. Nakano, Y. Umakoshi, Acta Mater. 51 (9) (2003) 2623-2637.

URL     [Cited within: 1]

Y. Cao, J. Zhu, Y. Liu, Z. Lai, Z. Nong, Physica B Condens. Matter 412 (2013) 45-49.

URL     [Cited within: 1]

/