Laser melting deposition of Inconel625/Ti6Al4V bimetallic structure with Cu/V interlayers

A bimetallic structure (BS) made of Inconel625 (IN625) nickel-base superalloy with excellent high temperature properties and Ti6Al4V (TC4) titanium alloy with a light weight and a high strength has broad application prospects in aerospace engineering. However, the integrated manufacturing of the IN625/TC4 BS is a difficult research topic in the industry. In this work, the laser melting deposition (LMD) technology was used to prepare an IN625/TC4 BS without cracks and other metallurgical defects by adding Cu/V interlayers. The results show that the IN625/TC4 BS structure from the IN625 side to the TC4 one can be divided into four regions: IN625 region (region A) → Interlayers/IN625 transition region (region B) → TC4/Interlayers transition region (region C) → TC4 region (region D). The phase compositions of these regions are: γ-Ni + laves → (Ni, Cu)ss + (V, Cr)ss + TiNi → α-Ti + β-Ti + Ti2Ni → α-Ti + β-Ti. The Vickers hardness distribution is uneven in all regions, and the highest value (about 590.0 HV) is achieved in region B. The tensile strength of the IN625/TC4 BS with Cu/V interlayers reaches nearby 514.5±9.5 MPa at room temperature, and fractures are initiated in region B.


Introduction
Bimetallic structure (BS) materials comprise two distinct metals or alloys; they possess exceptional properties and a wide range of applications. BS materials have been developed to meet the demands of the modern aerospace industry and other high-tech fields, catering to the need for repeated operations in harsh environmental conditions [1]. Thanks to their unique mechanical properties, oxidation resistance, and corrosion resistance at high temperature, the Inconel625 (IN625) nickel-based superalloy is widely used in the manufacturing of high-temperature components of aeroengines [2,3]. On the other hand, the high specific strength, good corrosion resistance, and creep resistance of the Ti6Al4V (TC4) titanium alloy make it indispensable in lightweight structural parts of aerospace vehicles [4,5]. The combination of nickel-based superalloys and titanium alloys into integrated BSs can fully exploit their respective characteristics. This enables one to enhance the service performance of structural parts but also reduce their weight, thereby improving the thrust-to-weight ratio of aircrafts [6,7]. However, due to the huge differences between the physical and chemical properties of the two metals, the direct combination of related constituents causes the emergence of numerous brittle intermetallic compounds (IMCs) and large internal stresses, which increase crack sensitivity in the transition region and cause the nucleation of metallurgical defects, such as cracks [8,9]. Therefore, being able to combine nickel-based superalloys and titanium alloys with minimal side effects remains a great engineering challenge. Laser melting deposition (LMD) is a new type of free-form fabrication method. The intelligence, high Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. efficiency, low cost, and ability to attain a precise structural control make LMD an important approach to fabricate critical parts in the aerospace field [10][11][12][13]. During LMD, the heat input can be properly controlled, thus reducing the thermal stress of the component. At the same time, the design of the components can be realized upon their forming, which prevents the formation of brittle phases [14][15][16], thus decreasing the crack sensitivity of the constituents. Therefore, LMD is an effective way to prepare nickel-base superalloy/titanium alloy BSs (Ni-/Ti-BSs).
Up to now, many attempts have been made to obtain Ni-/Ti-BSs with desirable characteristics. For instance, Onuike et al [17] fabricated a TC4/Inconel718 (IN718) BS using the LMD technology. Firstly, 12 TC4 layers were deposited on the titanium alloy substrate and then covered with IN718. However, delamination cracking occurred when two IN718 layers were deposited because of the generation of Ti-Ni IMCs. Domack et al [18] investigated a gradient TC4/IN718 BS prepared via LMD. According to their results, macroscopic cracking emerged in the 40 vol% TC4+60 vol% IN718 transition layer, which was due to the formation of abundant IMCs and the generation of severe internal stress. To address the above issues, selecting a suitable material as the interlayer between the nickel-based alloy and the titanium alloy was proposed as a reliable solution to inhibit the generation of Ti-Ni IMCs. In this regard, copper has been by far the most commonly used interlayer material. For instance, pure Cu interlayers were utilized by Shang et al [19] to prepare a TA15/IN718 BS via LMD by adding pure Cu interlayers. Although the Ti-Ni IMCs were successfully inhibited, numerous Ti-Cu IMCs were generated at the same time. Zoeram et al [20] studied an NiTi/TC4 BS with a Cu interlayer obtained through the YAG welding technology. It was shown that the formation of Ti-Cu IMCs seriously affected the mechanical properties of welded joints whose maximum tensile strength and elongation were about 300 MPa and 3.3% only. Therefore, finding an ideal interlayer between titanium alloy and Cu that could inhibit the nucleation of Ti-Cu IMCs has become urgently required in the manufacturing of the IN625/TC4 BS. Gao et al [21] reported a 4J36/ TC4 BS with Ni/Cu/V multi-interlayers fabricated via the diffusion binding technology. While solid solutions, such as (Fe,Ni), (Cu,Ni), (V,Cu), and (Ti,V), were formed on the surfaces of the constituents, no IMCs were detected. Additionally, the tensile strength of the welded joints reached 249.3 MPa, and the elongation was up to 10.3%, which significantly reduced the crack sensitivity of the specimen. This finding indicates that the addition of V interlayers effectively prevents the formation of IMCs between the titanium alloy and Cu, whereas Cu/V interlayers inhibit the generation of IMCs between nickel and titanium constituents.
In view of the above, the IN625 nickel-based superalloy and the TC4 titanium alloy-the most common materials used in the aerospace field-were selected in this work to prepare the BS via LMD, in which Cu and V served as the interlayers. The microstructure, element distribution, phase composition, Vickers hardness distribution, and tensile strength of the IN625/TC4 BS at room temperature were analyzed in detail, providing a theoretical basis for the preparation of metallurgical BSs via LDM.

Material preparation
The substrate material used was an IN625 alloy plate with dimensions of 200 mm × 200 mm × 20 mm. Prior to the test, the substrate underwent sandpaper polishing and was then subjected to a chemical treatment to eliminate the oxide film and oil from its surface. Subsequently, the substrate was cleaned by scrubbing with acetone and placed in a vacuum drying oven, where it was heated to 200°C for 1 h. The powder materials consisted of a spherical alloy and metal powders prepared using the plasma rotating electrode method. The particle sizes of the IN625 and TC4 alloy powders were in the range of about 45-106 μm. The chemical compositions of the powders are shown in table 1. The particles of pure Cu (>99.9%) and V (>99.8%) metal powders were as large as 45-150 μm. Prior to the test, all powders were vacuum dried by heating them to 100°C for 2 h.

Manufacturing of the IN625/TC4 BS
The LMD equipment (Model: LDM2500, Make: Raycham, China) comprised a YLS-10000W fiber laser, a fourway powder feeding processing head, an inert-atmosphere processing chamber, a three-axis CNC table, a water  figure 1(a). Figure 1(b) depicts the physical diagram of the IN625/TC4 BS along with the sampling mode of the tensile part. Additionally, figure 1(c) presents the dimensional image of the stretched part.

Characterization methods
The metallographic and tensile specimens were extracted along the building direction (BD) from the LMDprocessed specimens through electric discharge machining (EDM) wire cutting. The metallographic specimens were first ground with 400#-5000# sandpapers, then mechanically polished with an SiO 2 suspension, and    In figure 2(b), the microstructure below interface 1 (region A) is predominately dendritic, which is typical of IN625 (γ-Ni + laves) alloys [22]. The microstructure above interface 1 (the lower part of region B) is composed of fine equiaxed crystals (dark gray) and the microstructure (white bright) between them. According to the microstructure and EDS data, the area below interface 1 (region A) mainly includes Ni and Cr elements, which is consistent with the characteristics of region A. Above interface 1 (the lower part of region B), there is an increase in the amounts of V, Ti, and Cu elements, while those of Ni and Cr decrease. In the fine equiaxed crystals (dark gray), the waveforms of V and Cr are the same, and the Cr content increases markedly. Therefore, it is inferred that the fine equiaxed crystals might originate from the solid solution formed by the V and Cr elements. The white bright microstructure between the crystals is mainly composed of abundant Ni elements and scarce Cu and Ti elements; so it is presumed to be a solid solution formed by the Ni, Cu, and Ti elements. According to figure 2(c), the microstructure below interface 2 (the upper part of region B) mainly consists of dendritic crystals (dark gray) and precipitates (white and light gray). The microstructure above interface 2 (the lower part of region C) exhibits quasi-equiaxed crystals (dark gray) and white inclusions between them. Given the EDS line scanning results from the bottom (the upper part of region B) to the top (the lower part of region C) near interface 2, the Ti content increases significantly, while those of V, Ni, Cr, and Cu decrease drastically. Below interface 2 (the upper part of region B), the waveforms of V and Cr are basically the same, and the amount of dark gray dendrites clearly increases, which implies the formation of a solid solution by the V and Cr elements. The white precipitates mainly contain Ni, Ti, and Cu elements, which are assumed to be the corresponding solid solutions. The gray precipitates are composed of Ni and Ti elements, suggesting the formation of Ti-Ni IMCs. Above interface 2 (the lower part of region C), the quasi-equiaxed crystals (dark gray) comprise Ti and V elements, indicating the existence of the β-Ti phase. An increase in the Ni content in the white inclusions suggests the emergence of Ti-Ni IMCs.
As shown in figure 2(d), the structure below interface 3 (the upper part of region C) is mainly composed of coarse equiaxed crystals, the inclusions between them (white bright color), and a small amount of precipitates inside the crystals (white bright color). The area above interface 3 (region D) presents coarse columnar crystals and crystal boundaries, and the inner part of the grains exhibits short-rod-like phases (dark gray) and other phases (white) between them, which is typical of the TC4 microstructure (α-Ti + β-Ti) [4]. According to the EDS line scans from the bottom (the upper part of region C) to the top (region D) near interface 3, the quantity of Ti elements gradually increases, while those of V, Ni, Cr, and Cu elements decrease slightly. Below interface 3 (the upper part of region C), the coarse equiaxed crystals are mainly composed of Ti and V elements, implying the formation of the β-Ti phase. The increase in the content of Ni elements at the crystal boundary (white bright color) and a small amount of precipitates (white bright color) inside the crystals indicate the appearance of Ti-Ni IMCs. Above interface 3 (region D), the microstructure is mainly composed of abundant Ti elements and scarce V elements. Given the microstructure data, the short-rod-like structure (dark gray) is attributed to the α-Ti phase, and the inclusions (white) between the rods are ascribed to the β-Ti phase.
In general, the microstructure within the four regions of the IN625/TC4 BS with Cu/V interlayers continuously and uniformly transitions without the formation of any metallurgical defects, such as cracks, and metallurgical bonding is realized at all three interfaces.
It is worth noting that according to the LMD parameters (table 1), the thicknesses of the pure Cu and V layers should be about 0.6 mm, whereas the Cu/V interlayers should be 1.2 mm thick. However, the thickness of the transition region (region B + region C) was experimentally found to be about 5.6 mm. The reasons for the increase in the transition region thickness are as follows.
(i) Once the Cu interlayer is deposited on the IN625 substrate, the dilution rate of the latter increases due to the high laser output power. On the other hand, the atomic radii of Cu and Ni (127.8 and 124 pm, respectively) are quite close to each other, and both elements possess face-centered cubic crystal structures. According to the phase diagram of the Cu-Ni binary alloy [23], two elements can form a Cu-Ni solid solution. Therefore, during the deposition of the Cu layer, the convection of the molten pool and the similarity between the elements could have promoted the diffusion and migration of Ni, Cr, and other elements from IN625 to the Cu layer as well as the movement of Cu toward the IN625 layer, thus leading to the formation of the (Cu + IN625) transition layer.
(ii)During the deposition of the V interlayer, the dilution rate of the (Cu + IN625) transition layer increases owing to the superior melting point of the V elements and the high laser power. On the other hand, the difference between the atomic radii of V, Ni, and Cr (134, 124, and 128 pm, respectively) is quite small. According to the phase diagram of the ternary alloy [24], V can form a solid solution with Ni and Cr, and both V and Cr possess body-centered cubic lattices. Moreover, V is even able to form a continuous solid solution with Cr. In the course of the V layer deposition, the convection of the molten pool and the similarity between the elements promote the diffusion and migration of Ni, Cr, and other elements from the (Cu + IN625) transition layer to the V deposition layer as well as the diffusion and migration of V elements toward the (Cu + IN625) transition layer.
(iii) As soon as TC4 is deposited after the V interlayer deposition, it dilutes the V interlayer to a certain extent, causing the diffusion of V elements to TC4. On the other hand, V is an inherent element of the TC4 alloy, possessing a stable isomorphic β-phase, which can be infinitely solid-soluble in the β-Ti phase through replacement [25].
Once TC4 is deposited, many Ti elements can easily diffuse toward the V intermediate layer or even through it toward region B.
The above circumstances result in the greater thickness of region B + region C (5.6 mm) relative to the theoretical thickness of the Cu/V interlayers (1.2 mm).

Phase composition of the IN625/TC4 BS
The XRD data on the LMD IN625/TC4 BS with Cu/V interlayers are shown in figure 3. With the introduction of the interlayers, a variety of complex phases are precipitated in the BS transition region. Along the region A → region B → region C → region D direction, the corresponding phase compositions are γ-Ni + laves→(Ni, Cu) + (V, Cr) + TiNi → α-Ti + β-Ti + Ti 2 Ni → α-Ti + β-Ti. Moreover, Ti-Ni IMCs still exist in the transition region, which might be caused by the insufficient thickness of the Cu/V interlayers, which results in the impossibility to completely block the diffusion and migration of Ni and Ti to regions C and B, respectively.
The typical microstructures and magnified images of the four regions of the LMD IN625/TC4 BS with Cu/V interlayers are illustrated in figure 4, and the EDS results for each point in the figure are given in table 3. Figure 5 displays the ternary phase diagrams of the Ni-Cr-V and Ni-Cu-Ti alloys [24]. Figure 4(a) depicts the microstructure within region A, revealing the presence of dendrites (dark gray) and precipitates (white) between them. The corresponding EDS data were acquired at points 1 and 2, respectively. At point 1, the main elements are Ni (63.42%) and Cr (22.16%), whereas other alloying elements are less abundant. Given the XRD results, the dendrites are ascribed to the γ-Ni solid solution [22]. Compared with point 1, the contents of Ni and Cr at point 2 are lower, while those of Mo, Nb, Fe, and Ti are higher, and the atomic ratio of the elements basically conforms to the (Nb, Ti, Mo)(Ni, Cr, Fe)2 structure, so it is inferred that the inclusions correspond to the laves phase [22].     Figure 4(d) displays the microstructure of the lower part of region C, highlighting the presence of quasiequiaxed crystals (light gray) and crystal boundary precipitates (dark gray), and the relevant EDS analysis locations are points 8 and 9, respectively. The main element composition at point 8 is 76.29% Ti, 8.09% V, and 5.59% Ni, which is attributed to the β-Ti phase according to the XRD analysis. The microstructure at point 9 is mainly composed of 59.77% Ti and 32.58% Ni, and the atomic ratio of Ti and Ni is about 2:1. Therefore, the precipitates at the grain boundaries are ascribed to the Ti 2 Ni phase. Figure 4(e) depicts the microstructure of the upper part of region C, revealing the coarse equiaxial crystals (light gray) with the short acicular structures (dark gray) and precipitates (white) inside the crystals and at their boundaries. The corresponding EDS locations are points 10-13, respectively. The microstructure at point 10 is mainly composed of 75.2% Ti, 7.97% V, and 5.73% Ni, which is attributed to the β-Ti phase according to the XRD analysis. The investigation at point 11 reveals the presence of Ti (81.47%) and Al (9.08%), which indicates the existence of the α-Ti phase. The local structure at point 12 consists of 57.93% Ti and 29.6% Ni with an atomic ratio of about 2:1, indicating the Ti 2 Ni composition. The same phase (Ti 2 Ni) is detected at point 13 with 60.81% Ti and 29.8% Ni. In conclusion, the LMD IN625/TC4 BS with Cu/V interlayers can be divided into four regions from the IN625 side to the TC4 one, whose phase compositions are γ-Ni + laves → (Ni, Cu)ss + (V, Cr)ss + TiNi → α-Ti + β-Ti + Ti 2 Ni → α-Ti + β-Ti. The microstructure of regions A and D is typical of the IN625 and TC4 alloys, respectively [4,22]. When the Cu/V interlayers are deposited onto the IN625 substrate, the dilution rate increases due to the high laser output power, which promotes the diffusion and migration of the Ni and Cr elements from the IN625 substrate to the interlayers as well as the diffusion of the Cu/V elements toward the substrate. As a result, a solid solution between Cu and Ni, as well as between V, Cr, and Ni, is formed. Due to the effect of the Cu/V elements on grain refinement, a finer grain structure is obtained in the transition region of the Cu/V interlayer. Once TC4 is deposited on the Cu/V interlayers, the latter is diluted to a certain extent, allowing the Ti elements to diffuse from region C to region B and causing the formation of Ti-Ni IMCs. At the same time, Ni elements also diffuse toward region C, forming Ti 2 Ni IMCs therein. The experimental results show that the formation of the Ti-Ni IMCs cannot be completely blocked by the deposition of a Cu layer and a V layer as interlayers. Thus, the interlayer thickness should be appropriately increased in future studies to further explore the microstructure and properties of the IN625/TC4 BS.    The difference in the Vickers hardness values of the two transition regions is mainly caused by their different microstructure morphologies, phase compositions, contents, and distributions. For example, with the introduction of the Cu/V interlayers, the (Ni, Cu)ss and (V, Cr)ss phases are first formed in region B near interface 1. The effect of the addition of the Cu and V elements (the atomic radii are 127.8 and 134 pm, respectively) varies from that of the matrix Ni and Cr elements (the atomic radii are 124 and 128 pm, respectively), resulting in the lattice distortion of the metal matrix. Additionally, the introduction of Cu/V slows down the diffusion of the matrix elements, resulting in solid solution strengthening. On the other hand, a fine equiaxed grain microstructure is present in this region, and the number of crystal boundaries is relatively large. In the process of vertical loading, the crystal boundaries hinder the dislocation movement, which results in the strengthening of the fine grains. In the upper part of region B, the contents of the Cu and V elements increase, resulting in an enhanced solid solution and fine crystal strengthening. At the same time, due to the increase in the Ti concentration, more Ti-Ni IMCs are precipitated from the matrix. These precipitated phases themselves have a high hardness and at the same time they prevent dislocation slip, thus contributing to precipitation strengthening and further increasing the hardness. As for region C, its lower part is mainly composed of the β-Ti and Ti 2 Ni phases, while its microstructure consists of quasi-equiaxed grains whose size is greater than that in region B. With the increase in Ti content, region C is mainly composed of α-Ti, β-Ti, and Ti 2 Ni phases with coarse equiaxed grains. Therefore, the Vickers hardness in this region is lower than that in region B but still exceeds the TC4 value.

Tensile properties of the IN625/TC4 BS
The room-temperature tensile test results conducted on the LMD IN625/TC4 BS with Cu/V interlayers are shown in figure 8. Figure 8(a) displays the stress-strain curve, figure 8(b) shows the fracture morphology, and figure 8(c) summarizes the EDS results taken at the points shown in figure 8(b). It can be seen from figure 8(a) that three specimens were evaluated in the tensile test. The room-temperature tensile strength of the BS is about