3.1. Vanadium deposition on titanium alloy
Figure 3a shows the panorama of the first three vanadium beads deposited on the titanium alloy. The resulting sample is characterized by a visible difference in etchability between the deposited beads, which may indicate different alloying of vanadium with titanium. It can be seen from the Fig. 3b distribution (Fig. 3b) that the hardness distribution in the vanadium layer is quite stable and ranges from 230–260 HV1. Such a range exceeds the hardness of pure vanadium, which also indicates solid-soluble hardening by titanium with varying concentrations. An increase in hardness up to 175 HV1 is observed on the titanium side near the fusion line, while the hardness of the base metal ranges from 120–125 HV1. This can be described by the formation of quenching structures in the heat affected zone.
When studying the chemical composition of the deposited vanadium layer using electron microscopy, the presence of a significant amount of titanium in the deposited beads was confirmed. This may be dangerous for further deposition of iron and nickel alloys. Vanadium has a higher melting point, and therefore significant melting of the titanium substrate is inevitable during deposition.
The fusion zone of titanium with a vanadium bead contains an interlayer with a higher etchability (Fig. 4a). Above the layer on the vanadium side there is a zone with etched grain boundaries which look like cracks, but they are not detected on non-etched microspecimen. The hardness of this zone is at a fairly high level, but is comparable to the hardness of the first vanadium bead. However, the interlayer itself has a low hardness in relation to the studied area and is about 260 HV0.05.
However, vanadium is a β-stabilizer for titanium and contributes to the formation of the ω-phase inside the β-phase in the transition interlayer at high concentrations.
Figure 4b displays a SEM-image in the fusion zone of titanium with the vanadium layer. The structure of the interlayer is characterized by elongated thin crystals with a characteristic liquation of one element in the interdendritic space. Probably, the interlayer, which is clearly distinguishable after etching on an optical microscope, is α’-martensite. While the interlayer with low etchability and slightly more vanadium content, is a β-phase with inclusions of a solid ω-phase. The presence of last has a negative effect on the strength characteristics of the compound due to a sharp decrease in plasticity. The most potentially dangerous ω-phase formation area characterized by vanadium concentration from 15 to 18% wt.
Figures 5a-c demonstrate the corresponding phase diagrams. Figure 5d reveals the dependence of the maximum vanadium content, at which the σ-phase presence has not yet been observed and the titanium content in the alloy. It can be seen from the calculation results that the maximum concentration of vanadium, at which the formation of a σ-phase is possible, decreases with an increase in the concentration of titanium in the alloy. Thus, to reduce the probability of the σ-phase formation at the fusion line of iron and vanadium, it is necessary to ensure a minimum titanium content in the last vanadium bead. This can be achieved by increasing the number of vanadium beads or reducing the penetration rate.
The obtained distributions of titanium and vanadium are presented in Fig. 6. Local zones with a low titanium content are visible and can be explained by insufficient mixing of the vanadium wire during welding It was done minimize overheating of the welding pool and reduce the penetration rate of titanium. In 4th bead, the titanium content is reduced to 2%, in 5th and 6th to 1% and < 0.5%, respectively.
To conclude, vanadium as an interlayer deposited on titanium makes it possible to provide a compound with a stable structure without sudden change in hardness and with acceptable manufacturability, despite a higher melting point. However, titanium dissolved from the substrate in the vanadium layer negatively affects the deposition of steels, nickel or iron-nickel alloy. In this work, the titanium content was reduced to 0.5% by deposition of 4 beads.
3.2. Sv-08G2S steel wire deposition on vanadium layer
Figure 7a presents a panorama of the deposited low-carbon wire directly on the vanadium. To prevent a contact of titanium with iron, 7 vanadium beads were deposited. Thus, an intermetallic layer was not form at the boundary of the iron and vanadium fusion.
Figure 7b indicates the microstructure of the iron-vanadium fusion zone. As can be inferred from the Fig. 7b, a layer of about 15–20 microns thick is formed on the fusion line. There is a consumption that it consists of a brittle σ-phase. However, due to the small thickness, destruction and even defects are not detected in this area. The hardness of this interlayer reaches a maximum value of 378 HV0.01.
In this way, deposition of Sv-08G2S steel wire onto a vanadium layer with a titanium content of less than 0.5% made it possible to obtain a structure of the interlayer without sudden change in hardness.
3.3. Nickel deposition on vanadium layer
A panorama of the deposited nickel on the vanadium layer is shown in Fig. 8a. As can be seen from the figure, an area of increased etchability is observed in the transition region between the two layers. The microstructure of the first bead consists of large grains formed by columnar crystallites (Fig. 8c). The upper layer of deposited nickel has a coarse-grained structure (Fig. 8b) with a tendency to decrease the etchability of grain boundaries as it moves away from the vanadium layer.
Figure 8d represents the microstructure of the interlayer between vanadium and nickel without etching. It is obvious that there are a large number of cracks in the light gray area near the fusion line along the width of the entire joint, as well as in the area of large crystals on the nickel side. The occurrence of such a defect is due to the high content of titanium in the last vanadium bead, which is metallurgically incompatible with nickel and forms brittle intermetallic compounds of various types.
Figure 9 displays the hardness distribution in the resulting deposited nickel layer. At the fusion line with vanadium, the nickel layer has an abrupt increase in hardness to a value of 480.9 HV1, which confirms the presence of a solid intermetallic layer in this area. The maximum microhardness of 910 HV0.01 is achieved near the vanadium fusion line. A further decrease in hardness as it moves away from the fusion line with vanadium is accompanied by a decrease in the concentration of vanadium in nickel, which forms dispersed strengthening intermetallides and strengthens a solid solution. The average hardness of the last deposited nickel bead is about 120 HV1.
Figure 10 illustrates electron images of the vanadium-nickel interlayer at various magnifications. The layer under consideration is characterized by a cellular-dendritic structure with a eutectic mixture along the boundaries of crystallites. The average chemical composition in this area is 56.49% nickel, 39.31% vanadium and 3.98% titanium, and the chemical composition of individual zones is shown in Fig. 10b. Based on the chemical composition, it can be assumed that intermetallic phases of alloyed Ni2V type (spectrum 3 and 4), triple intermetallic Ni-V-Ti system (spectrum 1) and their mechanical mixtures along the boundaries of crystallites (spectrum 2) are formed in the interlayer. Taking into account the nonequilibrium solidification conditions and presence of impurity elements, other phases may be contained in the interlayer. According to the Ti-V-Ni triple diagram, titanium narrows the solubility of vanadium in nickel, which apparently leads to an increase in the width of the brittle layer, the appearance of triple intermetallic phases and an increment of cracking. In addition, titanium is found even in the 4th layer of nickel (0.65% wt.), which indicates the active dissolution and diffusion of titanium in nickel. On the one hand, it can lead to undesirable interaction of titanium with iron, and on the other hand, cause the hardening of the deposited beads.
The distribution of vanadium and titanium in deposited nickel beads is detected in Fig. 11. It is worth noting that the relatively high content of vanadium and titanium is registered even in the 1st layer of nickel (21% V and 2.4% Ti). This clearly leads to the formation of brittle compounds and a growth in the probability of cracking in the fusion zone. Therefore, it is necessary to reduce the concentration of titanium in the vanadium layer on which the nickel layer is deposited. Decreasing the concentration of vanadium in the nickel layer turned out to be technically difficult to implement due to the high solubility of vanadium in nickel and active diffusion.
Figure 12a reveals a panorama of the deposited nickel layer on 8 beads of vanadium. In this case, the 1st deposited bead of nickel has a higher etchability than the subsequent ones, which may indicate a high content of vanadium in it. It is also worth mentioned that at the vanadium-nickel fusion zone, there is a layer with a reduced etchability relative to the 1st nickel bead.
The results of microhardness measuring in the interlayer are provided in Fig. 12b. The high hardness of this zone can be explained by the formation of double nickel and vanadium intermetallides, concerning the 1st bead by the release of strengthening phases of the Ni3V type. A sharp decrease in hardness in the 1st bead as it removes from the fusion line can be explained by the ingress of the indenter into a vanadium-depleted structure formed due to incomplete mixing of the filler material under conditions of minimal overheating.
The 1st and 2nd nickel beads are clearly distinguishable because of their different etchabilities (Fig. 12а). This can be explained by the fact that the 2nd vanadium bead is fully in solid solution, and a single-phase structure has been formed.
Obviously, the deposition of more vanadium beads makes it possible to isolate titanium from nickel in order to avoid the formation of cracks. However, a decrease in the width of the fusion zone and a decline in its hardness could not be achieved.
3.4. Iron-nickel alloy deposition on vanadium layer
Figure 13a shows a panorama of the resulting deposited layer of iron-nickel alloy MSG NiFe-1 wire to vanadium. Similarly to the previous cases, the first deposited bead has increased etchability in relation to the other beads.
In addition, the sample under consideration also contains the interlayer between deposited materials with a thickness of about 150 microns (Fig. 13b,c). Figure 14 provided the results of measuring microhardness in this zone. As in the case of pure nickel, the hardness of the interlayer reaches 1000 HV0.01 from the side of the vanadium fusion line. On the side of deposited layer of MSG NiFe-1 wire, the hardness of the light grains is about 347 HV0.01. Moreover, between these areas there is a mechanical mixture zone of these two phases with a hardness of 841–926 HV0.01. It is also significant that cracks present in the interlayer with a length of 100 microns and a width of 1.5 microns. This fact confirms the fragility of this layer.
3.5. 316L steel wire deposition on vanadium layer
Figure 18a represents panorama of the 316L austenitic steel layer deposited on vanadium. This compound broke down during the deposition process of the 5th bead. It may be caused by an increase in the level of mechanical stresses during the cooling process and the release of the σ-phase under the action of the thermal cycle from the deposition of subsequent beads. It can be seen from Fig. 15b that the destruction occurred along the fusion line. It is of high importance that from the side of austenitic steel there is the interlayer of reduced etchability with a thickness of about 30 microns with typical cracks and some spalling (Fig. 15b). On the vanadium side, the presence of metal rupture parallel to the fusion line is noted (Fig. 15с).
The microhardness (Fig. 15d) of the studied austenitic steel interlayer is near 1301 HV0.01, which confirms the brittle nature of the compound fracture. At the same time, the average hardness of the steel bead ranges from 305–390 HV0.01. Apparently, an area with a high proportion of σ-phase could form on the fusion line. The content of the last increased during the deposition of the following beads due to the high heating temperature. This is primarily related to the high chromium content in 316L steel, which can contribute to the formation of the σ-phase at lower vanadium concentrations.
3.6. Tensile test results
Figure 16 shows the tensile tests results of each earlier studied compounds. It can be seen that all the specimens destroyed in the elastic section of the stress-strain curve, which may indicate the fragile nature of the destruction. The greatest ultimate tensile strength is achieved in specimen V–MSG NiFe-1 (1), which is a compound of titanium and steel through interlayers of vanadium and iron-nickel alloy, and constitutes 251 MPa. The minimum ultimate tensile strength is provided by the compound through the low-carbon steel deposited on vanadium and reaches only 182 MPa. The most stable results of ultimate tensile strength are provided by samples obtained through interlayers of vanadium and nickel NP1.
Figure 16 also demonstrates panoramas of the destroyed specimens. Most of the compounds broke down along the vanadium fusion boundary with a subsequent layer of low–carbon steel, nickel and iron-nickel alloy. However, the specimen V–NP1 (2) destroyed along the Ti-V interlayer, which apparently contains a brittle ω-phase. Brittle cracks were found in the interlayer between titanium and vanadium on all specimen after destruction, parallel to fusion line. These cracks appear in the zone where the dispersed ω-phase exists. The formation of this phase can be prevented by reducing the width of the transition zone with an undesirable concentration of vanadium or by alloying titanium with aluminium.
Abrupt changes in residual stresses caused by the difference in coefficients of linear thermal expansion of the materials contribute to destruction along fusion lines in the elastic region. For example, this coefficient for titanium, vanadium, iron and nickel is 8.41, 8.3, 11.8 and 13.8 µm/m·K [41], respectively, while for 316L steel is 16 µm/m·K [42]. The data above indicates that the difference in linear expansion between titanium and vanadium is minimal, but destruction in this area could occur due to the formation of brittle ω-phase. At the same time, the combination of vanadium and 316L steel has the greatest difference in thermal expansion, which leads to the formation of high residual stresses during the cooling stage of deposition process. Probably all these facts together with the formation of σ-phase are the reasons of sample destruction during the deposition of subsequent beads. Simultaneously, deposition of iron, nickel and iron-nickel alloys on vanadium does not cause the destruction, but the difference in physical properties can lead to the formation of significant residual stresses. Their occurrence in the possible high hardness interlayer leads to brittle fracture along the fusion zone during tensile tests.