Laves Phase Formation and Segregation of Nb in Ni–Cr–Mo Superalloy over 316L by Hot Wire (HW) TIG Cladding Process

Abstract

The Ni-based alloys are widely used in power plant components such as boiler and piping systems that require good creep strength and resistance to corrosion. The present study focuses on metallurgical and mechanical properties of Ni–Cr–Mo (ERNiCrMo-3) alloy cladded over 316L plate by hot wire TIG process. This alloy is prone to microfissuring and also reduced corrosion resistance due to microstructural changes caused by weld solidification. The microstructure and phasic characterization were identified using scanning electron microscopy and X-ray diffraction techniques. From the microstructural observation, it was observed that Nb present in Ni–Cr–Mo alloy tends to segregate during solidification of weld thereby promoting the Nb-rich Laves phase formation. The growth of dendrites enhanced the segregation of Nb. The coarse dendritic microstructure was due to the effect of high heat input that decreases solidification rate near the interface zone. Dimples and quasi-cleavage fracture morphology was observed on the fractured samples of the tensile test. The resistance to corrosion was evaluated using potentiodynamic polarization method using 3.5 wt% NaCl solution. The results confirmed that Ni–Cr–Mo alloy acts as an efficient clad for 316L substrate from corrosion attack.

1 Introduction

With the advancement of science and technology, a gradual change had taken place from the concept of novel bulk materials to the processing of products that contain dissimilar metallurgical clads [1]. Because of its difficulty in meeting wide range of requirements such as microchemical, physical, mechanical, and thermal properties for a bulk material, clad metal constituting two or more materials bonded together had been developed to meet the unique requirements in their properties [2,3,4,5]. In the current scenario, there is an increase in demand for the material that can meet the combined requirements of metallurgical, corrosion, and mechanical aspects. As an alternative for bulk material, the Ni-based superalloy possessing high strength and good corrosion resistance can be cladded with 316L austenitic stainless steel substrate. This kind of clad was widely used in heat exchangers, power plant, boilers, reservoirs, and decorative trade. Variety of techniques are available in cladding of dissimilar materials, and some of them are diffusion bonding, cold metal arc transfer (CMT) cladding [6], laser cladding, roll bonding, hot wire TIG (HW TIG) metal deposition technique [7,8,9]. Nb present in the Inconel 625 assists in phase formation of δ (Ni₃Nb), γ (Ni₃Nb), MC carbides, and Laves ((Ni, Fe, Cr)₂(Nb, Mo and Ti)) phases. Hence, Nb segregation leads to the formation of Nb-rich precipitation known as Laves phase [10].

Generally, in traditional TIG cladding, geometric dilution and coarsening of microstructure are caused by the high heat input used during deposition process [11,12,13,14,15,16]. Many researches were conducted on cladding of Ni-based superalloy by TIG process. Abioye et al. in their study used Inconel 625 cladded over AISI 304 substrate using laser welding process. Generally, the microstructure of Ni–Cr–Mo deposited alloy consists of MC carbides, Laves phase in randomly oriented γ dendrites.

Related work done by Dinda et al. identified that clad region had cellular structure and columnar dendrites. Xu et al. observed that an unmixed area was observed nearer to the fusion zone with few Fe-rich precipitates that decreased hardness in this zone.

Verdi et al. observed that Cr₃C₂ precipitates in the Inconel 625 cladded region. These ceramic particles Cr₃C₂, being hard precipitates, improved the wear resistance of the clad portion [17,18,19,20]. The formation of γ phase during crystallization of Ni–Cr–Mo alloy during conventional TIG cladding results in the following phase change according to Dupont [21].

$$L \to \gamma \to L + \gamma \to L + \gamma + {\text{Laves}} \to \gamma + {\text{Laves}}$$

By using preheated filler wire, it is possible to produce continuous clad beads with low heat input with improvement in productivity. The addition of Nb to Ni–Cr–Mo increased the strength of the alloy system due to increase in solid saluting strengthening effect provided by the combined interaction of Ni and Mo that results in quasi-cleavage morphology after the bond test [22]. The γ″ precipitate Ni₃Nb had either body-centered tetragonal (BCT) or Ni₃Nb orthorhombic structure mainly due to the addition of Nb. In the present study, apart from the common microstructural features for the Ni-based alloy, a rare intermetallic cuboidal-shaped precipitate was observed that is not reported in the existing studies. The existing literature revealed that limited research had been conducted on dissimilar cladding of Ni–Cr–Mo alloy over 316L using hot wire tungsten inert gas (HW TIG) welding process. Hence, the present study focused on the microstructure, mechanical properties, and corrosion resistance correlation of hot wire TIG-deposited Ni–Cr–Mo (ERNiCrMo-3) alloy at various heat inputs.

2 Experiment

2.1 Experimental Setup

Type 316L stainless steel plate of dimension 500 × 100 × 10 mm was used as the substrate for deposition of Ni–Cr–Mo. The deposition was carried out using filler wire of diameter 1.2 mm. The chemical composition of 316L and Inconel 625 is listed in Table 1. The polishing was done using SiC grit papers of size 220, 400, 600, 800, 1000, and 1200 using standard metallographic techniques. Further acetone was used to clean the surface to remove oxidation and other contaminants.

Table 1 Chemical composition of ERNiCrMo-3 (Ni–Cr–Mo) filler wire and 316L materials

The preheated filler wire in HW TIG improved the process efficiency and appearance of the clad, compared to traditional TIG cladding process. The contact tube with diameter of 1.2 mm carrying filler wire was subjected to action of preheating was placed nearer to welding torch. Jet nozzle and tungsten electrode diameters were 12.5 mm and 3.2 mm, respectively. The ERNiCrMo-3 phase diagram is shown in Fig. 1. The preheated ERNiCrMo-3 (Ni–Cr–Mo) filler wire entered into the TIG arc region by wire feeding motion shown in Fig. 2. The selected process parameters are welding current (120–180 Amps), constant voltage (17 volts), torch angle (60°–80°), and travel speed (125–175 mm/min) as shown in Table 2 [23]. Commercially pure argon with a flow rate of 15 l/min was used as the shielding gas during deposition of filler wire.

Fig. 1

ERNiCrMo-3 phase diagram [24]

Fig. 2

Hot wire (HW) TIG set up

Table 2 Experimental parameters

The ERNiCrMo-3 phase diagram is shown in Fig. 1. The melting point was found to be 1369 °C. The Ni₃Ti, Ti₄C₂, M₃P, and MNS were the identified precipitates at equilibrium condition. At high temperature, Ti₄C₂S₂ was found to be stable. Ni₃Ti was found to be stable at low temperature, and it got dissolved at 650 °C and nucleation of Ti-rich precipitates again emerged at 950 °C.

Samples were named as T1, T2, and T3 samples and longitudinally cross-sectioned into dimensions of 500 × 10 × 10 mm using high- precision microwire cut EDM machine. The samples (T1, T2, and T3) were further sliced into dimensions of 20 × 10 × 10 mm using abrasive cutting machine, and final polishing was done by diamond paste. For 316L stainless steel, the chemical etchant used was aqua regia 3HCl/HNO₃ for 2 min, and Ni–Cr–Mo clad was electrolytically etched using 10% of ammonium persulphate solution at 3.5 V for 5 s. The microstructures were examined using LYNX model Stereozoom microscope (5 MP CMOS Camera) and an Image Capture Software was used for analysis. The composition and microstructure of each clad, interface and base area were investigated by scanning electron microscope (SEM). X-ray diffraction (XRD) was used to identify the phases present in the clad, fusion boundary zone, and substrate of samples T1, T2 and T3.

The hardness of clad, interface, and substrate region were measured using Vickers microhardness indentation machine as per ASTM E384 (Make: Wilson Hardness 402 MVD) specification (500gf of applied load and 10 s dwell time). Hardness evaluation across the transverse cross-section from the Ni–Cr–Mo region toward substrate region was done. Using Instron universal testing machine (UTM), the uniaxial tensile tests were performed on the T1, T2, and T3 as per ASTM: E8/8 M standard.

The measurements were taken on the sliced portion of clad, interface and base region each separately from three different samples (Table 2) and detailed geometric specification of the samples are shown in Fig. 3. The strain rate during testing was maintained at 1 × 10⁻³ as per the normal practice. The sliced samples parallel to the direction of cladding accounts to 1 mm thickness.

Fig. 3

Geometric specifications of tensile samples: a sample sliced into clad, interface and base region. b Dimension details of tensile specimen

Samples of 20 × 10 × 10 mm dimensions were taken for potentiodynamic polarization tests at a room temperature (24 °C) in 3.5%NaCl solution according to ASTM G61-86 specification using IVIUM electrochemical workstation. The counter electrode used is graphite, working electrode is sample and the reference electrode is saturated calomel electrode. For achieving OCP, initial delay of 10 s is taken. The corresponding Tafel plot was calculated between − 1v/SCE to + 1v/SCE for T1, T2, and T3 at a scan rate of 0.8 mv/s. The corrosion potential (E_corr), corrosion current density (I_corr) and polarization resistance (R_p) are calculated using the IVIUM soft electrochemistry software. The rate at which the corrosion takes place and corrosion current density are formulated by the following equation:

$${\text{Corrosion}}\;{\text{rate}} = 0.13 \times I_{\text{corr}} \times \frac{E.W}{A \times d}$$

(1)

The equivalent weight of 316L substrate is found to be of 25.50 g/eq and density is 8 g/cm³. Similarly, equivalent weight of Inconel 625 filler wire found to be of 25.573 g/eq and density is 8.44 g/cm³. Before start of corrosion tests, chemical composition of 316L base and Ni–Cr–Mo filler wire were noted. The obtained microhardness values were 235 ± 10 HV_0.5, 198 ± 10 HV_0.5, 185 ± 10 HV_0.5 and corrosion density comes around 8.73E−06 A/cm². The SEM analysis followed by EDAX is carried out on corroded surfaces.

3 Results and Discussion

3.1 Clad Bead Geometry

Figure 4 shows the appearance of typical clad bead deposited by HW TIG process. The samples T1, T2, and T3 with corresponding parameters used for cladding is listed in Table 3.

Fig. 4

Typical appearance of the clad bead

Table 3 Heat input given for sample T1, T2, and T3

3.1.1 Depth of Penetration

From the clad bead geometry of the samples (T1, T2, and T3), depth of penetration was measured from the center of clad portion to the substrate as shown in Fig. 5a–c. It has been inferred that the changes in the value of depth of penetration, measured using stereo zoom macroscope at a magnification of 20x, can be correlated with heat input used in the cladding of samples.

Fig. 5

Stereo macrographs of a T1, b T2, c T3

As welding current increases from 120 to 160 Amps, the heat input also increases from 0.699 to 1.305 kJ/mm. Heat input (H) in kJ/mm was calculated using the following relation [25]

$$H = \frac{{\left( {0.06 \times V \times I} \right)}}{S}$$

(2)

By varying the travel speed and welding current, there was change in the dimension of the bead and also in depth of penetration [26]. It is clearly seen from Fig. 5a–c, that the cladded sample T1 (Fig. 5a) with the lowest heat input (0.699) holds an inverse relationship with travel speed. The reason for the highest heat input for sample T3 (1.305) was due to increment in depth of penetration (800 μm). Therefore from the parameters considered, the sample “T1” had depth of penetration (0.683 mm). The welding process parameters are varied to have a gradual variation in the heat input to study its influence on microstructure and mechanical property characteristics.

Using optical microscope, the weld bead geometry measurements were made for various trails of welding current (I), travel speed (S), voltage (V), and torch angle (°). The results obtained from the experimental trials are shown in Table 4:

Table 4 Deposition process parameters and clad bead dimensions

(1) As the welding current increases, bead width and height of reinforcement also increase due to increase in deposition rate.

3.2 Microstructural Inferences

Porosity and defect-free cladding are obtained from Ni–Cr–Mo alloy for samples T1, T2, and T3 shown in Fig. 6. From the previous studies, it was observed that CMT and PTA claddings show continuous bands with free form was found. The HW TIG cladded sample shows (sample (T3): Fig. 5c), a clear HAZ up to minimum ~ 220 μm was noticed [27]. Due to high heat input, HW TIG cladding exhibits wide HAZ zone. Due to optimized parameters, the samples (T1, T2, and T3) were free from defects such as microfissuring and lack of fusion.

Figure 7a–e shows Ni–Cr–Mo alloy clad zone, interface and substrate for sample T3 (higher heat input, larger weld bead penetration). From the microstructural inferences, it was clear that the primary dendrites together with the columnar dendrites grow perpendicular to partially melted zone (PMZ) [28]. The solidification begins epitaxially from the bulk metal, which is parallel to heat extraction direction.

Fig. 6

Microstructures of clad, interface and substrate a T1, b T2 and c T3

Fig. 7

a–e Micrographs of Ni–Cr–Mo superalloy over 316L stainless steel

The hardness in the clad region comes around 235 ± 10 HV_0.5, and HAZ hardness comes around 198 ± 10 HV_0.5. Figure 14 shows the optimal pass of HW TIG cladding. From the observations, it was inferred that the thickness of the clad is five times higher than that of other clads prepared by laser cladding method [29]. The increased thickness of HAZ leads to weakening of metallurgical bond and fading of adhesion between clad and base metal, leading to intergranular microfissuring cracks at the fusion boundary zone [16].

During HW TIG cladding, chances for crack formation were possible on the sensitive zone between grain boundaries and primary dendrites. In HW TIG cladding, normally the thermal stresses are more than local tension stresses, and therefore, possibility for shrinkage cracks occurs during cooling of solid phase shown in Fig. 8 [30]. During HW TIG process, the filler rod and tungsten arc gets melted and heat input gets developed, and thus the entire base material is heated, and the calculated power density comes around 105 W/cm^2, which is less than that of laser cladding. But when it comes to the heat input, HW TIG offers high heat input for T3 (1.305 kJ/mm), this has an influence on mechanical properties.

Fig. 8

SEM image of sample T3 with HAZ zone

The clad layer height was found to be 1769 ± 68 μm from the optical microscopy measurements. Visual defects were not observed T1, T2, and T3 samples after penetrant examination. Solidification cracks were observed near the fusion boundary zone, i.e., at the HAZ region of the substrate. A typical solidification crack of 400 μm length is noticed from Fig. 9. During cladding, liquated phase is formed as a result of low melting compounds on the grain boundaries of HAZ. The formation of ϒ′ precipitates and development of tension stresses during cooling may be the reason for the development of hot cracks.

Fig. 9

Intergranular cracks on sample T3

According to solidification theory, the ratio of the temperature gradient (G) to growth rate (R) influences the solidified microstructure of the deposit. The formation of columnar dendrites occurs at a higher G/R ratio, whereas equiaxed growth occurs at a lower G/R ratio [31]. From Gaumann et al. studies on laser cladding, it was observed that the transition between columnar to equiaxed was observed at a ratio of G^3.4/V, where V denotes solidification velocity at the front side of columnar dendrite. The evolution of columnar dendrite was derived from the following relationship:

$$\frac{{G^{3.4} }}{V} > 2.7 \times 1024\left( { \frac{{G^{3.4} }}{{(m^{4.4} \times s)}}} \right)$$

(3)

It was inferred from Ni–Cr–Mo alloy that the growth mode of dendrites exhibits a transition from columnar to equiaxed structure, by increasing the cooling rate for a static temperature gradient.

This dendrites growth accelerates the segregation of alloy such as Nb and also beginning of the laves phase during solidification. The main reason for the occurrence of Laves phase is the presence of columnar dendrite. This in turn relies on the phase transformation due to the segregation of Nb, which is crucial for the occurrence of Laves phase under the growth of dendrite mode in Fig. 10. It has been revealed that microstructural changes can be controlled, through solidification condition [32].

Fig. 10

SEM micrographs revealing segregation of Nb in sample T3

From EDS spectra, the presence of elements like Fe, Cr, Ni, Mo, Si, Al were identified (Fig. 11). The element Nb enriched in the interdentritic regions are known as Laves phase amounts to about 10.98%, whereas Nb contained in the austenitic matrix, as γ phase contains only 0.08%.

Fig. 11

SEM image along with EDS spectra showing segregation of Nb in sample T3

The average amount of Nb content present in the chemical composition accounts to about 3.4%, and thus possibility for segregation of Nb in Ni–Cr–Mo alloy is examined. From Dupont et al. studies, it was revealed that during the solidification time, the phase transformation can be controlled by Nb segregation [33]. The high percentage of Nb present in the interdendritic zone leads to Nb-rich Laves phase formation. ϒ-Ni phase which is the prime phase of Ni–Cr–Mo alloy is identified from the results of XRD. Due to the overlapping of MC carbides over ϒ, ϒ′, ϒ″ phases, only two phases ϒ′, ϒ″ can be detected by XRD in Fig. 12.

Fig. 12

XRD curve for samples T1, T2, and T3

From Fig. 13a, the dark gray regions (B) and light gray regions (A) are enclosed by dark gray regions. The region (B) is found to be rich in Ni, Cr, Nb and Mo whereas light gray region (A) is found to be rich in Nb, Ni, Ti, Cr and Mo. Figure 13b shows dendrite (E) like structure rich in Nb, Ni, Cr.

Fig. 13

a, b SEM micrographs showing elemental composition and Nb segregation

Taking EDS results into consideration, dark gray region (B) holds 58 wt% Ni, 19.3 wt% of Cr, 10.98 wt% Nb, 9.8 wt% Mo, 17.3 wt% Cr, and elements like W, Fe, and Ti are also included. The ϒ phase continuous in structure made up of non magnetic Ni-based alloy having FCC structure containing Fe, Cr, Mo, and W. Hence ϒ phase is considered as a prime phase for Ni–Cr–Mo alloy. The ϒ phase present in (B) region is the reason behind appearance of blade like structure in region (E) as weight percent of elements in (E) are close to (B) shown in Fig. 13b. Significance of carbide phase and its corresponding segregation are identified through region (C). Similarly in (A) 47.2 wt% Ni, 13.6 wt% Mo, 17.5 wt% Cr, and 12.8 wt% Nb were observed. Due to the continuous solidification action of Ni–Cr–Mo alloy, dendritic phase ϒ appears, whereas Nb tends to segregate and at a particular period slightly deviate away from the center. Elements like Fe, Ni and Cr migrate to ϒ dendrites. Thus at the end of solidification, the Nb element increases and forms the γ/Laves phase. Hence, ϒ + Laves + NbC is formed [34].

3.3 Microhardness

Vickers microhardness (500 gf of applied load and dwell time of 10 s) test was used to measure the hardness value of samples T1, T2 and T3 respectively. Generally, for PTA, GMAW there exists a slight decrease in hardness value nearer to the fusion boundary zone, but in case of HW TIG, the drop is lesser due to the presence of HAZ zone. Average hardness value of the clad, interface, and base corresponding to T1, T2, T3 are 235 ± 25 HV_0.5, 198 ± 10 HV_0.5, 185 ± 20 HV_0.5 respectively shown in Fig. 14.

Fig. 14

Microhardness of samples T1, T2 and T3

In specific, sample T3 exhibits reduction in hardness, due to its high heat input (1.305 kJ/mm) and also due to the coarser dendrites. Also it has been inferred that increase of welding current, increases the microhardness value. When it comes to FBZ, higher heat input allows better fluidity of molten metal. Therefore, interface region exhibits a cliff in the hardness value for the samples [35].

In specific, sample T3 exhibits reduction in hardness, due its high heat input (1.305 kJ/mm) and also due to the coarser dendrites. Also it has been inferred that increase in welding current increases the microhardness value. When it comes to FBZ, higher heat input allows better fluidity of molten metal. Therefore, interface region exhibits a cliff in the hardness value for the samples [35].

The Ni–Cr–Mo alloy region has higher hardness with the range between 200 and 250 HV_0.5. Hall–Petch relationship states that, as grain size increases, hardness value decreases at room temperature. In Ni–Cr–Mo superalloy, the presence of elements like Nb and Mo increases the hardness. From the SEM observations, the segregation of Nb which was the reason behind the formation of Laves phase attributes to the increased hardness in the clad layer. In addition, carbide precipitation also contributes to the variation in hardness value.

3.4 Tensile Test

The stress v_s strain curve of the samples T1, T2 and T3 is plotted in Fig. 15. The ultimate tensile strength, yield strength, and percentage of elongation are evaluated from tensile curve and the results are listed in Table 5. The Ni–Cr–Mo clad zone exhibits higher tensile strength (919 MPa), followed by the interface zone (810 MPa) and least strength is exhibited by the base material (698 MPa). Considering the elongation percentage into account, the clad zone reveals lower elongation of (32.9%) than the base metal (66.1%). The base metal exhibits a steep decrease in the tensile strength when compared to interface and clad zone. In specific, the clad zone exhibiting a tensile strength of (919 MPa) reveals good mechanical properties. This ensures that the bonding strength between the substrate and clad are excellent. The unmixed area possessing inadequate hardness does not affect the total mechanical property [20].

Fig. 15

Tensile curve for the sample T1, T2, and T3

Table 5 Tensile results of the clad, interface, and substrate for T1, T2, and T3 samples

The formation of precipitated compounds and Laves phases were identified around the dimple structure shown in Fig. 16a–c. The metal carbides and Laves phases constitute the precipitated phases, which are responsible for the fracture. The Laves phase plays a key role in the formation of main nucleation points for the construction of microscopic holes during tensile testing. As the microvoids initiate and cracks lends to elongate on the direction of interface, the Laves phase gets weakened.

Fig. 16

Fractographic observation of dimples, tear ridges, fatigue striation shown a T1, b T2 and c T3

From Table 5, C denotes Clad portion, I denotes Interface portion and B denotes Base portion.

3.5 Fractographic Observation

The fractographic behavior of Ni–Cr–Mo alloy was examined under SEM. The sample T1, T2, and T3 show ductile mode of fracture. The elongated linear fracture initiation exhibiting columnar dendrites reveal Mo- and Nb-rich precipitates, along the grain boundaries shown in Fig. 15b.

Generally, fracture takes place through ductile and brittle mode. The fracture characterized as ductile exhibits fibrous pattern with refined dimples. Similarly, fracture also begins with the formation of micro voids extending to the initiation of cracks. From Fig. 15a, tear ridges with river markings is revealed. The brittle mode of fracture is characterized by cleavage fracture. Usually, the presence of carbides and intermetallic precipitates serve as the source of crack growth. When it propagates the grains and the crystals results in series of plateaus and thus river markings occur [33].

From Fig. 16a–c, the tensile surfaces reflect transgranular behavior indicated by deep, fine, and uniform equiaxed dimples within which the Ni₃Nb particles were randomly oriented. The Nb-rich carbides were observed, when rapid cooling takes place during cladding, it suppresses the γ/NbC eutectic action. The Nb-rich carbides appear darker due to the presence of heavier elements compared to γ elements. It is noteworthy that Nb tends to segregate in order to form Nb-rich compounds which are present on the grain boundaries or interdendritic regions during the process of solidification.

The Ni₃Nb needle-like platelet located on the underneath of the dimples is the reason for the initiation of crack growth and as the result, decohesion occurs on the interface region shown in Fig. 17. The sufficient amount of Nb present around the Laves phase becomes a favorable site for nucleation to occur leading to the formation of needle like structure.

Fig. 17

Initiation of crack along the grain boundary

3.6 Potentiodynamic Polarization Test on the Cladded Surface

The corrosion rate (C_Rate), corrosion potential (E_corr) and corrosion current density (I_corr) are evaluated from the potentiodynamic polarization tests. The solution used was 3.5% NaCl solution. Table 6 lists corrosion potential (E_corr), corrosion current density (I_corr) and corrosion rate (C_Rate) of samples. Figure 18 exhibits the nature of the potentiodynamic polarization curve of Ni–Cr–Mo alloy, fusion boundary zone, and 316L base area in 3.5% NaCl solution. Taking the samples T1, T2, and T3 into account T3 offers high corrosion rate compared to the other two. The difference between the base, interface and clad comes round 60 mV.

Table 6 Comparison of corrosion parameters and their corresponding results

Fig. 18

Potentiodynamic Polarization curves for samples (T1, T2 and T3)

From Table 6, comparison of corrosion parameters and their corresponding results are measured. Compared with the T1 and T2 sample, T3 offers higher corrosion rate. The Cl-rich surrounding offers pitting corrosion, from Fig. 17 of corrosion curve, the sample T1 and T3 contains pitting corrosion compared to the two, no pitting corrosion is found for sample T2. Current density increases with no change in potential and is only possible after narrow passive density. Thus T1 and T3 are less resistant to crevice corrosion than T2, though having slight difference in corrosion potential.

The clad region has dendritic microstructure. From Fig. 19a, b, the subgrain boundaries were seen as bright portion, the columnar dendritic portion is revealed as dark portion. The reason for the bright portion in subgrain boundary is mainly due to the presence of NbC type precipitates. For sample T3, the surface gets corroded, the boundary becomes visible and also indicated through valley like structure in and around the subgrain boundary.

Fig. 19

SEM micrograph revealing the grain structure of a T2 and b T3 taken before corrosion test

The pits are formed around NbC type precipitates in the corroded region as observed in Fig. 20b. The corrosion action originates from the grain boundary region and growth takes place toward the interior of the grain. In sample T3, the oxides are noticed at the subgrain boundaries.

Fig. 20

SEM micrograph revealing a T2 and b T3 after corrosion tests

Sample T3 shows that corrosion occurred and it is revealed as bright spots around the grain boundary. Nb and Mo are seen along the subgrain boundary, NbC carbides precipitation is commonly observed. In contrast, sample T2 is free from corrosion attack. It is confirmed through the presence of oxides precipitated on the grain boundary. For sample T3, composition of elements like Ni and Cr was similar to those present in the transgranular region from Fig. 21.

Fig. 21

Mo and Nb get precipitated on the subgrain boundaries in sample T3

Nb content showed higher percentage, and it was accounted to be six times higher at the grain boundary than at the surrounding. Confirmation for the elements at the grain boundary was revealed in elemental mapping (Fig. 22).

Fig. 22

SEM EDS mapping of corroded area of sample T2

In case of sample T2, Ni content was very less at the grain boundary, thus the region appeared dark. Due to the increased amount of oxygen present at the surface, corrosion occurred. The carbon transmission from the substrate, gets diluted, and mixed with Cr forming CrC at the subgrain boundary. Beforehand percentage amount of Cr was slightly less at the grain boundary, and thus Cr-deficient zone was formed. The passive layer of Cr oxidation is minimal; Ni reacts with O resulting in corrosion. Important inference is that appearance of long rod like shaped secondary phase composed of Laves phase are noticed in the sample T3. The dilution of Nb leading to the formation of Laves phase gets precipitated along the grain boundaries. The precipitates are cuboidal precipitates which are found to be rich in N, Ti and niobium shown in Fig. 23.

Fig. 23

Sample T3 containing Cuboidal precipitated rich in Nb, Ti, N

4 Conclusion

By using hot wire TIG (HW-TIG) process, defect and pores free clads were produced. Among the samples (T1, T2, and T3), sample ‘T3’ had high depth of penetration (Heat input: 1.305 kJ/mm) of 1.113 mm whereas the sample ‘T1’ had lowest depth of penetration (Heat input: 0.699 kJ/mm) of 0.683 mm. The clad region had primary dendrites in γ matrix with secondary precipitates were distributed at the columnar and cellular dendrites at the region closer to the fusion boundary zone. Further, the Ni–Cr–Mo clad showed a cellular structure with refined grains in fusion boundary zone. Niobium segregation at the grain boundaries formed the Laves phase. Average hardness value of the clad, interface, and substrate is 235 ± 25 HV_0.5, 198 ± 10 HV_0.5, 185 ± 20 HV_0.5. The fracture surfaces exhibited transgranular fracture mode showing deep, fine and uniform equiaxed dimples. From the potentiodynamic polarization test, the sample T3 exhibited highest corrosion rate of 3.832 mm/yr. This was due to enhanced segregation of Nb and Mo elements at the subgrain boundaries of matrix.