Experimental investigation on microstructural characterization and mechanical properties of plasma arc welded Inconel 617 plates

1 Introduction

Inconel 617, solid solution nickel-based alloy, has been extensively used in the elevated temperature applications due to its tremendous oxidation resistance, better mechanical properties and phase stability at higher temperature [1]. This alloy is mainly used in aircraft and land-based gas turbine engines for combustor, transition ducting and exhaust system components [2]. Super alloy 617 is also used for the manufacture of machinery to operate the Next Generation Nuclear Plant service temperatures from 650 to 1,000°C [3]. Inconel 617 maintains its tensile properties with slight change, at a wide range of temperatures (77–1,093 K) in various environments like salt/vacuum exposures for long periods [4]. Nickel-based super alloys are considered to be very difficult to weld and repair because of their vulnerability to heat-affected zone (HAZ) and weld metal cracking during fabrication, post-weld heat treatment (PWHT) and subsequent operation [5]. Janaki Ram et al. [6] conducted experiments on 2 mm thick Inconel 718 using electron beam welding. They reported microstructures of bead on plate (BoP) welding, high temperature tensile properties and stress rupture properties. Shah Hosseini et al. [7] had welded the Inconel 617 and SS 310 materials by using gas tungsten arc welding. Furthermore, they investigated the zonular mechanical properties of weldment by performing shear punch test. Henderson et al. [8] described the characteristic defects observed in the welding of Ni-based super alloys. They also analysed the weldability of nickel-based super alloys using gas tungsten arc, electron beam, laser welding and friction or inertia bonding. Richards and Chaturvedi [9] presented the effect of minor elements like C, B, S, P and others on the weldability of nickel-based super alloys. Moreover, Fontana et al. [10] described the high-power CO₂ and Nd:YAG laser welding of super alloys for manufacturing of aero engines and power plant components. They observed solidified microstructures in the fusion zone and the micro-fissures in the HAZ of Inconel 718. Furthermore, they also suggested the laser beam welding for the welding of super alloys rather than tungsten inert gas welding and electron beam welding. Shinozaki et al. [11] studied the weld cracking phenomenon by conducting experiments on 2 mm thick Inconel 718 plates using laser beam welding. Later on, Ren et al. [12] compared the performance of fibre laser welding and CO₂ laser welding of Inconel 617. They reported that the improvement in the hardness values in HAZ was due to the increased heat inputs and smaller values of secondary dendritic spacing. In addition to the aforementioned studies, Kim et al. [13] also studied the effect of PWHT of gas tungsten arc welded Inconel 617 and 263, and observed a change in the microstructure and mechanical properties. Jalilian et al. [14] investigated the effect of process parameters on Inconel 617 at transient liquid phase on micro-structural characteristics. They used gas tungsten arc welding to join 1 mm thick Inconel 617 sheets for the experimentation. Liu et al. [15] performed simulation of the HAZ using five types of thermal cycles to study the grain size, micro-hardness and ultimate tensile strength of the HAZ. They concluded that the grain size and micro-hardness values of simulated HAZ were equal to the base metal, but the ultimate tensile strength (UTS) decreased slightly at elevated temperatures. McCoy [16] examined the mechanical properties of gas tungsten arc-welded Inconel 617 and compared them with the mechanical properties of base metal. It was observed that the fracture toughness was responsible for the reduction in the ductility of Inconel 617 weldments. Later on, Pavan et al. [17] conducted experiments to join the Inconel 617 and SUS 304H dissimilar metals by utilizing gas tungsten arc welding. They reported that the fusion zone of the weldment was enriched with Cr, Mo, C and Si in its inter-dendritic regions. Furthermore, it leads to the increase in hardness of the fusion zone when compared with the base metal, i.e. Inconel 617. Fink and Zinke [18] performed butt welding of Inconel 617 plates with a modified dip arc welding process. They observed a crack phenomenon in the fusion zone and recommended the dip welding process as an alternative for pulsed gas metal arc welding. Adamiec and Kocurek [19] investigated the microstructural and mechanical properties of the Inconel 617 alloy using autogenous laser welding. They studied the weld penetration and hardness values of the welded zone for different powers of the laser beam.

From the literature it has been observed that many of the researchers had attempted to characterize the weld zone and its effect on the mechanical properties of Ni-based super alloys. Particularly, they focussed on the gas tungsten arc welding, electron beam welding, laser beam welding, CO₂ gas welding and modified deep arc welding methods. To the best of the authors’ knowledge, no work is reported on the melt-in-mode plasma arc welding (PAW) of Inconel 617. In this article, an attempt has been made to analyse the effect of PAW on 2 mm thick Inconel 617 plates. Weld bead microstructure and grain size were examined for different process parameters like welding current and welding speed. Furthermore, the variation of mechanical properties, namely ultimate tensile strength, yield strength, micro-hardness and flexural strength, was also studied.

2 Experimental details

Super alloy Inconel 617 plates of dimensions 150 mm × 100 mm × 2 mm were used to perform welding by using melt-in-mode PAW. The chemical composition of Inconel 617 is shown in Figure 1. Before performing BoP trials, the plates of Inconel 617 were cleaned with the help of a wire brush and acetone solution to remove the layers of oxidation.

Figure 1

Chemical composition of alloying elements in Inconel 617.

The PAW experiments were conducted by using Fronius Magic wave 4000 power source (Figure 2), which consists of two modules namely main module and plasma module.

Figure 2

Schematic diagram showing the experimental setup used for PAW.

The main module was used to set the welding current and voltage. The flow rate of shielding gas was kept fixed at the suitable level. The plasma module was used to set the plasma gas flow rate and current for the pilot arc generation. From the extensive literature study, it has been observed that the welding current and welding speed were found to have significant contribution on the quality of the welds produced using PAW [20]. Furthermore, the torch standoff distance was maintained constant and its value was kept at 8 mm. The movement of the torch and the welding speed were controlled with the help of a CNC X–Y slide. The suitable range of welding current and welding speed was determined by conducting a pilot study. The trial experiments were conducted by performing BoP trails as shown in Table 1. During pilot study, the welding current and welding speed were varied in the range of 70–110 amps and 250–300 mm·min⁻¹, respectively. Several BoPs were made with different combinations of welding current and welding speed. However, only a few extreme cases are shown in Table 1 to visualize the impact of setting the input process parameters at the extreme conditions.

Table 1

Weld bead surface obtained at various combinations of inputs

S. no.	Welding current (amp)	Welding speed (mm·min−1)	Weld bead surface
1	70	250
2	92.5	275
3	110	300
4	90	250
5	110	250

From Table 1, it is observed that either low current or high welding speed could not produce the desired weld penetration. On the other hand, high current or low welding speed resulted in increased convexity of the weld, which is not desirable. Therefore, in the present study to perform the butt welds, the welding current and welding speed are kept at the range given in Table 2. The plasma gas and shielding gas flow rates are kept fixed at 2.5 and 12 lpm, respectively. Furthermore, the torch standoff distance was kept fixed at 8 mm and 2% thoriated tungsten rod with diameter 1 mm was used as electrode.

A total of nine butt welded joints of Inconel 617 alloy were prepared by varying the said parameters at three levels as shown in Table 3. After the joints were made, the transverse section of the weld was cut by using wire cut EDM (BAOMA, BMW-3000). To obtain the macro-graph of the welded structure, the sample was polished by using various grades of emery papers. In order to view the welded micro-structure clearly, the solution of 90% HCL + 10% H₂O₂ was used as etchant medium. Conventional optical microscope was used to capture the macro-graphs. A Welding Expert System (Struers, Austria) with a magnification range of 20–240× was used to capture the macro-graphs. The welded surface of the butt welds and their macro-graphs are shown in Table 3. In Table 3, the symbols I, N, V and HI represent welding current, welding speed, welding voltage and heat input to the welding process, respectively. Further BW and BH indicate bead width and bead height, respectively.

Table 3

Weld bead surface and the corresponding macro-graphs of butt welds

Sample no.	Input parameters	Face side and root side of butt joint weld	Macro-graphs	Bead parameters (mm)
1	I = 80 amp; N = 250 mm·min−1			BW = 5.638
	V = 17.6 V; HI = 203 J·mm−1			BH = 2.287
2	I = 80 amp; N = 275 mm·min−1			BW = 5.613
	V = 17.5 V; HI = 183 J·mm−1			BH = 1.492
3	I = 80 amp; N = 300 mm·min−1			BW = 5.739
	V = 18.2 V; HI = 175 J·mm−1			BH = 2.379
4	I = 92.5 amp; N = 250 mm·min−1			BW = 6.573
	V = 17.5 V; HI = 233 J·mm−1			BH = 3.393
5	I = 92.5 amp; N = 275 mm·min−1			BW = 6.339
	V = 17.8 V; HI = 216 J·mm−1
				BH = 2.592
6	I = 92.5 amp; N = 300 mm·min−1			BW = 6.084
	V = 17.5 V; HI = 194 J·mm−1
				BH = 2.683
7	I = 105 amp; N = 250 mm·min−1			BW = 7.358
	V = 18.0 V; HI = 272 J·mm−1			BH = 3.689
8	I = 105 amp; N = 275 mm·min−1			BW = 7.381
	V = 18.2 V; HI = 250 J·mm−1			BH = 3.862
9	I = 105 amp; N = 300 mm·min−1			BW = 7.347
	V = 18.0 V; HI = 227 J·mm−1			BH = 3.531

From Table 3, it can be observed that when the value of welding current increases, the bead width increases in a proportional manner. It might have happened due to the fact that the increase in the value of current increases the energy available for melting of base metal in the welding zone. It has also been observed that the variation in the welding speed had resulted in the slight reduction in the value of bead height. This may be due to the reason that when the welding speed increases, the time available for the energy to penetrate into the base metal reduces and results in the reduction in the bead height. From the above study, it is obvious that various combinations of welding current and welding speed are required to achieve the desired bead width and bead height. Therefore, it is concluded that the combination of low current and low traverse speed is the optimal choice to achieve high depth to width ratio in PAW welding process that reduces HAZ, residual stresses and angular distortion.

3 Results and discussion

Once the butt welds were prepared with the selected PAW input process parameters, specimens of required size and dimension were cut to test various mechanical properties, such as tensile strength and root bend test. Moreover, the micro-hardness of the weldments at base metal, HAZ and fusion zone is measured. In addition to the micro-hardness, microstructure of the welded joint at the above specified locations was also examined with the help of optical microscope. Furthermore, the weldments were examined for their grain size in the fusion zone.

3.1 Weld microstructure analysis

The variations of microstructures for plasma arc welded butt joint of Inconel 617 plates at various locations, such as base metal, HAZ and fusion zone, are shown in Figure 3.

Figure 3

Microstructures at various locations on weld zone: (a) base metal, (b) HAZ and (c) fusion zone.

During the welding process, when the molten metal pool is moved across the centreline of the weld, the temperature gradient and grain growth rate vary significantly. Particularly in autogenous welding, the solidification of the weld pool occurs suddenly by the epitaxial growth on the partially melted grains. This solidification process depends on the speed of welding and the grain growth rate. During growth of the solid in the weld pool, the shape of the solid–liquid interface controls the development of microstructural features. The nature and the stability of the solid–liquid interface depend on the thermal and constitutional supercooling. Based on these conditions, the interface growth may occur as planar, cellular or dendritic. It is also referred in the literature [21] that these changes in solidification morphology are directly related to welding conditions.

In the present study, it was observed that the microstructure in the fusion zone shows the formation of dendritic structure after solidification. In Figure 3(a), black spots in the microstructure of base metal reveal the presence of carbides. Moreover, the microstructure of HAZ and fusion zone has shown the formation of γ ¹ (gamma prime) phase with short and long grain boundaries, respectively, confirmed in SEM analysis (Figure 4). It also exhibited the directional solidification in which aligned grain structures and grain boundaries were formed. This fact was also validated from the measurement of average grain size in all the three zones (Figure 3). It clearly shows the grain refinement phenomenon from the base metal to the fusion zone. Further SEM/EDS analysis has been conducted in the fusion zone for the three current and speed ranges, which are presented in Figure 4(a)–(c).

Figure 4

Microstructures and SEM/EDX spectra matrix at fusion zone locations for: (a) 80 amp, 250 mm·min⁻¹; (b) 92.5 amp, 275 mm·min⁻¹ and (c) 105 amp, 300 mm·min⁻¹.

The phase analysis conducted for three different parameters with varying current and weld speed at three levels is presented in Figure 5. The patterns indicate various peaks corresponding to molybdenum (Mo), nickel (Ni), cobalt (Co) and chromium (Cr). Interestingly, all the peaks corresponding to the aforementioned elements are getting weakened at the higher current that the molybdenum addition generally reduces the stacking fault energy and makes it a suitable candidate at the elevated temperature. The exceptional resistance of Inconel 617 to oxidation is due to the presence of chromium and aluminium elements. The phenomenal decrease of these elements in different welding conditions (as observed in Figure 4) compared to parent metal composition may cause the oxidation of the joints up to certain extent particularly high welding current conditions.

Figure 5

X-ray diffraction patterns at the fusion zone.

3.2 Analysis of weld mechanical properties

In this section, the study related to the micro-hardness, tensile strength and bending strength of the plasma arc welded butt joints is presented. Tensile strength and flexural strength were determined by testing three samples fabricated at particular welding conditions. However, the micro-hardness was measured at three different locations on the same sample to eliminate manufacturing uncertainties. The average of the said three readings is reported in this analysis section.

3.2.1 Weld micro-hardness analysis

Here, the Vickers micro-hardness values in the transverse direction of the butt welds measured using micro-hardness tester (Omni tech, Model: S. Auto) are presented. The micro-hardness tests were conducted by applying 500 g of load for a dwell time of 10 s. The hardness values were measured by finding the depth of penetration of the indent made onto the specimen. The hardness values were measured on both sides of the weld centre at an interval of 1 mm till the base plate. The variation of micro-hardness values across the weld zone for a combination of various currents and welding speeds is shown in Figure 6. It has been observed that a certain variation of hardness values exists from fusion zone to the base metal. At the centre of the fusion zone, the hardness values were seen to be high when compared with the HAZ and base metal zone. This might have happened due to the sudden cooling of the fusion zone from very high temperature to the room temperature when compared with the HAZ and base metal. Moreover, it has also been observed that at low heat input values (i.e. I = 80 amp, N = 250 mm·min⁻¹ and HI = 203 J·mm⁻¹) the cooling rate was high and resulted in a high hardness value when compared with high heat input values (i.e. I = 105 amp, N = 300 mm·min⁻¹and HI = 227 J·mm⁻¹). The smaller secondary dendritic spacing observed in the microstructure at low heat input may be the reason for the high hardness values at the fusion zone. The percentage increase in the hardness values was equal to 20.7, 12 and 13% for 80, 92.5 and 105 amp currents, respectively. From these data, it is evident that the welded zone has provided a better hardness value than the base metal.

Figure 6

Micro-hardness of the Inconel 617 butt welded joints measured across the weld bead.

3.2.2 Analysis of weld tensile strength

The strength of the welded joints is greatly influenced by the reformation of the grains that occurred due to the phase transformation in various zones, such as fusion zone, HAZ and base metal of the welding area. The tensile test specimens were prepared as per ASTM E8/E8M standard with 100 mm length and 10 mm width at the gripping (Figure 7). The gauge length and width of the specimen are kept equal to 25 and 6 mm, respectively.

Tensile tests were carried out on the universal testing machine (Tinius Olsen; H50KL). The stress is applied in a direction perpendicular to the weld line. Figure 9 shows the stress–strain curves for all the nine samples of butt welds. Furthermore, Figure 8 shows the fractured specimen in the jaws of the universal testing machine.

Figure 7

Schematic diagram showing the tensile test specimen.

Figure 8

Schematic diagram showing the fractured tensile specimen in the gripper.

Figure 9

Stress–strain curves for various samples of Butt welds.

From the tensile test data acquisition system, Young’s modulus, yield stress, ultimate tensile strength and percentage of elongation were determined. The crack was observed on the weld zone due to the oxidation effect on some samples. The ultimate tensile strength values obtained for sample 3 at 80 amp current and speed 300 mm·min⁻¹ were seen to be higher and nearer to the ultimate tensile strength of the base metal. This might have happened due to the fact that the weld bead and width values of this sample were seen to be minimum, when compared with other samples that are having full depth penetration and minimum weld sag. Moreover, the heat input value was found to be low for this sample when compared with others. It was also observed that at higher speeds, the cooling rate increased and helped in the formation of fine grains [22]. This fact was also confirmed by the average grain size of sample 3, which was found to be the minimum when compared with the others. Alongside, it had also been observed that sample nos. 2 and 4 exhibited the lowest value of UTS. It might have happened due to the partial penetration and undercut with more weld sag observed in samples 2 and 4, respectively. Finally, it has been observed that the UTS of the base metal was more when compared with welded specimens, and most of the failures occurred in the weld zone and heat effected zone of the welding area.

Furthermore, the yield strength of the welded joints was determined by drawing a line parallel to the elastic zone, which was having an offset of 0.2% and intersects the stress–strain curve. It was observed that the yield strength of both the base metal and welded joints was lower than the UTS. It might have happened due to the noticeable strain hardening beyond the yield point. It has also been observed that the percentage of elongation decreased with the increase in the heat input. This has happened due to the increase in the values of fracture toughness of the welded joint attained by the reduction in its ductility [16].

Furthermore, scanning electron microscope is used to observe the fractured surface of the tensile test specimens as shown in Figure 10. The SEM images reveal that weld metal has failed under ductile mode with fine dimples and micro-voids in the fractured surface. At higher heat inputs, the void space is observed to be higher which leads to the reduction of tensile strength compared to other specimens. Also high carbon content (30.8 wt%) is found in EDX spectra matrix for the high current sample. It may also cause the weld beads to be more susceptible towards solidification cracking. It is also attributed to the dissolution of Mo elements to the matrix during the welding, which is confirmed through the XRD patterns. However, with the increase in the values of ultimate tensile strength, these voids may develop cracks and it leads to failure.

Figure 10

Scanning electron micrograph of fractured surface after tensile test at room temperature: (a) I = 80 A, 250 mm·min⁻¹; (b) I = 92.5 A, 275 mm·min⁻¹ and (c) I = 105 A, 300 mm·min⁻¹.

3.2.3 Analysis of weld bending strength

Three-point bend test was performed to check the soundness and ductility of the weld. The root bend test was conducted on the universal testing machine (Tinius Olsen; H50KL) with suitable fixtures. The specimens were prepared as per ASTM E190/E192 standards (Figure 11) with length and width equal to 100 and 10 mm, respectively. The variation of flexural strengths for various samples is shown in Figure 12. It has been observed that the flexural strength of sample 3 was high when compared with other samples. This is due to the fine grain structure formed by the low heat input. Grain refinement and formation of larger number of small nuclei due to the fine grain structure may lead to the increase in flexural strength. Furthermore, no visible tears or cracks and open discontinuities were observed on the test samples even after 90° bending angle.

Figure 11

Schematic diagram of the bending specimen before and after bending.

Figure 12

Schematic diagram showing the variation of flexural strength for various specimens.

4 Conclusions

In the present study, an attempt was made to perform the microstructural and mechanical properties of melt-in-mode PAW of Inconel 617 super alloy. The following conclusions were drawn from the said experimental investigations:

• The BoP trails indicated that the depth of penetration decreased with the decrease in the value of current and increase in the value of speed.

• Furthermore, the butt welding of plates suggested that the bead width increased with the increase in the value of weld current.

• The microstructural study revealed a coarse grain structure in the fusion zone due to the formation of γ¹ phase with directional solidification. It was also observed that the grain coarsening increased from the parent metal to the fusion zone of the weld.

• XRD patterns revealed the dispersion of molybdenum to the matrix at the higher current and speed.

• The micro-hardness test illustrated a minimum value of hardness in the base metal and this was observed to be increasing rapidly towards the centre of the fusion zone. This happened with the increase in the cooling rates from the base metal to the fusion zone.

• The tensile and bending tests showed an increase in the value of its strength at low heat input values due to the formation of fine grain structure.