Hydrogen and molybdenum control on laves phase formation and tensile properties of inconel 718 GTA welds

Hydrogen and molybdenum control on laves phase formation and tensile properties of inconel 718 GTA welds

Materials Science & Engineering A 773 (2020) 138874 Contents lists available at ScienceDirect Materials Science & Engineering A journal homepage: ht...

11MB Sizes 0 Downloads 8 Views

Materials Science & Engineering A 773 (2020) 138874

Contents lists available at ScienceDirect

Materials Science & Engineering A journal homepage: http://www.elsevier.com/locate/msea

Hydrogen and molybdenum control on laves phase formation and tensile properties of inconel 718 GTA welds N. Anbarasan a, S. Jerome a, *, S.G.K. Manikandan b a b

Department of Metallurgical and Materials Engineering, National Institute of Technology, Tiruchirappalli, 620015, Tamil Nadu, India Indian Space Research Organization Propulsion Complex, Mahendragiri, India

A R T I C L E I N F O

A B S T R A C T

Keywords: Hydrogen addition Molybdenum GTAW Laves phase Tensile test

Gas Tungsten arc welding (GTAW) process was employed for welding of Inconel 718 with two different shielding gases, namely argon (Ar) and argon with a 5 vol% hydrogen mixture (ArH) and two fillers viz., ERNiCrMo-10 and ERNiCrMo-4. The effects of gas composition and filler wires on the laves phase formation were studied in detail. The results revealed that hydrogen addition through ArH shielding gas mixture resulted in better grain refine­ ment in the welds than pure Ar. The hydrogen addition induced a steep thermal gradient in the weld, which lowered the segregation of elements like Niobium (Nb) and Molybdenum (Mo) at the interdendritic regions. The laves phase formation in Mo-rich filler addition welds was minimized due to restriction of Nb segregation by Mo at the interdendritic region. Tensile test results indicated that the strength and ductility of the joints of both autogenous and filler added welds of Ar were higher than the ArH shielded welds. In the case of filler added welds, higher Mo content filler exhibited better tensile properties in both shielding gas combinations due to solid solution strengthening of Mo. Nano-sized hydrogen assisted cracks observed in ArH autogenous welds caused a reduction of strength and ductility.

1. Introduction Aerospace-grade Inconel 718 is used in the gas turbine and liquid propulsion rocket engine components extensively. The alloy exhibits high strain age cracking resistance than other Ni-based super alloys. Hence, it possesses good weldability. However, in certain welding con­ ditions, the solidified weld metal is blemished with defects like micro fissuring, solidification cracking, and formation of laves phase. Laves phase (Ni, Cr, Fe)2(Nb, Mo, Ti) is a topologically closed packed structure (TCP), it depletes the solid-solution strengthening elements such as Nb and Mo from the matrix [1,2]. Moreover, the phase aids for crack initiation during service and leads to a reduction of tensile strength, fatigue, creep resistance, and toughness of the weld. It thus reduces the mechanical properties of Inconel 718 welds. The low melting point laves phase favors the retention of the liquid film between advancing den­ drites that contribute to the solidification cracking [3]. Nb has a low partition coefficient value (k) as compared to other alloying elements. Therefore, it tends to segregate at the interdendritic region during so­ lidification. Few studies reported that in addition to Nb, other solute elements like Fe, Cr, and Si could also promote the formation of laves phase in superalloys [4,5]. Radhakrishnan and Prasad Rao [6] suggested

that replacing of Iron (Fe) by Cobalt (Co) and Nb by Tantalum (Ta) could reduce the laves phase formation. However, during welding, alteration of the chemical composition of the weld can be achieved through the filler wires. Filler wires having higher partition coefficient elements such as Mo, Co, W, Ta, and Re are preferable to reduce the laves phase formation [7]. Studies [8,9] proved that Molybdenum (Mo) - based filler wires could effectively mitigate the Nb segregation. It has a similar atomic size and having a high partition coefficient value than the Nb. Hence, it effectively reduces the Nb segregation, and further, it improves the solid-solution strengthening of welds [10,11]. Manikandan et al. [7] observed by using Mo-based fillers restrict the Nb segregation and laves phase formation in the fusion zone of Inconel 718. The cooling rate also plays a significant influence on the lowering of laves phase segregation [8,12]. Laser and electron beam welding pro­ cesses could reduce the laves phase segregation effectively due to its extreme cooling rate during solidification [13,14]. However, those processes could not be practically applicable due to the intricate design of aerospace components. Therefore, the manual mode of the GTAW process is a preferable welding process for joining of Inconel 718. For achieving a higher cooling rate in a GTAW process, few authors adopted dynamic grain refinement techniques such as arc oscillation and current

* Corresponding author. E-mail address: [email protected] (S. Jerome). https://doi.org/10.1016/j.msea.2019.138874 Received 30 October 2019; Received in revised form 22 December 2019; Accepted 23 December 2019 Available online 28 December 2019 0921-5093/© 2019 Elsevier B.V. All rights reserved.

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Table 1 Elemental composition (wt.%) of Base material and filler wires. Material

Ni

Mo

Cr

Fe

Mn

Si

C

Al

Ti

Other

Inconel 718 Filler 1 Filler 2

Rem Rem Rem

3.0 14 17

17.4 21.4 16.5

20 4.4 5.8

– 0.14 0.5

0.06 0.07 0.01

0.05 0.01 0.01

.45 – –

1 – –

Nb þ Ta- 5.5 Co-0.1, W-3.1 Co-0.5, W-4.0

Table 2 Welding parameters. Shielding gas

Filler Material

Voltage

Peak Current

Background current

Pulse on Time

Welding speed

Pulse frequency

Heat Input

(l/min) Ar Ar Ar ArH ArH ArH

(Dia-1mm) Autogenous Filler 1 Filler 2 Autogenous Filler 1 Filler 2

(V) 10.5 10.6 10.6 15.4 15.9 16.3

(A) 110 110 110 80 80 80

(A) 40 40 40 20 20 20

60% 60% 60% 60% 60% 60%

mm/s 1 1 1 1 1 1

Hz 20 20 20 20 20 20

kJ/mm 0.615 0.621 0.621 0.619 0.639 0.656

specific heat of Helium. The hydrogen facilitated in transferring more heat to the anode as compared to pure argon [21]. The addition of hydrogen with argon improves the arc efficiency of the GTAW arc [22] and reduces the surface oxidation [23]. Up to 5% of hydrogen addition with argon is used for welding of nickel based alloys and stainless steel. Present authors experimented with a mixture of Argon and 5 vol % of Hydrogen shielding gas mixture in their previous investigation and achieved the reduction of volume fraction of the laves phase [24]. Based on the literature studies, the present investigation aims to study the following objectives. Firstly, the reduction of laves phase formation by employing steep temperature gradient via Argon hydrogen gas mixture shielding and Molybdenum filler wires. Secondly, to un­ derstand the effects of shielding gases and Mo on the mechanical properties of the Inconel 718 welds. Fig. 1. Schematic representation of Tensile specimen and TEM specimen extraction region.

2. Experimental procedure Two millimeters thick Inconel 718 sheet was used as the base ma­ terial. The sheet was solution annealed at 980 � C for 1 h before the welding. The chemical composition of the base material and two Morich fillers, namely ERNiCrMo-10 (Filler 1) and ERNiCrMo-4 (Filler 2) are listed in Table 1. Two shielding gases, such as Argon (Ar) (99.9%) and premixed Argon (95%) plus 5% of hydrogen (ArH), were used at a flow rate of 15 l/min. The gas mixture containing 5% hydrogen in argon is classified as R1(reducing gas mixture) according to AWS A5.32 (ISO 14175) [25]. Square butt joint was made with a dimension of 150 � 150 � 2 mm. Pulsed mode of GTAW was used for all welding experiments, and the welding parameters are listed in Table 2. For comparative purposes, the same arc energy was maintained for Ar and ArH shielding gases. For metallography and mechanical properties studies, samples were sectioned by the wire cut electric discharge machining process (WEDM). For metallography studies, samples were electrolytically etched in a 10% oxalic acid solution in water. Optical microscopy (OM) was carried out using Leica DM2700 to observe the microstructures of the welds. Field Emission Scanning Electron Microscopy (FESEM) with Energy Dispersive X-Ray Spectroscopy (EDS) was used for analyzing the segregation constituents in the interdendritic region and fractography studies. The phase analysis was carried out by X-Ray Diffraction (XRD) with Cu-Kα radiation (1.54056A). Rigaku Ultima III diffractometer was used, and the samples were scanned at the rate of 1� /min over two ranges of 30–100� . The ternary phase diagram and Scheil solidification simulations were analyzed by ThermoCalc® with a thermodynamic database for Ni-based alloys. Instron 8801 universal tensile testing machine was used to evaluate the tensile strength as per ASTM E8/8M 11 at room temperature. JEOL JEM-2100 electron microscope at 200 kV was used for Transmission Electron Microscope (TEM) analysis.

pulsing [15,16]. Janaki Ram et al. [9] reduced the volume fraction of the laves phase in the weld by adopting the current pulsing method [12,17]. Manikandan et al. [18] employed an alternative approach to enhance the weld cooling rate by inducing a steep thermal gradient across the weld pool using a Helium (He) gas shielding. Inducing the steep thermal gradient during welding improves the grain refinement process and reduces the Nb segregation in the interdendritic regions [19,20]. Helium has a higher ionization potential, and it transfers a hotter arc than the pure Argon. However, due to its low density, the excessive gas flow rate is required for proper protection of the weld pool. J.F Key reported that Argon þ5% Hydrogen mixture had similar thermal conductivity and

Fig. 2. Base Metal microstructure of solution annealed Inconel 718. 2

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Fig. 3. Macrostructures of Ar and ArH shielded welded joints. Table 3 Bead width of autogenous and filler welds with Ar and ArH.

3.2. SEM/EDS analysis of the weld

Shielding gas

Autogenous

Filler 1

Filler 2

Ar ArH

mm 7.1 7.5

mm 6.4 6.6

mm 6.2 6.6

Fig. 7 a-f show the SEM micrographs captured at the center of the weld. The Secondary Dendritic Arm Spacing (SDAS) distance ‘λ2’ is estimated as per the methodology proposed by Rocha et al. [27]. The ‘λ2’ value quantifies the amount of grain refinement achieved by using ArH shielding gas and filler wires. Empirical relation (1) proposed by Vish­ wakarma et al. [8] was used to predict the cooling rate of the Inconel 718 using the values of SDAS.

Specimens for TEM were extracted from the fusion zone of the deformed tensile specimens, as shown in Fig. 1.

λ2 ¼ 141(έ)

3. Results

0.4

(1)

The cooling rate (έ ¼ GxR) controls the grain size of the weld, and it is inversely proportional to the SDAS. The faster cooling rate will in­ crease the solute concentration ahead of the dendritic tip and promotes equiaxed dendrites growth. The SDAS values are tabulated in Table 4 and found that the welds made with ArH shielding gas have higher cooling rates than the Ar shielding. Figs. 8–11 shows the SEM micrographs with EDS point analysis re­ sults at the dendritic core and interdendritic regions of the welds. Fig 8a and 10a reveal the presence of the laves phase at the interdendritic re­ gion in both autogenous welds. The formation of the laves phase attri­ butes to the segregation of elements like Nb and Mo at the interdendritic regions. EDS line scan analysis confirmed the segregation of the ele­ ments of Nb and Mo (Figs. 9a and 11a) over the other alloying elements. A point EDS analysis was performed at the dendritic core and inter­ dendritic regions, and the values are tabulated in Table 5. In autogenous welds, the concentration of Nb and Mo at the interdendritic regions is more than the dendritic core. However, a similar observation was not found in filler added welds. The microsegregation behavior of Nb and Mo across the HAZ and fusion line were analyzed with an EDS line scan (Fig. 12). In autogenous welds, no significant changes in the concentration gradient of both Nb and Mo are observed in HAZ (Fig. 12 a&d). A similar trend is observed in the PMZ due to the liquation of NbC [28]. In filler added welds of both Ar (12c&d) and ArH (12e&f), the concentration of Mo increases grad­ ually from the fusion line to the weld center, and the concentration of Nb is reduced.

3.1. Microstructural characterization of welds by OM Solution annealed Inconel 718 base metal microstructure is shown in Fig. 2. Austenitic microstructure with annealing twins and MC pre­ cipitates at the grain boundaries are observed. Fig. 3 a-f show macro­ structures of all weldments. No visible defects like cracks and porosities are observed in both fusion zone (FZ) and Heat affected zone (HAZ). The welds obtained by ArH appear wider than the Ar shielded welds in both autogenous and filler addition welds. Shielding gases have a significant effect on the bead dimension, and the same are listed in Table 3. Figs. 4 and 5 show the FZ microstructures of autogenous, Filler 1, and Filler 2 welds of Ar and ArH shielding. Welds are solidified as an austenitic phase with distinct solidification structures [26]. Solidifica­ tion grain boundaries (SGB) and Solidification subgrain boundaries (SSGB) are visible in the weld microstructures. SGBs are formed due to the intersection of a group of subgrains (dendrites or cells) [26]. SSGBs separate the adjacent dendrites or cells. In general, solidification cracks normally appear in the SGBs due to solute dumping. However, no such cracks are observed in any of the weld microstructures. Fig. 6 shows the HAZ microstructures of autogenous and filler added welds. Autogenous Ar (Fig. 6a) and ArH (Fig. 6d) welds have a narrow partially melted zone (PMZ) and in the filler added welds (Fig. 6b&c) and (Fig. 6e&f), an unmixed zone (UMZ) appears adjacent to the PMZ. The UMZ is a remelted base metal with a negligible mixing of fillers. Width of the UMZ appears narrower in ArH welds than the Ar shielded welds.

3

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Fig. 4. FZ Microstructures of Ar shielded welds Autogenous (a–b), Filler 1(c–d), and Filler 2(e–f).

4

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Fig. 5. FZ Microstructures of ArH shielded welds Autogenous(a-b), Filler 1(c–d), and Filler 2 (e–f).

5

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Fig. 6. Optical Microstructures at HAZ of a) Ar autogenous b) Ar-Filler 1 c) Ar-Filler 2 d) Ar autogenous e) ArH-Filler 1 and f) ArH-Filler 2.

Fig. 7. SEM images of a) Ar autogenous b) Ar-Filler 1 c) Ar-Filler 2 d) ArH autogenous e) ArH-Filler 1 and f) ArH-Filler 2.

6

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

3.3. XRD analysis of welds

Table 4 Solidification Parameters at the FZ of weldments. Specimen

SDAS (λ2)

Cooling Rate (έ ¼ GxR)

Computed Thermal Gradient (G)

Ar-Autogenous ArH-Autogenous Ar Filler 1 Ar Filler 2 ArH Filler 1 ArH Filler 2

(μm) 9.491 5.590 9.111 8.954 5.103 5.012

(K/s) 850.6 3195.3 942.1 984 4013.1 4197.1

(K/mm) 425.3 1597.6 471 492 2006.5 2098.5

Fig. 13 shows the X-ray diffraction pattern of all welds. Austenitic (Ni) peaks are dominant in all welds. However, the laves phases (L) peaks are not observed in filler added welds, but weak peaks are seen in the autogenous welds. The amount of NbC formed during solidification is insignificant due to the lower carbon content in the welds; hence, no peaks are detected [9]. Mo-rich filler reduces the formation of NbC. The lattice parameters of the γ peaks of autogenous and filler metal welds are calculated, and the values are listed in Table 6. The value of the lattice parameter of filler metal welds is larger than autogenous welds due to the enrichment of Mo. The measured lattice parameters in the current

Fig. 8. SEM image of laves phase segregation and EDS point scan at Dendritic Core (DC) and interdendritic region (ID) of a) Ar autogenous, b) Ar-Filler-1, and c) ArFiller-2. 7

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Fig. 9. EDS line scan Nb and Mo across DC and ID of a) Ar autogenous, b) Ar-Filler-1, and c) Ar-Filler-2.

8

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Fig. 10. SEM image of laves phase segregation and EDS point scan at Dendritic Core (DC) and interdendritic region (ID) of a) ArH autogenous, b) ArH-Filler-1, and c) ArH-Filler-2.

9

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Fig. 11. EDS line scan Nb and Mo across DC and ID of a) ArH autogenous, b) ArH-Filler-1, and c) ArH-Filler-2.

10

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

The ratio of the thermal gradient (G) to the growth rate (R) de­ termines the mode of the solidification structure of the weld. Since ArH addition increases the G of the weld pool, it influences the dendrite morphologies. All welds showed packets of equiaxed, and columnar dendrites feature due to the competitive grain growth during solidifi­ cation. During solidification, the dendrites grow towards the center of the weld from the fusion line. Pulsed GTAW process was employed for welding, due to the fluctuation of the arc intensity, the weld pool dis­ lodges the tip of the growing dendrite ahead of the solidification front. The fragmented dendrites act as heterogeneous nucleation sites and promote the formation of SGB [17,18]. Hence, equiaxed dendritic grains observed at the center of the weld in Fig. 4 and 5. Fine grains with smaller SSGB areas are preferred since they provide a smaller inter­ dendritic space for the segregation of elements like Nb and Mo. From the results (Table 4), it is evident that the welds by ArH have smaller SDAS values than the Ar welds. The calculated cooling rate values indicate that the combination of filler materials and ArH shielding have a significant influence on the higher cooling rates than the Ar and autogenous welds. The higher cooling rates promote a nonequilibrium mode of solidification and stimulate the fine grain struc­ ture. The higher concentrations of Mo in Filler 1 and Filler 2 endorse faster cooling rates. Molybdenum has a low diffusivity than the Nb. The concentration gradient of Mo at the interdendritic liquid encourages the constitutional supercooling [34,35]. The solute pileup at the dendritic root resulted in dendritic fragmentation and led to subsequent grain refinement. Thus, the use of Mo-rich fillers aided in the formation of a fine grain structure with smaller SDAS compared to autogenous welds. EDS results (Table 5) show that the concentration of Nb in the interdendritic region of autogenous Ar (29.14%Nb) is higher than the ArH (21.09%) welds. As per the Ni–Nb binary system, due to low liq­ uidus temperature favors laves phase formation as reported by Hume Rothery [29]. The segregation ratios (concentration in liquid/­ concentration in solid) of Nb and Mo have been calculated (Table 7) as per the technique proposed by Wang et al. [36]. The segregation ratios of Nb was much higher than Mo due to the lower partition coefficient of Nb. The measured segregation ratios also indicate that the ArH weld exhibits lower segregation of Nb and Mo between the dendritic core and the interdendritic region [18,24]. Thus, in the case of autogenous welds, a significant reduction in the Nb segregation and laves phase formation was obtained due to hydrogen addition to the shielding gas. In the case of the segregation behavior of welds made with Filler 1 and Filler 2, the results are not consistent with autogenous welds. Because the size of the laves phase observed in the interdendritic regions of Ar shielded Filler 1 (Fig. 8b) and Filler 2 (Fig. 8c) welds are low. The laves phase observed in ArH shielding (Fig. 10b and c) is negligible. The EDS point and line scan (Fig. 9b&c) results indicate that the interden­ dritic regions of Ar shielded filler metal welds are more abundant with Mo and low in Nb concentration than ArH shielded welds (Fig. 11b&c). Notably, concentrations of solute elements are much lower as compared to autogenous welds, and it is attributed that the Mo rich fillers induce faster cooling rates (Table 2). The probability of secondary phase for­ mation in Mo rich fillers is higher. Perricone et al. [37] suggested that Filler 1, and Filler 2 used in the present study solidifies as L→Lþγ→ Lþγ þP→ Lþγ þP þσ→ γ þP þσ (Filler 1) and L→Lþγ→ Lþγ þP→γ þP (Filler 2). However, these phases not observed. From Table 5, Filler 2 welds have higher Mo concentration than the Filler 1 due to higher Mo wt.% (17%) in Filler 2 as compared to Filler 1(14%). From the Ni–Mo binary system [38], the formation of metastable phases such as Ni4Mo and Ni2Mo requires a minimum of 16 wt% of Mo [39]. The Mo content in both filler materials is more than 14 wt % and the possibility of for­ mation of the above-mentioned metastable phases are high. However, due to the low dilution of Mo, concentrations of Mo at the interdendritic, as well as dendritic core regions are less than <16% (Table 5). Hence those phases are not observed in the FZ. The segregation ratio (Table 7) of Mo in both Filler 1 and Filler 2 weld found to be the same irrespective of the shielding gases. However, the values are much lower than the

Table 5 EDS results showing wt.% composition at Interdendritic (ID) and dendrite core (DC). Specimen

Ti

Cr

Fe

Ni

Nb

Mo

W

Ar-DC Ar-ID ArH-DC ArH-ID ArFiller 1-DC ArFiller 1-ID ArFiller 2-DC ArFiller 2-IC ArH Filler 1-DC ArH Filler 1-ID ArH Filler 2-DC ArH Filler 2-ID

0.74 1.16 0.6 0.98 0.98 1.11 0.83 0.98 0.51 0.36 0.3 0.49

18.75 12.03 16.11 13.92 21.2 19.49 18.25 17.84 19.48 17.97 18.13 17.92

17.49 10.37 17.59 10.03 13.16 10.82 13.16 10.33 10.54 8.16 12.76 11.21

56.54 40.7 60.74 49.79 52.67 48.69 52.48 50.17 56.35 59.67 57.49 56.96

3.69 29.14 3.24 21.09 2.46 8.97 3.32 6.85 2.7 3.17 2.32 3.28

2.79 6.6 1.72 4.19 8.79 9.88 11.6 13.08 9.75 10 8.5 9.68

– – – – 0.74 1.04 0.36 0.75 0.67 0.67 0.5 0.46

investigation are correlated with similar studies [7,29]. In ArH welds, due to the retention of Nb and Mo in the dendritic core, the lattice parameter values are higher than the Ar welds. Filler 2 welds have a larger lattice parameter welds than the Filler 1 due to higher concen­ tration of Mo [30]. 3.4. Thermocalc simulations The present investigation is concerned about the segregation and evolution of phases involving elements like Nb, Mo, and Ni. A ternary phase diagram of the Ni–Mo–Nb system was simulated using Thermo­ calc. The simulation enables to predict the phases evolved at various temperatures. Fig. 14 shows the phases which evolve in 1400, 1300, 1200, 1100, 1000, and 900� C. Secondary phases such as the P phase, μ phase, σ phase, and Ni3Nb intermetallic formations observed in the Ni–Mo–Nb system. However, in the ternary diagrams, the focus is mainly on the phases which are evolved at the equivalent composition of the Nb, Mo, and Ni, a pointer in Fig. 14, represents the same. Scheil solidification simulator was used to predict the fraction of the solids formed in the FZ [31]. Fig. 15 shows the calculated solidification path of Inconel 718, Filler 1, and Filler 2. The simulations show the possibility of laves, P, and σ phase formations in the welds. 3.5. Tensile testing of welds Ultimate Tensile strength (UTS) and ductility of the welds are eval­ uated using the tensile test performed at a constant strain rate of 1 � 10 3s 1. Fig. 16 shows the images of the tensile test specimen. Obser­ vation from the fractured samples indicates that the Ar shielded filler metal welds failed away from the FZ while all the other welds failed at the FZ. The engineering stress-strain curve of the tensile test is shown in Fig. 17. Overall, the UTS of Ar shielded welds are higher than the ArH welds. The tensile strength of Ar shielded Filler 2 weld is highest than the Filler 1 and autogenous welds (Filler 2 > Filler 1 > autogenous). Similarly, in the ArH shielding, Filler 2 showed the highest value than the Filler 1 and autogenous welds. 4. Discussion 4.1. Effect of hydrogen addition and molybdenum rich fillers on laves formation The addition of 5 vol% of hydrogen to argon gas, increases the arc temperature considerably by the dissociation and recombination of hydrogen molecules (H2↔H þ H 422kJ) [32]. It also increases the thermal conductivity and increases the ionization potential of the arc, which in turn increases the arc voltage between the workpiece and the electrode [22]. Hence, the volume of the weld pool is increased, and it resulted in wider welds (Table 3) than the Ar shielding [32,33]. 11

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Fig. 12. SEM/EDS at HAZ of a) Ar autogenous b) Ar-Filler 1 c) Ar-Filler 2 d) Ar autogenous e) ArH-Filler 1 and f) ArH-Filler 2.

12

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Fig. 13. X-ray diffraction peaks of autogenous and filler metal welds of a) Ar and b) ArH. Table 6 Lattice parameters of γ peaks in autogenous and filler welds with Ar and ArH. Shielding gas Ar ArH

Autogenous

Filler 1

Filler 2

Å

Å

Å

3.60959 3.61082

3.61198 3.62522

3.61957 3.62621

4.3. Hydrogen and molybdenum effects on the tensile properties of the welds The welds of Filler 1 and Filler 2 in both shielding gases exhibited better tensile properties than autogenous welds. The UTS of Filler 2 has the highest values in both Ar and ArH environments due to the higher concentration of Mo content (Table 5). The enhancement of the tensile strength of filler added welds was due to the solid solution strengthening of Mo. The higher value of lattice parameters listed in Table 6 supports the strengthening effect of Mo fillers. Jang [10] reported a similar observation and informed that the increase in strength as well as ductility of Inconel 718 due to the addition of Mo content. However, fine-grained autogenous and filler ArH welds with lower Nb and Mo exhibited lower UTS and ductility (10%) as compared to Ar shielded welds (40%). The reduction of the tensile strength due to the hydrogen-assisted cracking. Fractography studies on the fracture sur­ faces of Ar welds show dimples and voids, which indicate a ductile mode of fracture [8]. ArH fracture surface (Fig. 18) revealed flat facet features indicating a quasi-cleavage mode of fracture, and it confirms the pos­ sibility of hydrogen-assisted cracking [40]. At higher magnification (Fig. 19), the fracture surface of ArH autogenous welds revealed the slip bands, which is consistent with the hydrogen enhanced localized plas­ ticity theory [41,42]. David et al. [43] and Li et al. [23] also reported similar hydrogen assisted cracking in Inconel 718. The TEM image is shown in Fig. 20 reveals the accumulation of dislocations around the laves phase in ArH welds and the resulting in a crack initiation around it. Generally, the laves phase acts as stress con­ centration sites, which attract hydrogen accumulation. It leads to a reduction of cohesive strength localized manner [44,45]. In addition to the laves phase, the presence of high angle SGBs in the welds can also act preferential sites for hydrogen-assisted cracking [41]. Oriani et al. [46] notified that the hydrogen lowers the cohesive force between the atoms around trapping sites. During the tensile test, the local stress exceeded the cohesive force by the hydrogen and resulted in premature failure. Moreover, the fine-grained microstructure of ArH welds provides a larger grain boundary area, and the finer laves phase acts as potential trapping sites for hydrogen entrapment [47,48]. Thus, the autogenous ArH weld failed with the least elongation due to hydrogen entrapment. The severity of hydrogen in the filler metal welds are minimum, the UTS

autogenous welds. It confirms that Mo tends to be less sensitivity to cooling rates [37]. A low Nb segregation ratio is observed in ArH filler welds due to the absence of Nb in the filler and high cooling rate experienced during solidification. Similar to the autogenous welds, hydrogen addition, in combination with the use of Mo rich fillers, resulted in reducing the detrimental laves phase. 4.2. Analysis of thermocalc simulation Ternary phase diagrams (Fig. 14) of the Ni–Nb–Mo system shows that at 1400� C, a complete liquid phase exists at the equivalent pro­ portion, and a γ þ Liquid phase evolves as the temperature drops to 1300� C. As the temperature of the ternary system is getting reduced, the P and σ phases evolve with the γ solidification structure. However, no evidence of such phases observed from the microstructural studies and XRD results. The absence of these phases attributes to the nonequilibrium solidification of the weld pool. As per the Scheil simula­ tion (Fig. 15), at the terminal stage of solidification, a wider solidifica­ tion range with the formation of the laves phase is observed. Whereas, the solidification ranges of Filler 1 and Filler 2 are found to be smaller. However, the Scheil solidification predictions are closer to the Ar, and ArH autogenous welds with the observation of the laves phase in both microstructural and XRD results. Again, the P or σ phases predicted by the Scheil simulation are not observed in the experimental results. The mechanism of microstructural development can support the inconsis­ tency between simulation and experimental results during solidification. Dilution influences the composition of the filler added welds, which subsequently reduces the feasibility of the secondary phase formations as predicted by the simulation. Also, when the solidification rates determine the development of these phases, the narrow solidification range reduces the probability of the formation of these phases.

13

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Fig. 14. Phase Diagram of a Ni–Nb–Mo system at 1400� C, 1300� C, 1200� C, 1100� C, 1000� C and 900� C and the pointer (✦) indicates the composition region where the made with Mo rich fillers welds.

14

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Table 7 Segregation ratios of Nb and Mo in autogenous and filler welds with Ar and ArH. Shielding Gas Ar ArH

Autogenous

Filler 1

Filler 2

Nb

Mo

Nb

Mo

Nb

Mo

7.89 6.51

2.36 2.44

3.65 1.17

1.12 1.02

2.06 1.41

1.12 1.13

TEM images in Fig. 21 illustrates the dislocation pileup in the SGBs of Ar and ArH shielded welds. ArH welds reveal nano-size cracks (Fig. 21 d, e &f) since a high concentration of hydrogen resulted in an agglomer­ ation of dislocations around the laves phases and SGBs. As a result, extensive hydrogen-dislocation interaction at these regions leads to the formation of nanocracks. These nanocracks weaken the laves phase/ matrix interface, and SGBs of ArH shielded welds resulting in crack initiation and propagation. Hydrogen enhances dislocation mobility, which initiates cracking and propagates along the grain boundaries having high dislocation densities. Stenerud et al. [50] also reported that the hydrogen in Inconel 718 reduces the energy needed for dislocation nucleation and the mobility of the dislocations. Moreover, the interac­ tion between the twins and grain boundaries can lead to premature failure of the weld, as observed in austenitic stainless steels [51]. The fine-grained autogenous ArH welds provide more grain boundary area for the dislocations to accumulate, and it leads to a reduction of me­ chanical properties. It proves that the Hydrogen mixing with Argon gas is beneficial in reducing the laves phase size; however, the hydrogen assisted cracking phenomenon makes it detrimental to the tensile

Fig. 15. Scheil solidification curves of Inconel 718, Filler 1, and Filler 2.

of autogenous ArH (908 MPa) is only 4% lower value than Ar weld (943 MPa). The absence of laves phases led to lesser hydrogen trapping sites and further reduced the hydrogen-assisted cracking in the filler added ArH shielded welds. SGBs and twins which evolve during plastic deformation and hinder the movement of dislocations due to the pileup of dislocations at the grain boundaries [49].

Fig. 16. Tensile test samples after the test.

Fig. 17. Engineering stress vs. strain graph of a) Ar shielded welds b) ArH shielded welds. 15

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Fig. 18. Fractured surfaces of a) Ar autogenous b) Ar-Filler 1 c) Ar-Filler 2 d) ArH autogenous e) ArH-Filler 1 f) ArH-Filler 2.

welds. Fig. 22a shows three regions (1, 2 &3). Region 1 is similar to the dynamic recovery regime (Stage III) of polycrystalline material [51,52]. Regions 1 and 2 represent a steady stage regime (Stage IV) in this stage; a balance between softening and hardening is reached [53]. Region 2 represents the starting point of twinning, and the dislocation pile-ups in SGBs and stress concentration regions of the laves phase [54]. The work hardening rate becomes steady in Region 2 and increases slightly at the end. Region 3 marks the barrier for twinning, and it followed by slip based deformation. A sudden drop in the work hardening rate at higher flow stress after Region 3 is attributed to the dislocation gliding [52]. Except for autogenous ArH, all other welds exhibited the work hard­ ening regimes, as observed in the base material. ArH shielded welds with Filler 1 and Filler 2 having a slope of 1.60 and 1.77 while Ar shielding welds slopes are 1.02 and 1.05 in Region 2 of Fig. 22c. The higher slope values of ArH indicate that the faster work hardening rate has occurred in the filler welds than the Ar welds. An increase in work hardening is attributed to the increase in the lattice parameter of the γ phase, as observed in Table 6 [55]. From Fig. 22c, it is found that the onset of Region 2 is evolved earlier in ArH welds than the Ar welds. The ArH filler welds have higher twin stress in Region 2 (Filler 1–383 MPa & Filler 2–398 MPa) than Ar welds (Filler 1–340 MPa & Filler 2–339 MPa). Thus, the use of ArH and Mo-rich Fillers enhanced the work hardening behavior of the welds. However, the twinning range (Region 2 to 3) of ArH welds are shorter than the Ar welds. The

Fig. 19. Slip bands in fracture surface of autogenous ArH weld.

properties in the as-welded condition. The use of Mo rich fillers has significantly reduced laves phase segregation, and it also provided resistance to hydrogen assisted cracking. 4.4. Work hardening behavior of welds Kocks-Mecking plot analyzes the work hardening behavior of the

Fig. 20. a) Dislocation concentration around laves phase b) Formation of nanocracks in AH shielded welds. 16

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

Fig. 21. TEM images of a) Ar autogenous b) Ar-Filler 1 c) Ar-Filler 2 d) ArH autogenous e) ArH-Filler 1 f) ArH-Filler 2.

Fig. 22. Kocks-Mecking plots of a) Base metal b) Autogenous Ar and ArH c) Ar and ArH Filler metals.

17

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

behavior in the autogenous welding mode; however, the work hard­ ening is improved in filler metal added welds. 5. Conclusions The effect of Hydrogen addition (5%) to the Argon shielding gas and Mo-rich fillers on the laves phase formation and tensile properties on Inconel 718 pulsed GTAW welds have been investigated. The following conclusions are drawn from the present investigation. ● ArH shielding gas, along with current pulsing, enhanced the cooling rates of the welds and resulted in fine grain structure and reduction in laves phase formation. ● ArH shielding mixture reduced the Nb and Mo segregation at the interdendritic region as compared to Ar welds. ● The use of Mo rich fillers reduced Nb segregation and resulted in a reduction of laves phase. ● Thermocalc simulations showed good agreement with experimental results in predicting laves phase formation in case of autogenous welds. ● The tensile strength and elongation of Ar shielded welds are marginally higher than ArH welds. The supply of Hydrogen through shielding gas degraded the tensile properties of ArH autogenous welds. ● The tensile strength of Filler 2(Ar-977MPa, ArH-908MPa) is higher than Filler 1(Ar-952MPa, ArH-880MPa), due to solid solution strengthening of Mo. ● ArH filler welds achieved higher work hardening rates than Ar welds.

Fig. 23. a) Deformation Twins observed in TEM analysis b) Schematic repre­ sentations of Dislocation Silp Bands (DSBs) and Deformation Twins.

reduction in the twinning range attributes to the presence of hydrogen in the weld. Also, it accelerates the dislocation pileup in the newly formed twin boundaries, thus reducing the twinning range. Fig. 23 shows the deformation twins formed, and a schematic dia­ gram representing the interaction between the dislocation slip planes and the deformation twins. Zhenbo Zhang et al. [56] also reported a similar correlation during the deformation of hydrogen charged Inconel 718. Studies by Chateau et al. [57] indicates that the hydrogen induces new dislocation slip planes to weaken the stress field between disloca­ tions and barriers in region 3. Fig. 24 shows the Dislocation Slip Bands (DSB) of all the welded samples, and the high-density dislocations are observed in ArH shielded samples than the AR shielded samples. Cracks always propagate along planar DSBs in the presence of hydrogen [56]. The presence of nano cracks in Fig. 21 supports the phenomenon, and the coalescence of these nanocracks leads to cracking. It resulted in premature failure of ArH filler added welds as compared to Ar filler added welds. Thus, the flow stress of ArH welds is lower than Ar welds. Hydrogen addition to shielding gas has resulted in poor work hardening

Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Fig. 24. Dislocation Silp Bands (DSBs) observed in a) Ar autogenous b) Ar-Filler 1 c) Ar-Filler 2 d) ArH autogenous e) ArH-Filler 1 f) ArH-Filler 2.

18

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

CRediT authorship contribution statement [21]

N. Anbarasan: Conceptualization, Methodology, Investigation, Writing - original draft. S. Jerome: Supervision, Resources. S.G.K. Manikandan: Project administration, Funding acquisition.

[22] [23]

Acknowledgment

[24]

The authors would like to thank the Indian Space Research Organi­ zation Propulsion Complex, Mahendragiri, for supporting this research work under the RESPOND project grant. The authors are very grateful to the School of Mechanical Engineering, Vellore Institute of Technology, Vellore, for generously assisting in the Tensile Testing Facility.

[25] [26] [27]

Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.msea.2019.138874.

[28]

References

[29]

[1] C. Radhakrishna, K.P. Rao, S. Srinivas, Laves phase in superalloy 718 weld metals, J. Mater. Sci. Lett. 14 (1995) 1810–1812, https://doi.org/10.1007/BF00271015. [2] R. Cort�es, N.K. Rodríguez, R.R. Ambriz, V.H. L� opez, A. Ruiz, D. Jaramillo, Fatigue and crack growth behavior of Inconel 718–AL6XN dissimilar welds, Mater. Sci. Eng. A 745 (2019) 20–30, https://doi.org/10.1016/j.msea.2018.12.087. [3] M.C. Maguire, J.R. Michael, Weldability of alloy 718 , 625 and variants, Superalloys 718 (1994) 881–892, 625, 706 Var. Deriv. [4] K.-M. Chang, H.-J. Lai, J.-Y. Hwang, Existence of laves phase in Nb-hardened superalloys, Superalloys 718, 625, 706 var. Deriv (1994) 683–694, https://doi.org/ 10.7449/1994/Superalloys_1994_683_694. [5] S.M. Jones, D.J. Radavich, S. Tian, Effect of composition on segregation microstructures and mechanical properties of cast alloy 718, in: Superalloys 718 Metall. Appl, 1989, pp. 589–597, https://doi.org/10.7449/1989/Superalloys_ 1989_589_597. TMS. [6] C. Radhakrishnan, K. Prasad Rao, The formation and control of Laves phase in superalloy 718 welds, J. Mater. Sci. 32 (1997) 1977–1984, https://doi.org/ 10.1023/A:1018541915113. [7] S.G.K. Manikandan, D. Sivakumar, K. Prasad Rao, M. Kamaraj, Laves phase in alloy 718 fusion zone — microscopic and calorimetric studies, Mater. Char. 100 (2015) 192–206, https://doi.org/10.1016/j.matchar.2014.11.035. [8] K.R. Vishwakarma, N.L. Richards, M.C. Chaturvedi, Microstructural analysis of fusion and heat affected zones in electron beam welded ALLVAC® 718PLUSTM superalloy, Mater. Sci. Eng. A 480 (2008) 517–528, https://doi.org/10.1016/j. msea.2007.08.002. [9] G.D. Janaki Ram, A. Venugopal Reddy, K. Prasad Rao, G. Madhusudhan Reddy, Control of Laves phase in Inconel 718 GTA welds with current pulsing, Sci. Technol. Weld. Join. 9 (2004) 390–398, https://doi.org/10.1179/ 136217104225021788. [10] C.H. Jang, Weld Development for Incoloy Alloy 908 , A Low Thermal Expansion Superalloy (PhD Thesis), MASSACHUSETTS Inst. Technol., 1995. [11] M.M. Tavakkoli, S.M. Abbasi, Effect of molybdenum on grain boundary segregation in Incoloy 901 superalloy, Mater. Des. 46 (2013) 573–578, https://doi.org/ 10.1016/j.matdes.2012.11.002. [12] S. Manikandan, D. Sivakumar, M. Kamaraj, High temperature mechanical properties of cryogenically cooled alloy 718 weldments, Mater. A. T. High. Temp. 33 (2016) 257–269, https://doi.org/10.1080/09603409.2016.1182960. [13] G.D.J. Ram, a.V. Reddy, K.P. Rao, G.M. Reddy, Microstructure and mechanical properties of Inconel 718 electron beam welds, Mater. Sci. Technol. 21 (2005) 1132–1138, https://doi.org/10.1179/174328405X62260. [14] K. Sivaprasad, S. Ganesh Sundara Raman, Influence of weld cooling rate on microstructure and mechanical properties of alloy 718 weldments, Metall. Mater. Trans. A 39 (2008) 2115–2127, https://doi.org/10.1007/s11661-008-9553-y. [15] K. Sindo, Welding Metallurgy, 2007, https://doi.org/10.1016/j. theochem.2007.07.017. [16] T. Yuan, Z. Luo, S. Kou, Grain refining of magnesium welds by arc oscillation, Acta Mater. 116 (2016) 166–176, https://doi.org/10.1016/j.actamat.2016.06.036. [17] G.D. Janaki Ram, A.V. Reddy, K.P. Rao, G.M. Reddy, Improvement in stress rupture properties of inconel 718 gas tungsten arc welds using current pulsing, J. Mater. Sci. 40 (2005) 1497–1500, https://doi.org/10.1007/s10853-005-0590-2. [18] S.G.K. Manikandan, D. Sivakumar, K.P. Rao, M. Kamaraj, Effect of weld cooling rate on Laves phase formation in Inconel 718 fusion zone, J. Mater. Process. Technol. 214 (2014) 358–364, https://doi.org/10.1016/j.jmatprotec.2013.09.006. [19] Q.Y. Pan, W.D. Huang, Y.M. Li, X. Lin, Y.H. Zhou, The solidification microstructures and microsegregation behaviour of a nickel-base superalloy under ultra-high temperature gradient conditions produced by a laser beam, J. Mater. Sci. Lett. 15 (1996) 2112–2114, https://doi.org/10.1007/BF00241142. [20] W. Zhang, L. Liu, X. Zhao, T. Huang, Z. Yu, M. Qu, H. Fu, Effect of cooling rates on dendrite spacings of directionally solidified DZ125 alloy under high thermal

[30]

[31]

[32] [33] [34]

[35] [36]

[37] [38]

[39] [40] [41] [42]

[43] [44] [45]

[46]

19

gradient, Rare Met. 28 (2009) 633–638, https://doi.org/10.1007/s12598-0090121-4. J.F. Key, Anode/cathode geometry and shielding gas interrelationships in GTAW, Weld. J. 59 (1980), 364-s to 370-s. J.J. Lowke, R. Morrow, J. Haidar, A.B. Murphy, Prediction of gas tungsten arc welding properties in mixtures of argon and hydrogen, IEEE Trans. Plasma Sci. 25 (1997) 925–930, https://doi.org/10.1109/27.649597. G.A. Young, C.K. Battige, N. Liwis, M.A. Penik, J. Kikel, A.J. Silvia, C.K. McDonald, Factors Affecting the Hydrogen Embrittlement Resistance of Ni-Cr-Mn-Nb Welds, Niskayuna, NY, 2001, https://doi.org/10.2172/821694. N. Anbarasan, S. Jerome, N. Arivazhagan, Argon and argon-hydrogen shielding gas effects on the laves phase formation and corrosion behavior of Inconel 718 gas tungsten arc welds, J. Mater. Process. Technol. 263 (2019) 374–384, https://doi. org/10.1016/j.jmatprotec.2018.07.038. A.W.S. (AWS) A.C. on F.M. and A. Materials, Welding Consumables - Gases and Gas Mixtures for, second ed., American Welding Society (AWS) A5, 2011. https://www .iso.org/standard/39569.html. J. Lippold, Welding Metallurgy and Weldability, 2015. O.L. Rocha, C.A. Siqueira, A. Garcia, Heat flow parameters affecting dendrite spacings during unsteady-state solidification of Sn-Pb and Al-Cu alloys, Metall. Mater. Trans. A 34 (2003) 995–1006, https://doi.org/10.1007/s11661-003-02293. I. Woo, K. Nishimoto, K. Tanaka, M. Shirai, Effect of grain size on heat affected zone cracking susceptibility.Study of weldability of Inconel 718 cast alloy (2nd Report), Weld. Int. 14 (2000) 514–522, https://doi.org/10.1080/ 09507110009549223. R.E.W. Casselton, W. Hume-Rothery, The equilibrium diagram molybdenumnickel, J. Less Common. Met. 7 (1964) 212–221. J.D. WhITTENBERGER, A review of: "superalloys II", in: C.T. Sims, N.S. Stoloff, W. C. Hagel (Eds.), A Wiley-Interscience Publication John Wiley & Sons, vol. 7, 1992, pp. 463–468, https://doi.org/10.1080/10426919208947432. New York, NY 615 pages, hardcover, 1987, Mater. Manuf. Process. Y.L. Hu, X. Lin, X.B. Yu, J.J. Xu, M. Lei, W.D. Huang, Effect of Ti addition on cracking and microhardness of Inconel 625 during the laser solid forming processing, J. Alloy. Comp. 711 (2017) 267–277, https://doi.org/10.1016/j. jallcom.2017.03.355. J.W. Hooijmans, G. den Ouden, Hydrogen absorption during arc melting, Mater. Sci. Technol. 12 (1996) 81–85, https://doi.org/10.1179/mst.1996.12.1.81. A. Savas¸, V. Ceyhun, Finite element analysis of GTAW arc under different shielding gases, Comput. Mater. Sci. 51 (2012) 53–71, https://doi.org/10.1016/j. commatsci.2011.07.032. S.G.K. Manikandan, D. Sivakumar, K. Prasad Rao, M. Kamaraj, Microstructural characterization of liquid nitrogen cooled Alloy 718 fusion zone, J. Mater. Process. Technol. 214 (2014) 3141–3149, https://doi.org/10.1016/j. jmatprotec.2014.07.022. B. Dutta, M. Rettenmayr, An experimental investigation on the kinetics of solute driven remelting, Metall. Mater. Trans. A 31 (2000) 2713–2720, https://doi.org/ 10.1007/BF02830330. L. Wang, Y. Yao, J. Dong, M. Zhang, Effect of cooling rates on segregation and density variation in the mushy zone during solidification of superalloy inconel 718, Chem. Eng. Commun. 197 (2010) 1571–1585, https://doi.org/10.1080/ 00986445.2010.493101. M.J. Perricone, J.N. Dupont, Effect of composition on the solidification behavior of several Ni-Cr-Mo and Fe-Ni-Cr-Mo alloys, Metall. Mater. Trans. A 37 (2006) 1267–1280, https://doi.org/10.1007/s11661-006-1078-7. N. El-bagoury, T. Matsuba, K. Yamamoto, H. Miyahara, K. Ogi, Influence of heat treatment on the distribution of Ni 2 Nb and microsegregation in cast inconel 718 alloy NbC, Mater. Trans. 46 (2005) 2478–2483, https://doi.org/10.2320/ matertrans.46.2478. A. Mookerjee, T. Saha-Dasgupta, I. Dasgupta, A. Arya, S. Banerjee, G.P. Das, A firstprinciples thermodynamic approach to ordering in binary alloys, Bull. Mater. Sci. 26 (2003) 79–89, https://doi.org/10.1007/BF02712791. M.L. Martin, J.A. Fenske, G.S. Liu, P. Sofronis, I.M. Robertson, On the formation and nature of quasi-cleavage fracture surfaces in hydrogen embrittled steels, Acta Mater. 59 (2011) 1601–1606, https://doi.org/10.1016/j.actamat.2010.11.024. M.L. Martin, B.P. Somerday, R.O. Ritchie, P. Sofronis, I.M. Robertson, Hydrogeninduced intergranular failure in nickel revisited, Acta Mater. 60 (2012) 2739–2745, https://doi.org/10.1016/j.actamat.2012.01.040. S. Jothi, S.V. Merzlikin, T.N. Croft, J. Andersson, S.G.R. Brown, An investigation of micro-mechanisms in hydrogen induced cracking in nickel-based superalloy 718, J. Alloy. Comp. 664 (2016) 664–681, https://doi.org/10.1016/j. jallcom.2016.01.033. D.J. Senor, K.L. Peddicord, J.P. Strizak, Effects of hydrogen on the fracture morphology of Inconel 718, Mater. Lett. 11 (1991) 373–378, https://doi.org/ 10.1016/0167-577X(91)90137-U. A.R. Troiano, The role of hydrogen and other interstitials in the mechanical behavior of metals, Metallography, Microstruct.Anal. 5 (2016) 557–569, https:// doi.org/10.1007/s13632-016-0319-4. M. Yamaguchi, K.-I. Ebihara, M. Itakura, T. Kadoyoshi, T. Suzudo, H. Kaburaki, First-Principles study on the grain boundary embrittlement of metals by solute segregation: Part II. Metal (Fe, Al, Cu)-Hydrogen (H) systems, Metall. Mater. Trans. A 42 (2011) 330–339, https://doi.org/10.1007/s11661-010-0380-6. R.A. Oriani, P.H. Josephic, Equilibrium aspects of hydrogen-induced cracking of steels, Acta Metall. 22 (1974) 1065–1074, https://doi.org/10.1016/0001-6160 (74)90061-3.

N. Anbarasan et al.

Materials Science & Engineering A 773 (2020) 138874

[47] G.M. Pressouyre, Trap theory of Hydrogen embrittlement, Acta Metall. 28 (1980) 895–911, https://doi.org/10.1016/0001-6160(80)90106-6. [48] G.M. Pressouyre, I.M. Bernstein, A kinetic trapping model for hydrogen-induced cracking, Acta Metall. 27 (1979) 89–100, https://doi.org/10.1016/0001-6160(79) 90059-2. [49] X. Li, J. Zhang, E. Akiyama, Q. Fu, Q. Li, Hydrogen embrittlement behavior of Inconel 718 alloy at room temperature, J. Mater. Sci. Technol. 35 (2019) 499–502, https://doi.org/10.1016/j.jmst.2018.10.002. [50] G. Stenerud, R. Johnsen, J.S. Olsen, J. He, A. Barnoush, Effect of hydrogen on dislocation nucleation in alloy 718, Int. J. Hydrogen Energy 42 (2017) 15933–15942, https://doi.org/10.1016/j.ijhydene.2017.04.290. [51] P. Zhou, M.X. Huang, On the mechanisms of different work-hardening stages in twinning-induced plasticity steels, Metall. Mater. Trans. A 46 (2015) 5080–5090, https://doi.org/10.1007/s11661-015-3100-4. [52] U.F. Kocks, H. Mecking, Physics and phenomenology of strain hardening: the FCC case, Prog. Mater. Sci. 48 (2003), https://doi.org/10.1016/S0079-6425(02)000038, 171–273.

[53] Y.C. Lin, M.-S. Chen, J. Zhong, Prediction of 42CrMo steel flow stress at high temperature and strain rate, Mech. Res. Commun. 35 (2008) 142–150, https://doi. org/10.1016/j.mechrescom.2007.10.002. [54] G. Sjoberg, D. Cornu, Hydrogen embrittlement of cast alloy 718 effects of homogenization, grain size and delta-phase, in: Superalloys 718, 625, 706 Var, Deriv., TMS, 2001, pp. 679–690, https://doi.org/10.7449/2001/Superalloys_ 2001_679_690. [55] S. Martin, S. Wolf, U. Martin, L. Krüger, D. Rafaja, Deformation mechanisms in austenitic TRIP/TWIP steel as a function of temperature, Metall. Mater. Trans. A 47 (2016) 49–58, https://doi.org/10.1007/s11661-014-2684-4. [56] Z. Zhang, G. Obasi, R. Morana, M. Preuss, Hydrogen assisted crack initiation and propagation in a nickel-based superalloy, Acta Mater. 113 (2016) 272–283, https://doi.org/10.1016/j.actamat.2016.05.003. [57] J.P. Chateau, D. Delafosse, T. Magnin, Numerical simulations of hydrogen–dislocation interactions in fcc stainless steels, Acta Mater. 50 (2002) 1507–1522, https://doi.org/10.1016/S1359-6454(02)00008-3.

20