Effect of thermal aging on microstructure, hardness, tensile and impact properties of Alloy 617

Materials Science & Engineering A 710 (2018) 47–56

Contents lists available at ScienceDirect

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

Effect of thermal aging on microstructure, hardness, tensile and impact properties of Alloy 617

MARK

⁎

Aditya Narayan Singh, A. Moitra , Pragna Bhaskar, G. Sasikala, Arup Dasgupta, A.K. Bhaduri Metallurgy and Materials Group, Indira Gandhi Centre for Atomic Research, HBNI, Kalpakkam, Tamil Nadu 603102, India

A R T I C L E I N F O

A B S T R A C T

Keywords: Alloy 617 Thermal aging Tensile strength Microstructure Fractographs

Influence of thermal aging on microstructural evolution and its effect on hardness, tensile and impact properties have been studied for Alloy 617. Aging was carried out at 1023 K up to 20,000 h duration on a solution annealed as-received alloy. Detailed microscopic investigation of the alloy revealed the precipitation of γ´ phase (rich in Al and Ti) along with M23C6 carbides during aging treatments. Thermal aging has imparted increased hardness and strength, however with decreased ductility and toughness. Yield stress (YS), though initially increases up to 5000 h of aging, and has shown a decreasing trend from 5000 to 10,000 h. Further aging up to 20,000 h has shown an anomalous increase in YS. Charpy impact test for 20,000 h aged sample showed a reduction in fracture energy (~ 85%) w.r.to as-received material along with a shift in fracture mode from transgranular ductile to intergranular brittle. The change in hardness and YS in aged Alloy 617 has been attributed to the aging-induced evolution of γ´ precipitates. The intergranular fracture associated with lower impact energy, as observed after prolonged aging duration, has been attributed to the grain boundary embrittlement originating from precipitation of M23C6 carbides at the grain boundaries.

1. Introduction Globally, the fossil-fired power plants are facing technological challenges to maintain a critical balance between the ever increasing demands of electricity generation while simultaneously addressing the global warming issues by ensuring lower greenhouse gas (CO2) emission. Towards doubling the electrical energy generation in coming decades [1] with aforesaid constrains, steam temperature between 973–1033 K and a pressure of 35 MPa have been envisaged in fossilfired power plants. It is imperative that higher the steam temperature, the higher is the efficiency [2,3] of the plants. To this end, proper selection of the materials for the high-temperature components is mandatory to ensure component's integrity during the service conditions. In this domain, the Ni-based superalloys as a group of materials are showing acceptable physical, mechanical and chemical properties. One such potential candidate material is Alloy 617, which can adequately operate at temperatures well above 1073 K [4]. Alloy 617 is a class of solid-solution strengthened, austenitic [5] and tungsten-free superalloy [6]. It is expected to retain its creep strength well above 1073 K /35 MPa [7] and has excellent oxidation-reduction resistance [8] even at very high-temperature. Alloy 617 has a face-centered cubic (FCC) structure [9], which is believed to draw its strength mainly from the solid solution

⁎

strengthening mechanisms owing to elements like Mo and Co. However, during the exposure at higher temperature regime, formation of intermetallic compounds of type (Ni3 (Ti, Al)), commonly known as γ´ precipitates have been widely reported in literatures [4,10]. In addition to this, there are reports for the formation of Cr-rich M23C6 carbides, often segregated at the grain boundaries after longer duration of aging [11]. This indicates towards aging induced microstructural evolutions, which would ultimately influence the mechanical properties of this alloy after long term exposures to the service temperatures. Though this alloy might possess acceptable mechanical properties in the as-received condition, the effect of thermal aging on important mechanical properties in consequence to the microstructural evolution is always a matter of concern for the designers. Benz et al. [12] has reported that this alloy has a very high propensity for precipitation and microstructures evolution occurs even in aging for one hour at 1023 K. Though there have been efforts in literature to evaluate the mechanical properties, they are often restricted to the shorter duration of aging treatments [4,13–16]. Kirchhöfer et al. [13] has reported an increase in the hardness from 160 to 240 HV after an aging treatment on 973 K/ 1000 h. Ren et al. [14] has also reported that this alloy achieved the same level of hardness even for a shorter duration of aging, i.e for 973 K/100 h. For the tensile tests, Guo et al. [15] has reported an increase in yield stress (YS) up to 250 MPa for the aging condition on

Corresponding author. E-mail address: [email protected] (A. Moitra).

http://dx.doi.org/10.1016/j.msea.2017.10.078 Received 20 June 2017; Received in revised form 8 September 2017; Accepted 24 October 2017 Available online 25 October 2017 0921-5093/ © 2017 Elsevier B.V. All rights reserved.

Materials Science & Engineering A 710 (2018) 47–56

A. Narayan Singh et al.

2. Experimental

Table 1 Chemical composition (wt%) of Alloy 617 employed in this study. Elements wt% Elements wt%

Ni Bal. Cu < 0.01

Cr 22.1 Si < 0.1

Co 11.6 S 0.001

Mo 9.4 Ti 0.4

Fe 0.12 N 0.005

Mn < 0.01 Nb 0.02

Al 1.2 Vd < 0.05

2.1. Materials

C 0.06 B 0.0036

The chemical makeup of the Alloy 617 used in this study is given in Table 1. The as-received alloy was cold finished, solution annealed at 1443 K and water quenched to retain a single phase FCC structure prior to aging. The specimens were received in tubular form with following dimensions: inner diameter (ID) of 28.1 ± 0.25 mm, a thickness of 11.9 ± 0.25 mm and a length of 700 mm. The thermal aging of the tubes was carried out in a muffle furnace. The aging conditions were maintained at 1023 K ± 2 K for 1000, 5000, 10000 and 20,000 h respectively followed by air cooling. 2.2. Characterization of microstructure, hardness, tensile and impact energy Microstructural studies were carried out on as-received and thermal aged materials by light microscope (LM). For LM studies, the samples were prepared using 1 µm diamond polishing after the conventional mechanical polishing to obtain a mirror finished surface. For etching the polished surfaces, Aqua-regia (3HCL:1HNO3) was used for the asreceived one. For the aged material, electrolytic etching with a solution of 60% HNO3 and 40% H2O was carried out at room temperature (RT), using 1.5 V DC for time durations of 5–120 s. Thin foils up to 80 µm thicknesses used for carrying out transmission electron microscope (TEM) studies of precipitates evolved during the aging treatments were also sectioned from the Charpy impact tested specimens with an initial dimension of 10 × 10 mm2. Subsequently, discs of diameter 3 mm coupons were punched out to follow twin-jet electropolishing. The electrolyte used for jet thinning was 20% Perchloric acid (HClO4) in a Struers Tenupol-5® at a temperature of about 243 K ( ± 1 K) at 15 V. The identification of phases and chemical characterization of precipitates were carried out by a combination of selected area electron diffraction (SAED) and energy dispersive spectroscopy (EDS) analysis.

Fig. 1. Schematic representation of a miniaturized tensile specimen with given dimensions in mm.

1033 K/1000 h, with a decreasing trend for further aging treatments up to 3000 h. Nanstad et al. [16] reported a drop of 36% degradation of impact energy even after aging on 1023 K/ 200 h w.r. to 250 J for asreceived specimen. Guo et al. [15] have also reported a significant reduction in the impact energy for the first 300 h of aging at 1033 K as compared to the as-received material. In spite of these efforts, an unequivocal conclusion regarding the role of microstructural evolution and the relevant mechanical properties is yet to be reached, especially after the long aging treatment. This paper envisages interpreting the mechanical properties like hardness, tensile behaviour, and Charpy impact energy of an Alloy 617 in the light of the microstructural evolution after aging treatments up to 20,000 h.

Fig. 2. Schematic representation of a Charpy impact specimen with given dimensions in mm.

Fig. 3. Light micrograph of Alloy 617: (a) Microstructure of as-received condition, (b) Frequency distribution of the grain size in Alloy 617 vs ASTM number for the as-received condition.

48

Materials Science & Engineering A 710 (2018) 47–56

A. Narayan Singh et al.

Fractographic analysis on broken specimens after the Charpy impact tests was carried out using scanning electron microscope (SEM) backed by energy dispersive spectroscopy (EDS) in secondary electron (SE) mode. The samples were thoroughly washed with acetone and dried completely before examining in the SEM.

Hardness of the alloy in as-received and aged conditions were measured using Vickers hardness (HV) tester using a load of 10 kg as per the standard procedure described in ASTM E92-16 [17]. Sub-size flat tensile test specimens (Fig. 1) were fabricated in compliance with ASTM E8 [18] standards. Tensile tests were carried out in the air in a floor model Hung Ta-2402 universal testing machine (UTM). Tests were performed at 298 K ( ± 2 K) using nominal strain rate of 3× 10−3 s−1. Impact toughness for the as-received and aged alloys were evaluated using a pendulum type 358 J capacity Charpy impact testing machine at the RT (298 K). Full size Charpy V-notch specimens (W x B x L = 10×10×55 mm) with a 2 mm 45° V-notch (Fig. 2) of root radius of 0.25 mm were used in this regard following the standard procedures described in ASTM E-23-16b [19].

3. Results and discussion 3.1. Microstructure 3.1.1. Light Microscope results 3.1.1.1. As-received Alloy 617. Figs. 3(a) and (b) shows light micrograph of the as-received (Solution treated condition) Alloy 617 and the corresponding frequency distribution profile of the grain size

Fig. 4. Microstructure of Alloy 617 obtained at 100X for specimens aged at 1023 K (a) 1000 h, (b) 5000 h, (c) 10,000 h, (d) 20,000 h, and (e) Frequency distribution vs ASTM grain size number for different aged specimens.

49

Materials Science & Engineering A 710 (2018) 47–56

A. Narayan Singh et al.

3.1.2. TEM results for the as-received and thermally aged Alloy 617 Figs. 5a to c shows the TEM micrograph of the as-received alloy, the selected area electron diffraction (SAED) pattern and EDS spectrum from precipitates respectively. The TEM investigations of these precipitates in the Alloy 617 were found to be Cr-rich M23C6. These results were also in conformity with results obtained by Mankins et al. [4] The results have been validated by the SAED pattern, as shown in Fig. 5(b). The EDS spectrum in Fig. 5(c) has confirmed the reported elemental composition. TEM study for thermally aged specimen for 10,000 h has earlier been carried out by the authors and reported by Aditya et al. [11], however for the new readers to appreciate the evolution of shape, size and morphology of γ′ from as-received to aged one and also to maintain the continuum they are shown in Figs. 6(a) to (d). The results confirm the aging induced occurrence of fresh precipitates of Ni3 (Al, Ti) type, which is commonly known as γ′ precipitates [20]. The morphology of γ′ precipitates in 1000 h aged condition has been noted to be spheroidal and is within the size range of ~ (40 – 70 nm) (Fig. 6(a)). A similar kind of observation has also been reported in work carried out by Krishna et al. [21]. It is well appreciated that the presence of γ′ precipitates in the matrix of an aged alloy of this material plays a crucial role in imparting strength at higher temperature [22]. It has been concluded from the TEM investigations that the morphology of γ′ precipitates hardly changes during the aging treatments. However, there are indications that it only grows in size due to thermal aging, but to a limited extent. In the Fig. 6(b), after the 10,000 h aging treatment, a marginal growth

respectively. It shows polygonal grains structure with a duplex nature in grain size distribution. Generally the matrix contains grain sizes varying from 40 to 100 µm. However, there has been occurrence of smaller grains with sizes varying from 30 to 40 µm and often in the triple point region. The frequency distribution chart reveals that majority of grain size lies in the range of ASTM number 4 − 6. Grain boundaries have been found to be thicker with little serrations. There is presence of annealing twins, as indicated Fig. 3(a). Similar observations on existence of annealing twins in this class of materials have also been reported in literature [8,19].

3.1.1.2. Microstructures of thermally aged Alloy 617. Light micrographs of the thermally aged Alloy 617 for 1000 h, 5000 h, 10000 h and 20,000 h are shown in Figs. 4a to d respectively. Towards studying the effect of aging on the grain size, a detailed statistical study of the grain size distribution up to 10,000 h of aging has recently been carried out by the authors and reported by Aditya et al. [11]. However, for the sake of continuity and convenience of the readers, the frequency distribution w.r.to the ASTM grain size numbers for all the above mentioned aging treatments including the 20,000 h aging are shown in Fig. 4(e). It may be observed from the frequency distribution chart that the aging treatments have only marginally influenced the grain size distributions for the Alloy 617 and still the majority of grains lies in ASTM number 4–6. The annealing twins, as observed in the as received material, have been found to be absent in the aged samples.

Fig. 5. Bright field TEM image of Alloy 617: (a) Bright field TEM micrograph of as-received condition, (b) SAED pattern confirming that they correspond to M23C6, and (c) The corresponding EDS spectrum for (a).

50

Materials Science & Engineering A 710 (2018) 47–56

A. Narayan Singh et al.

Fig. 6. Bright field TEM image and SAD pattern of Alloy 617 aged at 1023 K (a) Spheroidal γ′ in 1000 h aged specimen (b) Morphology in 10,000 h aged samples still remains same but with a larger size (c) Corresponding SAED pattern for Ni3 (Al, Ti), and (d) The corresponding EDS spectrum of Ni3 (Al, Ti).

in the size of the γ′ precipitates (to 80–110 nm) has been noticed. It could be appreciated that even this marginal change in the precipitate size may have significant influence on the deformation characteristics of this material. The results have been validated by the SAED pattern, as shown in Fig. 6(c). The EDS spectrum in Fig. 6(d) has confirmed the elemental composition of γ′ to be Ni-rich Ni3 (Al, Ti). TEM results of thermally aged specimens for 10000 h has also shown the M23C6 precipiates of size 70–80 nm as shown in Fig. 7(a). SAED pattern along the [011] zone axis as shown in Fig. 7(b), confirms the presence of Cr-rich M23C6. EDS analysis shows that the precipitates are rich in Cr and Mo.

displayed its lowest hardness value of ~ 196 HV, whereas the aged alloy shows improved hardness even after aging duration of 1000 h. The hardness of this alloy increases consistently for the aging treatment up to 5000 h. This could be attributed to the simultaneous occurrence of the M23C6 and γ´ precipitates leading to the overall strengthening of the matrix. There is a marginal drop in hardness between the aging treatment of 5000 h and 10,000 h. For the aging treatment from 10,000 to 20,000 h, it is tending towards saturation at around 250 HV. The marginal drop in hardness between the 5000 h and 1000 h aging durations needs further attention. It is well known that the matrix strength is a function of precipitate type, size, shape and morphology. In the present circumstances, the strength of the matrix would be ultimately governed by the simultaneous presence of M23C6 and Ni-rich Ni3 (Al, Ti) or γ′ precipitates, their respective sizes, shapes and distributions. The TEM results, as reported earlier, indicate towards an aging induced increase in size for the γ′ precipitates though a marginal, whereas the size of the M23C6 has not shown to be increased. Thus it may be concluded that the aging induced strength of the matrix is ultimately dependent on the size and distribution of the γ′ precipitates alone. After the 10,000 h aging treatment, a marginal growth in the size of the γ′ precipitates (from 40 −70 nm to 80–110 nm) has been noticed, as discussed earlier. It is expected that unless fresh precipitates

3.2. Hardness of Alloy 617 Hardness of the Alloy 617 in its as-received and aged conditions were measured at RT by a Vickers hardness tester with a dwell time of 15 s at a 40X magnification, and the results are reported in Fig. 8. A minimum of 15 points were considered for each hardness measurement for different aged conditions and error bands were calculated. The data reported in the manuscript falls in 3% error band and hence it can be clearly seen from the hardness data that there exists a clear trend in the hardness of Alloy 617 over aging. In the as-received condition, alloy 51

Materials Science & Engineering A 710 (2018) 47–56

A. Narayan Singh et al.

Fig. 7. Bright field TEM image and SAED pattern of Alloy 617 aged at 1023 K for 10,000 h:(a) Bright field TEM image showing precipitates of size 80–100 nm (b) SAED pattern confirming that they correspond to M23C6 (c) The corresponding EDS spectrum for (a).

form, the increase of size would lead to higher inter particle spacing, leading to loss of matrix strength [23]. This attributes to the marginal drop in hardness between the aging treatments of 5000–10,000 h. 3.3. Tensile properties of as received and aged alloy 617 The tensile tests on the Alloy 617 for the as-received and thermally aged conditions have been carried out at the RT (298 K) at a nominal strain rate of 3×10−3 s−1. The Engineering stress vs engineering strain plot and the true stress vs true plastic strain plots for the as received and aged conditions are shown in Fig. 9(a) and (b) respectively. From the Fig. 9(a) and (b), it is evident that the thermal aging has made significant influences towards the deformation characteristics of this alloy. Tensile tests was carried out on minimum of three samples and the data reported was in 3% error band. Due care has been taken for the data measurements and the scatter band. Variations of yield stress (YS) and % tensile elongation (TE) with aging duration of the Alloy 617 in both as-received and aged conditions are shown in Fig. 9(c) and (d) respectively. Yield stress (YS), as shown

Fig. 8. Profile of Hardness of the Alloy 617 with aging time.

52

Materials Science & Engineering A 710 (2018) 47–56

A. Narayan Singh et al.

Fig. 9. Tensile curves for Alloy 617 (a) Showing the variations in Engineering Stress-Strain curve for different aged conditions (b) True Stress-Strain plot from yield point (yp) to ultimate tensile strength (uts) for different conditions of aging duration (c) Variation of yield strength with aging duration, and (d) Variation of % of elongation.

The yield stress is defined as the nominal stress level at which bulk plastic deformation is initiated in the material, and is commonly attributed to the initiation of gross dislocation movements in the crystal lattice. It is a well acceptable fact that the type, morphology, shape and distribution of the precipitates influence the flow of dislocation to a significant extent, thus making the yield stress a very sensitive deformation parameter to their changes. Thus, the observed variation of yield strength due to the different aging durations needs to be discussed in this light. Further, the effect of stacking fault energy (SFE) often plays a crucial role in the deformation mechanisms, mostly reflected in the variation of yield stress. Lower stacking fault energy would lead to a lower probability of cross slip of dislocations, and rather confining the movements of dislocations on the slip planes alone. The Alloy 617, in solution annealed condition, can be termed as an alloy with low stacking fault energy (SFE) with SFE ~ 30 − 40 mJ/m2 [24–26] The presence of annealing twins in the as received alloys also indicates towards its lower SFE [27]. This indicates that the Alloy 617 is more prone to planner slip in deformation, where the motion of dislocations are more strongly influenced by the nature and distribution of the of precipitates, as compared to the higher SFE materials with higher tendency for cross slipping. However, the absence of annealing twins in the aged microstructures (Figs. 4a to d) indicates towards an increase in SFE due to aging induced precipitation and re distribution of different precipitates. However this needs further validation and is beyond the scope of this paper. The TEM results shows that the evolution of new precipitates in the form of γ′ has started at the initial stages of the aging itself. Considering

Fig. 10. Effects of thermal aging on variation of Charpy impact energy for the Alloy 617.

in Fig. 9(c), initially increases up to 5000 h of aging, then decreases in the 5000 – 10,000 h aging duration and finally shows a saturation/ marginal increase in the regime of 10,000 – 20,000 h aging duration. It has been observed that unlike the yield stress variation, the tensile ductility, as measured as the % total elongation (TE) consistently decreases with the aging duration, as shown in Fig. 9(d).

53

Materials Science & Engineering A 710 (2018) 47–56

A. Narayan Singh et al.

Fig. 11. SEM fractographs of failed as-received Alloys 617 during Charpy impact test (a) Near to notch tip (b) Middle of specimen, and (c) Far end of notch tip.

Fig. 12. SEM fractographs of failed specimen unveiling the effect of aging duration on Charpy impact test for the sample aged at 1023 K for 1000 h (a) Near to notch tip (b) Middle of specimen, and (c) Far end of notch tip.

Fig. 13. SEM fractographs of failed specimen unveiling the effect of aging duration on Charpy impact test for the sample aged at 1023 K for 5000 h (a) Near to notch tip (b) Middle of specimen, and (c) Far end of notch tip.

Fig. 14. SEM fractographs of failed specimen unveiling the effect of aging duration on Charpy impact test for the sample aged at 1023 K for 10,000 h (a) Near to notch tip (b) Middle of specimen, and (c) Far end of notch tip.

54

Materials Science & Engineering A 710 (2018) 47–56

A. Narayan Singh et al.

Fig. 15. (a) Micrographs of 1000 h aged specimen showing the presence of carbides decorating the grain boundaries.(b) EDS analysis of the precipitates confirming the presence of Cr and Mo in the M23C6.

strength. It has been that after a aging treatment of 5000 h, grain boundaries gets embrittled and this has been attributed to the aging induced M23C6 precipitations in the grain boundary region [11]. Correlating to the above mentioned results, it can be concluded that the grain boundary embrittlement is the major cause in aging induced loss of ductility.

the material to be prone to planner slip due to moderate SFE, the fresh precipitates are expected to inhibit the planner motion of dislocations to a significant extent. This is expected to increase the matrix strength, and raise the yield strength to a significant extent. This explains the stiff rise of yield stress (up to ~625 MPa) of this material up to the aging duration of 5000 h. However, further aging treatment up to 10,000 h shows an anomalous behaviour, as the yield stress drops to ~ 525 MPa. Roy et al. [28], Wang et al.[29], and Wang et al. [30] have also reported the similar behaviour of yield stress variations during similar aging treatments. It is interesting to note that during this aging duration, there has been a decreasing trend in the hardness too. As discussed earlier, after the 10000 h aging treatment, a marginal growth in the size of the γ′ precipitates (from 40 −70 nm to 80–110 nm) has been noticed. Coarsening of existing precipitates is always known to be accompanied with the loss in solid solution strength of the material. Further, in the absence of formation of fresh precipitates, it can be argued that the increase of size would lead to higher inter particle spacing. This means, for the planner slip to continue, the dislocations would receive lesser resistance from the matrix as the shear stress required to move a dislocation is known to vary inversely with the inter particle spacing [23]. It has also been noted earlier that the absence of annealing twins in the aged micrographs indicates towards an aging induced increase in SFE. This would give rise to a change in mode in the dislocation movement, i.e. pro planner slip mode for the as received one to pro cross slip mode for the aged ones. This indicates that even if fresh precipitates form during the aging treatments during 5000–10,000 h, their influence on the dislocation movements would be less significance as the dislocations would follow an easy path to escape to parallel planes by cross slipping. Thus it is reasonable to argue that during the aging duration of 5000–10,000 h the material undergoes a softening owing to the combined effect of increase in size of the γ′ precipitates and the increase in SFE in this regime. This explains the anomalous drop in yield strength of this material during the aging treatment of 5000 h and 10,000 h. Further, a saturation/marginal increase of yield stress up to 20,000 h of aging can be attributed to a saturation of strength due to the trade off between the ongoing hardening and softening process. The loss in ductility with the aging duration is clearly seen from the Fig. 9(d). It is generally argued that in alloys yield strength of a material bears inverse relation with the ductility. This argument also satisfactorily applies to this alloy till 5000 h aged conditions. This inverse relationship fails to follow in the aging duration between 5000 and 10,000 h where decrease in yield strength is also followed by a simultaneous decrease in ductility. This indicates that the loss of ductility is governed by a separate mechanism other than influencing the yield

3.4. Impact properties The variation of Charpy impact energy (J) with aging duration (h) is shown in Fig. 10. The Charpy impact energy values falls in the scatter band of less than 3% which is well above acceptance, a systematic trend of degradation with the aging time can be noticed. The material shows high impact energy (~300 J) in the as-received condition, but a sharp reduction in impact energy even after 1000 h of aging have been noticed. The toughness values of this alloy for further aging have shown a gradual drop, and after 20,000 of aging, a reduction of ~ 86% w. r. to as-received material has been noticed. To report on the Charpy impact energy a minimum of 10 data values were obtained and they were further analysed with their error band before reporting in the manuscript. From the experimental data obtained from Charpy impact energy one can clearly perceive that aging has profound impact on Alloy 617. 3.4.1. Fractographic investigation of impact specimen In order to corroborate the impact failure observation of the Alloys 617 on thermal aging, fractographic investigation have been carried out using SEM for the regions a) near to notch, b) middle of the specimen and c) far away from notch tip. The results as obtained from the as received and aged materials up to 10,000 h, are shown in Figs. 11–14. The fractographic observations from Figs. 11–14 clearly reveal a change of fracture mode from predominantly ductile one to predominantly intergranular fracture, as the aging time progresses from asreceived to 10,000 h. For the as received one, the presence of substantial fibrous dimples in all the regions corroborates well with its high impact energy. The brittle intergranular characteristics get more dominant on aging as seen from Figs. 12–14. This observation corroborates well with the severe loss of impact energy after aging. It is wellestablished fact that the Charpy impact fracture energy at room temperature of any material, as determined by conventional Charpy test, is a function of two variable components as a) energy required for crack initiation, b) energy required for crack propagation. Both of these components to fracture energy are significantly influenced by the inherent microstructural features evolved during aging condition. The crack initiation energy is by and large affected by the nature, size, and 55

Materials Science & Engineering A 710 (2018) 47–56

A. Narayan Singh et al.

embrittlement of Alloy 617.

shape of the precipitates which are continually being evolved during the aging at high-temperature for this material, whereas the crack propagation energy is often controlled by the grain size distribution and the grain boundary strength. Further, the strength of the matrix often plays a crucial role, as it is argued that a stronger matrix keeps a propagating crack sharper, thus demanding less energy to propagate. Figs. 12(a) and 13(c) shows the initiation of cracks near the grain boundaries. The intergranular nature of crack propagation is even more prominent in the 5000 and 10,000 h of aging. Recent investigations have revealed that the segregation of M23C6 carbides at the grain boundaries is the major reason towards the aging induced embrittlement of Alloy617 [11]. Fig. 15(a) shows the SEM micrographs taken from the polished and etched specimens of 1000 h aged condition. The EDS pattern, as shown in Fig. 15(b) confirms the precipitates to be rich in Cr and Mo, and it is concluded that the observed precipitates are M23C6 type carbides. Thus it is evident that even after this short aging duration, a network of M23C6 precipitates forms in the grain boundary region, and it is the major reason for embrittling the Alloy 617 after longer aging durations. Another important contribution towards fracture energy variation usually comes from the grain size distribution over aging. In our material Alloy 617, we found no or little change (Fig. 4(e)) in grain size distribution over aging [11]. So for simplicity we have ommitted out the discussion pertaining to grain size on fracture energy degrdation over aging.

Acknowledgement The authors highly acknowledge the experimental support provided by Ms. Paneer Selvi, Shri R. Balakrishnan and Smt. M. Sreevidya of IGCAR. The author Aditya Narayan Singh gracefully acknowledges the junior research fellowship awarded by IGCAR. References [1] D. Tytko, P.-P. Choi, J. Klöwer, A. Kostka, G. Inden, D. Raabe, Acta Mater. 60 (2012) 1731–1740. [2] C.C. Laranjeiras, S.I. Portela, Phys. Ed. 51 (2016) 055013. [3] J. Butler, Nature 116 (1925) 607–608. [4] W. Mankins, J. Hosier, T. Bassford, Metall. Trans. 5 (1974) 2579–2590. [5] C.T. Sims, A history of superalloy metallurgy for superalloy metallurgists, 1984. [6] R.L. Plaut, C. Herrera, D.M. Escriba, P.R. Rios, A.F. Padilha, Mater. Res. 10 (2007) 453–460. [7] W.-G. Kim, S.-N. Yin, G.-G. Lee, Y.-W. Kim, S.-J. Kim, Int. J. Pres. Ves. Pip. 87 (2010) 289–295. [8] K. Hrutkay, Nuclear Engineering, University of South Carolina, 2013. [9] R.T. Dewa, S.J. Kim, W.G. Kim, E.S. Kim, Metals 6 (2016) 178. [10] J. Hosier, D. Tillack, Met. Eng. Quart. 12 (1972) 51–55. [11] A.N. Singh, A. Moitra, P. Bhaskar, G. Sasikala, A. Dasgupta, A.K. Bhaduri, Metall. Mater. Trans. A 48 (2017) 3269–3278. [12] J. Benz, T. Lillo, R. Wright, Report INL/EXT-12-27974, 2013. [13] H. Kirchhöfer, F. Schubert, H. Nickel, Nucl. Technol. 66 (1984) 139–148. [14] W. Ren, R. Swindeman, ASME 2006 Pressure Vessels and Piping/ICPVT-11 Conference, American Society of Mechanical Engineers, 2006, pp. 489–500. [15] Y. Guo, B. Wang, S. Hou, Acta Metall. Sin. (Engl. Lett.) 26 (2013) 307–312. [16] R.K. Nanstad, M.A. Sokolov, X.F. Chen, Energy Technol. 2012: Carbon Dioxide Manag. Other Technol. (2012) 341–356. [17] ASTM E92-16, Standard Test Methods for Vickers Hardness and Knoop Hardness of Metallic Materials, ASTM International, West Conshohocken, PA, 2016. [18] ASTM E8/E8M-16a, Standard Test Methods for Tension Testing of Metallic Materials, ASTM International, West Conshohocken, PA, 2016. [19] ASTM E-23-16b, Standard Test Methods for Notched Bar Impact Testing of Metallic Materials, ASTM International, West Conshohocken, PA, 2016. [20] J.C. Lippold, S.D. Kiser, J.N. DuPont, Welding metallurgy and weldability of nickelbase alloys, John Wiley & Sons, 2011. [21] R. Krishna, H.V. Atkinson, S.V. Hainsworth, S.P. Gill, Metall. Mater. Trans. A 47 (2016) 178–193. [22] T.M. Smith, B.D. Esser, N. Antolin, A. Carlsson, R.E.A. Williams, A. Wessman, T. Hanlon, H.L. Fraser, W. Windl, D.W. McComb, M.J. Mills, Nat. Commun. 7 (2016) 13434. [23] G.E. Dieter, D.J. Bacon, Mechanical metallurgy, McGraw-Hill New York, 1986. [24] F. Pettinari, J. Douin, G. Saada, P. Caron, A. Coujou, N. Clement, Mater. Sci. Eng.: A 325 (2002) 511–519. [25] C. Cui, C. Tian, Y. Zhou, T. Jin, X. Sun, Dynamic strain aging in Ni base alloys with different stacking fault energy, Superalloys 2012, TMS Seven Springs, PA, USA, 2012, pp. 715–722. [26] C. Wang, C.-Y. Wang, Surf. Sci. 602 (2008) 2604–2609. [27] L. Tan, K. Sridharan, T.R. Allen, J. Nucl. Mater. 371 (2007) 171–175. [28] A.K. Roy, V. Marthandam, Mater. Sci. Eng.: A 517 (2009) 276–280. [29] J. Wang, L. Zhou, L. Sheng, J. Guo, Mater. Des. 39 (2012) 55–62. [30] C. Wang, Y. Guo, J. Guo, L. Zhou, Mater. Sci. Eng. A 670 (2016) 178–187.

4. Conclusions Based on the detailed studies of the effect of thermal aging on the hardness, tensile behaviour and Charpy impact of Alloy 617 and their correlation with microstructure, following conclusions have been made: 1. Thermal aging of Alloy 617 is accompanied with precipitation of γ′ precipitates, which strongly influence the hardness and tensile properties of this alloy after different aging durations. 2. The absence of annealing twins in the aged specimens indicates towards an aging induced increase of stacking fault energy. 3. The anomalous decrease in yield stress between 5000 h and 10000 h aging duration is attributed to the ease of dislocation movement in this regime owing to the combined effect of coarsening of γ′ precipitates and the rise in stacking fault energy. 4. Thermal aging has shown a substantial decrease in the impact energy, and this is attributed to the aging induced embrittlement of the alloy. This same has been reasoned to explain the loss of ductility after aging treatment. 5. The segregation of M23C6 carbides at the grain boundaries has been identified as the major reason towards the aging induced

56