Microstructure evolution and fracture mechanism of a novel 9Cr tempered martensite ferritic steel during short-term creep

Author’s Accepted Manuscript Microstructure evolution and fracture mechanism of a novel 9Cr tempered martensite ferritic steel during short-term creep Bo Xiao, Lianyong Xu, Lei Zhao, Hongyang Jing, Yongdian Han, Zhengxin Tang www.elsevier.com/locate/msea

PII: DOI: Reference:

S0921-5093(17)31254-6 http://dx.doi.org/10.1016/j.msea.2017.09.086 MSA35553

To appear in: Materials Science & Engineering A Received date: 9 June 2017 Revised date: 18 September 2017 Accepted date: 19 September 2017 Cite this article as: Bo Xiao, Lianyong Xu, Lei Zhao, Hongyang Jing, Yongdian Han and Zhengxin Tang, Microstructure evolution and fracture mechanism of a novel 9Cr tempered martensite ferritic steel during short-term creep, Materials Science & Engineering A, http://dx.doi.org/10.1016/j.msea.2017.09.086 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Microstructure evolution and fracture mechanism of a novel 9Cr tempered martensite ferritic steel during short-term creep Bo Xiaoa,b, Lianyong Xua,b*, Lei Zhaoa,b, Hongyang Jinga,b, Yongdian Hana,b, Zhengxin Tanga,b a School of Materials Science and Engineering, Tianjin University, Tianjin 300072, China; b Tianjin Key Laboratory of Advanced Joining Technology, Tianjin 300072, China. *Corresponding author. Tel./fax: +86 22 27402439. E-mail address: [email protected].

Abstract In this work, the microstructure evolution and fracture mechanism of a novel 9% chromium tempered martensite ferritic steel G115 were investigated over the temperature range of 625 to 675 °C using uniaxial creep tests. The creep curves consist of a primary transient stage followed by an apparent secondary stage, and an accelerated tertiary creep regime. The relationship between the minimum creep rate and the applied stress followed Norton's power law. Based on the EBSD analysis, there were no obvious textural features formed after creep deformation, and with the increase in creep time, the number of subgrains slightly increased, and then sharply increased, indicating dynamic recrystallization (DRX) occurs after creep deformation. In addition, three types of precipitates can be observed after creep deformation: W-rich Laves phase, Nb-rich MX, and Cu-rich precipitates. The Nb-rich MX with a square shape and Cu-rich precipitates with an ellipsoidal shape remain very stable. However, the W-rich Laves phases distributed mainly on the grain boundaries have rod-like, chain-like, and bulky shape, which are coarsened significantly. Representative fractographs of the G115 steel after creep deformation exhibit significant necking with an elliptical shape. A dense array of deep and equiaxed dimples appear in the central region under the tested creep conditions. Ductile fracturing is the dominant fracture mechanism during short-term creep deformation. Keywords: G115 steel; Microstructure evolution; Precipitates; Fracture mechanism. 1. Introduction Traditional coal-fired power plants lead to the continual consumption of fossil resources and the deterioration of the environment through significant pollution [1–4].

To

conserve

energy

and

protect

the

environment,

the

construction

of

ultra-supercritical (USC) power plants can contribute to mitigating the pollution caused by power plants by means of increasing steam temperature and pressure [5,6]. However, traditional martensitic creep-resistant steels such as P92 [7–9] and P122 steels [10–12] cannot be used under such severe conditions. Hence, novel materials with superior creep properties and oxidation resistance are urgently needed. To meet these challenges, a novel tempered martensitic creep-resistant 9% chromium steel G115 was developed with excellent creep properties at elevated temperatures, which was first developed by the China Iron and Steel Research Institute (CISRI) [13–17]. Compared to P92 steel, Mo was replaced by W and the proportion of B and N was optimized to suppress coarsening of the M23C6 precipitates. Cu was also added to further enhance the precipitation strengthening effect. Thus, this steel is expected to be used in USC power plants which can reach operating temperatures of up to 650 °C. To date, many studies have been performed on G115 steel, investigating the effect of heat treatments on the strength [13,14], evolution of the microstructure, toughness during aging [15,16], the tensile behavior during the high-temperature tensile test [17], and many other properties [18]. Furthermore, to accurately evaluate the structure integrity of the plants, the creep behavior of this steel must be considered. However, few studies have been performed on creep behavior of G115 steel. Based on the aforementioned issues, the aim of this study is to investigate the relationship between microstructure and creep rupture behavior under various creep conditions. Based on the experimental results, the creep behavior of G115 steel was analyzed in detail. Furthermore, to understand the creep rupture mechanism of G115 steel, the fracture surface characterizations and the microstructures of the fracture frontiers of the ruptured specimens were also studied. In addition, the morphology, distribution, and size of the precipitates after creep at 625, 650, and 675 °C were also investigated. 2. Experimental procedure 2.1. Material

In this study, a novel 9Cr martensite ferritic steel, namely G115, was chosen as the experimental material. The chemical composition of this steel is listed in Table 1. The G115 steel was normalized at 1100 °C for 1 h, followed by tempering at 760 °C for 3 h. The mechanical properties of G115 steel at room and elevated temperatures are given in Table 2. 2.2. Microstructural characterization To investigate the microstructure evolution after creep, the ruptured specimens were cut along the axial direction for microstructure analysis. The microstructure examinations were conducted via optical microscopy (OM, Axio Vert. A1), scanning electron microscopy (SEM, ZEISS EVO 18), and transmission electron microscopy (TEM, Tecnai G2 F30) coupled with energy-dispersive X-ray spectroscopy (EDS). The samples for the OM and the SEM examinations were polished mechanically and then etched in a mixed solution of FeCl3 (5 g), HCl (15 mL), and H2O (80 mL) for 35 s. The samples for TEM observations were mechanically ground to produce thin plates 50 μm in thickness and then were twin-jet electro-polished in a mixed solution of 5% perchloric acid and 95% ethanol. Furthermore, to clarify the creep deformation mechanism of G115 steel, the microstructure and texture of ruptured samples were characterized with electron backscattered diffraction (EBSD) technique. A cross-sectional sample in the rolling direction (RD)-transverse direction (TD) plane was prepared for EBSD measurement, as schematically shown in Fig. 1. The EBSD data were collected with a step size of 2.0 µm. Based on the EBSD results, the inverse pole figure (IPF), image quality (IQ) map, pole figure (PF), and misorientation map were analyzed. The dislocation structure and distribution of the precipitates were also studied. 2.3. Creep testing Uniaxial creep tests were performed at different stress levels in the range of 120– 220 MPa under temperatures of 625, 650, and 675 °C using creep machines (RDJ 50). Tests were conducted on standard creep specimens with 5.0 mm in gage diameter and 40.0 mm in gage length. Fig. 1 shows the dimensions of the specimens used for the

creep tests. The tested temperatures along the specimen were controlled to within ± 2 °C. The holding time prior to the tensile test was approximately 60 min. 3. Results and discussion 3.1. Microstructural characterization of the as-received steel The microstructure of the as-received G115 steel is composed of tempered martensitic structure inside the prior austenite grain with a grain size ranging from 30 to 50 μm (Fig. 2). Fig. 3 shows large intergranular carbides and fine carbonitrides in the body-centered cubic (BCC) matrix. Continuous Cr23C6 precipitates were located at grain boundaries, which have an ellipsoidal shape and sizes ranging from 100 to 200 nm (Fig. 3a). The MX precipitates showing quadrilateral features are NbN, and were distributed mainly within the martensite laths and along the lath boundaries (Fig. 3b). A dislocation cell structure was observed due to dislocation entanglement (Fig. 3c). Martensitic laths with high dislocation density appeared during the tests and can be seen in Fig. 3d. 3.2. Creep behavior of the G115 steel Typical short-term creep curves of G115 steel at 625, 650 and 675 °C are presented in Fig. 4. The creep behavior of this steel shows similar features. It can be seen from Fig. 4 that the steady-state creep rates (  min ) are relatively high compared to the long-term creep. Furthermore, creep failure strain improved significantly at 675 °C, which is related to the creep deformation and fracture mechanism. Based on the shape of creep strain–time curves, the creep curves can be divided into three stages [19], as shown in Fig. 5. Three regions can be characterized: a primary transient stage (from point O to point A) followed by an apparent secondary stage (from point A to point B), and an accelerated tertiary creep regime (from point B to point C). In the primary transient stage, the creep rate (  ) decreases prominently with time due to work hardening caused by dislocation multiplication and other interactions. Afterwards, the creep rate reaches a minimum value in an apparent secondary stage where there is a balance between work hardening and softening. Eventually, rapid creep deformation occurs in the tertiary creep stage until fracture

due to the reinforcement of the creep recovery processes and coarsening of precipitates and cavities [20]. In addition, the typical creep curves for the G115 steel at 650 and 675 °C under various stresses are shown in Fig. 6. The creep curves exhibit similar features. In addition, the steady-state creep rate (  min ) is referred to as the minimum creep rate, and the stress dependence of the steady-rate creep rate can be described by Norton’s power law [21]:

 min  A n ,

(1)

where A is the stress coefficient and n is the stress exponent. Moreover, the stress exponent n represents the creep deformation mechanism for a material [22]. The value n of the G115 steel can be obtained via the slope of the ln  – ln  plot, as shown in Fig. 7. The stress exponent n decreases from 12 to 11 as temperature increases from 650 to 675 °C. The stress exponent n in this work is high due to the high stresses employed. These results suggest that the power-law breakdown (PLB) regime is the dominant creep deformation mechanism for the G115 steel under high stress, which can be attributed to the formation of excess cavities [23]. To understand the effect of creep parameters on the local plastic deformation, the cross-sections of the specimens ruptured under various creep conditions are shown in Fig. 8. From Fig. 8 it can be seen that the necking elongation increases slightly from 3.7 to 4.2 mm, and then increases to 4.7 mm as the creep temperature increases from 625 to 675 °C. The tertiary creep regime can be ascribed to the local plastic deformation (necking). The result indicates that the local deformation can occur more easily at higher temperature. 3.3. Microstructure evolution 3.3.1. Change in EBSD under different creep conditions Orientation and grain boundary maps are shown in Fig. 9 after creep at 220 MPa/625 °C (a, b), 200 MPa/650 °C (a, b), and 160 MPa/675 °C. It can be seen from Figs. 9a, c, and g that the crept specimens do not show an obvious texture. Grain boundary maps of the G115 steel after various creep deformation conditions are shown in Figs. 9b, d, and f, where black lines denote high-angle grain boundaries

(HAGBs) for misorentation angles > 15° and red lines represent low-angle grain boundaries (LAGBs) for values between 2° and 15° [24]. Furthermore, the grains can be classified into three types: substructured grains, recrystallized grains, and deformed grains. If the average angle in a grain exceeds the minimum angle to define a subgrain (1°), the grain is classified as being a “deformed grain”. Some grains consist of subgrains whose internal misorientations are below 1°, but the misorientation from subgrain to subgrain is above 1°. In that case, the grains are classified as “substructured grains”. All remaining grains are classified as “recrystallized grains” [25]. The distribution of different types of grains for G115 steel at 625 °C under 220 MPa, 650 °C under 200 MPa, and 675 °C under 160 MPa are shown in Fig. 10. An increased number of polygonized subgrains were observed after creep. It can be seen from Fig. 10 that with the increase in creep time, the number of subgrains slightly increased, and then sharply increased, which indicates dynamic recrystallization (DRX) occurs after creep deformation. This result suggests that the DRX behavior occurs significantly at longer creep time. Texture analysis was used to investigate the deformation mechanism [26]. Liu et al. [27] reported the deformation behavior of Ti55 titanium alloy by using texture maximum intensity. For a material with significant texture, the maximum texture intensity decreases with further deformation, suggesting a weakening and randomization of the initial texture. In addition, the retention of the texture peaks under various strains may slow the process of grain rotation during tensile deformation. It can be proposed that the changes in texture and intensity are closely related to grain boundary sliding and grain rotation, which could weaken the texture intensity [27,28]. According to Liu et al. [27], the texture analysis can be applied to a material with initial significant texture. However, the as-received G115 steel used in this study has no obvious texture, as presented in Fig. 11(a). Furthermore, it is clear that the maximum texture intensities of the specimens after creep deformation are relatively mild (Fig. 11), which suggests that there is no significant textural feature after creep deformation. Therefore, based on the above analysis, such a slight texture evolution before and after creep deformation has no significant effect on creep

deformation in this study. 3.3.2. Change in dislocation structure under different creep conditions The dislocation structures and their evolution on the G115 steel during short-term creep deformation were observed by TEM. Fig. 12 shows the dislocation structure of the specimen after creep deformation at 625 °C under 220 MPa. The dislocation cells were observed and it could be seen that the dislocation structures were tangled, forming dislocation walls. Further, dislocation cells were composed of dislocation walls formed by high density dislocation tangles and dislocation channels with low dislocation density [29]. In addition, dislocation pile-ups with various impenetrable walls were found, which are usually formed at grain boundaries and bimetallic interfaces [30]. Understanding the dislocation pile-up feature is difficult. Interestingly, a continuum model [31] was developed to describe the dislocation pile-ups problem, which suggests that the near-lock piles distribution is dependent on the total number of dislocations and the stacking fault energy is positively correlated with the length of an equilibrated pile-up. Fig. 13 shows the dislocation structure of the specimen after creep deformation at 650 °C under 200 MPa. The dislocation patterns evolved from dislocation tangles and aligned dislocation arrays. In addition, the interaction of precipitates and dislocations was observed along with dislocation dipoles (Fig. 13b), which implies that the dislocation movement during creep deformation is strongly affected by the solute atoms and precipitates, due to the restriction of solute atoms and precipitate atmospheres [32,33]. Fig. 14 illustrates the dislocation structures of the specimen after creep deformation at 675 °C under 160 MPa. Some simple dislocation patterns such as dislocation tangles and lines were observed (Fig. 14a). Further, dislocation pile-ups were also found as shown in Fig. 14b. The distance between each dislocation is influenced by the combination of mutually repulsive/attractive dislocation interactions and the externally applied loads [31]. 3.3.3. Change in the precipitates under different creep conditions

Precipitation strengthening is the dominant creep strengthening mechanism for the 9Cr tempered martensite ferritic steel G115 [34], and the macroscopic mechanical properties are strongly affected by the behavior of the precipitates. Therefore, the types and sizes of precipitates should also be considered. Figs. 15–17 show the precipitates of the G115 steel after creep deformation with various creep conditions. After creep deformation, three types of precipitates can be distinguished based on EDX analysis: Cu-rich precipitates, W-rich Laves, and Nb-rich MX particles. Fig. 15 shows the precipitates of the G115 steel after creep deformation at 625 °C under 220 MPa. From Fig. 15a it can be seen that the W-rich Laves particles have a block shape and a size of ~110 nm. The Cu-rich precipitates show an ellipsoidal shape and are ~100 nm in size, however, the Nb-rich MX particles still exhibit a square shape and a size of ~40 nm, which is stable from the initial state [17]. In addition, it can be found from Fig. 15b that the W-rich Laves particles with chain and rod-like features were observed and are distributed mainly on the grain boundaries. Fig. 16 shows the precipitates of the G115 steel after creep deformation at 650 °C under 200 MPa. In Fig. 16a, the W-rich Laves phase particles have a rod-like shape and a size of ~100 nm. The Cu-rich precipitates still show an ellipsoidal shape and are ~80 nm in size. And the Nb-rich MX particles are very stable, which hinder the movement of the dislocations and help with the storage of dislocations [35,36]. Furthermore, the W-rich Laves phase particles, showing chain-like characteristics, are located on the subgrain boundaries (Fig. 16b). Fig. 17 shows the precipitates of the G115 steel after creep deformation at 675 °C under 160 MPa. The W-rich phase particles show mainly a chain-like shape, and a small amount of rod-like shapes were observed, which are mainly distributed along the subgrain boundaries. The growth of the Laves phase particles can be divided into two stages. In the first stage, it grows into a rod-like shape along the subgrain boundary and finally morphs into a chain-like shape, and a small amount of bulky shape particles [37]. Based on the above results, the schematic of the growth of Laves phase particles is shown in Fig. 18. Laves phase particles strongly influence both the microstructure and mechanical properties of the steel [38]. The precipitation of Laves phase is considered to be harmful for creep

strength in this 9Cr martensite steel because the nucleation and growth of Laves phases will promote diffusion of tungsten from the matrix to the Laves phase and weaken the effect of solid solution strengthening [39,40]. 3.3.4. Change in microcavities and microcracks under different creep conditions After creep deformation, the damage of the specimens occurs in the form of microcavities and microcracks. To understand the damage evolution of the G115 steel at various creep conditions, Figs. 19–21 show the damage in the G115 steel after creep deformation. Fig. 19 shows the microstructures of ruptured specimens at the fracture frontier at 625 °C under 220 MPa. Martensite cracking [41] was observed, which is generally characterized as thin cracks formed at the martensite lath boundary (Fig. 19a). The martensite is brittle, and the martensite lath boundaries sliding results in stress concentrations at the lath boundaries, where the local stress is sufficient to nucleate microcracks [37]. Consequently, microcracks grow by lath boundary sliding, and martensite cracking occurred. In addition, microcavities were found in the necking, which is due to the fracture of the martensite laths. Fig. 20 shows the microstructures of ruptured specimens at the fracture frontier at 650 °C under 200 MPa. Some microcavities appear at the fracture frontier, and are indicated by green arrows. Furthermore, microcavities were clearly observed at high magnification, which is due to the precipitates easily pinning the movement of the dislocations, consequently forming dislocation pile-ups, which result in stress concentration near the precipitates, and the formation of microcavities. Fig. 21 shows the microstructures of the ruptured specimens at the fracture frontier at 675 °C under 160 MPa. Lots of martensite cracking occurs, but its size is relatively small compared with Fig. 19a. 3.4. Fracture mechanism analysis To understand the fracture mechanism of the G115 steel during creep deformation, the morphology of the fracture surfaces was observed. Figs. 22–24 exhibit SEM images of ruptured specimens in the fracture surface under various creep deformation conditions, exhibiting obvious necking features, with an elliptical cross-section.

Fig. 22 shows the fracture surface of ruptured specimens after creep deformation at 625 °C under 220 MPa. It can be observed that a cup-like feature appears (Fig. 22a), where the feature consists of the shear-lip zone at the outer periphery, a fiber zone at the central region, and a radiation zone, indicating a typical ductile fracture mode. In addition, a flat surface can be seen on the shear-lip zone (Fig. 22b) and radiation zone (Fig. 22d), showing shallow and parabolic dimples. These are referred to as shear dimples, which are usually formed under shear stress [17]. Moreover, a dense array of deep and equiaxed dimples can be observed in the central region (Fig. 22c), which indicates that dimple rupture leads to ductile fracture of the specimen under this creep deformation condition. Fig. 23 shows the fracture surface of the ruptured specimen after creep deformation at 650 °C under 200 MPa, which illustrates significant necking behavior. However, the shear-lip zone is larger than that shown in Fig. 22a. Dense packing of non-uniform dimples can be observed in the central region at a higher magnification (Fig. 23c), which is a fully ductile facture. Fig. 24 shows the fracture surface of the ruptured specimen after creep deformation at 675 °C under 160 MPa, and the macro view shows an obvious elliptical necking (Fig. 24a). However, a smaller shear-lip zone can still be found. In addition, equiaxed dimples were still observed, indicating fully ductile fracturing. The above results suggest that ductile fracturing is the dominant fracture mechanism during short-term creep deformation under the tested creep conditions.

4. Conclusions In this work, the microstructure evolution and fracture mechanism of the G115 steel were investigated during short-term creep deformation. From the results, it can be concluded that: (1) Under creep tests, the creep curves are significantly affected by temperature and stress for all creep deformation conditions, which consist of a transient primary stage, an apparent secondary stage, and an accelerated tertiary creep regime. In addition, the stress exponent n decreases from 12 to 11 as the temperature increases from 650 °C to 675 °C. The power-law breakdown

(PLB) regime is the dominant creep deformation mechanism for the G115 steel under high stress. (2) The crept specimens do not show an obvious texture, and with an increase in creep time, the number of subgrains slightly increased, and then sharply increased, which indicates DRX occurs after creep deformation. (3) Three types of precipitates can be distinguished after creep deformation: W-rich Laves phase, Nb-rich MX, and Cu-rich phase particles. The Nb-rich MX with a square shape maintains its stability and remains in the initial state. The Cu-rich phase particles show an ellipsoidal shape, and the W-rich Laves phase has rod-like, chain-like, and bulky shapes, which are distributed mainly on the grain boundaries. (4) Representative fractographs of the G115 steel after creep deformation exhibit significant necking with an elliptical shape. A dense array of deep and equiaxed dimples appear at the central region under the tested creep conditions. Ductile fracturing is the dominant fracture mechanism during short-term creep deformation.

Acknowledgments This work was financially supported by The National Natural Science Foundation of China [grant number 51475326].

Reference [1] F. Masuyama, ISIJ Int. 41 (2001) 612–625. [2] F. Sun, Y.F. Gu, J.B. Yan, Y.X. Xu, Z.H. Zhong, M. Yuyam, J. Alloys Comp. 687 (2016) 389–401. [3] Q.Z. Gao, X. Dong, C. Li, Z.X. Lin, X.X Yang, M. Dai, J. Alloys Comp. 651 (2015) 537–543. [4] R.N. Hajra, A.K. Rai, H.P. Tripathy, S. Raju, S. Saroja, J. Alloys Comp. 689 (2016) 829–836. [5] K. Kimura, H. Kushima, K. Sawada, Mater. Sci. Eng. A 510-511 (2009) 58–63. [6] M. Nie, J. Zhang, F. Huang, J.W. Liu, X.K. Zhu, Z.L. Chen, L.Z. Ouyang, J.

Alloys Comp. 588 (2014) 348–356. [7] L. Maddi, G.S. Deshmukh, A.R. Ballal, D.R. Peshwe, R.K. Paretkar, K. Laha, M. D. Mathew, Mater. Sci. Eng. A 668 (2016) 215–223. [8] L. Zhao, L.Y. Xu, Y.D. Han, H.Y. Jing, Int. J. Mech. Sci.94 (2015) 63–74. [9] L. Zhao, H.Y. Jing, L.Y. Xu, Y.D. Han, J.J. Xiu, Mater. Sci. Eng. A 558 (2012) 119–128. [10] C. Schroer, J. Konys, T. Furukawa, K. Aoto, J. Nucl. Mater. 398 (2010) 109–115. [11] M. Yoshizawa, M. Igarashi, K. Moriguchi, A. Iseda, Hassan Ghassemi Armaki, K. Maruyama, Mater. Sci. Eng. A 510 (2009) 162–168. [12] S.Y. Bae, H.G. Kang, H.S. Yun, C.W. Kim, D.B. Lee, B.S. Lim, Mater. Sci. Eng. A 499 (2009) 262–266. [13] P. Yan, Z.D. Liu, W. Liu, H.S. Bao, Y.Q. WENG, J. Iron Steel Res. Int. 20 (9) (2013) 73–79. [14] P. Yan, Z.D. Liu, H.S. Bao, Y.Q. Weng, W. Liu, Mater. Sci. Eng. A 597 (2014) 148–156. [15] P. Yan, Z.D. Liu, H.S. Bao, Y.Q. Weng, W. Liu, Mater. Sci. Eng. A 588 (2013) 22–28. [16] P. Yan, Z.D. Liu, Mater. Sci. Eng. A 650 (2016) 290–294. [17] B. Xiao, L.Y. Xu, L. Zhao, H.Y. Jing, Y.D. Han, Mater. Sci. Eng. A 690 (2017) 104–119. [18] P. Yan, Z.D. Liu, H.S. Bao, Y.Q. Weng, W. Liu, Mater. Des. 54 (2014) 874–879. [19] Y. Zhang, H.Y. Jing, L.Y. Xu, L. Zhao, Y.D. Han, Y.X. Zhao, Mater. Sci. Eng. A 686 (2017) 102–112. [20] M.E. Kassner, Butterworth Heinemann, 2015. [21] W. Liu, X. Liu, F.G. Lu, X.H. Tang, H.C. Cui, Y.L. Gao, Mater. Sci. Eng. A 644 (2015) 337–346. [22] O.D. Sherby, E.M. Taleff, Mater. Sci. Eng. A 322 (2002) 89–99. [23] D. Lesuer, C. Syn, O. Sherby, TMS, Warrendale, PA (2000), P. 81–94. [24] F. Cruz-Gandarilla, A.M. Salcedo-Garrido, R.E. Bolmaro, T. Baudin, N.S.D. Vincentis, M. Avalos, J.G. Cabanas-Moreno, H. Mendoza-Leon, Mater. Charact.

118 (2016) 332–339. [25] Z.Q. Zhang, H.Y. Jing, L.Y. Xu, Y.D. Han, Z.Q. Gao, L. Zhao, J.L. Zhang, Appl. Surf. Sci. 413 (2017) 327–343. [26] H. Watanabe, K. Kurimoto, T. Uesugi, Y. Takigawa, K. Higashi, Mater. Sci. Eng. A 558 (2012) 656–662. [27] Z.G. Liu, P.J. Lia, L.T. Xiong, T.Y. Liu, L.J. He, Mater. Sci. Eng. A 680 (2017) 259–269. [28] P. Lin, Z.B. He, S.J. Yuan, J. Shen, Mater. Sci. Eng. A 556 (2012) 617–624. [29] H. Wang, H.Y. Jing, L. Zhao, Y.D. Han, X.Q. Lv, L.Y. Xu, Mater. Sci. Eng. A 690 (2017) 16–31. [30] E. Pacheco, T. Mura, J. Mech. Phys. Solids 17 (1969) 163–170. [31] X.H. Zhang, Acta Mater. 128 (2017) 428–439. [32] G. Facheris, M.S. Pham, K.G.F. Janssens, S.R. Holdsworth, Mater. Sci. Eng. A 587 (2013) 1–11. [33] M.S. Pham, S.R. Holdsworth, Mater. Sci. Eng. A 556 (2012) 122–133. [34] Eggeler G. Acta Metall 37 (1989) 3225–3234. [35] O. Prat, J. Garcia, D. Rojas, G. Sauthoff, G. Inden, Intermetallics 32 (2013) 362– 372. [36] Y.T. Xu, X.Y. Zhang, Y.B. Tian, C. Chen, Y.J. Nan, H. H, M.J. Wang, Mater. Charact. 111 (2016) 122–127. [37] M.I. Isik, A. Kostka, G. Eggeler, Acta Mater. 81 (2014) 230–240. [38] F. Abe, Mater. Sci. Eng. A 319 (2001) 770–773. [39] J.S. Lee, H.G. Armaki, K. Maruyama, T. Muraki, H. Asahi, Mater. Sci. Eng. A 428 (2006) 270–275. [40] K. Miyahara, J.H. Hwang, Y. Shimoide, Scr. Mater. 32 (1995) 1917–1921. [41] J.C. Zhang, H.S. Di, Y.G. Deng, R.D.K. Misra, Mater. Sci. Eng. A 627 (2015) 230–240.

Tables Captions Table 1. Chemical composition (wt.%) of the as-received G115 steel. Table 2. The mechanical properties of the as-received G115 steel at room and elevated temperatures.

Figures Captions Fig. 1. The dimensions of the creep specimens and schematic diagram of samples used in the EBSD examination. Fig. 2. The typical tempered martensitic microstructure of G115 steel for the as-received specimens: (a) OM, and (b) SEM image. Fig. 3. TEM images of G115 steel for the as-received specimens. (a) precipitates of Cr-rich type M23C6, located at grain boundaries, (b) precipitates of Nb-rich type MX within grains, (c) subgrain formed by dislocation entanglement, and (d) typical tempered martensite laths with high dislocation density. Fig. 4. The typical creep curves of G115 steel under different conditions. (a) 625 °C/ 220 MPa, (b) 650 °C/ 200 MPa, and (c) 675 °C/ 160 MPa. Fig. 5. Creep curves at a temperature of 675 °C and a stress of 140 MPa with the three dominant stages (Transient, steady, and accelerated stage). Fig. 6. Creep strain-time curves (a) at 650 °C under stresses of 200 MPa, 160 MPa and 140MPa, and (b) at 675 °C under stresses of 160MPa, 140 MPa and 120MPa. Fig. 7. Stress dependence of the steady-state creep rate of G115 steel at 650 °C and 675 °C. Fig. 8. Cross-sections of the specimens ruptured under different creep deformation conditions. The non-uniform strain was measured in these cross-sections. Fig. 9. Orientation and grain boundaries maps of the crept specimens at 625 °C under 220 MPa (a, b), 650 °C under 200 MPa (c, d), and 675 °C under 160 MPa (e, f). (Black lines denote HAGBs for misorentation angles > 15° and red lines represent LAGBs for values between 2° and 15°). Fig. 10. The distributions of different types of grains for G115 steel at (a) 625 °C

under 220 MPa, (b) 650 °C under 200 MPa, and (c) 675 °C under 160 MPa. (Recrystallized grains in blue, substructured grains in yellow, and deformed grains in red.) Fig. 11. Inverse pole figures (IPFs) before and after creep deformation at (a) the as-received state, (b) 625 °C under 220 MPa, (c) 650 °C under 200 MPa, and (d) 675 °C under 160 MPa. (The intensities corresponding to the IPFs are calculated from the plane parallel to the rolling plane). Fig. 12. Dislocation structure of the G115 steel during creep deformation (625 °C and 220 MPa): (a) dislocation cells, dislocation tangles, and dislocation lines, (b) dislocation pile-ups. Fig. 13. Dislocation structure of the G115 steel during creep deformation (650 °C and 200 MPa): (a) dislocation arrays, dislocation tangles, and the interaction of precipitates and dislocations, (b) dislocation pile-ups and dislocation dipoles. Fig. 14. Dislocation structure of the G115 steel during creep deformation (675 °C and 160 MPa): (a) dislocation lines and dislocation tangles, (b) dislocation pile-ups and dislocation tangles. Fig. 15. Precipitates of the G115 steel during creep deformation at 625 °C under 220 MPa: (a) precipitates of W-rich Laves, Cu-rich phase, and Nb-rich MX, and (b) W-rich Laves with chain-like and block shapes were nucleated at grain boundaries. Fig. 16. Precipitates of the G115 steel during creep deformation at 650 °C under 200 MPa: (a) dispersion precipitates of Cu-rich phase, W-rich Laves and Nb-rich MX, and (b) W-rich Laves with chain-like shapes were nucleated at grain boundaries and Cu-rich phases appear within the subgrains. Fig. 17. Precipitates of the G115 steel during creep deformation at 675 °C under 160 MPa: (a) precipitates of the Cu-rich phase, Nb-rich MX, and W-rich Laves with chain-like, and (b) chain-like W-rich Laves appear at subgrain boundaries. Fig. 18. A schematic of the growth of the Laves phase particles during creep deformation. Fig. 19. Damage of the G115 steel during creep deformation on the fractured frontier at 625 °C under 220 MPa: (a) martensite cracking, and (b) microcavities. (Arrows in

white indicate martensite cracking.) Fig. 20. Damage of the G115 steel during creep deformation on the fractured frontier at 650 °C under 200 MPa: (a) microcavities and microcracks, (b) microcavities at high magnification. (Arrows in green indicate microcavities and arrows in yellow indicate microcracks.) Fig. 21. Damage of the G115 steel during creep deformation on the fractured frontier at 675 °C under 160 MPa: (a) microcavities and martensite cracking, (b) martensite cracking at high magnification. (Arrows in white indicate microcavities and arrows in yellow indicate microcracks.) Fig. 22. SEM fracture surface images of the ruptured specimen at 625 °C under 220 MPa: (a) macro view exhibiting elliptical necking, (b) magnified image showing shear dimples characteristic on the shear-lip zone, (c) magnified view of the fiber zone showing equiaxed dimples, and (d) magnified view of the radiation zone. Fig. 23. SEM fracture surface images of ruptured specimen at 650 °C under 200 MPa: (a) macro view presenting elliptical necking, (b) magnified image showing shear dimples characteristic on the shear-lip zone, (c) magnified view of the fiber zone showing equiaxed dimples, and (d) magnified view of the radiation zone. Fig. 24. SEM fracture surface images of ruptured specimen at 675 °C under 160 MPa: (a) macro view presenting elliptical necking, (b) magnified image showing shear dimples characteristic on the shear-lip zone, (c) magnified view of the fiber zone showing equiaxed dimples, and (d) magnified view of the radiation zone.

Table 1. Chemical composition (wt.%) of the as-received G115 steel. Element

C

Cr

W

Co

Cu

Mn

Si

V

Nb

N

B

Fe

Amount

0.08

8.8

2.8

3.0

1.0

0.5

0.3

0.2

0.06

0.008

0.014

Bal.

Table 2. The mechanical properties of the as-received G115 steel at room and elevated temperatures. Temperature (°C)

Rm (MPa)

R0.2 (MPa)

A (%)

Z (%)

RT

755±3

618±4

21±2

67±1

625

334±6

313±5

19±3

83±4

650

281±2

267±7

20±1

83±3

675

232±4

217±5

30±3

86±1

Fig. 1. The dimensions of the creep specimens and schematic diagram of samples used in the EBSD examination.

Fig. 2. The typical tempered martensitic microstructure of G115 steel for the as-received specimens: (a) OM, and (b) SEM image.

Fig. 3. TEM images of G115 steel for the as-received specimens. (a) precipitates of Cr-rich type M23C6 located at grain boundaries, (b) precipitates of Nb-rich type MX within grains, (c) subgrain formed by dislocation entanglement, and (d) typical tempered martensite laths with high dislocation density.

0.20

(a)

Temperature = 625℃ Stress = 220 MPa

Creep strain

0.15

0.10

Steady creep rate=3.501E-04 s

0.05

-1

0.00 0

50

100

150

200

250

300

Time (h) 0.20

(b)

Temperature = 650℃ Stress = 200 MPa

Creep strain

0.15

0.10

0.05

Steady creep rate =3.113E-04 s

-1

0.00 0

20

40

60

80

100

120

Time (h) 0.35

(c)

Temperature = 675℃ Stress = 160 MPa

0.30

Creep strain

0.25 0.20 0.15 0.10

Steady creep rate =5.032E-04 s

-1

0.05 0.00 0

50

100

150

200

Time (h)

Fig. 4. The typical creep curves of G115 steel under different conditions. (a) 625 °C/ 220 MPa, (b) 650 °C/ 200 MPa, and (c) 675 °C/ 160 MPa.

0.25

Temperature = 675℃ Stress = 140 MPa 0.20

Accelerate stage

Steady stage Creep strain

C

0.15

0.10 Transient stage

0.05

B A

0.00 0

100

200

300

400 Time (h)

500

600

700

Fig. 5. Creep curves at a temperature of 675 °C and a stress of 140 MPa with the three dominant stages (Transient, steady, and accelerated stage).

0.15

0.25 (a)

Temperature=650℃

(b)

140 MPa 160 MPa 200 MPa

Temperature=675℃

120MPa 140MPa 160MPa

0.20

Creep strain

0.10

Creep strain

0.15

0.10

0.05

0.05

Interrupt

0.00

0.00 0

500

1000

1500 2000 Time (h)

2500

3000

3500

0

500

1000 Time (h)

1500

2000

Fig. 6. Creep strain-time curves (a) at 650 °C under stresses of 200 MPa, 160 MPa and 140MPa, and (b) at 675 °C under stresses of 160MPa, 140 MPa and 120MPa.

650℃ 675℃

Steady-state creep rate/h

-1

1E-3

1E-4

n=11

1E-5

n=12

1E-6 120

140

160

180

200

Stress/MPa

Fig. 7. Stress dependence of the steady-state creep rate of G115 steel at 650 °C and 675 °C.

Fig. 8. Cross-sections of the specimens ruptured under different creep deformation conditions. The non-uniform strain was measured in these cross-sections.

Fig. 9. Orientation and grain boundaries maps of the crept specimens at 625 °C under 220 MPa (a, b), 650 °C under 200 MPa (c, d), and 675 °C under 160 MPa (e, f). (Black lines denote HAGBs for misorentation angles > 15° and red lines represent LAGBs for values between 2° and 15°).

Fig. 10. The distributions of different types of grains for G115 steel at (a) 625 °C under 220 MPa, (b) 650 °C under 200 MPa, and (c) 675 °C under 160 MPa. (Recrystallized grains in blue, substructured grains in yellow, and deformed grains in red.)

Fig. 11. Inverse pole figures (IPFs) before and after creep deformation at (a) as-received state, (b) 625 °C under 220 MPa, (c) 650 °C under 200 MPa, and (d) 675 °C under 160 MPa. (The intensities corresponding to the IPFs are calculated from the plane parallel to the rolling plane).

Fig. 12. Dislocation structure of the G115 steel during creep deformation (625 °C and 220 MPa): (a) dislocation cells, dislocation tangles, and dislocation lines, (b) dislocation pile-ups.

Fig. 13. Dislocation structure of the G115 steel during creep deformation (650 °C and 200 MPa): (a) dislocation arrays, dislocation tangles, and the interaction of precipitates and dislocations, (b) dislocation pile-ups and dislocation dipoles.

Fig. 14. Dislocation structure of the G115 steel during creep deformation (675 °C and 160 MPa): (a) dislocation lines and dislocation tangles, (b) dislocation pile-ups and dislocation tangles.

Fig. 15. Precipitates of the G115 steel during creep deformation at 625 °C under 220 MPa: (a) precipitates of W-rich Laves, Cu-rich phase, and Nb-rich MX, and (b) W-rich Laves with chain-like and block shapes were nucleated at grain boundaries.

Fig. 16. Precipitates of the G115 steel during creep deformation at 650 °C under 200 MPa: (a) dispersion precipitates of Cu-rich phase, W-rich Laves and Nb-rich MX, and (b) W-rich Laves with chain-like shapes were nucleated at grain boundaries and Cu-rich phases appear within the subgrains.

Fig. 17. Precipitates of the G115 steel during creep deformation at 675 °C under 160 MPa: (a) precipitates of the Cu-rich phase, Nb-rich MX, and W-rich Laves with chain-like, and (b) chain-like W-rich Laves appear at subgrain boundaries.

Fig. 18. A schematic of the growth of the Laves phase particles during creep deformation.

Fig. 19. Damage of the G115 steel during creep deformation on the fractured frontier at 625 °C under 220 MPa: (a) martensite cracking, and (b) microcavities. (Arrows in white indicate martensite cracking.)

Fig. 20. Damage of the G115 steel during creep deformation on the fractured frontier at 650 °C under 200 MPa: (a) microcavities and microcracks, (b) microcavities at high magnification. (Arrows in green indicate microcavities and arrows in yellow indicate microcracks.)

Fig. 21. Damage of the G115 steel during creep deformation on the fractured frontier at 675 °C under 160 MPa: (a) microcavities and martensite cracking, (b) martensite cracking at high magnification. (Arrows in white indicate martensite cracking and arrows in green indicate microcavities.)

Fig. 22. SEM fracture surface images of the ruptured specimen at 625 °C under 220 MPa: (a) macro view exhibiting elliptical necking, (b) magnified image showing shear dimples characteristic on the shear-lip zone, (c) magnified view of the fiber zone showing equiaxed dimples, and (d) magnified view of the radiation zone.

Fig. 23. SEM fracture surface images of ruptured specimen at 650 °C under 200 MPa: (a) macro view presenting elliptical necking, (b) magnified image showing shear dimples characteristic on the shear-lip zone, (c) magnified view of the fiber zone showing equiaxed dimples, and (d) magnified view of the radiation zone.

Fig. 24. SEM fracture surface images of ruptured specimen at 675 °C under 160 MPa: (a) macro view presenting elliptical necking, (b) magnified image showing shear dimples characteristic on the shear-lip zone, (c) magnified view of the fiber zone showing equiaxed dimples, and (d) magnified view of the radiation zone.