Journal Information
Share
Share
Download PDF
More article options
Visits
0
Original Article
DOI: 10.1016/j.jmrt.2019.09.053
Open Access
Available online 10 October 2019
Effect of carbides on high-temperature aging embrittlement in 12%Cr martensitic heat-resistant steel
Visits
0
Junru Lia,
Corresponding author
lijunru_qdu@163.com

Corresponding author.
, Pengfei Zhanga,b, Tian Hea, Lianjun Chenga, Liwei Wanga, Hong Lia
a School of Electromechanic Engineering, Qingdao University, Qingdao 266071, China
b ITEC College of Engineering, University of Louisiana at Lafayette, LA 70503, United States
This item has received
0
Visits

Under a Creative Commons license
Article information
Abstract
Full Text
Bibliography
Download PDF
Statistics
Figures (25)
Show moreShow less
Tables (2)
Table 1. Chemical composition of the experimental steels (in mass percentage).
Table 2. Heat treatment process of the experimental steels.
Show moreShow less
Abstract

It is well-known that the 9–12%Cr martensitic heat-resistant steels own excellent mechanical properties; yet, in the case of aging embrittlement mechanism, little has been documented with carbides effect. In this study, to understand the aging embrittlement mechanism in steel 10Cr12Ni3Mo2VN, we focused on high-temperature short-term aging embrittlement by investigating the relationship between the heat treatment, microstructures, and aging embrittlement. It is mainly interested in the effect of carbides on the impact toughness of the steel sample after a heat treatment. The results show that the steel appears a reversible short-term aging embrittlement at temperatures within 500–600℃. It was found that carbides play an important role in the aging embrittlement, especially the M7C3-type carbides. Carbides formed at boundaries cause boundaries weakening and induce tempered martensite matrix to fracture in quasi-cleavage mode, which leads to the aging embrittlement. It was also found that the aging embrittlement can be eliminated by re-tempering at a higher temperature due to the transformation of carbides from M7C3-type to M23C6-type. Aging embrittlement arises again but becomes weak after a re-aging treatment. In addition, by refining prior austenite grains, increasing temper temperature and reducing carbon content, it can improve the resistance to aging embrittlement in the sample as refining grains increase the resistance to brittle fracture; what’s more, it could reduce the precipitation of carbides at boundaries during aging by increasing temper temperature and reducing carbon content.

Keywords:
Martensitic heat-resistant steels
Aging embrittlement
Reversible brittleness
Precipitates
Carbides
Full Text
1Introduction

Because of its excellent mechanical properties, 9–12%Cr martensitic heat-resistant steel has been widely used for making turbine blades of ultra-supercritical (USC) coal-fired power [1,2]. However, the martensitic heat-resistant steel is highly susceptible to embrittlement by various causes, such as temper embrittlement and aging embrittlement [3–5]. Among them, aging embrittlement is one of the most common causes for steel embrittlement. In previous studies, precipitation of carbides and segregation of harmful chemical elements were considered as the most common reason causing aging embrittlement [4–8]. Angeliu et al. pointed out that the long-term aging embrittlement below 500℃ in steel M152 was primarily attributed to the combination of alpha prime formation and segregation of stannum on boundaries [4]. In addition, aging embrittlement at 595℃ in steel M152 was the result of segregation of phosphorus and M23C6-type carbides precipitation along boundaries [5]. In a 9Cr-1.4W-0.06Ta-0.22V martensitic steel, the coarsening of M23C6 and precipitation of Laves phase were considered as main factors to cause embrittlement at 200–400[7]. In low-alloy steels, aging embrittlement was caused, for the most part, by segregation of impurity elements, precipitation of precipitates, and coarsening of grains [7,8]. In short, it suggests that aging embrittlement can be caused by various reasons under different aging treatments in different steels. Thus, to avoid brittle failure of the final-stage turbine blades, it is necessary to understand the aging embrittlement mechanism in martensitic heat-resistant steel.

The purpose of this study was to understand the aging embrittlement mechanism in martensitic heat-resistant steels. The sample investigated here was the 10Cr12Ni3Mo2VN steel, which has been used for making the final-stage turbine blades. According to a preliminary investigation, it was found that the high-temperature short-term aging embrittlement in 10Cr12Ni3Mo2VN steel can be related to the precipitation of carbides. Thereafter, the effect of carbides on the high-temperature short-term aging embrittlement is the major focus in this study.

2Experimental materials and methods

The investigated 10Cr12Ni3Mo2VN Steel samples were labeled as No.1 and No. 2. The sample chemical compositions are listed in Table 1 as in mass percentage. They were prepared by forging from electroslag remelting ingots. As compared with sample No. 1, the No. 2 contains more carbon but has similar contents of other elements.

Table 1.

Chemical composition of the experimental steels (in mass percentage).

Steel  Mn  Si  Sn  Cr  Ni  Mo  Fe 
No. 1  0.09  0.82  0.20  0.018  0.020  0.001  11.77  2.60  1.76  0.33  0.032  Bal. 
No. 2  0.13  0.85  0.19  0.019  0.021  0.001  11.85  2.58  1.73  0.31  0.034  Bal. 

To investigate the influence of high-temperature short-term aging on the toughness, specimens were machined from the forging bars in longitudinal directions with sizes of 12mm square section and 60mm long. The heat treatment and aging process on samples were designed as shown in Table 2. The sample No. 1 was furtherly divided into twelve groups of specimens in order to investigate the effect of temperature and the reversibility of aging embrittlement. For instance, specimen S1-1 was quenched after austenitizing for 45min at 1040℃ and then tempered for 120min (i.e., 2h) at 675℃ but no aging treatment; specimen S1-2 was aged for 100h at 500℃ after the same heat treatment (i.e., quenching and tempering) as for specimen S1-1; specimens S1-11 and S1-12 were undergone re-tempering and/or re-aging after heat treatment and aging treatment; and so on. All specimens underwent quenching and tempering processes prior to any tests. Specimens were heat-treated using resistance box-type furnaces. The heat treatment process was carried out by a box-type heat treatment furnace. The decarburization layer was removed before tests. The impact tests were carried out using the standard Charpy V-notch impact tester at room temperature (20℃). The hardness tests were carried out using Brinell hardness tester at room temperature (20℃). The impact energy and hardness were determined based on the collected test data.

Table 2.

Heat treatment process of the experimental steels.

Steel  Specimens  Quenching  Tempering  Aging  Re-tempering  Re-aging 
No. 1S1-1  1040℃–45min675℃–2h 
S1-2  500℃–100
S1-3  550℃–100
S1-4  600℃–100
S1-5  575℃–2h 
S1-6  550℃–100
S1-7  980℃–45min675℃–2h 
S1-8  550℃–100
S1-9  1100℃–45min 
S1-10  550℃–100
S1-11  1040℃–45min550℃–100675℃–2 
S1-12  550℃–100675℃–2550℃–100
No. 2S2-1  1040℃–45min 
S2-2  550℃–100

The virgin specimens were first ground and polished and subsequently etched in a solution (5g CuSO4, 70mL HCl, and 100mL H2O). Microstructure characterization was carried out by using a ZEISS-AXIOSCOPEA1 optical microscopy (OM) and a high-resolution SIGMA500 field emission SEM (FESEM) on the prepared specimens. An JSM-6390LV scanning electron microscopy (SEM) was used to characterize the fracture surface of impact specimens and the microvoids near fracture surface. A JEM-2100(HR) transmission electron microscope (TEM) was used to perform TEM analyses of the precipitates on carbon extraction replicas. An X-ray energy dispersive spectroscopy (EDS) was also used to analyze the composition of the precipitates on carbon extraction replicas. To prepare the carbon extraction duplicates, the virgin specimens were furtherly processed by depositing a carbon film on the etched surface. After that, the carbon film was removed using a solution containing 5g CuSO4, 70mL HCl, and 100mL H2O.

3Results3.1Impact toughness

Aging embrittlement increased with aging treatment temperature. It is worth noting that embrittlement is a loss of impact toughness of steel. The percentage value in the impact energy change shows aging embrittlement, i.e., higher value means the increase of aging embrittlement and vice versa. From Fig. 1 it can be seen that the impact energy decreases along with the increase of aging treatment temperature. For example, the impact energy decreases by 6.7% after aging for 100h at 500℃. As aging temperature increased to 550℃, the impact energy of specimen S1-3 is 30.4% less than the control sample’s (i.e., specimen S1-1). And impact energy further reduces by a factor of 39.4% as the aging temperature increased to 600℃. Therefore, increasing aging temperature promotes the aging embrittlement in the sample.

Fig. 1.

Impact toughness of Steel No. 1 after aging at different temperature.

(0.2MB).

The treatment of re-tempering and re-aging can affect the steel’s aging embrittlement. It is interesting to see, in Fig. 2, that the impact energy of specimen S1-11 is 10.6% higher than the control sample’s but over 40% higher than the impact energy of specimen S1-3. The increase in impact energy means the steel becomes tougher and owns higher resistance to impact damage. It suggests that the aging embrittlement was completely eliminated after re-tempering for 2h at 675℃. However, aging embrittlement appeared again in specimen S1-12 after re-aging for 100h at 550℃. The results above show the aging embrittlement is reversible in the investigated steel sample.

Fig. 2.

Influence of re-tempering and re-aging on the aging embrittlement in Steel No. 1.

(0.21MB).

Besides, austenitizing (i.e., quenching) and temper temperature could also affect the steel’s aging embrittlement. For example, as shown in Fig. 3, following observations can be drawn: (1) impact energy for all tested specimens decreases as the quenching temperature increases from 980°C to 1100°C; (2) impact energy for all tested specimens also decreases along with the increase of temper temperature; (3) percentage values increase as quenching temperature changed from 980°C to 1100°C but lessen as temper temperature increases from 575°C to 675°C. In short, the aging embrittlement becomes more serious once austenitizing temperature increases and temper temperature decreases.

Fig. 3.

Influence of austenitizing and temper temperatures on aging embrittlement in Steel No. 1.

(0.35MB).

What’s more, the element carbon plays an important role in determining both impact toughness and aging embrittlement. From Fig. 4, the impact energy of specimen S1-1 was about 158J but became about 110J for specimen S2-1. It can be seen that the impact energy of sample No. 2 is much lower when compared with the sample No. 1. It worth noting that the sample No. 2 contains more carbon than sample No. 1 has. The comparison of impact energy between samples of No. 1 and No. 2 suggests that the aging embrittlement was stronger in sample No. 2 as that the impact energy decreased by more than 50% after aging for 100h at 550℃. It indicates sample No. 2 exhibited lower resistance to high-temperature aging.

Fig. 4.

Impact toughness of Steels No. 1 and No. 2.

(0.15MB).
3.2Hardness tests

Keep in mind that the hardness of experimental steel was tested in this study rather than the strength. From the Fig. 5, it can be seen the hardness of samples slightly increases after samples aging at 500℃ and 550℃, but slightly decreases after samples aging at 600℃. After re-tempering for 2h at 675℃, the hardness remarkably decreases by 18.6 HBW (in Fig. 6). And it increases again by 2.8 HBW after samples re-aging at 550℃.

Fig. 5.

Hardness of Steel No. 1 after aging at different temperature.

(0.21MB).
Fig. 6.

Hardness after re-tempering and re-aging in Steel No. 1.

(0.2MB).

Fig. 7 shows the influence of austenitizing and temper temperatures on the hardness of steel sample No. 1. The results show hardness slightly decreases by 1.7 HBW after the aging of specimen S1-8, which was austenitized at 980℃. However, it increases by 3.4 HBW and 4.0 HBW after the aging under austenitizing at 1040℃ and 1100℃ respectively. Besides, it exhibits a big difference in the changing of hardness after the aging under different temper temperatures. The hardness is slightly increased by 3.4 HBW after the aging of specimen S1-3 which was tempered at 675℃. However, the hardness remarkably decreases by 26.8 HBW after the aging of specimen S1-6.

Fig. 7.

Hardness under different austenitizing and temper temperatures in Steel No. 1.

(0.35MB).

The effect of chemical elements on steel hardness is displayed in Fig. 8. For the sample No. 2, the hardness increases remarkably with an increment up to 10.1 HBW after the aging. As compared, the increment in the hardness of sample No. 1 is only about 3.4 HBW.

Fig. 8.

Hardness of Steels No. 1 and No. 2.

(0.19MB).
3.3Fractograph of impact specimens

Aging embrittlement mechanism was studied by investigating specimen fracture surfaces. Generally, the fracture surface of Charpy specimen consists of an initiation zone, a stable crack propagation zone, an unstable crack propagation zone, and an arrest zone of unstable crack propagation and a shear lip. The unstable crack propagation zone always closes to the arrest zone [3,9,10]. In Figs. 9–14, the SEM results show that all the specimen fractures contained an initiation zone, a stable crack propagation zone, an arrest zone, and a shear lip. But the unstable crack propagation zone was observed only on specimens S1-3, S1-4, S1-10, S1-12, S2-1, and S2-2. From those SEM micrographs, it is interesting to see that the unstable crack propagation zones were fractured in quasi-cleavage mode. Keep in mind that unstable crack propagation zone is usually the signal of embrittlement. Therefore the unstable crack propagation zone is the major focus in this study.

Fig. 9.

SEM fractograph of specimens S1-1(a), S1-2(b), S1-3(c), and S1-4(d) in the stable crack propagation zone.

(1.41MB).
Fig. 10.

SEM fractograph of specimens S1-1(a), S1-2(b), S1-3(c), and S1-4(d) near the arrest zone.

(1.31MB).
Fig. 11.

SEM fractograph of Steel S1-11 (a) and S1-12 (b) near the arrest zone.

(0.72MB).
Fig. 12.

SEM fractograph of specimens S1-5 (a) and S1-6 (b) near the arrest zone.

(0.68MB).
Fig. 13.

SEM fractograph of specimens S1-7(a), S1-8(b), S1-9(c) and S1-10(d) near the arrest zone.

(1.47MB).
Fig. 14.

SEM fractograph of specimens S2-1 (a) and S2-2 (b, c) in the unstable crack propagation zone (a, b) and the stable crack propagation zone (c).

(0.98MB).

Unstable crack propagation zone does not appear in specimen S1-1 according to the fractograph study. In Fig. 9a, it shows that the fracture in stable crack propagation zone is appeared as dimples in specimen S1-1. The fracture surface becomes flat near the arrest zone (in Fig. 10a), but no brittle cracks are observed. After the aging for 100h at 500℃ and 550℃, the fracture is still appeared as dimples, but becomes more flat (in Fig. 9b and c) compared with specimen S1-1. With increasing the aging temperature to 600℃, some biggish cleavage planes are observed in stable crack propagation zone (in Fig. 9d). No brittle zone is observed on specimen S1-2 which was aged at 500℃. But brittle zone is observed on specimens S1-3 and S1-4 which were aged at 550℃ and 600℃ respectively. Although the fracture surface becomes more flat near the arrest zone on specimen S1-2 (in Fig. 10b), the fracture is still appeared primarily in shallow dimples. Quasi-cleavage cracks are observed in the unstable crack propagation zone of specimens S1-3 and S1-4 (in Fig. 10c and d). Furtherly, biggish cleavage planes are observed in the unstable crack propagation zone of specimen S1-4. The percentage of the brittle fracture zone approximately increases from 15% to 25% along with the increase of aging temperature from 550℃ to 600℃. That indicates increasing aging temperature leads to the increase in the brittleness of the experimental steel.

Brittle fracture zone disappears after the re-tempering of specimen S1-11 for 2h at 675℃ (in Fig. 11a). It is also disappeared on specimen S1-12 which was re-aged for 100h at 550℃. But the fracture surface becomes more flat near the arrest zone (in Fig. 11b), indicating the fracture is a result of embrittlement after re-aging.

No brittle fracture zones are observed on both specimens S1-5 and S1-6 which were tempered at 575℃. According to the Fig. 12a, it shows that the specimen of S1-5 fractures with bigger dimples in the zone near the arrest zone. What’s more, as shown in Fig. 12b, the specimen of S1-6 fractures with a mixture of dimples and quasi-cleavage cracks after the specimen aging for 100h at 550℃. But dimples played a major role in determining fracture mode.

Fracture surfaces of specimens under different austenitizing temperatures are presented in Fig. 13. No brittle fracture zones are observed on specimens of S1-7 and S1-8 which were austenitized at 980℃. But the fracture surface is more flat near the arrest zone on specimen of S1-8 after the aging for 100h at 550℃ as compared with S1-7 (in Fig. 13a and b). Also no brittle cracks are observed on specimen S1-9 which was austenitized at 1100℃ (in Fig. 13c). But 23% brittle fracture zone appeared as quasi-cleavage cracks is observed on specimen S1-10 (in Fig. 13d). Furtherly, by comparing with specimens of S1-1 and S1-3 which were austenitized at 1040℃, it can be found that fractures are much thicker after austenitizing at higher temperatures. All of above show that aging embrittlement gets worsening with the increase in the austenitizing temperatures.

Brittle fracture zone appeared as quasi-cleavage crack is observed both on specimens of S2-1 and S2-2 (in Fig. 14a and b). The percentages of brittle fracture zone on specimens of S2-1 and S2-2 were measured as 11% and 46% respectively. It is worth noting that there are less tear ridges in the specimen S2-2 after the sample aging for 100h at 550℃ as compared with the specimen S2-1. That indicates more brittleness after an aging process. Besides, from the Fig. 14, it can be seen that some cleavage planes appears in the stable crack propagation zone of the specimen S2-2.

To understand the brittleness mechanism, microstructures near the fractures were investigated in specimens S1-1 and S1-3 (in Fig. 15) by using SEM micrographs. The SEM results show that there are lots of microvoids near the fracture surface on both specimens of S1-1 and S1-3. But the morphologies of these microvoids are different between specimens S1-1 and S1-2. Most of the microvoids observed appear as circular holes (in Fig. 15a) on the specimen S1-1 which was not under an aging treatment. But some cracks and chains of microvoids at boundaries, between prior austenite grains and martensite laths, are observed on the specimen S1-3 which was aged for 100h at 550℃. That indicates the boundaries become weaker after aging.

Fig. 15.

SEM micrographs near the fracture surface in specimens S1-1 (a) and S1-3 (b).

(0.5MB).
3.4Microstructure characterization

Microstructure plays a crucial role in determining mechanical properties of steel. Thereafter, to understand the aging embrittlement mechanism, microstructures of the steel sample after heat treatment were characterized. The metallographic images show that the matrixes of all samples consist of typical temper martensite laths. But the morphologies of the precipitates varies a lot across samples. The FESEM micrographs show that lots of spherical precipitates formed within the steel sample of S1-1 (in Fig. 16a). The spherical precipitates form along the martensite lath and prior austenite grain boundaries after tempered for 2h at 675℃ were M23C6-type [3]. From the Fig. 16b, it can be seen that there are more precipitates within the specimen of S1-3 after an aging for 100h at 550℃. But the precipitates formed during aging are difficult to be distinguished by FESEM due to the massive amounts of precipitates. But the precipitates formed during the aging can be clearly distinguished for specimens tempered at 575℃ (in Fig. 17). A small amount of acicular precipitates form within martensite laths in the specimen of S1-5, which was quenched and tempered at 575℃ (in Fig. 17a). These acicular precipitates are M7C3-type [3]. The boundaries of martensite laths and prior austenite grains are relatively distinct in the specimen of S1-5 before an aging. More precipitates are observed in the specimen of S1-6 after aging. And some precipitates are observed at the boundaries in the specimen S1-6 (in Fig. 17b). To investigate the precipitates formed during the aging, the microstructure of these precipitates was furtherly examined using TEM on carbon extraction replicas (in Fig. 18). And their chemical composition was analyzed by EDS (in Fig. 19). From the TEM micrographs, it can be seen that spherical precipitates formed at boundaries are isolated and finer precipitates formed at boundaries are continuous in the specimen S1-6 after an aging treatment (in Fig. 18). And the finer precipitates are also observed in martensite matrix. Upon the diffraction patterns and EDS analytical results, the isolated spherical precipitates are the (Cr, Mo, Mn, V)-rich M23C6-type carbides and the continuous finer precipitates are the (Cr, Mo, Mn, V)-rich M7C3-type carbides.

Fig. 16.

FESEM micrographs of specimens S1-1 (a) and S1-3 (b).

(0.43MB).
Fig. 17.

FESEM micrographs of specimens S1-5 (a) and S1-6 (b).

(0.42MB).
Fig. 18.

TEM micrographs of precipitates formed at boundaries in specimen S1-6.

(0.48MB).
Fig. 19.

EDS spectra of spherical precipitates (a) and finer precipitates (b) at boundaries in specimen S1-6.

(0.12MB).

The effect of aging temperature on microstructure of steels was investigated according to FESEM micrographs. Fig. 20 shows the microstructure of steels under different aging temperatures. It can be seen that less carbides precipitate in the specimen of S1-2 after an aging at 500℃ (in the Fig. 20a) but more carbides precipitate in the specimen of S1-4 after an aging at 600℃ (in the Fig. 20b). That indicates increasing aging temperature promotes the precipitation of carbides.

Fig. 20.

FESEM micrographs of specimens S1-2 (a) and S1-4 (b).

(0.41MB).

The effect of re-tempering and re-aging on microstructures was also investigated, as shown in Fig. 21. It seems little difference in morphology of precipitates between specimens of S1-3 (after aging, in Fig. 16a), S1-11 (after re-tempering), and S1-12 (after re-aging). Maybe some changes appear after re-tempering and re-aging. But, due to the large quantity of precipitates, it is difficult to distinguish the precipitate type based on the FESEM results.

Fig. 21.

FESEM micrographs of specimens S1-11 (a) and S1-12 (b).

(0.41MB).

The effect of austenitizing temperature on grain size was investigated in this study using OM. In Fig. 22, it can be clearly seen that the prior austenite grain size increases with the increase in austenitizing temperatures from 980℃ to 1100℃. The length of martensite laths concomitantly increases [9], as shown in Fig. 23. The size change of prior austenite grains was measured and shown in Fig. 24, in which it shows that the grain size is about 20μm at temperature 980℃ but it becomes about 105μm at temperature 1100℃. The grain size increases linearly along with austenitizing temperatures from 980℃ up to 1100℃.

Fig. 22.

OM micrographs of the prior austenite grains under the austenitizing temperature of 980℃ (a), 1040℃ (b), and 1100℃ (c).

(0.54MB).
Fig. 23.

OM micrographs of specimens S1-8 (a), S1-2 (b) and S1-10 (c).

(0.74MB).
Fig. 24.

The average size of prior austenite grains under different austenitizing temperature.

(0.05MB).

There are more precipitates in the steel sample No. 2 when compared with the steel sample No. 1. Fig. 25 shows the microstructure of the steel sample No. 2. The number of precipitates remarkably increases for the specimen of S2-2 (in Fig. 24b) after an aging when compared with the specimen of S2-1 (in the Fig. 24a). The size of prior austenite grains of specimen S2-1 was measured as 66.2μm that is slightly bigger than that of the specimen S1-1 under the same heat treatment.

Fig. 25.

FESEM micrographs of specimens S2-1 (a) and S2-2 (b).

(0.66MB).
4Discussion

Significant aging embrittlement was observed in the steel sample after aged for 100h at 500–600℃. According to the SEM fractograph of the fractured Charpy impact specimens, the fracture in specimens after aging tends to occur in quasi-cleavage mode. From the microstructures near the fracture surface, it can be seen that the microvoids near the fracture surface appear as singular circular hole in the specimen of S1-1 which was not aged. But cracks and chains of microvoids at boundaries between prior austenite grains and martensite laths were observed in the specimen of S1-3 after aging for 100h at 550℃. That indicates the boundaries become weaker after an aging. From previous studies, segregation of impurity elements (such as phosphorus and stannum) [4,5,8] and precipitation of compounds (such as carbides and Laves) [3,5,7,8] were known as the reason causing boundaries weakening and leading to aging embrittlement in steels. However, no segregation was observed by line scanning using EDS. Of course, it may also be that the segregation of impurity elements cannot be detected by EDS due to their low contents. But the segregation of phosphorus or stannum cannot explain why there is a stronger aging embrittlement in the sample Steel No. 2, which has the similar contents of phosphorus or stannum compared with the sample Steel No. 1. That indicates the aging embrittlement may not be caused by the segregation of impurity elements. In this study, it is mainly interested in the effect of carbides on the aging embrittlement.

Results obtained show that isolated spherical M23C6-type carbides and finer continuous M7C3-type carbides form at boundaries after aging. From the previous study on temper embrittlement in the steel sample, precipitation of M23C6-type carbides at boundaries was found to be harmful to the toughness. M23C6-type carbides form at boundaries weakened the binding force of boundaries. In addition, it plays a nucleating role in the development of cracks in steel, in which quenched martensite matrix is not effectively recovered. But further precipitation of M23C6-type carbides has little influence on the toughness in steel which was tempered for 2h at 675℃ [3]. Thus, the precipitation of M23C6-type carbides may play an important role in the aging embrittlement in the specimen of S1-7 which was tempered at 575℃ and in which quenched martensite matrix was not effectively recovered. But there is also a strong aging embrittlement in specimens which were tempered for 2h at 675℃. So we think that the precipitation of finer M7C3-type carbides may play a more important role in the aging embrittlement due to that. Finer M7C3-type carbides are continuously distributed along boundaries, while M23C6-type carbides are isolated. Obviously, the continuously distributed M7C3-type carbides have stronger harmful effect on boundaries than the isolated M23C6-type carbides. Therefore, continuously distributed finer M7C3-type carbides may play a primary role in aging embrittlement.

A previous study shows that M7C3-type carbides precipitated at low temper temperature 575℃ will transformed to M23C6-type carbides at higher temper temperatures above 625℃ [3]. Thus the recovery of toughness after re-tempering at 675℃ should be attributed to the transformation of M7C3-type carbides to M23C6-type carbides. The harmful effect of M7C3-type carbides could be eliminated after they transform to isolated M23C6-type carbides which have little influence on the toughness in steel tempered for 2h at 675℃. M7C3-type carbides, precipitated again after a re-aging, leads to the reappearance of aging embrittlement. But the embrittlement becomes weaker after a re-aging. That should be attributed to less M7C3-type carbides precipitated after a re-aging. The SEM micrographs show that there are more carbides after a re-tempering. That indicates less carbon and other alloying elements dissolve within matrix after a re-tempering. The lower contents of dissolved carbon and other alloying elements cause less precipitation of M7C3-type carbides during an aging. That’s why the aging embrittlement becomes weaker after a re-aging.

The stronger aging embrittlement at a higher aging temperature should be attributed to more precipitation of carbides. Generally, the precipitation rate of carbides increases with temper temperatures. More M7C3-type carbides will precipitate after aging for the same time at a higher aging temperature. That leads to stronger harmful effect on toughness and stronger aging embrittlement at a higher aging temperature.

The stronger aging embrittlement in specimens tempered at 575℃ should be attributed to two reasons. (1) More carbon and other alloying elements dissolve within matrix after tempering at 575℃ when compared with those after tempering at 675℃. The higher contents of dissolved carbon and other alloying elements induce more M7C3-type carbides precipitated. (2) Further precipitation of M23C6-type carbides has little influence on specimens tempered at 675℃ but has significant effect on specimens tempered at 575℃. Thus M23C6-type carbides formed after an aging has a significant harmful effect on specimens tempered at 575℃.

The lower toughness of the sample Steel No. 2 after quenching and tempering is attributed to their higher carbon content. It is well known that carbon usually decreases the toughness in steel. The stronger aging embrittlement in the Steel No. 2 should also be attributed to more precipitation of carbides due to its higher carbon content.

Martensite matrix is always recovered after aging which appears as a softening mechanism and always leads to the decrease in hardness. But the hardness in specimens of S1-2, S1-3, S1-8, S1-10, S1-12, and S2-2 was increased after an aging treatment. That indicates some hardening mechanism also arose during the aging. Precipitation hardening is the only possible hardening mechanism in this study. Increase of hardness after an aging should be attributed to the precipitation of finer M7C3-type carbides which are precipitated both at the boundaries and in the martensite laths. The precipitation hardening mechanism may cause significant effect on hardness than the softening mechanism in specimens of S1-2, S1-3, S1-8, S1-10, S1-12, and S2-2, which leads to the increase of hardness after an aging. The decrease of hardness after an aging in the specimen of S1-7 should be attributed to the recovery (of martensite matrix)-induced softening effect. Of the specimen S1-11 after the re-tempering for 2h at 675℃, the hardness decreases with 15.2 HBW compared with the specimen of S1-1. In a previous study, the hardness justly decreased 2.8HBW with extending temper time from 2h to 4h at 675[3]. That indicates the decrease of hardness after the re-tempering is not caused only by re-tempering. It should be also attributed to the mixing effect of aging and re-tempering, i.e., recovering of martensite matrix by aging and elimination of precipitation hardening by re-tempering. The eliminating of precipitation hardening after re-tempering is resulted from the transformation of carbides from finer M7C3-type to thick M23C6-type. The difference in hardening level after aging is determined by the amount of precipitation. For example, bigger increase of hardness after an aging in the sample Steel No. 2 indicates more carbides precipitated. Conclusively, increasing temper temperature and reducing carbon could decrease the precipitation of carbides and improve the resistance to aging embrittlement for the steel sample.

The improvement of resistance to aging embrittlement by decreasing austenitizing temperature should be attributed to the refining of prior austenite grains. Grain boundaries can effectively prevent the propagation of cleavage cracks in the experimental steel [9]. Therefore, by decreasing austenitizing temperature and refining grains, it could also improve the resistance to aging embrittlement for the steel sample.

5Conclusions

From the studies of the short-term high-temperature aging embrittlement in martensitic heat-resistant steel 10Cr12Ni3Mo2VN, several important results are summarized as follows.

  • (1)

    The aging embrittlement in the martensitic heat-resistant steel 10Cr12Ni3Mo2VN at 500–600℃ is caused by the precipitation of carbides at boundaries between prior austenite grains and martensite laths, especially the continuously distributed M7C3-type carbides.

  • (2)

    The aging embrittlement in steel 10Cr12Ni3Mo2VN is a reversible brittleness. It can be eliminated by the re-tempering at a higher temperature due to the transformation of carbides from M7C3-type to M23C6-type. And the aging embrittlement becomes weak after re-aging.

  • (3)

    The increase of hardness after an aging should be attributed to the precipitation hardening of finer M7C3-type carbides.

  • (4)

    By decreasing the austenitizing temperature, refining prior austenite grains, increasing temper temperature, and decreasing carbon content, it can effectively improve the resistance to aging embrittlement.

Conflicts of interest

The authors declare no conflicts of interest.

Acknowledgment

Authors gratefully acknowledge the support from the National Natural Science Foundation of China (No. 51701100).

References
[1]
P. Mayr, T.A. Palmer, J.W. Elmer, E.D. Specht, S.M. Allen.
Formation of delta ferrite in 9wt.% Cr steel investigated by in-situ X-ray diffraction using synchrotron radiation.
Metall Mater Trans A, 41 (2010), pp. 2462-2465
[2]
R. Viswanathan, J.F. Henry, J. Tanzosh, G. Stanko, J. Shingledecker, B. Vitalis, et al.
US program on materials technology for ultra-supercritical coal power plants.
J Mater Eng Perform, 14 (2005), pp. 281-292
[3]
J.R. Li, C.L. Zhang, Y.Z. Liu.
Influence of carbides on the high-temperature tempered martensite embrittlement of martensitic heat-resistant steels.
Mater Sci Eng A, 670 (2016), pp. 256-263
[4]
T. Angeliu, E.L. Hall, M. Larsen, A. Linsebigler, C. Mukira.
The long-term aging embrittlement of Fe-12Cr steels below 773 K.
Metall Mater Trans A, 34 (2003), pp. 927-934
[5]
G. Yang, X.Q. Liu, M.X. Yang, C. Wang, Z.D. Liu.
Research on brittleness of heat resistant steel 1Cr12Ni3Mo2VN (M152).
Spec Steel Technol, 15 (2009), pp. 14-24
[6]
K.C. Sahoo, J. Vanaja, P. Parameswaran, V.D. Vijayanand, K. Laha.
Effect of thermal ageing on microstructure, tensile and impact properties of reduced activated ferritic-martensitic steel.
Mater Sci Eng A, 686 (2017), pp. 54-64
[7]
W. Wang, S. Liu, G. Xu, B. Zhang, Q. Huang.
Effect of thermal aging on microstructure and mechanical properties of China low-activation martensitic steel at 550℃.
Nucl Eng Technol, 48 (2016), pp. 518-524
[8]
R. Xing, D. Yu, G. Xie, Z. Yang, X. Wang, X. Chen.
Effect of thermal aging on mechanical properties of a bainitic forging steel for reactor pressure vessel.
Mater Sci Eng A, 720 (2018), pp. 169-175
[9]
J. Li, C. Zhang, B. Jiang, L. Zhou, Y. Liu.
Effect of large-size M23C6-type carbides on the low-temperature toughness of martensitic heat-resistant steels.
J Alloys Compd, 685 (2016), pp. 248-257
[10]
H. Kim, J. Park, M. Kang, S. Lee.
Interpretation of Charpy impact energy characteristics by microstructural evolution of dynamically compressed specimens in three tempered martensitic steels.
Mater Sci Eng A, 649 (2016), pp. 57-67
Copyright © 2019. The Authors
Journal of Materials Research and Technology

Subscribe to our newsletter

Article options
Tools
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.