Abstract
Tube bundles are the most critical components of steam generators in pressurized water plants. They are the heat exchange interface between primary and secondary circuits, acting as an important safety barrier to prevent the radioactive leakage. Stress corrosion cracking (SCC) initiation behavior of the worn alloy 690 tubes was investigated by slow strain rate tests (SSRT) in the high-temperature alkaline solution. The scanning electron microscope, electron backscatter diffraction and transmission electron microscope were used to analyze the fretting wear and SCC initiation behavior of alloy 690 tubes. Results show that SSRT specimens show typical transgranular SCC characteristics. The number of initiated cracks and their average depth are increased with increase in the wear degree, suggesting a wear-induced SCC initiation. This may have a relationship with the grooves, delamination, micro-cracks and residual strain layer with tens of micrometers in thickness left on the worn surface. Furthermore, a preliminary analysis on the possible procedure of wear-induced SCC initiation and further propagation during SSRT tests was carried out based on slip-dissolution/oxidation mechanism.
Science Press
Steam generator (SG) is one of the key equipments in the nuclear power plants, which is used to transfer the heat in the reactor coolant circuit to the secondary side water, and further to generate saturated steam for the secondary circuit power device. Tube bundles are the most critical components of SG. On the one hand, they are the heat exchange interface between primary and secondary circuits. On the other hand, the pressure boundary area of SG tube bundles accounts for about 80% of the total area of the primary pressure boundary, acting as an important safety barrier to prevent the radioactive leakage. Reactor coolant containing boric acid and lithium hydroxide, and weak alkaline solution containing ammonia and diamine are in the primary side and secondary side of SG tubes, respectively. Under the normal operation condition, the temperature and pressure in the primary side and secondary side of SG tubes are 293~327 °C, 15.5 MPa and 226~283 °C, 8.6 MPa, respectively.
SG tubes used in nuclear power plants are found to be susceptible to fretting wear[1-7] and stress corrosion cracking (SCC) [8-17]. The SCC is a synergic effect of material, mechanics and environment, which are directly related to the material surface state such as scratches[18-23]. The fretting wear between the SG tube and the anti-vibration bar (AVB) or the tube support plate (TSP) caused by flow-induced vibration is unavoidable, which consequently causes a lot of defects on the surface, and leads to the initiation of SCC. Alloy 690 is used frequently as the SG tube material. As such, investigation on the effects of fretting wear on the SCC of alloy 690 is helpful to clarify the mechanism of the SCC initiation of the steam generator tubes.
In this study, the slow strain rate tensile test (SSRT) method was used to investigate the effects of wear on the SCC initiation behavior of alloy 690 tube in high-temperature alkaline solution. The wear and SCC behavior were revealed by characterizing the surface state and the cracks formed during the SSRT. The correlation between the wear and SCC was discussed.
Commercial alloy 690 tubes with an outer diameter of 19.05 mm and a wall thickness of 1.09 mm were used for the fretting wear and SCC tests. The chemical composition of alloy 690 is listed in Table 1. Specimens were cut from the as-received materials.
Table 1
Chemical composition of alloy 690 tube (wt%)
Cr | Fe | Mn | Ti | S | P | C | Si | Cu | Al | Ni |
29.31 |
10.26 |
0.31 |
0.25 |
0.002 |
0.009 |
0.025 |
0.11 |
0.04 |
0.39 |
Bal. |
X6Cr13 AVB specimens with a dimension of 34 mm×12.5 mm×3.5 mm were used as the contact material to alloy 690 tubes in the fretting wear test. The illustration of fretting wear between the alloy 690 tube and X6Cr13 AVB is shown in Fig.1. The vertical reciprocating relative friction movement of X6Cr13 AVB sheet against alloy 690 tube under the normal force was performed during the fretting wear test in deionized water at room temperature, in order to get fretting wear damaged alloy 690 tube specimens for the subsequent slow strain rate tensile tests. Fretting wear tests were performed under the displacement amplitude of 200 μm with a frequency of 20 Hz until the wear depth ratio of 20%, 40% and 60% (the ratio of wear depth to the tube wall thickness) were obtained. Alloy 690 tubes were cleaned ultrasonically in ethanol after the fretting wear tests. The worn surface morphology was observed by a TESCAN VEGA 5136XM scanning electron microscopy (SEM), and the residual strain of the worn surface was measured using a ZEISS Sigma 300 SEM equipped with electron backscatter diffraction (EBSD) in connection with the Channel 5 software. A step size of 0.5 μm at a voltage of 25 keV was used in the EBSD residual strain analysis. The microstructure of the cross-section specimen along the depth direction was observed by a Tecnai G2 F20 S-TWIN transmission electron microscope (TEM) equipped with an energy dispersive X-ray spectroscopy (EDS). The TEM specimens were prepared by a FEI-Scios focus ion beam (FIB) system.
Fig.1 Illustration of fretting wear between alloy 690 tube and X6Cr13 AVB
SCC behavior of the alloy 690 tube specimens was tested at a strain rate of 1×10-6 s-1 in a 10wt% sodium hydroxide alkaline solution at 290 °C (7.5 MPa) by simulating the secondary side water on a SERT-5000DP-9L tensile testing machine (TOSHIN KOGYO Co., Ltd, Japan) equipped with a static autoclave. The alkaline solution was prepared using high purity water and NaOH (99wt%). The dissolved oxygen (DO) in the alkaline solution was controlled at <5 µg/L by introducing the high-purity nitrogen gas for 3 h. Characteri-zations of the cracks formed on alloy 690 tubes were then conducted to clarify the SCC behavior. The surface morphology of each specimen was observed using the SEM. Then each specimen was sliced into three pieces along the longitudinal direction for counting the number of cracks and measuring the depth of cracks on the cross-section of the specimens within a distance of 1 mm near the fracture. To reveal the characteristics of the grain boundary near the cracks, electron backscatter diffraction (EBSD) analysis was conducted with a step size of 0.5 μm at a voltage of 20 keV using a ZEISS Sigma 300 SEM equipped with a camera in connection with the TSL software. Specimens for the EBSD analysis were prepared through grinding mechanically by SiC papers up to 2000 grit, polished by a diamond paste of 1 μm and the 40 nm colloidal silica slurry successively.
Fig.2 shows the typical SEM images of the worn surface of alloy 690 tube specimen with a wear depth ratio of 40%. A mixed characteristic of grooves and delamination as well as a few microcracks can be observed, revealing a complicated fret-ting wear mechanism of abrasive wear and delamination wear.
Fig.2 SEM images of worn surface of alloy 690 tube specimen with a wear depth ratio of 40%: (a) 50× and (b) 2000×
Fig.3 shows the EBSD observation of the strain layer on the cross-section of the worn area of alloy 690 tube specimen with a wear depth ratio of 60%. The red and yellow areas reveal a rather higher residual strain, while the blue areas show a less residual strain. The thickness of the residual strain layer can be accordingly measured to be about 10 μm near the worn surface.
Fig.3 EBSD observation of cross-section of alloy 690 tube specimen with a wear depth ratio of 60%
Fig.4 shows the TEM image of the wear center area of the alloy 690 tube specimen with a wear depth ratio of 60%. A large amount of dislocations and slips that are accumulated beneath the worn area can be observed obviously, indicating a severe plastic deformation induced by fretting wear.
Fig.4 TEM image of microstructure of the worn area on the alloy 690 tube specimen with a wear depth ratio of 60%
The load-displacement curves for SSRT of unworn and worn alloy 690 specimens with different wear depth ratios are shown in Fig.5. The maximum loads obtained from the load-displacement curves are listed in Table 2. It is found that the maximum load of alloy 690 specimens decreases significantly from 4018 N to 2117 N with increase in wear depth ratio from 0% to 60%.
Fig.5 Slow strain rate tensile curves of alloy 690 specimens with different wear depth ratios in the high temperature alkaline solution
Table 2
Maximum load for SSRT of alloy 690 specimens
Ratio of wear depth to tube wall thickness/% | Maximum load/N |
0 |
4018 |
20 |
3508 |
40 |
2705 |
60 |
2117 |
The morphologies of worn area near the fracture of alloy 690 SSRT specimens with different wear depth ratios are shown in Fig.6. It is found that the surface of unworn specimen is covered with Fe-Ni-Cr oxides with a size less than 10 μm. Some micro cracks on the surface of alloy 690 specimen with a wear depth ratio of 20% perpendicular to the tensile stress direction are observed. The cracks become denser and the size of the cracks becomes larger, when the wear depth ratio increases from 20% to 60%.
Fig.6 Surface morphologies of worn area near the fracture of alloy 690 specimens with different wear depth ratios after SSRT: (a) 0%, (b) 20%, and (c) 60%
Fig.7 shows the longitudinal cross-section morphologies of worn area near the fracture of the alloy 690 specimens with wear depth ratios of 0%, 20%, 40% and 60% after SSRTs. The cracks with a depth of 5~15 μm can be observed on the unworn specimen and worn specimen with wear depth ratio of 20%. The number and the average depth of the initiated cracks increase with further increase in the wear depth ratio from 20% to 60%. In order to clarify the characteristic of cracks, EBSD analysis for longitudinal cross-section of the worn area near the fracture of alloy 690 tube specimens with different wear depths was performed. Fig.8 shows a typical image of the cracks detected by EBSD for the specimen with a wear depth ratio of 60%, revealing obvious transgranular SCC characteristics.
Fig.7 Longitudinal cross-section morphologies of worn area near the fracture of alloy 690 specimens with different wear depth ratios after SSRT: (a) 0%, (b) 20%, (c) 40%, and (d) 60%
Fig.8 SEM image (a) and EBSD observation (b) of longitudinal cross-section of the worn area near the fracture of alloy 690 specimen with a wear depth ratio of 60%
The statistical results of the number of cracks with different depth observed on the longitudinal cross-section specimens with wear depth ratios of 0%, 20%, 40% and 60% are summarized in Fig.9. The corresponding maximum depth and average depth of cracks are listed in Table 3. The total number of cracks of the longitudinal cross-section specimens with wear depth ratios of 0%, 20%, 40% and 60% is 20, 18, 22 and 56, respectively, indicating an increase in the total number of cracks with increasing the wear depth. In addition, it is found that the unworn specimen shows the smallest maximum crack depth of 15.8 μm and the smallest average crack depth of 8.4 μm. The depth of most cracks is within 20 μm. The maximum depth and the average depth of the cracks show a rapid growth of 3 times with rising the wear depth ratio from 0% to 40%. The maximum depth and the average depth of the cracks reach the similar value with the increase of the wear depth ratio from 40% to 60%.
Fig.9 Number and depth of the cracks observed on the longitudinal section specimens with different wear depth ratios: (a) 0%, (b) 20%,
Table 3
Maximum depth and average depth of the cracks
Ratio of wear depth to thickness of tube wall/% | Maximum depth of cracks/μm | Average depth of cracks/μm |
0 |
15.8 |
8.4 |
20 |
33.1 |
10.7 |
40 |
56.3 |
25.9 |
60 |
51.5 |
26.9 |
The above results indicate that wear promotes the initiation of SCC, which is probably induced by the change of the surface state of material caused by fretting wear. Fig.10 shows the schematic diagram to illustrate the effect of wear on the SCC initiation of alloy 690 tube in a high-temperature alkaline solution. The alloy 690 tube surface is damaged severely during fretting, and a mixed characteristic of grooves and delamination as well as a few micro-cracks can be observed on the worn surface as shown in Fig.2. This indicates the combined wear mechanism of abrasive wear and delamination wear[24,25]. In addition, EBSD and TEM results prove a residual strain layer with proximately 10 μm in thickness and a high level of residual stress under the wear surface as shown in Fig.3 and Fig.4. During the fretting wear process, abrasive wear occurs. The cyclic load is applied on the surface and subsurface of the specimens during the reciprocating wear process. This fatigue process leads to the accumulation of defects such as dislocations and stacking faults. Furthermore, the slips and even micro-cracks occur in the substrate due to the complex stresses mentioned above. Some micro-cracks are further propagated and combined, which finally causes the removal of material.
Fig.10 Schematic diagrams illustrating the effect of wear on the SCC of alloy 690 tube in a high-temperature alkaline solution: (a) worn surface, (b) SCC initiation, and (c) SCC propagation
According to the slip-dissolution/oxidation mechanism, the surface state and the residual strain play important roles in the SCC initiation process[26,27]. When the surface of alloy 690 tube is destroyed by fretting wear, the number of the SCC initiation positions increases, such as the micro-cracks, the delamination and the grooves. Meanwhile, more dislocations in the deformation layer of the worn tubes move along the slip plane to form large slip steps, leading to the stress concentration on the worn surface under the applied stress. The surface oxide film near the slips exposed in high-temperature solution is broken and thus the new matrix metal is exposed to the high-temperature alkaline solution and dissolved. A passivation film will be formed again. The dislocations will slip and thus the oxide film will be broken again under the stress. This process is repeated over and over, causing cracks to grow. Accordingly, the cracks continue to propagate along the sliding steps to the inside of the matrix metal. The increase in the degree of wear leads to an increase in local stress concentration, which in turn promotes an increase in the number and depth of SCC initiation.
1) Alloy 690 tubes show a complex morphology after fretting wear, including grooves, delamination, micro-cracks, and residual strain layer with tens of micrometers in thickness.
2) Alloy 690 specimens with a wear depth ratio of 0%~60% after SSRT tests show the characteristic of predominant transgranular SCC.
3) The increased degree of surface damage and local stress concentration caused by wear promote the SCC initiation in high-temperature alkaline solution according to the slip-dissolution/oxidation mechanism.
References
2 Guerout F M, Fisher N J. Journal of Pressure Vessel Technology[J], 1999, 121(3): 304 [Baidu Scholar]
3 Fisher N J, Chow A B, Weckwerth M K. Journal of Pressure Vessel Technology[J], 1995, 117(4): 312 [Baidu Scholar]
5 Kong J K, Kim I S. Wear[J], 2003, 255(7-12): 1174 [Baidu Scholar]
7 Jeong S H, Cho C W, Lee Y Z. Tribology International[J], 2005, 38(3): 283 [Baidu Scholar]
9 Li Xiaohui, Huang Fa, Wang Jianqiu et al. Acta Metallurgical Sinica [J], 2011, 47(7): 847 (in Chinese) [Baidu Scholar]
10 Kuang W J, Was G S. Corrosion Science[J], 2015, 97: 107 [Baidu Scholar]
11 Tang Z M, Meng F J, Zhang P Z et al. Rare Metal Materials and Engineering[J], 2019, 48(11): 3541 [Baidu Scholar]
12 Park I G, Lee C S, Hwang S S et al. Metals and Materials International[J], 2005, 11: 401 [Baidu Scholar]
13 Ahn S J, Rao V S, Kwon H S et al. Corrosion Science[J], 2006, 48(5): 1137 [Baidu Scholar]
15 Memg F J, Wang J Q, Han E H et al. Corrosion Science[J], 2009, 51(11): 2761 [Baidu Scholar]
16 Shoemaker C E, Cordovi M, Stein A A. Workshop on Thermally Treated Alloy 690 Tubes for Nuclear Steam Generators: Proceedings, EPRI-NP-4665M-SR[R]. Palo Alto: EPRI, 1986 [Baidu Scholar]
17 Castano-Matin M L. Proceedings of the 6th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems -Water Reactors[C]. San Diego: The Minerals, Metals & Materials Society, 1993 [Baidu Scholar]
18 Staehle R W, Gorman J A. Corrosion[J], 2003, 59(11): 931 [Baidu Scholar]
19 Zhao L Y, Cui Y H, Yang F Q et al. Rare Metal Materials and Engineering[J], 2018, 47(5): 1399 [Baidu Scholar]
20 Sykes L J. Analysis of Steam Generator Tubing from Oconee Unit 1 Nuclear Station, TR-106484[R]. Palo Alto: EPRI, [Baidu Scholar]
1997 [Baidu Scholar]
21 Kuchirka P J, Blaszkiewicz, Byers W A et al. Oconee 2 Steam Generator Tube Examination, TR-106863[R]. Palo Alto: EPRI, 1997 [Baidu Scholar]
22 MacDonal P E, Shah V N, Ward L W et al. Steam Generator Tube Failures, NUREG/CR-6365[R]. Washington: NRC, [Baidu Scholar]
1996 [Baidu Scholar]
23 Rochester D P. Water Chemistry of Nuclear Reactor Systems[M]. London: Thomas Telford Publishing, 2001 [Baidu Scholar]
25 Lim M K, Oh S D, Lee Y Z. Nuclear Engineering and Design[J], 2003, 226(2): 97 [Baidu Scholar]
26 Han Y L, Mei J N, Peng Q J et al. Corrosion Science[J], 2015, 98: 72 [Baidu Scholar]
27 Han Y L, Mei J N, Peng Q J et al. Corrosion Science[J], 2016, 112: 625 [Baidu Scholar]