Characterization of interfacial reactions and oxide films on 316L stainless steel in various simulated PWR primary water environments

Journal of Nuclear Materials 489 (2017) 137e149

Contents lists available at ScienceDirect

Journal of Nuclear Materials journal homepage: www.elsevier.com/locate/jnucmat

Characterization of interfacial reactions and oxide films on 316L stainless steel in various simulated PWR primary water environments Junjie Chen a, b, Qian Xiao a, b, Zhanpeng Lu a, b, c, *, Xiangkun Ru a, Hao Peng a, Qi Xiong a, Hongjuan Li a a

Institute of Materials Science, School of Materials Science and Engineering, Shanghai University, Mailbox 269, 149 Yanchang Road, Shanghai, 200072, China State Key Laboratory of Advanced Special Steels, Shanghai University, 149 Yanchang Road, Shanghai, 200072, China c Shanghai Key Laboratory of Advanced Ferrometallurgy, Shanghai University, 149 Yanchang Road, Shanghai, 200072, China b

h i g h l i g h t s
Long-term EIS measurements of 316L SS in simulated PWR primary water.
Highest charge-transfer resistance and oxide film resistance in oxygenated water.
Highest electric double-layer capacitance and oxide film CPE in hydrogenated water.
Similar compositions, different shapes of oxides in deaerated/hydrogenated water.
Inner layer Cr-rich in hydrogenated/deaerated water, Ni-rich in oxygenated water.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 September 2016 Received in revised form 4 February 2017 Accepted 19 March 2017 Available online 22 March 2017

The effect of water chemistry on the electrochemical and oxidizing behaviors of 316L SS was investigated in hydrogenated, deaerated and oxygenated PWR primary water at 310  C. Water chemistry significantly influenced the electrochemical impedance spectroscopy parameters. The highest charge-transfer resistance and oxide-film resistance occurred in oxygenated water. The highest electric double-layer capacitance and constant phase element of the oxide film were in hydrogenated water. The oxide films formed in deaerated and hydrogenated environments were similar in composition but different in morphology. An oxide film with spinel outer particles and a compact and Cr-rich inner layer was formed in both hydrogenated and deaerated water. Larger and more loosely distributed outer oxide particles were formed in deaerated water. In oxygenated water, an oxide film with hematite outer particles and a porous and Ni-rich inner layer was formed. The reaction kinetics parameters obtained by electrochemical impedance spectroscopy measurements and oxidation film properties relating to the steady or quasisteady state conditions in the time-period of measurements could provide fundamental information for understanding stress corrosion cracking processes and controlling parameters. © 2017 Elsevier B.V. All rights reserved.

Keywords: 316L stainless steel Pressurized water reactor Corrosion Water chemistry Electrochemical impedance spectroscopy Transmission electron microscope

1. Introduction Austenitic stainless steels (SS) are widely used in pressurized water reactor (PWR) nuclear power plants as important structural materials due to their combined good mechanical properties and corrosion resistance. The corrosion resistance due to the passive

* Corresponding author. Institute of Materials Science, School of Materials Science and Engineering, Shanghai University, Mailbox 269, 149 Yanchang Road, Shanghai, 200072, China. E-mail address: [email protected] (Z. Lu). http://dx.doi.org/10.1016/j.jnucmat.2017.03.029 0022-3115/© 2017 Elsevier B.V. All rights reserved.

film formed on the surfaces of austenitic alloys in high-temperature water environments during the reactor operation is of great importance. The nucleation and propagation of localized corrosion, and stress corrosion cracking (SCC) in particular, are generally recognized to be related to the properties of the oxide film formed on the metal surface [1e8]. Water chemistry exerts a significant influence on the oxide film properties and is related to the oxidation kinetics [9e15]. The hydrogenated water chemistry in the PWR water has been applied to maintain a relatively low electrochemical corrosion potential to reduce the SCC susceptibility of the materials [10] The normal range of the dissolved hydrogen (DH)

138

J. Chen et al. / Journal of Nuclear Materials 489 (2017) 137e149

concentration in operating PWR plants is 25e50 cm3 (STP) H2/kg H2O [16]. Oxygen could be introduced into the primary loop system by adding aerated water, or by adding oxygen or H2O2 during the plant shutdown process for PWRs. This would stabilize the oxide and thus decrease the release of radioactive species into the coolant [10]. It has been reported that the duplex oxide film formed on stainless steels in high-temperature water contains more Cr in the inner layer and more Fe in the outer layer. Several different mechanisms of high-temperature aqueous oxidization were proposed to rationalize this phenomenon [17e19]. Stellwag [17] suggested that the formation of the inner layer could be related to a solid-state growth mechanism. The formation of the outer, coarse-grained oxide layer was because of the precipitation of dissolved metal ions. The enrichment of chromium in the inner layer could be explained by preferential dissolution of Fe and Ni during passivation and to the low diffusivity and solubility of chromium in the spinel lattice. Ziemniak et al. [18] thought that corrosion occurred in a non-selective manner and that the formation of the corrosion film consisted of two spinel oxide layers with similar compositions implied that immiscibility played an important role in the phase separation process. Robertson [19] found that the corrosion rate of SS in high temperature was controlled by the solid-state diffusion of Fe ions along grain boundaries in the oxide layer. A duplex layered oxide was formed with the inner layer growing by the ingress of water along oxide micropores and the outer layer growing by the diffusion of metal ions. The location of an alloy component across the oxide layer depended on its diffusion rate in the oxide. The oxide films formed on austenitic SS in high-temperature water environments are closely related to water chemistry [10e15]. Kim [11] found that the oxide film formed on SS in cyclic normal and hydrogen water chemistries mainly consisted of two layers: an outer layer with large spinel particles and small hematite particles and a fine-grained inner layer of chromium-enriched spinel. Kumai and Devine [12] suggested that the outer oxide layer formed on 304 SS was composed of M3O4 when dissolved oxygen (DO) was below a critical value; as DO increased, M2O3 particles gradually formed with roughening of the preformed M3O4 particles. Kuang [13] investigated the characteristics of the oxide films formed on 304 stainless steel at different DO concentrations in 290  C water. Less hematite and more spinel oxides formed on a fresh sample as the DO decreased. Meanwhile, the relative Cr content in the film increased. Terachi et al. [20] found that the corrosion rate of 316L SS increased slightly with increasing DH concentration in simulated PWR primary water. Han et al. [21,22] found that the oxide film of 316L SS formed in hydrogenated and deaerated PWR water was Cr-rich in the inner layer and Fe-rich in the outer layer. Xu et al. [10] reported that in cyclic hydrogenated and oxygenated water chemistry in high temperature water, the film resistance during the periods of DO was much higher than that during the periods of DH for 316 SS. The SCC behaviors of austenitic alloys in high-temperature water are also affected by water chemistry. It was found that for the 316L HAZ in a boiling water reactor (BWR) environment, switching the water chemistry from oxygen-bearing water to hydrogen-bearing water drastically decreased the electrochemical potential and the crack growth rate [23]. It was reported that on cold-rolled 316L, more extensive intergranular stress corrosion cracks were found in deaerated PWR water than in hydrogenated PWR water [24]. The crack growth rate (CGR) decreased with the changing water chemistry from 2 ppm DO in oxygenated water to de-oxygenated water with DO < 5 ppb, and decreased further after changing to hydrogen-saturated water for 3D cold-rolled 316NG SS in BWR water [25]. Andresen et al. [26] found that there was a peak

in crack growth rate vs. dissolved hydrogen in BWR water. In a PWR environment, it was found that the CGRs of 316L HAZ increased with increasing DO, and CGR in hydrogenated water was approximately one order of magnitude slower than in oxygenated water [27]. Meng et al. [28] found that the CGRs of cold-rolled 316L SS in simulated PWR water were not affected by the DH when it was lower than 5 cm3 (STP) H2/kg H2O and that the SCC growth rate decreases with increasing DH in the range of 5e50 cm3 (STP) H2/kg H2O. It was also found that the corrosion potential of stainless steel was weakly dependent on the DH, while the crack growth rate showed a significant decrease with increasing DH [28]. The possibility of predicting SCC from electrochemical measurements has aroused considerable interest in the nuclear industry [29]. Electrochemical monitoring can obtain in-situ information on water chemistry and corrosion conditions relevant for light water reactor systems. The applications of electrochemical monitoring in nuclear power environments include: optimization of hydrogen addition during hydrogen water chemistry in BWRs, hydrogen control in the primary system of PWRs and oxygen control and detection of redox transients in PWR secondary systems [30]. Electrochemical corrosion potential (ECP) monitoring is an important factor in operating BWR environments [31]. Reduction of ECP of the BWR vessel and internal stainless steel surfaces is essential for mitigating intergranular stress corrosion cracking (IGSCC) [32]. Andresen et al. [5] reported that the CGRs of cool worked 300 series SS in high-temperature pure water containing 2000 ppb DO decreased with a decreasing corrosion potential. In PWR environments, measuring ECP is an effective way to assess irradiation-assisted stress corrosion cracking (IASCC) which is influenced by high-temperature water environment, mechanical stresses and the presence of irradiation. It determines whether it is likely that IASCC occurs (a high value of the potential) or not (a low value) [33]. Arioka et al. [6] found that the IGSCC growth rate of non-sensitized, cold-worked 316 SS in a simulated PWR environment decreases gradually with a decrease in the electrochemical potential. It is necessary to clarify the relationship between electrochemical behavior and oxide film properties of SS in hightemperature high-pressure aqueous environments for a mechanistic understanding of the corrosion and SCC behaviors of SS in nuclear plant coolant environments. The objective of this work is to investigate the effects of various water chemistries on the electrochemical behavior and oxide film properties of 316L SS in simulated PWR primary water environments including hydrogenated, de-aerated and oxygenated water. Long-term electrochemical impedance spectroscopy (EIS) measurements of the interfacial kinetics are used in combination with the characterization of oxide films by Raman spectral analysis, scanning electron microscopy (SEM) and transmission electron microscopy (TEM) observation. 2. Experimental The elemental composition of the 316L SS used in the present work is listed in Table 1. Samples were cut from a 22-mm-thick plate. The plate was solution-annealed (SA) at 1100  C for 2 h followed by a water quench. The size of the specimen for exposure and electrochemical test was 10 mm  10 mm  3 mm. They were mechanically abraded Table 1 Chemical composition (wt.%) of 316L SS. C

Si

Mn

S

P

Cr

Ni

Mo

Fe

0.019

0.320

1.60

0.006

0.027

16.39

10.21

2.12

Bal.

J. Chen et al. / Journal of Nuclear Materials 489 (2017) 137e149

with emery paper successively up to #1500 and degreased with ethanol before exposure and electrochemical tests. All tests were conducted in a 316 SS autoclave equipped with water circulating loop system. To reduce the effect of autoclave body on the oxide formation on the specimens, the autoclave has been operated in the test solution for more than one week to stabilize the autoclave walls before the immersion tests. The testing solution was simulated PWR primary water (B: 1200 ppm (ppm in this paper refers to weight percentages) as H3BO3, Li: 2.0 ppm as LiOH, flow rate: ~5 L/ h, Pressure ~ 12.2 MPa) at 310  C. The pH of the test solution at 310  C was 6.99 as calculated using the pHSC4 software developed by Duke Power Company, Nuclear Chemistry. The normal PWR primary water chemistry or hydrogenated water chemistry with a DO level <5 ppb and a DH level of 2.65 ppm (30 ml/kg water STP) was achieved by H2 purging in the make-up water tank and maintaining an H2 overpressure of approximately 0.065 MPa. The deaerated water chemistry with a DH and DO level of less than 5 ppb was achieved by continuously bubbling N2 in the make-up water tank. The oxygenated water chemistry had a DO of 8 ppm and DH < 5 ppb. A three-electrode system was used for the electrochemical measurements. A polytetrafluoroethylene- (PTFE) insulated 316L SS lead wire was welded to the 316L electrode. The reference electrode was an Ag/AgCl pressure-balanced external reference electrode. The reference solution was 0.1 mol/L KCl. The reference electrode was maintained at 25  C and system pressure via the reference solution bridge. All electrode potentials in the present work have been converted to standard hydrogen electrode (SHE) values according to the following relationship [34]:

139

HR800 Raman spectrometer with a laser wavelength of 532 nm. The cross-sections of the oxide films formed on the surface of specimens were examined via TEM. For analysis, thin foil specimens were prepared using a FEI Helios Nanolab 600i dual-beam focused ion beam (DB-FIB) with Ga ion sputtering after a protective Pt strap was deposited on the oxide films. TEM and selected area electron diffraction analyses (SAED) were conducted using a JEOL JEM-2010F TEM equipped with an energy dispersive spectrometer (EDS) system operating at 200 kV. High-angle annular dark-field (HAADF) images were obtained.

3. Result and discussion 3.1. Electrochemical behaviors Fig. 1 shows the evolution of OCP for 316L SS vs. time in oxygenated, hydrogenated and deaerated PWR primary water. The OCP measurement began immediately at the time when the temperature reached 310  C. In the hydrogenated and deaerated environments, the OCP moved to negative with increasing time. In the hydrogenated environment, the OCP decreased rapidly approximately 0.28 V in the first 30 h. Then, the rate of change of OCP

ESHE ¼ Eobs þ 0:2866  0:001ðT  T0 Þ þ 1:745  107 ðT  T0 Þ2  30:3  109 ðT  T0 Þ3 (1) where ESHE represents the electrode potential vs. SHE, Eobs the measured electrode potential, T the experimental temperature (in  C) and T is 25  C. 0 The EIS was measured at OCP in potentiostatic mode in the frequency range from 100 Hz to 10 mHz with an AC amplitude of 20 mV (rms). All electrochemical measurements were carried out with a Gamry Reference 600 system. The surface morphologies of specimens after exposure tests were observed by CamScan Apollo 300 thermal field SEM. The surface layer of the corrosion oxides was analyzed on a LabRam

Fig. 1. The evolution of OCP for 316L SS vs. time in various simulated PWR primary water environments.

Fig. 2. The typical EIS diagrams at various immersion periods for 316L SS in hydrogenated PWR primary water. (a) Nyquist plot, (b) Bode plot.

140

J. Chen et al. / Journal of Nuclear Materials 489 (2017) 137e149

slowed down and the OCP gradually reached a nearly steady situation. The final OCP at 500 h of 316L SS in hydrogenated environment was approximately 0.40 V lower than that in the initial time. In the deaerated environment, the OCP decreased approximately 0.19 V in the first 192 h, then slowly moved in the negative direction and reached the final OCP at 1046 h, which was approximately 0.20 V lower than the initial value. In an oxygenated environment, the OCP of 316L SS gradually increased with time during the whole exposure period, and the OCP shifted approximately 0.24 V in the positive direction from the initial value to the final. Figs. 2e4 show the typical EIS diagrams at various immersion periods for 316L SS in hydrogenated, deaerated and oxygenated PWR primary water. The Nyquist diagrams in Figs. 2(a), 3(a) and 4(a) for 316L SS in hydrogenated, deaerated and oxygenated water showed the characteristics of capacitance reactance arcs. In all the test environments, the radius of the capacitance reactance arc increased with increasing exposure time. At a similar immersion time, the radius of the capacitance arc of EIS for 316L SS was the smallest in hydrogenated water and the greatest in oxygenated water. In the three types of water chemistry, the impedance magnitude jZj of the Bode plots for 316L SS increased with immersion time, and the phase angle of the Bode plots decreased with

the immersion time at lower frequency, which presented a variation tendency similar to the variation of Nyquist plots as influenced by water chemistry, as shown in Figs. 2(b), 3(b) and 4(b). A highfrequency region in EIS diagrams should mainly reflect the electrochemical characteristics of the oxide film and the lowerfrequency region should reflect the characteristics of the electric double layer and Faraday processes [14]. An equivalent circuit R-(Q(RW))-(CR) was used by Bosch et al. [35,36] to investigate the electrochemical corrosion behavior of AISI 304 and AISI 316 in primary water at 300  C. Macak et al. [37] used the R-(Q(RW))-(RQ) to analyze the EIS results of austenitic stainless steel 08CH18N10T in high-temperature water. Qiu et al. [14] adopted R-(Q(RW))-(QR)-(CR) to fit the EIS results of Alloy 600 in high-temperature water with different dissolved hydrogen concentrations. Xu et al. [38,39] used R-(Q(RW))-(RQ) to fit the EIS results of Alloy 52 and Alloy 182 in high-temperature water with a cyclic hydrogenated and oxygenated water chemistry. In the equivalent circuits mentioned above, R is solution resistance, (RC) or (RQ) is adopted to represent the EIS plots in high frequency region; (Q(RW)) represents the EIS plots in low frequency region, where Q is the constant phase element (CPE). A CPE was introduced to represent a non-ideal capacitance. The formation of a duplex oxide film has been reported on the surface of austenitic stainless

Fig. 3. The typical EIS diagrams at various immersion periods for 316L SS in deaerated PWR primary water. (a) Nyquist plot, (b) Bode plot.

Fig. 4. The typical EIS diagrams at various immersion periods for 316L SS in oxygenated PWR primary water. (a) Nyquist plot, (b) Bode plot.

J. Chen et al. / Journal of Nuclear Materials 489 (2017) 137e149

steel exposed to high-temperature water [4,18,40e45]. The nonideal semi-circle in Nyquist diagrams might be affected by many features such as surface roughness, frequency dispersion of time constants, porosity mass transport effects and relaxation effects [46]. The complex structure and chemical compositions of the oxide film formed in high-temperature water might lead to the deviation of ideal circuit element [47e49]. A CPE has an impedance equal to 1/Y0(ju)n, where Y0 is the CPE coefficient and n is the CPE exponent. Accordingly, the equivalent circuit R-(Q(RW))-(RQ) is thought to be more appropriate than R-(Q(RW))-(RC) for the systems in the present work. The circulating test solution might attribute to the deviation of ideal Warburg impedance. The electric double layer at the interface between the oxide on the metal surface and the solution could be combined into the outer oxide layer [50]. Depending on the fitted equivalent circuits in the literatures [14,35e39] and the actual fitting in the present work, a simple equivalent circuit R-(Q(RW))-(RQ) used to simulate the EIS results is illustrated in Fig. 5. The meaning of each element is described

141

below: RS is the solution resistance; Rct is the charge-transfer resistance of the Faraday processes; Cdl is used as the electric double layer; Rf represents the resistance for the oxide film and CPEf is the constant phase element for the outer oxide film; W is the Warburg impedance. All the EIS results were fitted using Gamry Echem Analyst 6.32 software. To analyze the change in the oxide film with time in different water chemistry environments, some parameters from the fitting results are shown in Fig. 6. The Rct and Rf of 316L SS exposed in all the test environments increased with increasing immersion time. After the stabilization of the electrode surface state, the Rct and Rf were the highest in oxygenated water and the lowest in hydrogenated water. In all the test environments, the Rct was higher than the Rf, the Cdl was higher than the CPEf, and the Cdl and CPEf decreased with increasing immersion time. The Cdl and CPEf were the highest in the hydrogenated water and the lowest in the oxygenated water. 3.2. Oxide film morphologies and characterizations

Fig. 5. The equivalent circuit for EIS simulation.

Fig. 7 shows the Raman spectra of the oxide films formed on the surfaces of 316L SS immersed in various simulated PWR primary water environments. In hydrogenated and deaerated water environments, spinel with Raman peaks at 328, 478, 572 and 690 cm1 was identified. In oxygenated water, spinel with Raman peaks at 334, 495, 662 and 700 cm1 and hematite with Raman peaks at 225, 244, 295, 412, 495 and 612 cm1 were identified.

Fig. 6. Change in (a) Rct, (b) Cdl, (c) Rf and (d) CPEf as a function of immersion time for 316L SS in hydrogenated, deaerated and oxygenated PWR primary water environments at 310  C.

142

J. Chen et al. / Journal of Nuclear Materials 489 (2017) 137e149

Fig. 7. Raman spectra of the oxide films formed on 316L SS immersed in various simulated PWR primary water environments.

Fig. 8 shows the SEM morphologies of the oxide films formed on 316L SS exposed to various simulated PWR primary water environments. For the specimen in the hydrogenated environment, the distribution of the oxide particles was inhomogeneous, whereby some large oxide particles were dispersed on the surface and small oxide particles were observed in the sub-surface layer, as shown in Fig. 8a. For the specimen in the deaerated environment, the faceted oxide particles with straight sides and planar faces were loosely dispersed on the surface, as shown in Fig. 8b. The size of the oxide particles was not uniform. Under the outer layer of oxide particles, the specimen surface was covered by a compact layer. In oxygenated water, the surface of the specimen was predominantly covered by equiaxial oxide particles with curving edges and blunt angles, as shown in Fig. 8c. Fig. 9a shows the TEM cross-sectional morphology of the oxide film grown on 316L SS after exposure to hydrogenated PWR primary water at 310  C for 500 h. The thickness of the inner layer of oxide film determined by TEM was approximately 72 ± 16 nm. The outer layer was composed of large oxide grains and the inner layer was a type of compact oxide film. Fig. 9b and c shows the elemental concentration profiles determined across the oxide film at different places marked in Fig. 9a. The outer oxide layer was enriched in Fe but depleted in Cr with respect to the inner oxide layer. The Cr content in the inner oxide layer was close to that in the matrix, whereas the Fe content was much lower than that in the matrix. Fig. 10a shows other TEM images of the oxide film formed on the 316L SS after exposure to hydrogenated PWR primary water at 310  C for 500 h. Fig. 10b and c shows the associated SAED patterns of the different oxide particles marked A and B in the outer oxide layer. The diffraction pattern of the outer oxide particles indicated that oxide particles A and B had a spinel-type crystal structure, likely in the form of (Ni, Fe)Fe2O4, consistent with the Raman results shown in Fig. 7. Moreover, EDS quantification of the outer oxide particle layer revealed that the atomic ratio of Fe/Cr/Ni was approximately 93.82: 2.60: 3.57 for particle A and 92.86: 2.98: 4.16 for particle B. The SAED pattern of region C from the inner layer showed that the inner layer consisted of fine spinel oxides. TEMEDS analysis revealed that the atomic percentage values of Fe, Cr, and Ni were non-stoichiometric and varied across the areas. Fig. 11a shows the cross-sectional morphology of the oxide film formed on 316L SS after exposure to deaerated PWR primary water at 310  C for 1170 h. The thickness of the inner oxide layer

Fig. 8. SEM (secondary electron image) morphologies of the oxide films formed on 316L SS exposed in various simulated PWR primary water environments: (a) hydrogenated, 500 h, (b) deaerated, 1170 h, and (c) oxygenated, 1012 h.

determined by TEM was approximately 143 ± 90 nm. The outer layer was composed of large oxide grains and the inner layer was a type of compact oxide film. Fig. 11b and c shows the elemental concentration profiles determined across the oxide film at different places marked in Fig. 11a. The result showed that the outer oxide layer was enriched in Fe, but it was depleted in Cr with respect to

J. Chen et al. / Journal of Nuclear Materials 489 (2017) 137e149

143

Fig. 9. (a) STEM-HAADF image of the oxide film grown on 316L SS exposed in hydrogenated PWR primary water for at 310  C 500 h; (b) EDS line scan profile of the oxide film shown in (a), marked as line 1, and (c) EDS line scan profile of the oxide film shown in (a), marked as line 2.

the inner oxide layer. The Cr content in the inner oxide layer was similar to that in the matrix, whereas the Fe content was much lower than that in the matrix. Fig. 12a shows other TEM images of the oxide film formed on the 316L SS after exposure to deaerated PWR primary water at 310  C for 1170 h. Fig. 12b and c shows the associated SAED patterns of the different oxide particles marked A and B in the outer oxide layer. The diffraction pattern of the outer oxide particles indicates that oxide particles A and B had a spinel-type crystal structure, likely in the form of (Ni, Fe)Fe2O4, consistent with the Raman result shown in Fig. 7. Moreover, EDS quantification of the outer oxide particle layer revealed that the atomic ratio of Fe/Cr/Ni was approximately 87.44:1.50:11.06 for particle A and 96.61: 0.76: 2.63 for particle B. The SAED pattern of region C from the inner layer showed that the inner layer consisted of fine spinel oxides. The atomic percentage values of Fe, Cr, and Ni were non-stoichiometric according to TEMEDS analysis. Fig. 13a shows the cross-sectional morphology of the oxide film grown on 316L SS after exposure to oxygenated PWR primary water at 310  C for 1012 h. The thickness of the inner layer oxide film determined by TEM was approximately 165 ± 80 nm. The outer layer was composed of large oxide grains. The inner layer was a type of less compact oxide film. Fig. 13b, c and d shows the elemental concentration profiles determined across the oxide film at different places marked in Fig. 13a. The outer oxide layer was enriched in Fe. The Ni content in the inner oxide layer was similar to that in the matrix, whereas the Cr and Fe contents were much lower than those in the matrix. Fig. 14a shows other TEM images of the oxide film formed on 316L SS after exposure to oxygenated PWR primary water at 310  C for 1012 h. Fig. 14c, d and e show the associated SAED patterns of

the different oxide particles marked A, B and C in the outer oxide layer. The diffraction patterns of the outer oxide particles indicated that oxide particles A, B and C had a hematite-type crystal structure, consistent with the Raman result shown in Fig. 7. Moreover, EDS quantification of the outer oxide particle layer showed that the atomic ratio of Fe/Cr/Ni was approximately 94.56: 5.54: 0 for particle A, 84.56: 14.93: 0.51 for particle B and 89.37: 9.46: 1.17 for particle C. The SAED pattern of region D from the inner layer combined with the EDS results showed that the inner layer consisted of non-stoichiometric fine spinel oxides, as shown in Fig. 14f. 3.3. Effect of water chemistry on interface reaction and oxide film properties Electrochemical measurements and characterization of oxide films show that the water chemistry and corresponding electrochemical conditions have a significant effect on the oxidation behaviors of 316L SS in simulated PWR water. The OCP in the deaerated water was approximately 0.05 V higher than that in the hydrogenated water, which was approximately 0.66 V lower than that in the oxygenated water, as shown in Fig. 1. OCP is one of the key electrochemical parameters for evaluating the corrosion behavior of materials in high-temperature water. The EIS test is an effective way of investigating the electrochemical behaviors of austenitic alloys in high-temperature water. The EIS results of Alloy 600 in simulated PWR water indicated that the oxide film had a duplex structure [14]. In a simulated BWR environment with a highly oxidizing agent, an oxide film of higher electric resistance formed on 304 SS and was observed via the EIS test [51]. Wang et al. [52] found that in simulated primary water at

144

J. Chen et al. / Journal of Nuclear Materials 489 (2017) 137e149

Fig. 10. TEM images and SAED patterns of 316L SS exposed in hydrogenated PWR primary water at 310  C for 500 h: (a) TEM (bright field) image, (b) SAED pattern of oxide particle A marked in (a), (c) SAED pattern of oxide particle B marked in (a), (d) SAED pattern of oxide region C marked in (a) and (e) high resolution TEM (bright field) image of the boundary of the inner layer and the outer particles.

DO < 0.01 ppm, the electrochemical impedance of Alloy 690TT at low frequency was higher than that of Alloy 800SN, whereas at DO > ¼ 0.1 ppm, the result was reversed. The capacitance associated with the oxide film can be expressed as C ¼ εrε0/d where d is the film thickness, εr is the relative dielectric constant of the oxide and ε0 is the permittivity of a vacuum [53]. Accordingly, the capacitance is inversely proportional to the film thickness. A decreasing film capacitance will reflect increasing film thickness. The EIS fitting results in the present work showed that CPE values of the oxide films formed in the three water chemistry environments decreased with time, indicating that the oxide film grew continuously during the test period, as shown in Fig. 6d. The increasing of the oxide film resistances in the three water chemistry environments with increasing time also reflects the thickening of the oxide films with increasing time, as shown in Fig. 6c. According to the fitting results, the Rf in the oxygenated water was the highest: Rf (oxygenated water) > Rf (deaerated water) > Rf (hydrogenated water). The CPEf was the lowest in the oxygenated water and the highest in the hydrogenated water: CPEf (oxygenated water) < CPEf (deaerated) < CPEf (hydrogenated water). These EIS results indicate that the thickness of the oxide film in oxygenated water is the greatest. It has been reported that the capacitance of the electric double layer is higher than that of the oxide films [14]. The Cdl obtained from the fitting results ranged from 102 to 104 S sn/cm2, and the CPEf ranged from 104 to 109 S sn/cm2, as shown in Fig. 6b and d. The Cdl was higher than the CPEf. This can be attributed to the porous outer layer with a high surface area [54]. The effects of water chemistry on the electrochemical properties

of oxide films formed on other austenitic alloys have been reported in the literatures [10,14,15]. Qiu et al. [14] reported that increasing DH content results in the decreasing of inner oxide film resistance and increasing of inner oxide film capacitance which indicated a thinner inner-layer oxide film formed on Alloy 600. Xu et al. [15] found that the oxide film resistance of Alloy 182 increased with decreasing DH in 290  C simulated PWR primary water. It was reported that [10] in cyclic hydrogenated and oxygenated water chemistry in high temperature water, the film resistance in oxygenated periods was much higher than that in hydrogenated periods for Alloy 182, Alloy 52 and 316 SS. Comparing the results in present work to those in literatures, it is found that the oxide film resistance of austenitic alloys is lower in hydrogenated environment and higher in oxygenated environment. The characteristics of the oxide films formed on an alloy in a high-temperature aqueous environment, such as elementary composition, microstructure, thickness and so on, can influence the protection of the oxide film and the corrosion behavior of the alloys [55e58]. The formation of a duplex oxide film has been reported on the surface of austenitic stainless steel exposed to hightemperature water [4,18,40e45]. Several studies [4,18,40e45] described these crystallites of the duplex oxide layer as spinel, which were of the AB2O4 crystallite type (with A ¼ Fe (II) or Ni (II), B ¼ Fe (III) or Cr (III)). The Cr-rich inner layer of fine-grained oxide was compact and very adherent to the base metal, and thus was generally nonporous and very protective. The Fe-rich external layer was composed of non-uniform large grains that had precipitated from the solution.

J. Chen et al. / Journal of Nuclear Materials 489 (2017) 137e149

145

Fig. 11. (a) STEM-HAADF image of the oxide film formed on 316L SS exposed in deaerated PWR primary water at 310  C for 1170 h, (b) EDS line scan profile of the oxide film shown in (a), marked as line 1, and (c) EDS line scan profile of the oxide film shown in (a), marked as line 2.

As shown in Fig. 8, the surface morphology of 316L SS in hightemperature water is strongly affected by water chemistry, i.e. DH, DO and the resultant OCP. In the hydrogenated and deaerated water environments, the large outer oxide particles showed straight edges and planar faces. For the specimen in the hydrogenated environment, the distribution of the oxide particles was inhomogeneous, whereby some large oxide particles were dispersed on the surface and small oxide particles were observed in the sub-surface layer. For the specimen in the deaerated water, the faceted oxide particles were loosely dispersed on the surface. Under the oxide particles, the specimen surface was covered by a compact layer. In the oxygenated water, the surface of the specimen was predominantly covered by equiaxial oxide particles with curving edges and blunt angles. Some literature [18,55,59e61] noted that spinel and hematite particles could be distinguished by their distinct difference in morphology. The spinel particles mainly appear faceted while the hematite particles are mostly equiaxial with curving edges and blunt angles. According to the TEM results shown in Figs. 10 and 12, the outer oxide particles were assigned to a spinel morphology for 316L SS in the hydrogenated and deaerated water, and hematite particles were found for 316L SS in the oxygenated water. The results are consistent with the Raman spectroscopy results shown in Fig. 7. Kumai and Devine [12] revealed that M2O3 formed by the oxidation of the outer surface of M3O4 as the DO concentration gradually increased. The TEM-EDS analysis shown in Figs. 9 and 11 indicated that the inner oxide films for 316L SS in the hydrogenated and deaerated water were Cr-rich. The inner oxide film was compact and no pores

were observed. The Cr enrichment in the inner oxide film for 316L SS in hydrogenated and deaerated water was caused by selective dissolution of Ni and Fe. The diffusion rates of the metallic cations in the oxides decrease in the order of Fe2þ > Ni2þ [ Cr3þ [19,62]. Cr is more easily oxidized at lower electrode potentials [63,64] and its oxide has a lower solubility than Ni and Fe oxides in hydrogenated or deaerated water, resulting in the formation of the Cr-rich oxide on the surface [21,22]. For 316L SS in the oxygenated water, the inner oxide film was Ni-rich. This phenomenon might be due to the highly oxidizing environment with high DO. Miyazawa et al. [65] have reported similar results. The inner layer of 304 SS exposed in the more-oxidizing BWR environment consisted of very fine Nirich spinel-type magnetite, while the inner layer consisted of fine Cr-rich spinel-type magnetite in the less-oxidizing environment. The TEM results show that the inner layer formed on 316L SS in the oxygenated water was more porous than those in the hydrogenated and deaerated water environments, as shown in Fig. 14. The reason for the formation of the porous and Ni-rich inner layer is that Cr is stable as a soluble ion at a high electrochemical potential [63,66] and the Cr in the oxide would dissolve into solution [11,67,68]. The porous base layer mentioned above should be related to the release of Cr. The formation of various oxide films on 316L stainless steel in various high-temperature water environments can be interpreted with the E-pH diagrams. E-pH diagrams for pure Ni, Cr, Fe and the ternary alloy Fe-Cr-Ni at 300  C have been reported by B. Beverskog et al. [63,64,69,70], which are referred to in the following analysis. The calculated pH for the test solutions at 310  C was about 7. The

146

J. Chen et al. / Journal of Nuclear Materials 489 (2017) 137e149

Fig. 12. TEM image and SAED patterns of 316L SS exposed in deaerated PWR primary water at 310  C for 1170 h: (a) TEM (bright field) image, (b) SAED pattern of oxide particle A marked in (a), (c) SAED pattern of oxide particle B marked in (a), (d) SAED pattern of oxide region C marked in (a) and (e) high resolution TEM (bright field) image of the boundary of the inner layer and outer particle.

measured OCP for 316L SS at the final stage of tests was 0.017 V(SHE) in the oxygenated water, 0.641 V(SHE) in the deaerated water and 0.695 V(SHE) in the hydrogenated water. These OCP values are used in the analysis with the potential-pH diagrams by Beverskog et al. [63,64,69,70]. The OCP value for 316L SS in the oxygenated water was in the Fe2O3, NiO and HCrO4 zone, which means that Fe and Ni can form stable oxides while Cr is stable as a soluble ion. The OCP values for 316L SS in the deaerated and hydrogenated water were in the Fe3O4, NiO and Cr2O3 zone, which means that Fe, Cr and Ni can form stable oxides. These oxides can also react to form other types of oxides. The oxides formed on 316L stainless steels are also related to the kinetics for the formation of each oxide. The thermodynamics analysis based on the potentialpH diagram was consistent with the characterization results by SEM and TEM. The EIS results provide kinetics information for the oxidation. Combining the EIS measurements and oxide characterization is expected to be helpful in understanding the oxidation behavior and the SCC processes. The results in the present work showed that the oxide film formed in hydrogenated or deaerated water with lower open circuit potentials and low oxide film resistances was composed of compact Cr-rich inner oxide layer and Fe/Ni-rich outer oxide layer. The oxide film formed in oxygenated water with high open circuit potential and high oxide film resistance was composed of porous Ni-rich inner oxide layer and Fe-rich outer oxide layer. According to the mixed-conduction model (MCM) proposed by Bojinov et al. [71], oxide film exhibits different semi-conductor properties in different potential ranges, showing different electrochemical properties. In a

low potential range, the main charge carrier is oxygen vacancy, showing n-type conductivity of oxide film with low film resistance. While in a high potential range, the cation vacancy would play an important role, showing p-type conductivity of oxide film with high film resistance [38]. The corrosion resistance of Fe-Ni-Cr alloys in high-temperature pressurized water relies mainly on the Crbearing oxides [52]. In oxygenated water at higher potentials, Cr is easier than Fe and Ni to form soluble ion, and cation vacancy is the main charge carrier in oxide film. The transient oxidation rate during SCC is affected by the oxidation kinetics and the film degradation parameter such as film rupture strain [1,2,72] or degradation strain [8,72,73]. The enhancement of crack tip oxidation can be realized via physical degradation mode [1,2,8,72], physicalechemical degradation mode or both [8,72]. The SCC growth rate is determined by both the crack tip oxidation rate kinetics and the crack tip mechanical fields [1,2,8,72,73], as shown in Equations (2) and (3). m da ¼ ka $ðε_ct Þ dt

(2)

ε_ct ¼ εd =td

(3)

ε_ct is crack tip strain rate, ka is oxidation rate constant, εd is film degradation strain, td is the time for the onset of film degradation. The definitions of ka and m depend on the rate-determining step for the crack tip oxidation process. For slip-dissolution/oxidation SCC mechanism [1,2], there is

J. Chen et al. / Journal of Nuclear Materials 489 (2017) 137e149

147

Fig. 13. (a) STEM-HAADF image of the oxide film grown on 316L SS exposed in oxygenated PWR primary water for 1012 h at 310  C, (b) EDS line scan profile of the oxide film shown in (a), marked as line 1, (c) EDS line scan profile of the oxide film shown in (a), marked as line 2 and (d) EDS line scan profile of the oxide film shown in (a), marked as line 3.




Ma $i0 ka ¼ z$r$F$ð1  mÞ



t0 εf

!m (4)

where i0 is active surface oxidation current density, t0 is time for the onset of current decay, m is the slope of the current decay curve, and εf is defined as the film rupture strain for representing εd . For quasi-solid state oxidation SCC mechanism [8,72,73], there is

i h ka ¼ ðk1 Þð1mÞ $ðεd ÞðmÞ

(5)

The oxidation rate constant k1 and the slope for the film recovery processes m are represented by the quasi-solid state oxidation kinetics equation that relates oxide film thickness L with oxidation time t.

i h L ¼ ðk1 Þð1mÞ $ðtÞð1mÞ

(6)

Based on these SCC mechanisms and modeling equations, the transient oxidation processes and relevant parameters contribute to the SCC growth rate. Water chemistry can change the SCC growth rates via its effect on the oxidation rate and the oxide film toughness in terms of film degradation strain. The oxide film properties such as the geometry, chemical composition and microstructure would affect both the electrochemical properties and the film toughness. The reaction kinetics parameters obtained by EIS measurements are relevant to the quasi-steady state conditions at the time period of measurements, which are fundamental for the understanding of the SCC processes. Equations (2)e(6) show that the

transient oxidation parameters rather than the steady or quasisteady state oxidation parameters can be directly related to the SCC element processes [1,2,8,72,73]. It has been found that SCC growth rates of austenitic stainless steels generally increases with increasing dissolved oxygen or open circuit potential [1,2,73]. However, the EIS data after a long-time immersion showed that the oxide film resistance for 316L SS in the oxygenated water was higher than those in the deaerated water or in the hydrogenated water. This can be explained by the EIS data from continuous measurements, as shown in Figs. 2e4. The change of oxide film resistance is more significant in the oxygenated water than in the deaerated or hydrogenated water during the immersion period for about 500 h. It implies that the film degradation would result in an overall higher oxidation rate in the oxygenated water than in the deaerated or hydrogenated water, which needs further investigations focusing on short periods tests or transient oxidation rate tests [2,8,72]. Another margin between the electrochemical test results and the SCC behavior is the direct stress/strain effect on oxidation rate, which has been reported and analyzed [73,74] and more fundamental tests are required for obtaining quantitative data for alloys in high temperature water environments. Both experimental and plant data have shown the strong impact of water chemistry on the SCC behavior of austenitic stainless steels in high temperature water environments [1e6,25,28]. The detailed analysis of oxide film properties and the electrochemical parameters would provide the fundamental information for understanding the SCC processes, predicting the SCC behavior and the development of SCC mitigation techniques. In the process of nuclear power plant (NPP) operation, well-

148

J. Chen et al. / Journal of Nuclear Materials 489 (2017) 137e149

Fig. 14. TEM image and SAED patterns of 316L SS exposed in oxygenated PWR primary water for 1012 h at 310  C: (a) TEM (bright field) image; (b) high resolution TEM (bright field) image of the boundary of the inner layer and outer particle, (c) SAED pattern of oxide particle A marked in (a), (d) SAED pattern of oxide particle B marked in (a), (e) SAED pattern of oxide particle C marked in (a) and (f) SAED pattern of oxide region D marked in (a).

managed water chemistry has an important contribution. For example, during the shutdown of PWRs, oxygenation would be implemented to decrease the radiation effects on circuit components [9]. During the start-up of PWRs, the oxygen is removed and hydrogen is injected, but a certain amount of oxygen may be present, especially in the occluded volumes such as in low-flow regions. SCC of austenitic stainless steel in oxygen-bearing high temperature water has been observed in both laboratory tests [75,76] and in the plants [77]. SCC events on stainless steel initiating from the primary side occurred in occluded environments with high oxygen contents [78]. The transition of water chemistry during start-up and shutdown of NPPs would influence the oxide film formed on primary components and the resultant resistance to cracking. A suitable water chemistry control program helps NPPs to ensure the integrity of the reactor coolant system. 4. Conclusions Long-term EIS measurements and oxide film characterization by SEM, Raman spectroscopy and high-resolution TEM are used to investigate the oxidation behaviors of 316L SS in hydrogenated,

deaerated and oxygenated PWR primary water at 310  C. 1. The OCP of 316L SS in the deaerated water is approximately 0.05 V higher than that in the hydrogenated water, which is approximately 0.66 V lower than that in the oxygenated water. 2. Dissolved oxygen or hydrogen in high-temperature water significantly affects the EIS parameters. The Rct and Rf are the highest in the oxygenated water and the lowest in the hydrogenated water. The Cdl and CPEf were the highest in the hydrogenated water and the lowest in the oxygenated water. 3. The oxide film grew with immersion time in various hightemperature water environments. The Rct and Rf increased with increasing immersion time. The Cdl and CPEf decreased with increasing immersion time. 4. Oxide films formed in the hydrogenated water and in the deaerated water are similar in chemical composition but very different in morphology. The outer layer is composed of spinel oxides, and the inner layer is compact and Cr-rich. Outer oxide particles formed in the deaerated water were larger and more loosely distributed than those formed in the hydrogenated water.

J. Chen et al. / Journal of Nuclear Materials 489 (2017) 137e149

5. High DO in high-temperature water results in a large change in the composition and morphology of the oxide film. In oxygenated water, an oxide film with hematite outer particles and Nirich inner layer is formed. Equiaxial outer oxide particles with curving edges and blunt angles cover the surface of 316L SS in the oxygenated water, while outer oxide particles with straight edges and planar faces are found in the hydrogenated and deaerated water environments. 6. The results of electrochemical measurements and oxidation tests could evaluate the time-dependent evolution of reaction rate of materials under various high temperature water chemistry conditions. The variation of electrochemical parameters and oxidation behaviors relating to the steady or quasi-steady state oxidation at the time-period of measurements could provide the fundamental information for understanding the SCC processes. Acknowledgements This work has been supported by Shanghai Municipal Commission of Economy and Informatization (No. T-221715003), the National Natural Science Foundation of China (51571138), and the International Cooperative Project sponsored by the Science and Technology Commission of Shanghai Municipality, China (13520721200). The support from the Instrument Analytical and Research Center, Shanghai University is acknowledged. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]

[11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29]

F.P. Ford, Corrosion 52 (1996) 375e395. P.L. Andresen, F. Peter Ford, Mater. Sci. Eng. A 103 (1988) 167e184. B.T. Lu, J.L. Luo, Y.C. Lu, Electrochim. Acta 53 (2008) 4122e4136. T. Terachi, K. Fujii, K. Arioka, J. Nucl, Sci. Technol. 42 (2005) 225e232. P.L. Andresen, M.M. Morra, Corrosion 64 (2008) 15e29. K. Arioka, T. Yamada, T. Miyamoto, T. Terachi, Corrosion 67 (2011) 691e706. Z. Lu, T. Shoji, T. Dan, Y. Qiu, T. Yonezawa, Corros. Sci. 52 (2010) 2547e2555. Z. Lu, T. Shoji, H. Xue, C. Fu, J. Press. Vessel Technol. 135 (2013) 214021e21402-9. W. Kuang, X. Wu, E.-H. Han, Corros. Sci. 63 (2012) 259e266. J. Xu, S. Xia, X. Zhong, T. Shoji, N. Industry, C. Hatchery, A. Aoba, Z. Qu, The corrosion behavior of Nickel-based alloys 182 and 52 and 316 stainless steel in cyclic hydrogenated and oxygenated water chemistry in high temperature aqueous environment, in: 17th Int. Conf. Environ. Degrad. Mater. Nucl. Power Syst. e Water React., 2015. Ottawa, Ontario, Canada, p. ENVDEG129. Y.-J. Kim, Corrosion 51 (1995) 849e860. C.S. Kumai, T.M. Devine, Corrosion 63 (2007) 1101e1113. X. Liu, X. Wu, E.H. Han, Corros. Sci. 53 (2011) 3337e3345. Y. Qiu, T. Shoji, Z. Lu, Corros. Sci. 53 (2011) 1983e1989. J. Xu, T. Shoji, C. Jang, Corros. Sci. 97 (2015) 115e125. € m, Energy mater, Mater. Sci. Eng. Energy Syst. 3 (2008) K. Norring, J. Engstro 113e118. B. Stellwag, Corros. Sci. 40 (1998) 337e370. S.E. Ziemniak, M. Hanson, Corros. Sci. 44 (2002) 2209e2230. J. Robertson, Corros. Sci. 32 (1991) 443e465. T. Terachi, T. Yamada, T. Miyamoto, K. Arioka, K. Fukuya, J. Nucl. Sci. Technol. 45 (2008) 975e984. G. Han, Z. Lu, X. Ru, J. Chen, J. Zhang, T. Shoji, Corros. Sci. 106 (2016) 157e171. G. Han, Z. Lu, X. Ru, J. Chen, Q. Xiao, Y. Tian, J. Nucl. Mater 467 (2015) 194e204. Z. Lu, T. Shoji, Y. Takeda, Y. Ito, A. Kai, N. Tsuchiya, Corros. Sci. 50 (2008) 625e638. J. Chen, Z. Lu, Q. Xiao, X. Ru, G. Han, Z. Chen, B. Zhou, T. Shoji, J. Nucl. Mater 472 (2016) 1e12. Z. Lu, T. Shoji, F. Meng, Y. Qiu, T. Dan, H. Xue, Corros. Sci. 53 (2011) 247e262. P.L. Andresen, J. Hickling, A. Ahluwalia, J. Wilson, Corrosion 64 (2008) 707e720. R. Zhu, J. Wang, L. Zhang, Z. Zhang, E. Han, Corros. Sci. 112 (2016) 373e384. F. Meng, Z. Lu, T. Shoji, J. Wang, E. Han, W. Ke, Corros. Sci. 53 (2011) 2558e2565. J.-M. Le Canut, S. Maximovitch, F. Dalard, J. Nucl. Mater 334 (2004) 13e27.

149

[30] A. Molander, Electrochemical measurements in nuclear power environments, in: 15th Int. Conf. Prop. Water Steam, 2008. Berlin, Germany. [31] S. Hettiarachchi, R.M. Horn, Y.J. Kim, P.L. Andresen, Electrochemical corrosion potential (ECP) reduction and crack mitigation experiences with NobleChem™ and on-line NobleChemTM, in: 14th Int. Conf. Environ. Degrad. Mater. Nucl. Power Syst. e Water React., 2009, pp. 1012e1031. Virginia Beach, VA, USA. [32] S. Garcia, S. Williams, C. Custer, B. Colson, J. Giannelli, D. Morgan, J. Varnam, In-process electrochemical corrosion potential monitoring experience at boiling water reactors utilizing on-line Noblechem TM for mitigation of intergranular stress corrosion cracking, in: 17th Int. Conf. Environ. Degrad. Mater. Nucl. Power Syst. e Water React., 2015. Ottawa, Ontario, Canada, p. ENVDEG047. [33] M. Postler, R.W. Bosch, S. Van Dyck, The influence of corrosion potential on SCC of stainless steels in PWR primary coolant environment, in: 11th Int. Conf. Environ. Degrad. Mater. Nucl. Power Syst. d Water React., 2003, pp. 1e8. [34] D.D. Macdonald, J. Electrochem. Soc. 126 (1979) 908. ber, M. Vankeerberghen, J. Nucl. Mater 360 (2007) [35] R.W. Bosch, M. We 304e314. [36] R.W. Bosch, M. Vankeerberghen, Electrochim. Acta 52 (2007) 7538e7544. k, P. Sajdl, P. Ku [37] J. Maca cera, R. Novotný, J. Vosta, Electrochim. Acta 51 (2006) 3566e3577. [38] J. Xu, T. Shoji, J. Nucl. Mater 461 (2015) 10e21. [39] J. Xu, T. Shoji, Corros. Sci. 104 (2016) 248e259. [40] M.C. Bloom, G.N. Newport, W.A. Fraser, J. Electrochem. Soc. 111 (1964) 1343. [41] D.H. Lister, R.D. Davidson, E. McAlpine, Corros. Sci. 27 (1987) 113e140. [42] J.E. Castle, H.G. Masterson, Corros. Sci. 6 (1966) 93e104. [43] J. Robertson, Corros. Sci. 29 (1989) 1275e1291. [44] S. Lozano-Perez, D.W. Saxey, T. Yamada, T. Terachi, Scr. Mater 62 (2010) 855e858. [45] G.S. Was, S. Teysseyre, Z. Jiao, Corrosion 62 (2006) 989e1005. [46] R.M. Carranza, M.G. Alvarez, Corros. Sci. 38 (1996) 909e925. [47] A. Machet, A. Galtayries, S. Zanna, L. Klein, V. Maurice, P. Jolivet, M. Foucault, P. Combrade, P. Scott, P. Marcus, Electrochim. Acta 49 (2004) 3957e3964. [48] H. Lefaix-Jeuland, L. Marchetti, S. Perrin, M. Pijolat, M. Sennour, R. Molins, Corros. Sci. 53 (2011) 3914e3922. [49] M. Sennour, L. Marchetti, F. Martin, S. Perrin, R. Molins, M. Pijolat, J. Nucl. Mater 402 (2010) 147e156. [50] X. Liu, X. Wu, E.H. Han, Corros. Sci. 65 (2012) 136e144. [51] S. Uchida, T. Satoh, J. Sugama, N. Yamashiro, Y. Morishima, T. Hirose, T. Miyazawa, Y. Satoh, K. Iinuma, Y. Wada, M. Tachibana, J. Nucl. Sci. Technol. 42 (2005) 66e74. [52] J. Wang, X. Li, F. Huang, Z. Zhang, J. Wang, R.W. Staehle, Corrosion 70 (2014) 598e614. [53] J.-J. Park, S.-I. Pyun, S.-B. Lee, Electrochim. Acta 49 (2004) 281e292. [54] Z. Szklarska-Smialowska, W.-K. Lai, Z. Xia, Corrosion 46 (1990) 853e860. [55] W. Kuang, E.-H. Han, X. Wu, J. Rao, Corros. Sci. 52 (2010) 3654e3660. [56] M. Da Cunha Belo, M. Walls, N.E. Hakiki, J. Corset, E. Picquenard, G. Sagon, €l, Corros. Sci. 40 (1998) 447e463. D. Noe [57] L. Marchetti, S. Perrin, Y. Wouters, F. Martin, M. Pijolat, Electrochim. Acta 55 (2010) 5384e5392. [58] Z. Zhang, J. Wang, E.H. Han, W. Ke, Corros. Sci. 53 (2011) 3623e3635. [59] W. Kuang, X. Wu, E.-H. Han, Corros. Sci. 52 (2010) 4081e4087. [60] Y.-J. Kim, Corrosion 55 (1999) 81e88. [61] W. Kuang, X. Wu, E.-H. Han, L. Ruan, Corros. Sci. 53 (2011) 1107e1114. [62] R. Dieckmann, Solid State Ionics 12 (1984) 1e22. [63] B. Beverskog, I. Puigdomenech, Corrosion 55 (1999) 1077e1087. [64] B. Beverskog, I. Puigdomenech, Corros. Sci. 39 (1997) 43e57. [65] T. Miyazawa, T. Terachi, S. Uchida, T. Satoh, T. Tsukada, Y. Satoh, Y. Wada, H. Hosokawa, J. Nucl. Sci. Technol. 43 (2006) 884e895. [66] R. Lemire, G. McRae, J. Nucl. Mater 294 (2001) 141e147. [67] Y. Hemmi, Y. Uruma, N. Ichikawa, J. Nucl. Sci. Technol. 31 (1994) 443e455. [68] W. Kuang, X. Wu, E.-H. Han, J. Rao, Corros. Sci. 53 (2011) 3853e3860. [69] B. Beverskog, I. Puigdomenech, Corros. Sci. 38 (1996) 2121e2135. [70] B. Beverskog, I. Puigdomenech, Corros. Sci. 39 (1997) 969e980. [71] M. Bojinov, P. Kinnunen, G. Sundholm, Corrosion 59 (2003) 91e103. [72] T. Shoji, Z. Lu, H. Murakami, Corros. Sci. 52 (2010) 769e779. [73] Z. Lu, T. Shoji, S. Yamazaki, K. Ogawa, Corros. Sci. 58 (2012) 211e228. [74] L. Qiao, X. Mao, Acta Metall. Mater 43 (1995) 4001e4006. [75] K. Arioka, T. Yamada, T. Terachi, G. Chiba, Corrosion 62 (2006) 568e575. [76] K. Arioka, T. Yamada, T. Terachi, R.W. Staehle, Corrosion 62 (2006) 74e83. [77] T. Couvant, P. Moulart, L. Legras, P. Bordes, J. Capelle, Y. Rouillon, T. Balon, PWSCC of austenitic stainless steels of heaters of pressurizers, in: 13th Int. Conf. Environ. Degrad. Mater. Nucl. Power Syst. React., CNS, CDROM, 2007. [78] G.O. Ilevbare, P.A. CA, F. Cattant, N.K. Peat, SCC of stainless steels under PWR service conditions, in: Fontevraud 7-Contrib. Mater. Investig. To Improv. Saf. Perform. LWRs, SFEN, Avignon, France, 2010.