- Email: [email protected]
Investigation of passive films formed on the surface of alloy 690 in borate buffer solution
Journal of Nuclear Materials 465 (2015) 418e423
Contents lists available at ScienceDirect
Journal of Nuclear Materials journal homepage: www.elsevier.com/locate/jnucmat
Investigation of passive films formed on the surface of alloy 690 in borate buffer solution Lv Jinlong*, Liang Tongxiang, Wang Chen, Guo Wenli Beijing Key Laboratory of Fine Ceramics, Institute of Nuclear and New Energy Technology, Tsinghua University, Zhongguancun Street, Haidian District, Beijing 100084, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 20 March 2015 Received in revised form 10 June 2015 Accepted 15 June 2015 Available online 20 June 2015
The passive film formed on the surface of the alloy 690 in borate buffer solution was studied by potentiodynamic curves and electrochemical impedance spectroscopy. With the increasing of the passivation potential, the corrosion resistance of the alloy 690 reduced. Moreover, the corrosion resistance of the passive film was the lowest in the vicinity of 0.6 VSCE. These results were supported by XPS and MotteSchottky analyses. The corrosion resistance of the alloy 690 increased with the increasing of passivated potential in borate buffer solution with chloride ion. The chloride ion decreased corrosion resistance of the alloy 690 according to point defect model. © 2015 Elsevier B.V. All rights reserved.
Keywords: Nickel EIS XPS XRD Passive films
1. Introduction Nickel based alloy 690 was candidate materials for the storage of high level waste generated from reprocessing of spent nuclear fuel [1]. Moreover, the alloy 690 also was currently as an alternative to traditional austenitic stainless steel due to its high strength, lower nickel concentration and excellent corrosion resistance [2]. For example, the alloy 690 showed a better corrosion resistance than alloy 800 in simulated primary water [3], therefore, the alloy 690 was suitable to be used in aggressive environment. Due to high stress corrosion cracking and pit corrosion resistance in high temperature pressurized water reactor (PWR), the alloy 690 as the alternative material for the alloy 600, was widely used as stream generator tubing materials in PWR plants [4,5]. Electrochemical properties and growth mechanism of passive films on alloy 690 in high-temperature alkaline environments were investigated widely [6e8]. Moreover, the study showed that alloy 690 dissolved preferentially along the direction which had lower surface free energy in a solution of 0.1 M H2SO4 and 0.1 M NaCl at room temperature [9]. The passive film was formed rapidly and its growth decreased its
* Corresponding author. E-mail addresses: [email protected], (L. Jinlong). http://dx.doi.org/10.1016/j.jnucmat.2015.06.033 0022-3115/© 2015 Elsevier B.V. All rights reserved.
[email protected]
the corrosion rate [10]. The passive films played an important role in preventing corrosion. Therefore, it was very important to deeply understand the form processes of passive films for estimating the structural integrity [11]. The passivation characteristics and corrosion resistance of the passive films formed on the alloy 690 in borate buffer solution without and with chloride ion were investigated in the present study. 2. Experimental procedures Specimens from alloy 690 plate, with chemical composition (wt.%): 0.03C, 0.28 Si, 0.25 Mn, 29.0 Cr, 0.02 Cu, 9.5 Fe, 0.002 S and balance Ni, were cut to cuboid with a dimension of 10 mm 10 mm 2 mm for test. The specimens were solution annealed at 1100 C for 1 h in vacuum, then the sheets were subsequently quenched into water. The copper wires were welded to each specimen for electrical connection through the spot welding. All edges and surfaces of the test specimens were coated with epoxy resin prior to electrochemical test, leaving an active surface area of 10 mm 10 mm exposed to the solution. Before the electrochemical test, all the specimens were abraded with 1000, 2000 and 3000 grit silicon carbide paper and polished with 1.5 mm alumina slurry. The polished specimens were ultrasonically cleaned finally in acetone and ethanol. A conventional three-electrode electrochemical cell was
L. Jinlong et al. / Journal of Nuclear Materials 465 (2015) 418e423
used. A platinum counter electrode and a saturated calomel reference electrode (SCE) were connected to a CHI 660E electrochemical station (Chenhua instrument Co. Shanghai, China) controlled by a computer and software. To ensure good reproducibility, a minimum of three sets of measurements of each specimen were taken and an average value was considered in borate buffer solution (0.05 M H3BO3 þ 0.075 M Na2B4O7$10H2O) solution without and with chloride ion. The electrochemical impedance spectroscopy (EIS) measurements were carried out using a frequency range of 100 kHz to 10 mHz and with a 5 mV amplitude of the ac signal. The surface compositions of alloy 690 were measured by X-ray photoelectron spectroscopy (XPS) measurement. The XPS experiments were performed using PHI Quantera SXM (ULVAC-PHI, INC). Photoelectron emission was excited by monochromatic Al Ka radiation. The vacuum of the specimen chamber was 6.7 108 Pa. The C 1 s peak from adventitious carbon at 284.8 eV was used as a reference to correct the charging shifts. XPSPeak4.1 software was used to fit the XPS experiment data. 3. Results and discussion XRD result of the solution annealed alloy 690 is shown in Fig. 1a. From the XRD spectra, the austenitic phase is found in solution annealed specimen with random grain orientation. The average grain size of the solid solution specimen is 28 mm in Fig. 1b and very few twins are produced due to annealing process. The anodic polarization curve of the solid solution alloy 690 in borate buffer solution is shown in Fig. 2. In the anodic polarization curve, current density increases with applied potential and two obvious peak currents are observed at 0.6 VSCE and 0.8 VSCE, respectively. The current increases dramatically when the potential is higher than the transpassive potential. The previous study showed that the compacting and thickness of the passive film were important factors in determining the corrosion current density [12]. Fig. 3a and b shows the Nyquist plots and Bode plots of alloy 690 after passivation at the different potentials for 1 h. The diameter of semicircle in the Nyquist plot decreases firstly with the increasing of passivated potential in borate buffer solution. The smallest diameter of semicircle in the Nyquist plot is found at 0.6 VSCE, then the diameter of semicircle in the Nyquist plot increases due to higher passivated potential. Bode plots are shown in Fig. 3b. One time constant is observed. The equivalent circuit in Fig. 4 is proposed for fitting EIS data to quantify the electrochemical parameters [13]. In this equivalent circuit, Rs represents solution resistance; Q1 is double charge layer capacitance; R1 is the charge-transfer resistance. The electrochemical impedance parameters obtained from the fitting of
419
Fig. 2. The anodic polarization curves of the solution annealed alloy 690 in borate buffer solution.
electrochemical impedance spectroscopy (EIS) diagrams are shown in Table 1. The solution resistance changes from 32.49 to 34.17 U cm2 for the alloy 690, indicating no obvious change of the solution during the form of the passive film. In addition, the charge transfer resistance of alloy 690 decreases from 1.67 106 U cm2 to 3.49 104 U cm2 with passivation potential increasing from 0 VSCE to 0.6 VSCE, then the charge transfer resistance increases up to 1.34 105 U cm2 again at 0.8 VSCE. The charge transfer resistance of the alloy 690 is smallest at 0.6 VSCE, which indicates the changed composition and thickness of the passive film. So, in the present study, the MotteSchottky plots are used to explain the semiconductor properties of films on alloy 690. The linear region of the plots is attributed to the variation of the width of the space charge layer of the passive film on the specimen with the applied potential, according to 2 C 2 ¼ CH2 þ CSC ¼
2 kT E Efb εS ε0 qND e
(1)
2 C 2 ¼ CH2 þ CSC ¼
2 kT E Efb εS ε0 qNA e
(2)
where ε0 is the vacuum permittivity (8.854 1012Fm1), εs is the dielectric constant of the specimen, e is the electron charge (1.6 1019C), k is the Boltzmann constant (1.38 1023JK1), ND and NA are the donor or acceptor density, respectively, T is the absolute temperature and Efb is the flatband potential [14]. Fig. 5a shows the MotteSchottky plots for alloy 690 after passivation at the different potentials for 1 h. The MotteSchottky plots are measured by sweeping in negative direction. It can be seen that the passive film on the surface of alloy 690 possesses the same
Fig. 1. (a) The X-ray diffraction patterns and (b) microstructures of the solution annealed alloy 690.
420
L. Jinlong et al. / Journal of Nuclear Materials 465 (2015) 418e423
Fig. 3. (a) Nyquist plots and (b) Bode plots of passive films formed at different potentials in borate buffer solution for alloy 690.
Fig. 4. The electrochemical equivalent circuit for EIS fitting.
Table 1 Electrochemical impedance parameters obtained from the fitting of EIS results for the alloy 690 passivated in borate buffer solution. Potential (VSCE)
Rs (U cm2)
R1 (U cm2)
Q1 YO (U
0 0.2 0.4 0.6 0.8
33.13 32.49 32.74 32.85 34.17
2.28 2.83 3.69 4.48 3.17
1
cm
2
105 105 105 105 105
n
S )
n 0.93 0.90 0.88 0.86 0.89
1.67 1.17 1.03 3.49 1.34
106 106 105 104 105
semiconductor properties and turning potentials around 0.7 VSCE, 0.36 VSCE, 0.12 VSCE and 0.13 VSCE. The sweeping velocity is very high and its effect on growth of the passive film is little [15]. From 1.2 VSCE and 0.7 VSCE, the passive film on the alloy shows p-type semiconductor due to the negative slope. At the potential range between 0.7 VSCE and 0.36 VSCE, the plots also show a linear tendency. The passive film performs n-type semiconductor due to the positive slope. Such variation of p-type to ntype repeatedly appears at the narrow potentials from 0.36 VSCE to 0.13 VSCE and from 0.13 VSCE to 0.13 VSCE. The phenomenon is probably caused by the composition and structure of passive film. When the number of electrons in conduction band of passive film is
more than that of holes in valence band, the passive film is considered as n-type semiconductor such as Fe2O3, MoO3, Fe(OH)3 [16]. Conversely, it belongs to p-type semiconductor such as Cr2O3, MoO2, FeCr2O4, NiO [17]. So, the MotteSchottky plots can be explained by the following discussion. During the anodic polarization process between 0 VSCE and 0.13 VSCE, n-type passive films Fe2O3 and Fe(OH)3 [18] are probably formed on the specimen. At the potential range from 0.13 VSCE to 0.5 VSCE, p-type passive films Fe3O4 and NiO [19] are formed on the surface of the alloy 690. With the potential increasingly moving to positive direction, the passive film containing chromium oxide could appear. The donor and acceptor concentrations in the passive film formed on alloy 690 in borate buffer solution are presented in Fig. 5b. It can be seen that the most donor and acceptor concentrations at 0.6 VSCE deteriorate corrosion resistance of the passive film. XPS results is used to support results above. The different compositions of the passive film could be used to explain different corrosion resistance in the passive region. XPS measurements are used to directly characterize the composition of passive film formed on specimens after being passivated at 0.2 VSCE and 0.6 VSCE for 1 h. The Cr 2p3/2 are presented in Fig. 6a and b, respectively. At the lower potential, the following dissolution reactions of Cr may occur: Cr / Cr3þ þ 3e
(3)
2Cr3þ þ 6OH / Cr2O3 þ 3H2O
(4)
Cr3þ þ 3H2O / Cr(OH)3 (aq or s) þ 3Hþ
(5)
The anodic current density increases at 0.6 VSCE, and following reaction may occur:
Fig. 5. (a) The MotteSchottky plots for passive films of the alloy 690 obtained at different potentials in borate buffer solution, (b) The relationships between the acceptor concentration and donor concentration and passivated potential.
L. Jinlong et al. / Journal of Nuclear Materials 465 (2015) 418e423
421
Fig. 6. The detailed XPS spectra of (a, b)Cr 2p3/2 and (c, d) Ni 2p3/2 of the passive films formed on alloy 690 at (a, c) 0.2 VSCE and (b, d) 0.6 VSCE, respectively.
Cr3þ þ 6OH / CrO3 þ 3H2O þ 3e
(6)
Cr þ 6OH / CrO3 þ 3H2O þ 6e
(7)
According to reactions (6) and (7), Cr atoms are oxidized to higher valence Cr6þ in form of CrO3. Valence change of the chromium was rarely reported in borate buffer solution for nickel based alloy 690. However, Feng et al. [20] observed the valence variation of Cr in the passive region by the potentiodynamic polarization plot for 316L stainless steel in borate buffer solution. When the potential is slightly higher than 0.6 VSCE, the amounts of dissolved Cr increase. This suggests that the passive film immediately starts to dissolve at slightly higher 0.6 VSCE, the following reactions occur:
CrðOHÞ3 þ 5OH /CrO2 4 þ4H2 O þ 3e
(8)
This illustrates that the passive film formed at 0.6 VSCE is weaker due to the destruction of protective Cr(OH)3 layers. This explains why the corrosion resistance of the passive film formed at 0.6 VSCE is the lowest for the alloy 690, as shown in Fig. 3a. The analysis Ni 2p3/2 XPS signals in the passive film formed on alloy 690 at 0.2 VSCE and (c, d) 0.6 VSCE, are shown in Fig. 6c and d, respectively. Comparing with specimen passivated at lower potential, specimen passivated at higher potential contains more NiO. However the corrosion resistance of the alloy 690 is the smallest at 0.6 VSCE. This could be attributed to two reasons. Firstly, amounts of dissolved Cr increase at higher passivated potential, which deteriorates its corrosion resistance. Secondly, although more NiO is formed at higher passivated potential, the content of NiO in passive film could be low. The previous studies showed that the content of nickel oxide in passive film formed on austenite stainless steel was very low. For example, Abreu et al. [21] reported that a very small amount of nickel oxides were detected in passive films formed on the surface of 2205 duplex stainless steel and 304L stainless steel in alkaline solution. Moreover, Kocijan et al. [22] observed a small quantity of nickel oxides in passive film formed on the surface of
316 stainless steel in borate buffer solution. This is mainly due to the fact that diffusion rate of iron in passive film is faster than that of nickel. Fig. 7 show the polarization curves of the alloy 690 at room temperature in borate buffer solution with 0.2 mol NaCl. From the polarization curves, the passive range is determined. Comparing Fig. 7 with Fig. 2, it can be seen that the breakdown potential slightly reduces due to the addition of chloride ion. This indicating that the corrosion resistance of alloy 690 significantly reduced by chloride ion. The lower breakdown potential of the alloy 690 could be related to semiconductor characteristics of the passive film formed on its surface. To determine the stability of the passive film of the alloy 690, the EIS also is carried out in borate buffer solution with 0.2 mol NaCl. For this purpose, all the working electrodes were passivated at different potentials for 1 h to form the passive films, just at which the impedance spectra were measured. In Fig. 8a, the real impedance is plotted vs. the imaginary impedance at each frequency for the alloy 690 at different
Fig. 7. The anodic polarization curves of the solution annealed alloy 690 in borate buffer solution with 0.2 mol NaCl.
422
L. Jinlong et al. / Journal of Nuclear Materials 465 (2015) 418e423
Fig. 8. (a) Nyquist plots and (b) Bode plots of passive films formed at different potentials in borate buffer solution with 0.2 mol NaCl for alloy 690.
Table 2 Electrochemical impedance parameters obtained from the fitting of EIS results for the alloy 690 passivated in borate buffer solution with 0.2 mol NaCl. Potential (VSCE)
Rs (U cm2)
R1 (U cm2)
Q1 YO ( U
0.3 0.2 0.1 0 0.1
23.36 22.54 21.47 23.48 24.65
1.78 1.53 1.39 1.28 1.17
potentials in borate buffer solution with 0.2 mol NaCl. It can be noticed that the Nyquist plots are depressed semicircles. The Nyquist plots of the alloy 690 in different passivated potentials display similar features. The partial semi-circle above the real axis is associated with a capacitive effect. The diameter of semicircle in the Nyquist plot increases with the increasing of passivated potential in borate buffer solution with chloride ion. Fig. 8b shows the Bode plots of alloy 690 after passivation at the different potentials for 1 h in borate buffer solution with 0.2 mol NaCl. One time constant is observed. In addition, in middle frequency region, broadening plateau reveals protective passive films formed on the surface of the alloy. The electrochemical impedance parameters obtained from the fitting of EIS diagrams are shown in Table 2. It can be seen that the charge transfer resistance of the alloy 690 increases from 1.02 105 U cm2 to 7.89 105 U cm2 with passivation potential increasing from 0.3 VSCE to 0.1 VSCE. The charge transfer resistance values of the alloy 690 in borate buffer solution are higher than those in borate buffer solution with chloride ion at the same passivated potential. Fig. 9a shows the MotteSchottky plots for the passive films formed on alloy 690 at different potentials in borate buffer solution
1
cm
2
105 105 105 105 105
n
S )
n 0.86 0.88 0.89 0.90 0.91
1.02 1.23 2.84 3.46 7.89
105 105 105 105 105
with 0.2 mol NaCl. The results show that there are changes in the slopes of the plots at a potential around 0.6 VSCE. This indicates the formation of a mixture of iron and chromium passive film on the surface of the alloy 690 [23]. The passive films with duplex structures in the borate buffer solution with chloride ion are composed of a chromium-rich inner layer and an iron-rich outer layer with p-type semiconductor characteristics and n-type semiconductor characteristics, respectively [24]. The donor and acceptor concentrations in the passive film formed in the borate buffer solution with chloride ion for alloy 690 is calculated from the slopes of the MotteSchottky plots in Fig. 9a, and the results are presented in Fig. 9b. It can be seen that the concentrations of the donor and acceptor are of the order of magnitude of 1021 cm3. Moreover, the acceptor density is higher than donor density in the solution. Comparing Fig. 5b with Fig. 9b, it is found that the chloride ions increase significantly donor and acceptor concentrations in the passive film. In addition, the donor and acceptor concentrations decrease with the increasing of the passivated potential, which implies improved corrosion resistance of the passive film. The results of MotteSchottky analysis further support the conclusion obtained by EIS.
Fig. 9. (a) The MotteSchottky plots for passive films of the alloy 690 obtained at different potentials in borate buffer solution with 0.2 mol NaCl (b) The relationships between the acceptor concentration and donor concentration and passivated potential.
L. Jinlong et al. / Journal of Nuclear Materials 465 (2015) 418e423
According to the point defect model (PDM) [25,26], the passive film contains a number of point defects. The interstitial cations and oxygen vacancies are donors, while the cation vacancies are acceptors. The former injected at the metal/passive film interface, while the latter injected at the film/solution interface. The movement of point defects in passive films results in its formation and dissolution (or breakdown). Moreover, according to PDM, chloride ions in borate buffer solution could be incorporated in a passive film by occupying oxygen vacancies. The absorbed chloride ions would fill the anion vacancies, and the system responded to the loss of oxygen vacancies by generating cation vacancy: oxygen vacancy pairs via a Schottky-pair type of Reaction (9). 0
Null ¼ VxM þ 2x Vo
(9)
0
x and Vo represent cation vacancy and oxygen vacancy, where VM respectively; x is valence of metal cation. The newly generated oxygen vacancies could in turn react with additional chloride ions at the passive film/solution interface to generate more oxygen vacancies and cation vacancies. Therefore, the generation of cation vacancy was autocatalytic. This could decrease corrosion resistance of the passive film.
4. Conclusions The properties of the passive films formed on alloy 690 in borate buffer solution without and with chloride ion were investigated in the present study. Following results are obtained. (1) The corrosion resistance of passive film for the alloy 690 was the lowest in the vicinity of 0.6 VSCE in borate buffer solution. (2) The accelerated dissolving of the chromium at 0.6 VSCE promoted to form Cr6þ, which deteriorated corrosion resistance of passive film for the alloy 690. (3) The chloride ion decreased corrosion resistance of the alloy 690 due to autocatalytic reaction.
423
Acknowledgements This work was financially supported by National Natural Science Foundation of China (Grant No. 51302149).
References [1] P.K. Samantaroy, G. Suresh, R. Paul, U.K. Mudali, B. Raj, J. Nucl. Mater. 418 (2011) 27e37. [2] H. Li, S. Xia, B.X. Zhou, W.Q. Liu, Mater. Charact. 66 (2012) 68e74. [3] X.H. Li, J.Q. Wang, E.H. Han, W. Ke, Corros. Sci. 67 (2013) 169e178. [4] R.S. Dutta, J. Nucl. Mater. 393 (2009) 343e349. [5] J.M. Zagal, H.F. Lopez, O. Flores, J.L. Albarran, L. Martinez, Corros. Sci. 50 (2008) 3371e3377. [6] J.B. Huang, X.Q. Wu, E.H. Han, Corros. Sci. 52 (2010) 3444e3452. [7] J.B. Huang, X.Q. Wu, E.H. Han, Corros. Sci. 51 (2009) 2976e2982. [8] F. Huang, J.Q. Wang, E.H. Han, W. Ke, J. Mater. Sci. Technol. 28 (6) (2012) 562e568. [9] S.Y. Wang, J.Q. Wang, Corros. Sci. 85 (2014) 183e192. [10] J. Robertson, Corros. Sci. 32 (1991) 443e465. [11] J. Hu, Q. Duan, G.X. Cheng, Z.X. Zhang, X.D. Chen, B. Wang, Ann. Nucl. Energy 58 (2013) 25e32. [12] M.S. Venkatraman, S. Ivan, S. Cole, B. Emmanuel, Electrochim. Acta 56 (2011) 8192e8203. [13] Y.M. Qi, H.Y. Luo, S.Q. Zheng, C.F. Chen, D.N. Wang, Corros. Sci. 69 (2013) 164e174. [14] W.P. Gomes, D. Vanmackelbergh, Electrochim. Acta 41 (1996) 967e973. [15] J.B. Huang, X.Q. Wu, E. Han, Corros. Sci. 51 (2009) 2976e2982. [16] C. Sunseri, S. Piazza, F.D. Quarto, J. Electrochem. Soc. 137 (1990) 2411e2417. [17] S.R. Morrison, Electrochemistry at Semiconductor and Oxidized Metal Electrodes, Plenum Press, New York, 1980. [18] N.E. Hakiki, S. Boudin, B. Rondot, et al., Corros. Sci. 37 (1995) 1809e1822. [19] S. Maximovitch, Electrochim. Acta 41 (1996) 2761e2771. [20] Z.C. Feng, X.Q. Cheng, C.F. Dong, L. Xu, X.G. Li, Corros. Sci. 52 (2010) 3646e3653. bal, R. Losada, X.R. No voa, G. Pena, M.C. Pe rez, Elec[21] C.M. Abreu, M.J. Cristo trochim. Acta 49 (2004) 3049e3056. Donik, M. Jenko, Corros. Sci. 49( (2007) 2083e2098. [22] A. Kocijan, C. [23] Agatino Di Paola, Electrochim. Acta 34 (1989) 203e210. [24] N.E. Hakiki, S. Boudin, B. Rondot, M. Da Cunha Belo, Corros. Sci. 37 (1995) 1809e1822. [25] C.Y. Chao, L.F. Lin, D.D. Macdonald, J. Electrochem. Soc. 128 (1981) 1187e1193. [26] S.M. Moon, S. Pyun II, Electrochim. Acta 43 (1998) 3117e3126.
Contents lists available at ScienceDirect
Journal of Nuclear Materials journal homepage: www.elsevier.com/locate/jnucmat
Investigation of passive films formed on the surface of alloy 690 in borate buffer solution Lv Jinlong*, Liang Tongxiang, Wang Chen, Guo Wenli Beijing Key Laboratory of Fine Ceramics, Institute of Nuclear and New Energy Technology, Tsinghua University, Zhongguancun Street, Haidian District, Beijing 100084, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 20 March 2015 Received in revised form 10 June 2015 Accepted 15 June 2015 Available online 20 June 2015
The passive film formed on the surface of the alloy 690 in borate buffer solution was studied by potentiodynamic curves and electrochemical impedance spectroscopy. With the increasing of the passivation potential, the corrosion resistance of the alloy 690 reduced. Moreover, the corrosion resistance of the passive film was the lowest in the vicinity of 0.6 VSCE. These results were supported by XPS and MotteSchottky analyses. The corrosion resistance of the alloy 690 increased with the increasing of passivated potential in borate buffer solution with chloride ion. The chloride ion decreased corrosion resistance of the alloy 690 according to point defect model. © 2015 Elsevier B.V. All rights reserved.
Keywords: Nickel EIS XPS XRD Passive films
1. Introduction Nickel based alloy 690 was candidate materials for the storage of high level waste generated from reprocessing of spent nuclear fuel [1]. Moreover, the alloy 690 also was currently as an alternative to traditional austenitic stainless steel due to its high strength, lower nickel concentration and excellent corrosion resistance [2]. For example, the alloy 690 showed a better corrosion resistance than alloy 800 in simulated primary water [3], therefore, the alloy 690 was suitable to be used in aggressive environment. Due to high stress corrosion cracking and pit corrosion resistance in high temperature pressurized water reactor (PWR), the alloy 690 as the alternative material for the alloy 600, was widely used as stream generator tubing materials in PWR plants [4,5]. Electrochemical properties and growth mechanism of passive films on alloy 690 in high-temperature alkaline environments were investigated widely [6e8]. Moreover, the study showed that alloy 690 dissolved preferentially along the direction which had lower surface free energy in a solution of 0.1 M H2SO4 and 0.1 M NaCl at room temperature [9]. The passive film was formed rapidly and its growth decreased its
* Corresponding author. E-mail addresses: [email protected], (L. Jinlong). http://dx.doi.org/10.1016/j.jnucmat.2015.06.033 0022-3115/© 2015 Elsevier B.V. All rights reserved.
[email protected]
the corrosion rate [10]. The passive films played an important role in preventing corrosion. Therefore, it was very important to deeply understand the form processes of passive films for estimating the structural integrity [11]. The passivation characteristics and corrosion resistance of the passive films formed on the alloy 690 in borate buffer solution without and with chloride ion were investigated in the present study. 2. Experimental procedures Specimens from alloy 690 plate, with chemical composition (wt.%): 0.03C, 0.28 Si, 0.25 Mn, 29.0 Cr, 0.02 Cu, 9.5 Fe, 0.002 S and balance Ni, were cut to cuboid with a dimension of 10 mm 10 mm 2 mm for test. The specimens were solution annealed at 1100 C for 1 h in vacuum, then the sheets were subsequently quenched into water. The copper wires were welded to each specimen for electrical connection through the spot welding. All edges and surfaces of the test specimens were coated with epoxy resin prior to electrochemical test, leaving an active surface area of 10 mm 10 mm exposed to the solution. Before the electrochemical test, all the specimens were abraded with 1000, 2000 and 3000 grit silicon carbide paper and polished with 1.5 mm alumina slurry. The polished specimens were ultrasonically cleaned finally in acetone and ethanol. A conventional three-electrode electrochemical cell was
L. Jinlong et al. / Journal of Nuclear Materials 465 (2015) 418e423
used. A platinum counter electrode and a saturated calomel reference electrode (SCE) were connected to a CHI 660E electrochemical station (Chenhua instrument Co. Shanghai, China) controlled by a computer and software. To ensure good reproducibility, a minimum of three sets of measurements of each specimen were taken and an average value was considered in borate buffer solution (0.05 M H3BO3 þ 0.075 M Na2B4O7$10H2O) solution without and with chloride ion. The electrochemical impedance spectroscopy (EIS) measurements were carried out using a frequency range of 100 kHz to 10 mHz and with a 5 mV amplitude of the ac signal. The surface compositions of alloy 690 were measured by X-ray photoelectron spectroscopy (XPS) measurement. The XPS experiments were performed using PHI Quantera SXM (ULVAC-PHI, INC). Photoelectron emission was excited by monochromatic Al Ka radiation. The vacuum of the specimen chamber was 6.7 108 Pa. The C 1 s peak from adventitious carbon at 284.8 eV was used as a reference to correct the charging shifts. XPSPeak4.1 software was used to fit the XPS experiment data. 3. Results and discussion XRD result of the solution annealed alloy 690 is shown in Fig. 1a. From the XRD spectra, the austenitic phase is found in solution annealed specimen with random grain orientation. The average grain size of the solid solution specimen is 28 mm in Fig. 1b and very few twins are produced due to annealing process. The anodic polarization curve of the solid solution alloy 690 in borate buffer solution is shown in Fig. 2. In the anodic polarization curve, current density increases with applied potential and two obvious peak currents are observed at 0.6 VSCE and 0.8 VSCE, respectively. The current increases dramatically when the potential is higher than the transpassive potential. The previous study showed that the compacting and thickness of the passive film were important factors in determining the corrosion current density [12]. Fig. 3a and b shows the Nyquist plots and Bode plots of alloy 690 after passivation at the different potentials for 1 h. The diameter of semicircle in the Nyquist plot decreases firstly with the increasing of passivated potential in borate buffer solution. The smallest diameter of semicircle in the Nyquist plot is found at 0.6 VSCE, then the diameter of semicircle in the Nyquist plot increases due to higher passivated potential. Bode plots are shown in Fig. 3b. One time constant is observed. The equivalent circuit in Fig. 4 is proposed for fitting EIS data to quantify the electrochemical parameters [13]. In this equivalent circuit, Rs represents solution resistance; Q1 is double charge layer capacitance; R1 is the charge-transfer resistance. The electrochemical impedance parameters obtained from the fitting of
419
Fig. 2. The anodic polarization curves of the solution annealed alloy 690 in borate buffer solution.
electrochemical impedance spectroscopy (EIS) diagrams are shown in Table 1. The solution resistance changes from 32.49 to 34.17 U cm2 for the alloy 690, indicating no obvious change of the solution during the form of the passive film. In addition, the charge transfer resistance of alloy 690 decreases from 1.67 106 U cm2 to 3.49 104 U cm2 with passivation potential increasing from 0 VSCE to 0.6 VSCE, then the charge transfer resistance increases up to 1.34 105 U cm2 again at 0.8 VSCE. The charge transfer resistance of the alloy 690 is smallest at 0.6 VSCE, which indicates the changed composition and thickness of the passive film. So, in the present study, the MotteSchottky plots are used to explain the semiconductor properties of films on alloy 690. The linear region of the plots is attributed to the variation of the width of the space charge layer of the passive film on the specimen with the applied potential, according to 2 C 2 ¼ CH2 þ CSC ¼
2 kT E Efb εS ε0 qND e
(1)
2 C 2 ¼ CH2 þ CSC ¼
2 kT E Efb εS ε0 qNA e
(2)
where ε0 is the vacuum permittivity (8.854 1012Fm1), εs is the dielectric constant of the specimen, e is the electron charge (1.6 1019C), k is the Boltzmann constant (1.38 1023JK1), ND and NA are the donor or acceptor density, respectively, T is the absolute temperature and Efb is the flatband potential [14]. Fig. 5a shows the MotteSchottky plots for alloy 690 after passivation at the different potentials for 1 h. The MotteSchottky plots are measured by sweeping in negative direction. It can be seen that the passive film on the surface of alloy 690 possesses the same
Fig. 1. (a) The X-ray diffraction patterns and (b) microstructures of the solution annealed alloy 690.
420
L. Jinlong et al. / Journal of Nuclear Materials 465 (2015) 418e423
Fig. 3. (a) Nyquist plots and (b) Bode plots of passive films formed at different potentials in borate buffer solution for alloy 690.
Fig. 4. The electrochemical equivalent circuit for EIS fitting.
Table 1 Electrochemical impedance parameters obtained from the fitting of EIS results for the alloy 690 passivated in borate buffer solution. Potential (VSCE)
Rs (U cm2)
R1 (U cm2)
Q1 YO (U
0 0.2 0.4 0.6 0.8
33.13 32.49 32.74 32.85 34.17
2.28 2.83 3.69 4.48 3.17
1
cm
2
105 105 105 105 105
n
S )
n 0.93 0.90 0.88 0.86 0.89
1.67 1.17 1.03 3.49 1.34
106 106 105 104 105
semiconductor properties and turning potentials around 0.7 VSCE, 0.36 VSCE, 0.12 VSCE and 0.13 VSCE. The sweeping velocity is very high and its effect on growth of the passive film is little [15]. From 1.2 VSCE and 0.7 VSCE, the passive film on the alloy shows p-type semiconductor due to the negative slope. At the potential range between 0.7 VSCE and 0.36 VSCE, the plots also show a linear tendency. The passive film performs n-type semiconductor due to the positive slope. Such variation of p-type to ntype repeatedly appears at the narrow potentials from 0.36 VSCE to 0.13 VSCE and from 0.13 VSCE to 0.13 VSCE. The phenomenon is probably caused by the composition and structure of passive film. When the number of electrons in conduction band of passive film is
more than that of holes in valence band, the passive film is considered as n-type semiconductor such as Fe2O3, MoO3, Fe(OH)3 [16]. Conversely, it belongs to p-type semiconductor such as Cr2O3, MoO2, FeCr2O4, NiO [17]. So, the MotteSchottky plots can be explained by the following discussion. During the anodic polarization process between 0 VSCE and 0.13 VSCE, n-type passive films Fe2O3 and Fe(OH)3 [18] are probably formed on the specimen. At the potential range from 0.13 VSCE to 0.5 VSCE, p-type passive films Fe3O4 and NiO [19] are formed on the surface of the alloy 690. With the potential increasingly moving to positive direction, the passive film containing chromium oxide could appear. The donor and acceptor concentrations in the passive film formed on alloy 690 in borate buffer solution are presented in Fig. 5b. It can be seen that the most donor and acceptor concentrations at 0.6 VSCE deteriorate corrosion resistance of the passive film. XPS results is used to support results above. The different compositions of the passive film could be used to explain different corrosion resistance in the passive region. XPS measurements are used to directly characterize the composition of passive film formed on specimens after being passivated at 0.2 VSCE and 0.6 VSCE for 1 h. The Cr 2p3/2 are presented in Fig. 6a and b, respectively. At the lower potential, the following dissolution reactions of Cr may occur: Cr / Cr3þ þ 3e
(3)
2Cr3þ þ 6OH / Cr2O3 þ 3H2O
(4)
Cr3þ þ 3H2O / Cr(OH)3 (aq or s) þ 3Hþ
(5)
The anodic current density increases at 0.6 VSCE, and following reaction may occur:
Fig. 5. (a) The MotteSchottky plots for passive films of the alloy 690 obtained at different potentials in borate buffer solution, (b) The relationships between the acceptor concentration and donor concentration and passivated potential.
L. Jinlong et al. / Journal of Nuclear Materials 465 (2015) 418e423
421
Fig. 6. The detailed XPS spectra of (a, b)Cr 2p3/2 and (c, d) Ni 2p3/2 of the passive films formed on alloy 690 at (a, c) 0.2 VSCE and (b, d) 0.6 VSCE, respectively.
Cr3þ þ 6OH / CrO3 þ 3H2O þ 3e
(6)
Cr þ 6OH / CrO3 þ 3H2O þ 6e
(7)
According to reactions (6) and (7), Cr atoms are oxidized to higher valence Cr6þ in form of CrO3. Valence change of the chromium was rarely reported in borate buffer solution for nickel based alloy 690. However, Feng et al. [20] observed the valence variation of Cr in the passive region by the potentiodynamic polarization plot for 316L stainless steel in borate buffer solution. When the potential is slightly higher than 0.6 VSCE, the amounts of dissolved Cr increase. This suggests that the passive film immediately starts to dissolve at slightly higher 0.6 VSCE, the following reactions occur:
CrðOHÞ3 þ 5OH /CrO2 4 þ4H2 O þ 3e
(8)
This illustrates that the passive film formed at 0.6 VSCE is weaker due to the destruction of protective Cr(OH)3 layers. This explains why the corrosion resistance of the passive film formed at 0.6 VSCE is the lowest for the alloy 690, as shown in Fig. 3a. The analysis Ni 2p3/2 XPS signals in the passive film formed on alloy 690 at 0.2 VSCE and (c, d) 0.6 VSCE, are shown in Fig. 6c and d, respectively. Comparing with specimen passivated at lower potential, specimen passivated at higher potential contains more NiO. However the corrosion resistance of the alloy 690 is the smallest at 0.6 VSCE. This could be attributed to two reasons. Firstly, amounts of dissolved Cr increase at higher passivated potential, which deteriorates its corrosion resistance. Secondly, although more NiO is formed at higher passivated potential, the content of NiO in passive film could be low. The previous studies showed that the content of nickel oxide in passive film formed on austenite stainless steel was very low. For example, Abreu et al. [21] reported that a very small amount of nickel oxides were detected in passive films formed on the surface of 2205 duplex stainless steel and 304L stainless steel in alkaline solution. Moreover, Kocijan et al. [22] observed a small quantity of nickel oxides in passive film formed on the surface of
316 stainless steel in borate buffer solution. This is mainly due to the fact that diffusion rate of iron in passive film is faster than that of nickel. Fig. 7 show the polarization curves of the alloy 690 at room temperature in borate buffer solution with 0.2 mol NaCl. From the polarization curves, the passive range is determined. Comparing Fig. 7 with Fig. 2, it can be seen that the breakdown potential slightly reduces due to the addition of chloride ion. This indicating that the corrosion resistance of alloy 690 significantly reduced by chloride ion. The lower breakdown potential of the alloy 690 could be related to semiconductor characteristics of the passive film formed on its surface. To determine the stability of the passive film of the alloy 690, the EIS also is carried out in borate buffer solution with 0.2 mol NaCl. For this purpose, all the working electrodes were passivated at different potentials for 1 h to form the passive films, just at which the impedance spectra were measured. In Fig. 8a, the real impedance is plotted vs. the imaginary impedance at each frequency for the alloy 690 at different
Fig. 7. The anodic polarization curves of the solution annealed alloy 690 in borate buffer solution with 0.2 mol NaCl.
422
L. Jinlong et al. / Journal of Nuclear Materials 465 (2015) 418e423
Fig. 8. (a) Nyquist plots and (b) Bode plots of passive films formed at different potentials in borate buffer solution with 0.2 mol NaCl for alloy 690.
Table 2 Electrochemical impedance parameters obtained from the fitting of EIS results for the alloy 690 passivated in borate buffer solution with 0.2 mol NaCl. Potential (VSCE)
Rs (U cm2)
R1 (U cm2)
Q1 YO ( U
0.3 0.2 0.1 0 0.1
23.36 22.54 21.47 23.48 24.65
1.78 1.53 1.39 1.28 1.17
potentials in borate buffer solution with 0.2 mol NaCl. It can be noticed that the Nyquist plots are depressed semicircles. The Nyquist plots of the alloy 690 in different passivated potentials display similar features. The partial semi-circle above the real axis is associated with a capacitive effect. The diameter of semicircle in the Nyquist plot increases with the increasing of passivated potential in borate buffer solution with chloride ion. Fig. 8b shows the Bode plots of alloy 690 after passivation at the different potentials for 1 h in borate buffer solution with 0.2 mol NaCl. One time constant is observed. In addition, in middle frequency region, broadening plateau reveals protective passive films formed on the surface of the alloy. The electrochemical impedance parameters obtained from the fitting of EIS diagrams are shown in Table 2. It can be seen that the charge transfer resistance of the alloy 690 increases from 1.02 105 U cm2 to 7.89 105 U cm2 with passivation potential increasing from 0.3 VSCE to 0.1 VSCE. The charge transfer resistance values of the alloy 690 in borate buffer solution are higher than those in borate buffer solution with chloride ion at the same passivated potential. Fig. 9a shows the MotteSchottky plots for the passive films formed on alloy 690 at different potentials in borate buffer solution
1
cm
2
105 105 105 105 105
n
S )
n 0.86 0.88 0.89 0.90 0.91
1.02 1.23 2.84 3.46 7.89
105 105 105 105 105
with 0.2 mol NaCl. The results show that there are changes in the slopes of the plots at a potential around 0.6 VSCE. This indicates the formation of a mixture of iron and chromium passive film on the surface of the alloy 690 [23]. The passive films with duplex structures in the borate buffer solution with chloride ion are composed of a chromium-rich inner layer and an iron-rich outer layer with p-type semiconductor characteristics and n-type semiconductor characteristics, respectively [24]. The donor and acceptor concentrations in the passive film formed in the borate buffer solution with chloride ion for alloy 690 is calculated from the slopes of the MotteSchottky plots in Fig. 9a, and the results are presented in Fig. 9b. It can be seen that the concentrations of the donor and acceptor are of the order of magnitude of 1021 cm3. Moreover, the acceptor density is higher than donor density in the solution. Comparing Fig. 5b with Fig. 9b, it is found that the chloride ions increase significantly donor and acceptor concentrations in the passive film. In addition, the donor and acceptor concentrations decrease with the increasing of the passivated potential, which implies improved corrosion resistance of the passive film. The results of MotteSchottky analysis further support the conclusion obtained by EIS.
Fig. 9. (a) The MotteSchottky plots for passive films of the alloy 690 obtained at different potentials in borate buffer solution with 0.2 mol NaCl (b) The relationships between the acceptor concentration and donor concentration and passivated potential.
L. Jinlong et al. / Journal of Nuclear Materials 465 (2015) 418e423
According to the point defect model (PDM) [25,26], the passive film contains a number of point defects. The interstitial cations and oxygen vacancies are donors, while the cation vacancies are acceptors. The former injected at the metal/passive film interface, while the latter injected at the film/solution interface. The movement of point defects in passive films results in its formation and dissolution (or breakdown). Moreover, according to PDM, chloride ions in borate buffer solution could be incorporated in a passive film by occupying oxygen vacancies. The absorbed chloride ions would fill the anion vacancies, and the system responded to the loss of oxygen vacancies by generating cation vacancy: oxygen vacancy pairs via a Schottky-pair type of Reaction (9). 0
Null ¼ VxM þ 2x Vo
(9)
0
x and Vo represent cation vacancy and oxygen vacancy, where VM respectively; x is valence of metal cation. The newly generated oxygen vacancies could in turn react with additional chloride ions at the passive film/solution interface to generate more oxygen vacancies and cation vacancies. Therefore, the generation of cation vacancy was autocatalytic. This could decrease corrosion resistance of the passive film.
4. Conclusions The properties of the passive films formed on alloy 690 in borate buffer solution without and with chloride ion were investigated in the present study. Following results are obtained. (1) The corrosion resistance of passive film for the alloy 690 was the lowest in the vicinity of 0.6 VSCE in borate buffer solution. (2) The accelerated dissolving of the chromium at 0.6 VSCE promoted to form Cr6þ, which deteriorated corrosion resistance of passive film for the alloy 690. (3) The chloride ion decreased corrosion resistance of the alloy 690 due to autocatalytic reaction.
423
Acknowledgements This work was financially supported by National Natural Science Foundation of China (Grant No. 51302149).
References [1] P.K. Samantaroy, G. Suresh, R. Paul, U.K. Mudali, B. Raj, J. Nucl. Mater. 418 (2011) 27e37. [2] H. Li, S. Xia, B.X. Zhou, W.Q. Liu, Mater. Charact. 66 (2012) 68e74. [3] X.H. Li, J.Q. Wang, E.H. Han, W. Ke, Corros. Sci. 67 (2013) 169e178. [4] R.S. Dutta, J. Nucl. Mater. 393 (2009) 343e349. [5] J.M. Zagal, H.F. Lopez, O. Flores, J.L. Albarran, L. Martinez, Corros. Sci. 50 (2008) 3371e3377. [6] J.B. Huang, X.Q. Wu, E.H. Han, Corros. Sci. 52 (2010) 3444e3452. [7] J.B. Huang, X.Q. Wu, E.H. Han, Corros. Sci. 51 (2009) 2976e2982. [8] F. Huang, J.Q. Wang, E.H. Han, W. Ke, J. Mater. Sci. Technol. 28 (6) (2012) 562e568. [9] S.Y. Wang, J.Q. Wang, Corros. Sci. 85 (2014) 183e192. [10] J. Robertson, Corros. Sci. 32 (1991) 443e465. [11] J. Hu, Q. Duan, G.X. Cheng, Z.X. Zhang, X.D. Chen, B. Wang, Ann. Nucl. Energy 58 (2013) 25e32. [12] M.S. Venkatraman, S. Ivan, S. Cole, B. Emmanuel, Electrochim. Acta 56 (2011) 8192e8203. [13] Y.M. Qi, H.Y. Luo, S.Q. Zheng, C.F. Chen, D.N. Wang, Corros. Sci. 69 (2013) 164e174. [14] W.P. Gomes, D. Vanmackelbergh, Electrochim. Acta 41 (1996) 967e973. [15] J.B. Huang, X.Q. Wu, E. Han, Corros. Sci. 51 (2009) 2976e2982. [16] C. Sunseri, S. Piazza, F.D. Quarto, J. Electrochem. Soc. 137 (1990) 2411e2417. [17] S.R. Morrison, Electrochemistry at Semiconductor and Oxidized Metal Electrodes, Plenum Press, New York, 1980. [18] N.E. Hakiki, S. Boudin, B. Rondot, et al., Corros. Sci. 37 (1995) 1809e1822. [19] S. Maximovitch, Electrochim. Acta 41 (1996) 2761e2771. [20] Z.C. Feng, X.Q. Cheng, C.F. Dong, L. Xu, X.G. Li, Corros. Sci. 52 (2010) 3646e3653. bal, R. Losada, X.R. No voa, G. Pena, M.C. Pe rez, Elec[21] C.M. Abreu, M.J. Cristo trochim. Acta 49 (2004) 3049e3056. Donik, M. Jenko, Corros. Sci. 49( (2007) 2083e2098. [22] A. Kocijan, C. [23] Agatino Di Paola, Electrochim. Acta 34 (1989) 203e210. [24] N.E. Hakiki, S. Boudin, B. Rondot, M. Da Cunha Belo, Corros. Sci. 37 (1995) 1809e1822. [25] C.Y. Chao, L.F. Lin, D.D. Macdonald, J. Electrochem. Soc. 128 (1981) 1187e1193. [26] S.M. Moon, S. Pyun II, Electrochim. Acta 43 (1998) 3117e3126.