- Email: [email protected]
Effects of cold work and sensitization treatment on the corrosion resistance of high nitrogen stainless steel in chloride solutions
Electrochimica Acta 54 (2009) 1618–1629
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Effects of cold work and sensitization treatment on the corrosion resistance of high nitrogen stainless steel in chloride solutions Yao Fu a , Xinqiang Wu a,∗ , En-Hou Han a , Wei Ke a , Ke Yang a , Zhouhua Jiang b a State key Laboratory for Corrosion and Protection, Institute of Metal Research, South Campus, Chinese Academy of Sciences, 62 Wencui Road, Shenyang 110016, PR China b School of Materials and Metallurgy, Northeastern University, Shenyang 110004, PR China
a r t i c l e
i n f o
Article history: Received 31 July 2008 Accepted 24 September 2008 Available online 8 October 2008 Keywords: High nitrogen stainless steel Cold work Sensitization treatment Corrosion resistance X-ray photoelectron spectroscopy
a b s t r a c t The effects of cold work and sensitization treatment on the microstructure and corrosion resistance of a nickel-free high nitrogen stainless steel (HNSS) in 0.5 M H2 SO4 + 0.5 M NaCl, 3.5% NaCl and 0.5 M NaOH + 0.5 M NaCl solutions have been investigated by microscopic observations, electrochemical tests and surface chemical analysis. Cold work introduced a high defect density into the matrix, resulting in a less protective passive film as well as reduced corrosion resistance for heavily cold worked HNSS in a 3.5% NaCl solution. No obvious degradation in corrosion resistance took place in a 0.5 M H2 SO4 + 0.5 M NaCl solution, possibly due to the stability of the passive film in this solution. Sensitized HNSSs showed reduced corrosion resistance with increasing cold work level in both 3.5% NaCl and 0.5 M H2 SO4 + 0.5 M NaCl solutions due to a reduction in the anti-corrosion elements in the matrix during the cold workaccelerated precipitation process. The cold work and sensitization treatment had no influence on the corrosion resistance of the HNSS in the 0.5 M NaOH + 0.5 M NaCl solution even though the property of the passive film changed. The effects of cold work and sensitization treatment on the characteristics of passive films formed in the three solutions are discussed. © 2008 Elsevier Ltd. All rights reserved.
1. Introduction The beneficial effects of nitrogen on the properties of highalloyed steels have led to the widespread development of high nitrogen stainless steels (HNSSs) due to recent advances in processing technologies [1–3]. As an alloying element, nitrogen has several beneficial effects on the properties of steels, in particular, those related to high yield strength and toughness (good fracture resistance) [1]. Nitrogen is also a strong austenite-stabilizing element and can thus substitute for nickel, which is expensive and causes an allergic reaction in human skin. Moreover, nitrogen alloying, especially in combination with molybdenum, improves resistance to localized corrosion in general, and resistance to general corrosion in some environments [4,5]. One problem with HNSS is the formation of precipitates such as Cr-nitride, , and upon thermal exposure in the temperature range of 500–1050 ◦ C, leading to a significant reduction in corrosion resistance [6,7]. It has been reported that the deleterious effects of Cr2 N on pitting corrosion are associated with
∗ Corresponding author. Tel.: +86 24 23841883; fax: +86 24 23894149. E-mail address: [email protected] (X. Wu). 0013-4686/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2008.09.053
the formation of a Cr-depletion zone adjacent to Cr2 N precipitates [8]. It was also found that the lamellar Cr2 N was the most susceptible site for pitting corrosion when compared with any other heterogeneity or Cr2 N precipitate along the grain boundaries [9]. Stainless steels are subject to different levels of cold work during the final manufacturing stages of components for numerous applications in industry. Cold work might affect the corrosion resistance of stainless steels because deformed substructures like planar dislocation arrays [1,7] and deformation twinning [10] could be introduced. Peguet et al. [11] have reported the different roles of cold work on the pitting corrosion resistance at different pitting stages, including pit initiation, propagation and repassivation. Barbucci et al. [12] reported that the passive currents of 304 SS in sulfate + chloride solutions significantly increased with the degree of cold work. In addition, the pitting corrosion resistance was observed to decrease with increasing cold work in a 3.5% NaCl solution [13]. Efforts have also been made to clarify the relationship between cold work and the sensitization process. The cold work effect in Type 316 SS has been determined to be due to the higher diffusivity of chromium and the lower free energy barrier to carbide nucleation at grain boundaries in the deformed microstructure [14–17]. The acceleration of sensitization with cold work could also be related to the effects of
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
point defects and microstructural sinks on diffusion, as reported by Mansur [18]. The metallurgical changes caused by cold work and sensitization treatment have not been related clearly to the influence on the corrosion resistance in chloride solutions of different pH. Moreover, previous studies have not included the composition and structural evolution of passive films, which have direct influence on the corrosion resistance. In the present work, we examine the microstructure and corrosion resistance of a HNSS after cold work and/or sensitization treatment. The structural and chemical composition of the passive film induced by cold work and sensitization treatment were analyzed by electrochemical impedance spectroscopy (EIS), and X-ray photoelectron spectroscopy (XPS). The morphology of the pitting attack after electrochemical polarization was observed. The relationship between the microstructure evolution, changes in passive film characteristics and variations in the corrosion resistance of the HNSS are discussed. 2. Experimental 2.1. Microstructure observation The chemical composition of the HNSS investigated is listed in Table 1. The steel plates were cold rolled to a 8, 30, 49 or 60% reduction in thickness. Some of the specimens were then isothermally sensitized at 650 ◦ C for 2 h. 10 mm × 10 mm specimens were cut from the plates, with test surfaces parallel to the rolling direction. The specimens were polished to a diamond finish (1.5 m) and electrolytically etched in a 10% ethane diacid reagent at 12 V for 90 s to observe the changes in microstructure introduced by cold work. Specimens for transmission electron microscopy (TEM) analysis were mechanically polished and punched into 3 mm discs. Final
1619
Table 1 Chemical composition of the high nitrogen stainless steel in this work (in wt.%). C Cr Mn Mo N Si S P Fe
0.04 18.4 15.8 2.19 0.66 0.24 0.005 0.017 Bal
thinning was performed using a twin-jet electrolytic polisher operated at 20 V in an electrolyte of perchloric acid and alcohol at −30 ◦ C. The substructure of the cold worked and sensitized specimen was analyzed by an FEI Tecnai G2 F20 S-Twin microscope. 2.2. Electrochemical tests 2.2.1. Electrodes and electrolyte To prepare working electrodes for the electrochemical tests, the specimens were sealed with an epoxy resin to expose an area of 1 cm2 . The exposed faces were polished with 800 grit silicon carbide paper and rinsed with de-ionized water just before immersion. The specimens used to observe pitting attacks after polarization tests were polished to a diamond finish (1.5 m). After immersion for 5 min, the working electrode was passivated at the film formation potential for 20 min for EIS measurements and 1 h for XPS measurements, respectively. A three electrode cell featuring a Pt counter electrode and a saturated calomel reference electrode (SCE) was employed. All the potentials in the paper are given vs. this reference electrode. The solutions used in this investigation were: solution I, 0.5 M H2 SO4 + 0.5 M NaCl (pH ≈ 0.4); solution II, 3.5% (0.6 M) NaCl (pH ≈ 5.8); and solution III, 0.5 M NaOH + 0.5 M
Fig. 1. SEM images of (a) deformation bands in an 8% cold worked non-sensitized HNSS, (b) deformation bands in a 30% cold worked non-sensitized HNSS, (c) weakened grain, twin boundaries in an 8% and also (d) deformation bands in a 30% cold worked sensitized HNSS.
1620
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
Fig. 2. TEM images of the (a) the planar dislocation arrays, (b) dislocation pile-ups in the 8% cold worked non-sensitized specimen, (c) deformation twinning and high density of dislocation (the diffraction pattern on the bottom-right was taken from the junction of the twinning and matrix), (d) high density of defects and misorientation within the grain of the 49% cold worked non-sensitized specimen.
NaCl (pH ≈ 13.1). The electrolytes were prepared from concentrated sulfuric acid, analytical grade sodium chloride, sodium hydroxide and de-ionized water. The experiments were carried out at room temperature under naturally aerated conditions.
lution spectra of the following regions were recorded: Fe 2p, Cr 2p, Mn 2p, Mo 3d, N 1s, O 1s, and C 1s. The hydrocarbon C 1s peak was assumed to be at 284.6 eV and was used as an internal standard to determine the binding energy of other photoelectron peaks. A
2.2.2. Apparatus and procedures Polarization curves were conducted at a sweep rate of 0.1667 mV s−1 , using an EG&G 273A Potentiostat controlled by PowerSuite software. Impedance measurements were performed with a lock-in amplifier (EG&G 5210) coupled to a potentiostat (EG&G 273A), using a 10 mV (rms) perturbation from 100 kHz to 10 mHz. Fitting was performed with ZSimpwin software. The chemical composition profile for the passive film was obtained by XPS analysis using the Al K␣ X-ray source (1487 eV, 150 W) and a pass energy of 20 eV. Characterization of the passive films was performed at a takeoff angle of 45◦ . The depth profile of the chemical composition was obtained by sputtering with argon ions (base pressure: 1E−7 Torr, energy: 2 kV, current: 2.0 A cm−2 ) and a pass energy of 50 eV. A survey spectrum was first recorded to identify all the elements present at the surface. Then, high reso-
Fig. 3. Schematic illustration of the SAD pattern of the deformation twinning in Fig. 2c.
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
1621
Fig. 4. TEM images of (a) small particles of phase at the grain boundary and (b) within the grain of the 49% cold worked and sensitized specimen, (c) the SAD pattern with [0 1¯ 1] orientation and (d) the SAD pattern with [1¯ 0 4] orientation of the precipitates. (e) TEM-EDS analysis of the precipitates.
Shirley background subtraction was made to obtain the XPS signal intensity. 3. Results and discussion 3.1. Microstructure observation The cold worked specimens showed an increase in the deformation bands with increasing cold work level (Fig. 1a and b). TEM observation showed that the planar dislocation arrays were most prevalent at the beginning of the deformation (Fig. 2a). Dislocation pile-ups (Fig. 2b) were observed at the grain boundaries. At a higher deformation stage, the deformed microstructure was characterized by pronounced deformation twinning, and a high density of dislocation between twin lamellae (Fig. 2c). The alternating black
and white regions with high contrast in Fig. 2d suggests that the misorientation within the grain and the density of defects were rather high. The disordered multi-spots in the diffraction pattern indicated the existence of multiple fine subgrains with different orientations. The exact twinning plane could be determined from the combined analyses of the selected area diffraction (SAD) pattern and the transformation matrix [19]. A typical SAD pattern composed of matrix and deformation twinning is shown in Fig. 2c, where the matrix was tilted to its [1¯ 1 4] direction. The extra spots showed the characteristic arrangement of a SAD pattern of fcc crystals. According to the transformation matrix for deformation twinning in fcc materials calculated by Mahajan [20], it can be shown that ¯ planes transthe twinning on the (1 1 1), (1¯ 1 1), (1 1¯ 1) and (1 1 1) ¯ [1¯ 1 0], [7 7¯ 8] forms the [1¯ 1 4] matrix direction into the [1 1 5 4],
1622
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
Fig. 5. Potentiodynamic polarization curves of the cold worked non-sensitized HNSSs in (a) 0.5 M H2 SO4 + 0.5 M NaCl solution, (b) 3.5% NaCl solution, and (c) 0.5 M NaOH + 0.5 M NaCl solution.
and [5 1 1 4] twinning directions, respectively. Among these directions, only the [1¯ 1 0] direction is consistent with the observed diffraction pattern shown in Fig. 2c. Assuming that the [1¯ 1 0] orientation is twinning, the SAD pattern of Fig. 2c can be indexed as shown in Fig. 3. Based on the combined analyses of the SAD pattern with the transformation matrix, it can be concluded that the crystallographic twinning plane was the exact (1¯ 1 1) plane in this study. Fig. 1c and d are the microstructures of the sensitized HNSSs at different cold work levels. After metallographic etching, the 8% cold worked samples showed many corrosion pits at the grain and twin boundaries. The 30% cold worked samples, however, exhibited serious corrosion pits at deformation bands adjacent to the grain and twin boundaries. No precipitates were detected by X-ray diffraction; however, by TEM observation, the SAD pattern (Fig. 4c and d) and TEM-EDS (Fig. 4e) analysis indicated that small particles of phase had precipitated at the grain boundaries (Fig. 4a) and also within the grains (Fig. 4b), with a chemical composition similar to previous reports [21]. Here, extensive dislocation networks, mechanical twinning and the subgrain boundaries introduced by
Fig. 6. Potentiodynamic polarization curves of the cold worked and sensitized HNSSs in (a) 0.5 M H2 SO4 + 0.5 M NaCl solution, (b) 3.5% NaCl solution, and (c) 0.5 M NaOH + 0.5 M NaCl solution.
cold work provide more sites for rapid pipe diffusion and faster nucleation or more point defects and microstructural sinks on diffusion, as reported by Mansur [18]. Therefore, more phase particles were able to precipitate during sensitization treatment for the specimen with more cold work. 3.2. Potentiodynamic study Figs. 5 and 6 are the potentiodynamic polarization diagrams for the cold worked HNSSs before and after sensitization treatment in chloride solution with different pH levels. Fig. 7 shows the variation in critical pitting potential with cold work level. Each curve exhibited a range of passivity beyond the open-circuit potential. The specimens tested in the acid sulfate + chloride solution showed active dissolution before passivation. For the non-sensitized HNSSs, the critical pitting potential decreased and the passive regions became narrower in 3.5% NaCl solution with increasing cold work level (Fig. 5b). However, in the acid sulfate + chloride solution, no obvious change could be observed from the polarization
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
Fig. 7. The variation of the critical pitting potential as a function of the cold work level of (a) the cold worked non-sensitized HNSS and (b) the cold worked and sensitized HNSS.
1623
Fig. 8. SEM images of the pitted surface for the (a) 49% cold worked non-sensitized specimen and (b) 49% cold worked and sensitized specimen after potentiodynamic polarization in a 3.5% NaCl solution.
Table 2 Equivalent circuit parameters for the HNSSs, tested at the passive potentials. Rsol ( cm2 )
Q (−1 sn cm−2 )
n
R1 ( cm2 )
C (F cm−2 )
R2 ( cm2 )
8% (0.3) 30% (0.3) 60% (0.3)
1.85 1.60 1.70
4.47E−05 5.04E−05 4.71E−05
0.93 0.94 0.94
48,260 66,200 58,370
1.13E−04 2.96E−04 7.08E−05
142,400 102,600 101,000
3.5% NaCl
8% (0.3) 30% (0.25) 60% (0.17)
7.62 7.33 7.23
1.14E−05 1.20E−05 2.76E−05
0.88 0.83 0.95
3.56 4.01 115,600
6.53E−06 7.79E−06 8.38E−05
648,000 252,700 179,900
0.5 M NaOH + 0.5 M NaCl
8% (0.3) 30% (0.3) 60% (0.3)
2.85 2.14 2.33
4.18E−05 6.67E−05 5.86E−05
0.79 0.73 0.78
1.36 1.32 1.26
2.22E−05 1.86E−05 1.46E−05
227,400 124,000 27,600
8% (0.05) 30% (−0.16) 60% (−0.18)
1.55 2.10 2.36
7.31E−05 6.65E−05 6.60E−05
0.79 0.86 0.87
0.77 11.25 13,490
1.59E−05 1.96E−06 5.56E−04
52,050 32,780 14,290
8% (0.02) 30% (−0.12) 60% (−0.15)
7.57 6.17 6.26
1.37E−05 8.21E−05 9.51E−05
0.76 0.85 0.78
4.83 10.05 3.02
8.43E−06 1.26E−05 1.65E−05
165,200 136,300 128,800
2.48 3.09 2.86
8.72E−05 4.90E−05 7.11E−05
0.8 0.9 0.81
1.28 20,560 1.88
1.52E−05 6.46E−05 1.44E−05
57,790 29,650 109,900
Solution (a) Non-sensitized HNSS 0.5 M H2 SO4 + 0.5 M NaCl
(b) Sensitized HNSS 0.5 M H2 SO4 + 0.5 M NaCl
3.5% NaCl
0.5 M NaOH + 0.5 M NaCl
Cold work level (film generated potential, V)
8% (0.3) 30% (0.3) 60% (0.3)
1624
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
Fig. 10. Examples of (a) Nyquist and (b) Bode plots for the passive films of the HNSS formed at the passive potential, and (c) the equivalent circuit used for quantitative evaluation of EIS spectra.
Fig. 9. SEM images of the pitted surface for the (a) 8% and (b) 49% cold worked nonsensitized specimen and (c) the 49% cold worked and sensitized specimen after potentiodynamic polarization in a 0.5 M H2 SO4 + 0.5 M NaCl solution.
treatment changed the electrochemical polarization behavior of the HNSSs in alkaline chloride solution (Figs. 5c and 6c). SEM observations of the specimens after polarization to 1E−3 A cm−2 were performed to study the pit size and distribution. The non-sensitized specimens showed no significant changes in the size and distribution of corrosion pits with increasing cold work level in 3.5% NaCl. Several large pits with a diameter of roughly 20–50 m were found on the specimen surfaces after potentiodynamic polarization tests (Fig. 8a), irrespective of cold work level. Localized corrosion was mainly observed along the grain boundaries and deformation bands in the acid sulfate + chloride solution (Fig. 9a and b). Sensitization treatment increased the pitting attack sites in 3.5% NaCl for the HNSSs, since more pits were observed along the grain boundaries on the surface of sensitized specimens (Fig. 8b). Grain boundaries and deformation bands were severely corroded after polarization in the acid sulfate + chloride solution (Fig. 9c). All the specimens exhibited uniform corrosion in the alkaline chloride solution.
3.3. EIS results curves (Fig. 5a) and critical pitting potential (Fig. 7a) with cold work level. After sensitization treatment, the specimens exhibited reduced corrosion resistance in both the 3.5% NaCl and acid sulfate + chloride solutions, as confirmed by the narrower passive region and decreased critical pitting potential (Figs. 6a, b and 7b). The degradation of corrosion resistance appeared more obvious with increasing cold work levels. In the acid sulfate + chloride solution, the passive current density also generally increased with increasing cold work level (Fig. 6a), indicative of the reduced protection of the passive films. Neither cold work nor sensitization
To investigate the relative stability of the passive films formed on the cold worked HNSSs, impedance spectra were recorded at the passive potential after passive film generation. Fig. 10a and b shows three examples of the recorded EIS spectra. The impedance spectra were analyzed using the equivalent electrical circuit widely used for inhomogeneous or microporous electrode surfaces due to an outer precipitated layer (Fig. 10c). A good fit was obtained. The use of a constant phase element (CPE) was necessary [22,23] due to the distribution of relaxation times resulting from heterogeneities
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
1625
Fig. 11. The X-ray photoelectron spectra recorded from the passive film of the (a) 8%, and (b) 49% cold worked non-sensitized specimen at potentials of 0.3 and 0.2 V, respectively. The film was generated in a 3.5% NaCl solution.
at the electrode surface. The impedance of the CPE is given by:
ZCPE =
1 −n (jω) Q
3.4. Characteristic changes of passive film (XPS analysis)
(1)
Therefore, the total impedance is:
n
Ztotal = Rsol + Q (jω) +
1 + R2 Cjω R1 + R2 + R1 R2 Cjω
−1 (2)
where Rsol includes the electrolyte resistance and other ohmic resistances, R2 represents the film resistance, R1 corresponds to the resistance inside the film pores, Q corresponds to the pseudocapacitance of the film, expressed using the CPE, and C the double layer capacity [24]. The fitted parameters are listed in Table 2. The film resistance (R2 ) of the non-sensitized HNSSs decreased with increasing cold work level in both the 3.5% NaCl and alkaline chloride solutions. Little change in film resistance was observed for the films generated in the acid sulfate + chloride solution. Sensitized specimens showed different results. The film resistance decreased in both the 3.5% NaCl and acid sulfate + chloride solution. However, in alkaline solution, it only varied with cold work level. The passive film protection was higher in the 3.5% NaCl solution than in the other two solutions, as shown by the higher film resistance. The resistance inside the film pores (R1 ) was smaller than the film resistance (R2 ). This may be explained by the fact that the electrolyte solution in the micropores or interstitials was partly short-circuited [24].
To explore the influence of cold work, sensitization treatment and pH level on the constitution and other properties of the passive films, the chemical composition of the passive films formed on the HNSSs were first obtained by XPS analysis. Spectra deconvolution of the primary compounds in the HNSS passive film was performed based on their binding energies (BE) (Table 3). The quantitative analysis of passive film composition was performed using the following equation [25]: ox = CM
I ox /S
M ox M
met CM =
Ii /Si
I met /S
M met M Ii
/Si
(3)
(4)
ox and I met is the peak intenwhere CM is the atomic percentage, IM M sity corresponding to the area of the element M in the oxide state (including oxides and hydroxides) and metal state, and SM is the sensitivity factor based on the XPS instrument. In this study, the sensitivity factors of Cr, Fe and Mn were 7.96, 10.82, and 9.17, respectively. This semiquantitative analysis provides a convenient guide to the relative abundance of various species in the passive films. More accurate estimations taking into account the effect of the contamination layer and the uneven composition along the film depth have been developed in other reports [26]. Fig. 11 shows the main oxide composition of the passive films generated in the 3.5% NaCl solution of the 8 and 49% cold worked specimens. The oxide peaks are stronger for the 8% cold worked
1626
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
specimen in comparison with the 49% cold worked one. For the 8% cold worked specimen: ICr-ox /ICr-hy = 1.07, IFe-ox /IFe-hy = 4.7; for the 49% cold worked one: ICr-ox /ICr-hy = 0.97, IFe-ox /IFe-hy = 1.65. It is possible that the protection of the passive film in a heavily cold worked specimen is weaker due to the presence of hydroxides, which can lead to impairment of the resistance to localized corrosion [31]. The signal of metallic state of Fe and Cr from substrate was stronger for the heavily cold worked specimen, indicating a thinner passive film. A high density of dislocations, mechanical twinning and the presence of other defects introduced by cold work, as shown in Fig. 2, could be responsible for the changes in the constitution of the passive film [13], since a high defect density could cause severe atomic disorder, an increase in free energy and thus weaker atomic bonding in the matrix. This would directly influence the quality of passive films formed on the matrix of the heavily cold worked specimens. The less stable and protective passive film and thus lower film resistance (Table 2) resulted in the decreased critical pitting potential with increasing cold work level in the 3.5% NaCl solution (Fig. 7a). The highly defective areas could also be considered high stress regions and potential pitting sites with low barrier energy. Although enhanced surface diffusion due to an increased defect density could contribute to the formation of stable passive film and improvements in pitting resistance [32], it is obvious that the negative effect of cold work on corrosion resistance dominated here. The passive film of the sensitized specimen contained lower concentrations of Mo, Cr, and N in comparison with that of the nonsensitized specimen at the same depth from the surface (Fig. 12). The concentration-depth profile of these elements might be partly an artifact due to ion etching, but comparison between the element concentrations at the same depth level could still give useful information about the effect of sensitization treatment. The changes in concentration of the alloying elements in the passive film could be attributed to the precipitation of the phase during sensitization treatment (Fig. 4). The precipitation of the phase caused reduction of anti-corrosion elements such as Cr and Mo in the matrix and, in turn, in the passive film (Fig. 12). Such reduction of the anticorrosion elements in the film directly resulted in the degradation of corrosion resistance for the cold worked and sensitized specimens. Electrochemical tests clearly showed the decreased critical pitting potential (Fig. 7b) and film resistance (Table 2) for the sensitized HNSSs in comparison with the non-sensitized ones in both
Fig. 12. The X-ray photoelectron spectra depth profiles of (a) Cr, (b) Mo and (c) N recorded from the passive films of the 49% cold worked non-sensitized and sensitized HNSSs at potentials of 0.2 and −0.13 V, respectively.
Table 3 The binding energies of the primary compounds of the HNSS passive film obtained from XPS spectra deconvolution. Element
Peak
Species
Binding energy (eV)
References
Cr
2p3/2
Crmet Cr3+ -ox Cr3+ -hy
573.8 576.3 577.1
[25,26] [25,26] [25,26]
Fe
2p3/2
Femet Fe2+ -ox Fe3+ -ox Fe3+ -hy
706.8 709.0 710.3 711.2
[25,27] [25,27] [25,27] [25,27]
Mo
3d5/2
Momet Mo4+ Mo6+
227.4 229.2 232.2
[26] [26] [26]
O
1s
Oxide Hydroxide
530 531.4
[26,27] [26,27]
N
1s
N in solid solution NH3 -ligand NH4 +
397 399.7 400.3
[28,29] [28,29] [28,29]
Mn
2p3/2
Mnmet Mn2+ -ox Mn3+ -ox
638.8 641.0 641.4
[30] [30] [30]
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
1627
Fig. 13. The X-ray photoelectron spectra recorded from the passive film of the 8% cold worked non-sensitized specimen at a potential of 0.3 V. The film was generated in a 0.5 M H2 SO4 + 0.5 M NaCl solution.
3.5% NaCl and acid sulfate + chloride solutions. The reduction in corrosion resistance was more obvious for larger prior cold worked specimens due to an enhanced precipitation process after cold work [14,15]. Reduced corrosion resistance, particularly along the grain boundaries and deformation bands, could also be observed from the corrosion morphology (Figs. 8b and 9c). The passive film formed in the acid sulfate + chloride solution contained more hydroxides than oxides (Fig. 13): ICr-ox /ICr-hy = 0.99, IFe-ox /IFe-hy = 0.45. The metallic signal from the substrate was quite strong, possibly due to a thin passive film. The less protective and stable passive film might be responsible for its lower film resistance in comparison with that in the other two solutions before large cold work, as confirmed by EIS tests (Table 2). It was also observed that the passive current density was higher in the acid sulfate + chloride solution than in the other two solutions (Fig. 14). The measured passive current density i consists of the partial current densities of all the anodic or cathodic processes [33,34]: i = iox + iredox + icorr + iC
were present at concentration in the passive film even for the HNSS without cold work, the negative effect of cold work on the changes of passive film stability and thus corrosion resistance was not as obvious as in the 3.5% NaCl solution, as shown in electrochemical tests (Fig. 7a and Table 2). It is also possible that enhanced surface diffusion due to increased defect density could contribute to film formation more in an acid sulfate + chloride solution than in a 3.5% NaCl solution, since metal ions travel through the oxide film with a higher speed, as indicated by the higher passive current density (Fig. 14). Thus the transport of alloying elements from metal base to the metal/oxide interface played a larger role in the acid sulfate + chloride solution. The combined negative and positive effects of cold work might result in little overall change in the film constitution and resistance as well as the corrosion resistance.
(5)
where iox denotes the transfer of oxygen ions from the electrolyte into the oxide; icorr , the transfer of metal ions from the oxide into the electrolyte; iredox , electron transfer reactions, e.g. hydrogen or oxygen evolution; iC , capacitive charging. The anodic corrosion of a passive metal at steady state, i.e. at constant thickness such that iox = 0 is a straightforward process. In such a case, metal ions travel through the oxide film at a constant rate and are transferred in an ion transfer reaction (ITR) at the oxide/electrolyte interface. Thus i can approximate the dissolution rate of the passive film, icorr . Therefore, the higher dissolution rate of the passive film in the acid sulfate + chloride solution might be responsible for the reduced protection of the passive film (Fig. 13) and lower film resistance (Table 2) in comparison to the other two solutions. Since the stable oxides were present at low concentration while the hydroxides
Fig. 14. Current–time transients at passive potentials generated in three solutions for films on the 8% cold worked non-sensitized specimen.
1628
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
Table 4 Atomic percent of main oxides in the passive films generated in the three solutions. The film generated potentials are 0.3 V. 8% cold worked unsensitized HNSS
ox CCr
ox CFe
ox CMn
0.5 m/L H2 SO4 + 0.5 m/L NaCl 3.5% NaCl 0.5 m/L NaOH + 0.5 m/L NaCl
0.65 0.28 0.20
0.24 0.59 0.58
0.11 0.13 0.22
The cold work-dependent corrosion resistance for the HNSS in the 3.5% NaCl and acid sulfate + chloride solutions was much different from that for the N-bearing stainless steel investigated previously. It has been reported that an increase in the cold work level of type 316LSS (0.07% N) from 5 to 25% decreases the pitting resistance in acidic chloride medium (pH < 1). While in neutral chloride medium, the corrosion resistance improved with up to 20% cold work before being degraded [32]. This was due to the heavy deformation resulting in enhanced dissolution in acidic chloride medium, where metal dissolution plays a dominant role in the rate-control mechanism. This was not the case in a neutral chloride solution, where oxygen diffusion was the rate-controlling step. The reason for the above difference is unknown and could be partially due to differences in the materials and solution systems investigated. The passive film formed in the alkaline chloride solution contained the largest amount of O. The Cr oxide content was lowest and Mn oxide content the highest among the films formed in all the three solutions (Table 4). The accumulation of chromium oxide on the surface of the passive film was more obvious for the HNSS in the acid sulfate + chloride solution than in the alkaline solution (Table 4). This is reasonable as chromium oxide dissolves readily in alkaline solution while iron oxide, unlike chromium oxide, dissolves readily in acid sulfate + chloride solution [35]. This helps to explain the higher film protection in 3.5% NaCl solution, since the dissolution rates of both iron oxide and chromium oxide are comparably lower. It has been reported that the corrosion resistance of a FeCr alloy was better than pure Cr in 0.1 M NaOH [35]. This is due to the iron oxide present in the outer layer of the passive film, which increases the overvoltage for the transpassive dissolution of Cr. Thus, in alkaline solution, iron oxide enables the passive film on the FeCr alloy to be stable over a wider potential range than on pure chromium. Therefore, iron oxide contributes to the stability of the passive film more in alkaline solution, although the effect of Cr is not negligible. The influence of cold work and sensitization treatment on corrosion resistance would be much smaller if iron oxide is used to facilitate the transpassive dissolution instead of another alloying element. This could partly explain the small change in the polarization curve in alkaline chloride solution (Figs. 5c and 6c). It is also possible that the alkaline condition inhibited pitting before transpassive dissolution. Thus, the specimens exhibited similar polarization behavior and no pitting took place even if the film protection was weak. More detailed studies are necessary to clarify this problem. Mo6+ and NH4+ were detected on the surface of passive films formed in all three chloride solutions (Fig. 13), in agreement with the results of previous studies [28,29]. 4. Conclusions The effects of cold work and sensitization treatment on the microstructure of a HNSS were studied by SEM and TEM. The corresponding changes in corrosion resistance in chloride solutions of different pH have been investigated using potentiodynamic and potentiostatic polarization. The nature of passive films has been analyzed in situ by EIS measurements and ex situ by XPS analysis. The following conclusions can be drawn:
1. Increasing the cold work level (up to 60%) steadily decreased the corrosion resistance of the non-sensitized HNSS in a 3.5% NaCl solution. The passive film of the heavily cold worked HNSS contained fewer oxides and more hydroxides. The less compact and protective film could be attributed to the high defect density in the steel base introduced by the cold work. 2. phases were observed to precipitate both at the grain boundaries and also within the grains during sensitization treatment. The reduction of anti-corrosion elements in the matrix and passive film resulted in the degradation of the corrosion resistance of sensitized HNSSs in both 3.5% NaCl and 0.5 M H2 SO4 + 0.5 M NaCl solutions. The degradation in corrosion resistance was more dramatic with increasing prior cold work level. 3. The reduction in corrosion resistance by cold work was not obvious in a 0.5 M H2 SO4 + 0.5 M NaCl solution for the non-sensitized HNSS. This might be related to the stability of the passive film in this solution. 4. Although the film resistance changed, neither cold work nor sensitization treatment affected the electrochemical polarization behavior of the HNSS in an alkaline chloride solution. All the HNSSs exhibited uniform corrosion in alkaline chloride solution. 5. No obvious changes in pit size and distribution were observed for the non-sensitized HNSSs with different cold work levels in a 3.5% NaCl solution. Sensitized HNSSs showed more pits on the corroded surfaces in comparison with non-sensitized ones at the same level of cold work. In the 0.5 M H2 SO4 + 0.5 M NaCl solution, corrosion occurred at the most vulnerable positions, namely, at the grain and twin boundaries for small cold worked specimens and along the deformation bands for the large cold worked specimens. Acknowledgments This study was jointly supported by the Science and Technology Foundation of China (50534010), and the Innovation Fund of Institute of Metal Research (IMR), Chinese Academy of Sciences (CAS). References [1] J.W. Simmons, Mater. Sci. Eng. A 207 (1996) 159. [2] H. Hanninen, J. Romu, R. Ilola, J. Tervo, A. Laitinen, J. Mater. Process. Technol. 117 (2001) 424. [3] T. Briglia, G. Terwagne, F. Bodart, C. Quaeyhaegens, J. D’Haen, L.M. Stals, Surf. Coat. Technol. 80 (1996) 105. [4] X.Q. Wu, S. Xu, J.B. Huang, E.H. Han, W. Ke, K. Yang, Z.H. Jiang, Mater. Corros. 59 (2008) 676. [5] X.Q. Wu, Y. Fu, J.B. Huang, E.H. Han, W. Ke, K. Yang, Z.H. Jiang, J. Mater. Eng. Perform., (2008) doi:10.1007/s11665-008-9295-4. [6] A. Toro, W.Z. Misiolek, A.P. Tschiptschin, Acta Mater. 51 (2003) 3363. [7] Y.J. Oh, J.H. Hong, J. Nucl. Mater. 278 (2000) 242. [8] L.R. Scharfstein, in: D. Peckner, I.M. Bernstein (Eds.), Handbook of Stainless Steels, McGraw Hill, 1977 (Ch. 15). [9] H. Ha, H. Kwon, Electrochim. Acta 52 (2007) 2175. [10] T.-H. Lee, C.-S. Oh, S.-J. Kim, S. Takaki, Acta Mater. 55 (2007) 3649. [11] L. Peguet, B. Malki, B. Baroux, Corros. Sci. 49 (2007) 1933. [12] A. Barbucci, G. Cerisola, P.L. Cabot, J. Electrochem. Soc. 149 (2002) 534. [13] Y. Fu, X.Q. Wu, E.H. Han, W. Ke, K. Yang, Z.H. Jiang, J. Electrochem. Soc. 155 (2008) C455. [14] N. Parvathavarthini, R.K. Dayal, S.K. Seshadri, J.B. Gnanamoorthy, J. Nucl. Mater. 168 (1989) 83. [15] S.K. Mannan, R.K. Dayal, M. Vijayalakshmi, N. Parvathavarthini, J. Nucl. Mater. 126 (1984) 1. [16] C.L. Briant, Effect of nitrogen and cold work on the sensitization of austenitic stainless steels, Electric Power Research Institute Report EPRI-NP-2457, 1982. [17] L.E. Murr, A. Advani, S. Shankar, D.G. Atteridge, Mater. Char. 24 (1990) 135. [18] L.K. Mansur, J. Nucl. Mater. 83 (1979) 109. [19] N. Thompson, Proc. Phys. Soc. Lond. B 66 (1953) 481. [20] S. Mahajan, Metallography 4 (1971) 43. [21] P. Liu, Phase Analysis in Steel using Analytical Transmission Electron Microscopy, Sandvik Materials Technology, Sandviken, 2004.
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629 [22] A. Carnot, I. Frateur, S. Zanna, B. Tribollet, I. Dubois-Brugger, P. Marcus, Corros. Sci. 45 (2003) 2513. [23] C. Hitz, A. Lasia, J. Electroanal. Chem. 500 (2001) 213. [24] Z. Grubac, M. Metikos-Hukovic, J. Electroanal. Chem. 565 (2004) 85. [25] C.T. Liu, J.K. Wu, Corros. Sci. 49 (2007) 2198. [26] I. Olefjord, B. Brox, U. Jelvestam, J. Electrochem. Soc. 132 (1985) 2854. [27] I. Olefjord, L. Wegrelius, Corros. Sci. 38 (1996) 1203. [28] Y.C. Lu, M.B. Ives, C.R. Clayton, Corros. Sci. 35 (1993) 89. [29] C.-O.A. Olsson, Corros. Sci. 37 (1995) 467.
1629
[30] C.D. Wagner, J.F. Moulder, L.E. Davis, W.M. Riggs, Handbook of X-ray Photoelectron Spectroscopy, Perkin-Elmer, Norwalk, Conn, USA, 1989. [31] C.R. Clayton, Y.C. Lu, J. Electrochem. Soc. 133 (1986) 2465. [32] U. Kamachi Mudali, P. Shankar, S. Ningshen, R.K. Dayal, H.S. Khatak, B. Raj, Corros. Sci. 44 (2002) 2183. [33] B. Heine, R. Kirchheim, Corros. Sci. 31 (1990) 533. [34] J.W. Schultze, M.M. Lohrengel, Electrochim. Acta 45 (2000) 2499. [35] P. Schmutz, D. Landolt, Corros. Sci. 41 (1999) 2143.
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Effects of cold work and sensitization treatment on the corrosion resistance of high nitrogen stainless steel in chloride solutions Yao Fu a , Xinqiang Wu a,∗ , En-Hou Han a , Wei Ke a , Ke Yang a , Zhouhua Jiang b a State key Laboratory for Corrosion and Protection, Institute of Metal Research, South Campus, Chinese Academy of Sciences, 62 Wencui Road, Shenyang 110016, PR China b School of Materials and Metallurgy, Northeastern University, Shenyang 110004, PR China
a r t i c l e
i n f o
Article history: Received 31 July 2008 Accepted 24 September 2008 Available online 8 October 2008 Keywords: High nitrogen stainless steel Cold work Sensitization treatment Corrosion resistance X-ray photoelectron spectroscopy
a b s t r a c t The effects of cold work and sensitization treatment on the microstructure and corrosion resistance of a nickel-free high nitrogen stainless steel (HNSS) in 0.5 M H2 SO4 + 0.5 M NaCl, 3.5% NaCl and 0.5 M NaOH + 0.5 M NaCl solutions have been investigated by microscopic observations, electrochemical tests and surface chemical analysis. Cold work introduced a high defect density into the matrix, resulting in a less protective passive film as well as reduced corrosion resistance for heavily cold worked HNSS in a 3.5% NaCl solution. No obvious degradation in corrosion resistance took place in a 0.5 M H2 SO4 + 0.5 M NaCl solution, possibly due to the stability of the passive film in this solution. Sensitized HNSSs showed reduced corrosion resistance with increasing cold work level in both 3.5% NaCl and 0.5 M H2 SO4 + 0.5 M NaCl solutions due to a reduction in the anti-corrosion elements in the matrix during the cold workaccelerated precipitation process. The cold work and sensitization treatment had no influence on the corrosion resistance of the HNSS in the 0.5 M NaOH + 0.5 M NaCl solution even though the property of the passive film changed. The effects of cold work and sensitization treatment on the characteristics of passive films formed in the three solutions are discussed. © 2008 Elsevier Ltd. All rights reserved.
1. Introduction The beneficial effects of nitrogen on the properties of highalloyed steels have led to the widespread development of high nitrogen stainless steels (HNSSs) due to recent advances in processing technologies [1–3]. As an alloying element, nitrogen has several beneficial effects on the properties of steels, in particular, those related to high yield strength and toughness (good fracture resistance) [1]. Nitrogen is also a strong austenite-stabilizing element and can thus substitute for nickel, which is expensive and causes an allergic reaction in human skin. Moreover, nitrogen alloying, especially in combination with molybdenum, improves resistance to localized corrosion in general, and resistance to general corrosion in some environments [4,5]. One problem with HNSS is the formation of precipitates such as Cr-nitride, , and upon thermal exposure in the temperature range of 500–1050 ◦ C, leading to a significant reduction in corrosion resistance [6,7]. It has been reported that the deleterious effects of Cr2 N on pitting corrosion are associated with
∗ Corresponding author. Tel.: +86 24 23841883; fax: +86 24 23894149. E-mail address: [email protected] (X. Wu). 0013-4686/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2008.09.053
the formation of a Cr-depletion zone adjacent to Cr2 N precipitates [8]. It was also found that the lamellar Cr2 N was the most susceptible site for pitting corrosion when compared with any other heterogeneity or Cr2 N precipitate along the grain boundaries [9]. Stainless steels are subject to different levels of cold work during the final manufacturing stages of components for numerous applications in industry. Cold work might affect the corrosion resistance of stainless steels because deformed substructures like planar dislocation arrays [1,7] and deformation twinning [10] could be introduced. Peguet et al. [11] have reported the different roles of cold work on the pitting corrosion resistance at different pitting stages, including pit initiation, propagation and repassivation. Barbucci et al. [12] reported that the passive currents of 304 SS in sulfate + chloride solutions significantly increased with the degree of cold work. In addition, the pitting corrosion resistance was observed to decrease with increasing cold work in a 3.5% NaCl solution [13]. Efforts have also been made to clarify the relationship between cold work and the sensitization process. The cold work effect in Type 316 SS has been determined to be due to the higher diffusivity of chromium and the lower free energy barrier to carbide nucleation at grain boundaries in the deformed microstructure [14–17]. The acceleration of sensitization with cold work could also be related to the effects of
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
point defects and microstructural sinks on diffusion, as reported by Mansur [18]. The metallurgical changes caused by cold work and sensitization treatment have not been related clearly to the influence on the corrosion resistance in chloride solutions of different pH. Moreover, previous studies have not included the composition and structural evolution of passive films, which have direct influence on the corrosion resistance. In the present work, we examine the microstructure and corrosion resistance of a HNSS after cold work and/or sensitization treatment. The structural and chemical composition of the passive film induced by cold work and sensitization treatment were analyzed by electrochemical impedance spectroscopy (EIS), and X-ray photoelectron spectroscopy (XPS). The morphology of the pitting attack after electrochemical polarization was observed. The relationship between the microstructure evolution, changes in passive film characteristics and variations in the corrosion resistance of the HNSS are discussed. 2. Experimental 2.1. Microstructure observation The chemical composition of the HNSS investigated is listed in Table 1. The steel plates were cold rolled to a 8, 30, 49 or 60% reduction in thickness. Some of the specimens were then isothermally sensitized at 650 ◦ C for 2 h. 10 mm × 10 mm specimens were cut from the plates, with test surfaces parallel to the rolling direction. The specimens were polished to a diamond finish (1.5 m) and electrolytically etched in a 10% ethane diacid reagent at 12 V for 90 s to observe the changes in microstructure introduced by cold work. Specimens for transmission electron microscopy (TEM) analysis were mechanically polished and punched into 3 mm discs. Final
1619
Table 1 Chemical composition of the high nitrogen stainless steel in this work (in wt.%). C Cr Mn Mo N Si S P Fe
0.04 18.4 15.8 2.19 0.66 0.24 0.005 0.017 Bal
thinning was performed using a twin-jet electrolytic polisher operated at 20 V in an electrolyte of perchloric acid and alcohol at −30 ◦ C. The substructure of the cold worked and sensitized specimen was analyzed by an FEI Tecnai G2 F20 S-Twin microscope. 2.2. Electrochemical tests 2.2.1. Electrodes and electrolyte To prepare working electrodes for the electrochemical tests, the specimens were sealed with an epoxy resin to expose an area of 1 cm2 . The exposed faces were polished with 800 grit silicon carbide paper and rinsed with de-ionized water just before immersion. The specimens used to observe pitting attacks after polarization tests were polished to a diamond finish (1.5 m). After immersion for 5 min, the working electrode was passivated at the film formation potential for 20 min for EIS measurements and 1 h for XPS measurements, respectively. A three electrode cell featuring a Pt counter electrode and a saturated calomel reference electrode (SCE) was employed. All the potentials in the paper are given vs. this reference electrode. The solutions used in this investigation were: solution I, 0.5 M H2 SO4 + 0.5 M NaCl (pH ≈ 0.4); solution II, 3.5% (0.6 M) NaCl (pH ≈ 5.8); and solution III, 0.5 M NaOH + 0.5 M
Fig. 1. SEM images of (a) deformation bands in an 8% cold worked non-sensitized HNSS, (b) deformation bands in a 30% cold worked non-sensitized HNSS, (c) weakened grain, twin boundaries in an 8% and also (d) deformation bands in a 30% cold worked sensitized HNSS.
1620
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
Fig. 2. TEM images of the (a) the planar dislocation arrays, (b) dislocation pile-ups in the 8% cold worked non-sensitized specimen, (c) deformation twinning and high density of dislocation (the diffraction pattern on the bottom-right was taken from the junction of the twinning and matrix), (d) high density of defects and misorientation within the grain of the 49% cold worked non-sensitized specimen.
NaCl (pH ≈ 13.1). The electrolytes were prepared from concentrated sulfuric acid, analytical grade sodium chloride, sodium hydroxide and de-ionized water. The experiments were carried out at room temperature under naturally aerated conditions.
lution spectra of the following regions were recorded: Fe 2p, Cr 2p, Mn 2p, Mo 3d, N 1s, O 1s, and C 1s. The hydrocarbon C 1s peak was assumed to be at 284.6 eV and was used as an internal standard to determine the binding energy of other photoelectron peaks. A
2.2.2. Apparatus and procedures Polarization curves were conducted at a sweep rate of 0.1667 mV s−1 , using an EG&G 273A Potentiostat controlled by PowerSuite software. Impedance measurements were performed with a lock-in amplifier (EG&G 5210) coupled to a potentiostat (EG&G 273A), using a 10 mV (rms) perturbation from 100 kHz to 10 mHz. Fitting was performed with ZSimpwin software. The chemical composition profile for the passive film was obtained by XPS analysis using the Al K␣ X-ray source (1487 eV, 150 W) and a pass energy of 20 eV. Characterization of the passive films was performed at a takeoff angle of 45◦ . The depth profile of the chemical composition was obtained by sputtering with argon ions (base pressure: 1E−7 Torr, energy: 2 kV, current: 2.0 A cm−2 ) and a pass energy of 50 eV. A survey spectrum was first recorded to identify all the elements present at the surface. Then, high reso-
Fig. 3. Schematic illustration of the SAD pattern of the deformation twinning in Fig. 2c.
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
1621
Fig. 4. TEM images of (a) small particles of phase at the grain boundary and (b) within the grain of the 49% cold worked and sensitized specimen, (c) the SAD pattern with [0 1¯ 1] orientation and (d) the SAD pattern with [1¯ 0 4] orientation of the precipitates. (e) TEM-EDS analysis of the precipitates.
Shirley background subtraction was made to obtain the XPS signal intensity. 3. Results and discussion 3.1. Microstructure observation The cold worked specimens showed an increase in the deformation bands with increasing cold work level (Fig. 1a and b). TEM observation showed that the planar dislocation arrays were most prevalent at the beginning of the deformation (Fig. 2a). Dislocation pile-ups (Fig. 2b) were observed at the grain boundaries. At a higher deformation stage, the deformed microstructure was characterized by pronounced deformation twinning, and a high density of dislocation between twin lamellae (Fig. 2c). The alternating black
and white regions with high contrast in Fig. 2d suggests that the misorientation within the grain and the density of defects were rather high. The disordered multi-spots in the diffraction pattern indicated the existence of multiple fine subgrains with different orientations. The exact twinning plane could be determined from the combined analyses of the selected area diffraction (SAD) pattern and the transformation matrix [19]. A typical SAD pattern composed of matrix and deformation twinning is shown in Fig. 2c, where the matrix was tilted to its [1¯ 1 4] direction. The extra spots showed the characteristic arrangement of a SAD pattern of fcc crystals. According to the transformation matrix for deformation twinning in fcc materials calculated by Mahajan [20], it can be shown that ¯ planes transthe twinning on the (1 1 1), (1¯ 1 1), (1 1¯ 1) and (1 1 1) ¯ [1¯ 1 0], [7 7¯ 8] forms the [1¯ 1 4] matrix direction into the [1 1 5 4],
1622
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
Fig. 5. Potentiodynamic polarization curves of the cold worked non-sensitized HNSSs in (a) 0.5 M H2 SO4 + 0.5 M NaCl solution, (b) 3.5% NaCl solution, and (c) 0.5 M NaOH + 0.5 M NaCl solution.
and [5 1 1 4] twinning directions, respectively. Among these directions, only the [1¯ 1 0] direction is consistent with the observed diffraction pattern shown in Fig. 2c. Assuming that the [1¯ 1 0] orientation is twinning, the SAD pattern of Fig. 2c can be indexed as shown in Fig. 3. Based on the combined analyses of the SAD pattern with the transformation matrix, it can be concluded that the crystallographic twinning plane was the exact (1¯ 1 1) plane in this study. Fig. 1c and d are the microstructures of the sensitized HNSSs at different cold work levels. After metallographic etching, the 8% cold worked samples showed many corrosion pits at the grain and twin boundaries. The 30% cold worked samples, however, exhibited serious corrosion pits at deformation bands adjacent to the grain and twin boundaries. No precipitates were detected by X-ray diffraction; however, by TEM observation, the SAD pattern (Fig. 4c and d) and TEM-EDS (Fig. 4e) analysis indicated that small particles of phase had precipitated at the grain boundaries (Fig. 4a) and also within the grains (Fig. 4b), with a chemical composition similar to previous reports [21]. Here, extensive dislocation networks, mechanical twinning and the subgrain boundaries introduced by
Fig. 6. Potentiodynamic polarization curves of the cold worked and sensitized HNSSs in (a) 0.5 M H2 SO4 + 0.5 M NaCl solution, (b) 3.5% NaCl solution, and (c) 0.5 M NaOH + 0.5 M NaCl solution.
cold work provide more sites for rapid pipe diffusion and faster nucleation or more point defects and microstructural sinks on diffusion, as reported by Mansur [18]. Therefore, more phase particles were able to precipitate during sensitization treatment for the specimen with more cold work. 3.2. Potentiodynamic study Figs. 5 and 6 are the potentiodynamic polarization diagrams for the cold worked HNSSs before and after sensitization treatment in chloride solution with different pH levels. Fig. 7 shows the variation in critical pitting potential with cold work level. Each curve exhibited a range of passivity beyond the open-circuit potential. The specimens tested in the acid sulfate + chloride solution showed active dissolution before passivation. For the non-sensitized HNSSs, the critical pitting potential decreased and the passive regions became narrower in 3.5% NaCl solution with increasing cold work level (Fig. 5b). However, in the acid sulfate + chloride solution, no obvious change could be observed from the polarization
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
Fig. 7. The variation of the critical pitting potential as a function of the cold work level of (a) the cold worked non-sensitized HNSS and (b) the cold worked and sensitized HNSS.
1623
Fig. 8. SEM images of the pitted surface for the (a) 49% cold worked non-sensitized specimen and (b) 49% cold worked and sensitized specimen after potentiodynamic polarization in a 3.5% NaCl solution.
Table 2 Equivalent circuit parameters for the HNSSs, tested at the passive potentials. Rsol ( cm2 )
Q (−1 sn cm−2 )
n
R1 ( cm2 )
C (F cm−2 )
R2 ( cm2 )
8% (0.3) 30% (0.3) 60% (0.3)
1.85 1.60 1.70
4.47E−05 5.04E−05 4.71E−05
0.93 0.94 0.94
48,260 66,200 58,370
1.13E−04 2.96E−04 7.08E−05
142,400 102,600 101,000
3.5% NaCl
8% (0.3) 30% (0.25) 60% (0.17)
7.62 7.33 7.23
1.14E−05 1.20E−05 2.76E−05
0.88 0.83 0.95
3.56 4.01 115,600
6.53E−06 7.79E−06 8.38E−05
648,000 252,700 179,900
0.5 M NaOH + 0.5 M NaCl
8% (0.3) 30% (0.3) 60% (0.3)
2.85 2.14 2.33
4.18E−05 6.67E−05 5.86E−05
0.79 0.73 0.78
1.36 1.32 1.26
2.22E−05 1.86E−05 1.46E−05
227,400 124,000 27,600
8% (0.05) 30% (−0.16) 60% (−0.18)
1.55 2.10 2.36
7.31E−05 6.65E−05 6.60E−05
0.79 0.86 0.87
0.77 11.25 13,490
1.59E−05 1.96E−06 5.56E−04
52,050 32,780 14,290
8% (0.02) 30% (−0.12) 60% (−0.15)
7.57 6.17 6.26
1.37E−05 8.21E−05 9.51E−05
0.76 0.85 0.78
4.83 10.05 3.02
8.43E−06 1.26E−05 1.65E−05
165,200 136,300 128,800
2.48 3.09 2.86
8.72E−05 4.90E−05 7.11E−05
0.8 0.9 0.81
1.28 20,560 1.88
1.52E−05 6.46E−05 1.44E−05
57,790 29,650 109,900
Solution (a) Non-sensitized HNSS 0.5 M H2 SO4 + 0.5 M NaCl
(b) Sensitized HNSS 0.5 M H2 SO4 + 0.5 M NaCl
3.5% NaCl
0.5 M NaOH + 0.5 M NaCl
Cold work level (film generated potential, V)
8% (0.3) 30% (0.3) 60% (0.3)
1624
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
Fig. 10. Examples of (a) Nyquist and (b) Bode plots for the passive films of the HNSS formed at the passive potential, and (c) the equivalent circuit used for quantitative evaluation of EIS spectra.
Fig. 9. SEM images of the pitted surface for the (a) 8% and (b) 49% cold worked nonsensitized specimen and (c) the 49% cold worked and sensitized specimen after potentiodynamic polarization in a 0.5 M H2 SO4 + 0.5 M NaCl solution.
treatment changed the electrochemical polarization behavior of the HNSSs in alkaline chloride solution (Figs. 5c and 6c). SEM observations of the specimens after polarization to 1E−3 A cm−2 were performed to study the pit size and distribution. The non-sensitized specimens showed no significant changes in the size and distribution of corrosion pits with increasing cold work level in 3.5% NaCl. Several large pits with a diameter of roughly 20–50 m were found on the specimen surfaces after potentiodynamic polarization tests (Fig. 8a), irrespective of cold work level. Localized corrosion was mainly observed along the grain boundaries and deformation bands in the acid sulfate + chloride solution (Fig. 9a and b). Sensitization treatment increased the pitting attack sites in 3.5% NaCl for the HNSSs, since more pits were observed along the grain boundaries on the surface of sensitized specimens (Fig. 8b). Grain boundaries and deformation bands were severely corroded after polarization in the acid sulfate + chloride solution (Fig. 9c). All the specimens exhibited uniform corrosion in the alkaline chloride solution.
3.3. EIS results curves (Fig. 5a) and critical pitting potential (Fig. 7a) with cold work level. After sensitization treatment, the specimens exhibited reduced corrosion resistance in both the 3.5% NaCl and acid sulfate + chloride solutions, as confirmed by the narrower passive region and decreased critical pitting potential (Figs. 6a, b and 7b). The degradation of corrosion resistance appeared more obvious with increasing cold work levels. In the acid sulfate + chloride solution, the passive current density also generally increased with increasing cold work level (Fig. 6a), indicative of the reduced protection of the passive films. Neither cold work nor sensitization
To investigate the relative stability of the passive films formed on the cold worked HNSSs, impedance spectra were recorded at the passive potential after passive film generation. Fig. 10a and b shows three examples of the recorded EIS spectra. The impedance spectra were analyzed using the equivalent electrical circuit widely used for inhomogeneous or microporous electrode surfaces due to an outer precipitated layer (Fig. 10c). A good fit was obtained. The use of a constant phase element (CPE) was necessary [22,23] due to the distribution of relaxation times resulting from heterogeneities
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
1625
Fig. 11. The X-ray photoelectron spectra recorded from the passive film of the (a) 8%, and (b) 49% cold worked non-sensitized specimen at potentials of 0.3 and 0.2 V, respectively. The film was generated in a 3.5% NaCl solution.
at the electrode surface. The impedance of the CPE is given by:
ZCPE =
1 −n (jω) Q
3.4. Characteristic changes of passive film (XPS analysis)
(1)
Therefore, the total impedance is:
n
Ztotal = Rsol + Q (jω) +
1 + R2 Cjω R1 + R2 + R1 R2 Cjω
−1 (2)
where Rsol includes the electrolyte resistance and other ohmic resistances, R2 represents the film resistance, R1 corresponds to the resistance inside the film pores, Q corresponds to the pseudocapacitance of the film, expressed using the CPE, and C the double layer capacity [24]. The fitted parameters are listed in Table 2. The film resistance (R2 ) of the non-sensitized HNSSs decreased with increasing cold work level in both the 3.5% NaCl and alkaline chloride solutions. Little change in film resistance was observed for the films generated in the acid sulfate + chloride solution. Sensitized specimens showed different results. The film resistance decreased in both the 3.5% NaCl and acid sulfate + chloride solution. However, in alkaline solution, it only varied with cold work level. The passive film protection was higher in the 3.5% NaCl solution than in the other two solutions, as shown by the higher film resistance. The resistance inside the film pores (R1 ) was smaller than the film resistance (R2 ). This may be explained by the fact that the electrolyte solution in the micropores or interstitials was partly short-circuited [24].
To explore the influence of cold work, sensitization treatment and pH level on the constitution and other properties of the passive films, the chemical composition of the passive films formed on the HNSSs were first obtained by XPS analysis. Spectra deconvolution of the primary compounds in the HNSS passive film was performed based on their binding energies (BE) (Table 3). The quantitative analysis of passive film composition was performed using the following equation [25]: ox = CM
I ox /S
M ox M
met CM =
Ii /Si
I met /S
M met M Ii
/Si
(3)
(4)
ox and I met is the peak intenwhere CM is the atomic percentage, IM M sity corresponding to the area of the element M in the oxide state (including oxides and hydroxides) and metal state, and SM is the sensitivity factor based on the XPS instrument. In this study, the sensitivity factors of Cr, Fe and Mn were 7.96, 10.82, and 9.17, respectively. This semiquantitative analysis provides a convenient guide to the relative abundance of various species in the passive films. More accurate estimations taking into account the effect of the contamination layer and the uneven composition along the film depth have been developed in other reports [26]. Fig. 11 shows the main oxide composition of the passive films generated in the 3.5% NaCl solution of the 8 and 49% cold worked specimens. The oxide peaks are stronger for the 8% cold worked
1626
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
specimen in comparison with the 49% cold worked one. For the 8% cold worked specimen: ICr-ox /ICr-hy = 1.07, IFe-ox /IFe-hy = 4.7; for the 49% cold worked one: ICr-ox /ICr-hy = 0.97, IFe-ox /IFe-hy = 1.65. It is possible that the protection of the passive film in a heavily cold worked specimen is weaker due to the presence of hydroxides, which can lead to impairment of the resistance to localized corrosion [31]. The signal of metallic state of Fe and Cr from substrate was stronger for the heavily cold worked specimen, indicating a thinner passive film. A high density of dislocations, mechanical twinning and the presence of other defects introduced by cold work, as shown in Fig. 2, could be responsible for the changes in the constitution of the passive film [13], since a high defect density could cause severe atomic disorder, an increase in free energy and thus weaker atomic bonding in the matrix. This would directly influence the quality of passive films formed on the matrix of the heavily cold worked specimens. The less stable and protective passive film and thus lower film resistance (Table 2) resulted in the decreased critical pitting potential with increasing cold work level in the 3.5% NaCl solution (Fig. 7a). The highly defective areas could also be considered high stress regions and potential pitting sites with low barrier energy. Although enhanced surface diffusion due to an increased defect density could contribute to the formation of stable passive film and improvements in pitting resistance [32], it is obvious that the negative effect of cold work on corrosion resistance dominated here. The passive film of the sensitized specimen contained lower concentrations of Mo, Cr, and N in comparison with that of the nonsensitized specimen at the same depth from the surface (Fig. 12). The concentration-depth profile of these elements might be partly an artifact due to ion etching, but comparison between the element concentrations at the same depth level could still give useful information about the effect of sensitization treatment. The changes in concentration of the alloying elements in the passive film could be attributed to the precipitation of the phase during sensitization treatment (Fig. 4). The precipitation of the phase caused reduction of anti-corrosion elements such as Cr and Mo in the matrix and, in turn, in the passive film (Fig. 12). Such reduction of the anticorrosion elements in the film directly resulted in the degradation of corrosion resistance for the cold worked and sensitized specimens. Electrochemical tests clearly showed the decreased critical pitting potential (Fig. 7b) and film resistance (Table 2) for the sensitized HNSSs in comparison with the non-sensitized ones in both
Fig. 12. The X-ray photoelectron spectra depth profiles of (a) Cr, (b) Mo and (c) N recorded from the passive films of the 49% cold worked non-sensitized and sensitized HNSSs at potentials of 0.2 and −0.13 V, respectively.
Table 3 The binding energies of the primary compounds of the HNSS passive film obtained from XPS spectra deconvolution. Element
Peak
Species
Binding energy (eV)
References
Cr
2p3/2
Crmet Cr3+ -ox Cr3+ -hy
573.8 576.3 577.1
[25,26] [25,26] [25,26]
Fe
2p3/2
Femet Fe2+ -ox Fe3+ -ox Fe3+ -hy
706.8 709.0 710.3 711.2
[25,27] [25,27] [25,27] [25,27]
Mo
3d5/2
Momet Mo4+ Mo6+
227.4 229.2 232.2
[26] [26] [26]
O
1s
Oxide Hydroxide
530 531.4
[26,27] [26,27]
N
1s
N in solid solution NH3 -ligand NH4 +
397 399.7 400.3
[28,29] [28,29] [28,29]
Mn
2p3/2
Mnmet Mn2+ -ox Mn3+ -ox
638.8 641.0 641.4
[30] [30] [30]
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
1627
Fig. 13. The X-ray photoelectron spectra recorded from the passive film of the 8% cold worked non-sensitized specimen at a potential of 0.3 V. The film was generated in a 0.5 M H2 SO4 + 0.5 M NaCl solution.
3.5% NaCl and acid sulfate + chloride solutions. The reduction in corrosion resistance was more obvious for larger prior cold worked specimens due to an enhanced precipitation process after cold work [14,15]. Reduced corrosion resistance, particularly along the grain boundaries and deformation bands, could also be observed from the corrosion morphology (Figs. 8b and 9c). The passive film formed in the acid sulfate + chloride solution contained more hydroxides than oxides (Fig. 13): ICr-ox /ICr-hy = 0.99, IFe-ox /IFe-hy = 0.45. The metallic signal from the substrate was quite strong, possibly due to a thin passive film. The less protective and stable passive film might be responsible for its lower film resistance in comparison with that in the other two solutions before large cold work, as confirmed by EIS tests (Table 2). It was also observed that the passive current density was higher in the acid sulfate + chloride solution than in the other two solutions (Fig. 14). The measured passive current density i consists of the partial current densities of all the anodic or cathodic processes [33,34]: i = iox + iredox + icorr + iC
were present at concentration in the passive film even for the HNSS without cold work, the negative effect of cold work on the changes of passive film stability and thus corrosion resistance was not as obvious as in the 3.5% NaCl solution, as shown in electrochemical tests (Fig. 7a and Table 2). It is also possible that enhanced surface diffusion due to increased defect density could contribute to film formation more in an acid sulfate + chloride solution than in a 3.5% NaCl solution, since metal ions travel through the oxide film with a higher speed, as indicated by the higher passive current density (Fig. 14). Thus the transport of alloying elements from metal base to the metal/oxide interface played a larger role in the acid sulfate + chloride solution. The combined negative and positive effects of cold work might result in little overall change in the film constitution and resistance as well as the corrosion resistance.
(5)
where iox denotes the transfer of oxygen ions from the electrolyte into the oxide; icorr , the transfer of metal ions from the oxide into the electrolyte; iredox , electron transfer reactions, e.g. hydrogen or oxygen evolution; iC , capacitive charging. The anodic corrosion of a passive metal at steady state, i.e. at constant thickness such that iox = 0 is a straightforward process. In such a case, metal ions travel through the oxide film at a constant rate and are transferred in an ion transfer reaction (ITR) at the oxide/electrolyte interface. Thus i can approximate the dissolution rate of the passive film, icorr . Therefore, the higher dissolution rate of the passive film in the acid sulfate + chloride solution might be responsible for the reduced protection of the passive film (Fig. 13) and lower film resistance (Table 2) in comparison to the other two solutions. Since the stable oxides were present at low concentration while the hydroxides
Fig. 14. Current–time transients at passive potentials generated in three solutions for films on the 8% cold worked non-sensitized specimen.
1628
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629
Table 4 Atomic percent of main oxides in the passive films generated in the three solutions. The film generated potentials are 0.3 V. 8% cold worked unsensitized HNSS
ox CCr
ox CFe
ox CMn
0.5 m/L H2 SO4 + 0.5 m/L NaCl 3.5% NaCl 0.5 m/L NaOH + 0.5 m/L NaCl
0.65 0.28 0.20
0.24 0.59 0.58
0.11 0.13 0.22
The cold work-dependent corrosion resistance for the HNSS in the 3.5% NaCl and acid sulfate + chloride solutions was much different from that for the N-bearing stainless steel investigated previously. It has been reported that an increase in the cold work level of type 316LSS (0.07% N) from 5 to 25% decreases the pitting resistance in acidic chloride medium (pH < 1). While in neutral chloride medium, the corrosion resistance improved with up to 20% cold work before being degraded [32]. This was due to the heavy deformation resulting in enhanced dissolution in acidic chloride medium, where metal dissolution plays a dominant role in the rate-control mechanism. This was not the case in a neutral chloride solution, where oxygen diffusion was the rate-controlling step. The reason for the above difference is unknown and could be partially due to differences in the materials and solution systems investigated. The passive film formed in the alkaline chloride solution contained the largest amount of O. The Cr oxide content was lowest and Mn oxide content the highest among the films formed in all the three solutions (Table 4). The accumulation of chromium oxide on the surface of the passive film was more obvious for the HNSS in the acid sulfate + chloride solution than in the alkaline solution (Table 4). This is reasonable as chromium oxide dissolves readily in alkaline solution while iron oxide, unlike chromium oxide, dissolves readily in acid sulfate + chloride solution [35]. This helps to explain the higher film protection in 3.5% NaCl solution, since the dissolution rates of both iron oxide and chromium oxide are comparably lower. It has been reported that the corrosion resistance of a FeCr alloy was better than pure Cr in 0.1 M NaOH [35]. This is due to the iron oxide present in the outer layer of the passive film, which increases the overvoltage for the transpassive dissolution of Cr. Thus, in alkaline solution, iron oxide enables the passive film on the FeCr alloy to be stable over a wider potential range than on pure chromium. Therefore, iron oxide contributes to the stability of the passive film more in alkaline solution, although the effect of Cr is not negligible. The influence of cold work and sensitization treatment on corrosion resistance would be much smaller if iron oxide is used to facilitate the transpassive dissolution instead of another alloying element. This could partly explain the small change in the polarization curve in alkaline chloride solution (Figs. 5c and 6c). It is also possible that the alkaline condition inhibited pitting before transpassive dissolution. Thus, the specimens exhibited similar polarization behavior and no pitting took place even if the film protection was weak. More detailed studies are necessary to clarify this problem. Mo6+ and NH4+ were detected on the surface of passive films formed in all three chloride solutions (Fig. 13), in agreement with the results of previous studies [28,29]. 4. Conclusions The effects of cold work and sensitization treatment on the microstructure of a HNSS were studied by SEM and TEM. The corresponding changes in corrosion resistance in chloride solutions of different pH have been investigated using potentiodynamic and potentiostatic polarization. The nature of passive films has been analyzed in situ by EIS measurements and ex situ by XPS analysis. The following conclusions can be drawn:
1. Increasing the cold work level (up to 60%) steadily decreased the corrosion resistance of the non-sensitized HNSS in a 3.5% NaCl solution. The passive film of the heavily cold worked HNSS contained fewer oxides and more hydroxides. The less compact and protective film could be attributed to the high defect density in the steel base introduced by the cold work. 2. phases were observed to precipitate both at the grain boundaries and also within the grains during sensitization treatment. The reduction of anti-corrosion elements in the matrix and passive film resulted in the degradation of the corrosion resistance of sensitized HNSSs in both 3.5% NaCl and 0.5 M H2 SO4 + 0.5 M NaCl solutions. The degradation in corrosion resistance was more dramatic with increasing prior cold work level. 3. The reduction in corrosion resistance by cold work was not obvious in a 0.5 M H2 SO4 + 0.5 M NaCl solution for the non-sensitized HNSS. This might be related to the stability of the passive film in this solution. 4. Although the film resistance changed, neither cold work nor sensitization treatment affected the electrochemical polarization behavior of the HNSS in an alkaline chloride solution. All the HNSSs exhibited uniform corrosion in alkaline chloride solution. 5. No obvious changes in pit size and distribution were observed for the non-sensitized HNSSs with different cold work levels in a 3.5% NaCl solution. Sensitized HNSSs showed more pits on the corroded surfaces in comparison with non-sensitized ones at the same level of cold work. In the 0.5 M H2 SO4 + 0.5 M NaCl solution, corrosion occurred at the most vulnerable positions, namely, at the grain and twin boundaries for small cold worked specimens and along the deformation bands for the large cold worked specimens. Acknowledgments This study was jointly supported by the Science and Technology Foundation of China (50534010), and the Innovation Fund of Institute of Metal Research (IMR), Chinese Academy of Sciences (CAS). References [1] J.W. Simmons, Mater. Sci. Eng. A 207 (1996) 159. [2] H. Hanninen, J. Romu, R. Ilola, J. Tervo, A. Laitinen, J. Mater. Process. Technol. 117 (2001) 424. [3] T. Briglia, G. Terwagne, F. Bodart, C. Quaeyhaegens, J. D’Haen, L.M. Stals, Surf. Coat. Technol. 80 (1996) 105. [4] X.Q. Wu, S. Xu, J.B. Huang, E.H. Han, W. Ke, K. Yang, Z.H. Jiang, Mater. Corros. 59 (2008) 676. [5] X.Q. Wu, Y. Fu, J.B. Huang, E.H. Han, W. Ke, K. Yang, Z.H. Jiang, J. Mater. Eng. Perform., (2008) doi:10.1007/s11665-008-9295-4. [6] A. Toro, W.Z. Misiolek, A.P. Tschiptschin, Acta Mater. 51 (2003) 3363. [7] Y.J. Oh, J.H. Hong, J. Nucl. Mater. 278 (2000) 242. [8] L.R. Scharfstein, in: D. Peckner, I.M. Bernstein (Eds.), Handbook of Stainless Steels, McGraw Hill, 1977 (Ch. 15). [9] H. Ha, H. Kwon, Electrochim. Acta 52 (2007) 2175. [10] T.-H. Lee, C.-S. Oh, S.-J. Kim, S. Takaki, Acta Mater. 55 (2007) 3649. [11] L. Peguet, B. Malki, B. Baroux, Corros. Sci. 49 (2007) 1933. [12] A. Barbucci, G. Cerisola, P.L. Cabot, J. Electrochem. Soc. 149 (2002) 534. [13] Y. Fu, X.Q. Wu, E.H. Han, W. Ke, K. Yang, Z.H. Jiang, J. Electrochem. Soc. 155 (2008) C455. [14] N. Parvathavarthini, R.K. Dayal, S.K. Seshadri, J.B. Gnanamoorthy, J. Nucl. Mater. 168 (1989) 83. [15] S.K. Mannan, R.K. Dayal, M. Vijayalakshmi, N. Parvathavarthini, J. Nucl. Mater. 126 (1984) 1. [16] C.L. Briant, Effect of nitrogen and cold work on the sensitization of austenitic stainless steels, Electric Power Research Institute Report EPRI-NP-2457, 1982. [17] L.E. Murr, A. Advani, S. Shankar, D.G. Atteridge, Mater. Char. 24 (1990) 135. [18] L.K. Mansur, J. Nucl. Mater. 83 (1979) 109. [19] N. Thompson, Proc. Phys. Soc. Lond. B 66 (1953) 481. [20] S. Mahajan, Metallography 4 (1971) 43. [21] P. Liu, Phase Analysis in Steel using Analytical Transmission Electron Microscopy, Sandvik Materials Technology, Sandviken, 2004.
Y. Fu et al. / Electrochimica Acta 54 (2009) 1618–1629 [22] A. Carnot, I. Frateur, S. Zanna, B. Tribollet, I. Dubois-Brugger, P. Marcus, Corros. Sci. 45 (2003) 2513. [23] C. Hitz, A. Lasia, J. Electroanal. Chem. 500 (2001) 213. [24] Z. Grubac, M. Metikos-Hukovic, J. Electroanal. Chem. 565 (2004) 85. [25] C.T. Liu, J.K. Wu, Corros. Sci. 49 (2007) 2198. [26] I. Olefjord, B. Brox, U. Jelvestam, J. Electrochem. Soc. 132 (1985) 2854. [27] I. Olefjord, L. Wegrelius, Corros. Sci. 38 (1996) 1203. [28] Y.C. Lu, M.B. Ives, C.R. Clayton, Corros. Sci. 35 (1993) 89. [29] C.-O.A. Olsson, Corros. Sci. 37 (1995) 467.
1629
[30] C.D. Wagner, J.F. Moulder, L.E. Davis, W.M. Riggs, Handbook of X-ray Photoelectron Spectroscopy, Perkin-Elmer, Norwalk, Conn, USA, 1989. [31] C.R. Clayton, Y.C. Lu, J. Electrochem. Soc. 133 (1986) 2465. [32] U. Kamachi Mudali, P. Shankar, S. Ningshen, R.K. Dayal, H.S. Khatak, B. Raj, Corros. Sci. 44 (2002) 2183. [33] B. Heine, R. Kirchheim, Corros. Sci. 31 (1990) 533. [34] J.W. Schultze, M.M. Lohrengel, Electrochim. Acta 45 (2000) 2499. [35] P. Schmutz, D. Landolt, Corros. Sci. 41 (1999) 2143.