Pitting Corrosion Resistance of Ta-Bearing Duplex Stainless Steel

Abstract

The effect of Ta addition on the pitting corrosion resistance of duplex stainless steels was investigated in both cases without and with deoxidization/desulfurization by the addition of Al and Ca. The pitting corrosion resistance was improved by the addition of Ta with two proposed mechanisms. For steels without Al and Ca, the MnS inclusions which act as initiation sites of the pitting corrosion are modified to the electrochemically-stable (Ta,Mn) oxysulfides. For steels with Al and Ca, the pitting initiation sites (CaS and (Al,Ca)oxides) are coated with the stable Ta-containing nitrides resulting in the suppression of pitting corrosion propagation.

 

This Paper was Originally Published in Japanese in Zairyo-to-Kankyo 69 (2020) 237–245.

Fig. 11 Schematic illustration of pitting corrosion process of (a) steel without Ta, (b) Ta-bearing steel without Al and Ca, and (c) Ta-bearing steel with Al and Ca.

1. Introduction

Owing to their favorable combination of good mechanical properties and high corrosion resistances,^1–3) duplex stainless steels (DSS) have been applied in many different applications such as chemical plants,¹⁾ desalination plants,⁴⁾ and off-shore petroleum facilities.⁵⁾ For enhanced safety and prolonged service life of DSS components, it is important to suppress the pitting corrosion which is one of the most serious forms of corrosion in the stainless steels’ failure. Generally, the pitting corrosion resistance of DSS correlates linearly with pitting corrosion resistance equivalent value (PRE) or PREW⁶⁾ which is often defined as follows.   

\begin{equation*} \mathrm{PRE} = \%\text{Cr}+3.3\%\text{Mo}+16\%\text{N}\ (\%\ \text{by weight}) \end{equation*}

Thus, to improve the pitting corrosion resistance of stainless steels, there has been much investigations into the development of DSS with high PRE value such as super DSS¹⁾ and hyper DSS⁷⁾ whose PRE is higher than 40 and 48, respectively.

In addition to PRE, the pitting corrosion resistance of stainless steels is also strongly affected by the inclusions. For example, MnS inclusions are known to act as the pitting initiation sites in environments containing chloride ions.^8–18) It has been also reported that the non-metallic inclusions such as the other sulfides and oxides (e.g. MnCr₂O₄, (Mg,Al,Ca) oxides, and (Ti,Ca) oxides) have an influence on the pitting corrosion behavior.^19–23) Therefore, it is essential to elucidate the influences of MnS and other types of inclusions (e.g., oxide and sulfide) on the pitting corrosion for the development of DSS with higher pitting corrosion resistance. Several inclusion controls to enhance the pitting corrosion resistance of stainless steels have been proposed, such as elemental addition,^24–29) heat treatment,³⁰⁾ and the surface treatment.^31–33) For example, it has been reported that additions of Ti,²⁴⁾ Cu,²⁵⁾ and rare earth metals such as Ce^26,27) etc. are effective for the improvement of pitting corrosion resistance.

The authors have focused on the inclusion control by elemental addition and showed that the addition of Ta improves the pitting corrosion resistance and the crevice corrosion resistance of DSS model alloys.^34,35) As an example of studying the addition of Ta to corrosion-resistant alloys, Takizawa et al. showed that the Ta addition enhanced the crevice corrosion resistance of Ni-based alloy.³⁶⁾ However, there is little or no report on the effect of Ta addition on the formation of inclusions and the pitting corrosion resistance of DSS. For the practical stainless steels, both deoxidization and desulfurization were often conducted by the addition of deoxidizer and desulfurization agents such as Al and Ca.³⁷⁾ It is expected that the inclusions morphology and the corrosion resistance change depending on the type of deoxidizer and desulfurization agents, however, the pitting corrosion resistance of Ta-bearing DSS with deoxidization and desulfurization has not been investigated.

In this study, the effect of Ta addition on the pitting corrosion resistance of UNS S32750-type super DSS was investigated in both cases without and with deoxidization/desulfurization by the addition of Al and Ca. Additionally, in order to clarify the effect of Ta addition on the formation of sulfide inclusions, which can be the initiation site of localized corrosion, the pitting corrosion behaviors of DSS with varying S contents were investigated.

2. Experimental Procedure

2.1 Materials

Table 1 shows the chemical composition of the UNS S32750-equivalent super DSS. For steels B0, B1, and B2, no Ta was added. S content was changed to examine the effect of S on the inclusions formation and the pitting corrosion resistance. For steels T0, T1, and T2, Ta was added to investigate the effect of microalloying addition of Ta. In a similar way to steels without Ta, S content of steels with Ta was changed. The steels B0 and T0 were deoxidized and desulfurized by the addition of Al and Ca. The PRE values of steels as shown in Table 1 are almost the same. The steels were hot rolled and solution heat-treated at 1373 K. After the heat treatment, the specimens were quenched in water and cut into specimens parallel to the rolling direction.

Table 1 Chemical compositions of steels (mass%).

2.2 Pitting potential measurement

The pitting potential measurements were conducted by reference to procedures defined JIS G0577. Prior to the electrochemical measurements, the specimens were mechanically polished with SiC paper through 600 grid and cleaned ultrasonically with acetone. The surface area of the working electrode was determined to be 100 mm² using the resin to seal. The specimen was immersed for 600 s in 20 mass% NaCl solutions at 353 K under open air. The solution was prepared from ion-exchanged water and reagent grade chemicals: NaCl (Kanto Chemicals). After that, the potentio-dynamic polarization curve was measured by a potentiostat/galvanostat at a scan rate of 20 mV min⁻¹. The potential at which the current density exceeded 100 µA·cm⁻² was specified as the pitting potential, V′_C100. The pitting potential measurements were repeated at least three times. All potentials were measured vs. a saturated calomel electrode (SCE), and a platinum electrode was used as a counter electrode.

2.3 Pitting corrosion test

The pitting corrosion test was performed by reference to procedures defined in ASTM G48 method A and E. The specimens were mechanically polished with SiC paper through 600 grid and cleaned ultrasonically with acetone. The immersion test was carried out in a sealed test cell containing 600 mL of two test solutions (6 mass% FeCl₃ solution and 6 mass% FeCl₃ with 1 mass% HCl solution) at different temperatures for 24 h. The test solution was prepared from ion-exchanged water and reagent grade chemicals: FeCl₃·6H₂O (Kanto Chemicals) and HCl (Wako Pure Chemicals). After the test, the depth of pitting corrosion was evaluated by a digital optical microscope. The lowest temperature yielded pitting corrosion exceeding a depth 25 µm was measured as the critical pitting corrosion temperature (CPT).

2.4 Inclusion characterization

The morphology of the inclusions was observed by using a field-emission scanning electron microscope (FE-SEM). The compositions of the inclusions were analyzed by energy-dispersive X-ray spectroscopy (EDX). The cross-section images and the compositions of the inclusion were obtained by a transmission electron microscope (TEM) and EDX.

3. Results

3.1 Effect of Ta addition on pitting corrosion resistance

Figure 1 shows the results of the pitting potential measurements. In the anodic polarization curve of Ta non-additive steel B2 with relatively high S, a lot of current oscillations were observed at passive region (Fig. 1(c)). The current oscillations were due to the pitting initiation and the subsequent repassivation of the metastable pitting. A sharp increase in current density due to the initiation of stable pitting was observed at ca. 0.2 V. Thus, the pitting potential of steel B2 was determined to be 0.2 V. For steel T2 with Ta, the current oscillations due to the metastable pitting were suppressed and the transpassive region was observed from 0.5 V to 0.9 V despite the same S content of 0.002 mass%. Sharp increase in current density was observed at ca. 0.99 V. The pitting potential of steel T2 is 0.99 V which is nobler than that of steel B2. Comparing the polarization curves of Ta non-additive steels with different S content (B0, B1, and B2), it is seen that the current oscillations decrease and the pitting potential increases by decreasing S content from 0.002 mass% to 0.001 and <0.0005 mass%. By contrast, the polarization behaviors of Ta-added steels with S content of 0.002 mass% or less are almost the same regardless of S content; the transpassive region was observed and the current density increased above ca. 0.9 V. When the content of S was increased to 0.003 mass%, the current oscillations were observed at passivity zone even in steel with Ta, and the pitting potential decreased to 0.8 V.

Fig. 1

Anodic polarization curves of steels without and with Ta containing (a) <0.0005, (b) 0.001, (c) 0.002, and (d) 0.003 mass% S in 20 mass% NaCl aqueous solution at 80°C.

Figure 2 shows the relationship between the pitting potentials and S contents. For steels without Ta, the pitting potentials drastically declined with increasing of S content from 0.001 to 0.002 mass%. For steels with Ta, the pitting potential decreased when the S content increased from 0.002 to 0.003 mass%, but it was about the same value below 0.002 mass%. The pitting potentials of steels with Ta are higher compared to those of the steels without Ta at any S content. These results indicate that the increase of S content results in the decrease of pitting corrosion resistance and Ta addition enhances the pitting corrosion resistance.

Fig. 2

Pitting potentials of steels measured in 20 mass% NaCl aqueous solution at 80°C.

Figure 3 shows the results of the pitting corrosion tests of steels B0 and T0. For the pitting corrosion test using 6 mass% FeCl₃ solution (Fig. 3(a)), the corrosion rate of steel B0 without Ta increased at higher temperature and the pitting corrosion was observed at the temperature of 75°C and above. Thus, the CPT of steel B0 in 6 mass% FeCl₃ solution was determined to be 75°C. By the addition of Ta, the corrosion rate decreased and CPT increased from 75°C to 85°C. Improvement of pitting corrosion resistance by the addition of Ta was also observed for the pitting corrosion test using solution of 6 mass% FeCl₃ with 1 mass% HCl as shown in Fig. 3(b).

Fig. 3

Corrosion rate of steel B0 (<0.0005 mass% S, without Ta) and steel T0 (<0.0005 mass% S, with Ta) evaluated by the pitting corrosion test using (a) 6% FeCl₃ solution and (b) 6% FeCl₃ with 1% HCl solution at different temperatures for 24 h.

3.2 Effect of Ta addition on inclusions formation

Figure 4 shows the SEM images of typical inclusions observed in steels B2 (0.002 mass% S without Ta) and B0 (<0.0005 mass% S without Ta). Table 2 presents the relative compositions evaluated by EDX spot analysis at points 1 to 5 in Fig. 4. For steel B2 without Al and Ca, the sulfide containing Mn was observed adjacent to the (Cr,Mn) oxide as shown in Fig. 4(a). From EDX analysis of point 1 referring to (Cr,Mn) oxide, the atomic ratio of Cr, Mn, O was 2:1:6. Fe, Ni were derived from the information of the steel substrate. The EDX analysis at point 2 showed peaks of Mn and S in addition to Fe, Cr, Mo, and O, which included the information of the neighboring (Cr,Mn) oxide and the steel substrate (Fe, Cr, Mo). Considering that the molar ratio of (Cr,Mn) oxide (Cr:Mn:O = 2:1:6) and that the amount of O derived from (Cr,Mn) oxide is 35.3 mol%, the composition of Mn in the sulfide was estimated as 15 mol% by subtracting the contribution of Mn (5.9 mol%) due to the neighboring (Cr,Mn) oxide from the detected Mn (21.3 mol%). This is about the same value as S molar ratio, and the sulfide was identified to be MnS with a molar ratio of Mn:S = 1:1. In the steel B0 with deoxidization/desulfurization by the addition of Al and Ca, (Al,Ca) oxide and (Al,Mg) oxide were observed as shown in point 3 and 4. Although Mg was not intentionally added, it is considered to be derived from impurities from the furnace wall. The sulfide containing Ca was observed adjacent to the oxide (Point 5). In addition to Ca and S, the components derived from the matrix (Fe, Ni, Cr, Mo) and adjacent oxide (Al, Mg) were detected. Considering that the molar ratio of (Al,Ca) oxide (Al:Ca:O = 2:1:6), the composition of Ca in the sulfide was calculated to be about 19 mol% by subtracting the contribution of Ca due to (Al,Ca) oxide. The molar ratio of Ca is almost the same as that of S, indicating that the sulfide was CaS. Comparing inclusions in steels B2 and B0, (Cr,Mn) oxide were modified to (Al,Ca) oxide and (Al,Mg) oxide by the addition of Al and Ca. This is because Al, Ca, and Mg have higher deoxidizing ability than Cr and Mn. MnS was modified to CaS by the addition Ca which has high desulfurization ability.

Fig. 4

SEM images of inclusions observed in steels B2 (0.002 mass% S, without Ta) and B0 (<0.0005 mass% S, without Ta).

Table 2 Relative compositions of point 1 to 5 in Fig. 4 evaluated by EDX spot analysis (mol%).

Figure 5 shows the SEM and EDX mapping images of the typical inclusions observed in steels T2 (0.002 mass% S, with Ta) and T0 (<0.0005 mass% S, with Ta). For steel T2 without Al and Ca, the oxysulfide was formed around the oxide as shown in Fig. 5(a). Since the distribution of Mn and other composition was unclear in the EDX mapping image, the cross-sectional observation and point analysis of inclusions by TEM and EDX was conducted for detailed investigation as shown in Fig. 6 and Table 3. The oxide contains Ta, Mn, Cr, and the oxysulfide contains Ta and Mn. Thus, the core-shell complex inclusion was formed with the (Ta,Mn,Cr) oxide as the core and the (Ta, Mn) oxysulfide as the outer layer. For steel T2, more than 50 inclusions were analyzed, but none of the inclusions were identified as MnS which was observed in no-Ta-added steel B2 with similar S content. Thus, MnS was modified to (Ta,Mn) oxysulfides by the addition of Ta. For steel T0 with Al and Ca, Ta and N were observed around the (Al,Mg) oxide and the adjacent CaS as shown in Fig. 5(b). Comparing steel T0 and steel B0 without Ta (Fig. 4(b)), no modification of oxides and sulfides was observed and the Ta-containing nitride was formed by the addition of Ta. This is because Al and Ca, which have higher deoxidizing and desulfurizing abilities than Ta and Mn, preferentially formed oxides and sulfides. Additionally, Ta which did not form compounds with O and S partially reacted with N contained in the matrix at the interface between inclusion and matrix to form a core-shell complex inclusion.

Fig. 5

SEM images and EDX mapping images of the inclusion observed in steels T2 (0.002 mass% S, with Ta) and T0 (<0.0005 mass% S, with Ta).

Fig. 6

TEM cross-sectional image of the inclusion observed in steel T2 (0.002 mass% S, with Ta).

Table 3 Relative compositions of points 6 and 7 in Fig. 6 evaluated by EDX spot analysis (mol%).

4. Discussion

4.1 Role of Ta on the pitting corrosion resistance of steels without deoxidization/desulfurization by the addition of Al and Ca

In the previous report, we discussed the mechanism of the improved pitting corrosion resistance of DSS by the addition of Ta at 0.002 mass% S content.³⁵⁾ In this work, we discuss the role of Ta based on the pitting corrosion resistance that changes depending on the S content. To verify the mechanism for improvement of pitting corrosion resistance by the addition of Ta and reduction of S content, the initiation site of pitting corrosion was investigated. Generally, it is difficult to detect the initiation site by a mere surface observation of the specimens after the pitting corrosion tests because the pitting corrosion stably grows large and the information about initiation site disappears. Gao et al. reported that potentiostatic pulse technique (PPT) is useful for the determination of pitting initiation of the duplex stainless steel.³⁸⁾ By applying the square-wave potentiostatic pulse, the pitting corrosion initiates at an applied high potential and a subsequent low potential induces the repassivation of the pitting site, which suppress the spread of pitting corrosion. In this work, the pitting initiation was investigated by utilizing PPT in combination with the SEM observation of the same area before and after the corrosion test.

Figure 7 shows the SEM images of steels B2 and T2 before and after the PPT where the pulse sequence consisting of 0.50 V × 3 s and 0 V × 1 s was applied 500 times for the mirror-polished specimen in 20 mass% NaCl aqueous solution at 353 K. As shown in Fig. 7(a), MnS accompanied by (Cr,Mn) oxide before PPT in steel B2 dissolved and a pit initiation emerged after PPT. Muto et al. investigated the mechanism of pitting corrosion at MnS and reported that the stable pit grows to the pitting corrosion after dissolution of MnS and the formation of the trenches at the boundaries between MnS and steel matrix.^17,18) Similarly, in this research, it is estimated that MnS inclusions act as the initiation site of the pitting corrosion.

Fig. 7

SEM images of inclusions in steels (a) B2 (0.002 mass% S, without Ta) and (b) T2 (0.002 mass% S, with Ta) before and after PPT.

To clarify that MnS is the initiation site of pitting corrosion, the pitting potential measurement was conducted after PPT which is regarded as a surface treatment to remove MnS inclusions. Here, to eliminate the effect of passive film formed during PPT, the specimens was slightly polished after PPT, then, the pitting potential was measured. As shown in Fig. 8, the number of current spikes decreases and pitting potential becomes higher after PPF, indicating that the metastable and stable pitting derives from MnS dissolution. The decrease of the pitting potential with increase of S content can be explained by the formation of MnS which acts as the pitting initiation site. By contrast, as shown in Fig. 7(b), no appearance change was observed before and after PPT for the Ta-containing inclusion in steel T2. This indicates that Ta-containing inclusions composed of (Ta,Mn) oxysulfide and (Ta,Mn,Cr) oxide exhibit higher electrochemical stability than MnS. Thus, the improvement of the pitting corrosion resistance by the addition of Ta for steels without Al and Ca can be explained by the modification of MnS to the electrochemically-stable (Ta,Mn) oxysulfide.³⁵⁾ The pitting corrosion resistance decreased when the S content increased to 0.003 mass% even with Ta-bearing steel (Fig. 2). This is because the modification of MnS by Ta addition was insufficient and MnS inclusions were formed. From the SEM observation of steel T3 as shown in Fig. 9, the fine MnS was observed in addition to the core-shell complex inclusions composed of (Ta,Mn,Cr) oxide and (Ta,Mn) oxysulfide. Thus, Ta has the role of modifying MnS to (Ta,Mn) oxysulfide to improve the pitting corrosion resistance, which is especially effective when the S content is low.

Fig. 8

Anodic polarization curves of steel B2 (a) before and (b) after PPT measured in 20 mass% NaCl aqueous solution at 80°C.

Fig. 9

SEM images of inclusions observed in steel T3 (0.003 mass% S, with Ta).

4.2 Role of Ta on the pitting corrosion resistance of steels with deoxidization/desulfurization by the addition of Al and Ca

To identify the Ta-alloying effect on the pitting corrosion resistance of steels with deoxidization/desulfurization by the addition of Al and Ca, the initiation sites of the pitting corrosion in CPT test were examined. The SEM observation was performed for the mirror-polished specimens by 1 µm diamond paste before and after the CPT test using 6% FeCl₃ with 1% HCl solution at 70°C. Figure 10 shows SEM images of the steels B0 without Ta and T0 with Ta before and after the pitting corrosion test. Here, the corrosion test was conducted in a short time of 1 h to suppress the spread of pitting corrosion. As shown in Fig. 10(a), the dissolution of CaS and (Al,Ca) oxide was observed in steel B0 after the corrosion test. The pitting corrosion propagated from the site of the lost inclusion toward the matrix. It has been reported that CaS adversely affects the corrosion resistance³⁹⁾ and the pitting corrosion occurs due to the dissolution of (Al,Ca) oxide.²¹⁾ It is estimated that CaS and (Al,Ca) oxide act as the initiation site of pitting corrosion for steel B0. As shown in Figs. 10(b), CaS and a part of oxide inclusion in steel T0 dissolved away after CPT test. However, no noticeable change in appearance for Ta-containing nitride was observed, implying the high electrochemical stability of Ta-containing nitride. It seemed that the pit propagation from the dissolution of inclusion toward the matrix was suppressed by Ta-containing nitride shell.

Fig. 10

SEM images of inclusions in steels (a) B0 (<0.0005 mass% S, without Ta) and (b) T0 (<0.0005 mass% S, with Ta) before and after the corrosion test using 6% FeCl₃ with 1% HCl solution at 70°C for 1 h.

Based on the above results, the role of Ta on the pitting corrosion resistance is proposed as schematically drawn in Fig. 11. For the steels without Ta, the soluble inclusions such as MnS, CaS, and (Al,Ca) oxides dissolves, resulting in the exposure of the metal matrix, which allows the initiation and subsequent propagation of stable pitting corrosion (Fig. 11(a)). The mechanism for improving corrosion resistance by Ta addition changes depending on the presence or absence of Al and Ca. For the Ta-bearing steels without Al and Ca, MnS inclusions were modified to (Ta,Mn) oxysulfide with higher electrochemical stability (Fig. 11(b)). As a consequence, the initiation site of pitting corrosion was reduced, resulting in the improvement of pitting corrosion resistance. For the Ta-bearing steels with Al and Ca, the dissolution of CaS and oxides occurs; however, the Ta-containing nitride shell did not dissolve during the corrosion test. The exposure of the metal matrix followed by the pitting corrosion propagation is suppressed by the Ta-containing nitride shell as schematically drawn in Fig. 11(c). Namely, Ta-containing nitride shell acts as a barrier of the pitting corrosion propagation, which improves the pitting corrosion resistance.

Fig. 11

Schematic illustration of pitting corrosion process of (a) steel without Ta, (b) Ta-bearing steel without Al and Ca, and (c) Ta-bearing steel with Al and Ca.

5. Conclusion

In the present research, the effects of Ta addition on the pitting corrosion resistance were investigated for steels without and with deoxidization/desulfurization by the addition of Al and Ca. The following conclusions were drawn.

1. (1)    The pitting corrosion resistance improves by the addition of Ta in both stainless steels without and with deoxidization/desulfurization by the addition of Al and Ca. The mechanism for improving pitting corrosion resistance by adding Ta differs depending on the presence or absence of Al and Ca.
2. (2)    For steels without Al and Ca, MnS inclusions, which act as initiation sites of the pitting corrosion, are modified to the more electrochemically-stable (Ta,Mn) oxysulfides, resulting in the suppression of pitting corrosion initiation and the improvement of the pitting corrosion resistance.
3. (3)    For steels with Al and Ca, CaS and (Al,Ca) oxides, which act as initiation sites of the pitting corrosion, are partly coated with the more electrochemically-stable Ta-containing nitrides, resulting in the suppression of pitting corrosion propagation and the improvement of the pitting corrosion resistance.

Acknowledgments

The authors wish to thank Kobe Steel, Ltd. and Maruichi Stainless Tube Co., Ltd. for allowing publication of this study. The authors would like to express thanks to M. Moriichi, K. Mizohata, and G. Tanabe for their assistance in experimental works and helpful discussions.

REFERENCES

• 1)   J.-O.  Nilsson: Mater. Sci. Technol. 8 (1992) 685–700.
• 2)  J.-O. Nilsson and G. Chai: International Conference & Expo DUPLEX Associazione Italiana Di Metallurgia–AIM, Grado, Italy (2007).
• 3)   J.-O.  Nilsson and  A.  Wilson: Mater. Sci. Technol. 9 (1993) 545–554.
• 4)   J.  Olsson and  M.  Snis: Desalination 205 (2007) 104–113.
• 5)  A.C. Tomaselli, A. Valente and F.C. Camargo: OTC 22829 (2011) 1–10.
• 6)   K.  Ogawa,  H.  Okamoto,  M.  Ueda,  T.  Kobayashi and  T.  Mizuta: Sumitomo Kinzoku 46 (1994) 80–89.
• 7)  G. Chai, U. Kivisäkk, J. Tokaruk and J. Eidhagen: Stainless Steel World (2009) pp. 27–33.
• 8)   G.  Eklund: J. Electrochem. Soc. 121 (1974) 467–473.
• 9)   T.  Suter and  H.  Böhni: Electrochim. Acta 42 (1997) 3275–3280.
• 10)   E.G.  Webb,  T.  Suter and  R.C.  Alkire: J. Electrochem. Soc. 148 (2001) B186–B195.
• 11)   E.G.  Webb and  R.C.  Alkire: J. Electrochem. Soc. 149 (2002) B272–B279.
• 12)   S.H.  Jeon,  S.T.  Kim,  J.S.  Lee,  I.S.  Lee and  Y.S.  Park: Mater. Trans. 53 (2012) 1617–1626.
• 13)   H.  Krawiec,  V.  Vignal,  O.  Heintz,  R.  Oltra and  J.M.  Olive: J. Electrochem. Soc. 152 (2005) B213–B219.
• 14)   M.P.  Ryan,  D.E.  Williams,  R.J.  Chater,  B.M.  Hutton and  D.S.  McPhail: Nature 415 (2002) 770–774.
• 15)   Q.  Meng,  G.S.  Frankel,  H.O.  Colijn and  S.H.  Goss: Nature 424 (2003) 389–390.
• 16)   J.O.  Park,  S.  Matsch and  H.  Böhni: J. Electrochem. Soc. 149 (2002) B34–B39.
• 17)   I.  Muto,  Y.  Izumiyama and  N.  Hara: J. Electrochem. Soc. 154 (2007) C439–C444.
• 18)   A.  Chiba,  I.  Muto,  Y.  Sugawara and  N.  Hara: J. Electrochem. Soc. 160 (2013) C511–C520.
• 19)   D.E.  Williams,  M.R.  Kilburn,  J.  Cliff and  G.I.N.  Waterhouse: Corros. Sci. 52 (2010) 3702–3716.
• 20)   S.J.  Zheng,  Y.J.  Wang,  B.  Zhang,  Y.L.  Zhu,  C.  Liu,  P.  Hu and  X.L.  Ma: Acta Mater. 58 (2010) 5070–5085.
• 21)   S.  Zheng,  C.  Li,  Y.  Qi,  L.  Chen and  C.  Chen: Corros. Sci. 67 (2013) 20–31.
• 22)   H.Y.  Ha,  C.J.  Park and  H.S.  Kwon: Corros. Sci. 49 (2007) 1266–1275.
• 23)   M.A.  Baker and  J.E.  Castle: Corros. Sci. 33 (1992) 1295–1312.
• 24)   N.  Shimahashi,  I.  Muto,  Y.  Sugawara and  N.  Hara: J. Electrochem. Soc. 160 (2013) C262–C269.
• 25)   R.S.  Lillard,  M.A.  Kashfipour and  W.  Niu: J. Electrochem. Soc. 163 (2016) C440–C451.
• 26)   M.  Nishimoto,  I.  Muto,  Y.  Sugawara and  N.  Hara: J. Electrochem. Soc. 164 (2017) C901–C910.
• 27)   S.H.  Jeon,  S.T.  Kim,  M.S.  Choi,  J.S.  Kim,  K.T.  Kim and  Y.S.  Park: Corros. Sci. 75 (2013) 367–375.
• 28)   X.  Liu and  Y.  Bo: Adv. Mater. Res. 718 (2013) 29–32.
• 29)   H.Y.  Ha,  C.  Park and  H.  Kwon: Scr. Mater. 55 (2006) 991–994.
• 30)   M.  Tohjoh,  I.  Muto,  A.  Chiba,  Y.  Sugawara and  N.  Hara: ECS Trans. 58 (2014) 39–45.
• 31)   Y.T.  Zhou,  B.  Zhang,  S.J.  Zheng,  J.  Wang,  X.Y.  San and  X.L.  Ma: Sci. Rep. 4 (2014) 3604.
• 32)   N.  Hara,  Y.  Sugawara and  I.  Muto: Int. J. Corros. 2012 (2012) 1–6.
• 33)   A.  Chiba,  I.  Muto,  Y.  Sugawara and  N.  Hara: Corros. Sci. 106 (2016) 25–34.
• 34)   H.  Sakai,  K.  Sugahara and  H.  Saito: Zairyo-to-Kankyo 59 (2010) 222–227.
• 35)   M.  Kawamori,  J.  Kinugasa,  Y.  Yonenaga,  Y.  Fukuta,  M.  Shimamoto,  T.  Sugimura,  T.  Sato,  N.  Nishizawa and  M.  Nagao: Zairyo-to-Kankyo 64 (2015) 320–323.
• 36)   M.  Kawamori,  J.  Kinugasa,  Y.  Yonenaga,  Y.  Fukuta,  M.  Shimamoto,  T.  Sugimura,  T.  Sato,  N.  Nishizawa and  M.  Nagao: Zairyo-to-Kankyo 64 (2015) 497–500.
• 37)   K.  Mineura,  I.  Takahashi and  K.  Tanaka: ISIJ Int. 30 (1990) 192–198.
• 38)   J.  Gao,  Y.  Jiang,  B.  Deng,  Z.  Ge and  J.  Li: Electrochim. Acta 55 (2010) 4837–4844.
• 39)   I.I.  Reformatskaya and  L.I.  Freiman: Prot. Met. 37 (2001) 459–464.