Oxide properties and stress corrosion cracking behaviour for Alloy 600 in leaded caustic solutions at high temperature

Oxide properties and stress corrosion cracking behaviour for Alloy 600 in leaded caustic solutions at high temperature

Corrosion Science 53 (2011) 1247–1253 Contents lists available at ScienceDirect Corrosion Science journal homepage: www.elsevier.com/locate/corsci ...

571KB Sizes 0 Downloads 51 Views

Corrosion Science 53 (2011) 1247–1253

Contents lists available at ScienceDirect

Corrosion Science journal homepage: www.elsevier.com/locate/corsci

Oxide properties and stress corrosion cracking behaviour for Alloy 600 in leaded caustic solutions at high temperature Dong-Jin Kim ⇑, Hyuk Chul Kwon, Hyun Wook Kim, Seong Sik Hwang, Hong Pyo Kim Nuclear Materials Research Division, Korea Atomic Energy Research Institute, 1045 Daedeok-daero, Yuseong-gu, Daejeon, Republic of Korea

a r t i c l e

i n f o

Article history: Received 20 July 2010 Accepted 15 December 2010 Available online 22 December 2010 Keywords: A. Alloy B. EIS C. Passivity C. Stress corrosion

a b s t r a c t The stress corrosion cracking behaviour of Alloy 600 in caustic solutions with and without PbO at 315 °C was investigated by means of slow strain rate tension tests. The characterisation of the oxide that formed on Alloy 600 was derived from transmission electron microscopy, energy dispersive X-ray spectroscopy, X-ray photoelectron spectroscopy and electrochemical impedance spectroscopy. Lead was incorporated into the oxide in a metallic lead state and a lead oxide state, which degraded the passivity and induced PbSCC susceptibility. NiB was used as an inhibitor. It reduced the lead incorporation level in the oxide layer and decreased PbSCC susceptibility. Ó 2010 Elsevier Ltd. All rights reserved.

Introduction Nuclear power plants (NPPs) that use Alloy 600 as a heat exchanger tube in a steam generator (SG) have experienced a variety of corrosion problems such as pitting, intergranular attacks (IGA) and stress corrosion cracking (SCC). In spite of a considerable effort to reduce the material degradation, SCC remains an important problem to be overcome. Secondary water pH levels that reveal a shift from acidic to alkaline states can affect SCC behaviour to spread extensively in crevices. The pH depends on the water chemistry control, the water chemistry in crevices, plant-specific condition and so on. Specific chemical species accumulate in crevices leading to a specific condition of crevice chemistry. One of these chemical species is lead. It is one of the most deleterious species in reactor coolants that cause SCC in alloys [1–4]. Even Alloy 690 which is an alternative to Alloy 600 because of its outstanding robustness against SCC is also susceptible to lead in alkaline solutions [5–7]. Lead can be effectively detected in tubesheet samples, crevice deposits and surface scales removed from steam generators. Typical concentrations are 100–500 ppm, but some plants can detect concentrations as high as 2000–10,000 ppm [8]. The best way to prevent lead-induced SCC (PbSCC) is to eliminate the harmful lead from the secondary environment. However, this task is no feasible and most NPPs are already contaminated by lead. Moreover only a very low level of lead (in the sub parts per million ranges) can ⇑ Corresponding author. Tel.: +82 42 868 8387. E-mail address: [email protected] (D.-J. Kim). 0010-938X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.corsci.2010.12.016

affect PbSCC. Thus, a mechanism of PbSCC needs to be understood in order to control and mitigate PbSCC. Oxides on steel and Alloy 600 surfaces in an aqueous solution above 100 °C have a duplex film structure [9–13]. The oxide’s inner layer which is formed by the growth of an oxide layer on a metal surface is denser than its outer layer. The oxide’s outer layer which is formed by a dissolution and precipitation mechanism is less adhesive than its inner layer. The growth processes of the inner and outer layers occur at the metal/oxide and oxide/electrolyte interfaces, respectively. The growth rates are controlled by a transport of the layer forming species through the layer, that is, by an inward diffusion of oxygen including electrolyte species and an outward diffusion of metal cations. Cracks can initiate and propagate through unavoidable breakdowns and alterations of the surface oxide that forms naturally on Alloy 600 in an aqueous solution. Characterisation of the surface oxide properties would therefore be helpful for elucidating PbSCC mechanism. Hwang, et al. [14] investigated the oxide modification by Pb at 250 °C as a function of the pH value of a room temperature solution. Other reports have focused on the composition of oxide that forms in leaded acid and caustic solutions [15,16]. However the way Pb affects the oxide structure and SCC is unclear. The addition of lead to a solution is presumed to modify the oxide property and hence induce SCC susceptibility. However the PbSCC phenomenon could be reduced if this type of modification can be avoided through the discovery of a suitable inhibitor. The present work investigates SCC susceptibility in caustic solutions supplemented with lead as an accelerator and NiB as

1248

D.-J. Kim et al. / Corrosion Science 53 (2011) 1247–1253

Table 1 Chemical compositions of the Alloy 600 TT and Alloy 600 HTMA. Material

C

Si

Mn

P

Cr

Ni

Fe

Co

Ti

Cu

Al

B

S

N

Ce

Alloy 600 TT Alloy 600 HTMA

0.03 0.025

0.07 0.05

0.003 0.22

0.024 0.07

16.12 15.67

74.42 75.21

9.06 8.24

0.007 0.005

0.005 0.39

0.003 0.011

0.006 0.0014

0.001 0.001

0.0103 0.0103

0.003

0.15

Fig. 1. Design of the test specimen fabricated for SSRT test.

an inhibitor using a slow strain rate tension (SSRT) test. The oxides that form on Alloy 600 in aqueous solutions with and without lead were examined by means of a transmission electron microscopy (TEM), an energy dispersive X-ray spectroscopy (EDS), an X-ray photoelectron spectroscopy (XPS) and an electrochemical impedance spectroscopy (EIS). 2. Experimental The test specimens were fabricated from a 19.05 mm outside diameter Alloy 600 steam generator tubing, which was solution annealed at 975 °C for 20 min, followed by thermal treatment (TT) at 704 °C for 15 h or high temperature mill annealed (HTMA) at 1024 °C for 3 min. Table 1 lists the chemical composition (in weight percentage). High-purity water (18 M X cm at room temperature) was used as the reference solution. Table 2 lists the aqueous solutions used in this work. Reagent grade PbO was added as a lead source to a solution in amounts of 5000 ppm or 10,000 ppm. The performance of an NiB inhibitor was evaluated by adding

4 g/L of NiB to the leaded solution. Deaeration was conducted by purging with a high purity nitrogen gas to remove the dissolved oxygen for 20 h before the tests. As shown in Fig. 1, the SSRT tests were performed on uniaxial tension specimens fabricated from HTMA tubing in unleaded and leaded solutions, and in a leaded solution supplemented with NiB. The tests were carried out in 1.89 L nickel autoclaves at 315 °C and an equilibrium pressure. The test specimens were at an open circuit potential (OCP) without an impressed electrochemical current. The strain rate was 2  107 s1. After the SSRT tests, the surfaces were observed by means of a SEM (JSM6360) for the purpose of determining the SCC ratio. The electrochemical tests were performed on rectangular plate specimens (10 mm  10 mm) fabricated from the thermally treated tubing. The surface of the specimens was polished up to 1 lm using a diamond suspension. An Alloy 600 wire was spot welded to the specimen, and the wire was shielded with heat-shrinkable polytetrafluoroethylene (PTFE) tubing. The test specimens were immersed in a 3.78 L nickel autoclave at 315 °C for 14 days. The electrochemical impedance measurements were taken in a frequency range of 106– 103 Hz at the OCP with a perturbation level of 10 mV. A Solartron 1255 frequency response analyser was used with a Solartron 1287 electrochemical interface. Table 2 shows the experimental matrix. Separate autoclaves were used for the leaded and unleaded test solutions to avoid any cross contamination. After the immersion test, the plate specimens were examined. The surface oxide layer and its composition were examined by means of a field emission TEM, equipped with an EDS (JEM2100F, JEOL). The information on the chemical binding was obtained from an XPS (AXIS-NOVA, KRATOS Analytical). The spectra for Ni 2p, Cr 2p, O 1s and Pb 4f were recorded with AlKa radiation (hm = 1486.6 eV), at a pass energy of 20 eV. The take-off angle, the base pressure and the sputter rate for a depth profiling were 45°, 6.7  107 Pa and 0.04 nm/s in SiO2, respectively. An online database was used for analysis of the XPS results [18].

Table 2 Various aqueous solutions and their pHs for SSRT and immersion (electrochemical impedance) tests at 315 °C [17]. Environment

pH (315 °C) by MULTEQ

Remark

Experiment

H20 10,000 ppm PbO 10,000 ppm PbO + 4g/L NiB

5.8 7.9 7.9

Neutral pH increase Deaeration

SSRT test

0.01 M Na2S04 + 0.01 M NaHS04 10,000 ppm PbO 10,000 ppm PbO + 4g/L NiB

5.6 8.7 8.7

Acid pH increase Deaeration

SSRT test

0.01 M Na2S04 10,000 ppm PbO 10,000 ppm PbO + 4g/L NiB

5.8 8.6 8.6

Neutral pH increase

SSRT test

Ammonia 5,000 ppm PbO 5,000 ppm PbO + 4g/L NiB

6.3 7.9 7.9

Mild caustic pH increase Deaeration

Immersion test

0.1 M NaOH (Deaeration) 5,000/10,000 ppm PbO (Deaeration) 5,000/10,000 ppm PbO (Non-deaeration)

9.9 9.9 9.9

Caustic The same pH

SSRT, immersion test

10 wt% NaOH 10,000 ppm PbO

10.4 10.4

Strong caustic Deaeration

SSRT test

40 wt% NaOH 10,000 ppm PbO

10.9 10.9

Strong caustic Deaeration

SSRT test

1249

D.-J. Kim et al. / Corrosion Science 53 (2011) 1247–1253 Table 3 Elongation to rupture and SCC ratio obtained from SSRT test in various aqueous solutions at 315 °C. Environment

Elongation to rupture (%)

SCC ratio

pH (315 °C) by MULTEQ

Remark

H2O 10,000 ppm PbO addition 0.01 M Na2S04 + 0.01 M NaHS04 10,000 ppm PbO addition 0.01 M Na2S04 10,000 ppm PbO addition 0.1 M NaOH (Deaeration) 10,000 ppm PbO addition (Deaeration) 10,000 ppm PbO addition (Non-deaeration) 10 wt% NaOH (Deaeration) 10,000 ppm PbO addition (Deaeration) 40 wt% NaOH (Deaeration) 10,000 ppm PbO addition (Deaeration)

56 30 64 40 49 26 57 24 35 55 38 61 63

Little 83 16 57 Little 81 Little 78 48 17 31 Little 5

5.8 7.9 5.5 8.7 7.5 8.6 9.9 9.9 9.9 10.4 10.4 10.9 10.9

Neutral pH increase Acid pH increase Mild caustic pH increase Caustic Susceptible to PbSCC Strong caustic Susceptible to PbSCC Strong caustic Relatively less susceptible

3. Results and discussion As shown in Table 2, the pH at 315 °C increases when PbO is added. The increase seems to be caused by the PbO + H2O = Pb2+ + 2OH reaction [19]. Table 3 shows the elongation to rupture and the SCC ratio as a function of the aqueous solution with and without PbO for the HTMA Alloy 600. The SCC ratio is defined as the SCC area over the cross-sectional area of the specimen, which is obtained from observation of the fracture surface. The elongation to rupture was determined from the stress–strain curve. Both the elongation to rupture and the SCC ratio can be used as criteria for SCC susceptibility. Their usefulness as criteria is based on the fact that the yield strength and tensile strength decrease with the SCC leading to the lower elongation to rupture. The SCC susceptibility was greatly increased in H2O, 0.01 M Na2SO4 + 0.01 M NaHSO4, 0.01 M Na2SO4 and 0.1 M NaOH when PbO was added to the solutions. The SCC ratio was 78% in a deaerated 0.1 M NaOH solution with PbO but negligible in a 0.1 M NaOH solution without PbO. The PbO increased the SCC susceptibility to an SCC ratio of 31% in a highly caustic 10 wt.% NaOH solution (2.5 M NaOH). In this solution without PbO, the sample of Alloy 600 had an SCC ratio of 17%. However, the effect of PbO on SCC in a 40 wt.% NaOH (10 M NaOH) solution was much less than the corresponding effect in a 0.1 M NaOH solution. The SCC ratio decreased with the pH in a highly caustic solution. A neutral pH at 315 °C is 5.8. Accordingly, the HTMA Alloy 600 is more susceptible to a PbSCC in a mild caustic solution such as a 0.1 M NaOH solution rather than a strong caustic solution such as 2.5 M and 10 M NaOH solutions. The dependence of PbSCC susceptibility on the pH value may be closely related to stability of the oxide composed of Ni, Cr, Fe and Pb and to oxide solubility in a highly caustic solution. This dependence will be investigated more closely in the future. A series of TEM, TEM-EDS, EIS, XPS and thermodynamic analyses was conducted to investigate the effect of PbO on the SCC behaviour in a mild alkaline solution. Fig. 2 shows TEM micrographs of a surface oxide layer that formed on TT Alloy 600 specimens in (a) the unleaded 0.1 M NaOH and (b) the leaded 0.1 M NaOH solutions at 315 °C. A porous outer oxide layer and an inner oxide layer were observed in the unleaded solution, whereas only an oxide layer was observed in the leaded solution. Fig. 3 shows the results of the TEM-EDS analyses for the specimens tested in the unleaded 0.1 M NaOH solution (Fig. 3a) and in the leaded solution (Fig. 3b). Ni-rich oxide in the outer layer and a relatively Cr-rich oxide in the inner layer were formed in the un-

Fig. 2. TEM micrographs of the surface oxide layer formed on the TT Alloy 600 specimens in aqueous solutions at 315 °C: (a) 0.1 M NaOH and (b) 0.1 M NaOH + PbO.

leaded 0.1 M NaOH solution. The results of Figs. 2 and 3 for the unleaded 0.1 M NaOH solution show that a duplex oxide layer was formed at the surface; that is, a porous Ni-rich outer layer and a

1250

D.-J. Kim et al. / Corrosion Science 53 (2011) 1247–1253

(a) 110

(b)

100

100

Oxide

90

90

oxide

Composition / wt%

70

Oxygen cation Cr cation Fe cation Ni

60 50 40 30

70

50 40 30 20

10

10

0

0

-10

-10 0.1

0.2

0.3

0.4

Oxygen cation Cr cation Fe cation Ni cation Pb

60

20

0.0

Alloy 600

80

80

Composition / wt%

110

0.0

0.5

0.2

0.4

0.6

0.8

1.0

Distance / μm

Distance / μm

Fig. 3. TEM-EDS analyses of the surface oxide layer formed on the TT Alloy 600 specimens in aqueous solutions at 315 °C: (a) 0.1 M NaOH and (b) 0.1 M NaOH + PbO.

dense Cr-rich inner layer. These results are similar to the experimental results obtained in the unleaded ammonia solution [20]. Metallic ions in a solution are reportedly re-deposited to form a porous outer oxide layer [9]. In the leaded solution, a large amount of lead was observed at about 25 wt.% in the oxide. The Ni was depleted in the oxide layer. The duplex oxide layer observed in the unleaded solution was not observed in the leaded solution. The EIS results of the specimens immersed in the aqueous solutions with and without PbO must be analysed because the oxide passivity is closely related to the SCC behaviour. Fig. 4 presents Nyquist plots of the TT Alloy 600 specimens in (a) the unleaded 0.1 M NaOH and (b) the leaded 0.1 M NaOH solutions at 315 °C. The impedance spectra of the unleaded solution imply that the equivalent circuit is composed of a series of a solution resistance and a parallel of oxide capacitance and the oxide

(a) 5000

resistance, connected to a constant phase element related to a diffusion process of an electrolyte or a metallic cation. However, when lead oxide is added to the solution, the total impedance was considerably decreased indicating that the oxide passivity was greatly decreased. An inductive loop was observed, which may be related to lead incorporation into the oxide. The inductive loop may be attributed to the conduction path by the incorporated lead. The SSRT and EIS results suggest that better passivity improves the SCC resistance. In general, the repassivation of a more passive oxide is faster than that of a less passive oxide in the passive range. This tendency induces SCC resistance. Besides the passivity and repassivation kinetics, oxide ductility is reportedly degraded when lead is incorporated into the oxide [21]. Thus, lead aggravates the oxide passivity as well as the oxide ductility and consequently increases PbSCC susceptibility.

(b) 4.0 0.1 M NaOH

300

3.5

250

4000

200

3.0

-Imaginary Impedance / Ω cm

2

-Imaginary Impedance / Ω cm

2

150 100

3000

0.1 M NaOH + PbO

50 0 -50 -50

0

50

100

150

200

250

300

2000

1000

2.5 2.0 1.5 1.0 0.5 0.0 -0.5

0

0

1000

2000

3000

4000 2

Real Impedance / Ω cm

5000

-1.0 2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

7.0

2

Real Impedance / Ω cm

Fig. 4. Nyquist plots obtained from the electrochemical impedance measurements for the TT Alloy 600 immersed in the 0.1 M NaOH solutions at 315 °C: (a) unleaded 0.1 M NaOH and (b) leaded 0.1 M NaOH.

1251

D.-J. Kim et al. / Corrosion Science 53 (2011) 1247–1253

Pb for Alloy 600

7000

of the Ni species as a metallic state and an oxide state as a function of the sputtering time. The depletion of nickel as an oxide state and a metallic state was observed in the oxide that formed in the leaded ammonia solution. This observation is consistent with the Ni depletion in the oxide shown in Fig. 3. Consideration was given to the following eight reactions for the experiment in high temperature aqueous solution.

10sec sputtering

Intensity / CPS

6000 5000

PbO

4000

NiO þ Pb

3000

Pb

¼ PbO þ Ni 2þ

2000

Cr2 O3 þ 3Pb

1000

Fe3 O4 þ Pb





0 141

140

139

138

137

136

135

Binding Energy / eV

22000

NH4OH, PbO

18000

¼ Fe2þ þ PbO þ Fe2 O3





þ V2 Ni

ð9Þ



¼ PbNi

ð10Þ

NiNi ¼ Ni V2 Ni þ Pb

The following summation of reactions (9) and (10) shows that the Ni at the Ni site in the oxide dissolves to leave a Ni vacancy and that the lead ion is oxidised by a reaction with the Ni vacancy in the oxide. 2þ

NiNi þ Pb



¼ Ni

Ni(Oxide) Ni(Metal)

0 -2000 2000

3000

4000

5000

6000

Sputtering Time / s Fig. 6. Plot of XPS intensity of nickel as a metallic state and metal as an oxide state against sputtering time for the TT Alloy 600 specimen immersed in leaded ammonia solution.

þ PbNi

ð11Þ

As shown in Fig. 6, reaction (11) confirms that Ni as a Ni oxide state can be depleted. A strain field in the oxide which appears to originate from a lattice mismatch as a result of the lead that is incorporated into the oxide degrades the level of passivity. According to the high temperature E-pH diagram, pure cations such as Ni2+, Cr3+ and Fe3+ are unstable in an alkaline solution. Cations consequently react with water to produce aqueous forms such as HNiO2, CrO2 and HFeO2 producing H+. Reactions (1)–(3) can therefore be rewritten as follows in an alkaline solution. 2þ



þ 2H2 O ¼ PbO þ HNiO2 þ 3Hþ



4000

ð8Þ

Based on the thermodynamic data and solubility data [17,22], a forward reaction is possible for reaction (1) at 315 °C assuming that concentrations of nickel and lead ions are 106 M and 4.5  103 M, respectively. Defect chemistry can be used to rewrite this reaction as follows.

Fe3 O4 þ Pb

6000

ð7Þ

Pb þ H2 O ¼ PbO þ H2



8000

ð6Þ

þ H2 O ¼ PbO þ 2Hþ

Cr2 O3 þ 3Pb

10000

ð5Þ



Pb

ð3Þ ð4Þ

Fe2þ þ Pb ¼ Fe þ Pb

12000

1000

ð2Þ



NiO þ Pb

14000

0

¼ 3PbO þ 2Cr3þ

þ Pb ¼ Ni þ Pb

16000

2000

ð1Þ



Fig. 5 shows the XPS results for the specimen tested in the leaded ammonia solution [20]. Lead was incorporated into the oxide layer in a metallic state (136.7 eV) and as a lead oxide (138.9 eV) [18]. A comparison of the results in the ammonia solution and the 0.1 M NaOH solution shows that the oxide microstructure and its composition in the 0.1 M NaOH solution are similar to those of the ammonia solution [20]. Thus, from the experimental results, the oxide structure and its composition appear to depend mainly on the pH(T) value at high temperature of the mild caustic solution. According to the E-pH diagram, the pH(T) values for the ammonia solution and the 0.1 M NaOH solution are in a pH range where the stable phases are the same [22]. From this, the XPS results appear to be similar for different solutions. The thermodynamic analysis shows that the equilibrium electrochemical potential of Pb2+/Pb is higher than the equilibrium potential of Ni2+/Ni leading to increase of OCP of Alloy 600 in the leaded solution [19]. Hence, Pb can be deposited electrochemically in the leaded solution at the OCP. This outcome is consistent with the XPS detection of Pb metal. The origin of PbO detection is discussed later. Fig. 6 plots the XPS intensity of Ni as a metallic state and Ni as an oxide state against the sputtering time of a specimen immersed in the leaded ammonia solution. Fig. 6 is based on the integrated value of the XPS signal acquired around the binding energy level

20000

Ni



2Cr3þ þ 3Pb ¼ 2Cr þ 3Pb

Fig. 5. X-ray photoelectron spectrum of the surface oxide layer formed for the TT Alloy 600 in the leaded ammonia solution [20].

Intensity / A.U.



þ 4H2 O ¼ 3PbO þ 2CrO2 þ 6Hþ

þ 2H2 O ¼ PbO þ Fe2 O3 þ HFeO2 þ 3Hþ

ð12Þ ð13Þ ð14Þ

The thermodynamic data and solubility data [23,24] suggest that a forward reaction is possible for reaction (12) in a mild caustic solution. Unlike reaction (12), reaction (13) has no possibility of a forward reaction. That means there is no possibility of lead ions being substituted for the Cr site in the oxide. However, a forward reaction is possible for reaction (14). The direction of the reaction depends on the solubility, the phase stability and the pH(T) level. As expected from the potential difference in the equilibrium oxidation and reduction reactions, backward reactions (4)–(6) which describe the lead electrodeposition are available. Cations

D.-J. Kim et al. / Corrosion Science 53 (2011) 1247–1253

in a metallic state such as Ni, Cr and Fe can therefore be depleted in the leaded aqueous solution by lead electrodeposition. As mentioned above, pure cations react with water to form stable aqueous forms in an alkaline solution. According to thermodynamics of reactions (7) and (8), lead ions in an aqueous solution and electrodeposited lead are oxidised when the pH level at 315 °C is larger than 4.46 and the hydrogen partial pressure is as low as about 5 ppb in the secondary water of an NPP. Thus, as shown in Fig. 5, the PbO can be detected in the XPS analysis because of three factors: the lead incorporated into the oxide; the oxidation of the electrodeposited lead; and the PbO that is added to the solution. Table 3 shows that the SCC ratio which expresses the degree of PbSCC susceptibility for the HTMA Alloy 600 was larger to be 78% in the deaerated solution, compared to 48% in the non-deaerated 0.1 M NaOH solution. The Pb electrodeposition does not occur in the non-deaerated solution but occurs spontaneously in the deaerated solution. In deaerated solution, PbSCC is accelerated by the Pb electrodeposition and Pb ion incorporation into the oxide. In contrast, PbSCC is accelerated by only a Pb ion incorporation into the oxide in the non-deaerated solution. The fact that the passive oxide forms more easily in a non-deaerated condition than in a deaerated condition may feasibly reduce the PbSCC susceptibility. Nevertheless, it should be noted that both the Pb electrodeposition and the incorporation of Pb ions in the oxide induce the PbSCC. The EIS results showed that there is much less passivity in the oxide that forms in the leaded solution than in the oxide that forms in the unleaded solution as shown in Fig. 4. Thus, the incorporation of lead by electrodeposition and the incorporation of Pb ions in the oxide induce a single layer that is less passive, which in turn leads to PbSCC. This behaviour occurs instead of the formation of a passive oxide; that is, a porous outer layer of oxide and a dense inner layer of oxide, which is composed of Ni, Cr, Fe and O, are generally formed in an aqueous solution without Pb. Lead ions can reportedly be incorporated into an oxide producing dissolution of metal composed of Alloy 600 [25]. This can introduce a lattice mismatch which in turn degrades the passivity of the oxide. When incorporated into a film as an oxide, lead con also degrade passivity. This phenomenon was observed with a current transient technique under experimental conditions with and without Pb electrodeposition. The depletion of Ni which is a major metallic element of Alloy 600 was mainly observed as shown in Fig. 3. This depletion explains why the reaction kinetics should be combined with the thermodynamics. As shown in the high temperature E-pH diagram

[22], lead contaminants can generally be reduced and deposited on an oxide surface in a metallic form or be incorporated into the passive layer as lead ions or lead oxide. This behaviour depends on the OCP which is affected by the presence of lead contamination in the secondary side cooling water of an operating NPP. Fig. 7 presents the elongation to rupture as a function of the high temperature pH value of leaded solutions with and without NiB for the HTMA Alloy 600. The addition of NiB to the solution obviously enhanced the elongation to rupture as well as the failure strength (not shown here). In the leaded solution with the NiB inhibitor, the amount of lead in the oxide layer is significantly reduced and the Ni depletion in the oxide is decreased. These results are highlighted by a comparison of Fig. 8. and Fig. 3. The impedance value obtained in the leaded solution with NiB (Fig. 9) is significantly greater than that obtained in the absence of NiB (Fig. 4). The impedance spectrum shows a clear capacitive arc but there is no inductive loop that appears in the leaded 110 100

Oxide Alloyy 600

90 80

Composition / wt%

1252

70

Oxygen cation Cr cation Fe cation Ni cation Pb

60 50 40 30 20 10 0 -10

0.0

0.1

0.2

0.3

0.4

0.5

0.6

Distance / μm Fig. 8. TEM-EDS analysis of the surface oxide layer formed on the TT Alloy 600 specimen in leaded 0.1 M NaOH solution with NiB at 315 °C.

70

PbO+NiB

60

80

-Imaginary Impedance / Ω cm

2

50

Elongation to Rupture / %

70 60

less susceptible in strong caustic solution

50 40 30

40

30

20

10

20

0

Elongation to rupture (with PbO) Elongation to rupture (with PbO + NiB)

10

-10

0 6

7

8

9

10

11

12

13

o

pH(315 C) Fig. 7. Elongation to rupture as a function of the pH of leaded solutions with and without NiB for the HTMA Alloy 600.

0

10

20

30

40

50

60

70

2

Real Impedance / Ω cm

Fig. 9. Nyquist plot obtained from the electrochemical impedance measurements for the TT Alloy 600 immersed in the 0.1 M NaOH solution with PbO + NiB at 315 °C.

D.-J. Kim et al. / Corrosion Science 53 (2011) 1247–1253

solution. The impedance spectrum resembles the spectrum of the unleaded solution rather than that of the leaded solution. It was reported that the fast adsorption of NiB on the Alloy 600 surface occurs at an early stage of exposure in an aqueous solution and this behaviour can affect the oxide property and SCC resistance [20]. However, there is a need for further mechanistic study of the inhibitive effect of NiB. The passivity tendency in the EIS measurements is consistent with the SCC resistance tendency. 4. Conclusions (1) The HTMA Alloy 600 is more susceptible to PbSCC in a mild alkaline solution than in a strong alkaline solution. (2) The incorporation of lead in oxide degrades the oxide passivity and increases SCC susceptibility. A relatively passive duplex oxide layer of a porous Ni-rich outer layer and a dense Cr-rich inner layer forms in the unleaded solution but a duplex oxide layer is not observed and cations are depleted when lead is incorporated in the oxide layer that forms in the leaded solution. (3) The thermodynamic considerations and experimental results of the high temperature leaded solution indicate that the major element Ni of Alloy 600 can be depleted in a metallic state and in a Ni oxide state unlike Cr. Moreover matrix cations can be depleted by the Pb electrodeposition. The incorporated Pb prevents the formation of a passive oxide composed of Ni, Cr, Fe and O leading to PbSCC. (4) PbSCC is significantly reduced when NiB is added as an inhibitor. This result is consistent with the reduction in the amount of incorporated Pb and the increase in electrochemical impedance.

Acknowledgement This work was funded by Korea Ministry of Education, Science and Technology (MEST).

1253

References [1] J.M. Sarver, EPRI Workshop on Intergranular Corrosion and Primary Water Stress Corrosion Cracking Mechanisms, NP-5971, EPRI, Palo Alto, (1987) pp. C11/1. [2] M.L. Castano-Marin, D. Gomez-Briceno, F. Hernandez-Arroyo, Proc. of 6th Int. Symp. on Environmental Degradation of Materials in Nuclear Power SystemsWater Reactors, San Diego, CA, August 1–5 (1993) pp. 189–196. [3] M.D. Wright, M. Mirzai, Proc. of 9th Int. Symp. on Environmental Degradation of Materials in Nuclear Power Systems-Water Reactors, Newport Beach, CA, August 1–5 (1999) pp. 657–665. [4] R.W. Staehle, Proc. of 11th Int. Symp. on Environmental Degradation of Materials in Nuclear Power Systems-Water Reactors, Stevenson, WA, August 10–14 (2003) pp. 381–422. [5] F. Vaillant, D. Buisine, B. Prieux, D. Gomez Briceno, L. Castano, Eurocorr 96, Nice, (1996) pp. 13/1. [6] U.C. Kim, K.M. Kim, E.H. Lee, J. Nucl. Mater. 341 (2005) 169–174. [7] D.J. Kim, H.P. Kim, Proc. of 6th CNS Int. Steam Generator Conference, Toronto, Ontario, November 8–11 (2009) pp. 32. [8] K. Fruzzetti, Workshop of Effects of Pb and S on the Performance of Secondary Side Tubing of Steam Generators in PWRs, ANL, IL, May 24–27 (2005). [9] J. Robertson, Corros. Sci. 29 (1989) 1275–1291. [10] P. Marcus, J.M. Grimal, Corros. Sci. 33 (1992) 805–814. [11] A. Gebert, F. Schneider, K. Mummert, Nucl. Eng. Design 174 (1997) 327–334. [12] J.-J. Park, S.-I. Pyun, S.-B. Lee, Electrochim. Acta 49 (2004) 281–292. [13] D.-J. Kim, H.C. Kwon, H.P. Kim, Corros. Sci. 50 (2008) 1221–1227. [14] S.S. Hwang, U.C. Kim, Y.S. Park, J. Nucl. Mater. 246 (1997) 77–83. [15] T. Sakai, S. Okabayashi, K. Aoki, K. Matsumoto, Y. Kishi, CORROSION/90, NACE, Houston, (1990) paper No. 520. [16] A.M. Lanch, M. Garcia, M. Hernandez, Prod. Int. Conf. on Chemistry in Water Reactor, Nice, France, April 24–27 (1994) (SFEN, 1994). [17] MULTEQ calculation performed at ANL (2008). [18] http://www.lasurface.com. [19] M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, Pergamon Press Ltd., 1966. [20] D.-J. Kim, Y.S. Lim, H.C. Kwon, S.S. Hwang, H.P. Kim, J. Nanosci. Nanotechnol. 10 (2010) 85–91. [21] B. Lu, J. Luo, Y. Lu, Workshop on Detection, Avoidance, Mechanisms, Modeling, and Prediction of SCC Initiation in Water-Cooled Nuclear Reactor Plants, Beaune, Burgundy, France, September 7–12 (2008). [22] HSC Chemistry Database, 6.0. [23] EPRI report (NP-3137), 1983. [24] H. Hirano, H. Kawamura, H. Takamatsu, H. Takiguchi, Workshop of Effects of Pb and S on the Performance of Secondary Side Tubing of Steam Generators in PWRs, ANL, IL, May 24–27 (2005). [25] D.-J. Kim, H.P. Kim, S.S. Hwang, J.S. Kim, J. Park, Met. Mater. Int. 16 (2010) 259– 266.