Electrochemical characterisation of nickel-based alloys in sulphate solutions at 320 °C
Introduction
Stress corrosion cracking (SCC) is a form of corrosion that affects certain components of pressurised water reactors (PWR), and in particular the primary and secondary circuits of the steam generators.
This type of corrosion occurs on the secondary side, in confined zones where impurities of low volatility become concentrated because of limited water flow rates and differences in heat flows (hide-out phenomenom). Heat exchanger tubes are affected by cracking, caused by the simultaneous action of the chemical environment and the mechanical and thermal stress of the installation. Numerous studies have been performed to determine these conditions in secondary environment [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21].
It is very important to be able to predict stress corrosion in the steam generator tubes of PWR reactors so as to guarantee the safety of the installation and to plan maintenance tasks. Information on SCC encountered in nuclear installations is reproduced in laboratories in conditions as close as possible to real operating conditions. Such measurements are generally long and costly and the possibility of predicting/anticipating SCC by carrying out electrochemical measurements is of considerable interest. In fact, some stress corrosion mechanisms involve localised dissolution of the alloy when the oxide layer cracks and are thus directly related to the electrochemical properties of the alloy [22]. Stress corrosion occurs in potential ranges where the passive layers are not very stable, in particular around the Ni/NiO equilibrium and when the passivated surface is damaged by stress [10]. SCC depends on numerous factors, including the alloy composition and in particular the chromium content [6]. The tubes used in nuclear facilities are made from nickel-based (Inconel 600, Inconel 690) or iron-based (Incoloy 800) alloys. Alloy 600 has been the one most commonly used and is the most sensitive to SCC [4], [5], [8], [10], [11], [12], [13], [14], [18], [19], [20], [21]. This alloy is used after its original heat treatment (600MA) or may be subjected to an additional heat treatment (600TT) to improve its stress corrosion resistance. Alloy 690, which has a higher chromium content, is only very slightly susceptible to SCC in most environments that crack alloy 600 [6], [7].
On the secondary side of steam generators, earlier studies showed that in confined zones the environment was likely to become caustic with a high pH level, and most electrochemical studies have concerned this type of environment [1], [18], [19]. Some authors have modelled stress corrosion cracking in a caustic environment [23].
More recent analyses of secondary side environments, performed after hide-out return of the species dissolved at high temperatures, have shown the confined zones developed in SG secondary sides to be neutral or slightly alkaline sulphate environments [24], [25].
In these sulphate conditions, SCC was found to decrease as the pH level was increased from pH 5–10 [26]. SCC increased with increasing sulphate concentrations and was strongly dependent on the alloy type, being high for alloy 600MA, and lower for alloy 600TT, after being subjected to additional annealing. It was lower still for alloy 800 and was not observed for alloy 690.
Combrade et al. examined the local chemistry [16], [17] and showed the importance of surface films [27], [28], [29]. They showed that the presence of sulphate reduction products also promoted SCC [30]. The presence of sulphides, which are more stable than oxides, has been shown to inhibit the formation of the protective oxide layer [10]. In solutions with reducing properties, such as in the presence of hydrazine, the reduction of sulphates to sulphides and thiosulphates has been measured at high temperatures [31]. Thiosulphates are known for their depassivating effect even at ambient temperatures [32].
Far fewer studies have been performed in sulphate solutions than in caustic conditions, and aside from stress corrosion data, there are few electrochemical data available for this type of environment [33], [34], [35], [36]. Some authors have studied the semi-conductive nature of passive layers formed at different temperatures [35] or formed at high temperatures (although the measurements were made under ambient conditions) [37] and have completed these measurements by surface analyses (Auger, ESCA, SIMS, SDL) [36], [37], [38], [39], [40].
The aim of the present study was to obtain better knowledge of the electrochemical properties of nickel-based alloys at 320 °C in sulphate solutions. Different nickel-based alloys were studied, with different sulphate concentrations at neutral or slightly alkaline pH. The experimental results include corrosion potential measurements, polarisation curves and polarisation resistance measurements based on the polarisation curves and using Electrochemical Impedance Spectroscopy (EIS). The results were analysed in the light of the thermodynamic stability of the alloy components and solution constituents at 320 °C [41], [42], [43], [44], stress corrosion data [26] and the surface analyses in this type of solution [36], [37], [38], [39], [40].
Section snippets
Alloys studied
Several nickel-based alloys (600MA, 600TT, 690) and one iron-based alloy (800) were studied. The composition, heat treatment and microstructure of these alloys are given in Table 1. The standard, mill-annealed (MA) specimen is subjected a preliminary annealing at 1050 °C. The thermally-treated condition (TT) corresponds to an additional annealing for 16 h at 700 °C, which modifies the microstructure. This treatment promotes precipitation of carbides on the grain boundaries, which generally
Corrosion potential measurements
The corrosion potentials of the alloys (Fig. 1) measured after stabilisation correspond to mixed potentials from simultaneous oxidation of the metal elements and reduction of the electrolyte, in particular hydrogen evolution and possibly sulphate reduction. The platinum potential simply provided an indication of the redox activity of the solution. No significant differences were observed between the corrosion potential of the platinum and that of the alloys. Comparison of the corrosion
Conclusion
The thermodynamic stability of the passive layers of alloys 600, 690 and 800 at 320 °C, 12 MPa, in sulphate solutions at pH 5 to 9.5, was strongly influenced by pH. The potential–pH diagrams showed that the passive layer was very stable at neutral pH levels (pH 5–6) because of the presence of chromium oxide and nickel and iron oxide or sulphide. The oxide layer was less stable at alkaline pH levels (pH 8–9.5) because of chromium oxide solubility. In accordance with these data, the polarisation
Acknowledgements
The authors are grateful to Electricité De France for financial support and to O. De Bouvier, EDF, Centre de Recherche des Renardières, 77818, Moret-sur-Loing for fruitful discussion and collaboration.
References (45)
- et al.
Corros. Sci.
(1996) - et al.
J. Nucl. Mater.
(2000) Mater. Chem. Phys.
(1997)- et al.
J. Nucl. Mater.
(1996) - et al.
Corros. Sci.
(1996) - et al.
J. Nucl. Mater.
(1996) - et al.
Corros. Sci.
(2000) - et al.
Corros. Sci.
(1995) - et al.
Annales de Chimie Science des Matériaux
(2002) - et al.
Corros. Sci.
(1997)