Effect of temperature on corrosion and semiconducting properties of oxide films formed on M5 zirconium alloy

Abstract

Nuclear fuel cladding for pressurised water reactors is commonly manufactured with zirconium alloys. The M5 alloy is a relatively new cladding material for in-reactor used with enhanced performance compared to traditional zircaloys. In this work, the influence of temperature on the corrosion resistance and semiconducting properties of the passive film formed on the M5 alloy in a borate buffer solution has been evaluated. The electrochemical behaviour of the zirconium alloy was assessed by potentiodynamic polarisation tests, electrochemical impedance spectroscopy and Mott–Schottky plots. The results indicated that the corrosion resistance of the M5 alloy decreased with temperature due to the formation of a less stable and more defective passive film. The Mott–Schottky approach used in combination with polarisation tests and impedance measurements was effective to reveal the protective state of the passive film on the M5 alloy.

Keywords

M5 alloy Zirconium Passive film Mott–Schottky Temperature

Introduction

Zirconium alloys are the state of the art materials for nuclear fuel claddings in pressurised water reactors (PWRs) due to a suitable combination of properties such as corrosion and creep resistances and low neutron absorption cross-section.¹ Zr–Nb alloys are preferred over traditional zircaloys due to performance enhancement regarding corrosion resistance and hydrogen pickup.² The development of this new class of zirconium alloys is pursued by materials scientists in order to meet increasing demands of the nuclear industry for high burnup fuels.³ The traditional example of these so called advanced zirconium alloys is the M5 alloy. It consists of a Zr–Nb–O alloy that is currently employed as nuclear fuel cladding and structural components for nuclear core applications⁴ in PWRs. Attractive properties of the M5 alloy derive from controlled chemical composition and microstructure during manufacturing. The absence of tin increases corrosion resistance with respect to traditional zircaloys, and the fully recrystallised state leads to enhanced creep resistance for in-reactor use.⁵

Corrosion of fuel claddings is an important technological issue in the nuclear industry. Chen et al. ⁶ state that corrosion can lead to uniform thinning of fuel cladding and also to localised attack. Cox⁷ discusses several in-reactor corrosion mechanisms of zirconium alloys and pointed that the stability of the oxide film on the surface of the metallic substrate is closely related to the corrosion rate during operation. In this regard, the metal/oxide interface has been found to play an important role in the corrosion processes of Zr–Nb alloys.⁸ The stability of the oxide film can be affected by the niobium content in the alloy and also by annealing treatments during manufacturing of engineering components due to microstructural changes induced by different thermal cycles.⁹

The protective character of oxide films against corrosion can be assessed by their semiconducting properties. It is well known that passive layers on metallic surfaces behave as semiconductors. Corrosion phenomena are strongly related to the semiconductive properties of these films. Mott–Schottky approach applies well to characterise the semiconductive behaviour of oxide films in contact with an electrolyte and has been widely employed to study the stability of passive layers on stainless steels, titanium, aluminium and magnesium alloys.^10–14 The predominance of the type of point defects in the passive film accounts for a p-type or n-type semiconductor behaviour. Thus, when cation vacancies are in excess, the passive film generally behaves as a p-type semiconductor, whereas anion vacancies or cation interstitials lead to n-type behaviour.¹⁵ The concentration and distribution of these defects within the passive film determine its stability and influence the corrosion resistance of the underlying metallic substrate.¹⁶

Corrosion damage of zirconium alloys is related to operating temperatures.³ Although temperature has been also recognised as an important parameter for other passive metals in aqueous solutions,^17,18 studies devoted to Zr–Nb alloys are scarce in the literature. The main goal of the present work was to investigate the formation and growth of passive film in M5 zirconium alloy. In order to meet this goal, the Mott–Schottky approach was used to assess the concentration of point defects within the passive film formed under different temperatures in a phosphate buffered solution. This electrolyte was selected to provide a stable passive layer, giving the basis for the understanding of fundamental aspects of passivity in the zirconium alloy. Corrosion tests were also carried out to provide a complete characterisation of the M5 electrochemical behaviour at each testing temperature.

Experimental

The material used in this work was a tube cladding made of M5 zirconium alloy (nomical composition: 0.8-1.2 wt-% Nb, 1100-1700 ppm O, < 350 ppm S, bal. Zr). Samples were cut from the tube cladding in cylindrical pieces (1 cm long). Next, the samples were connected to a copper wire using a conductive adhesive. Then, the samples were embedded in cold curing epoxy resin, leaving an area of ∼0.3 cm² to be exposed to the electrolyte. Before the electrochemical tests, the samples were ground with SiC abrasive paper and polished with diamond paste.

The electrolyte for all electrochemical measurements was a buffer solution of composition H₃BO₃ (0.05 M) + Na₂B₄O₇.10H₂O (0.075 M) and pH 9.2. The advantage of choosing this electrolyte for the measurements is the stable condition of the oxide film formed on the metallic surface, allowing for reliable assessment of its semiconducting properties.¹⁹ The tests were performed at room temperature and at 40, 60 and 90°C using an Autolab PGSTAT 100 potentiostat/galvanostat. The experimental arrangement was composed of a conventional three-electrode cell with a platinum wire as the counter-electrode, saturated calomel electrode (SCE) as reference and the M5 alloy as the working electrode. All potentials mentioned in the text are given related to the SCE.

Two sets of experiments have been performed. In the first one, the open circuit potential (OCP) was monitored for 1800 s. Next, the sample was potentiodynamically polarised from − 300 mV versus the OCP up to +1.0 V. In the second one, the OCP was also monitored for 1800 s. Next, electrochemical impedance spectroscopy (EIS) measurements were performed at the OCP in the frequency range from 100 kHz to 10 mHz with an acquisition rate of 10 points per decade and an amplitude of the perturbation signal of ± 10 mV (rms). In the next step, capacitance measurements were carried out using the Mott–Schottky approach at a fixed frequency of 1 kHz. The samples were polarised in the cathodic direction in successive steps of 50 mV from − 1.0 V up to − 2.0 V. Three different samples were tested at each condition.

Results and discussion

Potentiodynamic polarisation curves

Potentiodynamic polarisation curves of the M5 alloy immersed in H₃BO₃ (0.05 M) + Na₂B₄O₇.10H₂O (0.075 M) and pH 9.2 at different temperatures are shown in Fig. 1. The corrosion potential (E _corr) and corrosion current densities (i _corr) were determined from these curves using the Tafel extrapolation method. The results are presented in Table 1. Passive current densities (i _pass) are also displayed in Table 1.

Potentiodynamic polarisation curves of M5 alloy immersed in H₃BO₃ (0.05 M) + Na₂B₄O₇.10H₂O (0.075 M) and pH 9.2 at different temperatures

Table 1

Electrochemical parameters determined from potentiodynamic polarisation curves shown in Fig. 1

Sample	E _corr/mV	i _corr/μA cm^{− 2}	i _pass/μA cm^{− 2}
RT	− 303 ± 92	0.07 ± 0.02	2.63 ± 0.97
40°C	− 192 ± 17	0.16 ± 0.04	5.45 ± 1.64
60°C	− 165 ± 40	0.51 ± 0.10	6.09 ± 2.57
90°C	− 154 ± 41	1.19 ± 0.45	10.61 ± 2.31

The polarisation curves show a similar trend at all temperatures, with the material reaching a stable passive condition above the corrosion potential. No breakdown potential could be perceived. Notwithstanding, some marked differences can be perceived. The corrosion potential increases with temperature. This behaviour has been observed for passive metals.^20,21 Temperature also affected the values of i _corr, which were shown to increase as the temperature rises. These two effects derive from the enhancement of the cathodic reactions in the surface of the electrode with temperature. The shift of the polarisation to higher potentials and current densities with increasing temperatures is clearly seen in Fig. 1. Similar behaviour has been reported for stainless steels in aqueous solutions.²²

The variations of E _corr and i _corr point to a reduction of the passive character of the oxide film formed on the surface of the M5 alloy with temperature. Thus, the anodic dissolution process is also affected by temperature. This can be promptly envisaged by considering the passive current densities displayed in Table 1. There is a steep increase of the passive current density with temperature, suggesting also that the growth of the passive film is enhanced at higher temperatures.²⁰ Moreover, the passive range shortens at higher temperatures, as denoted by the smaller difference between the final potential (1.0 V) and E _corr. Literature indicates that passive films that formed at lower temperatures are less defective and have higher resistance to breakdown than those formed at higher temperatures.²³ The results obtained in the present work confirm this trend and reveal that the corrosion resistance of M5 alloy can be adversely affected by temperature.

Electrochemical impedance spectroscopy

EIS spectra of the M5 alloy obtained in H₃BO₃ (0.05 M) + Na₂B₄O₇.10H₂O (0.075 M) and pH 9.2 at different temperatures are shown in Fig. 2. Nyquist plots (Fig. 2a) show a capacitive loop that becomes more flattened with increasing temperatures. The radius of the capacitive loop was severely affected by temperature. It is well known that the diameter of the capacitive loop is directly related to the polarisation resistance of the passive film, thus reflecting its corrosion resistance.^24,25 Following this reasoning, the highest corrosion resistance was observed at room temperature. A small decrease of the radius of the capacitive loop was observed up to 40°C. This trend was markedly intensified at 60°C and especially at 90°C. The plot obtained at 90°C present a very small radius compared to the other temperatures so that it is badly resolved in Fig. 2a. In order to more clearly distinguish between this condition and the others, the inset in Fig. 2a shows the Nyquist plots with an expanded scale, evidencing the small capacitive loop obtained at 90°C.

EIS spectra of M5 alloy immersed in H₃BO₃ (0.05 M) + Na₂B₄O₇.10H₂O (0.075 M) and pH 9.2 at different temperatures: a Nyquist plots; b bode plots; solid lines, fitted data

Bode plots (Fig. 2b) obtained at room temperature and at 40°C are typical of passive metals exhibiting high impedance modulus and capacitive behaviour with phase angles close to − 90°.^26,27 The impedances are progressively lower for increasing temperatures, reaching its minimum at 90°C. Similarly, phase plots present an extended plateau ranging from the high frequency to the medium frequency domain, denoting the capacitive character of the electrode surface with angles close to − 90°. In the low frequency region of the spectra, there is a steep decrease of the phase angles that can be related to charge transfer reactions at the metal/passive film interface. The plateau in the medium frequency domain was narrower for the plots acquired at 60 and 90°C. The decrease of phase angles is shifted to frequencies as high as 10 Hz at 60 and 90°C, suggesting that charge transfer reactions are facilitated at these temperatures. Additionally, the peak phases are more distant from − 90°, deviating from the typical pure capacitive behaviour.

The qualitative analysis of the impedance spectra points to a decrease of the protective character of the passive formed at the surface of the M5 alloy as the temperature increases, in agreement with the results obtained with the potentiodynamic polarisation tests described in the ‘Potentiodynamic polarisation curves’ section. Fitting of EIS experimental data with equivalent electrical circuits (EECs) is a useful procedure to give a more quantitative analysis of the events occurring at the electrode. Hence, interpretation of the EIS results was complemented by fitting the experimental data with the EEC shown in Fig. 3. This circuit is typically employed to model the electrochemical response of metals that develop passive films in aqueous solutions, presenting two time constants.^28,29 Constant phase elements (CPEs) were used instead of pure capacitors in order to represent the deviation from the pure capacitive behaviour as observed in the Bode phase plots of Fig. 2b. The impedance of a CPE (Z _CPE) is given according to equation (1), where ω is the angular frequency, j ^{− 2} = − 1 is the imaginary number and n is the exponent of the CPE. For an ideal capacitor, n = 1 and for an ideal resistor n = 0. When 0.5 < n < 1, the CPE describes frequency dispersion arising from local heterogeneities in the electrode.²⁷ (1) In the circuit shown in Fig. 3, R _s describes the electrolyte resistance, R ₁ is the oxide film resistance, CPE₁ represents the capacitance of the oxide film, and R ₂ and CPE₂ model the charge transfer resistance and the capacitive behaviour at the metal/oxide film interface respectively. The quality of the fitting procedure can be promptly assessed by comparing the fitted and experimental data in Fig. 2. The results of the fitting procedure are shown in Table 2.

Equivalent circuit used to fit EIS experimental data

Table 2

Equivalent circuit parameters obtained by fitting EIS experimental data with EEC shown in Fig. 3

Sample	R _s/Ω cm²	R ₁/kΩ cm²	CPE₁ (10^{− 4} Ω^{− 1} cm^{− 2} sⁿ)	n ₁	R ₂/kΩ cm²	CPE₂ (10^{− 6} Ω^{− 1} cm^{− 2} sⁿ)	n ₂
RT	23.7 ± 1.1	92.0 ± 3.6	0.08 ± 0.01	0.95 ± 0.02	523 ± 24	2.9 ± 0.7	0.96 ± 0.04
40°C	16.5 ± 3.8	97.8 ± 7.8	0.06 ± 0.02	0.93 ± 0.03	243 ± 39	7.8 ± 3.0	0.84 ± 0.15
60°C	16.6 ± 3.0	14.1 ± 2.9	0.13 ± 0.03	0.85 ± 0.06	96 ± 7	42.6 ± 8.6	0.78 ± 0.11
90°C	11.1 ± 2.2	9.0 ± 0.3	0.37 ± 0.04	0.81 ± 0.03	11 ± 1	365.1 ± 49.5	0.67 ± 0.02

The variation of R ₁ indicates that the resistance of the oxide film is severely affected by temperature, sharply decreasing when the temperature was increased up to 90°C. The capacitances of the oxide film (CPE₁) increased with temperature. This effect can be related to an increase of film defects with temperature. These results can be associated with those obtained with the polarisation tests (Fig. 1). As discussed in the ‘Potentiodynamic polarisation curves’ section the values of passive current density (i _pass) increased with temperature, as well as the values of corrosion current density (i _corr) and corrosion potential (E _corr). Such variations are related to a less stable passive film that grows faster as the temperature increases. Consequently, this rapidly grown passive film would be less protective and more defective. The variations of R ₁ and CPE₁ agree well with this reasoning, confirming the results obtained with the polarisation tests.

The second time constant models the electrochemical response at the metal/passive film interface at the defects of the passive layer. The high value of R ₂ indicates the high corrosion resistance of the M5 alloy at room temperature. In the same way, the low value of CPE₂ suggests that the area exposed to the electrolyte is small at this condition and the passive film has few defects.³⁰ However, this protective state is not sustained as temperature increases. R ₂ shows a steep decrease as temperature reaches 40°C, reaching its minimum at 90°C. The values of CPE₂ increased sharply with temperature, presenting an increment of two orders of magnitude at 90°C when compared to the room temperature result. It is also interesting to note the deviation of the n ₂ values from the pure capacitive behaviour (n = 1) with temperature, suggesting the increased heterogeneity of the electrochemical interface.

The variations of R ₂ and CPE₂ with temperature reveal the less protective nature of the passive film formed on the surface of the M5 alloy above room temperature. These results are of great relevance to the engineering applications of this material, since temperatures can reach >300°C depending on its intended in-reactor use.³¹ The defective nature of the passive film can be assessed by the variation of its semiconducting properties based on the Mott–Schottky approach. The next section deals with this interpretation, discussing the correlation between the semiconducting properties of the passive film and the corrosion tests carried out by potentiodynamic polarisation and EIS measurements.

Semiconducting properties of passive film

According to the Mott–Schottky theory, the capacitance (C ) of passive films that formed on metals varies the applied potential and can be described according to equation (2), where C _SC is the capacitance of the space charge layer of the oxide film (semiconductor) and C _H is the Helmholtz capacitance.³² (2) The space charge capacitance is negligible compared to the Helmholtz capacitance when the potential is applied at high frequency so that the potential drop occurs entirely inside the space charge layer. In this case, the measured capacitance would be due only to the space charge layer and is given by equations (3) and (4) for n-type and p-type semiconductors respectively. In these equations, ε is the dielectric constant of the oxide film (ε = 21 for ZrO₂ ³³), ε₀ is the vacuum permittivity (8.85 × 10^{− 14} F cm^{− 1}), N _d is the donor density and N _a is the acceptor density in the passive film, E is the applied potential, E _FB is the flat band potential, k is the Boltzmann constant, T is the absolute temperature, and e is the elementary charge (e = 1.602 × 10^{− 19} C).³⁴ (3) (4) Mott–Schottky plots (C^{− 2} versus E) of the M5 alloy immersed in H₃BO₃ (0.05 M) + Na₂B₄O₇.10H₂O (0.075 M) and pH 9.2 at different temperatures are shown in Fig. 4. All plots are characterised by a linear region with a positive slope between − 1.6 and − 2.0 V, denoting an n-type semiconductor behaviour.¹⁶ At higher potentials, the capacitance tends to be independent of the applied potential and Mott–Schottky behaviour is no longer observed. The n-type behaviour of ZrO₂ films has been reported by other authors.³⁵ The main charge carriers in oxide films with n-type semiconductivity are oxygen vacancies and cation interstitials.³⁶ It is reported that such dopants prevent the migration of cations from the metallic substrate to the electrolyte and also the incorporation of aggressive anions from the electrolyte into the passive film.³⁷ Cation vacancies are associated with passivity breakdown and the onset of pitting corrosion.³⁸ The protective character of the passive film against corrosion would thus be favoured under this situation.

Mott–Schottky plots of M5 alloy immersed in H₃BO₃ (0.05 M) + Na₂B₄O₇.10H₂O (0.075 M) and pH 9.2 at different temperatures

The donor density (N _d) in the passive film has been determined from the slope of linear regions of the Mott–Schottky plots. The results are presented in Table 3. The magnitude of N _d was ∼10¹⁹ cm^{− 3} independently of the testing temperature. Highly disordered passive films are reported to present doping densities in the range 10²⁰–10²¹ cm^{− 3}.^39,40 In this respect, the films formed on the M5 alloy can be considered relatively little disordered due to the low concentration of defects. However, the results displayed in Table 3 indicate that these films became more defective with temperature, since N _d gradually increased from room temperature to 90°C. The formation of a more disordered passive film at progressively higher temperatures is consistent with the results obtained from the potentiodynamic polarisation curves and EIS spectra. Therefore, it is possible to infer that the formation of disordered passive films leads to high passive current densities and are indicative of the growth velocity of the oxide layer. Electrochemical evaluation by polarisation tests and Mott–Schottky plots proved to be a valuable tool to identify this tendency and to evaluate the stability of the passive film on the M5 alloy.

Table 3

Donor densities (N _d) for M5 alloy immersed in H₃BO₃ (0.05 M) + Na₂B₄O₇.10H₂O (0.075 M) and pH 9.2 at different temperatures

Charge carrier concentration	RT	40°C	60°C	90°C
N _d (10¹⁹ cm^{− 3})	4.07 ± 1.45	7.09 ± 1.63	7.42 ± 2.06	11.2 ± 1.29

Conclusions

The effect of temperature on the corrosion resistance and semiconducting properties of the passive film formed on the M5 alloy has been evaluated. Potentiodynamic polarisation tests indicated that the corrosion potential, corrosion current density and passive current density increased with temperature, therefore suggesting the formation of a less stable and rapidly grown passive layer at high temperatures. EIS spectra revealed that the electrochemical response of the passive film could be successfully modelled by an equivalent circuit consisting of two time constants. The resistance of the oxide film decreased with temperature and its capacitance increased, supporting the results obtained with the polarisation tests. The passive films presented an n-type semiconductive behaviour according to Mott–Schottky plots. The donor density increased with temperature, showing that the passive film became more defective at high temperatures and, therefore, more prone to corrosion attack, confirming the results obtained by EIS measurements and polarisation tests. The results obtained in the present work show that the overall electrochemical behaviour of the M5 alloy is adversely affected by temperature. The formation of a more disordered passive film at higher temperatures accounts for the loss of stability of the passive film.

Footnotes

Acknowledgement

The authors gratefully acknowledge Indústrias Nucleares do Brasil (INB) for kindly providing the M5 alloy used in this work.

References

Azevedo

C. R. F.

: ‘Selection of fuel cladding material for nuclear fission reactors’ Eng. Fail. Anal., 2011 18 1943–1962. doi: 10.1016/j.engfailanal.2011.06.010

Tang

and Yang

: ‘Dissolution and precipitation behaviours of hydrides in N18, Zry-4 and M5 alloys’ Int. J. Hydrogen Energy 2009 34 7269–7274. doi: 10.1016/j.ijhydene.2009.07.018

Bojinov

Karastoyanov

Kinnunen

and Saario

: ‘Influence of water chemistry on the corrosion mechanism of a zirconium-niobium alloy in simulated light water reactor coolant conditions’ Corros. Sci., 2010 52 54–67. doi: 10.1016/j.corsci.2009.08.045

Peng

D. Q.

Bai

X. D.

and Chen

B. S.

: ‘Corrosion behavior of carbon-implanted M5 alloy in 1 M H₂SO₄’ Appl. Surf. Sci., 2005 245 215–222. doi: 10.1016/j.apsusc.2004.10.012

Mardon

J. P.

Garner

G. L.

and Hoffmann

P. B.

: ‘M5® a breakthrough in Zr alloy’. Proc. LWR Fuel Performance Meeting, Top Fuel, Orlando, FL, USA, September 2010.

Chen

Urquidi-Macdonald

and Macdonald

D. D.

: ‘The electrochemistry of zirconium in aqueous solutions at elevated temperatures and pressures’ J. Nucl. Mater., 2006 348 133–147. doi: 10.1016/j.jnucmat.2005.09.014

Cox

: ‘J. Some thoughts on the mechanisms of in-reactor corrosion of zirconium alloys’ J. Nucl. Mater., 2005 336 331–368. doi: 10.1016/j.jnucmat.2004.09.029

Müller

and Lanzani

: ‘Corrosion of zirconium alloys in concentrated lithium hydroxide solutions’ J. Nucl. Mater., 2013 439 251–257. doi: 10.1016/j.jnucmat.2012.07.030

Kim

H. G.

Park

S. Y.

Lee

M. H.

Jeong

Y. H.

and Kim

S. D.

: ‘Corrosion and microstructural characteristics of Zr-Nb alloys with different Nb contents’ J. Nucl. Mater., 2008 373 429–432. doi: 10.1016/j.jnucmat.2007.05.035

10.

BenSalah

Sabot

Triki

Dhouibi

Refait

Ph.

and Jeannin

: ‘Passivity of Sanicro28 (UNS N-08028) stainless steel in polluted phosphoric acid at different temperatures studied by electrochemical impedance spectroscopy and Mott-Schottky analysis’ Corros. Sci., 2014 86 61–70. doi: 10.1016/j.corsci.2014.04.056

11.

Taveira

L. V.

Montemor

M. F.

Da Cunha Belo

Ferreira

M. G.

and Dick

L. F. P.

: ‘Influence of incorporated Mo and Nb on the Mott-Schottky behaviour of anodic films formed on AISI 304L’ Corros. Sci., 2010 52 2813–2818. doi: 10.1016/j.corsci.2010.04.021

12.

Woldemedhin

M. T.

Raabe

and Hassel

A. W.

: ‘Characterization of thin anodic oxides of Ti-Nb alloys by electrochemical impedance spectroscopy’ Electrochim. Acta 2012 82 324–332. doi: 10.1016/j.electacta.2012.06.029

13.

Levine

K. L.

Tallman

D. E.

and Bierwagen

G. P.

: ‘Mott–Schottky analysis of aluminium oxide formed in the presence of different mediators on the surface of aluminium alloy 2024-T3’ J. Mater. Proc. Technol., 2008 199 321–326. doi: 10.1016/j.jmatprotec.2007.08.023

14.

de Oliveira

M. C. L.

Pereira

V. S. M.

Correa

O. V.

de Lima

N. B.

and Antunes

R. A.

: ‘Correlation between the corrosion resistance and the semiconducting properties of the oxide film on AZ91D alloy after solution treatment’ Corros. Sci., 2013 69 311–321. doi: 10.1016/j.corsci.2012.12.015

15.

Fattah-alhosseini

Golozar

M. A.

Saatchi

and Raeissi

: ‘Effect of solution concentration on semiconducting properties of passive films formed on austenitic stainless steels’ Corros. Sci., 2010 52 205–209. doi: 10.1016/j.corsci.2009.09.003

16.

Fernández-Domene

R. M.

Blasco-Tamarit

García-García

D. M.

and García-Antón

: ‘Passive and transpassive behaviour of Alloy 31 in a heavy brine LiBr solution’ Electrochim. Acta 2013 95 1–11. doi: 10.1016/j.electacta.2013.02.024

17.

Zhang

and Shoesmith

D. W.

: ‘Influence of temperature on passive film properties on Ni-Cr-Mo alloy C-2000’ Corros. Sci., 2013 76 424–431. doi: 10.1016/j.corsci.2013.07.016

18.

Yang

Guo

and Liu

: ‘The effect of temperature on corrosion behavior of SS316L in the cathode environment of proton exchange membrane fuel cells’ J. Power Sources 2011 196 5503–5510. doi: 10.1016/j.jpowsour.2011.02.070

19.

Modiano

Fugivara

C. S.

and Benedetti

A. V.

: ‘Effect of citrate ions on the electrochemical behavior of low-carbon steel in borate buffer solutions’ Corros. Sci., 2004 46 529–545. doi: 10.1016/S0010-938X(03)00170-7

20.

Blasco-Tamarit

Igual-Munhoz

García-Antón

and García-García

: ‘Effect of temperature on the corrosion resistance and pitting behaviour of Alloy 31 in LiBr solutions’ Corros. Sci., 2008 50 1848–1857. doi: 10.1016/j.corsci.2008.03.016

21.

Escrivà-Cerdán

Blasco-Tamarit

García-García

D. M.

García-Antón

Akid

and Walton

: ‘Effect of temperature on passive film formation of UNS N08031 Cr-Ni alloy in phosphoric acid contaminated with different aggressive anions’ Electrochim. Acta 2013 111 552–561. doi: 10.1016/j.electacta.2013.08.040

22.

Blasco-Tamarit

García-García

D. M.

and García-Antón

: ‘Imposed potential measurements to evaluate the pitting corrosion resistance and the galvanic behaviour of a highly alloyed austenitic stainless steel and its weldment in LiBr solution at temperatures up to 150°C’ Corros. Sci., 2011 53 784–795. doi: 10.1016/j.corsci.2010.11.013

23.

Abd Ed Meguid

E. A.

and Abd El Latif

A. A.

: ‘Critical pitting temperature for type 254 SMO stainless steel in chloride solutions’ Corros. Sci., 2007 49 263–275. doi: 10.1016/j.corsci.2006.06.011

24.

Zucchi

Grassi

Frignani

Monticelli

and Trabanelli

: ‘Electrochemical behaviour of a magnesium alloy containing rare earth elements’ J. Appl. Electrochem., 2006 36 195–204. doi: 10.1007/s10800-005-9053-3

25.

Souza

E. C.

Rossitti

S. M.

and Rollo

J. M. D. A.

: ‘Influence of chloride ion concentration and temperature on the electrochemical properties of passive films formed on a superduplex stainless steel’ Mater. Charact., 2010 61 240–244. doi: 10.1016/j.matchar.2009.12.004

26.

Valero Vidal

and Igual-Munhoz

: ‘Study of the adsorption process of bovine serum albumin on passivated surfaces of CoCrMo biomedical alloy’ Electrochim. Acta 2010 55 8445–8452. doi: 10.1016/j.electacta.2010.07.028

27.

Pan

Thierry

and Leygraf

: ‘Electrochemical impedance spectroscopy study of the passive film on titanium for implant application’ Electrochim. Acta 1996 41 1143–1153. doi: 10.1016/0013-4686(95)00465-3

28.

Fuchs-Godec

and Pavlovic

M. G.

: ‘Synergistic effect between non-ionic surfactant and halide ions in the forms of inorganic or organic salts for the corrosion inhibition of stainless-steel X4Cr13 in sulphuric acid’ Corros. Sci., 2012 58 192–201. doi: 10.1016/j.corsci.2012.01.027

29.

Kocijan

Merl

D. K.

and Jenko

: ‘The corrosion behaviour of austenitic and duplex stainless steels in artificial saliva with the addition of fluoride’ Corros. Sci., 2011 53 776–783. doi: 10.1016/j.corsci.2010.11.010

30.

Macdonald

D. D.

Ismail

K. M.

and Sikora

: ‘Characterization of the passive state on zinc’ J. Electrochem. Soc., 1998 145 3141–3149. doi: 10.1149/1.1838777

31.

Chen

Urquidi-Macdonald

and Macdonald

D. D.

: ‘Electrochemical impedance spectroscopic study of passive zirconium’ J. Nucl. Mater., 2008 379 162–168. doi: 10.1016/j.jnucmat.2008.06.011

32.

Hakiki

N. B.

Boudin

Rondot

and Da Cunha Belo

: ‘The electronic structure of passive films formed on stainless steels’ Corros. Sci., 1995 37 1809–1822. doi: 10.1016/0010-938X(95)00084-W

33.

Wang

D. P.

Wang

S. L.

and Wang

J. Q.

: ‘Relationship between amorphous structure and corrosion behaviour in a Zr-Ni metallic glass’ Corros. Sci., 2012 59 88–95. doi: 10.1016/j.corsci.2012.02.017

34.

Paola

A. D.

: ‘Semiconducting properties of passive films on stainless steels’ Electrochim. Acta 1989 34 203–210. doi: 10.1016/0013-4686(89)87086-0

35.

J-J.

Ling

Y-H.

Zhang

R.-Q.

Dai

and Bai

X-D.

: ‘Properties of passive nano films on zircaloy-4 affected by defects induced by hydrogen permeation’ Chin. Phys. B 2014 23 088106–1-088106-6. doi: 10.1088/1674-1056/23/8/088106

36.

Goodlet

Faty

Cardoso

Freitas

P. P.

Simões

A. M. P.

Ferreira

M. G. S.

and Da Cunha Belo

: ‘The electronic properties of sputtered chromium and iron oxide films’ Corros. Sci., 2004 46 1479–1499. doi: 10.1016/j.corsci.2003.09.022

37.

Carmezim

M. J.

Simões

A. M.

Figueiredo

M. O.

and Da Cunha Belo

: ‘Electrochemical behaviour of thermally treated Cr-oxide films deposited on stainless steel’ Corros. Sci., 2002 44 451–465. doi: 10.1016/S0010-938X(01)00076-2

38.

Macdonald

D. D.

: ‘Passivity: enabler of our metals based civilization’ Corros. Eng. Sci. Technol., 2014 49 143–155. doi: 10.1179/1743278214Y.0000000158

39.

Antunes

R. A.

de Oliveira

M. C. L.

and Costa

: ‘Study of the correlation between corrosion resistance and semi-conducting properties of the passive film of AISI 316L in physiological solution’ Mater. Corros., 2012 63 586–592.

40.

D. G.

Wang

J. D.

and Chen

D. R.

: ‘Influence of potentiostatic aging temperature and pH on the diffusivity of a point defect in the passive film on Nb in an HCl solution’ Electrochim. Acta 2012 60 134–146. doi: 10.1016/j.electacta.2011.11.024