- Email: [email protected]
Multilayered coatings: Tuneable protection for metals
Corrosion Science 52 (2010) 3847–3850
Contents lists available at ScienceDirect
Corrosion Science journal homepage: www.elsevier.com/locate/corsci
Letter
Multilayered coatings: Tuneable protection for metals T. Markley a, S. Dligatch b, A. Trinchi a, T.H. Muster a,*, A. Bendavid b, P. Martin b, D. Lau a, A. Bradbury a, S. Furman a, I.S. Cole a a b
CSIRO Division of Materials Science and Engineering, Private Bag 33, Clayton South, Victoria 3169, Australia CSIRO Division of Materials Science and Engineering, P.O. Box 218, Lindfield, NSW 2070, Australia
a r t i c l e
i n f o
Article history: Received 6 July 2010 Accepted 25 July 2010 Available online 2 August 2010 Keywords: A. Aluminium B. Polarization C. Electrochemical calculation C. Oxide coatings C. Passive films
a b s t r a c t The role of oxide bi-layers in controlling the onset of corrosion has been explored. A high-throughput electrochemical approach was employed to determine the breakdown potential of aluminium metal over-coated with combinations of silicon, titanium, aluminium and magnesium oxides. Bi-layered coatings consisting of two 100 nm thick metal oxide layers provided increased protection against breakdown, and combinations with vastly different iso-electric point of solid (IEPS) were found to exhibit improved barrier properties in comparison to single-component oxides. Furthermore, the most protective oxide bilayers were produced when a high IEPS oxide was deposited directly onto the metal surface and subsequently over-coated with a low IEPS oxide. The barrier properties of bi-layer coatings appear to be tuneable, with notable dependencies on surface charge and thickness. Crown Copyright Ó 2010 Published by Elsevier Ltd. All rights reserved.
1. Introduction All industrially relevant metals are thermodynamically unstable and rely upon the addition or formation of protective coatings to slow the kinetics leading to oxidation. Alloys that rely upon passive oxide layers, such as stainless steel and aluminium, tend to suffer from pitting corrosion at the sites of localised breakdown of the protective oxide. There are numerous works describing the importance of chloride ion transport through oxides and passive films and their role in nucleating pitting events [1,2]. Previous studies have also suggested that the surface charge of the oxide can alter the protective nature of the oxide [3]. The formation of oxide bilayers has also been shown to influence the barrier properties [4–7]. Previous work by McCafferty demonstrated that the pitting potential of oxide-covered aluminium in chloride-containing solutions can be affected by the iso-electric point of the oxide (IEPS) [3]. This led to a proposed model for pit initiation that relied upon the surface possessing a positive charge that would tend to attract chloride anions, thus leading to their transported to the metal interface via oxygen vacancies in the oxide. The initiation of pitting was therefore proposed by McCafferty to be controlled by three factors: (1) the oxide surface charge, (2) the heat of adsorption of chloride ions onto the oxide, and (3) the density of oxygen vacancies in the oxide coating.
* Corresponding author. Tel.: +61 3 95452874; fax: +61 3 95441128. E-mail address: [email protected] (T.H. Muster).
Sato [6] also explored the influence of oxide surface charge on controlling the transport of charged species through oxide layers. Oxide precipitation of metal multivalent oxyanions (i.e. metal phosphates/molybdates) can lead to fixed negative charges and the oxide becoming cation selective. In contrast, the precipitation of metal monovalent complexes (i.e. hydroxychlorides) lead to fixed positive charges and anion selectivity. In this manner Sato proposed that the specific adsorption of multivalent anions to the oxide–solution interface imposes a bipolar oxide structure. The bipolar structure possessing an anion-selective inner layer and cation-selective outer layer was demonstrated to block anodic ion transport. Hayashi et al. [4] studied changes in the electrochemical nature of stainless steel upon the deposition of various oxides. They demonstrated that oxides of approximate thickness of 1 lm led to decreased current densities, where oxides possessing a low reported point of zero charge (PZC)1 showed the best performance. In addition, the maximum anodic current density was studied for doublelayered oxides deposited onto stainless steel. In changing the sequence of the coatings, the passive current density was decreased for samples possessing an oxide of higher PZC near the metal surface, and an oxide with a lower PZC at the outer coating. The coating order was found to have a distinct effect on the passive current density, however no correlation was found between the critical passivation
1 Note: The definition of point of zero charge (PZC) is by definition the same as isoelectric point (IEP). IEPS denotes iso-electric point of solid.
0010-938X/$ - see front matter Crown Copyright Ó 2010 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.corsci.2010.07.028
3848
T. Markley et al. / Corrosion Science 52 (2010) 3847–3850
Fig. 1. Multielectrode array testing apparatus.
the aluminium array. The oxides were chosen due to their previously reported surface charge properties, which span the IEPS range from approximately pH 2 (SiO2) to pH 12 (MgO). The IEPS was characterised for the deposited oxides by measuring the zeta potential, f, as a function of pH by the streaming potential approach [9] and determining the pH at which f = 0. The flow of an indifferent electrolyte (0.001 or 0.01 M NaNO3 pH adjusted using HNO3 and NaOH) through a narrow capillary between two oxidecoated glass slides creates a measurable potential difference along the length of the capillary. The measurement of this potential as a function of the applied pressure enables the determination of the zeta potential (as the slope of DV/DP) according to Eq. (1):
f¼ Fig. 2. Zeta potential as a function of pH for 200 nm oxide coatings. IEPS is determined as the pH where f = 0.
Table 1 IEPS data for various oxides. Oxide
IEPS (this study)
IEPS (literature) [10]
SiO2 TiO2 Al2O3 MgO
pH pH pH pH
pH 0.5–3.7 pH 3.5–6.7 pH 5–9.4 pH 12
3.0 3.6 7.4 12.3
current and the PZC and no thickness values were reported for the bi-layer coatings. Recent work by Macdonald and coworkers, in expanding upon the point defect model (PDM), has also shown the importance of the oxide bi-layer in controlling passivation phenomena [8]. Whilst recognizing the bi-layer structure of oxides, the original PDM only considered the role of a thin-barrier layer at the metal interface. The mechanisms by which bi-layered oxides influence corrosion phenomena are not totally understood at present, but as we demonstrate in this work, can be used to good advantage and are amenable to implementation by industry. The present work demonstrates that the deposition of bi-layer coatings of differing IEPS may offer new opportunities to engineer protective coatings based upon two controlling factors: surface charge and thickness.
2. Material and surface charge characterisation Physical vapour deposition was utilised to deposit a circular array of aluminium metal electrodes onto a double glass-slide substrate (as seen on the right image of Fig. 1). Multilayer combinations of metal oxides were subsequently deposited over
DV gk DP e0 er
ð1Þ
where P = pressure (Pa), V = streaming potential (V), er = dielectric constant (80.2), e0 = relative permittivity (8.854 1012 J m1), g = solution viscosity (Pa s) and k = conductivity (S m1). Determined values of the f and IEPS in this study (see Fig. 22) in comparison to reported values in literature are given in Table 1. 3. Electrochemical characterisation and discussion The barrier properties of each oxide system were characterised by measuring their breakdown potential. The breakdown potential was determined by performing potentiodynamic polarization scans (1 VSCE to +1.5 VSCE at 1 mV s1) simultaneously for each electrode in the 30-electrode circular array. Each coating combination was placed over at least 15 individual electrodes, enabling a rapid assessment of the variation in the coating performance. Fig. 3 shows a sample of raw polarization data for oxide coatings over a small number of electrodes. The data shown as red is that of 6 replicate 100 nm MgO coatings on aluminium, demonstrating that breakdown potentials for MgO coatings were reproducible at 0.72 VSCE for this particular experiment. The black data shows that the addition of a 100 nm SiO2 coating, deposited over the top of the MgO coating significantly increased the breakdown potential. It is also noticeable in this instance that the spread in the breakdown potential is significantly increased upon the deposition of the additional coating. Reasons for the increase in the spread are presently unknown and will be examined in future works. The dotted line shows the average breakdown potential for the aluminium array in the absence of an additional oxide coating. In the case shown in Fig. 3 the addition of 100 nm MgO layer decreased the measured pitting potential, however, in Fig. 4 we demonstrate that the deposition of 100 nm layers of the various oxides 2 Note that values of f are unusually high for silica given that 0.01 M NaNO3 was used for this determination. Errors in the magnitude f however do not influence the position of IEPS and therefore the value of IEPS = 3.0 are thought to be accurate.
T. Markley et al. / Corrosion Science 52 (2010) 3847–3850
3849
Fig. 3. Potentiodynamic polarization data for bi-layer coatings of MgO and SiO2. Red: 100 nm MgO. Black: 100 nm MgO/100 nm SiO2. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 4. Measured breakdown potential of 100 nm oxides deposited onto aluminium electrode arrays plotted as a function of the IEPS determined using streaming potential.
was shown to influence the breakdown potential according to the measured IEPS. These findings are in agreement with previous works of McCafferty [3] but to our knowledge have not been reported for physically deposited oxides. Interestingly, for the case where the aluminium array was tested in the absence of any additional oxide, the breakdown potential coincided with that where a 100 nm aluminium oxide was deposited. This suggests that the 100 nm aluminium
Fig. 5. Breakdown potential plotted as a function of the difference in IEPS between inner oxide (IEPS1) and outer oxide (IEPS2). Error bars indicate standard deviation.
oxide coating did not offer any additional protection to the underlying aluminium. However, where the deposited aluminium oxide layer was increased to 200 nm the breakdown potential was shown to increase to +0.07 VSCE. In fact, for each of the oxides studied an increase in thickness to 200 nm led to increased breakdown potentials. For instance, 200 nm MgO coatings exhibited an average breakdown +0.17 VSCE. Also of interest is the ability of multilayered oxides to increase measured breakdown potential. Fig. 5 demonstrates that the combination of oxides with varying IEPS has a profound effect on the breakdown potential. The barrier properties of the deposited multilayers were found to increase where the two oxides had greater differences in their respective IEPS’s. Further, in all cases where the oxide adjacent to the metal (IEPS1) exceeded that of the oxide outer layer (IEPS2) by greater than four pH units, the barrier properties were found to increase. Fig. 6 demonstrates that improved barrier properties were found to occur when 100 nm oxide layers with high IEPS are deposited adjacent to aluminium and the low IEPS oxide layers are exposed to the aqueous environment. The results of this study demonstrate the potential advantage of engineering protective systems that have a bi-layer with controlled
Fig. 6. Breakdown potential (z-axis) plotted against IEPS1 (inner oxide) and IEPS2 (outer oxide).
3850
T. Markley et al. / Corrosion Science 52 (2010) 3847–3850
which will further suppress ion transport. If this last mechanism is correct it would follow that increasing the electric field at the bi-layer interface should lead to greater suppression of corrosion. In the context of this paper, higher fields at the interface would occur where the difference in the IEPS of the two oxides is the greatest. Acknowledgement Dr. Patrick Hartley is gratefully acknowledged for assistance with streaming potential measurements. References Fig. 7. Schematic demonstrating possible mechanism for creating high-performance barriers against corrosion.
surface charge properties. Fig. 7 shows the likely mechanism leading to the improved barrier properties of a multilayer oxide where IEPS1 > IEPS2. A high IEPS oxide near a metal surface is able to adsorb protons into cation vacancies, decreasing the ability of metal atoms to fill cation vacancies in the oxide. A low IEPS oxide exposed to the electrolyte decreases the ability of chloride anions to adsorb onto the surface of the oxide, hence are less likely to penetrate through the oxide [11–13]. This mechanism opens up new research opportunities to design coatings for corrosion control, where both oxide–oxide and oxide–polymeric bi-layers could be utilised to provide barriers against the mass transport of both cations and anions. The advantages of developing bipolar layers have been previously discussed by Sato [5]. While the mechanisms of formation of the bipolar layer are quite different to those of the present work, the mechanisms that control the bipolar performance may be similar. In addition to showing that bipolar layers could suppress anodic current flow, Sato also suggested that a high electric field would exist at the interface between the two layers, which can lead to dissociation of water and then deprotonation of the bipolar layer
[1] Z. Szklarska-Smialowska, Mechanism of pit nucleation by electrical breakdown of the passive film, Corros. Sci. 44 (2002) 1143–1149. [2] P.C. Pistorius, G.T. Burstein, Metastable pitting corrosion of stainless steel and the transition to stability, Philos. Trans. R. Soc. Lond. 341 (1992) 531–559. [3] E. McCafferty, A surface charge model of corrosion pit initiation and of protection by surface alloying, J. Electrochem. Soc. 146 (1999) 2863–2869. [4] Y. Hayashi, M. Masuda, J. Lee, M. Kojima, Study on corrosion properties of sputter coating on oxides of stainless steels, Mater. Sci. Eng. A198 (1995) 71– 74. [5] N. Sato, Toward a more fundamental understanding of corrosion processes, Corrosion 45 (1989) 354–368. [6] N. Sato, Interfacial ion-selective diffusion layer and passivation of metal anodes, Electrochim. Acta 41 (1996) 1525–1532. [7] N. Sato, The potentials of mixed electrodes of corrodible metal and metal oxide, Corros. Sci. 42 (2000) 1957–1973. [8] J. Ai, Y. Chen, M. Urquidi-Macdonald, D.D. Macdonald, Electrochemical impedance spectroscopic study of passive zirconium, J. Electrochem. Soc. 154 (2007) C43–C51. [9] J. Lyklema, Fundamentals of Interface and Colloid Science: Solid–Liquid Interfaces, Academic Press, San Diego, 1995 (Chapter 4). [10] G.A. Parks, The isoelectric points of solid oxides, solid hydroxides, and aqueous hydroxo complex systems, Chem. Rev. 65 (1965) 177–198. [11] T.H. Muster, I.S. Cole, The protective nature of passivation films on zinc: surface charge, Corros. Sci. 46 (2004) 2319–2335. [12] T.H. Muster, I.S. Cole, Corrigendum to: the protective nature of passivation films on zinc: surface charge, Corros. Sci. 47 (2005) 1861–1862. [13] P.M. Natishan, W.E. O’Grady, E. McCafferty, D.E. Ramaker, K. Pandya, A. Russell, Chloride uptake by oxide covered aluminium as determined by X-ray photoelectron and X-ray absorption spectroscopy, J. Electrochem. Soc. 146 (1999) 1737–1740.
Contents lists available at ScienceDirect
Corrosion Science journal homepage: www.elsevier.com/locate/corsci
Letter
Multilayered coatings: Tuneable protection for metals T. Markley a, S. Dligatch b, A. Trinchi a, T.H. Muster a,*, A. Bendavid b, P. Martin b, D. Lau a, A. Bradbury a, S. Furman a, I.S. Cole a a b
CSIRO Division of Materials Science and Engineering, Private Bag 33, Clayton South, Victoria 3169, Australia CSIRO Division of Materials Science and Engineering, P.O. Box 218, Lindfield, NSW 2070, Australia
a r t i c l e
i n f o
Article history: Received 6 July 2010 Accepted 25 July 2010 Available online 2 August 2010 Keywords: A. Aluminium B. Polarization C. Electrochemical calculation C. Oxide coatings C. Passive films
a b s t r a c t The role of oxide bi-layers in controlling the onset of corrosion has been explored. A high-throughput electrochemical approach was employed to determine the breakdown potential of aluminium metal over-coated with combinations of silicon, titanium, aluminium and magnesium oxides. Bi-layered coatings consisting of two 100 nm thick metal oxide layers provided increased protection against breakdown, and combinations with vastly different iso-electric point of solid (IEPS) were found to exhibit improved barrier properties in comparison to single-component oxides. Furthermore, the most protective oxide bilayers were produced when a high IEPS oxide was deposited directly onto the metal surface and subsequently over-coated with a low IEPS oxide. The barrier properties of bi-layer coatings appear to be tuneable, with notable dependencies on surface charge and thickness. Crown Copyright Ó 2010 Published by Elsevier Ltd. All rights reserved.
1. Introduction All industrially relevant metals are thermodynamically unstable and rely upon the addition or formation of protective coatings to slow the kinetics leading to oxidation. Alloys that rely upon passive oxide layers, such as stainless steel and aluminium, tend to suffer from pitting corrosion at the sites of localised breakdown of the protective oxide. There are numerous works describing the importance of chloride ion transport through oxides and passive films and their role in nucleating pitting events [1,2]. Previous studies have also suggested that the surface charge of the oxide can alter the protective nature of the oxide [3]. The formation of oxide bilayers has also been shown to influence the barrier properties [4–7]. Previous work by McCafferty demonstrated that the pitting potential of oxide-covered aluminium in chloride-containing solutions can be affected by the iso-electric point of the oxide (IEPS) [3]. This led to a proposed model for pit initiation that relied upon the surface possessing a positive charge that would tend to attract chloride anions, thus leading to their transported to the metal interface via oxygen vacancies in the oxide. The initiation of pitting was therefore proposed by McCafferty to be controlled by three factors: (1) the oxide surface charge, (2) the heat of adsorption of chloride ions onto the oxide, and (3) the density of oxygen vacancies in the oxide coating.
* Corresponding author. Tel.: +61 3 95452874; fax: +61 3 95441128. E-mail address: [email protected] (T.H. Muster).
Sato [6] also explored the influence of oxide surface charge on controlling the transport of charged species through oxide layers. Oxide precipitation of metal multivalent oxyanions (i.e. metal phosphates/molybdates) can lead to fixed negative charges and the oxide becoming cation selective. In contrast, the precipitation of metal monovalent complexes (i.e. hydroxychlorides) lead to fixed positive charges and anion selectivity. In this manner Sato proposed that the specific adsorption of multivalent anions to the oxide–solution interface imposes a bipolar oxide structure. The bipolar structure possessing an anion-selective inner layer and cation-selective outer layer was demonstrated to block anodic ion transport. Hayashi et al. [4] studied changes in the electrochemical nature of stainless steel upon the deposition of various oxides. They demonstrated that oxides of approximate thickness of 1 lm led to decreased current densities, where oxides possessing a low reported point of zero charge (PZC)1 showed the best performance. In addition, the maximum anodic current density was studied for doublelayered oxides deposited onto stainless steel. In changing the sequence of the coatings, the passive current density was decreased for samples possessing an oxide of higher PZC near the metal surface, and an oxide with a lower PZC at the outer coating. The coating order was found to have a distinct effect on the passive current density, however no correlation was found between the critical passivation
1 Note: The definition of point of zero charge (PZC) is by definition the same as isoelectric point (IEP). IEPS denotes iso-electric point of solid.
0010-938X/$ - see front matter Crown Copyright Ó 2010 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.corsci.2010.07.028
3848
T. Markley et al. / Corrosion Science 52 (2010) 3847–3850
Fig. 1. Multielectrode array testing apparatus.
the aluminium array. The oxides were chosen due to their previously reported surface charge properties, which span the IEPS range from approximately pH 2 (SiO2) to pH 12 (MgO). The IEPS was characterised for the deposited oxides by measuring the zeta potential, f, as a function of pH by the streaming potential approach [9] and determining the pH at which f = 0. The flow of an indifferent electrolyte (0.001 or 0.01 M NaNO3 pH adjusted using HNO3 and NaOH) through a narrow capillary between two oxidecoated glass slides creates a measurable potential difference along the length of the capillary. The measurement of this potential as a function of the applied pressure enables the determination of the zeta potential (as the slope of DV/DP) according to Eq. (1):
f¼ Fig. 2. Zeta potential as a function of pH for 200 nm oxide coatings. IEPS is determined as the pH where f = 0.
Table 1 IEPS data for various oxides. Oxide
IEPS (this study)
IEPS (literature) [10]
SiO2 TiO2 Al2O3 MgO
pH pH pH pH
pH 0.5–3.7 pH 3.5–6.7 pH 5–9.4 pH 12
3.0 3.6 7.4 12.3
current and the PZC and no thickness values were reported for the bi-layer coatings. Recent work by Macdonald and coworkers, in expanding upon the point defect model (PDM), has also shown the importance of the oxide bi-layer in controlling passivation phenomena [8]. Whilst recognizing the bi-layer structure of oxides, the original PDM only considered the role of a thin-barrier layer at the metal interface. The mechanisms by which bi-layered oxides influence corrosion phenomena are not totally understood at present, but as we demonstrate in this work, can be used to good advantage and are amenable to implementation by industry. The present work demonstrates that the deposition of bi-layer coatings of differing IEPS may offer new opportunities to engineer protective coatings based upon two controlling factors: surface charge and thickness.
2. Material and surface charge characterisation Physical vapour deposition was utilised to deposit a circular array of aluminium metal electrodes onto a double glass-slide substrate (as seen on the right image of Fig. 1). Multilayer combinations of metal oxides were subsequently deposited over
DV gk DP e0 er
ð1Þ
where P = pressure (Pa), V = streaming potential (V), er = dielectric constant (80.2), e0 = relative permittivity (8.854 1012 J m1), g = solution viscosity (Pa s) and k = conductivity (S m1). Determined values of the f and IEPS in this study (see Fig. 22) in comparison to reported values in literature are given in Table 1. 3. Electrochemical characterisation and discussion The barrier properties of each oxide system were characterised by measuring their breakdown potential. The breakdown potential was determined by performing potentiodynamic polarization scans (1 VSCE to +1.5 VSCE at 1 mV s1) simultaneously for each electrode in the 30-electrode circular array. Each coating combination was placed over at least 15 individual electrodes, enabling a rapid assessment of the variation in the coating performance. Fig. 3 shows a sample of raw polarization data for oxide coatings over a small number of electrodes. The data shown as red is that of 6 replicate 100 nm MgO coatings on aluminium, demonstrating that breakdown potentials for MgO coatings were reproducible at 0.72 VSCE for this particular experiment. The black data shows that the addition of a 100 nm SiO2 coating, deposited over the top of the MgO coating significantly increased the breakdown potential. It is also noticeable in this instance that the spread in the breakdown potential is significantly increased upon the deposition of the additional coating. Reasons for the increase in the spread are presently unknown and will be examined in future works. The dotted line shows the average breakdown potential for the aluminium array in the absence of an additional oxide coating. In the case shown in Fig. 3 the addition of 100 nm MgO layer decreased the measured pitting potential, however, in Fig. 4 we demonstrate that the deposition of 100 nm layers of the various oxides 2 Note that values of f are unusually high for silica given that 0.01 M NaNO3 was used for this determination. Errors in the magnitude f however do not influence the position of IEPS and therefore the value of IEPS = 3.0 are thought to be accurate.
T. Markley et al. / Corrosion Science 52 (2010) 3847–3850
3849
Fig. 3. Potentiodynamic polarization data for bi-layer coatings of MgO and SiO2. Red: 100 nm MgO. Black: 100 nm MgO/100 nm SiO2. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 4. Measured breakdown potential of 100 nm oxides deposited onto aluminium electrode arrays plotted as a function of the IEPS determined using streaming potential.
was shown to influence the breakdown potential according to the measured IEPS. These findings are in agreement with previous works of McCafferty [3] but to our knowledge have not been reported for physically deposited oxides. Interestingly, for the case where the aluminium array was tested in the absence of any additional oxide, the breakdown potential coincided with that where a 100 nm aluminium oxide was deposited. This suggests that the 100 nm aluminium
Fig. 5. Breakdown potential plotted as a function of the difference in IEPS between inner oxide (IEPS1) and outer oxide (IEPS2). Error bars indicate standard deviation.
oxide coating did not offer any additional protection to the underlying aluminium. However, where the deposited aluminium oxide layer was increased to 200 nm the breakdown potential was shown to increase to +0.07 VSCE. In fact, for each of the oxides studied an increase in thickness to 200 nm led to increased breakdown potentials. For instance, 200 nm MgO coatings exhibited an average breakdown +0.17 VSCE. Also of interest is the ability of multilayered oxides to increase measured breakdown potential. Fig. 5 demonstrates that the combination of oxides with varying IEPS has a profound effect on the breakdown potential. The barrier properties of the deposited multilayers were found to increase where the two oxides had greater differences in their respective IEPS’s. Further, in all cases where the oxide adjacent to the metal (IEPS1) exceeded that of the oxide outer layer (IEPS2) by greater than four pH units, the barrier properties were found to increase. Fig. 6 demonstrates that improved barrier properties were found to occur when 100 nm oxide layers with high IEPS are deposited adjacent to aluminium and the low IEPS oxide layers are exposed to the aqueous environment. The results of this study demonstrate the potential advantage of engineering protective systems that have a bi-layer with controlled
Fig. 6. Breakdown potential (z-axis) plotted against IEPS1 (inner oxide) and IEPS2 (outer oxide).
3850
T. Markley et al. / Corrosion Science 52 (2010) 3847–3850
which will further suppress ion transport. If this last mechanism is correct it would follow that increasing the electric field at the bi-layer interface should lead to greater suppression of corrosion. In the context of this paper, higher fields at the interface would occur where the difference in the IEPS of the two oxides is the greatest. Acknowledgement Dr. Patrick Hartley is gratefully acknowledged for assistance with streaming potential measurements. References Fig. 7. Schematic demonstrating possible mechanism for creating high-performance barriers against corrosion.
surface charge properties. Fig. 7 shows the likely mechanism leading to the improved barrier properties of a multilayer oxide where IEPS1 > IEPS2. A high IEPS oxide near a metal surface is able to adsorb protons into cation vacancies, decreasing the ability of metal atoms to fill cation vacancies in the oxide. A low IEPS oxide exposed to the electrolyte decreases the ability of chloride anions to adsorb onto the surface of the oxide, hence are less likely to penetrate through the oxide [11–13]. This mechanism opens up new research opportunities to design coatings for corrosion control, where both oxide–oxide and oxide–polymeric bi-layers could be utilised to provide barriers against the mass transport of both cations and anions. The advantages of developing bipolar layers have been previously discussed by Sato [5]. While the mechanisms of formation of the bipolar layer are quite different to those of the present work, the mechanisms that control the bipolar performance may be similar. In addition to showing that bipolar layers could suppress anodic current flow, Sato also suggested that a high electric field would exist at the interface between the two layers, which can lead to dissociation of water and then deprotonation of the bipolar layer
[1] Z. Szklarska-Smialowska, Mechanism of pit nucleation by electrical breakdown of the passive film, Corros. Sci. 44 (2002) 1143–1149. [2] P.C. Pistorius, G.T. Burstein, Metastable pitting corrosion of stainless steel and the transition to stability, Philos. Trans. R. Soc. Lond. 341 (1992) 531–559. [3] E. McCafferty, A surface charge model of corrosion pit initiation and of protection by surface alloying, J. Electrochem. Soc. 146 (1999) 2863–2869. [4] Y. Hayashi, M. Masuda, J. Lee, M. Kojima, Study on corrosion properties of sputter coating on oxides of stainless steels, Mater. Sci. Eng. A198 (1995) 71– 74. [5] N. Sato, Toward a more fundamental understanding of corrosion processes, Corrosion 45 (1989) 354–368. [6] N. Sato, Interfacial ion-selective diffusion layer and passivation of metal anodes, Electrochim. Acta 41 (1996) 1525–1532. [7] N. Sato, The potentials of mixed electrodes of corrodible metal and metal oxide, Corros. Sci. 42 (2000) 1957–1973. [8] J. Ai, Y. Chen, M. Urquidi-Macdonald, D.D. Macdonald, Electrochemical impedance spectroscopic study of passive zirconium, J. Electrochem. Soc. 154 (2007) C43–C51. [9] J. Lyklema, Fundamentals of Interface and Colloid Science: Solid–Liquid Interfaces, Academic Press, San Diego, 1995 (Chapter 4). [10] G.A. Parks, The isoelectric points of solid oxides, solid hydroxides, and aqueous hydroxo complex systems, Chem. Rev. 65 (1965) 177–198. [11] T.H. Muster, I.S. Cole, The protective nature of passivation films on zinc: surface charge, Corros. Sci. 46 (2004) 2319–2335. [12] T.H. Muster, I.S. Cole, Corrigendum to: the protective nature of passivation films on zinc: surface charge, Corros. Sci. 47 (2005) 1861–1862. [13] P.M. Natishan, W.E. O’Grady, E. McCafferty, D.E. Ramaker, K. Pandya, A. Russell, Chloride uptake by oxide covered aluminium as determined by X-ray photoelectron and X-ray absorption spectroscopy, J. Electrochem. Soc. 146 (1999) 1737–1740.