Printer Friendly

Surface analytical techniques become a valuable tool in corrosion research.

Road salt is seen to be the enemy, but the precise role of chloride ion in the corrosion process is not fully understood

Corrosion results from the chemical or electrochemical reaction between a material and its environment. All materials (metals, plastic, composites, ceramics) corrode to some degree. This article will consider only metallic corrosion and oxidation, although many of the surface-analytical techniques discussed can be applied to study the degradation of non-metals.

Metallic corrosion can take many forms, e.g. uniform attack, localized corrosion or stress corrosion cracking. The structure and composition of the |passive' oxide film which forms in aqueous in environments on all metals except gold is important in governing corrosion resistance. Metals become passive as a result of this oxide which can form when the potential is raised above a critical value, as indicated for metallic iron, Fig. 1. Once this protective film breaks down, corrosion is initiated and a corrosion product forms.

Modern surface-analytical techniques can provide useful information regarding the nature and chemical composition of these passive oxide films, leading to a better understanding of the corrosion processes that occur on an atomic scale. These techniques also provide useful information regarding transport processes in oxides grown at high temperature. Surface techniques can be of two-types: ex situ methods (where samples are removed from solution or a reaction vessel and placed in ultra-high vacuum systems), such as Auger electron spectroscopy (AES), secondary ion mass spectrometry (SIMS) and X-ray photoelectron spectroscopy (XPS), and in situ methods, such as infrared (IR) and Raman spectroscopy, Table 1 shows a summary of the first three techniques together with the information they provide. These three techniques have been widely used to examine passive films and corrosion products, whereas in situ techniques have been used more to study the role of corrosion inhibitors. This article will also concentrate on ex situ methods, and will illustrate the application of the various techniques to corrosion and oxidation research studies performed in our laboratory. Over the years our efforts have focused on the nature of passivity, its breakdown by aggressive ions such as chloride, and the transport processes which take place during oxide growth at high temperatures.

The |passive' oxide

film on iron

The nature of this common oxide film has been the subject of investigation since the days of Michael Faraday. While SIMS, Auger, XPS, etc., are ideally suited to examine the 1.6-nm-thick film, its precise nature and chemical composition are still a matter of some controversy. Our view is that the film is crystalline and consists, of an inner magnetite ([Fe.sub.3][O.sub.4]) layer and an outer magnetite ([gamma]-[Fe.sub.2][O.sub.3]) layer, Fig. 1, of which the outermost portion is cation deficient.

Others consider the film to be an amorphous gel with bridging di-oxy and di-hydroxy bonds. Surface-analytical data from our laboratory support the [Fe.sub.3][O.sub.4]/[gamma]-[Fe.sub.2][O.sub.3] duplex model[1]. In particular, SIMS can be used to detect the presence of H-containing species and the lack of hydroxyl ions within passive films has been confirmed in such experiments[2]. Figure 2 shows experimental SIMS data for a passive film together with |dry' [Fe.sub.2][O.sub.3] and |wet' FeOOH (lepidocrocite) standards. The film is removed by ion sputtering and the [FeO.sup.+] and [FeOH.sup.+] signals are measured through the film (depth profiling). The profile for the passive oxide film is very similar to that for the |dry' [Fe.sub.3] [O.sub.2] standard until the oxide/metal interface is reached after about

7 minutes of sputtering, (For |wet' FeOOH. the [FeOH.sup.+] signal is always higher than that of [FeO.sup.+.]) The hydroxyl content within the film as calculated from the SIMS data is zero ([+ or -] 0.1%); only a fraction of a monolayer of OH is adsorbed on the oxide surface as seen by the initially higher [FeOH.sup.+] signal. Conversion electron (back-scattered) Mossbauer spectroscopy (EMS) is much more surface sensitive than the conventional transmission technique and has been used to examine passive oxide films on iron. The data can be interpreted in terms of the film being a small particle size crystalline oxide, and both CEMS and complementary XPS data support the model that the passive film resembles [gamma]-[Fe.sub.2] [O.sub.3] [3].

Questions are sometimes raised regarding possible changes which may occur in passive films when specimens are removed from solution and then pumped in an ultra-high vacuum system for SIMS. Auger or XPS analysis. Performing electrochemical experiments in [O.sup.18]-containing solution and then monitoring the [O.sup.18] level of films using SIMS is a good method to determine the air stability of passive films. If films are unstable, they will pick up [O.sup.18] from the air, and their [O.sup.18] content will be less than that of the solution enrichment.

Passive films on iron (except at very low potentials in the passive range shown in Fig. 1) are found to be stable in the air. Rather surprisingly, films on iron-chromium alloys (more like stainless steels) do change on removal from solution, and the extent of change increases with increasing chromium content of the alloy [4]. [O.sup.18] /SIMS can also be used to indicate whether, for example, airformed films can be removed by a cathodic treatment to leave a bare metal surface prior to performing an electrochemical experiment to produce the passive films described above.

The technique can also be used to study the mechanism of growth of films at different potentials. Films can be formed at one potential in the passivee range, Fig. 1 and then the potential increased to a higher value still in the passive region, for additional growth in an [O.sup.18]-enriched solution. A subsequent SIMS profile through the oxide shows the [O.sup.18.] to be a maximum at the outer surface and then decreases throughout the entire thickness of the film [1]. This result indicates that the additional oxide has grown by inward oxygen transport through the film, likely via both lattice and grain boundary sites. (Oxide growth studies using [O.sup.18]/SIMS will be described later in more detail for high temperature corrosion.)

Role of chloride ion in

pit initiation

From a practical viewpoint, the influence of road salt on automobile corrosion has dramatic consequences, yet the precise role of the chloride ion in the corrosion process is not fully understood. For example, there is considerable disagreements as to whether chloride ion is incorporated into passive films as a precursor for the initiation of pitting corrosion of the metal. We have used both AES and SIMS to determine the extent of any chloride ion incorporated into films formed on nickel, iron and iron-chromium alloys.

As seen by the Auger profile, Fig. 3(a), passive films formed on nickel in 1M chloride solution are found to contain up to 4% chloride ion [5], whereas films formed on iron do not incorporate chloride ion even when formed in chloride-containing solution [6]. Chloride ion is also incorporated into films on iron-chromium alloys after electron polishing a surface in a perchloric/acetic acid mixture.

Figure 3(b) shows the Auger profile of such a film. Quantitative layer-by-layer Auger analysis [7] shows that chloride is distributed over the outer six layers of the 10 layer-thick oxide. This quantitative analysis is compared to the experimental data, in Fig. 3(b). From the analysis it is concluded that the inner part of the film is chromium-rich and the outer part is iron-rich. The presence of incorporated chloride in oxide films influences the open-circuit breakdown behavior of both nickel and iron-chromium alloys.

However, experiments with nickel show that the incorporated chloride is not a precursor for pit initiation, and that nickel is more resistant to pitting when chloride is incorporated in the film. The precise events which give rise to breakdown of passive films in chloride solution are still being investigated by many groups around the world.

The performance of materials at high temperature is often controlled by the protective qualities of the surface oxide which forms. It is important, therefore, to understand the oxide growth processes in order to develop more corrosion-resistant coatings or alloys. SIMS is ideally suited to examine these growth processes, in particular the extent of inward oxygen transport compared with outward cation transport through the oxide scale [8]. By this technique, in addition to examining to [O.sup.18] and [O.sup.16] signals, the negative polyatomic species M [O.sup.18.sub.2], M [O.sup.16.sub.2] and the isotopically mixed M [O.sup.18] [O.sup.16] can also be analyzed after sequential oxidation of a material in [O.sup.18.sub.2] and [O.sup.16.sub.2]. An example of such a polyatomic SIMS profile is shown for a thick oxide grown on a (100) single crystal chromium surface at 825[degrees] C, Fig. 4(a), (1.1 [mum] of oxide was produced in [O.sup.18.sub.2] followed by an outer 0.3 [mum] in [O.sup.16.sub.2]). It is clear from the profile that the latter oxide is on the outside of the scale, showing that the major transport process during oxidation is outward chromium diffusion.

However, the Cr [O.sup.18] [O.sup.16] profile is asymmetric about the [O.sup.18]-oxide/[O.sup 16]-oxide interface with an obvious oxygen penetration into the inner oxide. This is considered to be due to some inward oxygen diffusion down oxide grain boundaries. The amount of oxide created in this way within the existing [.sup 18O]-containing film, is however, small, accounting for only 1% of the total new oxide. There is also a small rise in the [Cr.sup 18 O.sup 16 O] and [Cr.sup 16 O.sub 2] signals in the vicinity of the metal/oxide interface which corresponds to the zone where the original 2-nm-thick, [O.sup 16]-oxide film formed during surface pretreatment would be situated.

Expansion of the [Cr.sup 16O.sub 2] profile confirms a small but clearly distinct peak in this region, Fig. 4(b), for polycrystalline chromium oxidized at 825 [degrees] C. This result demonstrates that the [sup.18 O]/SIMS technique is sufficiently sensitive to allow a relatively thin layer of oxide to be detected underneath a very much thicker external layer. This thin prior oxide film, in fact, serves as an inert marker confirming oxide growth primarily by outward cation diffusion from the metal to the outer oxide surface, and formation of new oxide mainly on top of the existing scale.

The addition of small amounts of so-called reactive elements such as cerium and yttrium increases the high-tempperature oxidation resistance of chromium and iron-chromium alloys. We have applied very thin (4nm) reactive element coatings by sputtering ceria (CeOsub.2) or ytttria ([Ysub.2 O sub.3]) directly onto chromium or iron-chromium substrates, and have used SIMS to determine the location of the reactive element within the oxide scale and to study its effect on transport processes [9]. For examples, SIMS profiles of a thin oxide (76 nm-thick) formed after short-term oxidation at 900 [degrees] of iron-26% chromium coated with 4 nm of ceria are shown in Fig. 5(a).

The cerium in the oxide layer is detectable and its maximum signal is observed to be away from the alloy/oxide interface. With prolonged oxidation the location of the cerium maximum moves out further towards the oxide/gas interface.

If the ceria coating can be considered to be a stationery marker, its location towards the oxide/gas interface indicates that inward oxygen diffusion is becoming more prominent in governing oxide growth. [Osup.18]/SIMS experiments confirm that oxygen transport is occuring. Figure 5(b) shows data for a 0.36 [micro]m-thick oxide formed sequentially in [16 sup.O].2 and then in [sup.18 O sub.2] at 900 [degrees] C. In this longer-term experiment the location of the cerium maximum is now in middle of the scale. There are three [O sup.18] maxima ([Cr sup.53 O sup.18 sup.+]): at the gas/oxide interface, within the scale coincident with the cerium maxima, at the alloy/oxide interface showing that oxygen diffusion has occurred during oxidation.

Examination of SIMS data. Fig. 5(b), in the form of the fraction of [O sup.18] associated with the cerium- and chromium-bearing species through the scale shows that a higher fraction is associated with the cerium-bearing species. If the cerium in the scale is located at oxide grain boundaries, the association of [O sup.18] with cerium strongly suggests that transport of [O sup.18] through the scale occurs along grain boundariees. Therefore, coating iron-chromium alloys (or chromium metal) with a small amount of ceria causes a change in oxide growth mechanism from predominantly cation diffusion for uncoated material, Fig. 4(a) to predominantly anion diffusion via oxide grain boundaries, Fig. 5(b).

SIMS lacks the spatial resolution to determine the precise location of the cerium or its form and distribution within the chromium oxidee layer. However, techniques such as transmission electron microscopy (TEM) with electron energy loss (EELS) analysis [10] can be used to identify any cerium-chromium oxide phases formed at oxide grain boundary regions. Figure 6 shows both an EELS image (a) and bright field image (b) from the same area of a 44 nm-thick oxide formed on ceria-coated iron-26% chromium at 900[degrees] and stripped from the substrate. Figure 6(a) shows the distribution of cerium-containing oxide as the light colored regions.

X-ray analysis in TEM shows the oxide particles to have a cerium to chromium ratio

1 suggesting that the particles are the oxide [CeCrOsub.3], and this has been confirmed by electron diffraction measurements [11]. It is likely that these particles and, perhaps cerium ions, retard the grain boundary diffusion of chromium cations to the extent that the mobility of oxygen anions along these grain boundaries exceeds that of the cations.

EELS microscopy provides complementary information to AES and SIMS data regarding metal corrosion and oxidation processes because of its high spatial resolution. One of the main disadvantages of techniques like AES and SIMS is their relatively poor spatial resolution compard with TEM. However, developments continue to improve the resolution of surface techniques. For example, with a liquid metal ion gun, it is now possible to obtain SIMS images with a spatial resolution better than 50 nm. This can be imagined from the SIMS images, Fig. 7, taken from an oxide formed at 1100 [deegrees] C, first in [sup.16 O sub.2] and then in [sup.18 O sub.2], on an iron-25% aluminum alloy.

Figure 7 (a) shows an [O sup.18] image near the outer oxide surface. The white patches
represent 50% [O sup.18], the white lines
 25% [O sup.18]: the grey areas 2% [O sup.18] the

black areas those where patches of oxide have spalled off. As we profile through the oxide towards the alloy surface, the white [O sup.18] rich patches disappear, some of the lines disappear and the low level in the grey areas in maintained. In addition, some areas of oxide containing 50% [O sup.18] appear as the oxide/alloy interface is approached. (Figure 7(b) shows an [O sub.16] image taken near the oxide/alloy interface indicating the high level of [O sup.16]. The [O sup.18] distribution is localized and non-uniform, and the [O sup.18] regions near the oxide/alloy interface are consistent with oxygen short-circuit diffusion and formation of new oxide grains at the alloy surface. Clearly, the SIMS images illustrate that growth mechanism of alumina ([alpha]-[Alsub.2 O sub.3] is much more complex than simply oxygen grain boundary diffusion.


[1.] M.J. Graham, J.A. Bardwell, R.Goetz, D.F. Mitchell and B. MacDougall, Corros. Sci., 31, 139 (1990). [2.] D.F. Mitchell and M.J. Graham. Electrochem. Soc., 133. 956 (1986). [3.] M.E. Brett. K.M. Parkin and M.J. Graham. J. Electrochem. Soc. 133, 2032 (1986). [4.] J.A. Bardwell, G.I. Sproule, D.F. Mitchell, B. MacDougall and M.J. Graham, J. Chem. Soc. Farad. Trans. 87, 1011 (1991). [5.] B. MacDougall, D.F. Mitchell, G.I. Sproule and M.J. Graham. J. Electrochem. Soc., 130, 543 (1983). [6.] R. Goetz. B. MacDougall and M.J. Graham. Electrochimica Acta, 31, 1299 (1986). [7.] D.F. Mitchell, Appl. Surf. Sci., 9, 131 (1981). [8.] R.J. Hussey, D.F. Mitchell and M.J. Graham, Werkstoffe und Korrosion, 38. 575 (1987). [9.] R.J. Hussey, P. Papaiacovou, J. Shen, D.F. Mitchell and M.J. Graham. Proc. Symp. on Corrosion and Particle Erosion at High Temp., Ed., V. Srinivasan and K. Vedula, The Minerals, Metals and Materials Soc., p. 567 (1989). [10.] D.F. Mitchell and M.J. Graham. Electrochem. Soc., 133, 2433 (1986). [11.] D.A. Downham, R.J. Hussey, D.F. Mitchell and M.J. Graham. Proc. Symp. on High Temp. Oxidation and Sulphidation Processes. 29th Annual Conf. of Metal-lurgists, Hamilton, Ontario, August 1990. Pergamon Press, Proc., Vol. 22. p. 101 (1990).
COPYRIGHT 1992 Chemical Institute of Canada
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Graham, M.J.
Publication:Canadian Chemical News
Date:Jul 1, 1992
Previous Article:Double your disk space - double your fun!
Next Article:W(h)ither chemical engineering.

Related Articles
The bugs of rust; researchers are trying to tease out how microorganisms influence the corrosion of metals.
Taking corrosion's magnetic pulse.
Appropriate levels necessary for reduction of corrosion.
Simple laser-based pipeline corrosion assessment system.
Practical outcomes, benefits of MIC research program.
What's New In Corrosion Control Products.
Identification of a smart bond coating for gas turbine engine applications.
Influence of surface cleaning by tools made of copper alloys on the painting performance.
Improvement of the properties of AZ91D magnesium alloy by treatment with a molten [AlCl.sub.3]-NaCl salt to form an Mg-Al intermetallic surface layer.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters