Corrosion is a branch of materials research that is, by its nature, “schmutzy” (dirty). It has a reputation as an unglamorous subject and certainly suffers from an image problem in the materials community: the guest editors would like to have a dollar for every time we sat down and discussed “How can we make corrosion sound positive?” Solving corrosion problems requires an understanding of corrosion mechanisms, and corrosion mechanisms are infinitely subtle and interesting. Corrosion processes can often encompass concepts from a variety of disciplines, including thermodynamics, surface science, electrochemistry, physical metallurgy, and solid-state and statistical physics. While corrosion of metals is the subject of this issue of MRS Bulletin, ceramics, polymers, and geological materials are all greatly affected by corrosion.

Corrosion of metals is an electrochemical process, composed of anodic (metal oxidation) and cathodic (reduction) reactions occurring on a surface. The rates of these reactions are measured as current densities (in units of A m−2). On a single piece of metal, the anodic and cathodic current densities can vary arbitrarily from point to point, as long as the total anodic and cathodic currents are equal. This variation, which is associated with local differences in pH, surface composition, or defects in a surface film, leads to various localized phenomena, such as pitting and stress-corrosion cracking, and also to intriguing oscillatory dynamics.

The electrochemical dissolution of clean, oxide-free metal surfaces in electrolyte solutions is a fundamental process in metal corrosion as well as in technological etching and metal-refinement processes. Due to fundamental interest and for industrial reasons, dissolution processes and the effect of organic corrosion inhibitors have been studied intensely, concentrating on the determination of macroscopic etch rates and the effect of varying chemical parameters upon the etch kinetics. In the 1920s, ideas for the microscopic mechanism of the dissolution and the reverse deposition reaction had already been put forward by Stranski, who suggested that dissolution (growth) is dominated by detachment (attachment) of atoms at low coordination sites of the crystal surface and at kink sites along monoatomically high steps (see the schematic model in Figure 1). These kink sites are preserved during the removal (or addition) of the basic building blocks of the crystal (e.g., single metal atoms for elemental metals) and hence can sustain a continuous dissolution reaction. However, the direct verification of these ideas was impossible for more than seven decades, due to the very complex (defect) structure of the surfaces of typical metal electrodes, the presumably dominating effects of these defects on the etch process, and the lack of structurally sensitive probes suitable for in situ structural characterization.

The corrosion properties of alloys are of enormous practical importance: modern life would be very different without stainless steels. Alloy corrosion is also an intriguing field of scientific study that combines electrochemical kinetics with fashionable aspects of the morphological evolution of surfaces and even a dash of ancient history, via the studies of Forty and Lechtman on “depletion gilding” practiced by Early Andeans during pre-Columbian times in South America.

The basic alloy corrosion process, as used by the metalsmiths to gold-coat artifacts, is de-alloying. This is defined as the selective electrolytic dissolution of one or more components from a metallic solid solution. For this to happen, there must be a significant difference in the equilibrium metal/metal-ion electrode potentials for the two metals, taking into account any complex ions that might be formed in the electrolyte. For example, we can expect de-alloying in Au-Cu alloys, but not in Au-Pt alloys.

De-alloying shows sharp parting limits, expressed as critical atom percentages of the more reactive component above which that component can be removed from the alloy by electrochemical dissolution in an oxidizing environment such as nitric acid. Parting limits range from about 20 at.% to 60 at.%. This concept is still used in noble metal technology to separate noble metals from base metals. For example, an alloy of 55 at.% gold and 45 at.% silver does not de-alloy, but if it is re-melted with additional silver so that the atom fraction of Ag is greater than 60%, the gold can be separated almost completely by nitric acid immersion.

Most metals are thermodynamically unstable in aqueous environments. They are kinetically stabilized by the formation of very thin “passive” oxide films that provide protection from corrosion. The technological importance of passive films has led to widespread investigation of their structure and chemistry. However, despite decades of work since Michael Faraday first described the existence of the passive film on iron, the atomic structure of these films is still poorly understood. This is because the films are so thin (typically only a few nanometers thick), making measurements difficult, and because the structure can only be investigated in the (wet) electrochemical environment in which these films form.

The electrochemical nature of the formation of these oxide films is shown in Figure 1, which presents the current response of an iron electrode in an aqueous solution to an applied voltage (potential). As the potential is slowly increased from negative (cathodic) to positive (anodic), the iron first begins to dissolve. This area of dissolution is known as the active region, indicating that the iron is corroding freely as ferrous ions (Fe2+). Then, at a sufficiently anodic (oxidizing) voltage, a thin (30 Å) oxide film forms at the iron surface, preventing further dissolution. This results in a marked decrease in the measured current, a phenomenon known as passivity. Over the passive region, the current is virtually independent of applied voltage and remains extremely low, indicating that the iron is protected from additional oxidation or corrosion.

Intergranular separation in polycrys-talline materials involves breaking metallic bonds along grain boundaries in response to stress. The surfaces created in this manner expose the grain facets on either side of the original boundary, as shown in Figure 1. This mode of fracture often occurs at much lower fracture stress and energy than cracking by ductile processes through the interior of grains. The exposure of specific materials to certain environments and stress can promote this low-energy, intergranular mode of separation, even when fracture of the same material in vacuum would occur along a ductile transgranu-lar path. Three types of environment-assisted intergranular cracking can occur in a wide variety of alloy/environment systems: intergranular stress-corrosion cracking (IGSCC), intergranular hydrogen embrittlement, and intergranular liquid-metal embrittlement.

Figure 1 shows an example of IGSCC. This type of cracking is a pervasive problem in many technological applications, leading to extensive repairs, loss of service function, and safety concerns. IGSCC occurs in the weld-heat-affected zones of stainless-steel pipes in high-purity primary coolant waters within nuclear power plants, and in nickel-based alloys utilized as heat-exchanger tubing when exposed to the high-purity primary as well as secondary coolant waters in power plants. It is also seen in Al-based alloys used for fuselage skins and structural components in military and commercial aircraft when exposed to humid atmospheric conditions. Ferrous alloys used in the oil and gas industry are also susceptible. For instance, IGSCC of mild steels used in buried gas-transmission pipelines is a widespread international problem, leading to explosions when leaking natural gas ignites.

Metal/polymer composites are used in numerous technical applications. For example, polymer coatings on metal surfaces are used for corrosion protection, metal films on polymers inhibit static buildup, and polymers between two metals can serve as a “glue” for connecting materials that cannot be welded. Polymer/metal composites also play an important role in modern electronics. In condensers, polymers serve as insulating layers between metallic leads and are used to encapsulate entire electronic circuits. In all circumstances, interfaces are formed between the two different materials, and since the chemistry and structure change abruptly, interfacial failure is frequently observed.

The cause of failure may just be mechanical (e.g., shrinkage of the polymer during curing), or the interface stability may be degraded by attack of aggressive species, resulting in delamination. More specifically, loss of adhesion is directly caused by interfacial electrochemical reactions that nucleate at a defect and progress into intact regions of the interface. This occurs for encapsulated electronic parts in humid atmospheres as well as for lacquers on automotive parts.

Thus the investigation of corrosion reactions at a buried interface is an important area of research, but it is made very difficult by the fact that most electrochemical methods do not give information on localized reaction kinetics at a buried (metal/polymer) interface. This situation has changed with the invention and development of the scanning Kelvin probe (SKP). This method allows, for the first time, local analysis of reactions occurring at a buried metal/polymer interface. Based on the results obtained with the SKP, a detailed reaction model for the delamination process has been developed. This understanding has led to the development of new approaches that protect the interface from delamination. The idea is to chemically modify the interface using Afunctional molecules that promote adhesion between metal and polymer surfaces.