The corrosion and passivity of iron (and carbon steel) in media with different alkalinity as well as iron depassivation have been studied extensively using electrochemical methods and nano-scale surface characterization studies. The electrochemical techniques provide valuable information about the average electrochemical behavior of relatively large metal surfaces, typically in centimeter-square scale or larger. However, electrochemical processes on these surfaces typically occur in nanometer scale. Although nano-scale surface characterizations provide valuable information about passive films that form on carbon steel and iron in alkaline electrolytes, and their chloride-induced depassivation, they cannot explain the dynamic processes that lead to their formation as they can only provide data at particular instants. Atomistic modeling techniques such as Reactive Force Field Molecular Dynamics (ReaxFF-MD) have the potential to fill this gap. This research used ReaxFF-MD simulations to answer fundamental questions on iron corrosion, passivation and chloride-induced depassivation in four interrelated thrusts:
Thrust 1: ReaxFF-MD was used to study the initial stages of iron corrosion in pure water. The simulations were performed on iron under various applied external electric fields and temperatures. Oxide film formation was accompanied by iron dissolution in water, indicating active corrosion and supporting the expected thermodynamic behavior of iron in pure water. Oxide film thickness and iron dissolution increased with increasing applied external electric field. Corrosion rates increased slightly with increasing temperature within the temperature range of this investigation (300–350 K). Critical stages of the iron corrosion process during the simulations were identified as dissociation of water to OH- and H+, adsorption of OH- on the iron surface, penetration of oxygen into iron to form iron oxides, and dissolution of iron into solution. Comparisons of the simulated charge distributions and pair distribution functions to those of reference oxides showed the formed oxide compositions were not pure phases, but rather a mixture of oxides.
Thrust 2: The applicability of the classical EDL models was investigated to study corrosion and passivity of iron in neutral and alkaline media using ReaxFF-MD. The performance of the classical EDL models were studied in the ReaxFF-MD simulations of iron exposed to neutral (pH = 7) and highly alkaline (pH = 13.5) electrolytes under applied electric fields. Although the Helmholtz model was able to produce iron corrosion in the neutral electrolyte, it did not result in passive film formation in the highly alkaline solution system. The Gouy-Chapman model was not capable of simulating passivity for iron in the highly alkaline solution system or active corrosion in the neutral electrolyte. The Stern model was the only model that could simulate passivity and corrosion of iron for highly alkaline and neutral electrolytes, respectively. This study showed that ReaxFF-MD simulations of iron in neutral and alkaline electrolytes should use the Stern model for representing the EDL.
Thrust 3: The passivity of Fe(110) in a 0.316 M NaOH solution (pH = 13.5) was investigated using ReaxFF-MD. The simulations were carried out under an applied electric field of 30 MeV/cm. The electrical double layer was modeled using the Stern model. Under these conditions, following the expected thermodynamic behavior, a protective passive film formed during 500 ps simulation time. The initial stages of passivation differed from simulations that had been carried out in neutral electrolytes such that the highly alkaline environment allowed the stabilization of the metal through the formation of an Fe(OH)2 layer on the metal surface. This created conditions for oxygen diffusion into metal without dissolution of iron atoms into the electrolyte. The passive film had a multiple oxide structure such that outer layers were in the form of Fe2O3, middle layers were in the form of Fe3O4, and the inner layer was in the form of FeO. This multi-layer structure is in agreement with theoretical passivity models that are based on an inner barrier layer that forms directly on the metal substrate (FeO), and outer layers (Fe3O4 and Fe2O3) that precipitates through the further oxidation of the iron ejected from the inner layer. A parallel XPS investigation confirmed the findings of ReaxFF-MD simulations.
Thrust 4: Chloride-induced depasivation of iron in pH 13.5 NaOH solution was studied using ReaxFF-MD, electrochemical tests and x-ray photoelectron spectroscopy (XPS). The breakdown of the passive film by chlorides initiates with iron dissolution from the first layer of the passive film into the electrolyte. Iron dissolution and corresponding iron vacancy formation in the first layer of the passive film take place in four stages that involves local acidification of the electrolyte adjacent to the metal surface, followed by iron dissolution into the electrolyte in the form of Fe(OH)Cl2 and FeCl3. Chloride in the electrolyte mainly acts as a catalyst and do not penetrate into the passive film. The four-step process for the initiation of the passive film breakdown was used to explain the concept of a critical chloride threshold and the well-documented electrochemical observation that critical chloride thresholds are higher in solution with solution with higher pH. The ReaxFF-MD simulations support the depassivation hypothesis that is described by the point defect model. Both ReaxFF-MD simulations and XPS analysis showed that chlorides increase the Fe+3/Fe+2 of the passive film, and this increase is more evident in the inner and middle layers of the film.
It is expected that these fundamental findings of this research will facilitate the development of new corrosion mitigation strategies such as customized corrosion inhibitors and inexpensive corrosion-resistant steels.