A Mechanistic Model for CO₂ Localized Corrosion of Carbon Steel

Localized corrosion of carbon steel in CO₂ environments is a long-standing challenge faced by the oil and gas industry. Extensive research efforts have been dedicated to elucidating the mechanisms governing localized corrosion of carbon steel. Recent studies have discovered pseudo-passivation of carbon steel under FeCO₃ layer-forming conditions where high saturation level of FeCO₃ and high temperature are typically involved, which appreciably increases the potential of metal. A galvanic coupling mechanism is then proposed for localized corrosion propagation of carbon steel. In this theory, FeCO₃ layer-covered surfaces are considered to undergo a substantial surface pH increase due to the mass transfer limiting effect, which could trigger the formation of pseudo-passive film and result in potential increase of metal surface. In a case where a small portion of this film-covered surface loses the protective film, an active surface that has a lower potential will be exposed. A galvanic cell can then be established between film-covered and active surfaces, which drives the active surface to corrode at a higher rate. Saturation level of FeCO₃ is found to play a critical role in determining whether a pit propagates or dies. It was found that pit propagation is predominant when the saturation level is around 1, a condition known as a "grey zone". ¹ Outside the "grey zone", pits are often captured due to formation of FeCO₃ on the pit surface.

Based on the experimental findings, a new transient mechanistic model is developed in this study to simulate the localized corrosion process of carbon steel in a CO₂ environment (MULTICORP V5). The model covers the physics governing both uniform and localized corrosion, including mass transfer, chemical reactions, electrochemical reactions, FeCO₃ layer formation, FeS layer formation (for uniform corrosion only), pseudo-passivation and pit propagation. Pit initiation is triggered using a statistical function, as mechanisms for pit initiation are still under investigation and not available at this stage of research. The model is able to provide detailed information on critical parameters involved in the corrosion process, such as water chemistry, potential and current distribution in the solution, particularly for those adjacent to the metal surface. This information will assist engineers in better understanding the corrosion process in order to make strategic decisions. The uniform corrosion model has been fully calibrated against and verified with a database that contains a large number of experimental results under various conditions in CO₂/H₂S environments. The localized corrosion model is calibrated against limited experimental data obtained from the artificial pit tests in CO₂ environments. Parametric study is performed for localized corrosion. It is shown that model predictions quantitatively match experimental results and qualitatively agree with the general understanding of the localized corrosion process.

To disseminate the knowledge and raise the level of understanding of corrosion within the broader range of the corrosion community, a stand-alone electrochemical model is developed using Microsoft Excel VBA. The model, named FREECORP, is exclusively based on the public literature and offered free of charge. The model is capable of predicting steady-state CO₂ and/or acetic acid (HAc) uniform corrosion and transient H₂S uniform corrosion for carbon steel. Apart from corrosion rate prediction, polarization curves can also be predicted for individual and overall electrochemical reactions in order to enhance understanding of corrosion mechanisms. In the case of H₂S corrosion, the concentration profile of H₂S across the inner and outer makinawite film and bulk solution is shown in order to give more meaningful information about H₂S corrosion. This model is written with the concept of object-oriented programming (OOP) which provides great flexibility to model calculations. Any reactions, including system-defined and user-defined reactions, can be added or removed from the system, a feature that allows for investigation into the effect of individual reactions on the corrosion process. This feature also permits the expansion of the model into a much wider range of environments than the model was originally designed for. The current model is fully calibrated and verified with a large number of in-house experimental data contained in the ICMT (Institute for Corrosion and Multiphase Flow Technology at Ohio University) database.