Reliable corrosion rate prediction is an essential aspect in design of transmission and processing infrastructure in natural gas and oil industry. Underestimation of this parameter may lead to health, safety, environmental hazards and financial losses due to corrosion failures. On the other hand, overestimations can redefine the economic feasibility of a project.

The most common corrosive scenario in oil and gas industry is the corrosion of mild steel in aqueous acidic solutions containing weak acids such as carbonic acid, hydrogen sulfide, or organic acids. In a general point of view, this corrosion process can be summarized by anodic solution of iron on one side and hydrogen evolution reaction (HER) on the other side.

The mechanism of cathodic reactions in this system has been subject of numerous studies in last four decades and in some cases it is still open to debate. Conventionally these weak acids were considered to be directly reduced at the metal surface¹, evolving hydrogen. However, more recent studies suggest that the direct reduction of these species is insignificant and they merely buffer the pH at the metal surface resulting in lower surface pH comparing to the case of strong acids^2,3. However, considering the former mechanism, Nordsveen et al. note that disregarding the direct reduction reactions the predicted corrosion rates suffer from significant inaccuracy¹. Considering that the direct reduction of these weak acids is experimentally shown to be insignificant, the failure of corrosion rate predictive models in proper corrosion rate prediction based on this mechanism could be sought in their simplistic treatment of the HER. 

While the complex behavior of the HER is well-known, the available mechanistic corrosion rate predictive models generally consider the Volmer step as the rate determining step for all hydrogen donor species, at all environmental conditions, and the whole pH range. Such an assumption has been previously confirmed for limited conditions. However, its universal expansion as done in the present models has no solid background¹.

The present work is an initial step in introducing the detailed mechanistic treatment of the HER into the corrosion rate predictive models. Here, the theoretical structure of such a model is presented. Due to complexity of corroding systems, the initial verification of this mechanistic model was performed using a gold electrode in order to eliminate the interference of the anodic reaction. However, the theoretical structure and the mathematical model developed here could be further adapted to more realistic conditions encountered in industrial applications.

The most commonly accepted mechanisms for the HER are generally discussed based on ion discharge step (Volmer), followed by electrochemical recombination (Heyrovsky) or chemical recombination (Tafel) step. Considering the atomic hydrogen coverage on the metal surface and their corresponding interaction the rate of these elementary steps can be calculated similar to what described elsewhere⁴. Using these charge transfer expressions, a comprehensive mathematical model was developed based on the same approach of Nordsveen et al.¹. 

In order to verify the mathematical model, a series of experimental data in the form of steady state voltammograms were obtained. The experiments were carries out on a three electrode cell with a gold rotating disk electrode in deaerated perchlorate acidic solutions.

The experimental data is presented in Figure 1 where two distinctive Tafel slopes of 60 mV in lower current densities and 120 mV in higher current densities is observed. This behavior is in good agreement with the previously reported data in the literature. The reaction order in this set of data is obtained based on the variation of current density with pH at a fixed potential which showed the value of -0.8.

In the final step the mathematical model was fitted to the experimental data. It was found that the only case where both Tafel slope of 60 mV and reaction orders higher than -1 could be observed is where the Tafel step is rate determining and the surface coverage of hydrogen atoms and their repulsive interaction is significant. The comparison of the model and the experimental data as presented in Figure 1 shows good quantitative and qualitative agreement, where both Tafel slops and the reaction order of -0.8 is captured by the model.

1.           Nordsveen, M., S. Nešić, R. Nyborg, and A. Stangeland, Corrosion59 (2003): pp. 443–456.

2.           Tran, T., B. Brown, S. Nešic, and B. Tribollet, Corrosion70 (2013): pp. 223–229.

3.           Remita, E., B. Tribollet, E. Sutter, V. Vivier, F. Ropital, and J. Kittel, Corros. Sci.50 (2008): pp. 1433–1440.

4.           Gennero de Chialvo, M.R., and A.C. Chialvo, Electrochim. Acta 44 (1998): pp. 841–851.