Stainless steels, renowned for their corrosion resistance due to a protective chromium oxide layer, are widely used in various applications. However, in chloride-containing acidic environments, with or without electrochemical bias, severe corrosion can occur. While the corrosion mechanisms of stainless steels have been extensively studied, a comprehensive nanoscale, spatially resolved, and quantitative understanding has only recently emerged. In situ atomic force microscopy (AFM), particularly electrochemical AFM (EC-AFM), provides high spatial resolution and allows for manipulation of solution parameters, making it ideal for studying corrosion processes at solid-liquid interfaces. Previous studies have utilized in situ AFM and EC-AFM to investigate the polishing process, corrosion under external stress, and the early stages of corrosion initiation in various steels, showing the preferential corrosion at grain boundaries and strain-hardened areas. However, a comprehensive quantitative understanding of the interplay between corrosion kinetics and applied potential remains needed. The corrosion susceptibility of austenitic stainless steels is dependent on crystallographic orientation, with (111) and (100) planes generally exhibiting higher resistance. The use of a high-purity model Fe-Cr-Ni alloy allows for a fundamental understanding of corrosion mechanisms by eliminating the complexities introduced by multiple alloying elements and metallurgical defects. This study uses in situ liquid EC-AFM combined with ex situ aberration-corrected STEM/TEM analysis to investigate the chloride corrosion kinetics and subsurface chemical diffusion in a model Fe-18Cr-14Ni alloy.

The literature review extensively covers previous research using in situ AFM and EC-AFM to study corrosion in various steels. Studies highlighting the role of grain boundaries, crystallographic orientation, and the influence of applied potential on corrosion rates are discussed. The authors emphasize the lack of quantitative, nanoscale, spatially resolved data needed to develop accurate predictive models for corrosion kinetics. Previous research has established the importance of crystallographic orientation and surface energy on corrosion behavior, showing that planar orientations (111) and (100) generally exhibit higher pitting corrosion resistance. The importance of understanding the early stages of corrosion initiation and the role of chloride ion adsorption in the formation of passive films and subsequent dissolution are also acknowledged. The limitations of previous studies in providing comprehensive, quantitative data on the effects of applied bias and the interplay between surface kinetics and diffusion are highlighted.

Fe-18Cr-14Ni alloys were fabricated via arc melting, casting, and homogenization, followed by cold rolling and annealing. Microstructural characterization was performed using SEM, EBSD, and synchrotron XRD, confirming a fully stabilized FCC austenite structure. In situ EC-AFM experiments were conducted using a Nanoscope 8 AFM in contact mode, employing a three-electrode system with a 0.5 M deuterium chloride (DCI) solution in D2O. Two samples were prepared: a nonbiased sample and a biased sample. For the nonbiased sample, a 5 µm × 20 µm area was scanned continuously for 1.5 h. The biased sample was initially corroded for 30 min before applying a -0.5 V bias for an additional hour, with a 10 µm × 40 µm area continuously scanned. After the in situ AFM experiments, SEM, site-specific sample preparation, and STEM/TEM analysis were performed using an aberration-corrected JEOL ARM200CF microscope to characterize the microstructure and composition of corrosion pits. Specific AFM parameters such as scan rate, number of lines, and aspect ratio are explicitly stated. The methods for sample preparation, including metallographic polishing and the use of fiducial marks, are also described in detail. The use of a CH Instrument Model 600E Series Electrochemical Workstation for bias control is noted.

In the nonbiased sample, vertical pit depth increased linearly with time, indicating a surface kinetics/diffusion hybrid mechanism. Lateral pit width growth followed a power law with an exponent close to 0.5, suggesting a diffusion-controlled process. Pits initiated preferentially at grain boundaries. The biased sample showed a significant increase in the corrosion rate primarily due to increased pit nucleation, rather than the lateral growth of existing pits. The applied bias did not significantly alter the vertical dissolution rate. Quantitative analysis of pit depth, width, and number density was performed, and power-law fitting parameters are reported for both the nonbiased and biased samples. TEM analysis revealed a dual-layer oxide scale with an outer Fe-rich layer and an inner Cr-rich layer, with Ni segregation near the metal-oxide interface. The [323] grain, close to the [111] orientation, exhibited the highest dissolution rate, both with and without bias. The linear relationship between total pitting area and corrosion time is observed for both biased and non-biased samples. The rate constants derived from the fitting of the data before and after the application of bias are reported separately in the Table 2. This finding shows a significant impact of potential bias on corrosion.

The findings show that the corrosion mechanism shifts from a combined nucleation and lateral growth in the nonbiased case to a predominantly nucleation-driven process under bias. This shift is attributed to changes in the surface concentration of ionic species, with a decreased population of negatively charged species like Cl- and an increased population of H+ under negative bias potentially inhibiting lateral growth while promoting nucleation. The faster dissolution rate of the [323] grain, near the [111] orientation, highlights the influence of crystallographic orientation on corrosion kinetics, consistent with previous findings in the literature. The observation of a continuous Cr2O3 oxide layer with underlying Ni segregation provides insights into the subsurface chemical changes during corrosion.

This study provides unprecedented insights into the corrosion kinetics of a model Fe-Cr-Ni alloy using in situ EC-AFM, revealing a shift in corrosion mechanisms driven by applied potential. The nonbiased corrosion is characterized by a surface kinetics/diffusion hybrid mechanism for vertical growth and a diffusion-controlled mechanism for lateral growth. Applying a negative bias shifts the mechanism to one dominated by increased pit nucleation. The results emphasize the importance of both crystallographic orientation and applied potential in determining corrosion behavior. Future studies could explore the effects of different ionic species, sample deformation, and various potentials to further refine our understanding of corrosion mechanisms and improve the predictive design of corrosion-resistant alloys.

The study focused on a model Fe-Cr-Ni alloy. The findings may not be directly generalizable to all commercial stainless steels which contain additional alloying elements and potentially have different microstructures or impurity levels that would affect the corrosion behavior. The experiments were conducted at room temperature, and the corrosion behavior may differ at elevated temperatures. While the study provides quantitative data, further research is needed to develop more sophisticated predictive models for corrosion kinetics across a broader range of conditions.