Viacheslav Shkirskiy

Transient surface states can control the reactivity of stainless steel yet remain invisible to conventional post-mortem characterization because they may disappear as soon as electrochemical control is interrupted. Here, in situ reflective microscopy (RM) was combined with electrochemistry and matched atomic emission spectroelectrochemistry (AESEC) measurements to follow surface film evolution on 316L stainless steel. The surface progressively darkened during anodic polarization, but release from the vertex potential to open-circuit potential produced an abrupt brightening. Under the matched electrochemical sequence, AESEC detected selective Cu dissolution after potential release corresponding to a 0.4 nm CuO-equivalent layer. A multilayer Fresnel model containing a retained mixed-oxide passive film, in which CuO leaves the interface, reproduces the magnitude of the brightening under the selected optical assumptions. Residual-charge calculations admit mixed Fe-valence scenarios consistent with the optical thickness estimate. Together with thermodynamic calculations, these results support selective removal of a transient Cu-containing surface component. Such short-lived states can escape detection during transfer and post-mortem analysis, which motivates coupled real-time monitoring.

In spite of substantial progress in localized methods, no reliable in situ spatially resolved technique exists to measure conversion coating thickness at individual intermetallic phases. Here, dynamic in situ reflective microscopy (RM) was combined with post-mortem cross-sectional transmission electron microscopy (TEM) to resolve spontaneous Zr conversion coating (ZrCC) growth on S-phase (Al₂CuMg) particles in AA2024 exposed to 200 ppm H₂ZrF₆. Normalized intensity images were interpreted using Fresnel equations and estimated optical constants to infer local coating thickness. RM measurements across nine micrometer-sized particles on the same surface showed nearly identical optical responses. TEM performed on a different S-phase particle from the same sample provided an independent representative benchmark, with measured thicknesses in close agreement with RM-derived values. The TEM-supported time-resolved RM analysis revealed distinct precipitation kinetics: immediate fast growth on the S-phase, an induction period on the Al matrix, and ultimately convergent growth rates consistent with a self-limiting mechanism. Overall, this work establishes a spatially and temporally resolved approach for quantifying ZrCC formation and provides a framework for extension to other conversion coating systems and intermetallics.

Corrosion is a trajectory, not an event, yet current machine learning maps fixed inputs to outputs and cannot model reliably how a corroding interface responds to intervention. We argue that world models, deep learning frameworks that learn latent physical dynamics from observation sequences, can close this gap. A physically interpretable latent space anchored to corrosion mechanisms can enable prediction, counterfactual reasoning, and discovery. We chart a path toward realising it.

L'inhibition de la corrosion d'un galvanisé par des agents encapsulés dans des hydroxydes doubles lamellaires (HDL) : méchanismes de lixiviation et réactions de corrosion: Le travail présenté essaie de comprendre les mécanismes de l’action d’un inhibiteur de corrosion présent dans un revêtement hybride sous forme de pigments intercalés dans les hydroxydes double lamellaires (HDL) pour la protection de l’acier galvanisé. Trois étapes clés ont été choisies pour ce travail : (1) l’identification d’un inhibiteur de corrosion hydrosoluble pour l’acier galvanisé avec une compréhension de sa réactivité, (2) la détermination des facteurs et des mécanismes contrôlant la libération de l’inhibiteur à partir d’HDL et (3) la compréhension des mécanismes de protection dans un système modèle avec le revêtement hybride contrôlé par la libération de l’inhibiteur et la réactivité d’inhibiteur. MoO42- a montré la meilleure efficacité d'inhibition comparable à CrO42- dans des solutions alcalines et neutres. L’effet inhibiteur de MoO42- a été associé à la formation d’un film riche en Mo(V). L'effet de cet anion sur la dissolution de l'acier à bas carbone a été également vérifié pour exclure la possibilité d'un effet d'accélération des espèces choisies. Les tests de lixiviation ont montré que la libération de MoO42- à partir d’HDL a été contrôlée par la nature des ions échangés à partir du support par un mécanisme d'échange d'ions à un pH neutre et par la dissolution du cadre de la LDH à un pH alcalin. La présence de seulement Cl- conduit à moins de 40% de libération de MoO42- après 24 h d'immersion alors que les additions des carbonates ont abouti à libération de 100% après 1 h. Les tests d'immersion ont montré léger effet d'inhibition du système de revêtement dans Cl- et une augmentation dans CO32- en accords avec le niveau plus élevé de MoO42- libéré.

Localized detection of hydrogen permeation in steel membranes is crucial for practical applications but remains challenging. We present a reflective microscopy (RM) approach combined with machine learning (ML)-driven image analysis to address this issue. Hydrogen permeation in press-hardened steel alters the Fe(II)/Fe(III) ratio in the surface layer, affecting reflected light intensity due to changes in the refractive index. Although initially near the optical detection limit, this subtle contrast was enhanced using time-series analysis with UMAP and K-means clustering. The detected regions corresponded to hydrogen-affected zones confirmed by Scanning Kelvin Probe measurements. This work establishes a novel RM-ML framework for detecting and quantifying hydrogen permeation, offering a promising and accessible alternative to conventional methods for localized analysis of metallic membranes.

Localized detection of hydrogen permeation in steel membranes is crucial for practical applications but remains challenging. We present a reflective microscopy (RM) approach combined with machine learning (ML)-driven image analysis to address this issue. Hydrogen permeation in press-hardened steel alters the Fe(II)/Fe(III) ratio in the surface layer, affecting reflected light intensity due to changes in the refractive index. Although initially near the optical detection limit, this subtle contrast was enhanced using time-series analysis with UMAP and K-means clustering. The detected regions corresponded to hydrogen-affected zones confirmed by Scanning Kelvin Probe measurements. This work establishes a novel RM-ML framework for detecting and quantifying hydrogen permeation, offering a promising and accessible alternative to conventional methods for localized analysis of metallic membranes.