Steel corrosion, costing trillions globally, frequently manifests as localized corrosion, which is difficult to predict and detect. Understanding localized corrosion kinetics is crucial for designing safer structures. While in situ microscopy techniques offer direct observation of morphological changes, they often can't penetrate corrosion product layers. In situ X-ray computed tomography (XCT) is superior but suffers from high-energy radiation affecting reaction kinetics and long measurement times. Electrochemical techniques, like using wire beam or scanning electrochemical microscopy, lack accuracy or distort the corrosion process. This study presents an advanced ultrasonic technique that overcomes these limitations, offering direct, non-intrusive, and corrosion product-immune 3D morphology monitoring. The authors' previous work demonstrated an ultrasonic technique for uniform corrosion monitoring with 10s of nm resolution; however, this new research extends it to localized corrosion by using a 2D multi-element piezoelectric transducer array to track thickness loss of micro-sections, enabling reconstruction of the entire substrate morphology.

The paper reviews several existing methods for in-situ monitoring of corrosion, highlighting their limitations. Techniques based on STM, AFM, confocal laser scanning microscopy, and optical microscopy struggle to see through corrosion product layers. XCT, while superior in some ways, is limited by its long measurement times and the potential impact of high-energy X-rays on corrosion kinetics. Electrochemical methods utilizing wire beam electrodes or scanning electrochemical microscopy (e.g., scanning reference electrode technique and scanning vibrating electrode technique) also have limitations related to accuracy and potential distortion of the corrosion process. The authors position their ultrasonic method as a superior alternative, addressing the shortcomings of previous techniques.

The proposed ultrasonic technique uses a 2D multi-element piezoelectric transducer array permanently attached to the non-working surface of the corrosion substrate. Each element tracks thickness loss of a micro-section. The instantaneous 3D morphology is reconstructed by interpolating thickness losses from all micro-sections. Lateral resolution is customizable, depending on transducer element dimensions, ultrasonic wave frequency, and substrate thickness; axial resolution is 100 nm. Experiments were conducted on NAK80 steel substrates in various electrolytes (NaCl solution, NaCl + NaOH solution, and NaCl + HCl solutions of various pH levels). Three different corrosion cells with varied electrode configurations (carbon bar, copper plates, copper mesh) were employed to induce different morphological evolution processes. The thickness (Th) of each micro-section is calculated using the formula Th = V(T)(ToA₂ - ToA₁)/2, where V(T) is the temperature-dependent shear wave velocity, and ToA₁ and ToA₂ are the time-of-arrival of the first and second reflected wave packets. The impact of ultrasound on corrosion kinetics was verified in separate experiments. Measurement accuracy was assessed using white light interferometry, and the results are compared with numerical predictions generated using finite element method (FEM) simulations based on the applied DC and electrode configuration. Further experiments were conducted to explore the role of corrosion products by analyzing corrosion product precipitates using X-ray diffraction (XRD).

Experiments revealed that in alkaline environments, the corrosion rate initially accelerates then decelerates. This behavior is linked to the formation of Fe3O4, which consumes electrons. The ultrasonic method accurately reconstructs 3D morphology, matching well with optical measurements (6.77% mean difference). The transient surges in corrosion rate coincide with transient drops in ultrasonic signal amplitude, attributed to corrosion product layer deposition. In highly acidic environments, this transient behavior is absent, confirming the link between Fe3O4 formation and the observed corrosion rate fluctuations. The experiments systematically investigated the influence of pH on the corrosion process, demonstrating that the formation of corrosion products and their subsequent transformation play a key role in the observed corrosion kinetics. The study provides direct, quantitative measurements of the corrosion process across different micro-sections, revealing spatial variations in corrosion rates and highlighting the complex interaction between metal dissolution, corrosion product formation, and the overall morphology.

The study successfully addresses the research question by demonstrating the capability of the new ultrasonic technique to accurately and non-intrusively monitor localized corrosion processes in real-time. The findings highlight the importance of considering corrosion product formation and transformation when modeling corrosion kinetics. The observed transient surges in corrosion rate, linked to Fe3O4 formation, have not been previously captured with the same level of detail by other in-situ techniques. The high precision and temporal resolution of the ultrasonic method provide valuable insights into the complex interplay of electrochemical and morphological changes during corrosion. The agreement between ultrasonic reconstructions and optical measurements validates the accuracy and reliability of the technique.

The research successfully developed a high-precision, in situ 3D ultrasonic imaging technique for localized corrosion monitoring. This technique surpasses existing methods in its ability to penetrate corrosion product layers and accurately measure morphological changes at high resolution. The results show a correlation between transient surges in corrosion rate, the formation of Fe3O4, and the deposition/collapse of the corrosion product layer. Future research could explore the application of this technique to a wider range of materials and corrosion environments, as well as further investigate the underlying mechanisms of localized corrosion.

The accuracy of the ultrasonic method might be affected in regions with steep surface gradients. The study primarily focused on NAK80 steel; further research is needed to evaluate the generalizability of the findings to other materials. While the authors showed that ultrasound does not affect the corrosion kinetics, further research into this area could strengthen the conclusions. The numerical predictions employed simplified assumptions; more sophisticated models may be needed to fully capture the complexity of the corrosion processes.