Intelligent Materials Design for Real-Time Corrosion Monitoring Systems

Thesis Abstract

This study addresses the escalating challenge of corrosion-induced failures in critical infrastructure by developing an intelligent materials design framework that enables real-time monitoring and adaptive corrosion mitigation. The aim is to integrate multifunctional smart coatings and embedded sensors within metallic substrates to provide continuous corrosion state assessment, early fault detection, and autonomous protection strategies. Specific objectives include (i) synthesizing and characterizing self-healing, corrosion-sensing coatings based on nanostructured halloysite nanotubes and polymeric matrices; (ii) developing a digital twin framework that fuses electrochemical, environmental, and mechanical data for real-time corrosion prognosis; (iii) validating sensorized samples under accelerated chloride-rich and high-temperature environments; (iv) establishing predictive models linking sensor signals to corrosion rate and type, and (v) formulating design guidelines for scalable deployment in pipelines and offshore structures. A mixed-methods methodology is employed in three stages. First, an experimental phase tests 120 coated steel specimens (AISI 1018) with varying smart-coating formulations, subjected to accelerated corrosion conditions (3.5 wt% NaCl, 60°C) for up to 2000 hours, with electrochemical impedance spectroscopy (EIS), linear polarization resistance (LPR), and localized micropropagation mapping performed at 100-hour intervals. Second, a data-driven phase constructs a digital twin by integrating electrochemical data, environmental sensors (temperature, humidity, salinity), and mechanical load histories using a Bayesian network and Kalman filtering to provide real-time corrosion state estimates. Third, a theoretical phase applies reliability-centered maintenance (RCM) and the Theory of Planned Behavior to model user interaction with monitoring systems, informing adoption and deployment scenarios. Data collection instruments include potentiostats for EIS and LPR measurements, high-resolution scanning vibrating electrode technique (SVET) for current density mapping, fiber-optic distributed temperature sensing (DTS) along test racks, and embedded wireless IoT sensors for environmental variables. Material characterizations employ X-ray diffraction (XRD), transmission electron microscopy (TEM), scanning electron microscopy (SEM) with energy-dispersive X-ray spectroscopy (EDS), and nanomechanical testing to assess coating integrity and self-healing efficacy. Data analysis utilizes multivariate regression to relate sensor outputs to corrosion rates, machine learning algorithms (random forest, gradient boosting) for feature selection and prognosis, and survival analysis to quantify time-to-failure distributions. Model validation includes k-fold cross-validation and sensitivity analyses to evaluate robustness under varying environmental scenarios. Theoretical grounding integrates the Electrochemical Theory of Corrosion, the Self-Healing Polymer Concept, and the Digital Twin paradigm, with reference to the Diffusion–Adsorption–Reaction mechanism for ion transport in porous coatings. Expected findings indicate that optimally formulated self-healing coatings reduce corrosion rate by 40–65% relative to conventional coatings under chloride exposure, while embedded sensors provide corrosion-state predictions with a mean absolute error below 12% for 24-hour ahead forecasts. The digital twin is anticipated to deliver real-time maintenance recommendations with 85–92% predictive accuracy across test conditions, and the integrated system is expected to extend time-to-failure by 1.5–2.5× under accelerated aging. The study will identify key sensor modalities and their synergistic contributions to early corrosion signals, establishing thresholds for alarming and automated mitigation actions. The study contributes to knowledge by (i) advancing a holistic intelligent material design that couples smart coatings with real-time sensing and prognosis, (ii) delivering a validated digital twin framework for corrosion monitoring in metallic assets, and (iii) generating design guidelines for scalable deployment in oil and gas pipelines, offshore platforms, and maritime infrastructure. It also articulates methodological best practices for integrating materials science experiments with data analytics, machine learning, and reliability engineering within a coherent, theory-informed framework. The principal conclusion posits that intelligent materials design, when coupled with robust data-driven prognostics and a digital twin, markedly enhances the visibility, predictive capability, and proactive management of corrosion, thereby lowering lifecycle costs and improving asset resilience. Recommendations include standardized sensor integration protocols, field-testing across diverse environments, development of open data formats to enable cross-platform interoperability, and policy guidance for adopting intelligent monitoring systems in critical infrastructure.