(372h) Understanding and Modeling Corrosion in CO2 Transportation Pipelines: Findings from the Carbon Adapt Project

Corrosion in pipelines has a huge economic impact on the pipeline infrastructure for CO₂ injection. Corrosion can reduce the wall thickness of pipelines, leading them to crack and corrode, causing uncontrolled leaks or damage to the entire pipeline system. The transportation pipelines and injection wells play a key role, as they serve as the connector between carbon capture and its utilization and storage (CCUS). The overarching goal of the Carbon Adapt project is to assess the impact of CO₂ injection on current oil and gas infrastructure or the installation of new ones for CO₂ applications. This project aims to cultivate detailed knowledge of the interaction between CO₂ (liquid or supercritical), impurities (e.g., water, H₂S, NO_x), process conditions, reservoir fluids, and the pipeline material required, all within the context of CCUS in the Danish North Sea. The study will measure the corrosion effect of a liquid or supercritical CO₂ in the presence of impurities and provide a mathematical model to predict corrosion on existing and new infrastructure. The investigation of the CO₂ impurity corrosion effect will provide fundamental and necessary information to frame future quality guidance and legislation for CO₂ injection, which is currently lacking. The study aims to discern between CO₂ and H₂S dominant corrosion processes, presenting polarization curves for CO₂ dominance and H₂S concentration profiles relative to the distance from the steel surface. Importantly, this study, in its latter stages, will try to validate corrosion conditions from the flow-loop testing facility, allowing to draw comparison with experimental conditions and testing the accuracy of the developed model. The goal is to allow user customization by enabling the inclusion of user-defined reactions or the exclusion of undesired reactions, thereby offering flexibility in the corrosion calculation process.

Experiments with varying impurity concentrations are being carried out at a flow-loop facility specifically designed for corrosion testing in a liquid and supercritical CO₂ environment. The flow-loop testing facility simulates corrosion conditions similar to those expected in CO₂ transportation pipelines and injection wells. These experiments provide data to study the effect of different impurities of CO₂ streams for injection, flow conditions, and reservoir fluids on materials, both on the existing type of wells and for new pipelines, in order to obtain a clear correlation between the observed corrosion and the impurities.

A mechanistic model is being developed and tested to predict the corrosion rate induced by CO₂, H₂S, and organic acids, specifically targeting the internal corrosion of mild steel pipelines. The primary objective of this model is to develop a straightforward and efficient theoretical model for assessing corrosion in CCUS systems. Comprehensive scientific literature has been used to populate the model, encompassing underlying theories, calibration data, and limitations, which have been studied to improve the model.

This study presents the ability to forecast uniform corrosion rates induced by various corrosive agents, such as CO₂, H₂S, and organic acids. Furthermore, it simulated capabilities for the growth of iron carbonate and iron sulfide films. Additionally, the study identified predominant corrosive species by quantifying their respective contributions.

Since transportation and injection of captured CO₂ is an essential part of any CCUS strategy, the project's impact is significant, as improper material use and corrosion prediction can lead to uncontrolled releases of CO₂, as well as of toxic impurities contained in it, during transportation or when storing CO₂ filled with brine forming a highly corrosive environment. The ability to model corrosion inducing environments will lead to enhanced operation and reduced maintenance cost and loss due to corrosion, thus having a positive economic impact on CCUS infrastructure.