Chloride-induced corrosion of concrete-steel tubular columns (CSTC) : a perspective structure for offshore wind turbines

Abstract

The use of steel monopiles for offshore wind turbine structures is gaining popularity due to their lightweight nature. However, steel exposed to oceanic environments is susceptible to chloride-induced corrosion, especially in the tidal and splash zone. Repairing and applying protective coating inside marine structures is tedious and economically unviable. Hence, a relatively new column form, called concrete-steel-tubular-column (CSTC), is proposed to protect the steel column from corrosion by providing an outer concrete layer. However, the concrete cover made with plain concrete could hardly meet these stringent requirements. Henceforth, a challenging task is to propose a CSTC with superior durability performance inside marine environments and a service life prediction model that can realistically predict chloride-induced corrosion processes. The corrosion propagation stage plays a vital role in the service life design of CSTCs. Generally, the chloride-induced corrosion initiation time is considered the end of the service life. Even though cover cracking and crack propagation time contribute significantly to the estimation of service life, especially for fibre-reinforced concrete (FRC). The main objective of this thesis herein is to investigate the chloride-induced corrosion performance of CSTCs, prepared with self-prestressing ultra-high performance concrete (UHPC), slag-based geopolymer concrete (GPC) with different metakaolin replacement levels and predict a service life prediction model which incorporates the corrosion propagation stage. The corrosion rate measurements of CSTCs with various novel concrete mixtures and cover thicknesses were evaluated using the impressed current density method (ICDM). Results showed that adding expansive agent (EA) and steel fibres (SF) into UHPC can delay the concrete cover cracking time and reduce the average corrosion rate. Self-prestressing UHPC, produced by the combined addition of EA and SF, delayed the concrete cover cracking time and reduced the average corrosion rate by 40.1% and 97.4%, respectively. The increase in thickness of the concrete cover resulted in a significant decrease in the average corrosion rate and vice versa. The electrochemical measurements performed during the alternative wetting and drying cycle showed that the self-prestressing UHPC had the lowest corrosion rate of 0.41 μA/𝑐𝑚2, which is 87.3% lower than the concrete mixture without EA and SF. It sustained maximum residual peak loading, experienced minimum strength loss, and demonstrated higher cracking resistance. The corrosion products characterized by using an energy-dispersive x-ray detector (EDX) and x-ray diffractometry (XRD) for self-prestressing UHPC were oxides with a volume ratio to parent iron of approximately 2, which is the lowest among all. A novel service life prediction model for CSTCs is proposed for chloride-contaminated UHPC and compared by experimental data. This model can reliably predict the corrosion initiation time, cover cracking time, and crack propagation time until the specified crack width. Since cover cracking and crack propagation time can significantly extend the service life by up to 20%, it is necessary to consider them in the service life design for FRC. The chloride-induced corrosion performance of slag-based geopolymer prepared with 15% metakaolin substitution was comparable to ordinary portland cement (OPC) concrete. It can play a significant role in preparing sustainable and durable construction materials.