Dataset on carbonation and chloride-induced steel corrosion in cementitious mortars

Machine learning (ML) has seen success in civil and structural engineering, but its application to forecasting corrosion of steel reinforcement in concrete structures is limited due to small datasets from isolated studies. Moreover, the existing corrosion dataset of reinforced concrete typically lacks sufficient and comprehensive material and environmental information that enables reliable corrosion prediction of reinforced concrete under complex corrosion scenarios. This work aims to bridge the gap by compiling and building a comprehensive corrosion dataset focusing on carbon steel in cementitious mortars. This dataset involves 46 distinct mortar mixtures with embedded steel bars. The samples first underwent accelerated corrosion testing (either by carbonation or chloride contamination), followed by investigating their corrosion behaviours under varying relative humidity (RH) conditions. Corrosion data were obtained during this period, in which all corrosion measurements were conducted in laboratory settings and the results are tabulated in spreadsheet format (.xlsx). The dataset encompasses mixture parameters, material properties, environmental parameters, and electrochemical parameters. This extensive dataset provides valuable corrosion data for training ML models to predict steel corrosion across various corrosion-related variables.