Defect identification of nano-cementitious composites, using statistical analysis of thermal images

Carbon-based nanomaterials have diverse applications, including in electronics, energy storage, and environmental remediation. Current research aims to enhance the multifunctionality of construction materials by integrating cement with carbon-based nanomaterials known for their high thermal and electrical conductivity. However, nano-cementitious composites may develop internal defects due to the agglomeration of nanomaterials during fabrication. This study aims to utilize the heat characteristics of nano-cementitious composites for detecting internal defect. Test specimens were prepared by incorporating 1 wt% Multi-Walled Carbon Nanotubes (MWCNTs) into the cement paste. Internal defect sizes were defined from 5 % to 25 % of the specimen's area and were positioned in the center of the specimen. A heat generation test was conducted to capture thermal images and temperature data. Quantitative analysis followed, examining the correlation between defect size and temperature through statistical analysis of the thermal images, using skewness to assess this correlation. The results of the skewness analysis indicated a linear correlation with defect size. A parametric study was conducted to analyze the impact of the number of pixel data on skewness, and it was derived that at least 1500 pixel data are required. Consequently, this study concludes that internal defects in nano-cementitious composites can be detected and their sizes can be inferred through the linear relationship between defect size and skewness.