A key challenge in experimental research on steel corrosion in reinforced concrete is the destructive nature of most experimental techniques, which limits the ability to monitor progressive deterioration over time. Advancements in research on steel corrosion in reinforced concrete, however, have been made through the application of non-destructive imaging techniques, particularly X-ray and Neutron Computed Tomography (XCT and NCT). These techniques enable qualitative and quantitative assessments of steel corrosion in concrete.

X-ray attenuation generally increases with the atomic number (density) of the material, making XCT particularly effective for identifying voids in the specimen. In contrast, neutrons interact with the atomic nuclei and are sensitive to light elements, such as hydrogen, making NCT advantageous for identifying moisture and corrosion products in the sample. Effective implementation requires meticulously experimental planning to ensure high-quality data and efficient post-processing.

Sophisticated image analysis methods, such as a multimodal approach that exploits the statistics of both the X-ray and neutron attenuation fields, facilitate phase segmentation of the sample, thereby enabling detailed insights into factors such as the corrosion morphology and the influence of interfacial voids on pitting corrosion. Additionally, time-resolved imaging combined with digital volume correlation enable for monitoring the evolution of steel corrosion and damage within the sample.

This paper aims to provide insights into the effective application of XCT and NCT for investigating steel corrosion in reinforced concrete, focusing on the challenges and opportunities these techniques present for continuous, non-destructive monitoring. Through illustrative examples, this paper demonstrates how these imaging techniques can improve the understanding of steel corrosion in reinforced concrete.

Micromechanical and time dependent properties of cement paste can be predicted based on the microstructure by using analytical or numerical models. Herein, we propose an alternative approach for predicting the creep modulus of cement paste based on deep convolutional neural network (DCNN). The DCNN is trained using numerical simulation data obtained by the microscale lattice model, resulting in a database with more than 18000 samples. Then, 3 different DCNN architectures are built to learn from (part of) the database. Finally, the accuracy of DCNN prediction is tested on unseen samples. The proposed DCNN architectures can achieve excellent accuracy on the testing set, with the R2 higher than 0.95. Furthermore, the distribution of creep moduli predicted by the DCNNs coincides with the original dataset. Further analyses of the feature maps show that the DCNNs can correctly capture the local importance of different microstructural phases on the predicted creep moduli. Therefore, it was concluded that a well-trained DCNN allows prediction of creep moduli based on microstructural images as input, which is computationally much more efficient compared to image segmentation and numerical simulation methods commonly used today. Of course, computational demands for training the network may be significant, but are needed in principle only once.

In previous research, the utilization of carbon fibers in composite materials has garnered significant attention due to their exceptional mechanical and electrical properties. One promising method for optimizing these properties is through the control of fiber orientation during the extrusion-based 3D printing process. In this study, the orientation of long carbon fibers, specifically measuring 6 mm and 12 mm in length, was successfully controlled using extrusion-based 3D printing techniques via a syringe. The investigation focused on the effects of fiber aspect ratio, volume fraction, and orientation on the electrical properties of the carbon fiber reinforced cementitious composites (c-FRCC) under three-point bending conditions. Flexural hardening was enhanced by the controlled fiber alignment. Electrical results reveal that the electrical resistance (R) of all specimens with aligned carbon fibers is more sensitive to flexural deflection than that of specimens with random fiber orientation before localized failure. The sensitivity in the fractional change in electrical resistance, (R-R₀)/R₀, improved with increasing aspect ratio and volume fraction of carbon fibers. Understanding these relationships is essential for developing advanced materials with tailored properties for crack self-sensing.

In traditional linear elastic fracture mechanics, it is assumed that the stress at the crack tip is infinite. However, in real engineering materials, the ultimate tensile stress is finite, which leads to the formation of a nonlinear micro-cracking zone at the crack tip under applied load. In the context of concrete fracture mechanics, this region is known as the fracture process zone, which plays a crucial role in controlling crack propagation. This study employs Digital Image Correlation technology to measure the full-field displacement and strain throughout the loading process under both natural and saturated humidity conditions. By comparing the load-crack mouth opening displacement obtained from Digital Image Correlation with results from clip-on extensometers, the accuracy of Digital Image Correlation in capturing crack opening displacement at different strain rates is validated. The results indicate that the fracture process zone size and crack opening displacement evolve continuously as loading progresses, with the fracture process zone length and crack opening displacement increasing gradually. Additionally, a more pronounced linear distribution is observed at higher loads. The fracture process zone initially expands with increasing load but begins to shrink as the load diminishes. Furthermore, higher strain rates lead to a shorter fracture process zone length and correspondingly higher loads, while under saturated humidity conditions, the fracture process zone forms earlier than under natural humidity.