Carbon nanotube-based hierarchical fillers modified cementitious composites for smart structures

Abstract

Taking advantage of the excellent mechanical properties and extraordinary electrical properties of carbon nanotubes (CNTs), self-sensing cementitious composite (SSCC) nanoengineered with CNTs has recently attracted much research interest aimed at improving the mechanical properties while simultaneously providing additional multifunctional and smart benefits such as enhanced electrical conductivity and, most importantly, self-sensing properties enabling the health condition monitoring of critical elements of civil engineering infrastructure and even smart elements themselves with integrated self- monitoring abilities. A critical factor determining the properties of SSCCs is the uniform dispersion of CNTs in the cementitious composites. Unfortunately, due to the high aspect ratio, specific surface energy and the van der Waals force, CNTs agglomerate too easily in the cement matrix, leading to the formation of defective sites inside the composite and compromised CNT filler efficiency. In-situ growth and ex-situ grafting of nanoscale CNTs onto microscale materials to fabricate CNT-based hierarchical fillers has been studied in this research as a promising approach toward providing high loadings of CNTs in the composites, while alleviating the dispersion problem as well as enhancing composite properties. This thesis is fundamentally centred around solving the issue of CNT dispersion in cementitious composites by studying three advanced CNT-based hierarchical fillers. Involving preparations, study of properties and application performances, a new generation of multifunctional SSCCs nanoengineered with CNT-based hierarchical fillers have been developed. The main contents and conclusions are as follows. (1) High-density CNTs directly on carbon fibres (CF-CNTs) have been successfully synthesized using chemical vapour deposition (CVD) method. The complexity of the in-situ CVD growth of the CF-CNT hierarchical fillers was systematically examined to deepen the understanding of CNT growth mechanisms. For the first time, the utilization of the as-synthesized CF-CNT hierarchical fillers in tailoring SSCC performance was then investigated. Experimental results showed that highly customized CF-CNTs with favourable morphologies and plausible properties can be well tailored by suitably tuning the CVD parameters. With the incorporation of optimized CF-CNT hierarchical fillers, the SSCCs exhibited impressive mechanical, electrical and self-sensing properties due to the homogeneous distribution of CNTs in the matrix and the enhanced fibre/cement matrix interfacial bond strength. However, due to thermally-induced performance degradation of the CFs and the high density of CNT coverage on the CFs, the mechanical properties of the SSCCs with CF-CNTs were inferior to the SSCCs with CFs, especially with respect to flexural properties. (2) In line with the understanding of in-situ CVD growth of CF-CNTs, a more straightforward approach to synthesize hierarchically structured filler by in-situ CVD growing CNTs on cement particles (CNT@C) were then explored. The effect on early-age hydration, mechanical properties, microstructures, electrical properties and self-sensing properties of the SSCCs with CNT@C were systematically investigated for the first time. Experimental results showed that the nest-like CNT@C structure can promote early-age hydration while slowing the later hydration rate and hindering the strength development of the SSCCs. The addition of CNT@C can be effective in tailoring the electrical microstructures to enhance the electrical conductivity and self-sensing sensitivity of the SSCC. The SSCC with CNT@C achieved a maximum stress sensitivity of 2.87 %/MPa with a gauge factor of 748 and demonstrated excellent repeatability and stability, an outstanding adaptability to various applied conditions and fast response and recovery, thereby, showing great potential for practical structural health monitoring applications. (3) Due to the inefficient production of the CF-CNTs and CNT@C hierarchical fillers using the traditional CVD system, a commercially-available and low-cost composite filler, namely CNTs ex-situ assembled on carbon blacks (CNT/CB) via electrostatically self-assembly (ESA) was then adopted. The composites filler was directly incorporated into cementitious composites without any treatment to develop a new type of SSCC with high sensing sensitivity and repeatability. The SSCCs were then integrated into a five-story building model using custom-made clamps to verify the feasibility of vibration-based damage detection. Experimental results showed that the SSCCs with CNT/CB processed satisfactory mechanical property and excellent self-sensing repeatability and stability. Clamped to the building model, the SSCCs performed satisfactorily when subjected to sinusoidal excitations in the frequency range from 2 to 40 Hz. In addition, the modal frequencies of the building model as identified by the SSCCs, and the changes caused by damage simulated through adding additional masses were favourably consistent with the counterparts acquired by accelerometers and strain gauges, indicating that the SSCCs have great potential for structural modal identification and damage detection applications. (4) In order to use the SSCCs with CNT/CB composite fillers in complex field environments and understand the temperature influencing mechanisms, the temperature-sensitive properties and temperature effects on the self-sensing properties of the SSCCs were studied and characterized. Experimental results showed that an increase in content of CNT/CB composite fillers can decrease the activation energy of the SSCCs and facilitate the transport of the charge carriers, thus attenuating SSCC temperature-sensitivity. Temperature increase can reduce the self-sensing sensitivity without any effect on SSCC repeatability. In view of the diminishing temperature effect on the self-sensing properties of the SSCCs, responses of the SSCCs under simultaneous temperature and loading excitations were then treated using Bayesian source separation method to reconstruct two independent sources. Regardless of CNT/CB composite filler content, the temperature-related reconstructed source always had a high correlation coefficient with the measured temperature, indicating that the Bayesian source separation method can well separate the electrical resistance variation induced by temperature variation from that induced by simultaneous temperature and loading excitations. However, the method was unable to remove temperature effect on the self-sensing sensitivity of the SSCC.