Piezoelectric composite cements: Towards the development of self-powered and self-diagnostic materials

Piezoresistivity is the most commonly used sensing principle in cement-based smart composites for strain-monitoring applications. Nonetheless, the need for external electric power to conduct electrical resistivity measurements restricts the scalability of this technology, especially when implemented in remote structures. To address this issue, this manuscript thoroughly analyzes the piezoelectric properties of cement composites doped with reduced graphene oxide (rGO) and evaluates their potential as self-powered strain sensors. To do so, a comprehensive methodology involving voltammetry measurements, open circuit potential determination, and uniaxial compression testing is developed to determine the piezoelectric coefficients of charge $d_{33}$ and voltage $g_{33}$. Furthermore, a novel circuital model for signal processing of the electromechanical response is developed and experimentally validated in terms of time series of output voltage, resistance, and the generated electric power. The developed methodology is applied to laboratory samples manufactured following two different filler dispersion methods. The presented results evidence that samples prepared by ultrasonic cleaner dispersion achieve optimal properties, with a piezoelectric charge coefficient of 1122.28$\mathrm{\pm}$246.67 pC/N, about 47 times greater than previously reported composites in the literature. Unlike piezoresistive cement-based composites, a remarkable nonlinear correlation between the fractional change in the intrinsic resistance of the material and the applied mechanical strain has been observed. Instead, a considerable linearity ($\mathrm{R^2}=0.96$) between the externally applied mechanical strain and the generated (piezoelectric) electric power has been found, which suggests the great potential of the latter for conducting off-the-grid strain monitoring applications.