Study on Dielectric Properties of Asphalt Mixtures Considering the Effects of Relative Humidity

Abstract

Ground penetrating radar (GPR) technology is used for nondestructive testing of roads. It relies on the processing and analysis of electromagnetic waves reflected due to the differences in dielectric properties encountered by electromagnetic pulses during propagation for road quality assessment and detects distresses. The moisture content of the asphalt mixture in asphalt pavement undergoes dynamic changes, affecting the pavement’s relative humidity (RH). This alteration in RH has a direct impact on the dielectric properties of the asphalt mixture, thus influencing the accuracy of GPR road detection. Therefore, it is essential to quantify the relationship between RH and the dielectric properties of asphalt mixtures. This study utilized electromagnetic field theory to establish the correlation between air dielectric constant and RH. Dielectric constant tests were performed on commonly used single-phase component materials (limestone, 70# asphalt, SBS-modified asphalt) of asphalt mixtures and different types (asphalt type, asphalt-aggregate ratio, nominal maximum aggregate size) of asphalt mixtures under varying RH. The study quantified the impact of RH on the dielectric properties of asphalt mixtures and developed a CRIM optimization dielectric model that incorporates the RH effect. The research results showed that the dielectric constants of air, limestone, asphalt, and asphalt mixtures increase with the increase in RH. Specifically, the dielectric constants of air and asphalt mixtures exhibit a linear relationship with RH. Moreover, the increase in the dielectric constant of 70# asphalt with RH is greater than that of SBS-modified asphalt. Under identical conditions, reducing the impact of RH on the dielectric properties of asphalt mixtures can be achieved by increasing the asphalt-aggregate ratio, decreasing the nominal maximum particle size of the aggregates, or utilizing SBS-modified asphalt. The CRIM optimization model yielded an overall average relative error of 3.33% for calculating the dielectric constants of different types of asphalt mixtures under the influence of RH, representing at least a 1.67% improvement compared to classical models and the improved model proposed from existing literature. The research outcomes provide significant guidance for enriching and developing the fundamental theory as well as detection evaluation technology of GPR.