Full-waveform inversion using a stepped-frequency GPR to characterize the tack coat in hot-mix asphalt (HMA) layers of flexible pavements

The use of tack coats in the construction of road structures is a technique that ensures pavement layer bonding. Unfortunately, existing techniques commonly used in France to estimate tack coat conditions require coring structures for ad hoc characterization in the laboratory. This paper investigates a non-destructive approach to characterize changes in geometric and dielectric properties of the tack coats present inside the hot-mix asphalt (HMA) layers of flexible pavements, with the use of a stepped-frequency radar combined with a mono-static, off-ground, ultra-wideband Vivaldi antenna. The principle of this method is to consider the flexible pavement as a multi-layered medium with changing properties (thicknesses, dielectric susceptibilities and dispersion parameters). Then a full-waveform inversion is applied, which allows for the extraction of the dielectric and geometric parameters of each layer composing the structure. Following numerical validations, the multi-layered forward model, namely, a Green's function model integrating the 2-parameter variant of Jonscher's model, combined with Lambot's et al. radar full-wave method was experimentally validated using calibrating media (up to three layered media). Then, the link between both the dielectric and the geometric parameters of the tack coats and the emulsion quantity (bonding) was analyzed. The results showed good agreement between measured and modeled radar data. Radar data inversions showed in particular that the dielectric susceptibility and thickness of the transition zone increase with the emulsion quantity. Future research will focus on the establishment of a reference database of calibration curves linking dielectric susceptibility and tack coat characteristics.