Evaluation of Novel Jointless Engineered Cementitious Composites Ultrathin Whitetopping (ECC-UTW)

Ricardo Hungria


Ultrathin Whitetopping (UTW) is a rehabilitation technique that consists of placing a thin layer of concrete over a damaged asphalt pavement. While UTWs have demonstrated satisfactory performance in field tests, the inherent brittleness of concrete has led to the appearance of distresses. To address these distresses, Engineered Cementitious Composites (ECC) have been proposed for UTW applications, thanks to their exceptional attributes, including tensile ductility, flexural performance, and fatigue resistance. In this study, an ECC-UTW was constructed to investigate its potential as a viable solution. The project comprised two 18.3 m ECC sections with different thicknesses, namely 63.5 mm and 101.6 mm, both designed to be completely jointless. Additionally, a 101.6 mm jointed concrete section was included for comparison. Once the project was built, compressive, uniaxial, flexural performance, failing weight deflectometer, and interlayer shear tests were performed to determine the in-situ properties of the overlays. Moreover, accelerated loading was applied to these UTWs to assess their performance. The results revealed that the 63.5 mm ECC jointless section failed after enduring 76,378 passes of 40 kN (representative of a dual tire half axle load of 80 kN), which corresponds to 76,378 ESALs. Similarly, the 101.6 mm jointless ECC-UTW failed at 151,454 passes of 40 kN and 14,310 passes of 71.2 kN, which represents 331,488 ESALs. The 101.6 mm jointed concrete UTW, in contrast, failed at 150,094 passes of 40 kN and 35,299 passes of 71.2 kN, which represents 594,191 ESALs. Furthermore, finite element modeling was conducted to determine the critical stresses upon loading. Finally, these stresses were integrated to an existing ECC fatigue model to develop a performance prediction model, which determines the cycles to failure based on the material thickness. This model was validated with the in-situ results yielding a difference of 5.09% between the experimental and the computational data.