Evaluation of fatigue characteristics of asphalt mixtures using Cracking Tolerance index (CTIndex)

Evaluation of the cracking potential of asphalt mix during the design phase can considerably limit the fatigue cracking and increase the service life of the flexible pavement. Thus, attention needs to be paid to employing a simple yet reliable performance test indicator. Indirect Tension Asphalt Cracking Test (IDEAL-CT) is one such emerging test method and is widely used in research across the world. The present study examines the potential of using the cracking tolerance index (CTIndex) as an appropriate parameter to explain the fatigue resistance of mixtures with both unmodified and modified binders. This test can be conducted using a Marshall testing machine or multispeed tester, routinely used for asphalt mixture design in India and many other countries. To evaluate the effectiveness of this parameter, fatigue resistance indicators for different binders, i.e., recovery at 3.2 kPa stress level (R3.2) from multiple stress creep and recovery (MSCR) test, fatigue life (TNf3.5) from the time sweep (TS) test at 3.5% strain, fatigue life at 2.5% (LNf2.5), and 5% (LNf5) strain from linear amplitude sweep (LAS) test, as well as for different mixtures, i.e., fatigue life from Indirect tensile fatigue test (ITFT) at 150 microstrain were considered. Two unmodified (viscosity grade 30, viscosity grade 40) and three modified (polymer, crumb rubber, sasobit) binders were evaluated for their fatigue parameters which were further correlated with CTIndex. The CTIndex exhibited good to excellent correlations (R2 ∼ 0.85 to 0.92) with the fatigue lives of the binders and the mixtures. To examine the effect of rheology and chemistry of the binder on the CTIndex, correlations of complex shear viscosity using Dynamic Shear Rheometer (DSR) at 60 °C and asphaltenes content from Thin-layer chromatography with flame ionization detector (TLC-FID) with CTIndex were explored. Viscosity, asphaltenes content, and elastic recovery of the binders depicted good correlations (R2 ∼ 0.88) with CTIndex, indicating higher cracking tolerance of stiffer yet elastic binders. Further, the sensitivity of CTIndex to binder type, binder content, aggregate gradation, and test temperature was examined. Overall, the values of the CTIndex displayed less variability (coefficient of variation < 10%) in this study. The observed sensitivity of CTIndex to asphalt mix parameters and the correlations of the CTIndex with fatigue resistance indicating parameters suggested that the CTIndex is an effective parameter for evaluating the fatigue characteristics of asphalt mixes.