Revisiting asphalt mixture cracking Performance: Development of a work of Fracture-Based crack propagation resistance Index from IDEAL-CT

Due to its simplicity and efficiency, the Indirect Tensile Asphalt Cracking Test (IDEAL-CT) has been widely adopted to characterize the cracking performance of asphalt mixtures. However, the Cracking Tolerance Index (CT-Index), a key parameter derived from IDEAL-CT, exhibits critical limitations due to its over-reliance on the slope value at 75 % peak load point in the post-peak stage of the Load-Displacement curve. This over-reliance on a single point slope falls short of effectively accounting for the cumulative cracking propagation resistance, resulting in inconsistencies in evaluating the effects of air voids, and polymer/fiber modifications. To address this limitation, this study includes a proposed Cracking Propagation Resistance Index (CPR-Index), derived from the Work of Fracture versus Displacement curve, to characterize the average cumulative energy required for crack propagation during the post-peak stage. A comprehensive experimental program was conducted on asphalt mixtures with varying design factors to compare the capacity of the CPR-Index and CT-Index parameters, including air voids, polymer and fiber modifications, binder content, recycled asphalt content, and aging conditions. The variability of the two indices was also compared. The following findings can be drawn from the analysis of results: (1) the CPR-Index decreases with increasing air voids, effectively capturing the expected impact of mixture density on cracking resistance; (2) CPR-Index demonstrates superior sensitivity to polymer and fiber modifications, accurately reflecting their roles in improving cracking resistance as validated through the binder cracking test; (3) CPR-Index shows trends consistent with the CT-Index for binder content, RAP/RAS content, and aging conditions; (4) Pearson’s correlation analysis shows a strong relationship (R² = 0.71–0.92) between the CPR-Index and binder fatigue life measured from the binder time sweep test, and; (5) the CPR-Index exhibits better repeatability, with a coefficient of variation about 10 %, thus effectively reducing variation between replicate measurements.