Deformation and Damage Mechanisms of Cement-Stabilized Reclaimed Asphalt Pavement (CSRAP) Mixtures: Experimental Insights and Constitutive Modeling

With the growing emphasis on sustainability and resource conservation in pavement engineering, cement-stabilized reclaimed asphalt pavement (CSRAP) has emerged as a promising solution for base and subbase recycling. However, the stage-dependent deformation and damage mechanisms of CSRAP mixtures remain insufficiently understood. This study systematically investigates the mechanical response, strain localization, and damage development of CSRAP mixtures under uniaxial compression through a combined experimental–theoretical–microscopic approach. Unconfined compressive strength (UCS) tests and two-dimensional digital image correlation (2D-DIC) were employed to evaluate the effects of reclaimed asphalt pavement (RAP) characteristics on stress–strain response, strain localization, and staged deformation behavior. The results show that, after 7 days, increasing the RAP content from 0 to 100% reduces peak stress by 27.5–60.2% while increasing peak strain by up to 26.8%. For 100% RAP mixtures, extending the curing time from 7 to 90 days increases peak stress by 3.45–5.00 MPa. A physically based damage constitutive model was developed by integrating the Logistic function with Lemaitre’s strain equivalence hypothesis and validated using 80 sets of experimental data, demonstrating good agreement with the measured stress–strain curves. Energy-dispersive spectroscopy (EDS) analysis further revealed that curing age and RAP particle size significantly influence the composition and densification of the interfacial transition zone (ITZ), and that a moderate aged asphalt film can promote secondary hydration and improve interfacial integrity. More importantly, this study reveals the multistage coupling among RAP characteristics, ITZ evolution, and pre-peak damage development in CSRAP mixtures, providing new insight into how material composition affects deformation accumulation and damage progression. The proposed model not only captures the pre-peak stress–strain and damage evolution behavior with high accuracy, but also offers physically interpretable parameters, making it useful for constitutive characterization, performance prediction, and optimization design of recycled pavement bases.