Predictive correlations and mechanical assessment of asphalt mixtures modified by sasobit redux and crumb rubber

Abstract

Abstract To satisfy advanced pavement design methodologies such as the Mechanistic-Empirical Pavement Design Guide (MEPDG), accurate characterization of asphalt mixtures through parameters like the dynamic modulus│E*│is critical. However, obtaining these parameters via advanced laboratory testing demands specialized equipment and complex configurations. This study addresses this limitation by developing empirical correlations to predict│E*│and shear triaxial resistance (STR) directly from conventional, highly accessible laboratory tests, specifically Marshall, unconfined direct compression, indirect tensile strength (ITS), and wheel tracking tests. The experimental program evaluated both Hot Mix Asphalt (HMA) as control mixes and Warm Mix Asphalt (WMA) prepared with diverse aggregate gradations. To enhance the mechanical and environmental performance of the mixtures, two additives were evaluated, Sasobit Redux (SR), blended into the bitumen utilizing the wet process at dosages of 1.0% to 5.0% by weight of bitumen; and Crumb Rubber (CR), incorporated into the total mix weight via the dry process at dosages of 0.5, 1.0, 1.5, 2.5, and 5%. Specimen testing was conducted across a comprehensive matrix of loading rates and environmental conditions. The study successfully established highly predictive empirical equations as primary outcomes. The experimental data revealed that incorporating SR and CR additives significantly altered and enhanced key mechanical properties, specifically enhancing Marshall stability, rutting resistance, unconfined compressive strength, and ITS. Quantitatively, the statistical analysis proved the strong predictive capability of conventional tests, yielding exceptional mathematical correlations within this specific dataset with an R² value of 0.9831 for the│E*│versus rut depth (RD) relationship and an R² value of 0.8751 for the│E*│versus STR relationship. Consequently, these newly developed empirical equations deliver a validated, cost-effective framework for accurately estimating complex pavement design inputs directly from routine laboratory testing data.