A logistic threshold model for asphalt-aggregate adhesion based on mineral composition

This study establishes a quantitative framework to predict asphalt-aggregate adhesion based on mineral composition. We systematically investigated the interfacial behavior using simulated minerals (SiO₂/CaCO₅ blends at 0, 30, 40, 50, 60, 70, and 100 % SiO₂ content) and five natural aggregates. Methods included FTIR-based adsorption tests and water immersion stripping experiments across three asphalt grades (A50, A70, A90). Results revealed a sigmoidal correlation between SiO₂ content and adhesion loss, effectively captured by a novel logistic threshold model. The model identified critical boundaries: a Safe Zone (SiO₂ < 45 %) with minimal stripping (<5 %), a Transition Zone (45–65 % SiO₂), and a Failure Zone (>65 % SiO₂) with severe damage. For natural aggregates, the combined SiO₂+Al₂O₃ content provided a similar logistic relationship, with zones at < 60 %, 60–85 %, and > 85 %, respectively. Asphalt properties (e.g., softening point, viscosity) significantly influenced these thresholds. These findings provide quantitative, practical criteria for aggregate selection and optimization of asphalt mixtures, especially when incorporating non-traditional or acidic aggregates.