Virtual strain loading method for low temperature cohesive failure of asphalt binder

Cohesive failure is one of the primary reasons for low-temperature cracking in asphalt pavements. Understanding the micro-level mechanism is crucial for comprehending cohesive failure behavior. However, previous literature has not fully reported on this aspect. Moreover, there has been insufficient attention given to the correlation between macroscopic and microscopic failures. To address these issues, this study employed molecular dynamics simulation to investigate the low-temperature tensile behavior of asphalt binder. By applying virtual strain, the separation work during asphalt binder tensile failure was calculated. Additionally, a correlation between macroscopic and microscopic tensile behaviors was established. Specifically, a quadrilateral asphalt binder model was generated based on SARA fractions. By applying various combinations of virtual strain loading, the separation work at tensile failure was determined. Furthermore, the impact of strain loading combinations on separation work was analyzed. Normalization was employed to establish the correlation between macroscopic and microscopic tensile behaviors. The results indicated that thermodynamic and classical mechanical indicators validated the reliability of the tetragonal asphalt binder model. The strain loading combination consists of strain rate and loading number. All strain loading combinations exhibited the similar tensile failure characteristic. The critical separation strain was hardly influenced by strain loading combination. However, increasing strain rate significantly enhanced both the maximum traction stress and separation work of the asphalt binder. An increment in the loading number led to a decrease in separation work. The virtual strain combination of 0.5%-80 provided a more accurate representation of the actual asphalt's tensile behavior trend.