A comprehensive constitutive model for asphalt concrete was calibrated that included viscoelasticity, viscoplasticity, and irreversible microstructural damage in unconfined compression. Three different types of laboratory tests were designed and performed to calibrate each of these response components. Small-strain dynamic modulus tests were used to calibrate the undamaged linear viscoelastic properties. Cyclic creep and recovery tests to failure were performed to calibrate the viscoplastic properties. Constant-rate-of-strain tests to failure were used to calibrate the damage behavior. These tests were performed at a wide range of temperatures, loading rates, and stress levels. Upon calibration of each individual response, the model was validated by predicting the results of other constant-rate-of-strain tests at temperatures and strain rates different from those used in the calibrations. The predictions for these different conditions indicate that the comprehensive model can realistically simulate a wide range of asphalt concrete behavior.

A study was performed to evaluate whether time-temperature superposition principles would continue to apply to the behavior of asphalt concrete beyond the commonly assumed small strain (<100 με) limits. A series of unconfined uniaxial compression constant crosshead displacement rate tests were performed to large-strain values. The measured axial stress versus axial strain data were cross-plotted to produce stress versus reduced time master curves and corresponding temperature shift functions at various strain levels to determine the maximum strain level at which time-temperature superposition remains valid. The results suggest that asphalt concrete remains a thermorheologically simple material well into the postpeak region (i.e., that time-temperature superposition is valid throughout the useful stress-strain response). The results further suggest that the temperature shift function aT may be only a weak function of strain level. For many practical engineering purposes, however, the differences between the small-strain and large-strain temperature shift relations may be of negligible importance.

No abstract

No abstract