Predictive Framework for Modeling Changes in Asphalt Mixture Moduli with Oxidative Aging

Abstract

This paper presents a predictive framework for asphalt mixture moduli as a function of aging time with two levels of sophistication. This work is built on the method currently implemented in Pavement mechanistic-empirical (ME) that uses an effective time/frequency concept based on time-aging superposition to model the effect of aging on a mixture’s modulus. Time-aging superposition implies that an asphalt mixture’s modulus mastercurves, corresponding to different aging levels, coincide when they are shifted horizontally on the log-frequency axis. This study improves the accuracy of the existing model by decoupling the time-temperature and time-aging shifts. The new framework also uses the binder dynamic shear modulus | G*| as an aging index instead of the viscosity, which is used in Pavement ME. The | G*| aging index is used to calculate an effective frequency at short-term aging (STA), which is then used in the asphalt mixture sigmoidal model to calculate the corresponding asphalt mixture modulus with aging. The pavement aging model introduced by NCHRP 09-54 predicts log | G*| at 64°C and 10 rad/s for a specific field-aged condition and pavement depth. The proposed framework can use the predicted log | G*| to predict the mixture’s corresponding dynamic modulus (| E*|) at that aging level and pavement depth. Level 1 of this framework requires characterizing the | G*| at STA and calibrating the NCHRP 09-54 pavement aging model as well as measuring the mixture | E*| at STA. Level 2 does not require any binder testing, providing relatively less accurate predictions but relieving some testing requirements.