Evaluation of Dynamic Modulus Prediction Models for Determination of Viscoelastic Properties of Hot, Warm and RAP-contained Asphalt Mixtures

One of the requirements for mechanistic-empirical design of asphalt pavement is determining the viscoelastic properties of asphalt mixture, including dynamic modulus and phase angle. Since the dynamic modulus test of asphalt mixture is time consuming and expensive, some prediction models have been proposed by various researchers. The purpose of the present study was to investigate some of the available predictive models to determine the viscoelastic properties of hot, warm and RAP-contained asphalt mixtures. For this aim, asphalt mixtures with two types of gradation (with a nominal maximum aggregate size of 9.5 and 19 mm) representing the mixtures prepared for surface and binder layers, as well as three kinds of bitumen, including neat bitumen, Sasobit-modified bitumen and reclaimed bitumen, were designed according to Superpave method. Dynamic shear rheometer and dynamic modulus tests were performed on the bitumen and the mixtures, respectively. The empirical model of NCHRP 1-40D and micro-mechanical models of Hirsch and its modified form were evaluated for estimating the dynamic modulus of asphalt mixtures using laboratory measurements. It was found that the empirical model of NCHRP performed very well after calibration (R2> 0.95) and then the semi-empirical and micromechanical models of modified Hirsch and Hirsch had a good performance. All models had a better predictability for the modulus of fine graded mixtures. It seems that the considered models are not successful in taking into account the resistance from aggregates interlocking. Application of dynamic modulus prediction models is suitable for implementation of the new design method at Level 2.