The Effect of High RAP and High Asphalt Binder Content on the Dynamic Modulus and Fatigue Resistance of Asphalt Concrete

Abstract

This thesis investigates the effects of using various percentages of RAP and asphalt binder contents on the dynamic modulus and fatigue resistance of asphalt concrete. Two RAP percentages (20% and 40%) and three binder percentages (plant-mixed, plant-mixed + 0.5%, and plant-mixed + 1.0%) were evaluated. A Superpave gyratory compactor and an asphalt vibratory compactor were used to prepare dynamic modulus samples and fatigue beam samples at 7% air voids. Three replicate samples for each percentage of RAP and asphalt binder content were prepared for testing purposes. An Interlaken Technology Corporation servohydraulic testing machine and a Material Testing System servohydraulic machine were used to determine the dynamic modulus and fatigue resistance of the asphalt samples. Analysis of variance (ANOVA) was used to determine if any of the factors (air voids, percent RAP, and percent asphalt binder) affected the performance criteria (dynamic modulus and fatigue life cycles). Results suggest that as the amount of RAP increases in asphalt concrete, both the dynamic modulus and fatigue life will increase. As per the literature, these results were expected for the dynamic modulus, but not for the fatigue life. It is suspected that the increase in fatigue life for the 40% RAP mixes may be due to the use of a softer binder (PG 64-22 instead of PG 70-22). It was also found that by increasing the amount of binder in the mixture, the stiffness of asphalt concrete will decrease, but the fatigue life will improve. The fatigue life results showed a strong trend of this improvement for the 20% RAP samples, however, the results for the 40% RAP samples were inconclusive. For dynamic modulus, it was found that the percent RAP, additional binder, frequency, and temperature were all statistically significant with 95% confidence. For the fatigue life, ANOVA showed that the percent RAP and additional binder were statistically significant with 95% confidence. These results suggest that by utilizing a higher percentage of RAP and asphalt binder, it is possible to meet or improve upon the dynamic modulus and fatigue life of the lower percentage of RAP samples.