Open Access
Chang, Chieh-Tang
Graduate Program:
Civil Engineering
Master of Science
Document Type:
Master Thesis
Date of Defense:
September 14, 2009
Committee Members:
  • Mansour Solaimanian, Thesis Advisor
  • accelerated pavement testing
  • moisture damage
  • hot-mix asphalt pavement
  • dynamic modulus
The sources delivering quality aggregate for use in hot-mix asphalt concrete have been depleting in PennDOT District 1, which has currently resulted in shortage of quality Type A aggregate in this district. The local aggregates don’t meet the required criteria on soundness or absorption level, or both. Furthermore, stripping and moisture damage in these gravel aggregates are of concern even though most of these aggregates exhibit excellent skid resistance and durability. As a result of this situation, research was conducted to evaluate resistance of moisture damage of hot-mix asphalt concrete using District 1-0 aggregates. Research included modifications that could be applied to asphalt mixtures to improve moisture resistance of such aggregates in hot-mix asphalt. Four Type C aggregates and one Type A aggregate, all sources located in District 1-0, were selected for evaluation. Mixes were prepared as control, with liquid anti-stripping agent, with lime, and with a gravel-limestone blend for the #8 material at equal proportions. The tests included PennDOT’s modified version of AASHTO test method T283 (Tensile Strength Ratio), one-third scale model mobile load simulator (MMLS3), and dynamic modulus test with repeated freeze-thaw cycles. Overall, it was concluded that two of the five aggregate sources without any treatment could pass the requirement on moisture damage resistance based on the PennDOT version of the AASHTO T283 test method. It was also found that the specific liquid anti-stripping agent used with these mixes improved the moisture damage resistance significantly. The study indicated improvement of moisture damage resistance using the limestone-gravel blend to a much lesser degree compared to the improvement gained through the usage of liquid anti-stripping agent. In testing with the MMLS3, only three control mixes were included, and testing was conducted under both dry and wet conditions. A higher rutting level was found in wet tested specimens compared to dry specimens. Only two of the five mixes were chosen for evaluation using dynamic modulus testing due to limited resources and time. Each specimen was planned to be tested four times: unconditioned, after it was conditioned once, conditioned twice, and conditioned three times. However, a few specimens failed after second cycle of conditioning and therefore dynamic modulus could not be obtained for all conditions. Results indicate that the modulus decreases as the testing frequency decreases. Furthermore, the modulus dropped after being conditioned at all frequencies for all the mixes, especially after the second cycle of conditioning.