Suggested Citation:
Alavi, Mohammad Zia, Yuan He, John T. Harvey, David Jones (2015) Evaluation of the Combined Effects of Reclaimed Asphalt Pavement (RAP), Reclaimed Asphalt Shingles (RAS), Different Virgin Binder Sources on the Performance of Blended Binders for Mixes with Higher Percentages of RAP and RAS. Institute of Transportation Studies, University of California, Davis, Research Report UCD-ITS-RR-15-28

This report summarizes the main findings from a project funded by the National Center for Sustainable Transportation (NCST) to investigate the use of higher percentages of reclaimed asphalt pavement (RAP) and reclaimed asphalt shingles (RAS) as a replacement for a percentage of the virgin binder in new asphalt mixes. The research focused on testing procedures that do not first require chemical extraction and recovery of the asphalt binders from the RAP and RAS. Five different asphalt binders covering two performance grades and sourced from three California refineries were evaluated. The influence of two different percentages of RAP (25 and 40% by binder replacement) and one percentage of RAS (15% by binder replacement) were evaluated. The effect of a petroleum-based rejuvenating agent in selected mixes was also investigated. Key observations and findings from this project include the following:

Asphalt binder extracted and recovered from RAS could not be tested due to its very high stiffness.

Methods for preparing and testing FAM specimens were developed as part of this preliminary testing phase to measure dynamic shear modulus at different temperatures and frequencies. Cylindrical specimens 0.5 in. (12.5 mm) in diameter cored from a Superpave gyratory-compacted FAM specimen were tested using a torsion bar fixture in a DSR. Preliminary testing of FAM mixes prepared with materials passing the #8 (2.36 mm) sieves indicated that this approach appears to be repeatable and reproducible, and produces representative results for characterizing the performance-related properties of composite binders at binder replacement rates up to 40 percent and possibly higher. Use of materials passing the #8 sieve is recommended.

Statistical analyses of the test results indicated that RAP and RAS content, asphalt binder grade and source, and rejuvenating agent all had an influence on FAM mix stiffness.

The FAM mixes containing RAS showed similar stiffnesses to the corresponding control mixes, suggesting that the RAS binder did not effectively blend with the virgin binder at the temperatures and mixing durations used.

The influence of rejuvenating agent on reducing the blended binder and FAM mix stiffnesses was evident.

Reasonable correlations were observed between the stiffnesses of asphalt binder and the stiffnesses of FAM mixes at testing frequencies ranging from 0.1 Hz to 10 Hz. Discrepancies between the two measured stiffnesses may be an indication that blending between the virgin and reclaimed asphalt binders was incomplete in the FAM mix, but was forced during the chemical extraction and recovery.

Based on the findings from this study, FAM mix testing is considered to be a potentially appropriate procedure for evaluating the properties of blended asphalt binder in mixes containing relatively high quantities of RAP and RAS. Further testing on a wider range of asphalt binder grades, asphalt binder sources, and RAP and RAS sources is recommended to confirm this conclusion and to develop models for relating binder properties determined from FAM mix testing to those determined from conventional performance grade testing.

UC Pavement Research Center Research Report UCPRC-RR-2015-06