Balanced Mix Design for Highly Recycled Asphalt Mixtures: Laboratory and Field-Testing in Nebraska

Abstract: Balanced mix design (BMD) is an asphalt mixture design approach that aims to incorporate multiple performance tests on conditioned specimens, potentially leading to more innovative, sustainable, and durable asphalt pavements. A primary challenge in BMD implementation is establishing an appropriate framework, which varies depending on factors such as mixture type, climate, and traffic conditions. This study aimed to contribute to developing a BMD framework for highly recycled asphalt mixtures containing polymer-modified binders commonly used in Nebraska. The main objectives were to define appropriate performance-based tests and their associated threshold criteria, as well as a suitable long-term aging (LTA) protocol, that could be sensitive to the different mixture behaviors observed in a vast array of Nebraska mixtures under laboratory and field conditions. To achieve these objectives, six asphalt mixture types used in a diverse selection of Nebraska projects were collected, varying in binder source and grade, RAP content, binder content, and additives. Recommended monotonic performance tests included IDEAL-CT, Illinois flexibility index test, Hamburg wheel tracking test, IDEAL-RT, and high-temperature indirect tensile test, while dynamic modulus and flow number tests were used as established dynamic-loading tests. To assess long-term performance, three LTA protocols were employed: the NCHRP 09-54 method (3 days oven aging at 95 ºC), critical aging method (8 hours at 135 ºC), and modified critical aging method (4 hours at 135 ºC). In addition, field core samples were extracted at 0-, 12-, 24-, and 36-months post-construction and subjected to laboratory testing. Concurrent with the coring operation, field data were collected, including the international roughness index, cracking percentage, and rutting depth. Results evaluated the sensitivity of different tests for characterizing rutting and cracking resistance across various loading and environmental conditions. Statistical analyses were conducted to identify correlations between index-based parameters obtained from both monotonic and dynamic tests, considering overall trend and ranking across mixtures. The variability and discrimination potential of the selected performance tests were further analyzed and compared. Finally, even though the studied field sections are still in their initial stage (less than three years), the outcomes from lab observations were verified through a comparison of trends from field core performance test results and field monitoring data over time. The ranking of different Nebraska mixtures based on overall field assessment corroborated with laboratory findings. Regarding aging conditioning, while the NCHRP and critical aging protocols yielded similar cracking resistance trends, their aging severity differed, with the critical aging method inducing increased cracking susceptibility among mixtures. Samples conditioned using the proposed modified critical aging method presented similar cracking results to those subjected to the NCHRP protocol. To further verify the efficacy of the proposed aging method to induce similar aging mechanisms as NCHRP one, rheological and chemical parameters (Glower-Rowe and carbonyl index) of extracted and recovered binders from the studied LTA protocols were comparatively analyzed. Finally, the preliminary threshold criteria for each specific test were determined by comparing the performance test indices with the predicted service life obtained from mechanistic-empirical pavement design methods.