Innovations in Mixture and Structural Design of Cold Recycled Pavements

Abstract

Cold recycled (CR) pavements, the combination of reclaimed asphalt pavement (RAP) and an asphalt stabilizing agent (foamed or emulsified asphalt binder) at ambient temperatures, can provide a sustainable method to rehabilitate, maintain, and construct roadways. The design, testing, and applications of CR pavements have advanced as the knowledge surrounding the topic grows; however, many of the specifications and design methods being implemented are based upon Hot Mix Asphalt (HMA)/Warm Mix Asphalt (WMA) mixture properties and characteristics. If CR pavement is to be used as an alternative to an HMA base or potentially as a surface course, assessment of the mixture design, laboratory testing, and structural design methods is necessary. This dissertation aims to fill this knowledge gap by evaluating existing structural design and laboratory design processes and assessing the shortcomings of the existing processes.

To assess existing structural design methods, the American Association of State Highway and Transportation Officials (AASHTO) AASHTOWare Pavement Mechanistic-Empirical (ME) Design software was used. Structural and mixture properties of 13 flexible pavement sections containing a CR base or intermediate layer were input into the software, and the predicted distresses produced by the bottom up and total rutting distress models were compared to the measured distresses. Results indicated that the global coefficients, all of which were established based on HMA/WMA data, were not sufficient in accurately predicting pavement distresses. Model coefficient calibrations were conducted for the bottom up cracking and total rutting models. The bottom up cracking model coefficient calibrations indicated that the recommended method of measured cracking aggregation led to distress overprediction, which may result in an over or under designed pavement that does not perform as intended. Per the AASHTO Local Calibration Guide (AASHTO, 2010), recommended aggregation methods state that all cracking should be considered bottom up unless forensics indicate otherwise. The total rutting model coefficient calibrations showed that in pavements with 5 inches of total asphalt thickness and a CR layer placed directly on a granular base, subgrade rutting dominated the predicted rutting model with the subgrade rutting alone being over 4 times the maximum measured rut depth.

CR mixture advancements were also evaluated through the mixture design and field placement of a CR mixture as a surface course. Mixture design phases explored rejuvenator and emulsion contents impact on mixture performance, resulting in an optimum design for field placement. The placement of the rejuvenated CR mixture as a surface course was like that of any other CR mixture. Evaluations of laboratory testing methods and performance criteria concluded that CR mixture performance may not be accurately captured using existing methods and criteria developed around HMA/WMA. Discussion of laboratory performance tests for CR mixtures highlighted the limitations based on current methods and recommended additional testing for CR mixtures intended for surface course applications.

The results from this dissertation aided in identifying the limitations of the existing processes surrounding laboratory performance testing and ME structural design via AASHTOWare Pavement ME Design (PMED) software and provided recommendations for considerations to be made with implementation of the findings and future research.