Abstract
Incorporating Reclaimed Asphalt Pavement (RAP) in asphalt mixtures offers remarkable environmental and economic benefits. However, proper design and evaluation of asphalt mixtures with RAP are needed to ensure satisfactory performance. This study examined and proposed several methods for proper design and evaluation of asphalt mixtures with and without RAP. The first phase of this study developed and evaluated several compaction indices as well as measured mechanical properties of asphalt mixtures prepared with RAP. The main objective of this phase is to identify methods that can be used to detect changes in mix composition (e.g., change in RAP content, binder content, binder grade, etc.) during production. The change in mix composition may adversely affect the performance of asphalt mixtures. For example, the use of higher percent of RAP or lower binder content than specified may result in stiffer mixes that are prone to fatigue cracking leading to premature failure of asphalt pavements. A mathematical model was developed to calculate a parameter called Gyratory Stability (GS) from the compaction curve data specific to the Pine Superpave gyratory compactor. The GS is the summation of shear energy increments exerted on the sample during compaction. The GS describes the ability of asphalt mixtures to resist rutting, and it can be determined during the mix design stage using the gyratory compaction data. In addition, this study examined the sensitivity of various compaction and stability indices including GS, construction densification index (CDI), locking point (LP), workability energy index (WEI), and laboratory compaction index (LCI) to changes in mix composition (e.g., RAP content, RAP source, binder content, binder grade, etc.). Such indices can be used as quick, indirect indicators of variation in mix production when placed in the field. The results demonstrated that GS, construction densification index (CDI), and laboratory compaction index (LCI) were sensitive to the binder content; however, all the compaction indices were less sensitive to the change in the RAP content and binder grade. The GS and CDI decreased with the increase in binder content while the LCI increased with the increase in binder content which indicates less energy is needed for compaction. There were fair correlations between the rut depth and both GS and CDI; however, the LCI showed better correlation with the rut depth.
There is a need to evaluate and implement a balanced (engineered) mix design concept for asphalt mixes prepared with high RAP content to ensure that such mixtures have adequate resistance to cracking and rutting. Increasing the amount of RAP in asphalt mixtures offers environmental benefits and cost savings. This study demonstrated that the mix design of asphalt mixtures with 50 percent RAP can be optimized by increasing the binder content to improve the cracking resistance. The results demonstrated the improvement on the cracking resistance when increasing the binder content which indicates the importance of implementing the balance mix design when incorporating RAP materials in asphalt mixes. Furthermore, this study examined the sensitivity of various cracking performance indicators to the change in mix composition as well as the variability and correlation between different performance indicators. The results demonstrated that the Indirect Tensile (IDT) modulus (IDTModulus) and Indirect Tensile (IDT) strength (IDTStrength) were able to capture the change in binder content, binder grade, and RAP content. These two parameters can be used to assess the variation in mix composition during field production and operations. The second phase of this study focused on the development of a regression model to predict the IDT load-displacement curve which is used to calculate many monotonic cracking performance indicators to assess the performance of asphalt mixtures. The cracking performance indicators include IDTStrength, fracture energy (Gf), WeibullCRI, IDEAL-CTIndex, Cracking Resistance Index (CRI), and Flexibility Index. This study examined the effect of mix composition on the IDT load-displacement curve and developed statistical regression model to describe the curve.
Asphalt mixtures with different compositions (e.g., binder grade, binder content, aggregate gradations) were prepared and tested using the IDT. The IDT regression model includes parameters that describe mix composition and properties such as binder content, aggregate gradation, complex shear modulus and phase angle of asphalt binders. The results of the proposed model correlated well with the experimental measurements. The monotonic cracking performance indicator calculated from the model also had strong correlations to the ones calculated from the measured IDT load-displacement curve. The proposed model can be used as a tool to check the preliminary performance of the asphalt mixtures during the mix design stage and mix composition can be adjusted to improve its performance (e.g., adjust binder grade, binder content, gradation, etc.).
The last phase of this study further developed a Multi-Stage Semi-Circular Bending Dynamic Test (MSSD) that has been recently proposed by the researchers at University of Idaho. The MSSD was found to overcome the limitations associated with using monotonic testing when evaluating the cracking performance of samples extracted from the field. The dynamic testing simulates the repeated traffic loading applied on asphalt pavements in the field; therefore, the dynamic testing provides accurate assessment of fatigue cracking performance. The MSSD applies a fixed loading sequence to test different asphalt mixture samples including laboratory-prepared samples and field cores.
The MSSD test was conducted with testing equipment that did not measure the crack opening and collected a limited number of data points which limited the analysis of the test results. There is a need to further develop the MSSD test and use an advanced system with enhanced capabilities that would enable the researchers to collect more data for advance the analysis of the test results.
The MSSD results demonstrated that mixtures with higher density (less air voids) exhibited better resistance to cracking or longer fatigue life. In addition, the results clearly demonstrate the limitations of monotonic cracking performance indicators to assess the cracking resistance of asphalt mixture samples with different air void. Caution should be exercised when using the IDT test. Finally, the correlation between the revised MSSD parameters and original MSSD parameters was found to have a strong relationship with R2 of 0.83 and 0.96, respectively.