Aggregate Packing Characteristics of Asphalt Mixtures
Voids in the mineral aggregate (VMA), as a main volumetric design parameter in the Superpave mixture design method, is an important factor to ensure asphalt mixture durability and rutting performance. Moreover, an asphalt mixture’s aggregate skeleton, related to VMA, is another important factor that affects critical asphalt mixture properties such as durability, workability, permeability, rutting, and cracking resistance. The objective of this study is to evaluate the effects of aggregate size distribution and shape parameters on aggregate packing characteristics (volumetric and compaction properties) of asphalt mixtures. Three tasks were undertaken to reach this goal.
The first task was to propose an analytical approach for estimating changes in voids in the mineral aggregate (VMA) due to gradation variation and determining the relevant aggregate skeleton characteristics of asphalt mixtures using the linear-mixture packing model, an analytical packing model that considers the mechanisms of particle packing, filling and occupation. Application of the linear-mixture packing model to estimate the VMA of asphalt mixtures showed there is a high correlation between laboratory measured and model estimated values. Additionally, the model defined a new variable, the central particle size of asphalt mixtures that characterized an asphalt mixture’s aggregate skeleton. Finally, the proposed analytical model showed a significant potential to be used in the early stages of asphalt mixture design to determine the effect of aggregate gradation changes on VMA and to predict mixture rutting performance.
As the second task, a framework to define and understand the aggregate structure of asphalt mixtures was proposed. To develop this framework, an analytical model for binary mixtures was proposed. The model considers the effect of size ratio and air volume between the particles on the aggregate structure and packing density of binary mixtures. Based on this model, four aggregate structures, namely coarse pack (CP), coarse-dense pack (CDP), fine-dense pack (FDP) and fine pack (FP), were defined. The model was validated using a series of 3D discrete element simulation. Furthermore, the simulation of multi-sized aggregate blends using two representative sizes for fine and coarse stockpiles was carried out to apply the proposed analytical model to actual aggregate blends. The numerical simulations verified the proposed analytical model could satisfactorily determine the particle structure of binary and multi-sized asphalt mixture gradations and could, therefore, be used to better design asphalt mixtures for improved performance.
The third task virtually investigated the effect of shape characteristics of coarse aggregates on the compactability of asphalt mixtures using a discreet element method (DEM). The 3D particles were constructed using a method based on discrete random fields’ theory and spherical harmonic and their size distribution in the container was controlled by applying a constrained Voronoi tessellation (CVT) method. The effect of fine aggregates and asphalt binder was considered by constitutive Burger’s interaction model between coarse particles. Five aggregate shape descriptors including flatness, elongation, roundness, sphericity and regularity and, two Superpave gyratory compactor (SGC) parameters (initial density at Nini and compaction slope) were selected for investigation and statistical analyses. Results revealed that there is a statistically significant correlation between flatness, elongation, roundness, and sphericity as shape descriptors and initial density as compaction parameter. Also, the results showed that the maximum percentage of change in initial density is 5% and 18% for crushed and natural sands, respectively. The results of analysis discovered that among all particle shape descriptors, only roundness and regularity had a statistically significant relation with compaction slope, and as the amount of roundness and regularity increase (low angularity), the compaction slope decreases. Additionally, the effect of flat and elongated (F&E) particles percentage in a mixture using a set of simulations with five types of F&E particles (dimensional ratios 1:2, 1:3, 1:4 and 1:5) and ten different percentage (0, 5, 10, 15, 20, 30, 40, 50, 80 and 100) with respect to a reference mixture containing particles with flatness and elongation equal to 0.88 was conducted. Results indicated that increase of F&E particles in a mixture (more than 15%) results in a significant reduction in the initial density of the mixture especially for lower dimensional ratio (1:4 and 1:5).