The United States has more than 2.7 million miles of paved roads, of which 94\% are surfaced with asphalt pavement. The resilience and durability of asphalt materials have important consequences for transportation safety. Previous research showed that the porosity, i.e. the fraction of air voids in an asphalt pavement, which is largely influenced by the compaction during the installation process, has a significant influence on the durability of installed asphalt pavements. Therefore, understanding the compaction process of asphalt mixtures has become an essential topic of research. However, the existing modeling approaches are mostly phenomenologically based. Very few studies have focused on developing a physics-based predictive model for the compaction of asphalt mixtures. The development of a physics-based computational model is complicated by the complexity and variability of the asphalt mixture. Asphalt mixtures consist of (1) aggregates (sand, pebbles, and rocks) up to 3\ cm in size, (2) fine aggregate mixtures or FAM consisting of the sand portion of the aggregates, asphalt binder, and other additives coats. During the compaction process, the FAM surrounds the coarser aggregates and ultimately as the mixture cools and solidifies, binds them like glue. The details of each component vary considerably across the country. Part of the difficulty in modeling the compaction of such a complex multiphase mixture is to developing reliable rheology for the constitutive behavior of the mixture. In this study, we developed a multi-scale discrete element method (DEM) model for compaction of asphalt mixtures. The model is anchored by the representation of the asphalt as a two-phase mixture: (1) liquid-like FAM and (2) individual gravel particles. On the macroscopic level, only coarse (large) aggregates are considered in the simulation as non-spherical particles. The interaction between these aggregates is mediated both by the coarse particle properties and the properties of the interstitial fluid-like slurry FAM. We derive the dependence of the FAM rheology to the fluid properties of the asphalt binder and the solid properties of the finer particles using discrete element model (DEM) simulations. We use larger scale DEM simulations with coarse aggregates and the modeled FAM to model the gyratory compaction process of hot mixed asphalt with different viscosity of asphalt binder and different aggregate size distributions. The results of the thesis are comprised of three primary components described in this thesis: (1) the small scale model of particles and fluid which provide more macroscale and particle scale information about slurry flow behavior; (2) the larger multi-scale model framework of the asphalt compaction process itself as a process. The results can provide a systematic method for improving the mix design of asphalt mixtures and the compaction procedures toward a more efficient compaction process.
University of Minnesota Ph.D. dissertation.April 2019. Major: Civil Engineering. Advisors: Jia-Liang Le, Kimberly Hill. 1 computer file (PDF); xi, 143 pages.
Rheology of Granular-Fluid Systems and Its Application in the Compaction of Asphalt Mixtures.
Retrieved from the University of Minnesota Digital Conservancy,
Content distributed via the University of Minnesota's Digital Conservancy may be subject to additional license and use restrictions applied by the depositor.