Validation of SHRP asphalt mixture specifications using accelerated testing

James Lincoln Stiady, Purdue University

Abstract

The overall objectives of this study were to validate various HMA aggregate specifications and volumetric relationships established by Superpave. Specifically, the study addresses the effects of voids in the mineral aggregate (VMA), fine aggregate angularity (FAA), and HMA mixture gradation on rutting performance of mixtures designed using the Superpave volumetric design procedure. Mixture designs were conducted using a single neat PG64-22 binder, two nominal maximum aggregate sizes (19 and 9.5mm), two coarse aggregate types (limestone and granite), three fine aggregate types possessing FAA values of 39, 44, and 50, and three gradation types (above, through, and below the restricted zone). The rutting performance of the mixtures were evaluated using the INDOT/Purdue Accelerated Pavement Test Facility (APT) and the Purdue Laboratory Wheel Track Test Device (PURWheel). In addition, triaxial testing was utilized to identify critical VMA. Test results indicate that APT, PURWheel, and Triaxial test are sensitive to material property changes. The observed critical VMA was higher than the Superpave minimum VMA requirements. VMA could not be correlated with performance when VMA resulted from deficiencies in mixture production and/or placement. Film thickness was identified as a robust mixture property that correlated better with rutting performance than traditional volumetric properties. Fine aggregate angularity was found to correlate well with performance, but incorporating very high fine aggregate angularity did not necessarily provide better performance. Gradation influences APT, PURWheel, and triaxial test results. However, the restricted zone alone may not be adequate to characterize gradation to ensure acceptable performance. The asphalt binder content corresponding to optimum triaxial strength and PURWheel rutting performance ranged from 0.0 to 0.5 percent less than the Superpave design asphalt binder content for most mixtures. Based on all test results and analyses, it is suggested that film thickness be included in mixture design procedures with an acceptable range of 7 to 9 microns. Despite the importance of VMA in describing the HMA structure, stringent VMA criteria are not suggested because many factors could impact VMA during the construction process, such as asphalt binder content and in-place density. These factors could deceive the essence of VMA as a durability or HMA structure parameter. The use of a typical value of FAA is suggested for acceptable rutting performance. The mixture design and performance data suggest that achieving 4.0 percent air voids, the level at which the design asphalt binder content is selected in the Superpave mixture design process, when using high FAA (50) blends can lead to excess asphalt and thus poor rutting performance. Despite the importance of gradation in building aggregate structure, the selection of gradation with respect to the Superpave restricted zone as a requirement for performance is not suggested because equally adequate performance can be achieved with gradations plotting above, through, and below the restricted zone. Finally, a range of dust proportion of 0.9 to 1.7 is suggested for optimum rutting performance.

Degree

Ph.D.

Advisors

White, Purdue University.

Subject Area

Civil engineering

COinS