Evaluation of the MMLS3 for Accelerated Wearing of Asphalt Pavement Mixtures Containing Carbonate Aggregates

The purpose of this study was to develop an accelerated wearing protocol for assessing the susceptibility of asphalt surface mixtures to polishing. This was the second phase of the study. The first phase focused on assessing the characteristics of selected carbonate aggregates available in Virginia that are normally classified as “polishing” and thus not considered suitable for use in pavements except for those roads with an average daily traffic of less than 750 vehicles per day. The selection of aggregates used in pavements is critical in producing surfaces that will continue to provide good skid resistance through a lengthy service life. The specifications of the Virginia Department of Transportation (VDOT) call for “non-polishing aggregate” for use in most surface layers. The study was aimed at making use of locally available polishing aggregates that can reduce the cost of asphalt mixtures while maintaining satisfactory wearing and skid characteristics of the pavements. The objectives of the research were (1) to evaluate the polishing/wear features of mixtures containing limestone aggregate in the laboratory using an accelerated method; (2) to compare friction properties of the laboratory-polished specimens with actual pavement friction measurements; and (3) to compare friction properties of mixtures containing carbonate rock or blends with those of mixtures with non-carbonate rocks. The study included three types of aggregates, i.e., limestone, quartzite, and granite, and blends of these aggregates. The surface mixtures studied were conventional SM-9.5 and SM-12.5 mixtures containing various percentages of limestone, limestone recycled asphalt pavement, and limestone-granite/quartzite blends. The suggested test protocol to evaluate the polishing of asphalt concrete specimens prepared in the laboratory was developed using the third-scale model mobile load simulator (MMLS3). The MMLS3 is capable of applying realistic rolling wheel contact stresses similar to those on highways from the moving traffic. The skid resistance, friction, and texture of actual pavement surfaces and laboratory-fabricated specimens were measured after different polishing intervals. Skid resistance and frictional characteristics were measured by the British pendulum tester, dynamic friction tester, and locked-wheel skid tester; the circular texture meter was used to measure surface macrotexture. Results showed that the MMLS3 can be used to simulate traffic wearing of asphalt concrete specimens of different shapes and sizes in the laboratory including core specimens removed from existing pavements and that the BPT is effective in characterizing changes in friction on specimens that are subjected to simulated trafficking via the MMLS3. Further, test specimens should have a high initial macrotexture and mixtures should have good stability so that the wearing effects are focused on the aggregates. The study recommends that the Virginia Center for Transportation Innovation and Research (VCTIR) work with Virginia Tech and VDOT’s western districts to design and conduct an experiment to explore a series of carbonate / non-carbonate aggregate blends for asphalt mixtures and that the mixture gradations be designed to prevent the absence of macrotexture from impacting the ability to measure the “polish” of the coarse aggregate structure of the experimental mixtures. VCTIR should purchase tires with different tread patterns and try them on the MMLS3 to evaluate the polishing rate of specimens in more detail.