| The thermal coefficient of contraction of asphalt mixtures is one of the required properties in the prediction of thermal cracking of asphalt pavements. Due to the significant influence of asphalt binders on asphalt mixtures' thermal properties, asphalt mixtures are expected to show a significant glass transition behavior. In the current practice, however, the thermal coefficient of contraction of mixtures is assumed to be linear, and determined from a simplistic relationship with binders that take into account volumetric and physical mixture properties. Although measuring thermal properties of asphalt mixtures has been a long-standing research topic for the last decade, a limited number of studies have explored the methods for the direct measurements of these properties. Using the findings of previous studies, a system was developed in this project for the direct measurement of change in linear length of mixtures as a function of change in temperature. The system was used to reveal the significant glass transition experienced by mixtures and to evaluate the effect of mix variables on the thermal properties.; In addition to the glass transition test device, the study was conducted to build a Thermal Restrained Stress Testing device, commonly known as 'Thermal Stress Restrained Specimen Test (TSRST)', to verify the need for thermal volumetric properties by measuring directly the thermal stress build-up in the laboratory. The results collected include the testing data of various mixtures produced with different aggregate sources, aggregate gradations, and different modified binders. The results show that the developed device provides a good indication of the low-temperature cracking resistance and such a device can be used to directly measure the low-temperature cracking characteristics of asphalt mixtures.; To summarize, the importance of this study lies on the use of a prediction model that includes realistic non-linear thermal contraction properties, which was found to match very closely the measured stress build-up and also cracking temperatures. Based on a series of observations, it could be concluded that the best prediction of low-temperature cracking requires the thermal contraction properties in addition to the mixture creep stiffness and failure properties. |
