Investigation Of The Fatigue Property And Internal Structure Of Asphalt Mixtures Based On Multiscale Methods

The asphalt mixture,consisting of coarse aggregates,mastic,mortar and air void,is an in-homogenously visco-elastoplastic material.The accumulation and development of the fatigue damage of asphalt mixtures is extremely complex,because of the irregular internal structure and complex mechanical properties of different compositions.The multiscale modeling of asphalt mixtures considers it not as a homogeneous mass,but rather as an assemblage of materials at different characteristic length scales.It explains the macroscopic mechanism of asphalt mixtures by the properties of materials at smaller length scale.However,the application of the multiscale model in asphalt mixtures has no uniform standard,including the definition of different characteristic scales for asphalt mixtures,the laboratory testing methods for asphalt mixtures at different length scales and the internal structures for the asphalt mixtures at each scale.Therefore,it is critical to investigate the testing methods for asphalt mixtures at different length scales as well as definition or quantification of the internal structure of asphalt mixtures as each scale.Furthermore,it is a fundamental work to build a multiscale framework for the evaluation of asphalt mixtures,understand the interaction between the fatigue behavior and internal structure of asphalt mixtures at different scales.At first,the separation of mastic,mortar and asphalt mixtures are defined.The materials design of mastic,mortar and mixtures are determined by the surface ratio calculated by the image analysis methods and the physical parameters of different compositions.Meanwhile,according to the properties of materials at different scales,a repeatable compaction method for mastic,mortar or mixtures is determined by controlling the air void content.Secondly,the definition or quantification of the internal structure of asphalt mixtures is discussed.The laser diffraction method,scanning electron microscope(SEM)and optical camera imaging are utilized to acquire the 2D image of mastic,mortar and mixtures based on sample cutting and scanning or virtual construction.The internal structure of mastic,mortar or mixtures are quantified by the thickness distribution curve of the dispersion phase at different scales based on image analysis methods.Indexes,including expected value,mode,peak value and uniformity index,are proposed with reliable error test.Thirdly,the fatigue test methods at different scales are explored and test conditions and fatigue indexes at different scales are determined.The results show that the fatigue and healing indexes of asphalt mixtures at intermediate temperature could be obtained by the non-notched semi-circular bending(SCB)test at monotonic load,cyclic load and cyclic load with a certain rest time.The master curve of asphalt mastic or mortar could be obtained by the time-temperature sweep(TTS)test conducted within the linear visco-elastic(LVE)zone.The fatigue parameters of asphalt mortar and mastic could be deduced based on the visco-elastic continuum damage(VECD)model and the corresponding linear amplified sweep(LAS)test conducted within certain strain ranges.Finally,the relationship between the visco-elastic parameters,fatigue parameters as well as the internal structure indexes at different scales are analyzed.The mechanism between the mechanic properties and internal structure properties from mastic,mortar and mixtures are correlated.The results show that the fatigue parameters of mastic could correlated with the visco-elastic parameters and the index from the binder thickness very well.According to the multi-linear regression method,the prediction models for the fatigue parameter of mortar based on fatigue index and thickness index of mastic,and the fatigue parameter of mixture based on fatigue index and thickness index of mortar could be established.Overall,this research provides a solid foundation for future research in exploring the fatigue property of asphalt mixture based on multiscale perspective.