Rationality and Applicability of High-Temperature Performance Indexes for Polymer-Modified Asphalts

Abstract

At present, the rationality and applicability of various high-temperature performance evaluation indexes for polymer-modified asphalts have not been comprehensively explored. This study aimed to fully explore their rationality and applicability. For this purpose, five asphalt binders including four representative polymer-modified asphalts and one neat asphalt (NA) were chosen, and many of their high-temperature performance indexes were first obtained by using a series of tests. Then the five corresponding asphalt mixtures were fabricated and their high-temperature performance indexes were also obtained by performing a wheel tracking test. On this basis, the high-temperature potential of asphalt binders was evaluated using the obtained indexes, and the correlation between the indexes of asphalt binders and their mixtures was analyzed via data processing software. Results show that the rankings of high-temperature performance of these five asphalt binders are not exactly consistent when all the high-temperature performance indexes—including penetration (P) at 25°C, penetration index (PI), equivalent softening point (T800), softening point (SP), dynamic viscosity (DV) at 60°C, rutting factor (G*/sinδ), improved rutting factor [G*/(sinδ)9], Shenoy’s nonrecoverable compliance (JShenoy), critical temperature of rutting factor (TG*/sinδ), nonrecoverable creep compliance (Jnr), creep modulus (GV), zero shear viscosity (ZSV), and generalized dynamic shear modulus (GDSM)—are utilized for evaluation. However, all the indexes can distinguish the polymer-modified asphalts from NA well. Among these indexes, the evaluation results of Jnr at different temperatures are consistent at the same stress level (0.1 or 3.2 kPa), and all the correlation coefficients between the Jnr and dynamic stability (DS) of asphalt mixtures exceed 0.9; thus, the Jnr is recommended as the dominant evaluation index. The G*/sinδ, TG*/sinδ, and GV correlate weakly with the DS, and the ZSV performs poor repeatability in the test; thus, their rationality needs to be further verified using more asphalt samples.