The importance of the creep and recovery times on the rheological behavior and the susceptibility of modified asphalt binders to rutting

• Autor(a) principal: Domingos, Matheus David Inocente
• Data de Publicação: 2017
• Tipo de documento: Tese
• Idioma: eng
• Título da fonte: Biblioteca Digital de Teses e Dissertações da USP
• Texto Completo: https://www.teses.usp.br/teses/disponiveis/18/18143/tde-22122022-182306/
• topic: creep time; especificação Superpave; grau de desempenho; ligantes asfálticos modificados; modelo de potência; modified asphalt binders; parâmetros de deformação permanente; performance grade; power model; recovery time; rutting parameters; Superpave specification; tempo de fluência; tempo de recuperação

The performance-related Multiple Stress Creep and Recovery (MSCR) test and its corresponding rutting parameter nonrecoverable compliance Jnr – have advantages over the dynamic oscillatory shear test and the parameter G*/sinδ, but some limitations such as the limited applicability of the standardized creep-recovery times to modified binders have been a matter of concern in the literature. In an attempt to overcome such deficiencies, this study was conducted to quantify and evaluate the creep-recovery behavior of several modified binders equally graded in the Superpave® high temperature specification (PG 76-xx) with respect to their responses at different loading and unloading times (1/9, 2/9, 4/9, 8/9, 1/240 and 1/500 s). Five temperatures were selected in the study (52, 58, 64, 70 and 76°C) and 12 formulations were prepared from a 50/70 original material: AC+PPA, AC+Elvaloy+PPA, AC+rubber, AC+rubber+PPA, AC+SBS, AC+SBS+PPA, AC+EVA, AC+EVA+PPA, AC+PE, AC+PE+PPA, AC+SBR and AC+SBR+PPA. Dense-graded mixtures were also prepared with these binders and tested at 60°C to obtain the flow number (FN) values. A modified power model with four parameters (A, B, n and α) was selected to fit the binder data at longer creep and recovery times. The levels of correlation between FN and the binder data were high either for Jnr or GV (viscous component of the creep stiffness), and the rankings of binders and mixtures are almost the same for both parameters. These correlations became even better when the creep-recovery times of 2/9 s and 4/9 s were used, and it is believed that the steady state phenomenon played a role on the findings of the binder. Despite the identification of specific limitations, the parameter α may be used to have an estimation of the amount of percent recovery (R) in the binder, especially at standardized and longer loading times. The use of PPA together with a major additive in the binder (SBS, EVA, PE and SBR) tends to increase the rutting resistance of the formulation when the unloading time is equal to 9 s, but the AC+rubber and the AC+rubber+PPA are exceptions. With exception of the AC+Elvaloy+PPA, the AC+rubber and the AC+EVA (decreases in A and B with temperature), the increases in the accumulated strain in the binders with increasing creep time may be explained by the presence of higher strain rates (B values) and, in some cases, the nonlinear parameter (n) also gives some contribution to it. In terms of the binders with low levels of elasticity, the recovery time of 240 s may not be appropriate for all types of modifiers (especially the AC+rubber, the AC+SBR and the AC+SBS). With respect to the unloading time of 500 s, the AC+EVA showed recoveries equal to or approaching 100% at 100 Pa and temperatures no greater than 64°C, and the variables of the DSR may have affected the outcomes of materials such as the AC+Elvaloy+PPA. A refinement in the Superpave® specification criteria for assigning a traffic level to the binder is proposed, and these new criteria could help in choosing the best material for a specific paving application.