Estudo do efeito da variação no teor de emulsão asfáltica na resistência ao cisalhamento e no comportamento viscoelástico de um solo misturado com elevados teores de emulsão

Abstract

The mixture of soils with asphalt emulsions is a well-established stabilization technique widely employed in pavement construction. The addition of high concentrations of asphalt emulsion in these mixtures has resulted in modifications of soil properties, enabling its application in various types of engineering projects, beyond those in which it was traditionally employed. This research was conducted to assess the effect of asphalt emulsion content on the rheological properties and shear strength of a sandy soil mixed with asphalt emulsion. The experimental methodology involved: (i) soil collection and characterization, and selection of the asphalt emulsion; (ii) mixing the materials at emulsion dosages of 16%, 22%, and 28% by mass; (iii) compacting the mixtures usingMCT and Superpave methods; (iv) triaxial compression; and (v) complex modulus and creep tests. The results of the strength tests revealed that as the emulsion dosage increased, the friction angles decreased, while cohesion values were higher for mixtures with 16%and 22%dosages. However, in the mixture with 28%emulsion, there was a reduction in its cohesive intercept due to the separation of particles caused by an excess of binder. The rheological model adopted to characterize this material was the 2S2P1D, with satisfactory fitting. Behavior prediction was conducted for mixtures with 16% emulsion, after its linear viscoelastic characterization through complex modulus tests with maximum strains of 67.5 m/m, as well as the corresponding comparison with the results obtained directly by the creep test, performed with deformations up to 20 times greater. Prediction errors were found ranging from 44% to 61% of difference, which are consistent with granular materials. Once the validity of the model was confirmed, the mixtures with 22% and 28% emulsion dosages were further characterized. It was observed that the temporal-dependent parameters (k, h,  E and of the 2S2P1D model remained the same for all three mixture dosages, indicating their association with the type of binder used. Regarding the values of E0 and E00, a clear correlation was observed between the amount of binder in the mixtures and these values. As a first approximation, a linear relationship was established among these values. Through this study, it becomes feasible to perform an initial determination of a characterization coefficient for the behavior of soil-emulsion mixtures without the requirement for extensive experimental investigations, thereby enabling prospective assessments regarding structural risks attributable to creep. In the forthcoming research, emphasis should be placed on evaluating the microstructure of these mixtures and on the formulation of a micromechanical analysis that corroborates the findings obtained.