Quantification of self-heating and its effects under cyclic tests on a bituminous binder

Abstract

This paper studies the reduction in complex modulus when a bituminous binder is subjected to the fatigue test, along with the ability of this material to recover such a loss in modulus. It has been demonstrated in the literature that during a fatigue test, a rapid and reversible decrease of complex modulus can be observed at the beginning of the test. This reversible decrease is not correlated with fatigue and has been explained by the existence of reversible phenomena, such as thixotropy, self-heating, nonlinearity, etc. The aim of this work is to quantify the self-heating and determine its share in the reversible reduction of complex modulus under cyclic loading. A strain sweep test has been developed with different loading steps, followed by rest periods after each loading phase. The loading level increases with each step, and the loading phases allow observing the modulus decrease, while the rest periods are introduced to observe modulus recovery after each loading level. A complete recovery of modulus means that the modulus diminution is due to a reversible phenomenon however a partial recovery of modulus notifies the presence of nonreversible phenomenon (such as damage). In using this approach, the influence of loading level on both complex modulus decrease and recovery can be examined. In order to quantify self-heating during testing, the sample temperature is measured by embedding a thermocouple probe inside the sample during its manufacturing. To obtain the relationship between complex modulus and temperature, the complex modulus test results have been interpreted herein. A thermoviscoelastic numerical simulation was conducted to determine the temperature field in the material and quantify the variation in complex modulus with respect to self-heating. Results show a temperature variation of 11°C inside the sample. A comparison between experimental and numerical results indicates that the numerical model perfectly predicts the variations in both temperature and complex modulus until a certain strain level, where the damage, thixotropy or other factors have not appeared yet. It is also observed that the reversible complex modulus decrease can be explained by self-heating.