A multiscale approach for predicting freeze-thaw damage in asphalt mixtures
Time: Mon 2021-06-14 13.00
Subject area: Civil and Architectural Engineering Structural Engineering and Bridges
Doctoral student: Lisa Lövqvist , Bro- och stålbyggnad
Opponent: Professor Markus Oeser, RWTH Aachen University
Supervisor: Professor Nicole Kringos, Bro- och stålbyggnad; Docent Romain Balieu, Bro- och stålbyggnad
Abstract
Cold and wet climates are a known threat to asphalt pavements, causing both an embrittlement of the material, making it more prone to failure, and increasing the risk of damage due to the presence of moisture in the material and freeze-thaw cycles causing expansion and contraction of the air voids as the moisture inside them freezes and thaws. This damage evolution can lead to an overall decrease of material properties and ultimately showcase itself as raveling, pot holes,and cracks. To minimize the effect of the combination of these damage mechanisms, it is important to be able to understand and characterize the damage initiation and evolution. While this is commonly done through empirical knowledge and experimental tests, these tests have limitations regarding both the possibility to separate the damage mechanisms from each other to identify the dominant parameters and the long time scales which are required to accurately reflect the environmental conditioning in the field. Instead, mathematical and numerical modeling can be a good supplement to better characterize the damage development and understand which parameters are most dominant. The work presented in this thesis has therefore focused on developing a numerical model which is able to accurately predict the damage development due to the combination of environmental effects such as moisture and freeze-thaw cycles and other loads from the traffic.
To enable an understanding of how the different material components affect the damage development, as well as a prediction of the damage development on the pavement scale, a multiscale model framework is developed which consists of two scales: a local microscale including mastic, aggregates, and air voids, and a global macroscale of the homogenized asphalt mixture. On the microscale, the expansion of the air voids is explicitly modeled and includes its effect on both the adhesive damage in the mastic-aggregate interface and the cohesive damage in the mastic itself. The macroscale does not explicitly model the expansion and contraction but accounts for its deteriorating effect on the global behavior of the asphalt mixture. In addition to the freeze-thaw damage, both models also account for the damage due to the long-term presence of moisture inside it and other external loads, i.e., the traffic, being applied on the material.The two scales are coupled by letting the damage behavior of the macroscale depend on the effective damage behavior of the microscale. The developed model is shown to be able to predict a realistic damage evolution and associated degrading effect on the material properties which is consistent with previously reported experimental data.
In addition to giving an extensive description of the developed model, this thesis also presents how the model can be used. In the first example, the developed model is used to evaluate an existing experimental standard method regarding the resistance of asphalt mixtures to de-icing fluids. The second example regards the importance of including the coupling between the different damage modes, which is highlighted through predicting the damage evolution for a set of different weather scenarios.