Tianhao Yan

1. Asphalt binder is a thermoplastic adhesive that binds pavement constituents—mainly mineral aggregates. Although it accounts for only about 5% of the weight in asphalt mixtures, the binder strong...

1. Fourier-Transform Infrared (FTIR) spectroscopy is a powerful technique for chemical characterisation of materials (Griffiths & Haseth, 2006). FTIR's rapid testing procedure and accessibility mak...

The field density of as-constructed asphalt pavements usually exhibits considerable variability. Understanding and quantifying this variability is of crucial importance for building more reliable and durable pavements. This study investigates the probability distribution of the field density based on an analytical model of gyratory compaction. The model is derived from mass balance and the transition state theory for the process of aggregate rearrangement. Meanwhile, the variability in the compaction effort is also taken into account. The model is first calibrated by the laboratory gyratory compaction experiments, and is then used to predict the probability distribution of the field density, which is compared with field measurements. It is shown that the probabilistic model can capture the left-skewed and leptokurtic features of the field density distribution. The present model surpasses the commonly used Gaussian distribution for characterising the field density distribution. A parametric study is performed to investigate the effects of the compaction properties of mixture and field compaction effort on the probability distribution of the field density. The model can potentially be used to guide the asphalt mix design for achieving the desired field density.

Gyratory compaction has widely been used to evaluate the compactability of hot asphalt mixtures. Existing efforts on modeling of gyratory compaction have largely been devoted to sophisticated high-fidelity numerical simulations. This paper presents a one-dimensional nonlocal model for gyratory compaction. The model is anchored by the principle of mass conversation, in which the local densification rate is formulated as a function of the nonlocal packing fraction. The nonlocal model involves a material characteristic length scale, which is independent of the specimen size. The nonlocality gives rise to strong effects of the specimen height on the overall compaction curve as well as on the profile of the local packing fraction. A set of gyratory compaction experiments is performed on specimens of different heights. It is shown that the model is able to capture the measured size effect on the compaction curves. A parametric study is carried out to investigate the effects of nonlocality and model parameters on the predictions of compaction curve and profile of packing fraction.

High field density helps in increasing the durability of asphalt pavements. In a current research effort, the University of Minnesota and the Minnesota Department of Transportation (MnDOT) have been working on designing asphalt mixtures with higher field densities. One critical issue is the determination of the N_design values for these mixtures. The physical meaning of N_design is discussed first. Instead of the traditional approach, in which N_design represents a measure of rutting resistance, N_design is interpreted as an indication of the compactability of mixtures. The field density data from some recent Minnesota pavement projects are analyzed. A clear negative correlation between N_design and field density level is identified, which confirms the significant effect of N_design on the compactability and consequently on the field density of mixtures. To achieve consistency between the laboratory and field compaction, it is proposed that N_design should be determined to reflect the real field compaction effort. A parameter called the equivalent number of gyrations to field compaction effort (N_equ) is proposed to quantify the field compaction effort, and the N_equ values for some recent Minnesota pavement projects are calculated. The results indicate that the field compaction effort for the current Minnesota projects evaluated corresponds to about 30 gyrations of gyratory compaction. The computed N_equ is then used as the N_design for a Superpave 5 mixture placed in a paving project, for which field density data and laboratory performance test results are obtained. The data analysis shows that both the field density and pavement performance of the Superpave 5 mixture are significantly improved compared with the traditional mixtures. The results indicate that N_equ provides a reasonable estimation of field compaction effort, and that N_equ can be used as the N_design for achieving higher field densities.

Compaction of asphalt mixtures is a critical component of the construction process of asphalt pavements. The quality of compaction has a significant consequence for the durability of asphalt pavements. Compaction of asphalt mixtures is a complex physical process, which has not been fully understood. In this study, we investigate the physical mechanisms of compaction, based on which we propose a new method to evaluate the compactability of asphalt mixtures. Two mesoscopic physical mechanisms are introduced. One is related to the jamming of aggregates, which governs the densification process of the mixture. The other is related to the binder-aggregate interaction, which is responsible for the change of shear resistance of mixtures during compaction. Based on these mechanisms, different indices are proposed to characterise the compactability of asphalt mixtures. The model is applied to analyse seven asphalt mixtures that were used to construct test sections at MnROAD research facility. Statistical analysis is performed to identify correlations between the compactability indices and material compositions, such as gradation and binder content. Based on the most significant correlations, multiple linear regression models are developed, which can be used to design more compactable mixtures.