Aggregate Fracture in Unbound Road Materials

QC 20260428

Abstract

Mineral aggregate fracture in unbound granular materials (UGMs) beneath asphalt pavement layers affect pavement performance and may accelerate distresses. Improving the understanding, characterization, and quantitative prediction of aggregate fracture in UGMs is therefore essential, particularly to support the use of marginal-quality aggregates in road construction and thereby reduce the demand for high-quality aggregates and lower the environmental impact of road infrastructure.

This thesis introduces a new numerical framework, combined with an experimental study, to predict aggregate fracture in UGMs and quantify its influence on UGM macro-mechanical behavior. The framework is based on the discrete element method (DEM) and enables evaluation of aggregate fracture for varying gradations, loading conditions, and aggregate types. To ensure general applicability, granular mechanics–based contact laws and statistical fracture models are developed and incorporated into DEM.

The model parameters are identified and the framework validated through laterally confined monotonic uniaxial compression tests on UGMs. The tested materials included different aggregate types and gradations and were subjected to different maximum compressive loads. For UGMs composed of crushed granite, the DEM model captures the effects of gradation and load magnitude on both macro-mechanical response and aggregate fracture. To extend the framework to a wider range of aggregates, particularly marginal-quality aggregates, a new particle fracture model is developed that accounts for aggregate shape variability and statistical volume effects on fracture force distributions. The model is evaluated using single-particle crushing tests on four aggregate types and compared with two widely used fracture models, showing improved agreement with measured aggregate strength. When implemented in the DEM framework, the new model improves fracture predictions for UGMs containing marginal-quality aggregates.

The feasibility of using DEM to assess how aggregate fracture affects elastic stiffness and permanent deformation resistance of UGM is evaluated. Emphasis is put on UGMs containing marginal aggregates and on the potential for optimizing pavement structural design to enable their use without excessive performance loss. Blended UGMs containing crushed granite and crushed brick are investigated using confined compression tests and X-ray CT, and the observations are incorporated into the DEM model to predict both macro-mechanical behavior and aggregate fracture.

A systematic analysis of aggregate fracture in UGMs subjected to confined monotonic compression tests is conducted to identify the governing factors of fracture. The results show that aggregate fracture is controlled by the coupled effects of aggregate strength, gradation, and applied load, with the strength and applied load identified as the dominant factors. The results further demonstrate that aggregate fracture in UGMs cannot be fully predicted using standard aggregate strength tests alone and should therefore be evaluated under field-representative gradations and loading conditions. To support this assessment, an approach combining the developed DEM models with experimental measurements is introduced, and a fracture evolution parameter is proposed to quantify the progression of aggregate fracture during compression. The parameter shows strong agreement with experimentally observed aggregate fracture and provides an effective means for characterizing fracture development in UGMs under loading.

In summary, the results demonstrate that the developed DEM framework can quantify how a wide range of aggregate types—including marginal-quality aggregates—along with gradation and loading, affect UGM performance and aggregate fracture, and can support performance-based material selection and pavement design to mitigate aggregate crushing.

Link to DiVA