SWUTC Research Project Description

Title of Project:  Physically-based Model for Predicting the Susceptibility of Asphalt Pavements to Moisture-Induced Damage

Project Number:  476660-00012

Principal Investigator:
Eyad Masad
(979) 845-8308
P.I. Affiliation:  Texas A&M University
emasad@civilmail.tamu.edu

Project Monitor:
Mikhail Magdy
TxDOT Cedar Park
Austin, Texas 
Phone: 512-506-5838
mmikhail@dot.state.tx.us

Project Status:  Active

Date Started:  9/1/08

Estimation Completion Date:  8/31/09

Estimated Cost - Current Fiscal:  $72,750

Estimated Cost - Total Planned:  $72,750

Project Summary:
Project Abstract:
The focus of the proposed research will be on the development of a physically-based model that can be used effectively by pavement engineers for predicting the susceptibility of asphalt pavements to moisture-induced damage under mechanical (traffic and restraint conditions) and environmental (moisture, external and internal relative humidity, thermal) loading conditions. This model will be developed based on the laws of thermodynamics and continuum damage mechanics (CDM) frameworks. CDM is a framework for modeling the nucleation, growth, and propagation of numerous micro-cracks and their evolution into macro-cracks that ultimately leads to failure. CDM can be effectively used in predicting the onset (site and time or where and when) of damage nucleation (cracking potential) and its evolution (crack propagation). Furthermore, a computational model that can effectively be used in estimating the moisture content, temperature, and relative humidity due to wet-dry cyclic conditions will be integrated into the damage model for better prediction of the level of moisture-induced damage. The following physio-mechanical damage processes will be incorporated into the developed model: (i) weakening of the mastic and the aggregate-mastic bond due to the existence and diffusion of moisture, and (ii) erosion of the mastic film due to water flow during traffic.

Project Objectives:
The primary objectives of this study are:

• Development of a computational model for the fundamental analysis of combined mechanical and moisture induced damage of asphalt pavements based on the laws of thermodynamics and CDM.
• Development of an integrated computational model for estimation of moisture, relative humidity, and temperature beneath asphalt pavements.
• Integrate the fundamental material properties into the developed damage model.
• Demonstrate the ability of the model to predict the performance of a selected group of asphalt pavements with known performance.
• Conduct a detailed parametric study of the influence of asphalt mix composition on their susceptibility to moisture-induced damage.

Task Descriptions:
Task 1. Literature Review and State-of-the-Art
A comprehensive review will be conducted in this task to identify the existing constitutive models and computational methods (i.e. fracture mechanics, continuum damage mechanics, thermodynamics, and micromechanics) that are used in predicting moisture-induced damage in asphaltic pavements and other materials. Moisture damage is a complex phenomenon that involves thermodynamic, chemical, physical, and mechanical processes. Thus, this task will identify the models that are currently used in assessing all of these aspects. Moreover, a review of existing theories on the adhesive bond between aggregates and asphalt binders and the effect of the presence of moisture at the interfaces will be conducted.

Task 2. Development of a Moisture-Induced Damage Model
Based on the frameworks of thermodynamics, continuum damage mechanics, fracture mechanics, and stochastic micromechanics, physio-mechanical constitutive models for asphalt pavements under combined mechanical and environmental (moisture and temperature) loading conditions will be developed.  CDM is a framework for modeling the nucleation, growth, and propagation of numerous micro-cracks and their evolution into macro-cracks that ultimately leads to failure [4]. CDM can be effectively used in predicting the onset (site and time or where and when) of damage nucleation (cracking potential) and its evolution (crack propagation).  Stochastic micromechanics is used to represent the known dependence on binder film thickness of the transition from adhesive to cohesive fracture and healing and their dependence on the fracture and healing bond energies.  These bond energies, which are thermodynamic properties of surfaces, depend upon the surface energies of the micro-crack surfaces and the water vapor pressure on them. Furthermore, a computational model that can be used effectively in estimating the moisture content, temperature, and relative humidity due to wet-dry, hot-cold cyclic conditions will be integrated into the damage model for better prediction of the level of moisture-induced damage. The following physio-mechanical damage processes will be incorporated into the developed model: (i) weakening of the mastic and the aggregate-mastic bond due to the existence and diffusion of moisture, both through the films surrounding the aggregate particles and along the particle-mastic interfaces and (ii) erosion of the mastic film due to jetting water flow imposed by passing traffic.

Finally, the developed mechanical and moisture-induced damage constitutive relations will be coupled to an existing viscoelastic-viscoplasticity model [5] that can be used to predict the time-dependent behavior of asphaltic pavements under various types of mechanical loading conditions. Therefore, through this coupling, a complete physically-based model that can be used to predict various aspects of the nonlinear, inelastic, damage, and time-dependent behavior of asphaltic materials will be developed.

Task 3. Implementation in the Finite Element Code Abaqus
The developed constitutive relations will be implemented into the well-know commercial finite element code Abaqus [3] via the user material subroutine UMAT. In order to achieve this task, a numerical algorithm that is based on viscoelastic predictor and viscoplastic-damage correct steps will be formulated and implemented. The numerical algorithms in [5] for viscoelastic-viscoplastic constitutive equations will be utilized, whereas the numerical algorithm in [6] for coupling to damage will be used.

Task 3. Calibration and Validation of the Proposed Model
The developed constitutive models and computational algorithms will be calibrated and validated against a large amount of available experimental data. Identification of material properties that will be required as inputs to the models include: (i) properties for mineral aggregates such as particle size distribution, particle surface area, surface free energy, elastic modulus, and fracture toughness; and (ii) properties for asphalt binders include viscoelastic properties, complex modulus, and surface free energy. In case, a material property of interest is not available from the existing experimental data, then this/these property/properties will be measured at TAMU.

Task 4. Prediction of Moisture-Induced Damage
Once the constitutive models and computational algorithms are calibrated and validated, a parametric study will be conducted to study the influence of various material properties and compositions on the susceptibility of asphaltic mixes to moisture-induced damage. Design recommendations will be developed to enhance the material fundamental properties and composites for mitigating moisture damage.

Task 5. Final Report
A final report will be prepared at the end of the project.  This final report will document the findings from this study including, literature review, a detailed description of the models developed to predict moisture-induced damage, and design recommendations for mitigating different types of asphalt distress.

Index Terms:
Asphalt pavements, Moisture damage, Moisture content, Pavement distress, Pavement performance, Pavement cracking, Microcracking, Continuum damage, Research projects