Civil Engineering ETDs

Publication Date

7-12-2014

Abstract

Dynamic modulus (E*) test has gained substantial acceptance in recent years for evaluating Hot Mix Asphalt (HMA) rutting and fatigue cracking performances. Indeed, the recently developed Mechanistic Empirical Pavement Design Guide (MEPDG) uses E* data for stress and strain calculation, and E*-based models for prediction fatigue and rutting performances of a HMA pavement. Unfortunately, neither HMA mix design method nor the MEPDG uses E* value to evaluate moisture damage characteristics of HMA pavements. In fact, moisture damage is a very complex problem, and there exists no models for inclusion in the MEPDG for predicting moisture damage performance of HMA pavements. Rather, moisture damage performance of HMA is evaluated during the mix design state using the AASHTO T 283 test. According to AASHTO T 283 method, indirect tensile strength ratio of wet to dry samples is used as an indicator of moisture damage performance of a HMA mix. The AASHTO T 283 is a simple method, but it is not very reliable. In some cases, it gives false negative and positive values about the moisture damage characteristic of an HMA sample. To this end, this study attempts to study moisture damage in Asphalt Concrete (AC) sample by determining the loss in dynamic modulus (E*) value based on continuum damage mechanics. The loss in E* is evaluated by calculating the ratio of wet E* to dry E* (DMR). In this study, E* testing is performed on five different asphalt mixtures. Three of the asphalt mixtures are plant produced Superpave mixes, and the other two are laboratory prepared HMA mix. The plant produced mixes include SP-II mix with asphalt binders Performance Grade (PG) PG 64-22, SP-III mix with PG 70-22, and SP-IV mix with PG 70-22. The laboratory mixes are SP-III mix with PG 64-22, and SP-II with binder PG 70-22. The E* test is conducted at five temperatures (-10, 4, 21, 37, 54 °C) and six frequencies (25, 10, 5, 1, 0.5, 0.1 Hz). The resulting E* data are used to generate time-temperature mastercurves. Also, the laboratory E* data is fitted to several models such as the viscosity (\u03b7) -based Witczak model, the dynamic shear modulus (G*) -based Witczak models, and the Hirsch model. However, all models are found to underpredict the laboratory E* values. Therefore, the \u03b7-based Witczak model is modified to improve E* prediction accuracy in this study and E* ratios are used for evaluating moisture damage in AC. As mentioned above, the most common test method for evaluating moisture damage susceptibility of HMA mix is the AASHTO T 283. According to the AASHTO T 283, a dry asphalt concrete sample is vacuum saturated and then subjected to one cycle of freezing and thawing. However, one cycle of freeze-thaw does not simulate moisture condition resulting from repeated pore water pressure buildup and scouring cycles that occur when vehicle tires pass over a saturated pavement. Recently, Moisture Induced Sensitivity Tester (MIST) equipment can apply repeated pore pressure cycles inside an AC sample in the laboratory. Therefore, MIST device is employed in this study to cause damage in AC samples. Moisture damage resulting from MIST conditioning is then evaluated using the Dynamic Modulus Ratio (DMR) of wet to dry samples. MIST conditioning is performed at three different pressures (40, 55, 70 psi), three temperatures (40, 50, 60 \xbaC) and three different numbers of cycles (3500, 7000 and 10,500). MIST conditioning is found to decrease the E* value of asphalt concrete indicating the presence of moisture damage. The average DMR is observed to decrease from 1.0 to 0.85, 0.46 and 0.56 for MIST conditioning at 3500, 7000 and 10500 cycles, respectively. When the MIST conditioning temperature is increased to 40, 50 and 60 \xbaC, the average DMR decreases from 1.0 to 0.90, 0.82 and 0.76, respectively. The DMR at 276, 376 and 483 kPa MIST conditioning pressures are determined to be 1.0, 0.91 and 0.72, respectively. Therefore, moisture damage increases with an increase in number of cycles, temperatures and pressures. However, it is shown that temperature affects moisture damage less significantly than the number of cycles and pressures in a MIST device. Therefore, number of traffic and magnitude of traffic loading should be considered carefully to design moisture damage free pavements.'

Keywords

Dynamic, Modulus, Modeling, Moisture, Damage

Sponsors

New Mexico Department of Transportation

Document Type

Dissertation

Language

English

Degree Name

Civil Engineering

Level of Degree

Doctoral

Department Name

Civil Engineering

First Committee Member (Chair)

Tarefder, Rafiqul

Second Committee Member

Maji, Arup

Third Committee Member

Taha, Mahmoud

Fourth Committee Member

Khraishi, Tariq

Share

COinS