Long-Term Aging of Asphalt Mixtures for Performance Testing and Prediction: Phase III Results (2021)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

1   The accurate characterization of the in situ aging of asphalt pavement materials over the service life of the pavement is of utmost importance to the implementation of mechanistic- empirical (ME) pavement design and analysis methods. Materials characterization for aging is a requirement for such mechanistic analysis, and research is warranted to improve current practice. A prominent shortcoming of current practice is the use of an accelerated laboratory procedure to characterize asphalt pavements’ long-term field conditions that result from oxida- tive aging. For example, the long-term aging procedure provided in AASHTO R 30, Mixture Conditioning of Hot Mix Asphalt (HMA), prescribes aging compacted mixtures for 5 days at 85°C, which is a time and temperature combination that, according to the original Strategic Highway Research Program (SHRP) research based on a limited field calibration, would reflect a critical duration of field exposure for as few as 5 or 7 years or as long as 10 years. However, this single time-temperature combination may not be applicable for all the different climates found throughout the United States. The long-term aging of compacted specimens can also lead to both radial and vertical oxidation gradients in the specimen, which can be troublesome if the aged, compacted specimens are to be used in performance testing (Houston et al. 2005, Kim et al. 2018). Moreover, the current AASHTO R 30 standard does not address the significant aging gradient that is observed in the field within the top inches of a pavement (Harrigan 2007). To address these problems, research is needed to develop a calibrated and validated procedure that can accurately simulate the aging of asphalt mixtures at different time periods and pave- ment depths for performance testing and prediction. Current pavement performance prediction models have different levels of sophistication for numerically simulating (1) the increase in pavement stiffness that is due to aging (e.g., the Global Aging System in Pavement ME Design) and (2) the competing phenomena of accumulated damage and deterioration that manifest as a reduction in the modulus (e.g., CalME). Significant research has been devoted to understanding and modeling the aging of asphalt binder. How- ever, relatively little research has been devoted to modeling the effects of changes in the asphalt binder properties that are due to aging on asphalt mixture performance. The accurate predic- tion of asphalt mixture properties as a function of pavement depth throughout the service life of a pavement is becoming more important as more robust pavement design and performance prediction methods are being implemented. The original NCHRP Project 09-54, “Long-Term Aging of Asphalt Mixtures for Performance Testing and Prediction,” developed a laboratory aging procedure that prescribed a set of labora- tory aging conditions to represent the long-term aged state of the asphalt mixture in pavement as a function of climate and depth (Kim et al. 2018). The project also yielded a pavement aging model (PAM) that predicted the evolution of asphalt binder rheology with aging. This labora- tory aging procedure and PAM were calibrated using original component materials and field C H A P T E R 1 Introduction

2 Long-Term Aging of Asphalt Mixtures for Performance Testing and Prediction: Phase III Results cores from in-service and test track pavements from eight states in the United States and the province of Manitoba in Canada. Although the original NCHRP Project 09-54 deliverables represent significant contributions to the study of asphalt mixture aging, several critical shortcomings have become evident. The mixtures that were used to calibrate and validate the laboratory aging procedure and PAM were selected from a pool of materials that included the original binders, aggregate, and field cores from selected pavement sections. The chosen sections covered a wide range of pavement design, climatic conditions, ages, binder and aggregate characteristics, air void contents, asphalt contents, and gradations. However, these sections consisted primarily of HMA mixtures. Other mixtures, such as warm-mix asphalt (WMA), mixtures containing reclaimed asphalt pave- ment (RAP), and polymer-modified asphalt (PMA) mixtures, were not thoroughly considered in the calibration of the laboratory aging durations. In addition, only a single field core was analyzed per project location due to time limits. Additional field core replicates could have greatly improved the reliability of the developed deliverables. Also, the PAM was established by calibrating a rheology-based oxidation kinetics model against the field core measurements. The calibration process ignored the effects of mixture morphology (e.g., air void and asphalt contents) on oxidative aging. The inputs to the kinetics model required testing extracted and recovered binder obtained from loose mixture aging or Universal Simple Aging Test (USAT) binder aging. To improve the feasibility and practicality of implementation, it would be advan- tageous if the inputs to the kinetics model could be approximated using standard, performance grade (PG) aging methods and data. In addition, the predicted changes in the binder properties with oxidative aging must be related to changes in the asphalt mixture properties in order to facilitate the integration of the PAM into ME pavement performance prediction frameworks. In short, improvements to the original NCHRP Project 09-54 outcomes were necessary to ensure the accurate characterization of the long-term aging of asphalt mixtures for performance testing and prediction. Figure 1 is a flow chart describing the main objectives tackled in this report, the tasks associated with each objective, and finally, the resulting outcomes/products. Objectives and Scope The objectives of this project are to: 1. Refine the laboratory aging procedure that was developed in the original NCHRP Proj- ect 09-54 project by including WMA, RAP, and PMA mixtures and field core replicates. 2. Field calibrate the PAM that was developed in the original NCHRP Project 09-54 as a func- tion of depth, including field core replicates and WMA, RAP, and PMA mixtures. 3. Develop procedures to estimate the PAM inputs using PG, standard binder aging methods, and associated data. 4. Develop a framework by which the predicted changes in asphalt binder properties that are due to oxidative aging can be related to corresponding changes in asphalt mixture performance. To achieve these objectives, first, the predefined aging durations and field calibration method that were used in the oxidation kinetics model that was established in the original NCHRP Proj- ect 09-54 were refined by incorporating WMA, RAP, and PMA mixtures along with field core replicates from projects that were included in the original NCHRP Project 09-54 effort. Then, a systematic study of the effects of asphalt morphology on pavement aging was conducted to incorporate those effects into the PAM. Alternatives to loose mixture aging to obtain inputs for the PAM were investigated; these alternatives included aging binders in both a rolling thin film oven (RTFO) and pressure aging vessel (PAV) prior to testing as well as using the PG to derive

Refinement of Climate- Based Laboratory Aging Procedure Refinement of the Pavement Aging Model (PAM) Development of Procedures to Estimate the PAM Inputs using Standard Binder Aging Methods & PG Development of a Framework to Predict Changes in Asphalt Mixture Performance Due to Oxidative Aging Consider WMA, RAP, & PMA field sections Include replicate field core data for old & new sections Simplify Climatic Aging Index (CAI) Generate new aging duration maps Consider WMA, RAP, & PMA field sections Include replicate field core data for old & new sections Include effects of asphalt mixture morphology on pavement aging Evaluate PAM against Global Aging System (GAS) model Estimate inputs for virgin binders using RTFO & PAV aging methods Estimate inputs for virgin binders using PG Estimate inputs for RAP binders using generalized PAM predictions & PG Estimate ‘blended’ inputs given virgin & RAP inputs Age mixtures & corresponding binders systematically using the refined laboratory aging procedure Test mixtures & corresponding binders to obtain mechanical properties Develop a framework to predict changes in linear viscoelastic properties due to changes in binder properties caused by aging Develop a framework to predict changes in fatigue properties due to changes in binder properties caused by aging Conduct pavement performance simulations with evolving mixture properties as a function of time and pavement depth Main Outcome Laboratory Long-Term Aging Procedure Age loose mixture in the oven at 95°C for a duration specified by a new simplified CAI or new aging duration maps Main Outcome Field Calibrated Pavement Aging Model (PAM) Predict |G*| at any field aging duration, pavement depth, and climatic condition Main Outcome Estimate inputs to PAM for virgin/RAP/blended materials using PG and standard binder aging methods Main Outcome Asphalt Mixture Aging-Cracking (AMAC) Model Translate changes in |G*| predicted using PAM to changes in mixture linear viscoelastic & fatigue propertiesN C H R P 0 9 - 5 4 P � � � � � � E � � � � � � � � R E P O R T Figure 1. Flow chart describing different components and associated outcomes of the report.

4 Long-Term Aging of Asphalt Mixtures for Performance Testing and Prediction: Phase III Results approximations. Lastly, the effects of asphalt binder aging on asphalt mixture cracking perfor- mance were investigated systematically to establish a methodology to facilitate the integration of the PAM into pavement performance prediction frameworks. Report Organization This report has six chapters. Chapter 1 introduces the work that was conducted under the NCHRP Project 09-54 extension and provides reasons for its necessity. Chapter 2 summarizes the findings and results of the original NCHRP Project 09-54. Chapter 3 presents the research that was conducted to refine the laboratory aging durations proposed in the original NCHRP Project 09-54 by utilizing modern mixtures (RAP, WMA, and PMA) and field core replicates. Chapter 4 presents the research undertaken to refine the PAM, including the recalibration of the model using RAP, WMA, and PMA mixtures, field core replicates, and mixtures with systematic changes in asphalt mixture morphology as well as the establishment of methods to estimate the material-specific PAM parameters using standard binder aging methods and PGs. Chapter 6 presents the systematic evaluation of the effects of oxidative aging on asphalt mixture perfor- mance, which was then used to develop a framework to predict the evolution of asphalt mixture performance that is due to aging based on changes in asphalt binder properties. Chapter 7 sum- marizes the conclusions and suggests future research. Three appendices present more in depth information on the research. The final attachment is the proposed standard method of test for long-term conditioning of hot mix asphalt for performance testing.