Field Verification of Proposed Changes to the AASHTO R 30 Procedures for Laboratory Conditioning of Asphalt Mixtures (2019)

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.

3 Introduction Background Asphalt pavements are subject to rheological changes in the form of stiffening over time in response to oxidation. As asphalt mixtures are produced in plants, they are subjected to heat and air during mixing and storage. They are further aged as they are transported and placed at the pavement con- struction site. After this initial short-term aging at elevated temperatures, oxidation continues at a slower rate over the long term in the field. The stiffening and embrittlement resulting from oxidation increases more rapidly in warm climates than in cold climates. These rheological changes of the mixture can lead to cracking in various ways that may be related to climate or traffic factors or combinations of these effects. NCHRP Project 09-52, “Short-Term Laboratory Conditioning of Asphalt Mixtures” (findings published in NCHRP Report 815) investigated the following factors that could affect the degree or rate of aging of asphalt mixtures (Newcomb et al. 2015): • WMA technologies • Plant mixing temperatures • Aggregate absorption of asphalt binder • Mixing plant type • Climate • RAP and/or RAS content • Asphalt source WMA technologies (either additives or water foaming systems) serve as compaction aids that allow producers to reduce mixing plant and placement temperatures to minimize emissions and aging. Plant mixing temperatures intuitively should be tied to the degree of oxidation occurring in asphalt mixtures because oxidation is enhanced at elevated tempera- tures. Aggregate absorption can affect the aging of asphalt mixtures by selectively absorbing the lighter ends of the binder, leaving the heavy, more brittle fractions in the mortar of mixtures. The method of manufacturing the asphalt mix [batch mix plant (BMP) or drum mix plant (DMP)] has been reported to result in a difference of stiffness in mixtures (Lund and Wilson 1984). Climate has long been associated with different rates of aging for warmer and colder temperatures. The inclusion of RAP and RAS has been shown to affect the stiffness of mixtures immediately after construction. Finally, different asphalt sources are known to have different aging behaviors—even if they are categorized as having the same grade—as a result of the differences in oxidation of different combinations of molecules and the subsequent rheological response to chemical changes. In conclusion, mixture components, mixture processing, and plant design may result in changes in mixture perfor- mance. It is important that the aging protocols used in con- ditioning laboratory samples realistically reflect the aging that takes place during the production of the mixtures in a plant. Likewise, it is important to know how mixtures age in the field to understand how much conditioning to apply to simulate long-term aging. The following NCHRP projects have addressed these issues: • NCHRP Project 09-43, “Mix Design Practices for Warm Mix Asphalt” published in NCHRP Report 691: Mix Design Practices for Warm Mix Asphalt • NCHRP Project 09-47A, “Properties and Performance of Warm Mix Asphalt Technologies” published in NCHRP Report 779: Field Performance of Warm Mix Asphalt Technologies • NCHRP Project 09-48, “Field versus Laboratory Volu- metrics and Mechanical Properties” published in NCHRP Report 818: Comparing the Volumetric and Mechanical Prop- erties of Laboratory and Field Specimens of Asphalt Concrete • NCHRP Project 09-49A, “Performance of WMA Technol- ogies: Stage II: Long-Term Field Performance” published in NCHRP Research Report 843: Long-Term Field Perfor- mance of Warm Mix Asphalt Technologies C H A P T E R 1

4 Previous Research— NCHRP Project 09-52 For NCHRP Project 09-52, the focus was to (a) develop a laboratory short-term aging protocol to simulate the aging and asphalt absorption of an asphalt mixture during pro- duction and transportation, and (b) develop a laboratory long-term aging protocol to simulate the aging of the asphalt mixtures after construction. The project was divided into two phases: Phase I evaluated short-term aging protocols for HMA and WMA to simulate the plant aging, and Phase II evaluated long-term aging protocols to quantify field aging after construction. Nine field sites in eight states were constructed to provide the materials necessary to complete this research. Each field site had one or more of the factors listed previously as its primary study component. During the construction of the field sites, mineral aggregates, recycled materials, and binders were sampled to fabricate LMLC specimens to replicate con- ditions at mixture design. Mixtures produced by the asphalt plant were also sampled and compacted on or near the job site to provide PMPC specimens, and roadway field cores were obtained immediately after construction and, if possible, at intervals up to 2 years after construction. To prepare LMLC specimens in this project, laboratory STOA protocols were adopted from NCHRP Project 09-49, “Performance of WMA Technologies: Stage I—Moisture Sus- ceptibility” (published in NCHRP Report 763) (Epps Martin et al. 2014). These STOA protocols condition loose mixtures for 2 hours at 275°F (135°C) for HMA and for 2 hours at 240°F (116°C) for WMA. Two common LTOA protocols for compacted LMLC specimens were evaluated including (a) 5 days at 185°F (85°C) per AASHTO R 30; and (b) 2 weeks at 140°F (60°C). Results were included in NCHRP Report 763 (Epps Martin et al. 2014). On a limited basis, an additional LTOA protocol of 3 days at 185°F (85°C) was also evaluated. In NCHRP Project 09-52, both the resilient modulus (MR) stiffness test per ASTM D7369 at 77°F (25°C) and the Hamburg wheel tracking test per AASHTO T 324 at 122°F (50°C) were used to evaluate the effects of both short- term and long-term aging on asphalt mixtures. Addition- ally, a limited amount of dynamic modulus (E *) testing per AASHTO TP 79-13 was conducted. The M R stiffness test was effective for evaluating the effects of aging because (a) the binder properties govern the stiffness, and (b) the test can be used directly to compare the stiffness of cores and LMLC and PMPC specimens. The presence of RAP and RAS in asphalt mixtures was found to increase the stiffness of the materials through the plant significantly. A lower initial stiffness of asphalt mixtures appeared when higher aggregate water absorption values were added, because they required more effective asphalt content as dictated by volumetric mixture design require- ments. The presence of WMA technology resulted in lower stiffness in mixtures. Asphalt source showed that the differ- ences between asphalt sources could produce substantial differences in behavior. The factor of plant type showed no significant effect on the aging characteristics, and the appli- cation of a difference in production temperatures of 30°F (17°C) resulted in no difference in stiffness. To capture aging in the field, CDD was introduced as a field metric. The metric allowed the analysis to account for both in-service temperature and time of construction. As shown in Equation (1-1), the CDD is the sum of the daily high temperature above freezing [32°F (0°C) base] for all the days from the time of construction to the time of core sampling. Figure 1 presents MR stiffness ratios (ratio of aged MR stiffness to the initial MR stiffness at construction) versus Figure 1. MR stiffness ratio versus CDD.

5 CDD values for post-construction cores tested and presented in NCHRP Report 815 (Newcomb et al. 2015); the data points represent the average property ratio values for each field site, and the adjusted line represents the predicted line for MR stiffness ratio using Equation (1-2). As shown, the MR stiffness ratio has a significant ascending trend as CDD values increase. 32 (1-1)∑( )= −CDD Tdmax = + α ∗ − β        γ M Stiffness Ratio exp CDD R 1 (1-2) Where: CDD = cumulative degree-days for cores after specific in-service times. α, β, and γ = fitting coefficients. Figure 2 illustrates the MR stiffness ratio correlation between field aging and laboratory LTOA protocols from NCHRP Project 09-52. The average MR stiffness ratio values for LMLC specimens after STOA [2 hours at 275°F (135°C)] and after STOA and LTOA [(STOA + 2 weeks at 140°F (60°C) or 5 days at 185°F (85°C)] were plotted as markers by cross- ing the curve of the MR stiffness ratio versus CDD values. Laboratory LTOA protocols of 2 weeks at 140°F (60°C) and 5 days at 185°F (85°C) were able to produce equivalent aging as in the field with 9,100 and 16,000 CDDs, respectively. The bars on the calculated MR stiffness ratios represent the range of results. Using this CDD concept, there were three different sce- narios that were considered possible for the long-term aging 1.0 1.5 2.0 2.5 3.0 0 10000 20000 30000 40000 M R S ti ff ne ss R at io Cumulative Degree Days (°F-days) 2h@135C + 2w@60C/3d@85C 2h@135C + 5d@85C Figure 2. MR stiffness ratio correlation between field aging and laboratory LTOA protocols (based on Figure 1). of HMA versus WMA mixtures. Scenario I presented in Figure 3(a) shows that the MR stiffness of the HMA cores was always higher than that of the WMA cores, and mixtures aged along parallel paths. Scenario II, presented in Figure 3(b), shows that the HMA cores had higher initial MR stiffness than the corresponding WMA cores, and that ultimately, the rates of aging caused the stiffness values to intersect. Finally, Scenario III, presented in Figure 3(c), shows that there is equivalent mixture stiffness for both HMA and WMA at con- struction, but WMA mixtures aged faster post-construction compared with HMA mixtures. The findings and applications from NCHRP Project 09-52A are presented in Chapter 3. Literature Review The aging of asphalt binders has long been a concern in the flexible pavement community. Several studies from the 1960s evaluated the short-term aging of asphalt mixtures in the field. High temperatures that occurred during pro- duction and construction were considered the reason for the short-term aging (Heithaus and Johnson 1958; Traxler 1961; Chipperfield and Welch 1967). More recent studies have shown that several factors could significantly affect the performance and the short-term aging of asphalt mixtures. These factors include binder source, binder type, aggregate absorption, plant type, production temperature, and silo storage time (Traxler 1961; Chipperfield and Welch 1967; Lund and Wilson 1984, 1986; Topal and Sengoz 2008; Zhao et al. 2009; Morian et al. 2011; Rashwan and Williams 2011; Mogawer et al. 2012; Daniel et al. 2014). Table 1 briefly summarizes these studies on the short-term aging of asphalt mixtures.

6 After short-term aging during production and construction, asphalt pavements age at a continually slower rate throughout their in-service life. The changes resulting from field aging in asphalt mixture properties should be accounted for with performance testing after LTOA. The standard practice for laboratory mixture design of asphalt mixtures is to simulate long-term field aging by conditioning compacted specimens for 5 days at 185°F (85°C), in accordance with AASHTO R 30. As presented in NCHRP Report 815, this protocol simulated asphalt field aging by about 2 years (Newcomb et al. 2015). A recent study by Rondon et al. (2012) evaluated the evo- lution of asphalt mixture properties under environmental exposure. A typical HMA was exposed to the environment for 42 months in Bogota, Colombia, prior to being tested in the laboratory to determine changes in mechanical proper- ties. Test results showed that there was an increase in mixture stiffness, rutting resistance, and fatigue resistance for the first 29 months of environmental exposure and that this could be attributed to the aging of the asphalt binder resulting from temperature and ultra-violet radiation. However, decreased stiffness was indicated between 30 and 42 months. The University of New Hampshire performed a study to evaluate the effects of laboratory LTOA on RAP mixtures (Tarbox and Sias Daniel 2012). Plant-produced mixtures with 0%, 20%, 30%, and 40% RAP were compacted into specimens in the laboratory and were then long-term aged for 2, 4, and 8 days at 185°F (85°C) prior to being tested for E*. In this study, the Global Aging System was also adopted to predict the changes in E* values at various long-term aging levels (Mirza and Witczak 1995). These predicted values were then compared with the measured values. Test results indicated that the laboratory LTOA was able to produce mixtures with a significant increase in mixture stiffness compared with the short-term aged mixtures, and that the stiffening effect was more pronounced for virgin mixtures that had 0% RAP compared with the RAP mixtures. The inclusion of already aged binders in the RAP caused reduced susceptibility to laboratory aging. The comparison between predicted and measured E* values illustrated that the Global Aging System model predicted higher stiffness than the actual measured stiffness from laboratory aged samples. A laboratory aging study by Safaei et al. (2014) was per- formed to evaluate the effect of long-term aging on HMA and WMA mixture stiffness and fatigue resistance. The protocol for laboratory specimen fabrication was 4 hours at 275°F (135°C) for HMA and 2 hours at 243°F (117°C) for WMA and LTOA of 2 days and 8 days at 185°F (85°C). These speci- mens were tested for E* and direct tension tests to quantify mixture performance-related property evolution for aging at different aging stages. Laboratory LTOA was able to produce both HMA and WMA mixtures with a significant increase in mixture stiffness over the short-term aged mixtures. A reduc- tion of the difference in mixture stiffness between HMA and WMA was observed after laboratory LTOA. Bell et al. (1994) focused on the correlation between mix- ture aging in the field compared with laboratory aging in terms of STOA and LTOA protocols. The field cores with a wide range of in-service times were collected and tested to determine their triaxial and diametral MR stiffness. Labora- tory specimens were fabricated using materials from the field sites and conditioned with the selected laboratory STOA and LTOA protocols before being tested. Test results showed that Scenario II Cumulave Degree Days Cumulave Degree Days Scenario III (a) (b) (c) Cumulave Degree Days Scenario I Figure 3. Mixture MR stiffness evolution with field aging for HMA versus WMA: (a) Scenario I, (b) Scenario II, (c) Scenario III.

7 Reference Short-Term Aging Major Finding Heithaus and Johnson 1958 Field Aging Most aging during production and construction through compactionTraxler 1961 Chipperfield and Welch 1967 West et al. 2014 • WMA less aging than HMA during production • Reduced difference between WMA and HMA with field aging • Equivalent binder true grade and binder absorption for WMA versus HMA after 2 years of field aging Traxler 1961 Factor on Field Aging Binder chemistry and aggregate absorption major effects Chipperfield and Welch 1967 Aggregate gradation no effect Lund and Wilson 1984 and 1986 Binder type and binder source significant effects Mogawer et al. 2012 • Production temperature, silo storage, inclusion of recycled materials, and reheating significant effects • Softer binder with RAP = harder binder without RAP Daniel et al. 2014 • Production temperature and silo storage significant effects • Reduced difference in virgin versus RAP mixtures after reheating Topal and Sengoz 2008 • Binder type and binder source significant effects • Reduced aging with polymers Zhao et al. 2009 Morian et al. 2011 • Binder type and binder source significant effects • Reduced aging with polymers • Aggregate absorption and gradation important effects Rondon et al. 2012 • Increased mixture stiffness, rutting resistance, and fatigue resistance for first 29 months due to environmental exposure • Opposite trend indicated between 30 and 42 months Epps Martin et al. 2016 • Long-term field performance of WMA • Established criteria to evaluate moisture susceptibility for both STOA and LTOA Zhang et al. 2017 • Long-term aging of WMA and HMA mixtures at 23 field sites • Climate effect was indicated but not evaluated in this project • Foaming WMA was found to be the binder that aged the slowest, while both HMA and Sasobit mixtures had an increasing trend of aging Bell et al. 1994 Field versus Lab Aging (4 days at 100°C; 8 days at 85°C) • STOA of 4 h at 275°F (135°C) = field aging during the construction process • Effect on mixture aging: LTOA temperature > LTOA time • STOA plus LTOA of 4 days at 100°C and 8 days at 85°C = 9 years of field aging in Washington State Tarbox and Sias Daniel 2012 Lab Aging (2 days at 85°C; 4 days at 85°C; 8 days at 85°C) • Increased stiffness with LTOA • Stiffening effect from LTOA: virgin mixture > RAP mixture • Global Aging System model > LTOA Epps Martin et al. 2014 2 and 4 h at Tc; 2 and 4 h at 275°F (135°C); 2 h at Tc + 16 h at 140°F (60°C) + 2 h at Tc • Effect on aging: STOA temperature > STOA time • Recommend 2 h @ 275°F (135°C) for HMA and 2 h @ 240°F (116°C) for WMA Safaei et al. 2014 Lab Aging (2 days at 85°C; 8 days at 85°C) • Increased stiffness with LTOA • Reduced difference in stiffness for HMA versus WMA with LTOA Tc: compaction temperature. Table 1. List of previous research on STOA.

8 a laboratory STOA protocol of 4 hours at 275°F (135°C) was representative of the short-term aging that usually occurs at production and construction. To achieve a more significant hardening effect of asphalt mixtures, LTOA at 185°F (85°C) and LTOA at a higher temperature of 212°F (100°C) for a shorter time were performed. According to the study, LTOA of 4 days at 212°F (100°C) or 8 days at 185°F (85°C) was representative of about 9 years of field aging for the climate conditions in the state of Washington. NCHRP Report 763 evaluated the performance of HMA and WMA with field and laboratory aging (Epps Martin et al. 2014). It provided preliminary results to understand the correlation between intermediate- or long-term field aging and laboratory LTOA protocols, in addition to their effects on mixture properties. For this project, the laboratory STOA protocol was to condition loose mixtures for 2 hours at 275°F (135°C) for HMA or 2 hours at 240°F (116°C) for WMA; the laboratory LTOA protocol was to condition compacted specimens for 1 to 16 weeks at 140°F (60°C). These protocols were used to fabricate laboratory long-term aged HMA and WMA specimens from two field sites in Iowa and Texas based on available literature. To represent mixtures experiencing intermediate- to long-term aging in the field, cores were taken at construction and at one or two additional times after construction. These long-term aged field cores and labora- tory specimens were tested for MR stiffness to evaluate the evolution of mixture stiffness with field and laboratory aging and to reveal the potential correlation between field aging and laboratory LTOA protocols. An increasing trend of mix- ture stiffness was observed for intermediate- or long-term aging, both in the field and in the laboratory. A severe aging of mixtures was detected in the summer compared with the winter, which was likely due to the higher in-service tempera- tures in the summer. The correlation between field aging and laboratory LTOA protocols was evaluated based on the MR stiffness. The laboratory STOA protocol of 2 hours at 275°F (135°C) for HMA or 2 hours at 240°F (116°C) for WMA on loose mixtures and LTOA protocols of 4 to 8 weeks at 140°F (60°C) on compacted specimens were found to be represen- tative of the first summer of field aging. Performance properties of WMA binders and mixtures were studied in NCHRP Project 09-47A and results published in NCHRP Report 779 (West et al. 2014). Field cores were sampled at six existing field sites and eight new field sites. Each field site had at least one HMA control section and one WMA section. Six of the field sites included multiple WMA technologies (26 WMA and 14 HMA mixtures). Loose mix was collected and samples for volumetric analyses and per- formance tests were compacted on site, without reheating to eliminate further aging during production. With the NCHRP Project 09-43 protocols for WMA, mixture verifications were performed; blends were produced to match the field-produced gradation determined from extraction tests rather than the reported job-mix formula. Samples were tested for moisture susceptibility, flow number, E*, and fatigue cracking. For the recovered binders taken at construction and short-term aged specimens made in the laboratory, the testing results showed that WMA mixtures had slightly less aging than HMA mix- tures, as indicated by lower binder grade and lower mixture stiffness. After 2 years, the testing results of the recovered binders showed properties that were not substantially dif- ferent between WMA and HMA binders. From construc- tion to 2 years later, there was very limited aging detected for recovered binders. Initially, the binder absorption was slightly lower for WMA, but after 2 years, there was not much differ- ence in absorption between the HMA and the WMA binders. In NCHRP Project 09-49B, the focus was on investigating the long-term field performance of WMA technologies and evaluating long-term aging of asphalt mixtures for perfor- mance testing and prediction. Results were published in NCHRP Report 817: Validation of Guidelines for Evaluating the Moisture Susceptibility of WMA Technologies (Epps Martin et al. 2016). A total of 64 WMA mixtures from 44 field sites with moisture susceptibility were available. NCHRP Project 09-49 established criteria for moisture susceptibility for both STOA and LTOA to discriminate between the results of WMA mixtures with good versus poor field and laboratory performance, and this result was verified by 64 additional WMA mixtures. Zhang et al. (2017) focused on the performance of WMA and HMA mixtures after long-term aging at 23 field sites. The climate had a significant impact on aging when a dry or freeze climate was involved. Foaming WMA was found to be the mixture that aged slowest. A continuously increasing trend of stiffness with aging was detected from HMA and Sasobit mixtures in the field. Project Objectives and Scope The objectives of NCHRP Project 09-52A were to provide a longer-term evaluation of field aging of asphalt mixtures from NCHRP Project 09-52, refine the long-term aging model that was developed using CDDs, and provide additional data on how the mixture and production factors studied affect the rate of long-term aging of field mixtures. In this project, field cores were taken from the same field sites used in NCHRP Project 09-52. The National Center for Asphalt Technology tested the materials from the Florida and Indiana field sites, and the Texas A&M Transportation Institute (TTI) tested materials from the remaining field sites. A reevaluation of the aging model proposed in NCHRP Proj- ect 09-52 and further discernment of aging behavior caused by the factors considered in that project were completed during the data analysis for NCHRP Project 09-52A.