Click for next page ( 36

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 35
35 Bottled Nitrogen Activated Charcoal SAFT Supelco Hydrocarbon Trap VESSEL HaySepQ Kolby Junior 100 mm BINDER Indicating Silica Gel Supelco Moisture Trap Molecular Sieve 5A Ambient Laboratory Air Figure 3-24. Final version of VCS-III. change and the combined mass of the material collected on the balance between the oxygen consumed by oxidation, loss the HayeSep Q and the molecular sieve. The negative value of of hydrocarbon volatiles, and water resulting from oxidation the RTFOT mass change is used so that a positive value or other chemical reactions. for the absorbents and the RTFOT indicate material lost dur- ing the conditioning procedure. From Figure 3-25 it appears that the mass of hydrocarbon volatiles collected on the 3.5 SAFT Optimization Study HayeSep Q filter is similar for all of the binders tested rang- ing from 0.00 to 0.07 percent of the original mass of binder. The objective of the SAFT optimization study was to This range is similar to the range reported for the SAFT pro- enhance the efficiency of the operating parameters for the totype VCS. The mass of water collected on the molecular commercial SAFT so that it more nearly reproduced the sieve is much lower for the five polymer-modified binders degree of aging that occurs in the RTFOT for neat binders. compared to the neat binders and the air blown binder. During the VCS study, it was observed that the commercial There is no apparent correlation between the mass lost version of the SAFT, when operated using the parameters rec- during RTFOT conditioning and the mass collected on the ommended for the prototype, produced significantly less individual or combined absorbents. This is illustrated graph- aging than the RTFOT. Recall, the developers of the SAFT ically in Figure 3-26 where the mass collected on the absorbents documented good agreement between the prototype SAFT is plotted versus the mass lost during the RTFOT condition- and the RTFOT. The likely cause of the lower aging in the ing. This is not surprising given that the RTFOT represents commercial SAFT is the improved temperature control pro- vided by this device. In the prototype, temperature control was provided by a heating mantle that was in direct contact Table 3-13. Mass collected on VCS-III filters and with the SAFT vessel. In the commercial SAFT, the vessel is RTFOT mass change. heated in an oven. The process controller limits the maximum temperature in the oven to 176C. The heating mantle, on the Mass Collected on Filters, % of Original Mass other hand, can reach very high temperatures, and because it Negative of Binder Combined RTFOT Mass is in direct contact with the steel SAFT vessel, the binder at the HayeSep Q HayeSep Q Molecular and Loss, % of wall of the vessel can also reach temperatures well above the test Original Mass Sieve Molecular temperature of 163C. More rapid aging of the binder occurs Sieve in areas of the SAFT vessel with the highest temperature. AAC-1 0.072 0.141 0.214 0.058 AAD-2 0.041 0.134 0.175 1.058 The SAFT optimization study included two experiments. AAF-1 0.043 0.114 0.157 0.008 The first was an experiment to verify that the heat-up phase ABM-2 0.024 0.113 0.137 0.349 of the test does not result in significant aging of the binder. ABL-1 0.049 0.129 0.178 0.654 AAM-1 0.014 0.085 0.100 -0.122 During the heat-up phase, the temperature of the binder is Citgoflex 0.020 0.052 0.072 0.196 increased from approximately 100C to the testing tempera- ALF 64-40 0.046 0.007 0.053 0.207 ture of 163C while nitrogen flows through the SAFT vessel. Airblown 0.020 0.061 0.081 -0.031 Elvaloy 0.000 0.003 0.003 0.173 Since the starting temperature (temperature of the binder EVA 0.020 0.005 0.026 0.132 after charging the SAFT vessel) cannot be accurately controlled, Novophalt 0.025 0.006 0.031 0.132 it is critical that the heat-up phase not contribute significantly

OCR for page 35
36 Hayesep Q Molecular Sieve 0.16 0.14 Percent of Original Mass, % 0.12 0.10 0.08 0.06 0.04 0.02 0.00 Figure 3-25. Mass collected on filters. to the degree of aging that occurs in the test. The second 3.5.1 Heat-Up Effects Experiment experiment was an experiment to determine the effects of impeller speed, airflow rate, and test duration on the degree of During the heat-up portion of the SAFT test, the tempera- aging measured by the rheology of the binder at high in-service ture of the binder is increased from approximately 100C to pavement temperatures. Operating parameters for the com- the testing temperature of 163C while nitrogen flows through mercial SAFT were selected from this experiment. the SAFT vessel. Since the starting temperature (temperature The SAFT optimization study is described in detail in Appen- of the binder after charging the SAFT vessel) cannot be accu- dix D (see the project webpage on the TRB website). Its key rately controlled, it is critical that the heat-up phase not con- findings are discussed in the following sections. tribute significantly to the degree of aging that occurs in the Hayesep Q 0.25 Mass Collected on Absorbents, % of Original Binder Molecular Sieve 0.20 Combined Hayesep Q & Molecular Sieve 0.15 Mass 0.10 0.05 0.00 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 Negative of the RTFOT Mass Change, % Figure 3-26. Material collected on molecular sieve and HayeSep Q versus RTFOT mass change.

OCR for page 35
37 Table 3-14. High-temperature DSR data from heat-up effects experiment. Tank After Heat Up Average Average Temp, G*, , G*/sin, G*/sin, G*, , G*/sin, G*/sin, Binder C Rep kPa deg kPa kPa kPa deg kPa kPa PG 58-28 58 1 1.18 87.1 1.18 1.24 1.20 87.0 1.20 1.25 2 1.30 86.9 1.30 1.29 87.0 1.29 PG 76-22 76 1 1.45 83.6 1.46 1.48 1.44 84.1 1.45 1.39 2 1.50 83.3 1.51 1.33 83.9 1.34 PG 82-22 82 1 1.54 66.3 1.68 1.71 1.54 65.9 1.69 1.71 2 1.59 66.1 1.74 1.58 65.6 1.73 test. To assess the change in the properties of the asphalt binders, listed in Table 3-15, were chosen for this experiment. binder during the heat-up period DSR, measurements at The binders, which have been studied extensively elsewhere, 10 rad/s were made at the high pavement temperature for represent a range in chemical composition, mass change, and three binders (PG 58-28, PG 76-22, and PG 82-22). DSR sensitivity to short-term aging. No modified binders were measurements were made for each binder in the tank condi- included because the purpose of the experiment was to deter- tion and on material removed from the SAFT after comple- mine the SAFT operating parameters that best mimic the tion of the heat-up phase. Initial testing showed that the aging that occurs in the RTFOT for neat binders. SAFT trapped air bubbles in the PG 82-22 binder; therefore, The experiment design for the operating parameters experi- for this experiment and the operating parameters experiment ment is presented in Table 3-16. It employed a Plackett-Burman discussed in the next section, both the tank and the materials design to simultaneously assess the effects of changes in impeller sampled at the end of the heat up-up period were exposed to speed, airflow rate, and conditioning time on the aging that the PAV vacuum degassing procedure to remove the trapped occurs in the commercial SAFT. Plackett-Burman designs are air before they were tested in the DSR. often used in ruggedness testing to assess the effect of changes The DSR data from this experiment are summarized in in multiple test parameters. These are extremely efficient Table 3-14. For a given binder, all of the DSR data are within the designs that allow the main effects to be determined with a AASHTO T315 single-operator precision of 9.5 percent for test- limited amount of testing. For the three variables included in ing of original binder. Since the complex modulus before this experiment only four test results are needed to assess the and after the heat-up phase varies by less than the single- main effects. ASTM E1169-02 (32) presents detailed informa- operator precision, this experiment found that the binder does tion on the design and analysis of the type of experiment used. not significantly stiffen during the SAFT heat-up phase. The rheological data from the operating parameters exper- iment are summarized in Table 3-17. Two statistical analyses were performed on the DSR data shown in Table 3-17. The 3.5.2 Operating Parameters Experiment first analysis was performed to determine if the PAV degassing The purpose of this experiment was to determine the effect procedure affects the DSR data obtained after short-term of three operating parameters--impeller speed, airflow rate, aging of neat binders in the SAFT. Recall that previous work and conditioning time--on the properties of asphalt binder with modified binders during the heat-up effects experiment aged in the commercial SAFT. Three different unmodified showed that degassing was a necessary step for modified Table 3-15. Binders for the optimization study. RTFOT Aging Index, PG Binder/Source Comments Mass 135C Grade Change, % Viscosity California Coastal SHRP core binder (AAD-2) except AAD-1 used 52-28 -1.058 2.86 during SHRP West Texas SHRP core asphalt Intermediate 64-16 +0.122 1.98 (AAM-1) California Valley Replacement for SHRP (ABM-2) core binder AAG-2, 58-16 -0.348 1.62 except AAG-1 used during SHRP

OCR for page 35
38 Table 3-16. Operating parameters was immediately cooled at ambient temperature, a second experiment design. sample was exposed to 170C for 40 minutes, and the third sample was exposed to the PAV degassing procedure. A paired Impeller Air Aging Binder Testing difference analysis was used to assess the effect of degassing. Speed Flow Time Source Plan (rpm) (L/Min) (Min) Details of the analysis are presented in Appendix D (see the 2 45 Yes 700 60 -- -- project webpage on the TRB website). The following three 4 45 -- -- differences were considered: 60 Yes AAD-2 2 45 -- -- 1. Difference between degassed and air-cooled binder, 1400 60 Yes 4 45 Yes 2. Difference between air-cooled binder and binder heated 60 -- -- per the degassing procedure but without the application 45 Yes 2 60 -- -- of vacuum (oven treatment), and 700 3. Difference between degassed binder (degassed) and 45 -- -- 4 60 Yes binder heated per the degassing procedure but without the AAM-1 45 -- -- 2 60 Yes application of vacuum (oven treatment). 1400 45 Yes 4 60 -- -- The results of this analysis are summarized in Table 3-18 2 45 Yes 60 -- -- and include the mean difference (d ), its standard deviation 700 4 45 -- -- (Sd), the calculated t statistic (T), and its percentage value (p). 60 Yes ABM-2 45 -- -- Low p values (bold and underlined in Table 3-18) indicate 2 1400 60 Yes significant differences. This analysis found that the degassing 45 Yes procedure significantly affects the rheology of neat SAFT- 4 60 -- -- aged binders, and that it is probably the additional exposure to high temperature that causes this additional aging. Since binders aged in the SAFT. For simplicity, the study team used degassing is needed to remove entrapped air from stiff mod- the vacuum degassing procedure currently specified in ified binders, this experiment shows that it must be included AASHTO R28-06, Standard Practice for Accelerated Aging of as part of the procedure for all binders, and that the degassing Asphalt Binder Using a Pressurized Aging Vessel (PAV). This must be standardized. procedure exposes the binder to 170C for 402 minutes. For The second analysis that was performed investigated the the last 301 minutes, the binder is exposed to vacuum with effects of impeller speed, airflow rate, and conditioning time a residual pressure of 152.5 kPa absolute. The effect of the on the degree of aging that occurs in the SAFT to determine degassing process was evaluated by performing DSR tests on the sensitivity of the SAFT aging to these parameters. Tem- three split samples of the SAFT-aged material. One sample perature was not included in the experiment because the Table 3-17. DSR data from the operating parameters experiment. DSR at 58C Operating Degassed Oven Treatment Air Cooled RTFOT, AVG Binder Order Parameters Speed Flow Time G*, , G*/sin, G*, , G*/sin, G*, , G*/sin, G*, , G*/sin, rpm L/Min Min kPa deg kPa kPa Deg kPa kPa Deg kPa kPa deg kPa AAD-2 1 700 4 60 1.72 81.5 1.74 1.69 81.6 1.71 1.68 81.5 1.70 AAD-2 2 1,400 4 45 4.16 74.6 4.31 4.20 74.6 4.36 4.03 75.0 4.17 AAD-2 3 1,400 2 60 4.73 73.2 4.94 4.71 73.4 4.91 4.37 73.6 4.56 2.60 79.0 2.65 AAD-2 4 700 2 45 1.30 83.4 1.31 1.36 83 1.37 1.32 83.0 1.33 AAM-1 1 700 4 60 5.43 82.4 5.48 5.31 82.3 5.36 5.04 82.5 5.08 AAM-1 2 1,400 2 60 8.99 78.7 9.17 9.22 78.6 9.41 8.52 78.7 8.69 AAM-1 3 1,400 4 45 7.37 79.9 7.49 7.57 79.9 7.69 7.03 80.0 7.14 5.94 81.9 6.00 AAM-1 4 700 2 45 4.13 83.8 4.15 4.37 83.5 4.40 3.77 83.2 3.80 ABM-2 1 1,400 4 45 4.39 88.8 4.39 4.27 88.8 4.27 4.44 88.8 4.44 ABM-2 2 1,400 2 60 5.24 88.5 5.24 5.04 88.5 5.04 5.39 88.5 5.39 ABM-2 3 700 2 45 2.45 89.4 2.45 2.50 89.5 2.50 2.54 89.5 2.54 3.14 89.5 3.14 ABM-2 4 700 4 60 2.85 89.4 2.85 2.83 89.3 2.83 2.97 89.3 2.97

OCR for page 35
39 Table 3-18. Summary of degassing effects. 7 6 Paired Differences G*/sin, kPa 5 AAD-2 Parameter Degassed--Air Oven Treatment-- Degassed--Oven Cooled Air Cooled Treatment 4 AAM-1 d 0.14 0.17 -0.03 3 Sd 0.23 0.34 0.14 ABM-2 2 T 2.11 1.75 -0.64 p 0.03 0.05 0.74 1 0 0 2 4 6 Air Flow, L/min review of the literature and research in progress indicated a strong desire to perform the short-term aging test at 163C, Figure 3-28. Effect of airflow on G*/sin for which reasonably simulates hot mix plant operating temper- SAFT-aged material. atures. Figures 3-27 through 3-29 show the effect of the three operational parameters on G*/sin measured after SAFT SAFT. The magnitude of these changes was estimated to be a aging and degassing. 100-rpm increase in impeller speed or a 5-minute increase From these figures it is clear that the aging in the SAFT is in conditioning time to increase G*/sin by 7 percent. After affected by impeller speed and conditioning time, but not by careful consideration, the longer conditioning time was airflow rate over the ranges studied. Using the slopes in these selected because of concern that further increases in impeller figures, the RTFOT G*/sin values measured for three binders, speed could result in significant quantities of binder splash- and assuming linear relationships, the impeller speed required to reproduce RTFOT aging at the average conditioning time of ing onto the lid of the SAFT with the potential for asphalt 52.5 minutes and the conditioning time required to reproduce droplets to exit into the VCS. Visual observation from com- RTFOT aging at the average impeller speed of 1,050 rpm can pleted tests indicates that at 1,000 rpm no asphalt splashes on be estimated. These are summarized for the three binders in the lid, but at 1,400 rpm a significant amount of asphalt splashes Table 3-19. onto the lid. Based on Table 3-19, it appeared that using an impeller Table 3-21 presents rheological data for the Citgo PG 58- speed of 1,000 rpm, an airflow rate of 2,000 mL/min, and a 28 binder where the conditioning time was increased from 45 conditioning time of 45 minutes would result in aging in the to 50 minutes. As shown, this change increased G*/sin by commercial SAFT that is approximately equivalent to that approximately 6.5 percent. With the new operating param- which occurs in the RTFOT. eters, the SAFT ages the Citgo PG 58-28 binder slightly more The analysis presented in this section assumes linear effects. than the RTFOT, but based on Table 3-20, this increased aging Developmental work on the VCS-II provided an opportunity should provide better agreement when a wide range of binders to assess the tentative operating parameters. Independent is considered. rheological data (average of duplicate runs) collected during The final operating parameters selected from the operating the VCS-II testing are presented in Table 3-20. These data parameters experiment for the commercial SAFT for use in show that the operating conditions listed above for the SAFT the verification study were underestimates the aging that occurs with the RTFOT in three of the four cases. Either the conditioning time or the impeller 163C aging temperature, speed can be increased to increase the degree of aging in the 2,000 mL/min airflow, 8 9 7 8 6 G*/sin, kPA 7 G*/sin, kPa 5 AAD-2 6 AAD-2 5 4 AAM-1 AAM-1 4 3 ABM-2 3 ABM-2 2 2 1 1 0 0 500 1,000 1,500 40 50 60 70 Impeller Speed, rpm Duration, min Figure 3-27. Effect of impeller speed on Figure 3-29. Effect of conditioning time on G*/sin for SAFT-aged material. G*/sin for SAFT-aged material.