Click for next page ( 16


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 15
15 CHAPTER 3 Findings 3.1 Introduction procedures based on microwave technology (913). Early work used a standard household microwave oven to age a 10-g sam- This chapter presents key findings from the five major stud- ple of binder. Based on the performance grading system high ies conducted in NCHRP Project 9-36 that were described in temperature parameter, G*/sin , 33 minutes of microwave Chapter 2: (1) identify candidate methods, (2) selection study, treatment without pressure was found to be equivalent to Thin (3) VCS study, (4) SAFT optimization study, and (5) verifica- Film Oven Test (TFOT) aging, but 63 minutes was required to tion study. Each of these studies is documented in detail in an reproduce RTFOT aging. A total of 158 minutes of microwave appendix to this report (see TRB website). Conclusions and treatment without pressure was found to be approximately proposals based on these findings are presented in Chapter 4. equivalent to the combination of TFOT plus PAV aging based on the performance grading system intermediate tempera- 3.2 Identify Candidate Methods ture parameter G*sin. In later work, a MARS5 scientific microwave produced by CEM Corporation with temperature 3.2.1 Post-SHRP Approaches Used for and pressure control was used for the long-term version of Laboratory Aging of Asphalt Binders the test. In the latest protocol for long-term aging, 66 g of binder The review of the literature and research in progress iden- are aged at 135C in a scientific microwave unit under 3,200 kPa tified three approaches for laboratory aging of binders that (460 psi) of air pressure for 190 minutes. Comparison of rheo- have been pursued by various researchers in the United States logical properties between long-term microwave aging and and abroad since the completion of the SHRP asphalt research RTFOT plus PAV aging has shown good agreement. program. The three approaches are For Project 9-36, the principal concern with basing improved binder aging methods on microwave technology is a lack of Microwave technology, which increases the energy level of understanding of the mechanism whereby microwave energy the binder molecules and the bulk temperature of the binder, accelerates the aging process. Bishara and McReynolds (9) thereby accelerating the oxidation process; describe how microwave energy is dissipated as heat in an Various techniques for producing thin films of binder and asphalt sample, but its effect on volatile loss and oxidation reac- thereby increasing the availability of oxygen and the oppor- tions has not been studied. For long-term aging, high pressures tunity for volatilization; and are needed to accelerate the aging process, which is one of the Air blowing, which increases the availability of oxygen and major criticisms of the PAV. Approximate cost of the CEM the opportunity for volatilization. microwave unit is $20,000. Throughput and the inability to measure volatile loss also are significant concerns for the This section discusses the viability of using each of these microwave approach. Because the microwave aging mechanism approaches in an improved aging procedure considering the is not well understood and microwave technology must rely on state of knowledge of the chemistry of binder aging and the high pressures to simulate long-term aging, it was not consid- requirements of an ideal aging test developed in this project. ered a viable approach for further development in Project 9-36. 3.2.1.1 Microwave Technology 3.2.1.2 Thin Films A significant amount of work has been completed by Bishara Laboratory aging of binders in thin films at elevated tem- and his colleagues on developing short- and long-term aging perature has been the method of choice of asphalt technologists

OCR for page 15
16 for over 60 years. An asphalt binder's exposure to air during air bubbles can be dispersed in highly viscous modified binders mixing and in asphalt concrete mixtures is in the form of a at temperatures considered reasonable for a long-term aging thin film. The current RTFOT was developed in California in test. There also is concern regarding the partial pressure of the response to the need for a short-term aging test (14). In an volatiles within the bubbles and whether the thermodynamics attempt to simulate long-term aging, Griffin and his colleagues of volatile transfer to the atmosphere is duplicated within the (15) performed extensive studies with very thin films and closed bubbling system. However, since air blowing produces developed a special viscometer to characterize the viscosity of oxidation reactions that are similar to those of thin films, air the binder (1617). More recently, Petersen (18) reported on blowing was considered a viable approach for Project 9-36. the Thin Film Aging Test in which thin films of asphalt binder In summary, the review of the literature and research in are exposed to the atmosphere at service temperatures. The progress identified significant post-SHRP research on binder disadvantage of these tests, which may be considered as ultra- aging methods using the three approaches of microwave tech- thin film tests, is that they yield a very small quantity of nology, thin films, and air blowing. Of these three, microwave material, making them impractical for specification use. Thin technology was eliminated because the mechanism whereby film aging is the approach used in the Thin Film Oven Test, microwave energy accelerates the aging process is not well (AASHTO T170), the Rolling Thin Film Oven Test (AASHTO understood, and high pressures are needed for simulation of T240), the Pressure Aging Vessel (AASHTO R28), and two long-term aging. The use of thin films for laboratory aging is European methods (the German Rotating Flask and the Rolling well established and accepted by asphalt technologists. Air Cylinder Aging Test). Welborn (19) presented an excellent blowing is similar to thin film reactions, provided the air bub- summary of the development of the various U.S. laboratory bles remain small. Candidate methods for an improved short- aging tests based on thin films. The primary difference in all term binder aging test based on thin films and air blowing are of these methods is the thickness of the film and the method discussed in the following section. used to obtain the thin film. Methods based on thin films have been used extensively to 3.2.2 Methods Based on Viable Approaches simulate both short- and long-term aging. Our current knowl- edge of the chemistry of binder aging is founded in the interpre- 3.2.2.1 German Rotating Flask tation of large amounts of data obtained using this approach The German Rotating Flask (GRF) is the common name (20). Clearly, improved methods based on thin films should be given to German Standard DIN 52 016, Testing the Thermal considered as viable approaches for NCHRP Project 9-36. Stability of Bitumen in a Rotating Flask. This test was developed in Germany as an inexpensive alternative to the RTFOT. It 3.2.1.3 Air Blowing uses a rotary evaporator similar to that used in ASTM D5404 and AASHTO T319 for recovery of binders after solvent Air blowing was one of the first approaches recommended as extraction. Figure 2-3 showed a schematic of the GRF. In a laboratory aging test for asphalt binders. In 1937, Nicholson the test, air is introduced into the rotating flask to age the (21), and Rashig and Doyle (22) proposed short duration, binder. A 100-g sample is aged in the rotating flask at 165C high temperature, high airflow tests for accelerated aging of for 160 min, 10 min without airflow, followed by an addi- asphalt binders. In these tests, a 250-g sample of asphalt was tional 150 min with air at ambient temperature flowing at the aged at 218C for 15 minutes with an airflow rate of 9 L/min. rate of 500 mL/min. Later, Skidmore (23) proposed a longer, lower temperature Three studies to evaluate the GRF as an alternative short- air blowing test that aged a 100-g sample at 177C for 2 hours term aging procedure have been conducted in the United using an airflow rate of 1 L/min. Although these tests reason- States. Sirin et al. (24) and Tia et al. (25) evaluated the test for ably reproduced the ductility and penetration of asphalts the Florida Department of Transportation for use with mod- recovered from recently constructed pavements, an air blow- ified asphalts that could not be properly aged in the RTFOT. ing test for simulating short-term aging was never standard- They concluded that the GRF could be used to simulate ized. Little work was done using this approach after 1940 short-term aging in hot mix plants and that various degrees until the development of the Stirred Air Flow Test (SAFT) by of aging could be obtained by varying the temperature, air- Glover et al. (1) in 2001. flow, duration, and sample weight used in the test. They rec- Laboratory aging using thin films and air blowing are ommended a cover for the oil bath and the use of a Morton mechanically similar, as long as the air bubbles remain small flask to better control temperature and provide uniform mix- and well dispersed. Smaller air bubbles increase the ratio of ing of modified binders. They also recommended testing con- surface area to volume of the asphalt, and the reaction becomes ditions to approximately reproduce the aging that occurred less oxygen-diffusion limited and behaves more like a thin in the TFOT (AASHTO T170) and the RTFOT. Reported film reaction. The primary concern with air blowing is whether mass changes for the modified test incorporating the cover

OCR for page 15
17 Table 3-1. Properties used in the MGRF optimization. ties and mass change revealed that the MGRF consistently aged binders less than the RTFOT. In the second and third Property Conditions Value G*/sin Short-term aged Temperature at 2.2 kPa phases the airflow was increased to 1 L/min, and 2 L/min, G*sin PAV aged after short-term aging Temperature at 5,000 kPa respectively. The results were better, but the degree of aging S PAV aged after short-term aging Temperature at 200 MPa was still less than in the RTFOT. Based on these results, the m-value PAV aged after short-term aging Temperature at 0.300 Mass Loss Short-term aged Report round-bottom flask was replaced with a 2-L Morton flask to increase the surface area of the binder during the aging process. All eight MRL asphalts plus three modified binders (Styrelf, and Morton flask, and using an airflow rate of 4 L/min, were Ultrapave, and Novophalt) were aged using the Morton flask greater than those for the RTFOT. at 2 L/min airflow during the fourth phase of testing. The At the same time as the Florida studies, the Western Research resulting rheological property data are shown in Figure 3-1 Institute under FHWA's contract, "Fundamentals of Asphalts for the neat binders. The agreement for the two methods is and Modified Asphalts," initiated a study to modify the Ger- excellent over a very wide range of temperatures. Mass change man Standard DIN 52 016 to simulate RTFOT aging (2). The also was considered in the study. Figure 3-2 compares mass study included four phases and ultimately led to the develop- change data from the MGRF and RTFOT. The MGRF pro- ment of a draft AASHTO standard test method. In each of duced less mass loss than the RTFOT, particularly for binders these phases the properties listed in Table 3-1 were compared with high mass losses. Table 3-2 summarizes the operating for binders short-term aged in various modifications to the conditions for the MGRF. GRF and the RTFOT. In the first phase, the method was Ramaiah and D'Angelo have conducted a study for FHWA altered to test 200 grams of binder using a 2-L round-bottom to further evaluate the MGRF (26). This project included stud- flask at 165C with a flow rate of 500 mL/min. Eight Materials ies to assess the effect of variations in the operating parameters Reference Library (MRL) asphalts (AAA-1, AAB-1, AAC-1, to establish initial tolerances, and a study to compare short- AAD-1, AAF-1, ABM-1, AAK-1, and AAM-1) were aged in term aging from the MGRF with the RTFOT for five polymer the Modified German Rolling Flask (MGRF) with split sam- modified binders, one air blown binder, and three neat binders. ples aged in the RTFOT. Comparisons of rheological proper- The tolerances established from this study are included in 75 G*/sin = 2.2 kPa 65 G*sin = 5000 kPa GRADING TEMPERATURE FOR MGRF RESIDUE, C S = 300 MPa 55 m = 0.300 Equality 45 35 25 15 5 -5 -15 -25 -25 -15 -5 5 15 25 35 45 55 65 75 GRADING TEMPERATURE FOR RTFOT RESIDUE, C Figure 3-1. Comparison of continuous grade temperatures from the MGRF and RTFOT for eight MRL asphalts.

OCR for page 15
18 0.2 AAB-1 0 AAM-1 ABM-1 AAA-1 AAF-1 -0.2 MGRF MASS CHANGE, % AAK-1 AAC-1 AAD-1 -0.4 -0.6 -0.8 -1 -1.2 -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 RTFOT MASS CHANGE, % Figure 3-2. Comparison of mass change data from the MGRF and RTFOT. Table 3-2. Figures 3-3 and 3-4 present rheological and mass temperatures. The long-term aging is performed at 95C or change data for the binders used in the FHWA study. The 103C depending on the grade of the binder. During the aging, rheological data in Figure 3-3 shows excellent agreement with oxygen is introduced into the flask at the rate of 7 L/h, and the the RTFOT similar to that from the Western Research Insti- duration of the test is 47 hours. Table 3-3 summarizes oper- tute study. The mass change data, on the other hand, is sig- ating conditions for the LTRF. nificantly different than that from the Western Research The degree of aging obtained with the LTRF was compared Institute study. The conclusion from the FHWA study is that to that obtained with the Rotating Cylinder Aging Test the mass changes in the MGRF are greater than those in the (described in the next section of this report) using nine RTFOT. This conclusion is in general agreement with that unmodified and nine modified binders. Srensen (27) reported from the Florida studies reported by Sirin et al. (24) and Tia only moderate correlation for various properties measured on et al. (25). binders aged in the two devices. Linear correlation coefficients Research was conducted in Germany to develop a long- ranged from 0.75 to 0.97. Properties considered included term aging procedure based on the GRF (27). This test, referred softening point, penetration, penetration index, Frass break- to as the Long-Term Rotating Flask (LTRF), uses the same ing point, ductility, complex shear modulus, creep stiffness, equipment as the GRF to long-term-age binders in an oxygen and m-value. atmosphere at lower temperatures. Three steel balls are intro- duced into the round-bottom flask from German Standard 3.2.2.2 Rotating Cylinder Aging Test DIN 52 016 to promote mixing of binders at the lower aging The Rotating Cylinder Aging Test (RCAT) was developed in Belgium to accurately simulate long-term aging (28). The Table 3-2. Recommended operating parameters device is shown in Figure 3-5. The main components are an for the MGRF. oven for temperature control, and a cylindrical vessel and Parameter Condition (2) Tolerance (26) rotating mechanism for aging the binder. The vessel includes Sample Size 200 g 1g a unique rotating shaft mechanism to keep the binder evenly Temperature 165C 1.5C dispersed in the vessel during the test. Although the device Airflow 2 L/min 0.04 L/min Rotational Speed 20 rpm 5 rpm was developed to simulate long-term aging, it can be used for Heat-Up Time 10 min with no air flow 1 min short-term aging using temperature and airflow rates that are Aging Time 200 min under air flow 1 min the same as those for the RTFOT.

OCR for page 15
19 80 70 GRADING TEMPERATURE FOR MGRF RESIDUE, C G*/sin = 2.2 kPa 60 m = 0.300 or S = 300 MPa Equality 50 40 30 20 10 0 -10 -20 -30 -40 -40 -30 -20 -10 0 10 20 30 40 50 60 70 80 GRADING TEMPERATURE FOR RTFOT RESIDUE, C Figure 3-3. Comparison of continuous grade temperatures from the MGRF and RTFOT from the FHWA study. 0.2 0 -0.2 MGRF MASS CHANGE, % ALF Control PG 70-22 SBS Radial -0.4 Grafted PG 58-22 Air Blown PG 70-28 -0.6 PG 58-22 EVA -0.8 Grafted -1 -1.2 -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 RTFOT MASS CHANGE, % Figure 3-4. Comparison of mass change data from the MGRF and RTFOT from the FHWA study.

OCR for page 15
20 Table 3-3. Operating parameters Table 3-4. Operating parameters for the LTRF. for the RCAT. Parameter Condition Parameter Condition Sample Size 100 g Test Sample Size 500 g Mixing 3, 30-mm diameter steel balls Short Term Temperature 163C Temperature 95 or 103C depending on binder grade Speed 1 rpm Rotational Speed 4 rpm Airflow Air at 4 L/min Atmosphere Oxygen at 7 L/h Aging Time 235 min Aging Time 47 h Long Term Temperature 90C Speed 1 rpm Airflow Oxygen at 4.5 L/hr Aging Time 140 hr Table 3-4 summarizes the operating conditions for the RCAT. Using the commercially produced version of the device, both short- and long-term aging can be performed with the same equipment. First, a 500-g sample of binder is short-term aging that occurs in the RTFOT and the long-term short-term aged. Upon completion of the short-term aging, aging that occurs in the PAV (29, 30). a portion of the sample is removed for physical property measurements. The remainder of the sample is then long- 3.2.2.3 Stirred Air Flow Test term aged. A unique concept included in the long-term aging procedure is the removal of samples at various times to allow Glover, et al. (1) proposed an air blowing test called the characterization of aging kinetics. Using the operating condi- Stirred Air Flow Test (SAFT) for short-term aging of asphalt tions listed in Table 3-4, the RCAT reasonably reproduces the binders. The test was shown schematically in Figure 2-2. In OVEN SHAFT AIR SUPPLY ROTATING VESSEL Figure 3-5. Rotating cylinder aging test.

OCR for page 15
21 the SAFT, air from a nozzle submerged in the binder is dis- Table 3-5. Proposed operating persed in the binder by an impeller mounted to an external parameters for the SAFT. motor. Through trial and error, operating parameters for Parameter Condition the device were determined that approximate the level of Sample Size 250 g aging that occurs in the RTFOT. This work included consid- Temperature 163C eration of the effects of sample size, impeller type, nozzle type, Airflow 2000 mL/Min Stirrer Speed 700 rpm airflow rate, and impeller speed on 60C viscosity measure- Heat-Up Time 15 Min under Nitrogen ments and carbonyl growth. Impeller speed had the greatest Aging Time 30 Min under Air effect on the viscosity and carbonyl growth measurements. Increasing the speed of the impeller significantly increased aging as measured by viscosity and carbonyl growth. The mass loss. It is less sensitive to binder source than in the other operating conditions considered had only a minor RTFOT mass change and the order of ranking varies. Glover, effect on the aging process. After appropriate operating et al. (1) also reported high variability for the condensed parameters were selected, changes in 60C complex viscos- volatile compounds measurements. Coefficients of variation ity and carbonyl growth were used to compare the device for multiple tests on the same asphalt were high, averaging and the RTFOT for several unmodified and modified approximately 35 percent. Glover, et al. (1) dismissed the binders. Figure 3-6 presents a typical comparison reported poor agreement between the RTFOT mass change and the for the SAFT. The proposed operating parameters are sum- mass of condensed volatile compounds collected in the SAFT marized in Table 3-5. A draft ASTM standard test method as being the result of errors in the RTFOT mass change meas- for the SAFT was developed. urements. They hypothesized that the high variability in the The SAFT also includes a volatile compound collection SAFT measurements may be the result of condensation of system that consists of a simple air-cooled condenser through volatile compounds on the lid caused by temperature gradi- which the exhaust air from the vessel passes before exiting to ents over the height of the vessel. They recommended using a the atmosphere. Volatile compounds produced during the heating mantel that heats the device over its entire height in short-term aging process become trapped in the condenser. an effort to reduce the potential for volatile compounds to They are removed by washing the condenser with solvent, condense on the lid. and then they are weighed after the solvent evaporates. Fig- Since its development, additional evaluation of the SAFT ure 3-7 presents a comparison of the mass change from the has been completed by the Texas Department of Transporta- RTFOT and the mass of volatile compounds collected with tion and FHWA. The Texas effort was directed at an inter- the SAFT as reported by Glover, et al. (1). As shown in Figure laboratory study to develop precision statistics for the SAFT 3-7, the mass of volatile compounds collected with the SAFT and to compare them to the RTFOT (31). Six laboratories is approximately an order of magnitude less than the RTFOT and six materials were used in the study. Precision statistics Figure 3-6. Comparison of viscosity for asphalts aged in the SAFT and RTFOT (1).

OCR for page 15
22 0.2 MASS OF CONDENSED VOLATILES IN SAFT, AC-20(2) AAA ABM AC-10 0.0 AAG AAF AAM AAS AC-20(3) AAD -0.2 % BY WEIGHT -0.4 -0.6 -0.8 -1.0 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 MASS CHANGE IN RTFOT, % BY WEIGHT Figure 3-7. Comparison of mass of condensed volatile compounds collected in the SAFT and the mass change from the RTFOT. were developed for G*/sin measured at the specified high- agreement with the RTFOT as short-term-aged binders from temperature grade of the material tested, and the mass of the MGRF. volatile compounds collected. For the degree of aging as meas- ured by G*/sin, the authors concluded that 3.2.3 Candidate Short-Term Binder 1. The degree of aging in the SAFT at the current operating Aging Procedures conditions is somewhat less than that in the RTFOT. The Based on the review of the literature and research in progress, aging time should be increased from 30 to 35 min to provide three candidate procedures were identified for further consid- better agreement between the two methods. eration in Project 9-36. They are the (1) MGRF, (2) RCAT, and 2. The precision statistics for the SAFT were slightly poorer (3) SAFT. This section discusses the advantages of adapting pos- than those for the RTFOT, but should improve as operators itive elements of the procedures in a hybrid procedure, and become more familiar with the device. presents the procedures that were recommended for further consideration in NCHRP Project 9-36. The authors also concluded that the volatile loss procedure in the SAFT worked very well and provided a significant improvement of the RTFOT mass change determinations. 3.2.3.1 Comparison of Candidate Tests This conclusion for the volatile loss procedure is suspect because three of the laboratories could not provide valid Table 3-6 compares pertinent details of the three candi- volatile loss data, and the corresponding data for the RTFOT date tests. This table is arranged to quickly compare the were not collected. three tests. The upper section presents general information The FHWA evaluation of the SAFT included a comparison about the test, such as the cost and complexity of the equip- of high- and low-temperature rheological property measure- ment, the quantity of material that can be aged, how easy it ments for four binders aged in the SAFT and the RTFOT (26). is to use and clean, and whether NCHRP Project 9-36 could The aging time in the SAFT test was 35 min as recommended take advantage of further development that was ongoing. by Glover, et al. (31). Figure 3-8 compares the continuous per- The second section presents specific information relative to formance grade temperature for binders that are short-term application of the test to short-term aging, and the last sec- aged in the two tests. As shown, rheological properties of short- tion presents specific information relative to long-term term-aged binders from the SAFT show a similar degree of aging.

OCR for page 15
23 80 70 G*/sin = 2.2 kPa GRADING TEMPERATURE FOR SAFT RESIDUE, C 60 m = 0.300 or S = 300 MPa Equality 50 40 30 20 10 0 -10 -20 -30 -40 -40 -30 -20 -10 0 10 20 30 40 50 60 70 80 GRADING TEMPERATURE FOR RTFOT RESIDUE, C Figure 3-8. Comparison of continuous performance grade temperatures for binders aged in the SAFT and RTFOT from the FHWA study. Table 3-6. Comparison of candidate methods. Considerations Rotating Cylinder Aging Test Modified German Rotating Stirred Air Flow Test Flask Neat and Modified Binders Yes Yes Yes Amount of Material 500 g 200 g 250 g Equipment Cost $18,000 $3,500 $5,000 estimated Equipment Complexity Moderately complex Simple Moderately complex General Availability Low High Low Test Complexity Reasonable Simple Reasonable Binder Recovery Easy Easy Easy Clean Up Solvent Ignition Oven Solvent Active Development Yes Yes Yes Published Procedure Yes Yes Yes Further Refinement Needed No No Yes Temperature 163C 165C 163C Duration 235 min 210 min 45 min Short- Atmosphere Air at 4 L/min Air at 2 L/min Air at 2 L/min Term Measure Volatility None--configuration of vessel Mass change--adaptable to Volatile recovery system Aging makes volatile recovery difficult volatile recovery Mimics Aged Binder Chemistry Yes Yes Yes Repeatability/Reproducibility Limited data Limited data Some data Adaptable to Long Term Yes--work completed Yes--steel balls to enhance Probably with new impeller mixing Published Procedure Yes Yes No Temperature 90C 95C neat Long- 103C modified Term Duration 140 hr 47 hr NA Aging Atmosphere Oxygen at 4.5 L/hr Oxygen at 7 L/hr Aging Kinetics Possible Yes Yes Yes Mimics Aged Binder Yes Yes Likely Chemistry Repeatability/Reproducibility Limited Limited None