Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging (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.

A-1   A P P E N D I X A Proposed Appendix X1 to AASHTO R 28, Modifications for 0.8 mm Thin Film Conditioning (12.5 g Binder Mass) and Commentary

A-2 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging 1. Proposed Appendix X1 to AASHTO R 28, Modifications for 0.8 mm Thin Film Conditioning (12.5 g Binder Mass) X1.1. PURPOSE X1.1.1. This appendix describes modifications to AASHTO R 28 equipment and procedures that are needed to successfully condition nominal 0.8 mm thin films of RTFOT residue in the pressurized aging vessel (PAV). X1.1.2. Research has shown the rheological properties of 0.8 mm film conditioned for 20 hours in the PAV to be approximately equivalent to standard 3.2 mm film conditioned for 40 hours at the same temperature in the PAV. X1.2. SUMMARY X1.2.1. A nominal 0.8 mm thin film is formed by weighing 12.5 g of RTFOT residue into a 140 mm diameter flat bottom stainless steel pan, then heating the level pan in a nitrogen atmosphere at 135°C to allow the binder to flow and cover the flat portion of the pan. X1.2.2. The pans are then placed in a level PAV and conditioned for 20 hours under 2.1 MPa air pressure at the temperature specified in AASHTO M 320 or AASHTO M 332 for the binder grade specified. Note X1 – The levelness tolerance required to form and condition 0.8 mm thin films is significantly tighter than the tolerance specified in Section 6.1.1 of AASHTO R 28. X1.2.3. The PAV-conditioned residue is then vacuum degassed before rheological property testing. X1.3. APPRATUS X1.3.1. Equipment Modifications X1.3.1.1. Stainless Steel Pans- Ten flat bottom stainless steel pans 9.5 mm deep with an inside diameter of 140 mm and having a thickness of 1.0 mm. Note X2 – Flat pans are needed to form and maintain a 0.8 mm thin film. Pans made from 1.0 mm stainless steel will not warp at the temperatures used in this practice. The pans may warp if subjected to higher temperatures for cleaning. The pans should be cleaned using mineral

Proposed Appendix X1 to AASHTO R 28, Modifications for 0.8 mm Thin Film Conditioning (12.5 g Binder Mass) and Commentary A-3   spirits, mineral oil, citrus-based solvents, toluene, or similar solvent followed by acetone to remove the solvent residue from the pans. X1.3.1.2. Pressure Aging Vessel System- A PAV system meeting the requirements of sections 6.1 of AASHTO R 28 except the pan holder when installed in the PAV must hold the pans level to within 0.025 degrees. The PAV system shall have a method for rapidly leveling the sample holder while it is installed in the PAV and verifying levelness using a precision machinist level. X1.3.1.3. Vacuum Oven- A vacuum oven to form the 0.8 mm thin film and degas the residue after conditioning in the PAV. The vacuum oven shall meet the following requirements. X1.3.1.3.1. Vacuum vessel- The vacuum vessel shall be designed for an internal pressure down to less than 1 kPa absolute and shall have: (1) a vacuum port, (2) a port for injecting nitrogen, (3) a vent port, and (4) a pressure relief valve. The pressure relief valve shall not permit pressure inside the vacuum vessel to increase significantly above atmospheric pressure. The vacuum vessel shall be sized to accept the loaded 10- pan holder used in the PAV system. The pan holder when installed in the vacuum vessel must hold the pans level to within 0.025 degrees. The vacuum vessel shall have a method for rapidly leveling the sample holder while it is installed in the vacuum vessel and verifying levelness using a precision machinist level. The vacuum vessel shall have a sight port and lighting for observing the film in the top pan of the pan holder. X1.3.1.3.2. Temperature control- The vacuum oven shall be capable of maintaining the temperature inside the vacuum vessel between 100°C and 180°C. Temperature shall be controlled by a temperature sensor, accurate to ±1°C, mounted inside the vacuum vessel. The temperature control system shall maintain the temperature at any point inside the vacuum vessel within ±5°C of the target temperature. Note X3 – Accurate temperature control cannot be maintained in the vacuum oven because the temperature will change when opening the vacuum vessel to load the pan holder and introducing nitrogen as required during film formation. X1.3.1.3.3. Vacuum System- A vacuum system consisting of: (1) a vacuum pump, (2) a vacuum gauge, and (3) piping, valves, and fittings. The vacuum system shall be designed to handle air and nitrogen at temperatures up to 180°C. The vacuum system shall be capable of reducing the pressure in the vacuum vessel to 15 kPa absolute or less in 2.5 minutes or less and maintaining the pressure at 15 kPa absolute ± 2.5 kPa for at least 30 minutes. The vacuum system shall have a vacuum gauge or

A-4 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging electronic pressure sensor with ±0.5 kPa accuracy and readable to 2 kPa or less. A flow valve shall be installed on the vent port to regulate the flow of nitrogen or air into the vacuum vessel. X1.3.2. Additional Equipment X1.3.2.1. Nitrogen Supply- Industrial grade nitrogen having a purity of 99.998 percent or better. Pressurized cylinders shall have appropriate regulators for safe introduction of nitrogen into the vacuum oven. X1.3.2.2. Precision Machinist Level- A precision machinist level for verifying the levelness of the pans in the PAV and the vacuum oven. The level shall have sensitivity of 0.00042 mm/mm (0.024 degrees) or better per division. Note X4 – Starrett No. 98 Precision Machinist Level 6 inches in length has been used for this purpose. X1.4. ADDITIONAL CALIBRATION AND STANDARDIZATION X1.4.1. Vacuum Oven Temperature Sensor- Standardize the vacuum oven temperature sensor to 1.0°C at least every 6 months using a calibrated thermometer. X1.4.2. Vacuum Gauge- Standardize the vacuum oven vacuum gauge or electronic absolute pressure sensor using a calibrated pressure sensor to ±0.5 kPa or a manometer to 4 mm Hg at least every 6 months. X1.4.3. Precision Machinists Level- Verify the level at least weekly, any time the level is mishandled, or when changes must be made after previous leveling of the equipment using the following procedure. X1.4.3.1. Select a stable, flat bearing surface. The bearing surface does not have to be perfectly level but must be level enough for the bubble to be within the measuring range of the level. Note X5 – The penetrometer stand is a good surface for this procedure. X1.4.3.2. Clean the bearing surface and the underside of the level to remove any asphalt binder, dust, or other material. X1.4.3.3. Place the level onto the bearing surface, mark the perimeter of the level base, and note the position of the bubble after it has settled. X1.4.3.4. Turn the level by 180° and place the level back onto the surface at the same location as X1.4.3.3. Note again the position of the bubble after it has settled.

Proposed Appendix X1 to AASHTO R 28, Modifications for 0.8 mm Thin Film Conditioning (12.5 g Binder Mass) and Commentary A-5   X1.4.3.5. The bubble should be in the same position in X1.4.3.3 and X1.4.3.4. If not, adjust the adjustment nut on the level and repeat the verification until the bubble is in the same position when the level is turned 180 °. X1.4.4. Vacuum Oven Temperature Offset – Determine the target control temperature needed for the vacuum oven to form 0.8 mm films at 135°C. Perform tests heating the vacuum oven to target temperatures between 135°C and 150°C. Open the vacuum vessel, insert the holder with empty pans into the vessel, close and seal the vessel, and perform the vacuum cycle purging as described in the procedure section below. Select the target temperature to achieve 135ºC at the end of the vacuum cycle purging. X1.5. MODIFIED PROCEDURE X1.5.1. Before each run, verify the levelness of the pans in the pan holder when installed in vacuum oven using the precision machinist level. Check levelness along perpendicular axes. Adjust as necessary to achieve a level plane within 0.025 degrees. X1.5.2. Before each run, verify the levelness of the pans in the pan holder when installed in the PAV using the precision machinist level. Check levelness along perpendicular axes. Adjust as necessary to achieve a level plane within 0.025 degrees. X1.5.3. Condition the asphalt binder in accordance with T 240 (RTFOT). X1.5.4. After combining the RTFOT residue into a single container and blending as specified in T 240, (1) pour the hot residue directly into the stainless steel pans for immediate conditioning in the PAV; or (2) pour the residue into the stainless steel pans, cover and set aside for conditioning at a later time; or (3) allow the residue to cool in the single container for conditioning at a later time. If the residue is stored in a single container, heat the residue, stir gently, and pour the heated residue into the pans. X1.5.5. Preheat the vacuum oven with the pan holder without the pans to the temperature determined in X1.4.4 to form 0.8 mm thin films at 135°C. X1.5.6. Select the conditioning temperature and preheat the PAV without the pan holder and pans to the selected aging temperature. Note X6 – If conditioning binder for conformance to M 320 or M 332, select the appropriate conditioning temperature from M 320 or M 332.

A-6 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging Note X7 – Preheating the vessel to 10°C to 15°C above the conditioning temperature can be used to reduce the drop in PAV temperature during the loading process and minimize the time required to stability the system after loading to attain the required temperature. X1.5.7. Place each stainless steel pan on the balance and add 12.5 ± 0.1 g of RTFOT residue to each pan. This amount of binder will yield a film thickness of approximately a 0.8 mm. Note X8 – For binder grading with ΔTc determination, condition eight 12.5 g pans for ΔTc determination and two 50.0 g pans for low- and intermediate-temperature grading. Only the 12.5 g pans should be subjected to the film formation described in X1.5.8 through X1.5.11. X1.5.8. After RTFOT residue has been added to all pans, remove the pan holder from the vacuum oven and load it with the pans having 12.5 g of RTFOT residue. Place the pan holder with loaded pans into the vacuum oven and close the vacuum oven. X1.5.9. Purge the vacuum vessel with nitrogen. Apply three purge cycles. Each purge cycle consists of: (1) using the vacuum system to reduce the pressure in the vacuum vessel to 15 ± 2.5 kPa absolute in 2.5 minutes or less; (2) introduce nitrogen through the vent port to return the pressure in the vacuum vessel to atmospheric pressure in 0.5 minutes or less. Note X9 – If the pressure gauge or sensor in the vacuum oven reads absolute pressure, use 15 ± 2.5 kPa. If the pressure gauge or sensor reads gauge pressure, use Table 1 in AASHTO R 28 to determine the gauge reading for the elevation of the laboratory. X1.5.10. Leave the pan holder and loaded pans in the vacuum oven under nitrogen at 135°C until the RTFOT residue forms a thin film covering the bottom of the pan. Use the sight port to determine when the RTFOT residue has fully coated the pan. Note X10 – RTFOT residue from neat binders may form a uniform 0.8 mm film before the end of the three cycles of vacuum cycle purging. Stop the film formation process when a uniform film is formed and continue with X1.5.11. RTFOT residue for heavily modified binders may require an additional 20 minutes at 135°C after vacuum cycle purging to spread out to cover the pan. X1.5.11. Open the vacuum oven and remove the pan and holder from vacuum vessel. If conditioning both 12.5 g and 50.0 g pans, add two 50.0 g pans to the holder.

Proposed Appendix X1 to AASHTO R 28, Modifications for 0.8 mm Thin Film Conditioning (12.5 g Binder Mass) and Commentary A-7   X1.5.12. Open the PAV. Place the holder loaded with pans into the PAV and close the PAV. X1.5.13. Connect the temperature transducer and air pressure supply to the PAV. X1.5.14. Wait until the temperature inside the PAV is within 20°C of the conditioning temperature, apply an air pressure of 2.1 ± 0.1 MPa, and then start timing the test. If the temperature inside the vessel has not reached the desired temperature for applying pressure within 2 hours of loading the pan holder and pans, discontinue the procedure and discard the samples. Note X11 – Pressures in excess of 2.1 MPa do not substantially increase the rate of aging. Therefore, higher pressures are not warranted. Note X12 – Once pressurized, the temperature inside the pressure vessel will equilibrate rapidly. The time under pressure, not to include any preheating time at ambient pressure, is the conditioning time. Relatively little aging occurs at ambient pressure during the time that the vessel is being reheated to the test temperature given that asphalt binder residue under test has been exposed to 163°C in the RTFOT. X1.5.15. Maintain the temperature and air pressure inside the pressure vessel for 20 hours ± 10 minutes. X1.5.16. Preheat the vacuum oven to 170°C ± 5°C. The vacuum oven should be preheated to allow the pan holder with PAV-conditioned residue to be transferred directly to the vacuum oven after PAV conditioning. X1.5.17. At the end of the 20-hour test period, slowly begin reducing the internal pressure of the PAV using the air pressure bleed valve. Adjust the bleed valve to an opening that requires 9 ± 1 minutes to equalize the internal and external pressures on the PAV, thus avoiding excessive bubbling and foaming of the asphalt binder. During this process it may be necessary to adjust the setting of the needle valve as the pressure drops in order to maintain an approximate linear rate of pressure decrease. Do not include the pressure release and equalization time as part of the 20-hour conditioning period. X1.5.18. If the temperature indicated by the temperature-recording device falls above or below the target conditioning temperature ±0.5°C for more than 60 minutes during the 20-hour aging period, declare the test invalid and discard the material. X1.5.19. Open the PAV, remove the pan holder with the PAV-conditioned residue from the PAV and immediately transfer it to the vacuum oven.

A-8 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging X1.5.20. Close the vacuum oven and maintain the temperature in the vacuum oven at 170°C ± 5°C for 15 ± 1 minutes without vacuum applied. X1.5.21. After 15 ± 1 minutes of equilibration, reduce the pressure in the vacuum vessel to 15 ± 2.5 kPa absolute and maintain this pressure for 30 ± 1 minutes. After 30 ± 1 minutes, vent the vacuum oven to atmospheric pressure. Note X13 – If the pressure gauge or sensor in the vacuum oven reads absolute pressure, use 15 ± 2.5 kPa. If the pressure gauge or sensor reads gauge pressure, use Table 1 in AASHTO R 28 to determine the gauge reading for the elevation of the laboratory. X1.5.22. Open the vacuum oven, remove the pan holder and pans with degassed PAV residue, and transfer the pans to an oven set at 170°C. Transfer the PAV residue by scraping it into containers for testing. Combine degassed PAV residue from 12.5 g pans into a single container followed by gentle stirring to blend the residue. Combine degassed PAV residue from 50.0 g pans to a single container followed by gentle stirring to blend the residue. X1.5.23. Prepare test specimens directly from the residue in the containers; subdivide the residues into smaller containers for future testing or set the containers aside for future testing. X1.6. MODIFIED REPORT X1.6.1. Report the following Information: X1.6.1.1. Sample Identification; X1.6.1.2. Time required from start of vacuum cycle purging to form 0.8 mm thin film, nearest 1 min; X1.6.1.3. Conditioning temperature, nearest 0.5°C; X1.6.1.4. Maximum and minimum conditioning temperature, nearest 0.1°C; X1.6.1.5. Total conditioning time, hours and minutes; and X1.6.1.6. Any deviations from the specified procedure.

Proposed Appendix X1 to AASHTO R 28, Modifications for 0.8 mm Thin Film Conditioning (12.5 g Binder Mass) and Commentary A-9   COMMENTARY FOR PROPOSED APPENDIX X1 TO AASHTO R28, MODIFICATIONS FOR 0.8 MM THIN FILM CONDITIONING (12.5 g BINDER MASS) INTRODUCTION The equipment and procedural changes needed to condition 12.5 g samples in the pressurized aging vessel (PAV) are described in this proposed Appendix X1 to AASHTO R 28. This commentary was prepared to summarize the research from National Cooperative Highway Research Program (NCHRP) Project 09-61, “Short- and Long-Term Binder Aging Methods to Accurately Reflect Aging in Asphalt Mixtures” that supports the practice described in this proposed appendix to AASHTO R 28. It is intended for those who will be responsible for adoption and future revision of the proposed appendix and AASHTO R 28. The sections in this commentary follow those in the proposed appendix. When appropriate, references are cited. If additional development work is needed for implementation, it is also described. X1.1. PURPOSE There is growing interest in adding ∆Tc on residue from 40-hr PAV conditioning as a specification criterion in AASHTO M 320 and AASHTO M 332 to address cracking related to asphalt binder aging (Asphalt Institute, 2019). Testing of several binders in NCHRP Project 09- 61 has shown that 12.5 g, 20-hr, 2.1 MPa PAV conditioning yields residue with rheological properties similar to 50.0 g, 40-hr, 2.1 MPa PAV conditioning, allowing residue for intermediate- and low-temperature grading and the 40-hr ∆Tc criterion to be conditioned in 20 hours using a single run of the PAV. With 10 pans available in the PAV, loading two pans with 50.0 g and eight pans with 12.5 g yields approximately 95 g of binder for low- and intermediate- temperature grading and 85 g to 90 g of binder for ∆Tc determination. This is sufficient residue for intermediate-temperature dynamic shear rheometer testing and for testing bending beam rheometer (BBR) beams at two different temperatures. In NCHRP Project 09-61, data for several binders were collected for 12.5 g, 20-hr, 2.1 MPa PAV conditioning and 50.0 g, 40-hr, 2.1 MPa PAV conditioning at the same temperature. Table A-1 summarizes Christensen-Anderson master curve parameters collected on 10 binders. The parameter, log (ωc), is the logarithm of the crossover frequency, which is a measure of the hardness of the binder. The rheological index, R, defines the shape of the master curve. More- aged binders have lower crossover frequencies and higher rheological indices resulting in a flatter master curve. Table A-2 presents creep stiffness and m-values from BBR testing of four binders. More-aged binders have higher creep stiffness and lower m-values. Figure A-1 shows a comparison of low-temperature continuous grade temperatures for 12.5 g, 20-hr, 2.1 MPa PAV conditioning and 50.0 g, 40-hr, 2.1 MPa PAV conditioning collected during a Canadian study (Erskine et al. 2012). The data are scattered around the line of equality indicating these two conditions are approximately equivalent. The data in these tables and figure support substituting 12.5 g, 20-hr, 2.1 MPa PAV conditioning for 50.0 g, 40-hr, 2.1 MPa PAV conditioning.

A-10 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging Table A-1. Comparison of Christensen-Andersen Master Curve Parameters for 12.5 g, 20- hr, 2.1 MPa PAV conditioning and 50 g, 40-hr, 2.1 MPa PAV conditioning. Binder Mass, g Time, hr R log ωc AZ1-1 50.0 40 3.14 −1.3912.5 20 3.11 −1.43 AZ1-2 50.0 40 2.66 −0.3412.5 20 2.49 0.01 AZ1-3 50.0 40 2.35 −0.2512.5 20 2.44 −0.65 AZ1-4 50.0 40 2.34 −0.6012.5 20 2.34 −1.00 MN1-2 50.0 40 3.38 −0.6612.5 20 3.28 −0.61 MN1-3 50.0 40 2.60 0.5712.5 20 2.66 0.39 MN1-4 50.0 40 2.96 0.2412.5 20 2.97 0.27 MN1-5 50.0 40 2.14 1.9412.5 20 2.09 1.94 AAC-1 50.0 40 2.49 0.6912.5 20 2.50 0.73 AAF-1 50.0 40 2.29 0.2612.5 20 2.34 0.06 Table A-2. Comparison of BBR Stiffness and m-value Data for 12.5 g, 20-hr, 2.1 MPa PAV conditioning and 50.0 g, 40-hr, 2.1 MPa PAV conditioning. Binder Mass, g Time, hr Test Temp., °C Creep Stiffness, MPa m- value AAC-1 50.0 40 0.0 90.0 0.357 12.5 20 0.0 90.3 0.362 AAF-1 50.0 40 −6.0 78.4 0.380 12.5 20 −6.0 80.6 0.371 AC 1928 PG 76-34 50.0 40 −12.0 44.1 0.304 12.5 20 −12.0 48.4 0.306 50.0 40 −18.0 83.1 0.288 12.5 20 −18.0 88.7 0.287 ME 3 PG 76-28 50.0 40 −18.0 240 0.296 12.5 20 −18.0 233 0.301

Proposed Appendix X1 to AASHTO R 28, Modifications for 0.8 mm Thin Film Conditioning (12.5 g Binder Mass) and Commentary A-11   Figure A-1. Comparison of PAV Film Thickness and Aging Time (Data from Erskine et al. 2012). X1.2. SUMMARY The main differences between PAV conditioning using 12.5 g samples and 50.0 g samples are: 1. The pan must remain flat and not warp. In NCHRP Project 09-61, thicker pans, 1.0 mm thick compared to 0.6 mm specified in AASHTO R 28, were used. The pans were cleaned with solvent. These pans did not warp through approximately 100 conditioning cycles using conditioning temperature from 85ºC to 120ºC. 2. The pan must be level during film formation and conditioning. There is a discrepancy in AASHTO R 28 concerning levelness of the pans in the PAV. Section 6.1.1 specifies “… the asphalt binder film thickness in the bottom of the pans does not vary by more than 0.5 mm across any diameter of the pan.” This is a tolerance that can be met by leveling with a precision bull’s-eye level having a sensitivity of 0.0035 mm/mm (0.2 degrees) or less. Note 1 in Figure 1 in AASHTO R 28 states: “…Provisions shall be made to ensure that the bottom of the vessel is leveled so that the thickness of the binder in the pan varies by no more than ± 0.05 mm across the diameter of the pan.” This is a tighter tolerance of 0.1 mm that requires using a level having a sensitivity of 0.0007 mm/mm (0.04 degrees) or less. In NCHRP Project 09-61, the pans were leveled in the PAV to using a 6-inch precision machinist level having a sensitivity of 0.00042 mm/mm (0.024 degrees). This was a tighter tolerance than either value specified in AASHTO R 28 and results in variation in film thickness of less than 0.06 mm. -34 -32 -30 -28 -26 -24 -22 -34 -32 -30 -28 -26 -24 -22 20 H ou r, 0. 8 m m T hi ck ne ss P AV L ow T em pe ra tu re Co nti nu ou s G ra de , ° C 40 hour, 3.18 mm Thickness PAV Low Temperature Continuous Grade, °C

A-12 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging 3. The 3.2 mm film in AASHTO R 28 is formed by weighing 50.0 g of RTFOT residue into the pan and then allowing it spread out to fully coat the pan during conditioning. For 0.8 mm film using 12.5 g of RTFOT residue, this same approach was used in NCHRP Project 09-61 with neat binders conditioned between 85ºC and 115ºC and resulted in uniform films that fully coated the pans. However, when some heavily modified binders were conditioned as part of a sensitivity experiment, it was observed that these binders did not fully coat the pan at 100ºC and 105ºC. Additional work on film formation determined that heavily modified RTFOT residue would form a uniform film coating the entire pan when heated to 135ºC for 30 minutes. This film formation step must be performed under nitrogen to minimize unwanted oxidation of the binder. 4. The more-aged residue obtained from 12.5 g PAV conditioning must be vacuum degassed. Air trapped in the binder cannot be removed by heating and stirring the residue. X1.3. APPRATUS This section of the appendix describes modifications to equipment specified in AASHTO R 28 and additional equipment needed to effectively condition 0.8 mm films in the PAV. X1.3.1. Equipment Modifications X1.3.1.1 Pans Section 6.2 of AASHTO R 28 specifies the dimensions of the pans to be used for PAV conditioning. This section specifies “… thickness of approximately 0.635 mm (0.025 in).” This language came from AASHTO T 179, Standard Method of Test for Effect of Heat and Air on Asphalt Materials (Thin-Film Oven Test). Note 2 in AASHTO 179 indicates 24 gauge stainless steel (0.6 mm thick) pans are acceptable; and 26 gauge stainless steel (0.5 mm thick) pans are also acceptable but have a greater tendency to warp. Note 2 states “… in no case shall the thickness of the metal be less than 0.381 mm (0.015 in).” Therefore, pans meeting AASHTO T 179 requirements do not necessarily meet AASHTO R 28 requirements. Through initial testing of 0.8 mm films in NCHRP Project 09-61, it was determined that these pans did not remain flat enough to consistently form a film coating the entire pan. Thicker, 1.0 mm thick pans were purchased from Infratest, USA and used for the 12.5 g PAV conditioning. When levelled to 0.00042 mm/mm (0.024 degrees) using a precision machinist level, 0.8 mm films were successfully formed using these pans. The set of 10 pans was used for approximately 100 cycles to condition 0.8 mm thin films at temperatures ranging from 85ºC to 120ºC without warping. The pans were cleaned with mineral spirits and then rinsed with acetone. X1.3.1.2. Pressure Aging Vessel System All PAV conditioning for NCHRP Project 09-61 was done in an ATS Pressure Aging Vessel 3 (PAV3). Through initial testing, it was determined that more precise leveling than normally performed for AASHTO R 28 conditioning was needed to form and maintain the 0.8 mm thin film. For each PAV run, the ATS pan holder was levelled using a certified 6-inch Starrett No. 98 Precision Machinist Level that has a sensitivity of 0.00042 mm/mm (0.024 degrees). Leveling

Proposed Appendix X1 to AASHTO R 28, Modifications for 0.8 mm Thin Film Conditioning (12.5 g Binder Mass) and Commentary A-13   the ATS PAV3 is somewhat cumbersome. The leveling system for the pan holder consists of set screws in a ring at the bottom of the PAV as shown in Figure A-2. The set screws cannot be accessed with the pan holder in place, and the resistance temperature detector (RTD) projects above the leveling ring making it difficult to verify the levelness of the pans. Additionally, the set screws have coarse threads. The method used in NCHRP Project 09-61 to verify the levelness was to check levelness by placing the machinist level across the holder at the top while the holder was in place. The holder was then removed, and small adjustments were made to the screws. The holder was replaced, and the levelness checked. This process was repeated several times until the holder was level within the sensitivity of the machinist level. It should be noted that ATS’s specifications for the PAV3 shows the slots holding the pans to be parallel to within 0.05 mm. Over the pan diameter of 140 mm, this is 0.00036 mm/mm (0.020 degrees), which is a tighter tolerance than the sensitivity of the level used to check levelness. This is one-half of the tolerance in Note 1 of Figure 1 in AASHTO R 28. PAV holders from other suppliers may not have shelves meeting this tolerance. Figure A-2. Photograph of ATS PAV3 Leveling Ring. With PAVs available from several suppliers (Prentex, ATS, Infratest, Matest, Controls, and Cox / Cooper), Appendix X1 for thin film conditioning specifies the PAV system must provide a Leveling Ring Adjustment Screw RTD

A-14 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging method to rapidly level the sample holder while in the PAV to within 0.025 degrees and verify the levelness using a precision machinist level. For the ATS PAV, this can be accomplished by modifying the leveling ring to move the leveling screws outside of the holder, using fine thread screws, and providing a long handle adjustment wrench. Providing this capability may be more difficult for other PAV manufacturers. Additionally, it may be possible to build a verification device that uses electrolytic tilt sensors or use these sensors to provide automatic leveling. One manufacturer has an electrolytic tilt sensor with ±3 degree range and ±0.001 degree accuracy/repeatability (Fredericks Company 2020). If 0.8 mm thin film conditioning is considered for adoption, PAV manufacturers should be surveyed to determine the cost of modifying existing equipment to meet the levelness requirement of 0.025 degrees in this appendix. Modified equipment from various manufacturers should then be evaluated to confirm compliance with the levelness requirement and evaluate the time and effort required to level the pans. X1.3.1.3. Vacuum Oven The vacuum oven specified in this section is needed for forming the 0.8 mm thin film under nitrogen and to degas the residue after conditioning. In the PAV operating parameters experiment conducted during NCHRP Project 09-61, RTFOT residues from eight binders were conditioned at 100ºC and 120ºC using 0.8 mm thin films. In the long-term calibration experiment in NCHRP Project 09-61, RTFOT residues from 25 additional binders were conditioned using 0.8 mm thin films at three temperatures: 85ºC, 100ºC, and 115ºC. These two experiments used mostly neat binders. Only two modified binders were used. One was a terpolymer-modified PG 58-34 and the other was an AC-10 modified with 3 percent latex. In all cases, when the pans were leveled using a precision machinist’s level, films that completely covered the 140 mm diameter pan were formed using the approach described in AASHTO R 28 where the binder is weighed into the pans and then allowed to spread out to form a film during the 20-hour conditioning period. During the sensitivity experiment in NCHRP Project 09-61, RTFOT residue from four modified binders and four neat binders were conditioned at temperatures ranging from 95ºC to 105ºC. During this study, it was observed that the RTFOT residue from three PG 76 binders did not fully coat the pan during the conditioning period at temperatures of 100ºC and 105ºC. Limited additional work was conducted to determine if a proper film could be formed at a higher temperature. This work was conducted using a general-purpose vacuum oven meeting the requirements in AASHTO R 28 with a 12 in by 8 in by 2 in thick, Grade B granite inspection block in the oven to provide a flat surface for the pans. The inspection block was leveled using a precision machinist level and shims. The films were formed two pans at a time in the oven. Vacuum cycle purging with nitrogen was used to minimize oxidation during film formation. The rule of thumb for vacuum cycle purging is given by Equation 1 (Branan 2002). (1) where C(n) = concentration after n cycles, Co = initial concentration, and P = absolute pressure, n = number of purge cycles.

Proposed Appendix X1 to AASHTO R 28, Modifications for 0.8 mm Thin Film Conditioning (12.5 g Binder Mass) and Commentary A-15   Using the absolute pressure of 0.15 atm specified in AASHTO R 28 for degassing, the 21 percent oxygen concentration of air reduces to less than 0.1 percent after three cycles. With the equipment used in NCHRP Project 09-61, each purge cycle required approximately 3 minutes; 2.5 minutes to evacuate and 0.5 minutes to allow nitrogen to refill the oven. Three cycles of purging required approximately 9 minutes. Industrial grade nitrogen was used in NCHRP Project 09-61. This approach for forming 0.8 mm thin films was used with four binders, PG 76-34, PG 76- 28, PG 76-22, and PG 82-22. For these binders, proper 0.8 mm films were formed in less than 30 minutes using a target temperature of 135ºC. It is difficult to accurately control temperature during film formation because of heat losses that occur when opening the oven to insert the loaded pans and during vacuum cycle purging. The initial temperature of the oven was set by trial and error to 141ºC to account for the heat losses. Testing of residue before and after 0.8 mm film formation was conducted on the PG 76-34 binder and the PG 82-22 binder after 30 minutes under nitrogen at 135ºC to confirm oxidation was minimized. The resulting data are presented in Table A-3. The difference in G*/sinδ for the thin film compared to the RTFOT residue after a total of 30 minutes from the start of vacuum cycle purging ranged from 3.3 to 6.3 percent of the average. Table A-3. Effect of 0.8 mm Film Formation Under Nitrogen. Binder Treatment TemperatureºC G* kPa δ degree G*/sinδ G*/sinδ Difference % of Average PG 76-34 RTFOT 88 1.47 44.1 2.12 6.40.8 mm Film 88 1.56 43.4 2.26 PG 88-22 RTFOT 94 1.77 48.2 1.77 3.30.8 mm Film 94 1.83 47.7 1.83 RTFOT 100 1.26 49.8 1.26 3.90.8 mm Film 100 1.31 49.2 1.31 The vacuum oven specified in Appendix X1 was not used in the NCHRP Project 09-61 research. The primary difference between the specified vacuum oven and the approach used in the NCHRP Project 09-61 research is the ability to form 0.8 mm thin films simultaneously on 10 pans in the holder to be used for PAV conditioning. It is also envisioned that the 0.8 mm thin films would be degassed in this vacuum oven by returning the holder loaded with conditioned residue to this oven for degassing. After degassing, the residue from the pans would be scraped into a container for subsequent testing. This will accelerate the film formation, conditioning, and degassing process, making it more acceptable to practicing technicians. If 0.8 mm thin film conditioning is considered for adoption, PAV manufacturers should be asked to develop prototype vacuum ovens and demonstrate their use. Additional research should be conducted to better define the conditions for forming 0.8 mm thin films. Care should be exercised not to over specify the temperature and vacuum control. Accurate control of temperature will not be possible

A-16 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging due to heat losses that occur when opening the oven to insert the loaded pans and during vacuum cycle purging. Based on Equation 1, high accuracy is not needed for the vacuum control or measurement. The temperature control, temperature measurement, vacuum control, and vacuum measurement tolerances provided in Appendix X1 are estimates based on AASHTO R 28 and the limited experience gained in NCHRP Project 09-61. X1.3.2. Additional Equipment X1.3.2.1. Nitrogen Supply Nitrogen is needed for forming the 0.8 mm thin film. Typical purity specifications for industrial-grade nitrogen is 99.998 percent nitrogen (Air Products 2020). As discussed above, the vacuum cycle purging reduces to the oxygen concentration in the oven to less than 0.1 percent. Industrial grade nitrogen at 99.998 percent purity has a maximum oxygen concentration of 0.002 percent, which is 50.0 times less than the purged concentration in the oven. X1.3.2.2. Precision Machinist Level The accuracy of a digital level or sensitivity of a glass vial level is the important characteristic. Many digital levels have resolution better than the sensitivity specified for the precision machinist level, but their accuracy is much poorer. The precision machinist level specified offers excellent sensitivity at a reasonable price. It can be purchased with certification that it complies with the sensitivity of 0.00042 mm/mm. X1.4. ADDITIONAL CALIBRATION AND STANDARDIZATION This section of Appendix X1 describes standardization of the temperature sensor and vacuum gauge for the vacuum oven. It also describes a procedure to verify the precision machinist level. This is a well-known procedure for verifying levels. It is not necessary to have a perfectly level surface to perform verification of the level. A flat, stable surface that is level to within the range of the level is needed. A penetrometer base, if available in the lab, is a good surface to perform verification of the level. This section also describes a procedure to determine an offset temperature for the vacuum oven for forming 0.8 mm thin films at 135ºC. The offset for the determined in this manner for the oven used in NCHRP Project 09-61 was 141ºC. X1.5. MODIFIED PROCEDURE This section of Appendix X1 describes the procedure for forming 0.8 mm thin films and then conditioning those films for 20 hours in the PAV. Steps that are identical to Section 10 of AASHTO R 28 are repeated here for clarity rather than referring to AASHTO R 28. Information from NCHRP Project 09-61 supporting the changes is discussed below. X1.5.1. Pans must be level in the vacuum oven to form a 0.8 mm thin film. The equipment may have been accidentally bumped since its last use.

Proposed Appendix X1 to AASHTO R 28, Modifications for 0.8 mm Thin Film Conditioning (12.5 g Binder Mass) and Commentary A-17   X1.5.2. Pans must be level in the PAV to maintain and condition a thin film. The equipment may have been accidentally bumped since its last use. X1.5.3. Taken from AASHTO R 28. X1.5.4. Taken from AASHTO R 28. X1.5.5. The vacuum oven must be heated to a temperature higher than the film formation temperature to account for heat losses that occur when opening the oven to insert the pan holder with pans and during the vacuum cycle purging. X1.4.4 describes how to determine an appropriate offset temperature. For NCHRP Project 09-61, the preheat temperature was 141ºC. X1.5.6. It is not necessary to preheat the holder. The holder will be heated in the vacuum oven before putting the holder and pans in the PAV during the film formation. Note X6 is from AASHTO R 28 modified to include AASHTO M 332. Note X7 is from AASHTO R 28. X1.5.7. Tolerance of ±0.1 g was determined from a film thickness study completed during NCHRP Project 09-61 that investigated the change in various rheological properties as a function of film thickness. That study included short-term and long-term conditioning of binder films between 0.8 mm and 3.2 mm for SHRP binders AAC-1 and AAF-1. Plots of rheological properties as a function of binder film thickness were prepared, and the slope at 0.8 mm was used to determine the change in film thickness that would result in rheological property changes equal to the within-laboratory coefficient of variation for various tests. The results are summarized in Table A-4. The allowable film thickness variation was then converted to an allowable mass variation by multiplying the allowable film thickness variation by the pan area and the density of the binder. From Table A-4, short-term conditioned binder is most sensitive to film thickness changes. Since ±0.1 g was determined to be achievable with reasonable effort in testing done during NCHRP Project 09-61, it was included as the tolerance in the proposed Appendix X1. Table A-4. Summary of Film Thickness Tolerance Calculations for Properties Measured on Short- and Long-Term Conditioned Residue. Property Binder Slope Property at 0.8 mm Film Thickness %/mm Allowable Variation Film Thickness, mm Mass, g G*/sinδ AAC-1 −2.050.0 kPa/mm 3.27 kPa −62.6 0.05 0.50 AAF-1 −3.485 kPa/mm 3.26 kPa −106.8 0.03 0.29 G·sinδ AAC-1 −526.4 kPa/mm 3143 kPa −16.8 0.29 2.83 AAF-1 −1705.1 kPa/mm 7156 kPa −16.4 0.30 2.90 Creep Stiffness AAC-1 −9.84 MPa/mm 91.5 MPa −10.8 0.23 2.25 AAF-1 −24.43 MPa/mm 80.4 MPa −30.4 0.08 0.80 m- value AAC-1 0.010400 /mm 0.362 2.9 0.35 3.37 AAF-1 0.024100 /mm 0.373 6.5 0.15 1.50

A-18 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging Note X8 describes how to load pans for the conditioning needed to perform binder grading and ΔTc determination. One hundred grams of binder will be conditioned at 3.2 mm thickness, and 100 g of binder will be conditioned at 0.8 mm film thickness. Accounting for binder loss during scraping, this will provide about 85 g to 90 g of binder for rheological testing, which is sufficient for BBR testing of two beams at two temperatures and 8 mm dynamic shear rheometer testing at multiple temperatures. X1.5.8. First step to form a 0.8 mm thin film. X1.5.9. Based on Equation 1 presented earlier, vacuum cycle purging to 15 kPa absolute or 0.15 atm reduces the oxygen concentration in the vacuum oven to less than 0.1 percent. Two binders were tested after 30 minutes under nitrogen to confirm that limited additional stiffening occurred during film formation. Note X9 refers to Table 1 in AASHTO R 28 for equipment that uses gauge pressure. The first column heading in Table 1 in AASHTO R 28 is incorrect. This heading should read “Corrected Gauge Readings (inHg).” X1.5.10. 0.8 mm films were successfully formed in NCHRP Project 09-61 in 30 minutes or less for four heavily modified binders: PG 76-34, PG 76-28, PG 76-22, and PG 82-22. X1.5.11. Last step of film formation. X1.5.12. Transfer pans and holder to PAV. X1.5.13. Taken from AASHTO R 28. X1.5.14. Taken from AASHTO R 28. Notes X11 and X12 taken from AASHTO R 28. X1.5.15. Taken from AASHTO R 28. X1.5.16. Work in NCHRP Project 09-61 found that vacuum degassing is necessary for 0.8 mm film conditioned in the PAV to remove the air from the stiffer film. The vacuum oven should be preheated before PAV conditioning is finished. X1.5.17. Taken from AASHTO R 28. X1.5.18. Taken from AASHTO R 28. X1.5.19. The vacuum oven specified permits direct transfer of the holder with conditioned PAV residue from the PAV to the vacuum oven for degassing. This should speed the overall conditioning process. Note, the vacuum oven specified for this process was not available during NCHRP Project 09-61. Additional development work is needed to verify the acceptability of the process described. X1.5.20. Taken from AASHTO R 28. It may be possible to reduce the time or eliminate this heating step because the PAV residue is being transferred directly from the PAV.

Proposed Appendix X1 to AASHTO R 28, Modifications for 0.8 mm Thin Film Conditioning (12.5 g Binder Mass) and Commentary A-19   X1.5.21. Taken from AASHTO R 28. Note X13 refers to Table 1 in AASHTO R 28 for equipment that uses gauge pressure. The first column heading in Table 1 in AASHTO R 28 is incorrect. This heading should read Corrected Gauge Readings (inHg). X1.5.22. Taken from AASHTO R 28. X1.5.23. Taken from AASHTO R 28. X1.6. MODIFIED REPORT This section of the appendix describes the reporting requirements. All except X1.6.1 are the same as the reporting in AASHTO R 28. X1.6.1 requires documentation of any deviations from the procedures in the appendix. REFERENCES Air Products. Industrial and Medical Gas Specifications (U.S.). http://www.airproducts.com/Products/Gases/gas-facts/industrial-and-medial-gas-specifications- us.aspx. Accessed 8/2/2020. Asphalt Institute. “Use of the Delta Tc Parameter to Characterize Asphalt Binder Behavior, IS- 240.” Available from: http://www.asphaltinstitute.org/engineering/delta-tc-technical-document/, 2019. Branan, C. Rules of Thumb for Chemical Engineers. Gulf Professional Publishing, 2002. Erskine, J.A., Hesp, S.A., and Kavej, F. “Another Look At Accelerated Aging Of Asphalt Cements In The Pressure Aging Vessel.” 5th Eurasphalt & Eurobitume Congress, 13–15th June 2012, Istanbul. Fredericks Company. 0703-0711-99 ±3° TrueTILT™ Single-Axis Electrolytic Tilt Sensor. https://www.frederickscompany.com/products/0703-0711-99/#1476383581128-d8e57e33-4ae7. Accessed 8/6/2020.