National Academies Press: OpenBook

Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging (2021)

Chapter: Chapter 3 - Findings and Application

« Previous: Chapter 2 - Research Approach
Page 155
Suggested Citation:"Chapter 3 - Findings and Application." National Academies of Sciences, Engineering, and Medicine. 2021. Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging. Washington, DC: The National Academies Press. doi: 10.17226/26089.
×
Page 155
Page 156
Suggested Citation:"Chapter 3 - Findings and Application." National Academies of Sciences, Engineering, and Medicine. 2021. Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging. Washington, DC: The National Academies Press. doi: 10.17226/26089.
×
Page 156
Page 157
Suggested Citation:"Chapter 3 - Findings and Application." National Academies of Sciences, Engineering, and Medicine. 2021. Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging. Washington, DC: The National Academies Press. doi: 10.17226/26089.
×
Page 157
Page 158
Suggested Citation:"Chapter 3 - Findings and Application." National Academies of Sciences, Engineering, and Medicine. 2021. Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging. Washington, DC: The National Academies Press. doi: 10.17226/26089.
×
Page 158
Page 159
Suggested Citation:"Chapter 3 - Findings and Application." National Academies of Sciences, Engineering, and Medicine. 2021. Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging. Washington, DC: The National Academies Press. doi: 10.17226/26089.
×
Page 159
Page 160
Suggested Citation:"Chapter 3 - Findings and Application." National Academies of Sciences, Engineering, and Medicine. 2021. Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging. Washington, DC: The National Academies Press. doi: 10.17226/26089.
×
Page 160
Page 161
Suggested Citation:"Chapter 3 - Findings and Application." National Academies of Sciences, Engineering, and Medicine. 2021. Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging. Washington, DC: The National Academies Press. doi: 10.17226/26089.
×
Page 161
Page 162
Suggested Citation:"Chapter 3 - Findings and Application." National Academies of Sciences, Engineering, and Medicine. 2021. Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging. Washington, DC: The National Academies Press. doi: 10.17226/26089.
×
Page 162
Page 163
Suggested Citation:"Chapter 3 - Findings and Application." National Academies of Sciences, Engineering, and Medicine. 2021. Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging. Washington, DC: The National Academies Press. doi: 10.17226/26089.
×
Page 163
Page 164
Suggested Citation:"Chapter 3 - Findings and Application." National Academies of Sciences, Engineering, and Medicine. 2021. Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging. Washington, DC: The National Academies Press. doi: 10.17226/26089.
×
Page 164
Page 165
Suggested Citation:"Chapter 3 - Findings and Application." National Academies of Sciences, Engineering, and Medicine. 2021. Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging. Washington, DC: The National Academies Press. doi: 10.17226/26089.
×
Page 165

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.

155   Findings Short-Term Conditioning The primary concerns with AASHTO T 240 are: (1) binders of different consistencies could be treated differently because the film thickness and its renewal vary with the consistency of the binder, and (2) some heavily modified binders crawl out of the container during conditioning. The evaluation of short-term conditioning procedures identified two options for improving short-term laboratory conditioning of asphalt binders: (1) adopt the container and mixing screw modifications to AASHTO T 240 that were implemented in the UK Ageing Profile Test (Hill et al. 2009), or (2) adopt thin-film conditioning like the USAT (Farrar et al. 2014) but using thicker film to provide more binder for current specification tests. During outreach to industry, a third option was identified: modify AASHTO T 240 to include a heating step where binder in the containers is heated to the conditioning temperature before hand rotating the container to coat it just before loading the container in the RTFO. A short-term conditioning selection experiment was designed to compare conditioning from the candidate approaches to AASHTO T 240 and oven-conditioned loose mix. The loose mix conditioning procedures that were used were those recommended in NCHRP Project 09-52, which were based on a field evaluation of numerous factors affecting the short-term aging characteristics of asphalt mixtures (Newcomb et al. 2015). Eight binders with a wide range of consistency were used in the evaluation, and the evaluation was conducted at temperatures to simulate HMA and WMA production. The major findings from the short-term conditioning selection experiment related to the AASHTO T 240 binder consistency concern were: 1. The ratio of the aging index from the short-term binder conditioning procedures to the aging index from binder recovered from short-term loose mix conditioning was not a function of the viscosity of the binder at the conditioning temperature for any of the laboratory conditioning procedures, including AASHTO T 240. 2. For simulated HMA aging, there was not a significant difference between the aging index from any of the laboratory short-term binder conditioning procedures and the aging index for binder recovered from short-term conditioned loose mix. 3. For simulated WMA aging, there was not a significant difference between the aging index from the mixing screw procedures and the aging index for binder recovered from short-term conditioned loose mix. However, aging indices for AASHTO T 240, AASHTO T 240 with preheated containers, and the static 0.8 mm film were significantly lower than the aging index for binder recovered from short-term conditioned loose mix. 4. When the binder viscosity at the mixing temperature exceeded about 0.55 Pa•s, the mixing screw procedures increased the binder aging relative to AASHTO T 240. Apparently, the mixing screws improve the exposure of viscous binders to air during laboratory conditioning. C H A P T E R   3 Findings and Application

156 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging 5. Heating the binder and the containers before AASHTO T 240 conditioning increased the binder aging index about 6 percent. 6. There is a rough correlation of mass change measurements between the various short-term laboratory conditioning procedures. The major findings from the short-term conditioning selection experiment related to the AASHTO T 240 binder leakage concern were: 1. Binder leakage did not occur during AASHTO T 240 conditioning for any of the binders conditioned at 163°C. Binder leakage did occur during AASHTO T 240 conditioning for the stiffest binder conditioned at 135°C. 2. Binder leakage occurred during conditioning with the mixing-screw procedures for several of the binders at 163°C and 135°C. This was likely the result of the increased binder depth in the container due to the volume occupied by the mixing screws. To better document the extent of the binder leakage issue, the Maine Department of Trans- portation sent a survey on behalf of NCHRP Project 09-61 to the members of the AASHTO Committee on Pavements and Materials. The online survey was directed to technicians performing binder testing and is reproduced in Figure 102. Thirty-three agencies responded to the survey. Table 83 identifies the agencies that responded and summarizes the responses to the questions aimed at determining the extent of the binder flow problem. Agencies estimate between 0 and 50 percent of the binders that they grade exhibit binder flow. The agencies that responded estimate that they conduct AASHTO T 240 conditioning on 10,693 samples per year and approximately 405 of these samples, or approximately 3.8 percent, exhibit binder flow. Only four agencies estimate that binder flow occurs in more than 5 percent of the samples tested: Utah at 50 percent, Oregon at 30 percent, Maine at 15 percent, and Georgia at 10 percent. The commentary provided by Utah indicates that the binder flow occurs during sample unloading, and since the mass change containers are removed first, flow rarely affects the mass change measurement. The other agencies with high percentages of samples exhibiting binder flow identified the only grades that were susceptible to the flow. An unexpected finding is the number of agencies identifying neat binders as being suscep- tible to binder flow. Seven agencies listed only neat binders as being susceptible to binder flow. Fifteen agencies listed only modified binders as being susceptible to binder flow. Of these, three agencies identified GTR-modified binder, and three agencies identified latex-modified binder as being susceptible to binder flow. Two agencies listed both neat and modified binders as being susceptible to binder flow. The survey included three questions aimed at confirming whether the laboratory routinely checks the levelness of the carriage and rotates the container to generate a film before cooling. The responses to these questions are summarized in Table 84. The higher percentages of flow reported do not appear to be associated with deviation from the test procedure for these items. The last survey question asked the respondents to list other issues with AASHTO T 240 that they thought need to be addressed. The responses are summarized in Table 85. Twelve agencies listed additional areas of concern. The most common, from five agencies, were associated with improvement to temperature measurement and verification. Four agencies listed improving unloading and binder recovery, and two agencies listed tilting the oven with modified binders. Issues identified by only one agency were: (1) stiffer binders tend not to roll, (2) more automa- tion is needed, and (3) correction for laboratory elevation. The survey documented that the AASHTO T 240 binder flow issue affects about 4 percent of the binders tested, and it occurs when conditioning both modified and unmodified binders. Binders modified with GTR and latex appear to be particularly susceptible to the binder flow issue.

Figure 102. AASHTO T 240 online survey.

158 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging Agency Tests per Year Binder Flow Susceptible Grades Percent Number 1 600 30 180 64-28, 70-22, 70-28. Both 70 grades are modified. 2 250 1 3 Probably the less viscous virgin binder. 3 100 0 0 We only see flow out of container if it is a proficiency sample at a grade of 58-22. Notmodified. 4 155 15 23 PG 64E-28. 5 150 1 2 PG 52-28. 6 250 5 13 58-28, 70-28, 76-28, yes. 7 300 5 15 PG 70-22, modified grade. 8 200 0 0 9 435 0 0 N/A 10 200 0 0 None. 11 350 1 4 58-22, 64-22, 67-22. No modified binders exhibit the flowing. 12 2800 1 28 Material with latex polymer. 13 50 5 3 Reported as PG 67-22, unmodified from perhaps one specific supplier, and PG 70-22 latex-modified (PG 67-22 with addition of plant-blended latex). 14 100 5 5 PG 76-22, yes. 15 200 0 0 N/A 16 125 5 6 PG 64-22. They are not required by us to be modified. 17 65 3 2 64S-22, unmodified. 18 150 0 0 None. 19 24 0 0 N/A 20 300 1 3 64-28, 76-28, both modified. 21 125 1 1 PG 70-28 (PG 64-28 modified with 2–3% SBR latex added at the HMA production facility). 22 700 1 7 Softer binder grades. PG 52 and 58 23 250 1 3 64-22V with GTR. 24 330 1 3 PG 64V-28 (PG 76-28). Modified. 25 100 50 50 26 350 0 0 27 360 1 4 28 700 5 35 29 250 0 0 30 50 5 3 31 104 10 10 32 350 0 0 33 220 1 180 PG 64-34 and PG 70-28. They are modified. The flow occurs when the carriage is stopped to unload a bottle. Generally, it stays on the lip of the bottle. Fifty percent on lip of at least one bottle after unloading process begins. Mass change bottles are first out and rarely have any outflow. We've only seen it on highly modified 70s or 76s. Even then it's only happened a few times through the years. PG 70-28, but very rarely does it happen. 76-22. Everything we use is modified and no issues with binder flowing out of the container. 76-22 (64V) – GTR-modified binder only. 76-22 and GTR-modify once. Our PMA binder is PG 76-22 (PG 64E-22). We never have a problem with it. On very rare occasions maybe a drop or two would dribble out, and that is not even once a year. We also have, on a few occasions, ran the HPTO asphalt, PG 82-22, or 76-28, and that is well behaved. Over the past 20 years, we have run 1500-2000 PG 76 samples, and binder flowing out of the bottle is a non-issue. PG 88-22, PG 76-22 (SBS). Total 10,638 405 Table 83. Summary of survey results addressing the extent of the binder flow problem.

Findings and Application 159   Agency Percent with Binder Flow Levelness Check Precoat Container Cool Horizontally 1 30 Every 6 months. Yes Yes 2 1 We calibrate it two times per year. Yes Yes 3 0 Every 6 months. Yes Yes 4 15 Annually or as needed. Yes Yes 5 1 Every 6 months. Yes Yes 6 5 Every 12 months. Yes Yes 7 5 Routinely every 6 months; more often if issues noted. Yes Yes 8 0 Once per year. Yes Yes 9 0 Every 6 months. Yes Yes 10 0 Every 12 months or if oven is moved for maintenance. Yes Yes 11 1 Before each test. Yes Yes 12 1 6 months. Yes Yes 13 5 No set frequency - should be every 6 months with bearing check. Yes Yes 14 5 Two times per year. Yes Yes 15 0 Every 12 months. Yes Yes 16 5 Every 6 months as a minimum. Yes Yes 17 3 One to two times per year. Yes Yes 18 0 Every 12 months. Yes Yes 19 0 Every 2 to 3 months. Yes Yes 20 1 Every 6 Months. Yes No 21 1 Six-month intervals. Yes Yes 22 1 Monthly. Yes Yes 23 1 1 time per year Yes Yes 24 1 6-month intervals. We use 3 RTFOTovens. Yes Yes 25 50 Six months. With modified binders, I use my free hand to tip the bottle down beyond horizontal to let it flow towards the opening, then rotate, attempting to coat as evenly as possible. Yes 26 0 Yearly. Yes Yes 27 1 Every 6 months. Yes Yes 28 5 Every 6 months. Yes Yes 29 0 Every 6 months or as needed, for example, if an oven is moved. Yes Yes 30 5 Every 6 months. Yes Yes 31 10 Yearly. Yes Yes 32 0 Annually. Yes Yes 33 1 Bi-annually. Yes Yes Table 84. Summary of the survey results addressing procedural factors affecting binder flow.

160 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging Agency Percent with Binder Flow Issues with AASHTO T 240 1 30 2 1 Not much to complain. 3 0 We have a thermometer (LIG) hanging in the oven per procedure - is this redundant, because we do not see a temperature drift during the bi-annual temperature verification. Second point - has testing been done to quantify the difference in results if the airflow is off during extraction of the containers from the oven? For the 2 minutes it takes to pull the jars from the oven, it seems like the airflow should not have to be on. That is more of a question. 4 15 N/A 5 1 Rolling issues with highly modified binders. They don't roll. Stiffer materials are more difficult to remove from the containers. 6 5 Another way to calibrate and verify temperature. 7 5 8 0 9 0 None. 10 0 None. 11 1 N/A 12 1 13 5 14 5 Permitting use of K-type thermocouples. 15 0 N/A 16 5 With higher PG grades (PG 76 and 82) there are some samples that we cannot get any binder out of the bottle, or at the least is very tough to get out. 17 3 Unloading the RTFO in <5 minutes is challenging. Two technicians are needed to complete in <5 minutes. Removing >90% of the material from the bottle is also a challenge, especially with modified binders. 18 0 I cannot recall any issues I have encountered with AASHTO T 240. 19 0 No Issues. 20 1 21 1 An easier way to recover aged binder from the bottles. More automation to the equipment (rotating carriage automatically, instead of technician starting the rotation manually). 22 1 None at this time. 23 1 None. 24 1 25 50 We use ductility end pieces under the front of the oven to prevent this as much as possible while keeping the level within tolerance of T 240. 26 0 None. 27 1 N/A 28 5 No other issues that we are concerned about. 29 0 Correction for laboratory elevation - see NCHRP Project 20-07/Task 400. 30 5 Always known to tilt oven back to prevent this from occurring with GTR-modified binders, the rubber swells and crawls out bottle openings. 31 10 Checking the temperature for calibration is always a problem. We need a better way of hanging the thermometer in the oven and also be able to see and read the result. 32 0 Please make the oven temperature sensor and the thermometer in the same location. The current design is just bad engineering. 33 1 Table 85. Summary of other AASHTO T 240 issues.

Findings and Application 161   Long-Term Conditioning The primary improvement to AASHTO R 28 identified by the evaluation of long-term conditioning procedures was to increase the amount of aging simulated by the PAV to better match the properties of binders near the surface of the pavement after approximately 10 years in service. Feasibility experiments found novel approaches of using acoustic, sonic, and ultra- sonic mixing to accelerate oxidation reactions were not successful; therefore, further devel- opment work investigated changing PAV operating parameters to accelerate aging. The PAV operating parameters experiment evaluated the change in rheological and chemical properties for RTFOT-conditioned binder that was further conditioned using a range of PAV conditioning temperatures, binder masses, and conditioning times. Response surfaces for the change in the crossover frequency, change in the carbonyl absorbance, and change in sulfoxide absorbance were developed for four binders from the ARC validation sections in Arizona and four binders from the ARC validation sections in Minnesota. The response surfaces were used to compare rheological and chemical properties of PAV-conditioned residue to binder extracted from field cores that were aged for 4 to 5 years and 9 to 11 years. This comparison provided the range of PAV operating parameters needed to simulate near-surface aging after approximately 10 years of service in warm and cool climates. The major findings from the PAV operating parameters experiment were: 1. Varying the conditioning temperature, mass of binder, and conditioning time in the PAV produces binder residue with a wide range of rheological and chemical properties. The crossover frequency, which is a measure of the hardness of the conditioned binder, had a range of 7 to 12 orders of magnitude for the binders tested. 2. The rheological and chemical changes produced using various PAV operating conditions were rational and produced smooth trends for the binders tested. The crossover frequency decreased, the rheological index increased, and the carbonyl absorbance increased with increasing severity of conditioning. The sulfoxide absorbance was relatively insensitive to changes in PAV operating conditions. Data from all PAV conditions collapsed onto smooth curves. 3. PAV conditioning 50.0 g of binder for 40 hours at 100°C was approximately equivalent to PAV conditioning of 12.5 g of binder for 20 hours at 100°C. 4. Rheological and chemical properties of field-aged binders generally followed the same trends as PAV-conditioned binders with increasing exposure; however, the data were more variable. The crossover frequency decreased, the rheological index increased, and carbonyl absorbance increased with increasing time in service. The crossover frequency increased, the rheological index decreased, and the carbonyl absorbance decreased with depth in the pavement. Irrational and erratic trends were observed for the field-aged properties for the terpolymer-modified binder. FTIR analysis confirmed that the binder extraction and recovery process was likely not recovering all polymer for this binder. 5. Response surfaces fitted to the change in the crossover frequency and the change in the carbonyl absorbance were most sensitive to changes in conditioning temperature for all binders. Sensitivity to conditioning time and binder mass were binder dependent. 6. The response surfaces were used to determine PAV operating parameters for the Arizona and Minnesota sites that closely matched near-surface aging after 9 to 11 years in service. The PAV exposure required to approximate the field-aged chemistry was generally greater than that required to match the field-aged rheology. Based on a combined chemical and rheological criterion, PAV conditioning for 20 hours using a binder mass of 12.5 g and a temperature of 100°C reasonably reproduces the aging after about 10 years in service at a depth of 0.75 in at both sites. To reproduce the aging that occurs at a depth of 0.25 in requires higher temperatures for the same conditioning time and binder mass—108°C for the Arizona site and 104°C for the Minnesota site.

162 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging Based on the findings of the PAV operating parameters experiment, the long-term condition- ing calibration experiment was conducted. The objective of this experiment was to determine appropriate conditioning temperatures for 12.5 g, 20-hour, 2.1 MPa PAV conditioning for various climates. The basic approach consisted of comparing rheological and chemical proper- ties of laboratory-conditioned binders to those for binder recovered from cores from in-service pavements. Twenty-six pavements for the LTPP program that had original binder and cores taken after 8 to 16 years in service stored in the LTPP MRL were included in the experiment. The pavements were in 23 states and two provinces with mean annual air temperatures ranging from 1.7°C to 20°C. Recovered binder properties were measured for three 0.5 in thick slices: (1) 0 to 0.5 in, (2) 0.5 in to 1.0 in, and (3) 1.5 in to 2.0 in. To determine appropriate PAV conditioning temperatures for each pavement, the original binder samples were first RTFOT conditioned. The RTFOT-conditioned residue was then further conditioned in the PAV using 12.5 g in PAV pans, 20 hours conditioning time, and 2.1 MPa pressure. Three temperatures were used: 85°C, 100°C, and 115°C. Rheological and chemical properties of the original binder, RTFOT residue, and the residue from the three PAV conditioning temperatures were measured. Equivalent 12.5 g, 20-hr PAV temperatures were determined by equating the properties of the PAV residue to the properties of the recovered binders. The major findings of the long-term calibration experiment were: 1. Residue from 12.5 g, 20-hr PAV conditioning reasonably reproduced the rheological properties and carbonyl absorbance of binder recovered from in-service pavements at conditioning temperatures between 85°C and 115°C. The temperature where the PAV- conditioned residue most closely approximated the recovered binder properties was called the equivalent PAV conditioning temperature. 2. 12.5 g, 20-hr PAV conditioning did not reasonably reproduce sulfoxide levels for several of the binders, resulting in extremely high or extremely low equivalent PAV conditioning temperatures based on sulfoxide absorbance. 3. There was a fair correlation between the equivalent PAV conditioning temperatures based on changes in carbonyl absorbance and the equivalent PAV conditioning temperature based on changes in rheology. The equivalent PAV conditioning temperatures based on carbonyl absorbance were consistently approximately 10 degrees higher than those based on rheology. 4. Standard PAV conditioning produced residue with chemical and rheological properties that were similar to 12.5 g, 20-hr PAV conditioning at 85°C. 5. A rational statistical model was developed to estimate equivalent PAV conditioning tempera- tures based on rheology. The model showed the equivalent PAV conditioning temperature is sensitive to (1) average pavement temperature, (2) in-place air void content, (3) sensitivity of the binder to changes in aging temperature, (4) age of the pavement, and (5) depth in the pavement. 6. The statistical model was used to calculate equivalent 12.5 g, 20-hr PAV conditioning temperatures for specification testing as a function of climate. These temperatures are based on rheological properties from the top 1 in of the pavement and vary from 85°C to 115°C. Applications Short-Term Conditioning The findings of the evaluation, laboratory testing and analysis, and survey conducted in NCHRP Project 09-61 support the continued use of AASHTO T 240 for short-term condition- ing of asphalt binders to simulate aging during HMA production. Although the film thickness and its renewal vary with the consistency of the binder, residue from AASHTO T 240 for binders with a wide range of consistencies has rheological properties that are similar to binder recovered from loose mix conditioned following the procedures developed in NCHRP Project 09-52 to

Findings and Application 163   simulate production aging. The survey of highway agency technicians revealed users are generally satisfied with AASHTO T 240, but they would like to see additional improvement, primarily in the areas of temperature measurement, calibration, and binder recovery. The survey docu- mented that binder leakage occurs in approximately 4 percent of tests, and binder leakage occurs with both modified and neat binders. Some agencies report no binder leakage while neighboring agencies report high instances of binder leakage. This suggests the need for specific equipment and technique evaluation rather than wholesale changes to AASHTO T 240. If agencies want to simulate WMA production at lower temperatures, AASHTO T 240 could be modified to use a temperature greater than 135°C or increase the conditioning time. These modifications are preferred over adding mixing screws and maintaining the temperature at 135°C and the conditioning time at 85 minutes because: (1) no new equipment is required, (2) binder recovery from the mixing screws is difficult, and (3) mass change measurements using heated containers as required when using mixing screws are highly variable. Long-Term Conditioning The findings of the evaluation and the laboratory testing and analysis conducted in NCHRP Project 09-61 show PAV conditioning is capable of reasonably simulating the properties of field-aged binders. The time in service simulated by PAV conditioning depends on the operating parameters used in the PAV. Aging is accelerated by increasing the temperature, increasing the conditioning time, decreasing the film thickness, and increasing the pressure. Temperature has the greatest effect. The effect of conditioning time and film thickness are similar. Based on models of the pressure dependency of the asphalt oxidation reactions, doubling the pressure from 2.1 MPa to 4.2 MPa would only change the reaction rate by 8 percent and 15 percent. To maintain the current AASHTO R 28 conditioning time, PAV conditioning was calibrated to reproduce the aging that occurs within the top 1 in of the pavement using 12.5 g binder mass, 20-hr conditioning time, and 2.1 MPa pressure. To facilitate specification testing, the PAV conditioning temperatures were expressed as a function of the average of the 98 percent reliability high and low pavement temperatures at the surface of the pavement from LTPPBind without adjustment for traffic. They range from 85°C in areas of Alaska and northern Canada to 115°C in areas of Arizona, California, Nevada, and Texas. Condition- ing temperatures between 90°C and 110°C cover most of the continental United States. The complete listing of 12.5 g, 20-hr, 2.1 MPa PAV conditioning temperatures was provided in Table 77. While encouraging, there are implementation issues associated with film formation and specification criteria that need to be addressed as discussed below. The nominal film thickness for 12.5 g of binder in a standard 140 mm diameter PAV pan is 0.8 mm. To form and maintain a uniform film, the pan must be flat and level during condi- tioning. The experimental work in NCHRP Project 09-61 found that commercially available 1.0 mm thick stainless steel pans that were cleaned with solvents after testing remained flat and did not warp. These pans are thicker than required in AASHTO R 28. Also, the level- ness tolerance specified in AASHTO R 28 of film thickness of 0.5 mm in a 140 mm diameter PAV pan is not acceptable. The text of AASHTO R 28 specified 0.5 mm film thickness, while Figure 1 of AASHTO R 28 specifies 0.05 mm. Pans leveled with a precision machinist level, having a sensitivity of 0.00042 mm/mm (0.058 mm in 140 mm), resulted in full coverage of the pan. Achieving this degree of levelness was extremely difficult using current PAV equip- ment. Improvements to the PAV equipment are needed to allow rapid leveling to the above tolerance if 12.5 g PAV conditioning is to be adopted. A second issue with film formation was identified during the sensitivity experiment: heavily modified, RTFOT-conditioned binders, when PAV conditioned at the temperatures in Table 77, may not form a film that fully covers the pan. For the sensitivity experiment, a proper film

164 Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging for these binders was formed by heating the pan leveled to 0.00042 mm/mm under industrial nitrogen for 30 minutes at 135°C before PAV conditioning. If 12.5 g PAV conditioning is to be adopted, PAV equipment should be modified and automated to allow this coating step to be performed automatically. The final issue that must be addressed is developing appropriate grading criteria for the greater amount of aging simulated by 12.5 g, 20-hr, 2.1 MPa PAV conditioning. Based on current criteria, the low-temperature continuous grade of the 10 binders tested in the sensitivity experiment increased an average of 5.9°C. The intermediate-temperature continuous grade increased an average of 4.2°C. Thus, the adoption of this long-term conditioning without modification of the performance grading criteria will change the low-temperature grade of binders approximately one grade level. Field performance data are needed to establish appropriate grading criteria. One possible application of 12.5 g, 20-hr, 2.1 MPa PAV is in conjunction with the adoption of DTc after 40-hr PAV conditioning as a specification criterion (Asphalt Institute 2019). The testing in this project showed 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 DTc 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 DTc determination. This is sufficient residue for testing BBR beams at two different temperatures. The issues discussed above concerning film formation need to be addressed before this approach can be considered for use in practice. Combined Short- and Long-Term Conditioning The laboratory testing and analysis also showed residue from static 0.8 mm film conditioning for binders with a wide range of consistencies has rheological properties that are similar to binder recovered from loose mix conditioned following the procedures developed in NCHRP Project 09-52. Using the same film thickness and equipment for short- and long-term condition- ing offers the potential to simplify laboratory conditioning. A very reasonable approach is to short-term condition the binder in a low-pressure oven in the standard 140 mm diameter pans at 163°C. Conditioning the thin film under a low pressure above atmospheric would eliminate the laboratory elevation effect that is known to be an issue with AASHTO T 240 (Advanced Asphalt Technologies LLC 2018). Upon completion of the short-term conditioning, one pan would be removed, cooled, weighed for mass change determination, and the residue tested for high pavement temperature rheological properties. The remaining pans would be trans- ferred to the PAV for long-term conditioning under 2.1 MPa air pressure for 20 hours at the appropriate temperature based on climate temperature. This approach has some advantages: (1) it removes the viscosity-dependent film renewal associated with AASHTO T 240; (2) it removes the laboratory elevation effect associated with AASHTO T 240; (3) it eliminates binder transfer loss between short- and long-term conditioning; and (4) the higher temperature for the short-term conditioning will improve the uniformity of the thin film for the subsequent long-term conditioning. To avoid needing equipment that controls both pressure and vacuum, a standard conditioning pressure somewhat above the maximum atmospheric pressure should be used. Atmospheric pressure varies with elevation and weather. The elevation effect, however, is much larger. Standard atmospheric pressure at sea level is 101.33 kPa (14.70 psi). Based on the relationship for the variation in atmospheric pressure with elevation (U.S. Department of Commerce 1963), the atmospheric pressure at 7,000 ft elevation, the elevation of the highest labs participating

Findings and Application 165   in the AASHTO re:source proficiency sample testing, is 78.18 kPa (11.34 psi). The historical range of maximum and minimum atmospheric pressures corrected to sea level recorded in the United States is from 93.40 kPa (13.56 psi) to 106.33 kPa (15.42 psi) (Weather Underground n.d.). Therefore, the short-term conditioning pressures should be set to 110.00 kPa (15.95 psi) absolute, which is a small gauge pressure of 8.67 kPa (1.25 psi) at sea level and 16.6 kPa (4.61 psi) at an elevation of 7,000 ft. Based on work completed in NCHRP Project 20-07/Task 400, a significant difference in AASHTO T 240 residue occurs when the elevation difference between labs exceeds about 1,000 ft, which corresponds to a pressure difference of about 3.50 kPa (0.51 psi). Therefore, pressure control should be more accurate than this value. Closed-loop electronic air pressure regulators and digital pressure gauges are available with an accuracy of 0.25 percent of full scale. A 210 kPa (30 psi) device with an accuracy of 0.25 percent of full scale is accurate to 0.52 kPa (0.075 psi). Based on this analysis, it is feasible to control the pressure in a low-pressure oven to an absolute pressure of 110.00 kPa (15.95 psi) ± 1.00 kPa (0.14 psi). The primary unanswered question for combined short- and long-term conditioning is how much will the low pressure during short-term conditioning suppress volatile loss? The changes to the binder that occur during AASHTO T 240 conditioning and the construction process include oxidation and volatile loss. The short-term selection experiment documented lower mass loss for the static 0.8 mm film compared to AASHTO T 240. Applying pressure to the thin film during short-term conditioning may suppress volatile loss. Additional experimental work is needed to determine if volatile loss suppression is significant.

Next: Chapter 4 - Conclusions and Recommendations »
Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Asphalt binders experience aging that occurs in two distinct stages under quite different conditions: (1) short-term during construction (plant mixing, storage, placement, and compaction) and (2) long-term during the service life of the pavement.

The TRB National Cooperative Highway Research Program's NCHRP Research Report 967: Asphalt Binder Aging Methods to Accurately Reflect Mixture Aging documents research conducted to improve laboratory binder conditioning methods to accurately simulate the short-term and long-term aging of asphalt binders, and to calibrate the improved procedures to the aging that occur during mixture production, transport, and placement as well as during the service life of the pavement structure.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!