Short-Term Laboratory Conditioning of Asphalt Mixtures (2015)

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.

C-1 Preliminary Laboratory Experiment A preliminary laboratory experiment was conducted at the beginning of this project to evaluate the effects of binder absorption, aggregate size, binder source, binder grade, and conditioning protocol on the stiffness and strength of asphalt mixtures subjected to various laboratory short-term oven aging (conditioning) protocols. Table C-1 summarizes the factors used in this experiment, and all factors had two levels. Three replicate test specimens for each mixture type were prepared using the Superpave gyratory compactor (SGC) to a target air void (AV) level of 7 ± 0.5 percent. The specimens were subjected to the resilient modulus (MR) and indirect ten- sile (IDT) tests. MR and IDT strength were the test parameters selected as indicators of the mixture stiffness and strength. A statistical analysis was performed on test results to identify the influence of the different factors on the test parameters. Asphalt absorption by the aggregate can have a significant effect on mixture stiffness and strength. High-absorptive aggregates have the potential of reducing the film thickness of asphalt between aggregate particles and make it more suscep- tible to certain kinds of distress. As noted in Table C-1, two types of aggregates were investigated in the laboratory study, a high-absorption and a low-absorption aggregate. The high- absorption aggregate was from Georgetown, Texas, with a reported level of water absorption of around 3.5 percent; the low-absorption aggregate was from Brownwood, Texas, with a water absorption value of approximately 0.7 percent. The nominal maximum aggregate size (NMAS) is also expected to play a role in mixture stiffness and strength as the size of the aggregate usually relates to the size of its pores. The pore size can affect both the quantity of asphalt absorbed and cause preferential absorption for certain asphalt fractions. In the laboratory study, two different NMAS were selected: 9.5 mm and 19 mm. In addition, binder percentages selected from mix designs were different for mixtures with different aggregates and NMAS. Thus, the effect of NMAS on mixture stiffness evaluated in this study might be attributed to different NMAS and binder contents. Two common binders (Binder A and Binder V) from two different crude oil sources and produced in different refiner- ies in Texas were used in the laboratory study. Binder V was produced from a South American petroleum and Binder A was from a West Texas crude. A previous study showed these binders had different hardening behaviors (Glover 2010). However, since all loose mixtures were short-term condi- tioned in the oven at 275°F (135°C) for 2 or 4 hours prior to compaction, the unaged stiffnesses of these two binders and the difference in hardening behavior with time could not be captured in this laboratory experiment. Binders with dif- ferent performance grades (PGs) were included to assess the effect of the polymer modification on the binder oxidation and hardening process. The continuous PG binder grading was not available with the PG 70-22 and PG 64-22 binders used in the study. However, it should be noted that the differ- ence in properties between these two binders may be less than that indicated by their grades. The laboratory short-term oven aging protocols for the loose mix followed the recommendations of NCHRP Project 9-49. In that study, a comparison of MR stiffness among laboratory- mixed, laboratory-compacted mixtures; plant-mixed, field- compacted mixtures; and field cores recommended 2 hours at 275°F (135°C) for HMA specimens. Additionally, 4 hours at 275°F (135°C) were found to be an acceptable protocol for some mixture types and, thus, was included in this labora- tory experiment, especially since that is the standard loose mix conditioning practice followed by some state DOTs. Ninety-six SGC specimens for 32 different factor/level combinations were fabricated and subjected to MR and IDT tests, as shown in Table C-2. All specimens were tested approx- imately one week after fabrication. Regression analysis was performed to evaluate the effects of different factors on mix- ture stiffness and strength. The response variables used in the study were MR and IDT strength. All factors listed in Table C-1 in addition to the two-way interactions among those factors were included. A P P E N D I X C

C-2 Factors Level Values Aggregate Type (Water Absorption) Abs-1: 0.7% Abs-2: 3.5% NMAS NMAS-1: 9.5 mm NMAS-2: 19 mm Binder Source Binder A Binder V Binder Grade PG-1: PG 64-22 PG-2: PG 70-22 Conditioning Protocol STOA-1: 2 h @ 275°F (135°C) STOA-2: 4 h @ 275°F (135°C) Table C-1. Factors used in the laboratory experiment. Mixture ID Binder Absorption NMAS, mm Binder Source Binder Grade Conditioning Protocol Low-9.5-A-70-2h Low-Brownwood 9.5 Binder A PG 70-22 2 h @ 275°F (135°C) Low-9.5-A-64-2h Low-Brownwood 9.5 Binder A PG 64-22 2 h @ 275°F (135°C) Low-9.5-A-70-4h Low-Brownwood 9.5 Binder A PG 70-22 4 h @ 275°F (135°C) Low-9.5-A-64-4h Low-Brownwood 9.5 Binder A PG 64-22 4 h @ 275°F (135°C) Low-19-A-70-2h Low-Brownwood 19.0 Binder A PG 70-22 2 h @ 275°F (135°C) Low-19-A-64-2h Low-Brownwood 19.0 Binder A PG 64-22 2 h @ 275°F (135°C) Low-19-A-70-4h Low-Brownwood 19.0 Binder A PG 70-22 4 h @ 275°F (135°C) Low-19-A-64-4h Low-Brownwood 19.0 Binder A PG 64-22 4 h @ 275°F (135°C) Low-9.5-V-70-2h Low-Brownwood 9.5 Binder V PG 70-22 2 h @ 275°F (135°C) Low-9.5-V-64-2h Low-Brownwood 9.5 Binder V PG 64-22 2 h @ 275°F (135°C) Low-9.5-V-70-4h Low-Brownwood 9.5 Binder V PG 70-22 4 h @ 275°F (135°C) Low-9.5-V-64-4h Low-Brownwood 9.5 Binder V PG 64-22 4 h @ 275°F (135°C) Low-19-V-70-2h Low-Brownwood 19.0 Binder V PG 70-22 2 h @ 275°F (135°C) Low-19-V-64-2h Low-Brownwood 19.0 Binder V PG 64-22 2 h @ 275°F (135°C) Low-19-V-70-4h Low-Brownwood 19.0 Binder V PG 70-22 4 h @ 275°F (135°C) Low-19-V-64-4h Low-Brownwood 19.0 Binder V PG 64-22 4 h @ 275°F (135°C) High-9.5-A-70-2h High-Georgetown 9.5 Binder A PG 70-22 2 h @ 275°F (135°C) High-9.5-A-64-2h High-Georgetown 9.5 Binder A PG 64-22 2 h @ 275°F (135°C) High-9.5-A-70-4h High-Georgetown 9.5 Binder A PG 70-22 4 h @ 275°F (135°C) High-9.5-A-64-4h High-Georgetown 9.5 Binder A PG 64-22 4 h @ 275°F (135°C) High-19-A-70-2h High-Georgetown 19.0 Binder A PG 70-22 2 h @ 275°F (135°C) High-19-A-64-2h High-Georgetown 19.0 Binder A PG 64-22 2 h @ 275°F (135°C) High-19-A-70-4h High-Georgetown 19.0 Binder A PG 70-22 4 h @ 275°F (135°C) High-19-A-64-4h High-Georgetown 19.0 Binder A PG 64-22 4 h @ 275°F (135°C) High-9.5-V-70-2h High-Georgetown 9.5 Binder V PG 70-22 2 h @ 275°F (135°C) High-9.5-V-64-2h High-Georgetown 9.5 Binder V PG 64-22 2 h @ 275°F (135°C) High-9.5-V-70-4h High-Georgetown 9.5 Binder V PG 70-22 4 h @ 275°F (135°C) High-9.5-V-64-4h High-Georgetown 9.5 Binder V PG 64-22 4 h @ 275°F (135°C) High-19-V-70-2h High-Georgetown 19.0 Binder V PG 70-22 2 h @ 275°F (135°C) High-19-V-64-2h High-Georgetown 19.0 Binder V PG 64-22 2 h @ 275°F (135°C) High-19-V-70-4h High-Georgetown 19.0 Binder V PG 70-22 4 h @ 275°F (135°C) High-19-V-64-4h High-Georgetown 19.0 Binder V PG 64-22 4 h @ 275°F (135°C) Table C-2. Factor/level combinations used in the laboratory experiment.

C-3 The statistical analysis results by JMP statistical package (SAS product) for MR as the response variable are presented in Figure C-1. As indicated by the “Effect Tests” table, among the 15 factor/level combinations considered, the main effects of binder absorption, NMAS, binder source, binder grade, and conditioning protocol on MR were statistically significant at a = 0.05. In addition, the effect on MR from the two-way inter- actions between NMAS and binder source and between NMAS and binder grade were significant. Comparisons in terms of least squares means for significant factors of interests are pre- sented in Table C-3. It can be observed from the table that, in general, mixtures with higher absorptive aggregate, larger NMAS, binder A, and longer short-term conditioning time may result in higher MR values. In the following five figures that present the difference in MR results of asphalt mixtures due to different levels of binder absorption, NMAS, binder source, binder grade, and condi- tioning protocol, each bar represents the average value of three replicate specimens, and the error bars represent ± one stan- dard deviation from the average value. Figure C-1. JMP output of fitting model for resilient modulus.

C-4 As illustrated in Figure C-2, for all mixtures, the stiffness of mixtures with higher absorptive aggregates was higher than or equivalent to that of mixtures with lower absorptive aggre gates. Since two types of aggregates had different levels of binder absorption, mixtures with those two aggregates were designed differently and, thus, with different optimum binder contents. A study at Virginia Polytechnic Institute and State University evaluated the effect of binder film thickness on mixture stiffness (Wang 2007). In the study, samples of two hemispheroid particles bonded by binder with different film thicknesses were fabricated and then tested under a direct compression condition. The vertical displacement and the resistance force of samples with different binder film thick- nesses were recorded during the test. Test results indicated that the resistance force of the sample with thinner binder film thickness was higher than that of the sample with thicker binder film thickness. Therefore, it was concluded that asphalt mixtures with a thinner binder film thickness were stiffer than those with a thicker binder film thickness. The effec- tive binder film thickness (FTbe) of all mixtures in this study were calculated based on mixture volumetrics and are shown in Table C-4. As inferred from the table, effective binder film thickness of mixtures with higher absorptive aggregates was thinner than that of mixtures with lower absorptive aggre- gates, when all other factors are kept constant. Therefore, the higher MR values of mixtures with higher absorptive aggre- gates might be attributed to the thinner effective binder film thickness in the mixture. The effect of NMAS on mixture MR results is shown in Fig- ure C-3. As illustrated, for the majority of the mixtures, the stiffness of mixtures with NMAS of 0.75 in. (19 mm) was higher than or equivalent to those of mixtures with NMAS of 0.37 in. (9.5 mm). Since the binder content used in mixtures with NMAS of 0.75 in. (19 mm) was lower than that used in mixtures with NMAS of 0.37 in. (9.5 mm), and the binder content had a significant effect on mixture stiffness (Hamzah and Yi 2008), it was anticipated that the observed differences in MR results had a stronger correlation with binder content. Factors Level Values Least Squares Mean (ksi) Binder Absorption Brownwood 559.9 Georgetown 667.7 NMAS 9.5 mm 570.2 19 mm 657.3 Binder Source Binder A 695.0 Binder V 532.5 Binder Grade PG 70-22 591.9 PG 64-22 635.6 Conditioning Protocols 2 h @ 135°C (275°F) 568.9 4 h @ 135°C (275°F) 658.6 Table C-3. Least squares means results for resilient modulus. 0 200 400 600 800 1000 R es ili en t M od ul us (k si) Mixture Type Low Absorptive Aggregate High Absorptive Aggregate 9.5 -B ind er A- 70 -2h 9.5 -B ind er A- 64 -2h 9.5 -B ind er A- 70 -4h 9.5 -B ind er A- 64 -4h 19 -B ind er A- 70 -2h 19 -B ind er A- 64 -2h 19 -B ind er A- 70 -4h 19 -B ind er A- 64 -4h 9.5 -B ind er V- 70 -2h 9.5 -B ind er V- 64 -2h 9.5 -B ind er V- 70 -4h 9.5 -B ind er V- 64 -4h 19 -B ind er V- 70 -2h 19 -B ind er V- 64 -2h 19 -B ind er V- 70 -4h 19 -B ind er V- 64 -4h Figure C-2. Effect of binder absorption on mixture MR stiffness.

C-5 Mixture Label Binder Absorption FTbe 9.5-A-70-2h Low-Brownwood 18.8 High-Georgetown 14.6 9.5-A-64-2h Low-Brownwood 16.8 High-Georgetown 14.4 9.5-A-70-4h Low-Brownwood 18.8 High-Georgetown 14.6 9.5-A-64-4h Low-Brownwood 16.8 High-Georgetown 14.4 19-A-70-2h Low-Brownwood 19.6 High-Georgetown 14.3 19-A-64-2h Low-Brownwood 20.0 High-Georgetown 13.6 19-A-70-4h Low-Brownwood 19.6 High-Georgetown 14.3 19-A-64-4h Low-Brownwood 20.0 High-Georgetown 13.6 9.5-V-70-2h Low-Brownwood 19.1 High-Georgetown 14.6 9.5-V-64-2h Low-Brownwood 18.9 High-Georgetown 14.4 9.5-V-70-4h Low-Brownwood 19.1 High-Georgetown 14.6 9.5-V-64-4h Low-Brownwood 18.9 High-Georgetown 14.4 19-V-70-2h Low-Brownwood 20.1 High-Georgetown 13.7 19-V-64-2h Low-Brownwood 19.0 High-Georgetown 14.4 19-V-70-4h Low-Brownwood 20.1 High-Georgetown 13.7 19-V-64-4h Low-Brownwood 19.0 High-Georgetown 14.4 Table C-4. Calculated effective binder film thickness results. 0 200 400 600 800 1000 R es ili en t M od ul us (k si) 9.5mm NMAS 19mm NMAS Mixture Type Lo w- Bin der A- 70 -2h Lo w- Bin der A- 64 -2h Lo w- Bin der A- 70 -4h Lo w- Bin der A- 64 -4h Lo w- Bin der V- 70 -2h Lo w- Bin der V- 64 -2h Lo w- Bin der V- 70 -4h Lo w- Bin der V- 64 -4h Hi gh -B ind er A- 70 -2h Hi gh -B ind er A- 64 -2h Hi gh -B ind er A- 70 -4h Hi gh -B ind er A- 64 -4h Hi gh -B ind er V- 70 -2h Hi gh -B ind er V- 64 -2h Hi gh -B ind er V- 70 -4h Hi gh -B ind er V- 64 -4h Figure C-3. Effect of NMAS on mixture MR stiffness.

C-6 The effect of binder source on mixture MR results is illus- trated in Figure C-4. All mixtures with Binder A were signifi- cantly stiffer than or equivalent to those made with Binder V. The effect of binder grade on mixture MR values, as indicated in Figure C-5, was not consistent for all mixtures. In some cases, the incorporation of higher PG binders in the mixtures increased the mixture stiffness significantly, while the opposite trend was observed for other cases. Therefore, it was antici- pated that the interaction between binder grade and other fac- tors had a significant effect on the mixture MR values. Figure C-6 presents the effect of conditioning protocol on mixture MR values. As anticipated, the stiffness of all mix tures conditioned with longer conditioning times were signifi- cantly higher or equivalent to those with shorter condition- ing times. The ranking of significant factors in terms of their influence effects on mixture MR values was proposed as an indicator of which factors had the biggest influence in stiffness. This is obtained from the “Magnitude of Parameter Estimate” output from JMP listed in Table C-5. 0 200 400 600 800 1000 R es ili en t M od ul us (k si) Binder A Binder V Mixture Type Lo w- 9.5 -70 -2h Lo w- 9.5 -64 -2h Lo w- 9.5 -70 -4h Lo w- 9.5 -64 -4h Lo w- 19 -70 -2h Lo w- 19 -64 -2h Lo w- 19 -70 -4h Lo w- 19 -64 -4h Hi gh -9. 5-7 0-2 h Hi gh -9. 5-6 4-2 h Hi gh -9. 5-7 0-4 h Hi gh -9. 5-6 4-4 h Hi gh -19 -70 -2h Hi gh -19 -64 -2h Hi gh -19 -70 -4h Hi gh -19 -64 -4h Figure C-4. Effect of binder source on mixture MR stiffness. 0 200 400 600 800 1000 R es ili en t M od ul us (k si) PG 70-22 PG 64-22 Mixture Type Lo w- 9.5 -B ind er A- 2h Lo w- 9.5 -B ind er A- 4h Lo w- 19 -B ind er A- 2h Lo w - 19 -B ind er A- 4h Lo w- 9.5 -B ind er V- 2h Lo w- 9.5 -B ind er V- 4h Lo w- 19 -B ind er V- 2h Lo w- 19 -B ind er V- 4h Hi gh -9. 5-B ind er A- 2h Hi gh -9. 5-B ind er A- 4h Hi gh -19 -B ind er A- 2h Hi gh -19 -B ind er A- 4h Hi gh -9. 5-B ind er V- 2h Hi gh -9. 5-B ind er V- 4h Hi gh -19 -B ind er V- 2h Hi gh -19 -B ind er V- 4h Figure C-5. Effect of binder grade on mixture MR stiffness.

C-7 The statistical analysis results for the IDT strength obtained by JMP are shown in Figure C-7. As indicated by the “Effect Tests” table in Figure C-7, among the 15 factor/level com- binations considered, the main effects of binder absorption, binder source, binder grade, conditioning protocols, and the two-way interactions between binder absorption and NMAS, NMAS and binder source, and NMAS and binder grade on IDT strength are significant at a = 0.05. Comparisons in terms of least squares means for significant factors of interests are presented in Table C-6. It can be observed from the table that, in general, higher absorptive aggregate, higher binder grade, and longer conditioning protocol may contribute to higher mixture IDT strength. Additionally, asphalt mixtures with Binder A are stronger than those that employ Binder V when all other factors are kept constant. In the following five figures that present the difference in IDT strengths of asphalt mixtures due to different levels of binder absorption, NMAS, binder source, binder grade, and condi- tioning protocol, each bar represents the average value of three replicate specimens, and the error bars represent ± one standard deviation from the average value. The effect of binder absorption on mixture IDT results, as shown in Figure C-8, was significant. More specifically, for the majority of the mixtures, those with higher absorptive aggregates are stiffer than those with lower absorptive aggre- gates, which is consistent with the trend obtained from the MR stiffness results. As illustrated in Figure C-9, the effect of NMAS on mix- ture IDT strength was not consistent for all mixtures. In some cases, the incorporation of aggregates with larger NMAS in 0 200 400 600 800 1000 R es ili en t M od ul us (k si) 2h@275F 4h@275F Mixture Type Lo w- 9.5 -B ind er A- 70 Lo w- 9.5 -B ind er A- 64 Lo w- 19 -B ind er A- 70 Lo w - 19 -B ind er A- 64 Lo w- 9.5 -B ind er V- 70 Lo w- 9.5 -B ind er V- 64 Lo w- 19 -B ind er V- 70 Lo w- 19 -B ind er V- 64 Hi gh -9. 5-B ind er A- 70 Hi gh -9. 5-B ind er A- 64 Hi gh -19 -B ind er A- 70 Hi gh -19 -B ind er A- 64 Hi gh -9. 5-B ind er V- 70 Hi gh -9. 5-B ind er V- 64 Hi gh -19 -B ind er V- 70 Hi gh -19 -B ind er V- 64 Figure C-6. Effect of conditioning protocol on mixture MR stiffness. Factors Magnitude of Parameter Estimate Influence Effect Ranking Binder Source 81.2 Significant 1st Binder Absorption 53.9 Significant 2nd Conditioning Protocol 44.9 Significant 3rd NMAS 43.5 Significant 4th NMAS & Binder Source 29.5 Significant 5th NMAS & Binder Grade 24.7 Significant 6th Binder Grade 21.9 Significant 7th Table C-5. Linear fitting results for resilient modulus.

C-8 Figure C-7. JMP output of fitting model for IDT strength. Factors Level Values Least Squares Mean (psi) Binder Absorption Brownwood 149.2 Georgetown 155.1 NMAS 9.5 mm 150.3 19 mm 153.9 Binder Source Binder A 172.0 Binder V 132.3 Binder Grade PG 70-22 158.9 PG 64-22 145.4 Conditioning Protocols 2 h @ 275°F (135°C) 144.2 4 h @ 275°F (135°C) 160.1 Table C-6. Least squares means results for IDT strength.

C-9 the mixtures increased the mixture stiffness, while the oppo- site trend was observed for other cases. Therefore, it was anticipated that a more significant effect of NMAS on mix- ture IDT strength might be associated with the interaction with other main factors. The effect of binder source on mixture IDT strength is illus- trated in Figure C-10. All mixtures prepared using Binder A were significantly stronger than those prepared with Binder V. The effect of binder grade on mixture IDT strength is illus- trated in Figure C-11. For the majority of the mixtures, the IDT strengths of mixtures with PG 70-22 were higher than those of mixtures with PG 64-22, for both Binder A and Binder V. It further verified the significant effect of binder stiffness on mixture property. Figure C-12 presents the effect of conditioning protocol on mixture IDT strength results. As expected, all mixtures with longer conditioning (except that with “Low-9.5-Binder A-64”) had IDT strengths that were higher or equivalent to those 0 50 100 150 200 250 ID T St re ng th (p si) Low Absorptive Aggregate High Absorptive Aggregate Mixture Type 9.5 -B ind er A- 70 -2h 9.5 -B ind er A- 64 -2h 9.5 -B ind er A- 70 -4h 9.5 -B ind er A- 64 -4h 19 -B ind er A- 70 -2h 19 -B ind er A- 64 -2h 19 -B ind er A- 70 -4h 19 -B ind er A- 64 -4h 9.5 -B ind er V- 70 -2h 9.5 -B ind er V- 64 -2h 9.5 -B ind er V- 70 -4h 9.5 -B ind er V- 64 -4h 19 -B ind er V- 70 -2h 19 -B ind er V- 64 -2h 19 -B ind er V- 70 -4h 19 -B ind er V- 64 -4h Figure C-8. Effect of binder absorption on mixture IDT strength. 0 50 100 150 200 250 ID T St re ng th (p si) 9.5mm NMAS 19mm NMAS Mixture Type Lo w- Bin der A- 70 -2h Lo w- Bin der A- 64 -2h Lo w- Bin der A- 70 -4h Lo w- Bin der A- 64 -4h Lo w- Bin der V- 70 -2h Lo w- Bin der V- 64 -2h Lo w- Bin der V- 70 -4h Lo w- Bin der V- 64 -4h Hi gh -B ind er A- 70 -2h Hi gh -B ind er A- 64 -2h Hi gh -B ind er A- 70 -4h Hi gh -B ind er A- 64 -4h Hi gh -B ind er V- 70 -2h Hi gh -B ind er V- 64 -2h Hi gh -B ind er V- 70 -4h Hi gh -B ind er V- 64 -4h Figure C-9. Effect of NMAS on mixture IDT strength.

C-10 Figure C-11. Effect of binder grade on mixture IDT strength. 0 50 100 150 200 250 ID T St re ng th (k si) PG 70-22 PG 64-22 Mixture Type Lo w- 9.5 -B ind er A- 2h Lo w- 9.5 -B ind er A- 4h Lo w- 19 -B ind er A- 2h Lo w- 19 -B ind er A- 4h Lo w- 9.5 - Bin der V- 2h Lo w- 9.5 - Bin der V- 4h Lo w- 19 -B ind er V- 2h Lo w- 19 -B ind er V- 4h Hi gh -9. 5-B ind er A- 2h Hi gh -9. 5-B ind er A- 4h Hi gh -19 -B ind er A- 2h Hi gh -19 -B ind er A- 4h Hi gh -9. 5-B ind er V- 2h Hi gh -9. 5-B ind er V- 4h Hi gh -19 -B ind er V- 2h Hi gh -19 -B ind er V- 4h Figure C-12. Effect of conditioning protocol on mixture IDT strength. 0 50 100 150 200 250 ID T St re ng th (p si) 2h@275F 4h@275F Mixture Type Lo w- 9.5 -B ind er A- 70 Lo w- 9.5 -B ind er A- 64 Lo w- 19 -B ind er A- 70 Lo w- 19 -B ind er A- 64 Lo w- 9.5 -B ind er V- 70 Lo w- 9.5 -B ind er V- 64 Lo w- 19 -B ind er V- 70 Lo w- 19 -B ind er V- 64 Hi gh -9. 5-B ind er A- 70 Hi gh -9. 5-B ind er A- 64 Hi gh -19 -B ind er A- 70 Hi gh -19 -B ind er A- 64 Hi gh -9. 5-B ind er V- 70 Hi gh -9. 5-B ind er V- 64 Hi gh - 19 -B ind er V- 70 Hi gh -19 -B ind er V- 64 Figure C-10. Effect of binder source on mixture IDT strength. 0 50 100 150 200 250 ID T St re ng th (p si) Binder A Binder V Mixture Type Lo w- 9.5 -70 -2h Lo w- 9.5 -64 -2h Lo w- 9.5 -70 -4h Lo w- 9.5 -64 -4h Lo w- 19 -70 -2h Lo w- 19 -64 -2h Lo w- 19 -70 -4h Lo w- 19 -64 -4h Hi gh -9. 5-7 0-2 h Hi gh -9. 5-6 4-2 h Hi gh -9. 5-7 0-4 h Hi gh -9. 5-6 4-4 h Hi gh -19 -70 -2h Hi gh -19 -64 -2h Hi gh -19 -70 -4h Hi gh -19 -64 -4h

C-11 with less conditioning time. Therefore, the extended short-term conditioning for loose mix was able to significantly increase the mixture stiffness and strength. The ranking of significant factors in terms of their influ- ence effects on mixture IDT strength was proposed based on the “Magnitude of Parameter Estimate” output from JMP listed in Table C-7. On the basis of the statistical analysis of the laboratory test results, it can be concluded that binder source has a profound effect on both MR and IDT strength. Specifically, mixtures with Binder A were stiffer than those with Binder V. There was also evidence that mixture MR and IDT strength were sensitive to the laboratory conditioning protocol of the loose mix. NMAS was verified to have a significant effect on mixture stiffness while it showed no significant effect on mixture strength, and the difference was likely due to the compound effect of NMAS and binder content in the mix. The effects of binder grade and binder absorption on mixture MR stiffness and IDT strength were significant. Additionally, two-way interactions between NMAS and binder source, and between NMAS and binder grade, also had a significant effect on mixture stiffness and strength, based on the statistical analysis. Based on the findings from the small laboratory paramet- ric study, there is strong evidence that mixture aging charac- teristics are strongly influenced by asphalt source and binder absorption by aggregate. NMAS and binder grade have lesser effects on mixture aging, and it is suspected that the effects of NMAS may be tied to mixture volumetrics. The labora- tory conditioning protocol is of particular importance to the mixture stiffness and strength. In the field trials, the research team made a particular effort to incorporate the mixture variables of binder source and aggregate absorption into the testing matrix as well as the plant characteristics of temperature and design. The resulting aging protocols for short-term and intermediate-term aging reflect the differences in these variables, which span a wide range of field and mixture conditions. References Glover, C. J. (2010). “Oxidation and Kinetics of Aging in Asphalt Binders (and So What?).” Presented at the International Workshop on Binders and Mastics, Madison, Wisconsin. As of March 1, 2013: https:// uwmarc.wisc.edu/files/binders2010/Oxidation_and_Kinetics_of_ Aging_in_Asphalt_Binders_-_Glover.pdf. Hamzah, M. O. and T. C. Yi (2008). “Effects of Temperature on Resilient Modulus of Dense Asphalt Mixtures Incorporating Steel Slag Sub- jected to Short Term Oven Ageing.” World Academy of Science, Engi- neering and Technology, Vol. 2, No. 10, pp. 171–175. Wang, D. (2007). “Binder Film Thickness Effect on Aggregate Contact Behavior.” Master Thesis, Virginia Polytechnic Institute and State University, Blacksburg, VA. Factors Level Values Least Squares Mean (psi) Binder Absorption Brownwood 149.2 Georgetown 155.1 NMAS 9.5 mm 150.3 19 mm 153.9 Binder Source WT 172.0 SA 132.3 Binder Grade PG 70-22 158.9 PG 64-22 145.4 Conditioning Protocols 2 h @ 275°F (135°C) 144.2 4 h @ 275°F (135°C) 160.1 Table C-7. Linear fitting results for IDT strength.