Using reclaimed asphalt pavement (RAP) in asphalt mixtures can provide significant economic savings to contractors and highway agencies by reducing the demand for both the virgin binder and aggregates in asphalt mixtures.
Unfortunately, some highway agencies are reluctant to increase RAP content due to the general perception that these mixtures may be more susceptible to various modes of cracking such as fatigue, thermal and reflection. This is due to the fact the RAP binder is aged, stiffer and less strain-tolerant than a virgin binder. Agencies want assurance that high-RAP-content mixes will provide satisfactory performance in the field before permitting their use.
The goal of several recent research efforts has been to increase the percentage of RAP without sacrificing performance. One potential method for increasing the durability of these mixtures is to change the grade of the virgin binder; however, some agencies do not want to adjust it more than one or two grades, as an incomplete blending of the recycled and new binders could lead to early distresses caused by soft areas in the pavement. Also, softer binder grades are generally more expensive, which would negate some of the economic savings realized from using RAP.
Recent research at the National Center for Asphalt Technology (NCAT) indicates the performance of RAP mixtures also is affected by the volume of the virgin binder in the mixture rather than just its performance grade.
This study was completed as a companion to NCAT Report 12-03, which examined the effects of changing the virgin binder grade for 25% and 50% RAP mixtures. The previous laboratory study showed that for 25% RAP mixtures, using additional virgin binder provided the greatest resistance to cracking without increasing rutting susceptibility. For 50% mixtures, using the softer grade of virgin binder gave the best mixture performance.
The objective of this study was to evaluate how increasing the volume of virgin binder, using a softer grade of virgin binder or using a WMA chemical additive affected the durability and performance of RAP mixtures.
Putting more in
For this study, 10%, 25%, and 50% RAP mixtures were designed in accordance with AASHTO R 35 Standard Practice for Superpave Volumetric Design for Hot-Mix Asphalt (HMA). Each mix was designed as 12.5-mm nominal maximum aggregate size mixture. The mixes contained three aggregate stockpiles and a locally available RAP stockpile.
As is standard practice, the RAP was characterized before being used in the asphalt mixtures. A PG 67-22 virgin binder was the base binder, and the softer binder was a PG 58-28 binder reduction in both the high and low temperature grades of the virgin binder. It should be noted that a lower virgin binder grade is not standard practice for 10% RAP mixtures; however, at the request of the research sponsor, mixtures were prepared at each RAP content with the PG 67-22 binder and the PG 58-28 binder. When used, a chemical WMA additive was added to the asphalt binder during mixing at a rate of 0.53% by weight of the asphalt binder.
Table 1 shows the RAP binder ratios for the experimental mixtures using the PG 67-22 virgin binder. The 10%, 25% and 50% RAP mixes were replicated with the WMA additive at the corresponding optimum asphalt content. Likewise, the 10%, 25% and 50% RAP mixes were replicated with PG 58-28 virgin binder at the corresponding optimum asphalt content.
Numerous laboratory tests were conducted to quantify variations in the durability of the various RAP mixtures. The linear amplitude sweep (LAS) was utilized to characterize the fatigue properties of the blended RAP and virgin binder. The LAS test is an accelerated binder fatigue test that accounts for damage resistance as well as traffic loading by using increasing cyclic load amplitudes to accelerate damage. The end result is a prediction of binder fatigue life as a function of strain magnitude.
Energy ratio testing, developed at the University of Florida, also was performed to evaluate each mixture’s resistance to top-down or surface cracking. Overlay tests (OT) were conducted to assess the resistance to reflection cracking of the mixtures. The rutting resistance of the most durable asphalt mixtures was assessed using the Asphalt Pavement Analyzer (APA) to ensure that increasing mixture durability did not cause the asphalt mixture to become susceptible to rutting.
Blends of the virgin and extracted RAP binders were created corresponding to the amounts of each binder in the 10%, 25% and 50% RAP mixture designs. The blends with the PG 67-22 virgin binder were adjusted to correspond with the increased virgin binder contents of 0.25% and 0.5% as shown in Table 1. Results showed these small incremental increases in virgin binder had little effect on the number of cycles of failure. However, using the softer virgin binder dramatically improved the LAS fatigue test results. The blend corresponding to 25% RAP and PG 58-28 had much longer fatigue life than the virgin PG 67-22 alone, and the blend corresponding to 50% RAP and PG 58-28 had essentially the same fatigue life as the PG 67-22 alone.
Fracture energy and energy ratio results were less conclusive. For the 25% RAP mixes, all of the modifications (i.e., added binder, softer binder or WMA additive) resulted in statistically lower fracture energies and energy ratios. For the 50% RAP mixes, the mix with the WMA additive and mixes 0.25% and 0.50% added AC improved the fracture energy compared with the standard mix design. The fracture energy for the mix with the PG 58-28 was not statistically different from the mix with the PG 67-22. Similarly, the energy ratio of the 50% RAP mix was improved by adding 0.5% more asphalt or adding the chemical WMA additive.
Statistical analysis of the OT results indicated that for each RAP content, none of the mix modifications yielded statistically different resistances to reflection cracking. This is most likely due to the poor repeatability of the test.
The mixture with the greatest rut depth was the 10% RAP mixture with the PG 58-28 binder. However, this mixture had an APA rut depth of slightly greater than 4 mm, which is considered satisfactory. APA rut depths for the 25% and 50% RAP mixes were not statistically affected by the modifications made to the mixes.
Based on this laboratory study, the following conclusions were made regarding modifications to RAP mixtures to improve cracking resistance:
1. Using a softer binder had the greatest impact on improving the fatigue life of all the RAP binder blends based on the LAS binder fatigue test;
2. Increasing the effective virgin binder content had little effect on the LAS test results;
3. Using 0.5% additional virgin asphalt and WMA technology improved the fracture energy of the 10% and 50% RAP mixtures, although no statistical difference was found for the 25% RAP mixtures;
4. The energy ratio decreased for the 25% RAP mixtures that used a softer virgin asphalt or had increased asphalt content. Using a 0.5% additional asphalt or WMA showed the greatest energy ratio increase for the 10% and 50% RAP mixtures;
5. Overlay test results were not statistically affected by added asphalt, softer virgin binder or WMA at any RAP content; and
6. For the 10% RAP mixture, using a softer binder increased the APA rutting results, but the results were less than 5 mm, which would be acceptable for most roadways. Higher-RAP-content mixtures were not detrimentally affected by using softer virgin asphalt, a higher asphalt content or WMA.
No questioning durability
Based on this limited study, technical and cost-effective options for enhancing the durability of high-RAP mixtures appear viable, although further work is needed to validate these solutions in the field. When using less than 25% RAP, using an additional 0.5% virgin asphalt or incorporating a WMA technology into the RAP mixture should provide additional durability. At 25% RAP, a softer binder or WMA technologies should be used to increase the mixture durability. At 50% RAP, an additional 0.5% most consistently had the greatest impact on mixture durability.
It should be noted that these results are inconsistent with the results of the previous NCAT RAP durability study (NCAT Report 12-03). In the previous study, using additional asphalt provided the best benefits at 25% RAP and using a softer binder was the most effective way of increasing mixture durability at 50% RAP. This shows there is not a one-size-fits-all method for ensuring mixture durability. Choosing between additional asphalt and softer binders may require highway agencies and contractors to assess which method is best on a case-by-case basis, using local materials incorporated in the mix designs.
When using alternative technologies to increase mixture durability, one must ensure the mixture will not become susceptible to rutting. These options should be validated in the field and further analyzed on a regional basis. Using a different WMA technology may change the performance of the mixtures.
Highway agencies and contractors should conduct cost analyses to determine if adding additional binder, a softer binder or using a WMA technology would provide the most cost-effective solution when similar results are seen as options. AT