RAP and RAS use in asphalt paving
As you hear from superheroes, “With much power, comes great responsibility.” Well we do have Superpave, but no superheroes that we know of in our industry. However, we do have “super” piles of Reclaimed Asphalt Pavement (RAP) and Reclaimed Asphalt Shingles (RAS). While we want to be good stewards and use these reclaimed materials with a good purpose, this task also comes with much responsibility.
During the last decade, the quality and life of our asphalt pavements in many areas of the U.S. have decreased. In 2002, one state reported a decrease in average pavement life of about 4 years. Similar reports of concerns of decreased pavement life are being heard from other areas in the U.S.
What is the cause of such a dramatic loss in quality and durability? Some believe that Superpave has dried out our mixtures by too much compaction in the lab, thus reducing the amount of asphalt binder. Others have raised questions about the amount of recycled product in our mixtures.
Around the same time that RAP was widely introduced, Superpave and QA/QC processes were being introduced to states in the U.S. RAS came into the picture a little later. We did make a lot of changes to our mixtures in the late 1990s to rid our asphalt pavements of rutting and shoving per the Superpave standards. We also added RAP and RAS. Now many states have a cracking problem.
The Superpave mix design system included materials selection and compaction levels (Ndesign or gyrations). Mix design volumetrics (volume relationships to properly balance the mixture components of aggregate, asphalt and air) were built around virgin mixtures and so were the initial trials such as I-43 in Wisconsin. Superpave never really had a chance to prove itself before we added other non-virgin, recycled components with many assumptions on how to use them.
We believe that many of our issues with cracking stem from the improper use of RAP/RAS. When used correctly, RAP/RAS can have economic benefits, but when not used properly these same materials can lead to low binder contents, artificially-inflated VMAs (voids in the mineral aggregate) and overly stiff or brittle mixtures. For RAS we are still trying to understand its long-term effects. While RAP is 95 to 96 percent aggregate, RAS is only 70 to 80 percent aggregate and the rest being a stiff asphalt formulated for roofing standards.
Despite widespread use of RAP and RAS in the U.S., there seems to be a general lack of awareness of the guidelines for the proper use of RAP and RAS that are available and have been since 2002.
In 2002, National Cooperative Highway Research Program (NCHRP) published “Recommended Use of Reclaimed Asphalt Pavements in the Superpave Mix Design Method” by Rebecca McDaniel and R. Michael Anderson. This report was based on the work from “NCHRP 9-12, Incorporation of Reclaimed Asphalt Pavement in the Superpave System.” More recently, further information has been published in the Asphalt Institute’s “MS-2, Asphalt Mix Design Methods.”
Even RAS has ever-changing specifications on how it should be used. AASHTO PP78 is a provisional standard that has undergone major recent changes as we learn about how to use RAS more responsibly. The latest change is the addition of D Tc (called delta Tc) as a quality indicator to reduce cracking potential of the recovered RAS asphalt binder.
A general lack of awareness of these usage guidelines and specifications, and other studies like them, has contributed to less understanding the impact of RAP and RAS on asphalt mixes. Consequently, RAP and RAS usage levels have continued to climb higher without a full understanding of the long-term effects on the life of asphalt pavements.
This is not to say that we should not be using RAP/RAS. The economic benefits are too great to be ignored. In addition, asphalt pavements continue to be among the most recycled products in the U.S., according to the National Asphalt Pavement Association (NAPA). RAP’s recycling percentage is tied with lead-acid batteries, 99 percent of which are reported to be recycled. As for other frequently recycled products, paper and aluminum fall to 70 and 80 percent respectively, according to Scientific American’s website. These RAP recycling percentages show that we are good stewards in our paving industry and that we have an active awareness of our impact on the environment.
Current practices
The problem: U.S. economics since 2001 have caused DOTs to become more innovative with the use of RAP and RAS. These budgets have created some critical long-term compromises.
Assumptions: We are designing asphalt mixes and building roads based on the assumption that RAP has 4 to 5 percent “aged” asphalt binder and that RAS has 20 to 30 percent “extremely aged” asphalt binder. As a matter of fact, much of the stiff binder may not even release from RAS, causing an “assumed” false asphalt content. We must adjust the mix properties to properly accommodate the RAP and/or RAS.
Current practices also assume that 100 percent of the RAP asphalt and 75 to 80 percent of the RAS asphalt will blend into the mix. RAP, for example, may have an average of PG 82-16 and RAS may have an average of PG 120+10. A typical asphalt mix can handle about 15 percent of the PG 82-16 but cannot handle PG 120+10. A question some asphalt mix designers are asking is: “Do we really want to blend this in our surface mixtures?”
Another question we designers need to be asking is “How much of the asphalt from the RAP and RAS is actually alive, working and effective? A best-case scenario is that 100 percent of the asphalt from the RAP blends into the mix. But more likely a much smaller percentage actually blends into the mix, such as 80 percent. As for RAS, it may be closer to 20 to 50 percent of the RAS asphalt that blends into the mix. This happens because the RAS is usually extremely stiff and will not even be fluid to pour unless elevated to 400°F (204°C).
Example: If we are using a mix that requires 20 percent RAP with 5 percent AC (liquid asphalt), we will credit the RAP with 1 percent AC [5% X 20 = 1%AC].
5% RAP binder X 20% RAP = 1% RAP AC in the total mix.
In reality, however, the RAP realistically may only provide let’s say, 0.8% binder.
80% release X 1% RAP AC in the total mix = 0.8%
That means we may be missing 0.2 percent AC, but on paper our design looks perfect.
Where’s the asphalt?
The issue is not just the stiffness of the RAP or RAS, but an insufficient amount of the needed virgin asphalt. The quality and durability of the pavement is contained in the virgin asphalt.
This missing asphalt stems from the above assumptions and inflated RAP aggregate gravities.
Let’s discuss how the process works. Gyratory compactors (or any compaction device) simply compact the mix in a mold making a “bulk” sample. From this we measure the bulk gravity and calculate density, air voids, VMA, etc. (volumetric properties). It is then the mix designer’s option to fill the voids with AC or dust. They will normally not add more liquid asphalt (AC) since this is the more expensive design component. Keep in mind that on paper, all looks good because we may have an inflated RAP aggregate gravity. So, an inflated RAP aggregate gravity will produce deceptively high VMAs.
(For those who are wondering, simply decreasing gyrations does not equal more asphalt. In fact, it makes VMA easier to meet, causing the opposite effect of potentially decreasing asphalt content.)
If we combine the missing 0.2 percent from above with a potential inflated RAP aggregate gravities, we starve the mix of even more AC. The result being that we have less asphalt binder than is needed to produce the mix, while on paper (the design) all looks great.
Problems with the current RAP and RAS pavement mix may not show up right away. When we make changes in mix design we usually won’t see the effects for maybe five-plus years. If the effect we see is unsatisfactory, we replace the pavement with the current mix design containing RAP and we won’t see that effect for another five-plus years. Right now, we are seeing the second and third generation of RAP use. Can you see how this problem builds?
What can we do?
How can we overcome the trend to use more and more RAP and RAS without compensating for the missing AC? Perhaps the first thing we can do is agree that we need more information and more understanding about the effect that various quantities of RAP and RAS have on asphalt mixes. There is a need to adjust mix properties based on proven assumptions and not on current assumptions. We can do several things to be more responsible.
Let’s look at the references listed at the end of the article and see what is recommended. Not just do what we think may work.
1. Calibrate our burn-off asphalt contents with a solvent extraction for RAP and RAS. The correction factor should not be assumed. Also, we need to know more certainly how much of the AC from the RAP or the RAS is blending into the mix or at least use a better (less) assumption than 100 percent.
2. We must be more cautious of typical RAP and RAS mix designs and be willing to add softer liquid asphalt in order to compensate for the stiffer RAP and RAS mixes. Follow the recommend tiered system of:
• 0-14 percent replacement binder ratio – no PG adjustment
• 15-25 percent replacement binder ratio – lower one PG (for example 64-22 to a 58-28)
• >25 percent replacement binder ratio – use a blending chart, see on AASHTO M323
3. Use more realistic RAP aggregate gravities. If local aggregates have a gravity of 2.650, then it is not likely that the local RAP aggregate will be 2.750.
4. While we did not discuss consistency of the RAP/RAS piles, it is well known that the use of material from a RAP pile where the material is consistent will produce a more consistent mix. The use of material from an “anything goes” RAP pile will produce an inconsistent, unreliable mix.
To wrap it up, we can use RAP/RAS responsibly. While we are not superheroes that we mentioned at the beginning, we all do have a great responsibility to take care of our nation’s highway infrastructure by being responsible in the design process, specifically the use of RAP and RAS.
Examples of available guidelines
•“AASHTO M323, Appendix: Superpave Volumetric Design”
• “AASHTO PP53: Design Considerations When Using Reclaimed Asphalt Shingles (RAS) in New Hot Mix Asphalt (HMA)”
• “AASHTO MP15: Use of Reclaimed Asphalt Shingles as an Additive in Hot Mix Asphalt”
• “AASHTO PP78: Standard Practice for Design Considerations When Using Reclaimed Asphalt Shingles in Asphalt Mixtures”
• “Asphalt Institute MS-2, Asphalt Mix Design Methods” chapter 11
John Davis and Phil Blankenship, P.E. collaborated on this article.