By Dan Krivit

The use and understanding of recycled asphalt shingles (RAS) as a value-added supplement in hot mix asphalt (HMA) continues to grow throughout the U.S. and Canada. Three presentations were given by expert practitioners at the 4th Asphalt Shingles Recycling Forum in Chicago in November 2009 as part of the panel on the “Value of Recycled Asphalt Shingles in Hot Mix Asphalt.”1 Each speaker provided valuable, new insights into the technical and economic issues that are key drivers in this rapidly growing technology of asphalt shingle recycling.

Paul Lum, director of the quality asphalt and paving division2 with Lafarge North America, presented “HMA Performance with Processed Manufacturer Shingles.” He provided a detailed summary of the extensive research originally led by the Sintra Group, a subsidiary of the Colas Group3 (for research projects conducted in Quebec), and also conducted by Lafarge (for work conducted in Alberta and Ontario).

Most of the Lafarge and Colas research has been conducted on manufacturers’ (pre-consumer) shingle scrap and dates back as early as 1999. In Quebec, these studies included both laboratory and field components. A series of tests were conducted including: low temperature cracking, fatigue cracking, dynamic modulus measurements, and resistance to rutting.  Manufacturers’ RAS mixes were prepared as both an “additive” (when the virgin AC is not adjusted) and as a “modifier” (when the virgin AC is reduced to account for the RAS binder effects). The RAS mixes were compared against a set of similar “control” mixes with no RAS component.

The thermal cracking tests have indicated that mixes with 2%, 4%, 6% and 8% RAS behave the same with no impact from the shingles, since the RAS was treated as an additive with no adjustment made to the virgin asphalt cement (AC) content (i.e., the total of virgin + RAS AC content was increased). In the next series of tests, the amount of virgin AC has reduced on the theory that the AC within the RAS would have some positive contribution to the effective, total binder in the mix. The thermal cracking results tests indicated the mixes at lower contribution of RAS AC had similar results as the control. The Lafarge research concluded that the shingle fiber and aggregate in the RAS, and not just the RAS bitumen, plays a role in mitigating impacts and can even improve the low temperature cracking resistance in the mixture depending on mix design. Similarly, the fatigue cracking tests indicated that mixes with RAS have increased resistance to fatigue cracking. Even when the virgin AC binder was reduced to account for the RAS AC contribution, fatigue cracking resistance was improved significantly in the RAS mixes.

The dynamic modulus tests indicated that at high temperatures, the RAS modified mixes were significantly more rut resistant than the control mix without RAS. At low temperatures the shingle modified mixes behaved similarly to the control mix. The RAS—modified mixes should have the same performance in low temperature cracking as the control mix. It is also important to note the Lafarge reported research concluded that there is an optimum amount of RAS that can be added (up to 6% given the Lafarge test designs) before there starts to be negative impacts shown by the complex modulus tests. They believe the negative impacts may be due in part to the changes caused in the mix gradation and rheological properties of the mix.

The rutting tests were performed using the “LCPC French Rutting Test.”4,5 The results of the Lafarge rutting tests indicating improved rut resistance with the 5% RAS mixtures compared to control are shown in Figure 1.

Lafarge also conducted a series of field validation demonstration projects to determine HMA pavement performance when using RAS. After two years of service, the RAS-modified HMA pavement displayed excellent behavior with no signs of rutting. Transverse and longitudinal cracking rates were very low. Also, the surface texture was uniform and in excellent condition: no surface raveling or stripping was evident.6,7

Lum noted that it is expected that tear-off shingles will be more aged than manufacturers’ shingle scrap. In either case, there is a keen need to monitor the consistency of RAS for AC content (percent) and PG grade impacts.

The next panelist, Ron Sines, vice president of HMA operations at Oldcastle Materials, presented “Use of Recycled Shingles in HMA Pavements.” Oldcastle Materials, Inc., a division of CRH, plc (the International Building Materials Group), is the largest HMA producer in the United States, producing 50 million tons of asphalt at more than 450 facilities.

Sines first message was shingles recycling, like many other recycled commodities, is a part of our new economy. Recycling is part of a very powerful tool toward greater efficiency which means doing the same amount of work, or more, with less inputs and emissions.

Sines stated that there are many reasons to use RAS in HMA, but the primary driver is economic savings. He presented one version of the historical price curve (2001 through 2008) for virgin liquid asphalt binder (See Figure 2). He stated that using RAS can actually save $3 to $5 per finished ton of HMA depending on the local circumstances. These savings will vary depending on the prevailing landfill tipping fees, the primary competitor to recycling asphalt shingles.

Sines also noted that RAS is often used in conjunction with recycled asphalt pavement (RAP). Yet the availability of RAP is dwindling because more and more of this recycled product is being used on-site for road base aggregate at the time of milling. Plus, RAP is not as readily available in rural areas, so the use of RAS as an alternative provides more recycling opportunities.

Finally, using RAS in HMA is just the right thing to do from an environmental and political perspective. But the bottom line is that the entire shingle recycling process can be carefully engineered such that a high quality, reliable, consistent RAS product can be produced that satisfies stringent HMA materials specifications. In the end, RAS-modified HMA pavements are equivalent in performance to traditional mix designs.

Potential benefits from the use of shingles in HMA include:
•    Improved resistance to pavement cracking (due to reinforcement from fibers);
•    Improved resistance to rutting (due to fibers and increased stiffness of binder);
•    Conservation of natural resources;
•    Reduced costs for the production of HMA;
•    Conservation of landfill space; and
•    Reduced costs for shingle waste disposal.

Most of the past and recent studies have concluded that the impact of RAS on HMA pavements is neutral. That is, there are no significant, negative engineering effects of using shingles in asphalt pavement when proper quality assurance/quality control (QA/QC) measures are used throughout the recycling and HMA production processes.

The recently amended shingle recycling standard and guidelines by the American Association of State Transportation Officials (AASHTO)8 provide for guidance about the use of RAS in HMA. The AASHTO specifications state that during mix design, it is important to note that the size of the RAS particles can be expected to affect the fraction of RAS binder that contribute to the total, final blended binder in the finished HMA product. It is well known that the finer the RAS grind size, the more the RAS binder becomes effective in the mix. Particles of undissolved RAS asphalt binder may act like aggregate particles that require more virgin asphalt binder to accomplish coating. Also, fibrous material present in RAS may also require additional virgin asphalt binder to accomplish coating. Additional testing is required if the RAS binder exceeds 0.75% of the final HMA mix. The AASHTO standard practice guidelines also allow for controlled pre-blending of the RAS with fine, virgin aggregate (e.g., bituminous aggregate) to prevent agglomeration in the RAS storage stockpiles. AASHTO guidelines provide a very clear definition of what may constitute deleterious foreign contaminants in the RAS, a standard set of maximum allowable levels and a uniform method of measurement. AASHTO specifications allow as much as 3% by weight of the total RAS sample.

Sines stated that Oldcastle’s experience indicates that the primary engineering concern about RAS is the particle size. Oversized RAS can negatively impact the utilization of the RAS binder, final HMA mat quality, and the consistency of the aggregate blending. Therefore, the finer the RAS size, the better.

A variety of equipment vendors are actively and successfully selling their shingle grinding machines to recyclers that need a durable, cost-effective processing system that will consistently produce a RAS product that meets engineering specifications.

Sines reviewed the experience of nine different Oldcastle HMA plants that have been using RAS in their mixes on a fairly regular basis (See Figure 3). RAS is often pre-blended with traditional aggregates (e.g. sand or RAP) to help maintain flowability and prevent reagglomeration due to the high asphalt content. The amount of traditional bituminous aggregate (e.g., virgin sand) used as a “carrier aggregate” to blend with the shingles has ranged from 0% to 100%. Five of these plants mixed RAS at ratios of 20%, 25%, 50% and 75%. Three of these plants did not mix RAS with a carrier aggregate and, instead, fed 100% RAS into the HMA plants via the normal cold feed bins. As a general rule, RAS is normally fed into the HMA drum as a part of the “recycled” line into the middle feed collar of the drum. Often, RAP and RAS are fed side-by-side via cold feed bins and then fed via conveyors at a specified blend ratio this recycle collar. The “RAS + sand” and “RAS + RAP” blends have tended to retain more moisture than the “RAS only” feedstock piles, plus the time to mix the two products is eliminated.

Sines said Oldcastle has conducted a series of lab and field research studies to determine the effects of using RAS on the performance grade (PG) of the final mix. The results indicated that there were no significant production or placement problems. The typical mix ratio has been 5% to 7% RAS of the total aggregate by weight. The actual percentage used at each plant is based on the pavement grade specified, mix type, and surface course of HMA vs. binder or base course. Both Marshall and Superpave designs have been developed. Use of RAS has been successful in batch and drum facilities.

A few outstanding materials handling and market development issues were noted by Sines. For example, the shingle “tabs” (“cut outs” or “slugs”) produced as a reject product from “three-tab,” asphalt shingle manufacturers’ can sneak through grinder. There is still a general lack of acceptance of this newer recycling practice. Because the demand for RAS-derived HMA pavement is not fully accepted, the HMA plants must retain the ability to use multiple recycled products at the same time (e.g., RAP and/or RAS). There is an ongoing challenge about the consistency of whole shingle supply. Also the uniformity of the grind of finished RAS product from various recyclers and grinding machines is a problem and may require the oversized particles to be screened after grinding.

There is still very active debate and different mix design assumptions about the effective contribution of the RAS binder into the final mixture’s total binder content. One Oldcastle plant discounts the RAS binder by 60% and another discounts it by 20%. Part of this concern is derived from the RAS binder content consistency. Sines also noted that there is increased wear on equipment due to RAS use.

In summary, Sines said the recycled asphalt shingles can be effectively used in HMA to produce a mix of equal or better quality. Binder savings in excess of those obtained from RAP use alone appear realistically achievable. However, there are a number of practical issues need that still need to be addressed, but that the future looks very bright for this new technology.

Ray Bonaquist, PhD and P.E., chief operating officer for Advanced Asphalt Technologies, LLC, presented “Recycled Asphalt Shingles in HMA: Effect on Binder Properties in Assessing Blending.” Advance Asphalt Technologies (AAT) is a specialized engineering and testing company that has conducted numerous national research studies on this topic and provides consulting services to government agencies, asphalt binder suppliers, HMA producers, paving contractors and civil engineering design firms.

Bonaquist said HMA mixes that use recycled binders must meet the same performance criteria and quality standards as all-virgin mixes. The resulting, final HMA product derived from RAS must meet the same standards for binder grade, strength/durability, and variability. RAS and RAP products must be as homogeneous as possible because of the high degree of mixing that takes place before and during HMA production.

There is general consensus within the mix design engineering community that there needs to be a limit on the amount of RAS incorporated into the HMA plant. Bonaquist said shingle binders have much different rheology than traditional paving grade (virgin) binders. One of the primary concerns is whether RAS will lead to pavements with a greater potential for premature cracking.

Lab test results by AAT have indicated that at 25% RAS binder, the low temperature grade is one level poorer and the high temperature grade is improved two levels. Thus, it is very important to employ proper virgin binder selection that account for these predicted impacts such that the final HMA product binder meets or exceeds the minimum PG grade specified. The actual field results of how well RAS binder is effectively incorporated into the final HMA binder properties is very plant specific. This is similar to traditional mixes. In addition to mix design, results of the effective utilization of the RAS binder will vary by the quality of the RAS; HMA plant type; and plant operations. It is possible, however, to obtain a high degree of effective mixing between the RAS binder and the virgin binder.

The dynamic modulus lab tests, when performed on plant-produced, finished HMA mixtures, should reflect these variables. HMA producers can thereby calibrate their mix designs by comparing the lab measured actual mix properties against the estimates from the dynamic modulus predictive model. Operational changes in the shingles processing and the HMA production plant may then be needed to use very high RAS contents in the mix.

In conclusion, Bonaquist said RAS can be used effectively in HMA. Use of RAS in HMA will require additional engineering analysis and QA/QC procedures. In the end, it may be necessary to select appropriate virgin binder grades and for higher RAS percentages. The extent of such virgin binder grade adjustments can be determined through tests and modeling such as the dynamic modulus.

Dan Krivit is a consultant with Foth Infrastructure & Environment, LLC and can be reached at This e-mail address is being protected from spambots. You need JavaScript enabled to view it or 651-288-8509. The author wishes to thank Ed Jakush, Louisiana Elastomer, LLC, for his review and comment on this article.

Footnotes
1    This article includes key highlights from each of the three speakers on this Forum panel. For more details and highlights of other speakers, go to www.ShingleRecycling.org .  Note also that ShingleRecycling.org has started a new subscription service to give access to the full Power Point presentations submitted by the Forum speakers.
2    Paul Lum has since moved into the Lafarge’s Aggregates division.  Much of the research that Lum presented was conducted by Hassan Baaj (Lafarge Research Center).
3    Paul Lum’s presentation contained selected private data on the use of manufacturers’ scrap RAS in HMA that originated from both Lafarge (for Alberta and Ontario) and Sintra Group, a subsidiary of Colas Group, (for Quebec). The Quebec research was also conducted in Collaboration with the “Ecole de Technologie Supérieure “ (ETS) at University of Quebec in Montreal) and the Quebec Ministry of Transportation.
4    H. Baaj, P. Dorchies, D. Perraton, B. Tessier “Module Complexe et Comportement à Basse Température des Enrobés Bitumineux Modifiés aux Bardeaux d’Asphalte” 49th Annual Conference of the CTAA Montreal QC, November 21–24, 2004, p. 315-240.
5    H. Baaj “Asphalt pavement modified with recycled roofing asphalt shingles in municipal environment” The 13th Congress of Urban Infrastructure (INFRA) Laval, QC, Canada, 5 to 7 November 2007
6    H. Baaj, M. Paradis “Use of post-Fabrication Asphalt Shingles in Stone Matrix Asphalt Mix (SMA-10): Laboratory Characterization and Field Experiment on Autoroute 20 (Québec)” 53rd Annual Conference of the CTAA Saskatoon SK, November 16–19, 2008, p. 365-384.
7    H. Baaj “Asphalt pavement modified with recycled roofing asphalt shingles in municipal environment” The 13th Congress of Urban Infrastructure (INFRA) Laval, QC, Canada, 5 to 7 November 2007.
8    AASHTO MP015-09-UL: Standard Specification for Use of Reclaimed Asphalt Shingles as an Additive in Hot Mix Asphalt (HMA). (For ordering information, link to: https://bookstore.transportation.org/item_details.aspx?ID=1420.) And AASHTO PP053-09-UL: Standard Practice for Design Considerations When Using Reclaimed Asphalt Shingles (RAS) in New Hot Mix Asphalt (HMA). (For ordering information, link to: https://book-store.transportation.org/item_details.aspx?ID=1421.)