Collective critical thinking from Asphalt Institute engineers
By Dr. Mark Buncher, Ph.D., P.E. and Danny Gierhart, P.E.
Going back to 1956 with the first “MS-2 Mix Design Manual”, the Asphalt Institute (AI) has published established, well-proven mix design techniques that promote the long-term quality performance of asphalt pavement. Given that more than $25 billion is spent in the U.S. annually on asphalt pavements, changes to those established mix design techniques must undergo critical evaluation to protect such a valuable public resource.
While the Superpave mix design system essentially resolved the rutting issues our industry was facing in the 1990s, it also led to later durability concerns, especially when combined with more frequent use of higher percentages of reclaimed asphalt pavement (RAP) and recycled asphalt shingles (RAS) in our mixes. The pendulum shifted from mostly being concerned about rutting to now being mostly concerned about durability and cracking.
As the asphalt industry turned its attention to addressing cracking and durability issues, an innovative approach began to take shape: balanced mix design (BMD). Although the full definition is given later, the BMD approach essentially utilizes performance testing to balance an asphalt mixture’s resistance to rutting and cracking.
The mix design process is not only about balancing resistance to distresses; long-term sustainability must also be considered. Agencies want to use local and recycled materials to the greatest extent possible without sacrificing performance. Therefore, it is important to not only consider RAP and RAS usage in the BMD process but also how to use it without diminishing mix durability. Long-term performance is a key element in holistic pavement sustainability.
Another new mix design concept that is growing rapidly in popularity is recycled binder availability (RBA). RBA recognizes that the binder in the RAP and RAS is significantly stiffer than the new binder being added, and as a result, not all of it is activated or blended with the new binder.
Both BMD and RBA are modifications to conventional Superpave mix design methods and both are gaining traction in our industry due to the need to improve the durability in our mixes. This article reflects the current collective critical thinking of our Asphalt Institute staff engineers regarding both BMD and RBA and how they fit together.
BMD overview
“AASHTO PP 105, Balanced Design of Asphalt Mixtures” defines BMD as “asphalt mix design using performance tests on appropriately conditioned specimens that address multiple modes of distress taking into consideration mix aging, traffic, climate and location within the pavement structure.”
Four different approaches to BMD that are in AASHTO PP 105 are summarized as follows:
• Approach A – Volumetric Design with Performance Verification consists of using existing volumetric mix design methods to select an optimum binder content (OBC). The resulting mix is then tested with selected performance tests. If the mix passes the performance tests, the design is complete. If not, the mix design process is repeated with new materials and combinations until all volumetric and performance criteria are satisfied.
• Approach B – Volumetric Design with Performance Optimization uses the existing volumetric mix design to determine a preliminary OBC. Performance tests are then conducted at the OBC and two or more additional binder contents. The binder content that satisfies all performance criteria is now designated as the design OBC. This approach is slightly more flexible than Approach A.
• Approach C – Performance-Modified Volumetric Design also begins with the volumetric mix design. The performance test results are then used to adjust either the preliminary asphalt binder content or mixture component properties and proportions. The resulting volumetrics may not meet the initial criteria. This approach is more flexible than Approach A and Approach B.
• Approach D – Performance Design establishes and adjusts mix components and proportions with limited or no reliance on volumetric parameters. Minimum requirements may still be set for binder and aggregate properties and mix volumetrics may still be checked in production. This is the most flexible approach, but also deviates the most from historically proven volumetric mix design methods.
Thoughts on BMD
BMD is a positive step forward in the pursuit of better-performing mixtures, allowing for the use of innovative materials and methods. Rutting, cracking, and other pavement distresses can be specifically targeted by the performance tests selected by an agency within their BMD system.
While BMD can move our industry forward, there are critical questions that remain around implementation, especially the cracking tests and validation of those tests with in-service performance. Volumetric mix design has served our industry well for many years and we should not be quick to throw out air voids, VMA and VFA until we are extremely confident in the performance tests chosen for all materials, mixes and applications. BMD is only as good as the performance tests utilized, which must accurately reflect field performance. Guardrails are still needed and volumetrics provides that. These are the reasons why most state agencies currently are reluctant to move all the way to Approach D, which disregards volumetrics.
Some of the important questions that remain around implementing BMD are:
• What is the right cracking test? Tests that agencies are considering include the IDEAL-CT, I-FIT, Texas Overlay Tester and the AMPT direct tension cyclic fatigue test. It’s important to understand that certain cracking tests are better at targeting certain types of cracking mechanisms.
• What criteria best ensure good field performance? The Consortium for Asphalt Pavement Research and Implementation (CAPRI) created an excellent document on this topic entitled “Guidelines and Recommendations for Field Validation of Test Criteria for BMD Implementation.”
• Does the test and criteria change based on mix type, binder grade, geographic location, etc.?
• Do the cracking tests we choose trend in the correct direction regarding using polymer-modified asphalt and air voids? In other words, do the things that we know improve cracking performance also improve the results of the cracking tests we choose? Unfortunately, the answer is sometimes no. As an example, Kentucky Transportation Cabinet (KYTC) recently adopted IDEAL-CT as their BMD cracking test, which happens to be the most common among state agencies. The minimum cracking index chosen for mixes with their polymer modified PG 76-22 is lower (less stringent) than for mixes with their unmodified PG 64-22. KYTC set their criteria after benchmarking mixes and seeing their polymer modified PG 76-22 mixes were producing lower CT indexes and would have difficulty meeting the higher number chosen for unmodified PG 64-22 mixes. This is counterintuitive as we know PMA typically improves cracking performance. Lab test results should reflect performance in the field.
• Cracking is a mid- or long-term distress in the field, yet agencies generally utilize their cracking test on mix samples that have only received short-term aging. How can we appropriately age samples to evaluate cracking resistance best?
• How exactly does the optimum binder content (OBC) get selected within the BMD process? If there is a range of binder contents that meet performance testing and volumetric criteria, how is the actual OBC selected?
Asphalt Institute Senior Regional Engineer Greg Harder provided some additional detail in his article in the 2023 summer edition of Asphalt magazine entitled “Balanced mix design: The rest of the story.”
Using empirical mixture testing as part of a mix design procedure is not new. We had Marshall Stability and Flow in the Marshall mix design method, and Hveem Stability in the Hveem method. Don’t forget about Superpave Level 3 with its mechanistic property tests, before the industry decided those tests were not ready to be incorporated into our industry’s standard mix design procedure.
The point is – it’s not having a mix performance test that moves our industry’s mix design procedure forward, but rather it’s the quality of that test in terms of predicting performance. For BMD, that means we need to utilize the best cracking test with the right criteria that best relates to performance under the applicable conditions and with the materials utilized in the mix.
The good news is that there is an abundance of research and other activities happening at the national and state levels to answer these critical questions. A BMD Implementation Working Group consisting of industry experts has recently been formed. The AIEI (Asphalt, Innovate, Enlighten, Implement) program with the Federal Highway Administration (FHWA) offers BMD onsite workshops and peer exchanges to state DOTs.
When using RAP, the RBA concept makes good sense
The Asphalt Institute supports using high RAP mixtures, provided they have been properly engineered to meet expected performance without durability issues. Dr. Grover Allen’s article in the spring 2024 edition of Asphalt magazine entitled: “Is RAP just black rock?” laid out an explanation of the RBA (recycled binder availability) concept and the volume of research covering the topic. RBA was defined by Dr. Amy Epps Martin (2021) as “the amount of recycled asphalt binder from RAP/RAS that activates and contributes to the total effective binder content in an asphalt mixture.”
Without using RBA, the default assumption is that the RAP binder is 100 percent available for blending, even though the research indicates the percentage is less than that. The “black rock” theory assumes the other extreme, where none (0%) of the RAP binder is available to coat the aggregate or to contribute towards effective binder in the mix. Dr. Allen explains that the body of research around this topic converges on “partial black rock” or an RBA factor for RAP probably being between 60 and 80 percent. The percentage depends on several factors, including mixing conditions, mixing temperature and recycled binder stiffness.
The logical conclusion is that we are under-asphalting our asphalt mixtures when we don’t use RBA (a factor less than 100 percent). The table below shows the degree to which mixes are presumed to be under-asphalted as a factor of both the % of RAP used and the assumed RBA factor.
When we incorrectly assume that 100 percent of the RAP binder contributes as an effective binder, we incorrectly assume the mix is not under-asphalted. RBA is an engineering approach that adjusts our conventional mix design process to ensure we are adding sufficient virgin binder, acknowledging that not all the binder in RAP or RAS is effective.
Since RBA is a volumetric concept, it fits well with BMD approaches A, B and C because they all begin with volumetric mix design. If BMD approach D is used, which strictly relies on performance test results, the use of RBA would be unnecessary.
Many states are paying attention. At the 2023 Southeast Asphalt User/Producer Group (SEAUPG) annual meeting there were multiple sessions and a panel discussion on RBA. The feedback from panel members was greatly in favor of utilizing RBA. Several DOTs have already instituted such a policy, with many others discussing the idea, as shown in the updated map from NCHRP project 09-58. Utilizing RBA in the mix design process should increase pavement durability and service life with minimal impact on the life cycle cost of the pavement. According to recently completed research by NCAT, it only takes about two months of additional service life to justify the cost of an additional 0.2 percent increase in virgin binder.
Other ways to increase durability when using high RAP contents
RBA is not the only way to properly engineer asphalt mixtures that incorporate high RAP contents. This section briefly discusses some of the other ways.
Recycling agents (RA), which are both rejuvenators and softeners, are additives with chemical and physical characteristics designed to restore at least some of the rheological properties of aged binders in recycled asphalt materials. The binder in the RAP, and especially RAS, is generally much more brittle than virgin binder. Basically, RAs seek to soften and at times rejuvenate the brittle RAP/RAS binder to reduce the cracking susceptibility in mixes with high recycled binder ratios (RBR). RAs also have the potential to increase RBA, since RAs typically soften RAP binder, and RBA is inversely correlated to RAP binder stiffness. NCHRP Report 927 is an excellent resource on the topic.
Increasing Pbe, the percent effective binder, in asphalt mixtures provides more of the softer virgin binder needed to mitigate the negative influence of stiffer RAP/RAS binders. Agencies have sought to increase Pbe in several ways.
At first, agencies began to lower the number of design gyrations in Superpave mixtures. This is effective if nothing else changes in the aggregate structure. However, the aggregate structure can be reproportioned in such a way that a lower number of gyrations does not increase Pbe.
VMA is the space between compacted aggregate particles. It is filled with Pbe and air. If the air void requirement remains the same, an increased VMA would necessarily increase Pbe. However, VMA is a calculated property, not a measured property. It is based on the combined aggregate bulk specific gravity (Gsb) – a property that is often difficult to consistently determine. If an agency does not verify designer-supplied Gsb values, the Gsb could be reported artificially high which would result in a higher calculated VMA than actually exists.
The regressed air voids concept calculates the optimum binder content at the 4.0 percent air void content specified in “AASHTO M 323, Superpave Volumetric Mix Design.” It then locks in the aggregate structure and assigns the binder content needed to reach a lower air void content, e.g. 3.0 percent. Wisconsin DOT uses regressed air voids. One rule of thumb suggests that for every 1.0 percent decrease in air voids, the binder content increases about 0.3 percent.
Polymer-modified binders used in asphalt mixtures have been demonstrated to result in both less cracking and less rutting (AI’s “IS-215 Quantifying the Effects of Polymer Modified Asphalt (PMA)”.) The referenced document concluded that the use of PMA reduced cracking in the surface and binder courses, on average, by about half, extending the pavement service life.
Softer binders are often used in accordance with Table 2 in AASHTO M 323. It specifies that no softer binder is needed if the RAP percentage is less than 15 percent. Between 15 and 25 percent, the virgin binder grade is dropped by one (both PGH and PGL, e.g. a mix using PG 64-22 and 20% RAP would instead use PG 58-28). For above 25 percent RAP, a blending chart is required, but mix designers need to remember to adjust the RAP binder percentage to account for the RBA.
Summary
The Asphalt Institute supports the BMD concept. As with any new concept that will be used on a multibillion-dollar asset like the U.S. national road network, we must take time to carefully implement and evaluate the ramifications prior to wholesale changes. Critical thinking is a healthy thing. Volumetric mix design has been successfully used for many decades and should not be phased out completely until the knowledge gaps with BMD are filled. BMD is structured so that most approaches still include some measure of volumetric analysis. Approach D is a great target to work toward, but the risk is high until several questions are conclusively answered.
We support a national move towards higher RAP in asphalt mixtures with an eye toward increased sustainability. Yet, we cannot forget pavements that last longer require less frequent usage of natural resources, a critical concept of cradle-to-grave sustainability. The research overwhelmingly tells us that not all the binder in RAP is activated and acts as an effective binder, so we should continue implementing the concept of RBA in our mix design procedures. To do so, each agency will need to have its RAP stockpiles evaluated to determine a reasonable RBA factor for its materials and climate.
The Asphalt Institute’s mission remains to promote the safe use, benefits and quality performance of petroleum asphalts in a unified voice for our members. That means taking a methodical engineering approach toward BMD and utilizing tools we have such as RBA that allow us to optimize RAP in mixtures without compromising durability.
Buncher is the Asphalt Institute Director of Engineering. Gierhart is the Asphalt Institute Deputy Director of Engineering. All Asphalt Institute Regional Engineers and Asphalt Institute Director of Research and Laboratory Services Mike Anderson contributed to this article.
