A Growing Need

By: Dale Decker, P.E.

The use of the Superpave design process for hot-mix asphalt (HMA) mixtures is growing across the U.S. In some areas of the country contractors have experienced problems, but there have been many projects that have been constructed successfully. In many cases, modifications to the original Superpave specifications and guidelines have been necessary in order to accomplish the goal of building a high-quality, long-lasting HMA pavement.

The variety of aggregate product sizes that are being required for the production of Superpave-designed mixtures has, in some cases, increased substantially. Some aggregate suppliers have expressed concern that it may be necessary to make modifications to the aggregate specifications in order to avoid producing excesses of particular aggregate sizes. This concern is highly dependent on the aggregate source and production capabilities.

Some mix designers are using intermediate coarse aggregate sizes to "sweeten" the mix in order to achieve volumetric properties. By doing this, other sized material is being wasted to get this single-sized fraction. Few operations in the U.S. currently have fractionation capabilities. It will take some time for this ability to be developed.

The Superpave system specifies the determination of percent flat and elongated particles for the aggregate blend. While the requirement is specifically stated to be on the aggregate blend as a whole, some agencies are incorrectly interpreting the requirement to mean that the aggregate producer must meet the requirement for each aggregate product in order to satisfy the blend requirements. The correct interpretation is that if one of the aggregate products does not meet the flat and elongated requirement, the product would be limited in its quantity of use to ensure that the aggregate blend meets the specifications. This means that if one of the aggregate products does not meet the flat and elongated requirements, it can still be used if the blend does meet the specified percentage.

Superpave requires that no more than 10% of the aggregate particles will have a flat and elongated ratio of 5:1. A change proposed by the FHWA Expert Task Group to the specification would require no more than 20% at a 3:1 ratio. In fact, there is no data to support either requirement. The 5:1 ratio eliminates almost no aggregate products, while the 3:1 eliminates a high percentage of materials that are known to perform well. Significant research efforts are under way to attempt to clarify this requirement and to provide appropriate recommendations. No changes should be made to the existing 5:1 specification until appropriate data justify a change in the requirement.

The use of volumetric properties to control the production of HMA has been fully endorsed by the Superpave system. While not new in some areas of the country, the use of volumetric controls requires that some existing practices may have to be re-evaluated. One such issue is the determination of the aggregate bulk specific gravity. Calculation of the mix volumetric properties can be greatly affected by variations of as little as 0.01 in the specific gravity of the aggregate.

The frequency of performing specific gravity tests is predicated on the variability of the aggregate source. In some areas, specific gravity values are established for the construction season and are not changed. This approach is inappropriate in an environment of volumetric control of mixtures for contractor payment. If problems achieving VMA are encountered, the bulk specific gravity of the aggregate should be one of the first parameters checked.

The Superpave system includes a totally new binder grading system that is based on the properties of the binder at both high and low temperatures. The system works quite well for "neat," or unmodified, asphalt binder systems. However, the use of modified binders is increasing, and the grading system was never designed for use with these modified asphalt systems. A specification appropriate to modified binder systems is being developed but will not be available for the construction industry for probably another two years.

It is important to note that not all binders of the same PG grade will perform the same. Depending on the crude source, the presence of additives/modifiers and the refinery processing techniques used, two binders of the same PG grade may respond very differently in the contractor’s operations. This forces the contractor to exercise particular care in blending or mixing of binders that may be of the same grade but from different binder suppliers. Since HMA plants generally have only two binder storage tanks, this issue can cause serious logistical problems for the contractor.

Before the Superpave system came into use, the determination of mixing and compaction temperatures was based on the evaluation of the viscosity properties of the asphalt cement. With viscosity-graded "neat" binders, this process worked reasonably well. However, with PG binders, particularly modified materials, the conventional process does not work well because of the increased stiffness of the materials. In addition, binders of the same grade may have different mixing and compaction temperatures if different suppliers provide the binder.

The Superpave system has been promoted as a national, standardized set of test procedures and specifications. However, vastly different approaches to the "standards" are evolving. In order to address specific agency needs and experience, variations in the Superpave system are being introduced by many states.

"Standard" test procedures should not be modified by agencies without national agreement.

Having said that, individual regions of the country have both the right and responsibility to vary the test criteria or specifications to accommodate the materials and processes that are available locally. However, this does not preclude the use of a nationally agreed upon set of standard test procedures. The local industry (public and private) should collectively agree on what the test criteria (i.e., test specification values) should be for the results of the standard test methods.

Recent changes to the N-design table have greatly simplified the process for selection of the number of gyrations in the Superpave gyratory compactor for the mixture. Likewise, the use of N-design rather than N-maximum has made the design process simpler. However, designers are finding that these changes may cause variations in the volumetric properties of the Superpave-designed mixes.

The major missing link at this time is an appropriate N-design value for low volume roads that account for the predominance of the 2.4 million miles of paved roads in the U.S. In the meantime, many mix designers are "backing" into design density and air void values for mixes that have good performance history.

A major gap in the Superpave mixture design technology is the lack of a practical, rapid performance-related mixture test to evaluate the mechanical properties of HMA. Although much work has been done, and is being done, there is still no universally accepted procedure for the performance evaluation of mixtures in the laboratory or in the field.

There is a variety of tests currently in use that will provide some measure of the performance of the mixture. Current research is focusing on the issue and results are expected in the next two to three years. But what can the contractor do today (for the 550 million tons of HMA to be produced this year) to get some sense of mixture performance?

As you would expect, many people are relying on tests with which they have experience. Rut testers, the static creep test and Hveem and Marshall stabilometers have all been used to provide an indication of mixture strength. With any test, it is important to note that appropriate validation of the laboratory test values to known field performance is critical in order to make sound decisions about materials engineering.

Depending on how HMA samples are stored and reheated, the results of the tests for volumetric properties can be dramatically different. A standard test procedure for storing and reheating samples of HMA needs to be developed and published as an AASHTO and/or ASTM test method.

An increase in the percentage of aggregate passing the 0.075 mm (#200) sieve can be particularly troublesome in maintaining the volumetric properties of the mixture. Often there is a generation of fines as the material passes through the hot-mix plant. One way to overcome this problem is to incorporate the baghouse fines in the mix design process.

One of the key differences between conventional mixes and coarse-graded Superpave-designed HMA is the increase in film thickness on the aggregate particles. With an increase in the percentage of coarse aggregate particles, the mixtures require less asphalt binder to properly coat the particles. With some materials, this change can have a significant impact on the volumetric properties of the mixture. In the evaluation process, it is important to recognize this change. The appearance of the mix may be very different (richer looking) compared with the traditional mixes that have been produced. Mixture adjustments should be made based on mixture properties and not on appearance alone.

Reclaimed asphalt pavement (RAP) has been successfully used for decades in HMA. Properly crushed and handled RAP materials have been shown to have less variability than virgin aggregates. Experience has shown that HMA can be successfully produced using RAP and that the resulting pavement can be expected to perform as well as, or better than, mixes made with all-virgin materials.

However, some agencies severely limit or exclude RAP in a Superpave-designed mixture. The principal change from a Marshall-designed mixture to a Superpave-designed mixture is the compaction method. Gradation specifications also have changed, however, and in some cases this development has greatly restricted the use of RAP. With the tremendous economic and environmental benefits to using RAP and the proven record of performance for HMA containing RAP, there is no justification for specifications that prohibit or severely limit the use of RAP.

RAP should be "engineered" into the Superpave mix design process, just like any other material. Generally, it is accepted that for less than 10% RAP, little if any change to the binder properties occurs. A recent NCHRP project has shown that for less than 15% RAP, no change in PG binder grade is required. For 10-20% RAP, the binder grade is usually dropped one grade. For RAP percentages greater than 20%, the effect of the RAP on the binder should be established in the mix design process. The process control would then consist of monitoring the RAP consistency and the consistency of the virgin binder. This process control can be accomplished using the ignition oven without having to use and dispose of expensive solvents.

Over 4 million tons of stone matrix asphalt (SMA) has been placed across the U.S. It has proven to be a very successful mix type for heavy traffic pavements. Performance of the mix has been excellent in all properly constructed applications.

The Maryland State Highway Administration has developed a specification for Gap-graded Superpave that simply translates SMA into the Superpave mix design system. This approach recognizes the quality and performance of the SMA and at the same time takes advantage of the Superpave mix design process.

A similar approach may be used to include other mix types in the Superpave mix design process. If it’s hot, black and sticky, the mix design should utilize the Superpave system. In order to avoid confusion, the contracting community wants only one mix design process. That mix design process must be flexible enough to accommodate all types of HMA mixes, from interstate highways to low-volume roads.

Experience has shown that coarse- and fine-graded Superpave-designed mixtures behave very differently. It is critical that local experience with aggregate materials dictate the size of materials used. Successful mixtures can be produced using materials above, below and through the restricted zone. For this reason, the restricted zone should be eliminated from the AASHTO specification. While the current wording indicates that the zone is a guideline not a specification, many agencies will not allow the contractor’s mix design to pass through the restricted zone. The fine aggregate angularity (FAA) and N-initial requirements should successfully identify and control the quantity of dirty, rounded natural sands that the restricted zone was designed to eliminate. Having all three requirements adds difficulties to the mix design process with no indication of benefit.

Aggregate particle size can have a significant effect on the density achieved if the maximum particle size is too large for the lift thickness of mix being placed. Furthermore, requiring high levels of density may not be the best solution (i.e., increasing the density requirement from 92% to 93%).

The NAPA Superpave Construction Guidelines recommend that a minimum lift thickness of three times the nominal maximum aggregate size be used. While most agencies verbally accept that recommendation, there continue to be instances where the lift thickness violates that size relationship. This results in extreme difficulties in achieving the desired density.

The change in definition of aggregate size to nominal maximum aggregate size (NMAS) has resulted in more coarse aggregate being retained for each aggregate size.

If a given lift thickness is desired, one solution is to reduce the maximum aggregate size in the mixture used. This, however, is not as easy as it may sound. Aggregate production is geared to specific sizes. Switching from 12.5 mm to 9.5 mm aggregate size can cause problems in the production operation. There have been numerous examples in which the lift thickness was increased by as little as 12.5 mm (1/2 in.) and difficulties in achieving density were eliminated. Clearly, meeting the 3:1 requirement is the most logical approach to solving this particular density issue.

Some contractors continue to experience the so-called "tender zone" during mixture placement. The tender zone is a temperature regime between 60-125û C in which the mix is very tender and cannot be adequately compacted. This condition is not true for all mixtures, but it has been observed for some Superpave-designed mixtures. The occurrence of the tender zone seems to be more prevalent with coarse-graded mixtures and may be affected by lift thickness, filler/ asphalt ratio, PG binder grade and binder properties.

The type and number of rollers required to compact a mix that is experiencing the tender zone are a function of the mix design and must be determined for each specific project. No specific type of roller has been found to be better or worse for the compaction of Superpave-designed mixtures. Therefore, specifications should not require specific roller types or restrict contractor ingenuity.

One of the biggest changes affecting the HMA producer in the last few years has been the implementation of quality control systems to control the production process. The systems include testing to control the process as well as certification of both people and materials in the production process.

The contractor currently does most mix designs in the U.S. Thus, the contractor must have qualified people to accomplish this work.

Federal regulations, effective July 2000, require technician qualification. This process is good, but it is important that hands-on proficiency evaluation is included in the process. The biggest problem in this arena is the lack of reciprocity between governmental agencies in the qualification and acceptance of test results.

It is imperative that test procedures be standardized on a national basis so that technicians can be trained once and accepted by all in the industry as qualified personnel. Local agencies have the right and responsibility to adjust specification criteria to meet the needs of local materials and construction processes.

The quality control process is moving forward quite rapidly in many areas. Various approaches to setting up programs have been developed across the country. The systems that seem to be working best are those in which the agency and industry developed the program in a joint effort. Generally, the approaches should include such things as the taking and retention of binder and aggregate samples, frequency of sampling, mix property tolerances and acceptance criteria.

The use of the Superpave system has raised questions regarding the volumetric tolerances placed on the mixture. The tolerances proposed were developed using the Marshall mix design process and may need to be adjusted to account for the change in the compaction process. Typically, the particle orientation of mixtures compacted in the Superpave gyratory compactor is different from that obtained from Marshall-compacted specimens. While the SGC does provide a more rational approach to mixture compaction, there may be a need to adjust the volumetric criteria to account for the change in compaction and the precision and bias of the procedures.

Overall, given the dramatic changes, mixes designed with Superpave have not presented insurmountable problems to the contractor. The Superpave system will provide the HMA industry with tools that will enable us to continually improve hot-mix asphalt pavements. The key issues that affect the contractor are:

• National standardized test procedures must be enforced to maintain a coherent system between the states;

• Local engineering judgment and expertise must be used to make adjustments to the system to make Superpave work with available materials; and

• Designs for low-volume roads and commercial applications must be developed based on past experience with materials and construction practices.