Asphalt Materials and Mix Design

3.2.1 Penetration Grading Systems

The penetration of an asphalt binder (AASHTO T 49, ASTM D946) is the penetration of a weighted needle into a binder sample measured in units of decimillimeters (0.1 mm) at 25 °C (77 °F). Asphalt mixtures with stiffer binders (i.e., with a lower penetration) will be stiffer at a given temperature than mixes with softer binders (i.e., with a higher penetration). For example, at a given temperature, a mix containing binder classified as 60–70 penetration grade will be stiffer and may require more compactive effort by the rollers to achieve the desired density than will a mix containing a 120–150 penetration grade asphalt binder. This empirical test is an indicator of stiffness at one temperature, but any relationship with overall mix performance can be significantly variable.

3.2.2 Viscosity Grading Systems

Grading of asphalt binders by viscosity (resistance to flow) is defined by a viscosity measurement at 60 °C (140 °F) on the material in its original (as received from the refinery) condition (termed AC, for asphalt cement) or on a binder considered to be comparable to the material after it has passed through the high temperatures of the plant production process (termed AR, for aged residue). In the AC grading system, a mix containing an AC-20 will be stiffer than a mix containing an AC-10. Similarly, in the AR grading system, a mix containing an AR-4000 will be stiffer than one containing an AR-2000 at the same temperature. The ASTM viscosity standard is D3381, while the AASHTO viscosity standard is M 226.

3.2.3 Superpave Performance-Graded System

Grading systems based on penetration and viscosity have worked reasonably well for many years and are still used in many countries because of the simplicity and portability of the testing equipment. However, these simpler grading systems can categorize binders within the same grade even though they may exhibit very different temperature and performance characteristics in the environment in which they are used.

The PG system was developed to provide an improved set of asphalt binder specifications (AASHTO M 320, ASTM D6373). This system endeavors to measure physical properties that can be related directly to field performance by engineering principles. The PG tests are performed at loading times, temperatures, and aging conditions that more realistically represent those encountered by in-service pavements. The PG specifications help in selecting a binder grade that will limit the contribution of the binder to low-temperature cracking, permanent deformation (rutting), and fatigue cracking of the asphalt pavement within the range of climate and traffic loading found at the project site.

An important difference between the PG specifications and those based on penetration or viscosity is the overall format of the requirements. For the PG binders, the performance criteria remain constant; however, the temperatures at which those properties must be achieved vary depending on the climate in which the binder is expected to serve. The binder is graded in 6 °C increments of pavement temperature. An example of a binder designation in this system is PG 64-22. For this example, the binder is selected to resist environmental conditions in which the average 7-day maximum pavement temperature is at least 64°C (147 °F) but lower than 70 °C (158 °F). On the low-temperature side, the binder is selected to perform in pavement temperatures from −22 °C (−8 °F) down to just above the next grade at −28 °C (−18 °F).

The PG specifications help in selecting a binder grade that will limit the contribution of the binder to low-temperature cracking, permanent deformation (rutting), and fatigue cracking of the asphalt pavement within the range of climate and traffic loading found at the project site.

While this concept worked well for conventional speed, moderate traffic-volume pavements and airfields, research indicated that it needed some refinement for pavements that had slow-speed loading and heavier loading. Rather than change criteria and/or test conditions to reflect a change in loading time and traffic volume, the architects of the PG system elected to simply adjust for traffic speed and volume by “grade bumping,” using stiffer grades than indicated by the climate alone. This was a simple way to ensure adequate support in high-volume and/or slow-loading conditions. Requirements for grade bumping can be found in the current AASHTO M 323 for roadways, and for airfields, the requirements can be found in the DoD UFGS 32 12 15.13 and FAA P-401/P-403 specifications.

3.2.4 Multiple Stress Creep Recovery Grading System

The MSCR specification (AASHTO M 332, ASTM D7405) endeavors to solve issues with the Superpave PG system, improving the way the high-temperature behavior of polymer-modified asphalts is addressed and removing the need to grade bump. It uses the creep and recovery test concept to evaluate the asphalt binder’s potential for permanent deformation. Using the dynamic shear rheometer, a 1-s creep load is applied to the rolling thin-film oven-aged asphalt binder sample. After the 1-s load is removed, the sample is allowed to recover for 9 s. The test begins with the application of a low stress (0.1 kPa) for 10 creep/recovery cycles, and then the stress is increased to 3.2 kPa and repeated for an additional 10 cycles.

In the MSCR test, two separate parameters can be determined during each loading cycle: non-recoverable creep compliance (J_nr) and percentage of recovery (MSCR Recovery). Figure 4 shows a typical result from the MSCR test. The test specimens are creep loaded at 0.1 kPa and 3.2 kPa. After each 1-s creep load, the binder is allowed to recover for 9  s. J_nr is a measure of the residual strain left in the specimen after repeated creep loading and recovery, relative to the amount of stress applied. This parameter has been shown to be better correlated with rutting potential than the G*/sin δ parameter used in the Superpave system.

Source: U.S. Department of Transportation
Figure 4. MSCR Stress and Strain Responses

Figure 5 shows the acceptance criteria of AASHTO R 92 for MSCR Recovery. After the average amount of recovery is calculated, the results are used in combination with Jnr to indicate whether a binder has a significant elastic component.

Source: AASHTO (AASHTO R 92, Figure 1)
Figure 5. MSCR Recovery Acceptance Curve

Unlike the AASHTO M320 system, the test temperature used for the MSCR test is selected based on actual high pavement temperatures with no grade bumping. For example, if a binder grade would need to perform in an environment with average high pavement temperatures of 64 °C and low pavement temperatures reaching −22 °C, the MSCR test would be performed on the binder at a high temperature of 64 °C regardless of the traffic speed and loading. Higher loading is accounted for by increasing the stiffness (reducing the compliance) required for the asphalt binder at the grade temperature.

Unlike the AASHTO M320 system, the test temperature used for the MSCR test is selected based on actual high pavement temperatures with no grade bumping.

The designations shown in Table 1 are based on information found in AASHTO M 332.

Table 1. MSCR Designations

Source: AASHTO
km/h = kilometers per hour Source: AASHTO
Note: Grade bumping is accomplished by using “H,” “V,” or “E” designations and not by increasing the PG hightemperature grade as recommended in AASHTO M 320.

For standard traffic loading, J_nr (determined at 3.2 kPa shear stress) is required to have a maximum value of 4.5 kPa⁻¹. Continuing the example, the subsequent grade would then be a PG 64S-22. As traffic increases to heavy and very heavy loading, the J_nr of the asphalt binder needs to be lower—allowing maximum values of 2.0 and 1.0 kPa⁻¹, resulting in binder grades of PG 64H-22 and PG 64V-22. For extremely heavy traffic loading, the J_nr of the asphalt binder could only have a maximum non-recoverable creep compliance of 0.5 kPa⁻¹, resulting in a PG 64E-22.

Although the MSCR grading system is specified by many State DOTs, at the time of this writing it is not specified for airfields, neither in the DoD UFGS 32 12 15.13 nor the FAA P-401/P-403 specifications.

3.2.5 Temperature-Viscosity Characteristics

The viscosity of an asphalt binder changes with temperature, with lower temperatures resulting in higher viscosity (higher “stiffness”). This temperature-viscosity relationship impacts several aspects of asphalt testing, production, placement, and compaction.

During lab testing of both asphalt binder and asphalt mixtures, the testing temperature is always specified and must be tightly controlled. Without this control, test results lose their meaning. For example, the test temperature for the Hamburg Wheel-Tracking Test might be 50±1°C for a particular agency. If the actual test temperature is lower, then the binder will be stiffer and the results will exhibit artificially low rutting. If the temperature is allowed to wander higher than the test temperature, the rutting results would falsely indicate that the mix is more rut-susceptible than it actually is.

During production, the binder viscosity must be low enough that the binder can be pumped and handled in the facility. Both PG and MSCR specifications require that the binder viscosity be lower than 3 Pa·s at 135 °C.

During mix placement, viscosity has a large influence on the workability of the mixture. Hand work—when an asphalt crew laborer is placing the mixture with a shovel or an asphalt rake (lute)—becomes increasingly difficult as the viscosity of the binder increases. The mix flows through the paver more easily when the binder viscosity is lower.

The viscosity of an asphalt binder changes with temperature, with lower temperatures resulting in higher viscosity (higher “stiffness”). This temperature-viscosity relationship impacts several aspects of asphalt testing, production, placement, and compaction.

During compaction, the total time available to achieve proper mat density is influenced by the viscosity of the binder. As the binder becomes stiffer or more viscous as the mixture cools, a greater compactive effort is required to achieve a given density. The binder viscosity is not only affected by the inherent binder properties, but also the thickness of the compacted layer and environmental conditions during construction. Thinner layers will cool more quickly, significantly affecting the allowable time for compaction. Cooler, windier weather can drastically shorten the time available to achieve proper compaction or even make it impossible to achieve.

PaveCool and MultiCool are public domain applications that estimate the time available for compaction operations based on a variety of user-input variables. Figure 6 and Figure 7 demonstrate the effect of decreasing lift thickness on time for compaction, with all other variables held constant. Figure 6 estimates 44 min available for compaction for a 3-inch (76-mm) mat given several constants, including 50 °F (10 °C) air and existing surface temperatures and a fine/dense-graded mix with a PG 64-22 binder. Figure 7 shows that when using the same variables for a 1-inch (25-mm) mat, the time available for compaction drops to 7 min.

Source: Minnesota DOT
Figure 6. Available Compaction Time Example for a 3-inch Mat

Source: Minnesota DOT
Figure 7. Available Compaction Time Example for a 1-inch Mat

The rate of change in viscosity with change in the temperature of a binder is referred to as the binder’s temperature susceptibility. A material that is highly temperature-susceptible is one that exhibits a large change in viscosity for a small change in temperature. Multiple binders that have the same penetration at 25°C (77 °F) may not necessarily have the same viscosity at 135 °C (275 °F) since their temperature susceptibility characteristics may vary. In the lab, reporting this temperature-versus-viscosity relationship is required for some mix design procedures since lab mixing and compaction temperatures are typically based on a prescribed viscosity level.

The rate of change in viscosity with change in the temperature of a binder is referred to as the binder’s temperature susceptibility.

3.2.6 Polymer Modification

Polymer-modified asphalts (PMAs) work to flatten the temperature-viscosity relationship to make the binders less susceptible to changes in temperature. One way in which polymers are classified is based on their physical properties. Depending on their behavior when stretched with sufficient force, polymers are classified as plastomers (plastics) or elastomers (elastics). When stretched, plastomers will yield and remain in their stretched position when the load is released. Elastomers will yield under load (stretch) but will return to their original shape when the load is released. Most polyolefins behave as plastomers, while styrene-butadiene copolymers behave as elastomers.

When blended into asphalt, polymers tend to behave in two different ways. If the polymer forms discrete particles in the asphalt binder, then it functions primarily as a thickener or filler. This increases the viscosity of the asphalt binder while having no significant effect on low-temperature properties. If the polymer forms a continuous network in the asphalt binder, it functions as a homogeneous blend. This blend may impart some of the physical characteristics of the polymer to the binder, which may affect both the high- and low-temperature properties of the asphalt binder.

The task of the designer is to determine whether the extra cost of using PMA mixtures is worth the anticipated extra performance. For example, using a PMA mixture on a walking trail in a park may be an unnecessary extra expenditure. The use of a PMA mixture on a high-volume taxiway or roadway may easily be a better investment.

3.3.1 Particle Size Distribution (Gradation)

One of the important properties of aggregates for use in pavements is the distribution of particle sizes, or gradation. Gradation test methods are specified as either dry (ASTM C136/AASHTO T 27) or washed (ASTM C117/AASHTO T 11). For asphalt mix designs, the aggregate must always be washed to properly evaluate the percentage of fine particles.

Aggregate gradations having different maximum particle sizes will result in asphalt mixtures with different characteristics. Unfortunately, different specifications may have slightly different definitions for maximum particle size. The Superpave method uses the following aggregate size definitions:

• Maximum aggregate size—one sieve size larger than the nominal maximum size.
• Nominal maximum aggregate size (NMAS)—one sieve size larger than the first sieve to retain more than 10 percent by weight.

Gradation is generally controlled by specifications that define the distribution of particle sizes. AASHTO M 323 specifies Superpave aggregate gradation control points based on NMAS. Airfield asphalt mix gradation requirements are shown in FAA’s P-401 and DoD’s UFGS 32 12 15.13. They are grouped from coarsest to finest by Gradation 1, Gradation 2, and Gradation 3. The grading chart in Figure 8 represents a visual way of displaying aggregate gradations—the 0.45 power plot. The abscissa is particle size plotted to a 0.45 power scale, while the ordinate is usually the percent by weight passing a given size on an arithmetic scale.

Source: Asphalt Institute
Figure 8. Gradation Chart, Exponential Scale (0.45)

On this chart, the maximum density grading (i.e., the tightest possible particle packing) corresponds to a straight line drawn from the origin to the selected maximum particle size. This line on the .45 power chart is known as the MDL. The MDL shown in Figure 8 represents the maximum density gradation for an aggregate with a 1.0-inch (25.0-mm) maximum size. It must be noted that the MDL is approximate but can serve as a useful reference in proportioning aggregates. The actual gradation at which maximum density occurs, for any particular aggregate in a compacted asphalt mixture, is greatly influenced by the shape, strength, and surface texture of the aggregates, and it may not fall on the MDL.

The particle size distribution delineates the general type of aggregate structure in the asphalt mixture in two ways. First, the general mix type (dense-graded, open-graded, or gap-graded) is defined by the gradation. Second, the gradation also shows the NMAS of the mix. The NMAS of the mix to be used is usually related to its location in the pavement structure.

The most common type of asphalt mixture by far is the dense-graded mix. These mixes have an aggregate gradation somewhat evenly distributed throughout the entire range of sizes used. They can be used in new construction and in overlays; in base, intermediate, and surface layers; and on roadways and airfields. They can be coarse- or fine-graded, as shown in Figure 9. Fine-graded mixtures generally have gradations that plot above the MDL, while coarse-graded mixtures generally have gradations that plot below the MDL. Airfield mixes are typically fine-graded.

The most common type of asphalt mixture by far is the dense-graded mix. These mixes have an aggregate gradation somewhat evenly distributed throughout the entire range of sizes used.

The NMAS for surface mixes is generally 3/8 inch (9.5 mm) or 1/2 inch (12.5 mm), but it could be as fine as No. 4 (4.75 mm) or as coarse as 3/4 inch (19.0 mm). The choice of NMAS is often predicated on the desired surface texture, with a finer maximum size aggregate producing a smoother, tighter surface. Intermediate lifts typically use larger aggregate particles than surface mixes, while base mixes typically use 3/4-inch (19.0-mm) NMAS or larger. Dense-graded mixes should have a lift thickness of at least four times the NMAS when compacted. Lift thicknesses for fine-graded mixtures should be at least three times the NMAS.

Source: Asphalt Institute
Figure 9. Dense-Graded Mixtures Plotted on a 0.45 Power Chart

Another basic type of asphalt mixture is an open-graded mix, graphed on a 0.45 power chart in Figure 10. These mixes are mainly used as a surface lift and primarily designed for safety. Their open aggregate structure is created by the predominance of uniformly sized aggregate particles, which creates air voids around 18 to 20 percent. Open-graded mixes allow rainwater to flow through the surface mix, then move laterally across the top of the layer below into the drainage system. This design greatly reduces the risk of hydroplaning, mitigates splash and spray from tires, reduces roadway glare, and dampens roadway noise. These mixes are not good candidates for airfield paving due to the increased risk of generating FOD. Open-graded surface mixes typically have a 3/8- to 1/2-inch (9.5- to 12.5-mm) NMAS and are specified in lift thicknesses two to three times the NMAS. The NMAS versus lift thickness guidance for dense-graded mixes does not apply to open-graded mixes because the mixes are merely bonded and securely seated to the underlying layer to preserve air void space. Overcompaction of these mixes would defeat their free-draining design purpose. Although traffic helps keep the mix from clogging, its permeability will continually reduce over time. Open-graded surfaces typically have a shorter lifespan than dense-graded mixtures.

Open-graded mixes can also be used as open-graded bases or as part of a permeable pavement. The concept of using an open-graded base (or asphalt treated permeable base, ATPB) along with a pavement edge drain system to move moisture out of the pavement structure has been around for decades. When using an open-graded base, a filter fabric must be placed on the existing subgrade or base to prevent fines migrating up and clogging the air voids in the open-graded asphalt layer. The edge drain system must be monitored to prevent crushing during construction and clogging due to rodent nesting or vegetation growth; otherwise, the pavement will hold water instead of drain water. These maintenance difficulties have resulted in fewer open-graded base systems being designed.

The concept behind porous pavements can also be applied to parking lots and low-volume residential streets for stormwater management. The typical design of this system consists of an open-graded surface mix that allows drainage through the pavement and into an underlying stone reservoir. A geotextile fabric is typically placed on the uncompacted subgrade to mitigate migration of fines up into the stone recharge bed.

These porous pavement stormwater management systems are intended to allow quick penetration of precipitation through the pavement surface and slow infiltration through soil on which the stone reservoir is built. It is commonly thought of as being environmentally friendly due to reduced runoff and potentially cooler surfaces. Losing functionality due to clogging is still a concern. However, the porous surface can be milled and replaced to renew functionality. Permeable pavement systems are more expensive to install than traditional pavements, and as a design with a reservoir for stormwater management, they are typically not suitable for highways or airport pavements.

Source: Asphalt Institute
Figure 10. Open-Graded Mixture Plotted on a 0.45 Power Chart

Gap-graded mixtures have high percentages of coarse particles and fine particles, but few intermediate-sized particles, which is reflected in Figure 11. The target air void content is typically 4 percent. SMA is the most common example of a gap-graded asphalt mixture. It is a premium mix with high rut resistance due to coarse aggregate interlock and high durability due to the mastic created by the high binder content (typically polymerized) and mineral filler. Due to the higher binder content, and fibers which are often used to prevent draindown of the binder from the aggregate structure, SMA mixes are typically more expensive than dense-graded mixes. They are almost exclusively used as surface mixes. Their NMAS is typically 3/8 to 1/2 inches (9.5 to 12.5 mm), although some agencies specify different particle sizes. At the time of this writing, SMA mixes are used exclusively on roadways.

Source: Asphalt Institute
Figure 11. Gap-Graded Mixture Plotted on a 0.45 Power Chart

3.3.2 Specific Gravity

Aggregate specific gravity is an important property used to delineate the relative density of different aggregates. The ratio of the density (mass per unit volume) of each aggregate source to the density of water at 73.4 °F (23 °C) is the dimensionless property “specific gravity.” A simple way to think of this property is “the number of times something weighs more than the equivalent volume of water.” AASHTO T 84 (ASTM C128) and T 85 (ASTM D127) are used to determine fine and coarse aggregate specific gravity, respectively. In the asphalt industry, this property is shown to three decimal places (the nearest 0.001). These same tests are also used to determine aggregate absorption (to the nearest 0.1).

The following are the most commonly used aggregate specific gravities:

• Apparent specific gravity (G_sa).
• Bulk (dry) specific gravity (G_sb).
• Effective specific gravity (G_se).

The ratio of the density (mass per unit volume) of each aggregate source to the density of water at 73.4 °F (23 °C) is the dimensionless property “specific gravity.” A simple way to think of this property is “the number of times something weighs more than the equivalent volume of water.”

Each aggregate specific gravity used in the asphalt industry uses the dry mass of the aggregate. As shown in Figure 12, the aggregate volumes used to calculate commonly used specific gravities are different. The volume used to calculate the apparent specific gravity is the smallest volume—the volume of the aggregate particle only. The volume used to calculate the bulk specific gravity is the largest volume—the volume of the aggregate particle plus the volume of the water-permeable voids in the aggregate particle. The volume used to calculate the effective specific gravity is in between—the volume of the aggregate particle plus the volume of the water-permeable voids in the aggregate particle, minus the asphalt-permeable voids. Therefore, G_sa is always a larger number than G_se, which is always a larger number than G_sb.

Source: Asphalt Institute
Figure 12. Aggregate Specific Gravities

3.3.3 Surface Texture and Shape

The aggregate’s surface texture is an important factor contributing to its frictional resistance. This characteristic also strongly influences the resistance of a mix to rutting. The rougher the texture of the aggregate, the better the rutting resistance of the mix will be.

The microtexture of fine and coarse aggregates has a significant effect on the skid resistance of asphalt pavements. Pavement microtexture and macrotexture, geology to resist erosion from acid rain, and aggregate polish resistance all contribute to skid resistance and are therefore important safety considerations. Most agency specifications require aggregates with one or more of these characteristics to be used in the surface lift.

The shape of the aggregate also influences the rutting resistance of a mix, with angular aggregate producing greater resistance than more rounded material. As with surface texture, the more angular the aggregate, the greater the compaction effort that will be required to produce a mix with a specified degree of density. When properly compacted, however, the resulting pavement will be more rut-resistant.

Three of the four Superpave consensus properties deal with particle shape: the coarse aggregate angularity test (ASTM D5821), the fine aggregate angularity test (AASHTO T 304, ASTM C1252), and the flat and elongated particles test (ASTM D4791). Generally, the acceptance criteria used for these parameters are more stringent as the amount of traffic increases and as the mix is placed closer to the pavement surface.

A particle is considered flat and elongated if the ratio of the longest dimension to the smallest dimension is greater than 5:1. Flat and elongated particles tend to break during mixing and handling, changing the properties of the aggregate skeleton. By placing a limit on the proportion of particles with these characteristics, the potential for aggregate fracture during production and construction is limited.

3.3.4 Absorption

The amount of binder that is absorbed by the aggregate can significantly affect the properties of the asphalt mixture. If the aggregate particles have high asphalt absorption, the asphalt content in the mix must be increased to compensate for binder material that is drawn into the pores of the aggregate and is therefore unavailable as part of the film thickness around those particles. This is accounted for in asphalt mix design through the use of AASHTO R 30, Mixture Conditioning of Asphalt Mixtures. In this procedure, the asphalt mixture is conditioned in the oven to allow for binder absorption before the mix is tested further.

The asphalt content in the mix must be increased to compensate for binder material that is drawn into the pores of the aggregate and is therefore unavailable as part of the film thickness around those particles.

If binder absorption is not accounted for, the resulting mix will have a lower effective binder content (the unabsorbed binder on the outside of the particles). This may not only cause the mix to be dry and stiff, but it may also facilitate raveling, which is the disintegration of a pavement surface due to the dislodgement of aggregates.

If absorptive aggregates with high moisture contents are used, challenges with drying during asphalt production may arise. If severe enough, these challenges can necessitate changes to production rate or other plant dryer settings. Not properly drying the aggregate can lead to compaction difficulties and long-term durability issues with the pavement.

3.3.5 Clay Content

Clay content, also known as sand equivalent (AASHTO T 176, ASTM D2419) is the fourth of the Superpave consensus properties. The presence of clay or plastic fines in the fine aggregate (material passing the 4.75-mm [No. 4] sieve) can have a detrimental effect on an asphalt mixture. For example, clay minerals coating aggregates can prevent asphalt binders from thoroughly bonding to the surface of aggregate particles, contributing to the loss of adhesion between the asphalt binder and aggregate and increasing the potential for water damage to the paving mixture. Higher fines will decrease the asphalt content needed to produce 4-percent air voids, resulting in lower film thickness and loss of mixture durability. Too many clay-like fines can cause check cracking in the mat and decrease mix stability.

Another measure of harmful clays and organic matter present in an aggregate is the methylene blue value (AASHTO T 330, ASTM C837). In this method, methylene blue solution is titrated in increments into distilled water containing sample material passing the 75-micron (μm) (No. 200) sieve. A small amount of water containing the sample material and titrated methylene blue is removed via a glass rod and dropped onto filter paper. When the clay fraction of the sample aggregate can no longer absorb more methylene blue, a blue ring forms on the filter paper. A high methylene blue value indicates a large amount of clay or organic material present in the sample.

3.3.6 Aggregate Toughness

Aggregate toughness would be considered a Superpave source property. The specific test procedures and critical values for source properties could not be agreed on by a national consensus. However, these properties can still be important in a given region.

Toughness tests are used to determine an aggregate’s potential for degradation (production of fines and loss of angularity) during the handling, production, and placement of HMA. The two most common toughness tests are the Los Angeles abrasion (AASHTO T 96, ASTM C131) and Micro-Deval (AASHTO T 327, ASTM D6928) tests. The Los Angeles abrasion test evaluates the potential degradation of the aggregate in the dry condition. Some aggregates might degrade differently in a dry condition versus a wet condition, so the Micro-Deval test is used to assess degradation potential in the presence of water.

3.3.7 Aggregate Soundness

Other source property tests often include soundness tests. Soundness tests (AASHTO T 104, ASTM C88) estimate the resistance of aggregates to in-service weathering, typically in areas where freezing and thawing occur. These tests simulate the absorption, freezing, and thawing action of water into the aggregate particles by immersing the aggregates in a salty sodium or magnesium sulfate solution, then repeatedly dehydrating and rehydrating them. Upon rehydration, the salts that have been absorbed into the aggregate expand to simulate the expansion of freezing water. Magnesium sulfate is more aggressive, so loss limits are typically greater than those for sodium sulfate solutions.

3.3.8 Deleterious Materials

Another source property that is often specified is a maximum allowable percentage of deleterious materials. These materials are unsuitable because they are indeed deleterious—causing harm or damage to the asphalt mixture. They are typically defined as the percent by weight of undesirable contaminants such as clay lumps, soft shale, coal, wood, or mica. The most common deleterious materials test is the clay lumps and friable particles test (AASHTO T 112, ASTM C142). Different agencies specify a wide range of maximum allowable percentages, from as little as 0.2 percent to as high as 10 percent. It should be noted that many deleterious materials have a low specific gravity, so specifications in the higher range may allow much more deleterious material in the aggregate than the agency might want.

3.3.9 Recycled Materials

The asphalt industry has a history of recycling its pavements and of using materials from other nontraditional sources within pavements. While several materials have found their way into asphalt pavements, this Handbook will only address the two most common, reclaimed asphalt pavement (RAP) and recycled asphalt shingles (RAS).

3.4.1 Asphalt Binder Content

Since only the binder on the outside of the aggregate particles is useful in binding them together, that portion of the total binder content is called the effective binder content. The remaining portion of the total binder content is the percent absorbed binder. Note that the absorption capacity of aggregate is finite, so adding more binder over and above the absorption capacity of the aggregate will not increase the percent absorbed binder.

3.4.2 Percent Air Voids

Air voids in a compacted asphalt mixture consist of the small air spaces between coated aggregate particles.

Technicians often use the terms “percent density” and “percent air voids” when discussing the same mix characteristic. When percent is expressed as a percentage of theoretical maximum specific gravity, the relationship between the two is shown in the following equations:

Percent density = 100 − percent air voids, or

Percent air voids = 100 − percent density

It is important not to confuse laboratory-molded air voids with mat core air voids. Laboratory specimens are compacted at specified rates using calibrated machine-loading in steel molds that provide consistent confinement. Mat specimens are cut from a mat that was compacted by rollers of varying types and sizes using a non-prescribed number of passes and relies on the surrounding asphalt mixture and tack coat for confinement. Laboratory air voids provide information about the quality of the asphalt mixture. Mat core air voids provide information about the quality of the compactive effort on the mat.

3.4.3 Voids in the Mineral Aggregate

The term voids in the mineral aggregate (VMA) is defined as the intergranular void space between the aggregate particles in a compacted paving mixture, expressed as a percent of the total volume. VMA represents the space that is available to accommodate the effective volume of binder (i.e., the binder not absorbed into the aggregate) plus the volume of air voids in the mix.

An asphalt mixture needs a minimum percentage of VMA to have enough volume to hold both the proper amount of air voids and the proper amount of binder. If the VMA gets too low, the mixture does not have enough void space to hold the proper amounts of effective binder and air voids.

If the mix has enough air voids but not enough binder, the asphalt film thickness is too thin and the pavement becomes less durable. If the mix has enough binder but not enough air voids, the mix becomes less stable.

3.4.4 Voids Filled with Asphalt

Voids filled with asphalt (VFA) is the percentage by volume of the VMA that is filled with the effective binder. VFA, like VMA, tends to increase as the mix becomes finer and the total aggregate surface area grows.

VFA is calculated to ensure that the effective asphalt portion of the VMA in a mix is in the proper range. If the VFA is too low, the mix is too dry and will exhibit poor durability. If the VFA is too high, the mix is too rich and may be plastic and unstable.

The acceptable range of VFA varies depending on the loading situation of the mat. Higher loading requires a lower VFA, because mixture strength and stability are more of a concern. Lower loading situations call for a mix with a higher VFA to increase asphalt pavement durability.

3.4.5 Dust Proportion

The dust-to-binder ratio of a paving mixture, sometimes referred to as the dust proportion, is the ratio of the percentage of aggregate passing the 0.075-mm sieve to the effective binder. The dust proportion property is usually calculated for dense-graded mixes only.

In general, this property addresses the workability of asphalt mixtures. A low dust proportion often results in a tender mix, which lacks cohesion and is difficult to properly compact in the field because it tends to readily deform and move laterally under the roller. A high dust proportion can lead to durability issues because of an increased aggregate surface area.

The goal of mix design is to select a unique and economical blend of aggregate and binder that will achieve a good balance of the desired properties.

3.5.1 Stability

The internal friction provided by the aggregate particles and the cohesion provided by the asphalt binder provide stability to the asphalt mixture. Inter-particle friction among the aggregate particles is related to the shape and surface texture of both the fine and coarse aggregate and the characteristics of the aggregate gradation. Cohesion results from the bonding ability and the stiffness characteristics of the asphalt binder. A proper degree of both inter-particle friction and cohesion in a mix prevents it from movement in response to the forces exerted by traffic.

Excessive amounts of rounded aggregates, such as natural sand or gravel, often lead to instability concerns such as rutting, shoving, and tenderness. Natural sands can also range significantly in fine aggregate angularity and are often restricted to a maximum of 15 percent to address these concerns. Stability increases with the use of more angular aggregate with rougher surface texture.

Using too much binder (relative to the optimum binder content determined during mix design) or too soft a binder grade (for the environmental and traffic conditions) can lead to rutting and shoving (shown in Figure 14). Stability will increase with the use of a stiffer binder and/or lowering the binder content back toward optimum.

Source: National Center for Asphalt Technology
Figure 14. Rutting Due to Unstable Surface Mix

3.5.2 Durability

The durability of an asphalt pavement is its ability to resist cracking and raveling (shown in Figure 15). Cracking and raveling are often due to aging of the binder, disintegration of the aggregate, and/or stripping of the asphalt film from the aggregate (see Figure 16). These factors are affected by in-place density, weather, traffic, and incompatibilities between the aggregate and binder.

Source: National Asphalt Pavement Association
Figure 15. Raveled Asphalt Pavement Surface

Source: Asphalt Institute
Figure 16. Stripping at the Lift Interface

Generally, during mix design the durability of a mixture is enhanced by the following:

• Sufficient binder (relative to the optimum asphalt content). Insufficient binder will lead to a dry mix that is prone to aging, premature cracking, and raveling. Binder content and total surface area of the aggregate dictate asphalt film thickness. Thick asphalt films do not age or harden as rapidly as thin ones do. It is important to note that creating room for sufficient binder content (i.e., thicker asphalt film) is a function of the VMA. As discussed in Section 3.4.3, this space is filled with air and binder. Adding more binder without increasing VMA simply replaces air void space with binder. Sufficient VMA allows room for both the proper air void content and the proper binder content.
• Sound, tough aggregate that resists disintegration under traffic loading.
• Compatible asphalt binder/aggregate combinations that help prevent moisture damage.
• Additives such as hydrated lime or liquid anti-stripping agents can be incorporated into the mix design to address stripping.
• Binder with an adequate low-temperature grade or MSCR grade, which will resist cracking as the pavement contracts in cold weather.

Figure 17 shows how a very durable fuel-resistant mix functions at an airfield. A V-22 Osprey is undergoing a “hot refuel” at Lynchburg Regional Airport in Virginia. The engines continue to run during the 35-min refueling procedure. The durability and fuel resistance of the mix mainly comes from a combination of factors working together: a modified asphalt binder (PG 82-28 or PG 88-22), high binder content with subsequent low laboratory air voids (1.5 to 3.5 percent), and a high mat density (minimum of 96 percent).

Source: Ron Corun
Figure 17. Fuel-Resistant Mix at Lynchburg Regional Airport

3.5.3 Impermeability

When an asphalt pavement is impermeable, the passage of air and water into or through the asphalt pavement is prevented or at least highly restricted. This characteristic is related to the air void content, the aggregate gradation, and the lift thickness of the compacted mixture.

Figure 18 uses data from National Center for Asphalt Technology (NCAT) Report 03-02, An Evaluation of Factors Affecting Permeability of Superpave Designed Pavements. For those agencies that specify a maximum permeability, a common upper limit of 125×10^-5 cm/s² is shown in the graph. The NCAT report concluded the following:

• Air void content of dense-graded asphalt mixtures has a significant effect on the in-place permeability of the mat. As in-place air voids decrease, impermeability increases.
• The NMAS of dense-graded asphalt mixtures has a significant effect on the in-place permeability of the mat. As the NMAS of the mix decreases, the impermeability increases.
• Laboratory tests on specimens with different thicknesses showed that impermeability increases with an increase in mat thickness.

Source: National Center for Asphalt Technology (NCAT Report 03-02)
Figure 18. Permeability Versus In-Place Air Voids by NMAS

3.5.4 Workability/Compactability

Workability describes the ease with which an asphalt mixture can be placed and compacted. Mixtures with good workability are relatively easy to place and compact; those with poor workability are difficult to place and compact. Workability is especially important when hand placement and raking (luting) around manhole covers, sharp radii, and other obstacles are required.

Several factors impact the workability of HMA, including the following:

• Temperature: As the temperature of the asphalt mix increases, its viscosity decreases, making the mix more workable. However, excessive temperatures can lead to problems such as binder degradation and increased emissions.
• Binder type: The type of asphalt binder used can significantly affect workability. Polymer-modified binders, for example, can be less workable at certain temperatures due to their increased viscosity.
• Aggregate properties: The shape, size, and gradation of the aggregate can influence the mix’s workability. Mixes with larger NMAS or with more angular aggregate shapes tend to be less workable.
• Mix design: The specific combination of binder, aggregate, and any additives in the mix design will influence workability. Certain mix designs, such as SMA or mixes with a high percentage of recycled materials, may present workability challenges.

Some factors that might make the mix less workable in the short term, such as polymer-modified binders and premium mix design types like SMA, typically improve long-term performance overall.

3.5.5 Fatigue Resistance

Fatigue resistance is the pavement’s ability to perform under repeated wheel loads (traffic) without deteriorating. Research shows that air voids, binder content, and binder condition have a significant effect on a mixture’s fatigue resistance. As the percentage of air voids in the pavement increases, either by design or inadequate compaction, fatigue resistance is drastically reduced. Likewise, a pavement containing asphalt binder that has aged significantly or has excessive amounts of RAP with no corresponding modification to the mix will have less resistance to fatigue. Specifying a polymer-modified asphalt binder can significantly improve the fatigue resistance of an asphalt mixture, as it can increase the effective binder content.

The most effective method to improve a pavement’s fatigue resistance is not to specify a higher quality mix but to increase the overall structural thickness at the pavement design phase.

3.5.6 Skid Resistance

Skid resistance is the ability of an asphalt surface to provide adequate friction to the tires for safe braking and steering, particularly when the surface is wet. For good skid resistance, the tire tread must maintain contact with the aggregate particles and not ride on a film of water trapped between the pavement surface and tire (hydroplaning). FAA airfield runways are grooved, as shown in Figure 19, to avoid hydroplaning of the aircraft. A pavement with adequate surface microtexture and macrotexture will have greater skid resistance relative to a polished surface. Microtexture consists of wavelengths of 1 µm to 0.5 mm (0.0004 to 0.02 inches), and macrotexture consists of wavelengths of 0.5 to 50 mm (0.02 to 2 inches).

Skid resistance is improved using hard, durable, and angular (crushed) aggregate with a rough surface texture. The coarse aggregates must resist polishing (smoothing) under traffic. Calcareous aggregates (limestones) polish more easily than siliceous aggregates (quartz).

Mixtures that tend to rut or bleed present serious skid-resistance problems.

Source: Gonuc
Figure 19. Grooving an Airfield Runway

3.6.1 Warm-Mix Asphalt

WMA includes both additives and plant technologies that allow asphalt mixtures to be produced at lower-than-normal temperatures while maintaining mix workability, extending compaction windows and often acting as a compaction aid.

3.6.2 Liquid Anti-Stripping Additives

Chemicals known as liquid anti-stripping (LAS) additives have been used to treat moisture susceptibility in asphalt pavements. They function by reducing the surface tension between the asphalt binder and the aggregates, promoting better adhesion. Some of the chemical classes of amines that are used as LAS additives include fatty amines, amidoamines, and imidazolines. They all contain a hydrocarbon that has similar properties to asphalt and one or more amine groups. In addition to their chemistry, LAS additives can be characterized by properties such as heat stability and storage, concentration, viscosity, and odor. LAS additives are often added at the refinery or supplier’s terminal if requested. The method and rate of incorporation during the mix design should conform to the recommendations of the binder and LAS suppliers. Regardless of the type of LAS, the actual dosage rate should be determined by testing. It is possible for an increased dosage to lower test results.

Compatibility issues often need to be resolved; certain aggregate geologies may perform better with certain LAS formulations. Most agencies maintain a qualified product list for LAS.

3.6.3 Hydrated Lime

Hydrated lime is sometimes used in asphalt mixtures, primarily to improve resistance to moisture damage by reacting with clay in or on aggregate particles. It can also be used to stiffen the binder for improved rut resistance and potentially reduce age hardening. USACE reports that hydrated lime is also used to reduce bacterial deterioration of asphalt in warm, wet climates.

The amount of hydrated lime needed to improve the moisture sensitivity of an asphalt mixture generally ranges from 0.5 to 1.5 percent by dry weight of aggregate. It is typically added either dry or mixed with water as a lime slurry.

The addition of hydrated lime has been known to impact volumetric properties in the mixture. It is recommended that the manner of introducing lime during the mix design process be similar in nature to the method used in the field during plant mix production. It is also recommended that the mix designer optimize the amount of lime to be added to the mix. The amount of lime required will depend on the binder and aggregate combination used in the mix.

For additional information on LAS additives and hydrated lime, see National Cooperative Highway Research Program (NCHRP) Synthesis 595, Practices for Assessing and Mitigating the Moisture Susceptibility of Asphalt Pavements.

3.6.4 Fibers

Fibers for asphalt mixtures can generally be grouped into two categories:

• Fibers intended to stop or mitigate draindown in SMA mixtures.
• Fibers intended to stabilize the mix and inhibit rutting, cracking, and shoving.

Fibers to mitigate draindown in SMA are typically either cellulose fibers or mineral fibers. Cellulose fibers are plant-based fibers made from woody plants, although some are finely shredded recycled newspaper. Cellulose fibers have high absorption, which helps them maintain high binder contents without draindown. Cellulose fibers can be provided in loose form or in pellets and are typically added at a rate of about 0.3 percent by weight of the total mix mass.

Mineral fibers (also called mineral wool or rock wool) are manufactured by melting minerals then physically forming fibers by spinning or extruding. Minerals used to create mineral fibers include slag or a mixture of slag and rock, basalt, brucite, and carbon. Mineral fibers are less common and can be provided in loose form or in pellets and are typically added at a rate of about 0.4 percent by weight of the total mix mass.

Fibers intended to stabilize the mix are typically polyester, polypropylene, aramid, or combinations of these. The fibers used for this purpose are strong, heat-resistant, and from 5 to 40 mm in length. Polyacrylonitrile fibers can have a softening point as low as 430 °F (220 °C), while aramid fibers have softening points more than 800 °F (425 °C). Although these fibers may not significantly increase the tensile strength of the asphalt mixture, research has shown them to significantly increase the fracture energy, thereby improving crack resistance and crack propagation. These fibers are most commonly added at a rate from 0.065 to 0.1 percent by weight of the total mix mass.

To produce an asphalt mix design, asphalt binder and aggregate are blended together in different proportions in the laboratory. The resulting mixes are evaluated using a standard set of criteria to permit selection of an optimum binder content (OBC).

3.7.1 Superpave Method

AASHTO M 323 Superpave Volumetric Mix Design and R 35 Superpave Volumetric Design for Asphalt Mixtures lay out the U.S. practices and specifications for the Superpave volumetric method of mix design. Many agencies supplement the volumetric method with additional testing to assess other properties such as the mix’s rut and cracking resistance. The Superpave gyratory compactor (see Figure 20) is used to compact asphalt mix specimens.

Source: National Asphalt Pavement Association
Figure 20. Superpave® Gyratory Compactor

The volumetric mix design is accomplished in four steps: 1) selection of component materials, 2) selection of design aggregate structure, 3) selection of design asphalt content, and 4) evaluation of moisture susceptibility. Selection of the component materials includes selection of the appropriate binder and aggregate that meet requisite quality characteristic parameters provided in the project specifications.

After component materials that meet quality requirements have been selected, the design aggregate structure must be selected. The broad band aggregate gradation is specified either by the mix NMAS in AASHTO M 323 or by gradation number in airfield specifications (the FAA P-401 and DoD UFGS 32 12 15.13).

The aggregate structure can simply be preselected if the designer has sufficient history/knowledge of the local materials. If the designer needs to select and evaluate aggregates from multiple sources or sources with which they lack sufficient familiarity, trial blends are normally designed and evaluated. The designer estimates the optimum asphalt content of each trial blend, then mixes and compacts specimens. The subsequent volumetric properties for each trial blend are evaluated to see if they meet specifications. Typically, the most cost-effective blend of aggregates that will meet both lab and field specifications is selected as the design aggregate structure.

Asphalt binder is then added to the design aggregate structure, typically anywhere from three to five binder contents at 0.5-percent intervals bracketing the estimated OBC. Typically, two identical specimens are prepared and averaged at each binder content so variability can be minimized. The OBC is typically selected as the binder content at a preselected air void content. For AASHTO M 323 specifications, the target air void content is 4 percent. FAA specifies 3.5 percent air voids. It is important to check local specifications for the target air void content. The volumetric properties at the OBC are determined, and the designer ensures that the asphalt mixture meets all volumetric specifications at that binder content.

The final step in the Superpave method is to determine the mixture’s moisture susceptibility. AASHTO M 323 specifically calls for AASHTO T 283, Resistance of Compacted Asphalt Mixtures to Moisture-Induced Damage, to pass with a tensile strength ratio of no less than 0.80. Other specifications may call for slightly different minimum tensile strength ratio values.

3.7.2 Marshall Method

The Marshall Method of mix design was historically the predominant method of mix design for dense-graded asphalt mixtures and is noted for the portability of its equipment. At the time of this writing, there is still one State DOT using the Marshall Method. It is an optional method for FAA and DoD mixtures, and it is still common outside the United States and Canada.

For a single selected aggregate gradation, specimens at five different asphalt contents are molded with manual or mechanical Marshall hammers and molds. FAA allows either hammer type in accordance with ASTM. DoD specifies the manual hammer or the mechanical hammer calibrated to the manual hammer. A given mass of asphalt mixture at a specified temperature is placed in the Marshall molds, and the hammer is used to deliver 35, 50, or 75 blows, depending on the specification. The mold is then flipped, and the same number of blows is delivered to the other side of the specimen. Typically, three identical specimens are prepared and averaged at each binder content so variability can be minimized.

The specimens are then tested for various volumetric criteria. In addition to mix volumetrics, the Marshall Method also applies mechanical testing in the form of stability and flow tests (AASHTO T 245, ASTM D6927), which are precursors to the Balanced Mix Design (BMD) tests that are now coming into common practice. The specimens are loaded in indirect tension using a compression tester. The stability value is basically the maximum load that can be supported by a compacted Marshall sample, and the flow value is the deformation corresponding to the maximum load. In most cases, the OBC should be selected for which the compacted specimen has 3.5 percent air voids (FAA) or 4.0 percent air voids while meeting Marshall stability and flow criteria.

3.7.3 Balanced Mix Design Method

Watch Video

Balanced Mix Design for Airfield Mixtures

Advancements in performance testing brought the concept of BMD to augment or even replace volumetric design. BMD, as defined in Transportation Research Board (TRB) Circular E-C280, Glossary of Terms for Balanced Design of Asphalt Mixtures is “an asphalt mixture design framework using mechanical tests correlated to field performance on appropriately conditioned specimens that address multiple modes of asphalt layer distress taking into consideration mixture aging, traffic, climate, and location within the pavement structure.” Although primary modes of distress considered by most practitioners of BMD are rutting and cracking, other distresses such as brittleness (evaluated by the Cantabro mass loss test) could also be considered in a BMD process. It is very important that any performance tests used are well-correlated to field performance.

Advancements in performance testing brought the concept of BMD to augment or even replace volumetric design.

As of this writing, the following are the four primary approaches to BMD for mixture design:

3.7.3.4 Approach A

Volumetric Design with BMD Verification. This approach starts with the volumetric mixture design method for determining an OBC. The asphalt mixture at the OBC is then tested with the selected mechanical tests to assess its resistance to distresses of interest. If the mix design meets the test criteria, the JMF is established and production begins. If the mix design does not meet the test criteria, the entire mix design is repeated using different mixture proportions or materials until all the volumetric and BMD test criteria are satisfied.

A good reference for more details is NAPA publication IS-143 “Balanced Mix Design Resource Guide.” Table 3 summarizes each approach.

Table 3. Summary of BMD Approaches

3.7.4 RAP Considerations

The mix design should be specific regarding the source of the RAP and whether it has been fractionated or not. Some contractors fractionate RAP by passing it over one or more screens to produce separate coarse and fine stockpiles for a more controlled gradation. Fractionating RAP can reduce the overall variability of the final mix when using higher percentages of RAP. Fractionating also helps mix designers because the coarse and fine RAP stockpiles have different properties, increasing the ways that the RAP can influence volumetric properties. The fine fraction of the fractionated RAP will have a higher binder content while the coarse fraction will have a lower binder content, relative to the same RAP if it were not fractionated.

Some agencies limit the amount of RAP in the mixture by specifying a maximum allowable percentage by weight. This is an acceptable approach when the percentage of asphalt binder in the RAP is relatively similar to the total asphalt binder content of the mixture. With the increased use of fractionated RAP, many agencies now recognize the need to specify the amount of RAP in terms of the RAP binder ratio—the ratio of the RAP binder in the mix divided by the mixture’s total binder content. DoD and FAA airfield projects do not currently allow the use of RAP for surface mixes, except on shoulders.

Because the RAP binder has been significantly aged and stiffened, mixes with large percentages of RAP tend to be stiffer, less workable, and more difficult to compact during construction compared to mixes with a smaller percentage of RAP. If the impact of the aged RAP binder is not addressed in the mix design, the durability of a high RAP mix will suffer as well. High RAP mixtures should be engineered with a softer binder, more binder, or recycling additives to improve the durability of the mixture. The Asphalt Institute’s MS-2 offers detailed guidance on binder grade adjustments due to the use of RAP, but in summary:

• Asphalt binder content and gradation must be determined for all RAP levels.
• At lower RAP levels, (less than 15 percent), the stiffer RAP binder has minimal effect on the total mix binder stiffness, so no grade adjustment is necessary.
• At moderate RAP levels (between 15 and 25 percent), select one grade softer than normal (e.g., select a PG 58-28 if a PG 64-22 would normally be used in a virgin mix).
• At high RAP levels (greater than 25 percent), the physical properties of the extracted asphalt binder will need to be determined so that blending charts or equations can be used to select the appropriate grade of virgin binder. Assuming the RAP binder blends with the new binder at mixing temperatures, a softer new binder is blended with the stiffer RAP binder, resulting in a binder blend that meets the required grade for the project.

Once the appropriate virgin asphalt binder grade and percentage of each RAP source has been selected, the normal mixture design process can proceed.BMD testing can greatly assist in optimizing the mix design process using RAP. Refer to the NAPA BMD Resource Guide for more information.