Annotated Literature Review for NCHRP Report 640 (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

87 1.21 Huber, G., “Performance Survey on Open-Graded Friction Course Mixes.” Synthesis of Highway Practice 284. National Cooperative Highway Research Program. Transportation Research Board. National Research Council. Washington, D.C. 2000. 1.21.1 General This synthesis documents the recent [as of 2000] performance of OGFC in both Europe and the US. Specific chapters within the synthesis deal with the use of OGFC; design and construction practices; and performance, maintenance and rehabilitation practices from around the world. Huber indicated that OGFC was developed in the U.S. during the 1940’s, when the California Department of Highways experimented with plant mix seal coats due to problems encountered with chip seals. Plant mix seal coat pavement sections tended to have aggregates tightly bonded to the road surface eliminating windshield damage; fewer climatic problems during construction (especially after a rain); and more construction related advantages due to being able to place plant mix seal coats with a paver and being able to place the mix at relatively thin lifts (15 to 20 mm). These plant mix seal coats became known as OGFCs. 1.21.2 Benefits of Permeable Asphalt Mixtures Huber states that the primary benefits realized with OGFCs are increased permeability and noise reduction. Permeability leads to improved wet weather frictional properties, improved wet weather visibility, and less splash and spray. Huber describes that water sits on the road surface during a rain event where it can be splashed or thrown into the air in the form of a mist when vehicle tires pass over the pooled water. Traffic mist can reduce visibility more severely than fog because the airborne particles within the mist are larger than the particles within fog. Porous pavements reduce (and almost alleviate) the droplets of water caused by vehicles passing over the roadway because the water infiltrates into the interconnected voids of the pavement layer. Noise reduction is also provided due to the porosity of OGFC pavements. Huber reports that noise reductions at highway speeds (comparing OGFC to typical dense-graded mixes) are on the order to 3.0 dB(A), which is a 50 percent reduction in noise pressure. A roadway test in Denmark showed that even after 5 years porous asphalt mixtures are 2 to 4 dB(A) quieter than typical dense-graded mixes. Huber states that in most instances, OGFC pavements are more desirable on high speed roadways. High speeds are needed in order to generate enough hydraulic action under vehicle tires to allow the pavement to be self-cleaning and, therefore, maintain porosity. The synthesis included a survey of all 50 US state highway departments and 10 provinces from Canada. As part of the survey, Huber requested each agency to identify the number one benefit sought with the use of OGFC. Figure 12 illustrates the results of this survey. This figure indicates that the number one benefit perceived by the survey respondents was improved frictional resistance. The second most perceived benefit was improved driver visibility during wet weather.

88 0 5 10 15 20 Noise Reduction Marking Visibility Driver Visibility Friction Number of Agencies Figure 12: Benefits of Open-Graded Mixes Cited by Agencies 1.21.3 Materials and Mix Design Huber described the 2000 standard mix design method of OGFCs. The standard, at that time, was based upon the FHWA Technical Advisory T5040.31. Material requirements contained within this Technical Advisory are summarized as follows: 1) polish resistant aggregate with 75 percent crushed two faces and 90 percent one face; 2) mineral filler meeting AASHTO M17; 3) aggregate gradation meeting Table 49; 4) AC-20 graded asphalt binder; and 5) antistrip additives. Table 49: FHWA Design Gradation Band Percent Passing Sieve, mm Minimum Maximum 12.5 100 --- 9.5 95 100 4.75 30 50 2.36 5 15 0.075 2 5 Huber summarized the design steps contained within the Technical Advisory as follows: 1) determine percentage of S.A.E. No 10 oil retained on the aggregates after soaking and draining; 2) estimate asphalt binder content based on the percentage of oil retained; 3) determine air void content in coarse aggregate by compacting dry coarse aggregate using

89 a vibratory hammer; 4) calculate amount of fine aggregate; 5) test for draindown; and 6) evaluate moisture susceptibility of the designed mix. Huber also described some recent advancements in OGFC mix design technology. Specifically, the inclusion of polymer modified binders and fibers to reduce the potential of draindown was incorporated into the materials selection process. The combination of fibers and modified binders provided the extra benefit of allowing higher production temperatures. Higher production temperatures aided in more effectively removing moisture from aggregates, thereby, reducing the potential for moisture damage. Another advancement mentioned by Huber was the increasing of air voids in the designed mixtures. Increased air void contents provided self-cleaning due to the hydraulic action of traffic. Huber also noted that some agencies had increased the nominal maximum aggregate size from 9.5 mm to 12.5 mm and 19mm. Tables 50 and 51 present a summary of mix design criteria used within the U.S. at the time this synthesis was published. Huber also describes mix design methods from several European countries and South Africa. Table 52 summaries various porous asphalt mix design requirements from other parts of the world. Of particular interest in Table 52 is that all of these mix design criteria contain a minimum air void requirement. All six sets of criteria include a minimum air void content of 18 percent of more. In contrast, none of the US specifications had minimum air void contents. An interesting mix design concept is utilized by Spain. Minimum asphalt binder content is selected based upon a durability test called the Cantabro Abrasion test. The Cantabro Abrasion test measures disintegration of compacted samples using the Los Angeles Abrasion drum. Samples are placed in the drum without steel spheres and subjected to 300 revolutions at room temperature. After the 300 revolutions, the percentage of mass loss is then determined.

90 Table 50: Summary of 9.5mm OGFC Mixture Designs used in United States Arizona Unmodified Arizona Rubber California Florida Nevada Nevada Wyoming New Mexico Georgia Min. Max. Min. Max. Min. Max. Min. Max. Min. Max. Min. Max. Min. Max. Min. Max. Min. Max. Gradation (mm) 12.5 100 100 100 100 100 100 100 100 100 9.5 100 100 90 100 85 100 95 100 90 100 97 100 90 100 85 100 4.75 35 55 30 45 29 36 10 40 40 65 35 55 25 45 25 55 20 40 2.36 9 14 4 8 7 18 -- -- -- -- -- -- 10 25 -- -- 5 10 2.00 -- -- -- -- -- -- 4 12 -- -- -- -- -- -- 0 12 -- -- 1.18 -- -- -- -- -- -- -- -- 12 22 5 18 -- -- -- -- -- -- 0.075 0 2.5 0 2.5 0 3 2 5 0 4 0 3 2 7 0 4 2 4 Binder Grade PG 64-16 PG 64-16 + 20% Rubber AR4000, AR8,000 or PDA-6 AC30 + 12% Rubber AC20P or AC30 AC20P or AC30 PG 64-22 or PG 70-28 PAC20 PG 67-22 Binder Content -- -- -- 5.5 – 7.0 6.5 typical 6.5 typical 6.3 – 6.8 -- 6.0 – 7.3 Aggregate Properties Specific Gravity -- 2.35 - 2.85 -- -- -- -- -- -- -- Water Abs. -- 2.5 max. -- -- 4 max. 4 max. -- -- -- Sand Equiv. 45 min. 55 min. -- -- -- -- -- -- -- Crushed Faces 70 min. 95 min. 90 min. 100 min. 90 min. 90 min. 95 min. 75% min. -- Flakiness Index 25 max. -- -- -- -- -- -- -- -- L.A. Abrasion -- 40 max. 40 max. -- 37 max. 37 max. 35 max. 40 max. -- Carbonates -- 30 max. -- 88 max -- -- -- -- -- Mg Soundness -- -- -- -- 12 max. 12 max. -- 12 max. -- Compaction Method Dbl. Plunger Dbl. Plunger -- -- Static Compression Static Compression -- -- Marshall Air Voids -- -- -- -- -- -- -- -- --

91 Table 51: Summary of 12.5mm and 19.0mm OGFC Mixture Designs used in the United States California Georgia OGFC Georgia PEM Oregon Oregon Min. Max. Min. Max. Min. Max. Min. Max. Min. Max. Gradation (mm) 12.5 100 100 100 100 100 9.5 100 100 90 100 85 100 95 100 4.75 35 55 30 45 29 36 10 40 40 65 2.36 9 14 4 8 7 18 -- -- -- -- 2.00 -- -- -- -- -- -- 4 12 -- -- 1.18 -- -- -- -- -- -- -- -- 12 22 0.075 0 2.5 0 2.5 0 3 2 5 0 4 Binder Grade PG 64-16 PG 64-16 + 20% Rubber AR4000, AR8,000 or PDA-6 AC30 + 12% Rubber AC20P or AC30 Binder Content -- -- -- 5.5 – 7.0 6.5 typical Aggregate Properties Specific Gravity -- 2.35 - 2.85 -- -- -- Water Abs. -- 2.5 max. -- -- 4 max. Sand Equiv. 45 min. 55 min. -- -- -- Crushed Faces 70 min. 95 min. 90 min. 100 min. 90 min. Flakiness Index 25 max. -- -- -- -- L.A. Abrasion -- 40 max. 40 max. -- 37 max. Carbonates -- 30 max. -- 88 max -- Mg Soundness -- -- -- -- 12 max. Comp. Method Dbl. Plunger Dbl. Plunger -- -- Static Compression Air Voids -- -- -- -- --

92 Table 52: Summary of Non-North American Porous Asphalt Mixtures British (20 mm) British (10mm) Spanish (P-12) Spanish (PA-12) Italian South Africa Min Max Min Max Min Max Min Max Min Max Min Max Gradation (mm) 28 100 --- --- --- --- --- --- --- --- --- --- --- 20 95 100 --- --- 100 --- 100 --- 100 --- --- --- 19 --- --- --- --- --- --- --- --- --- --- 100 --- 14 55 80 100 --- --- --- --- --- 75- 100 --- --- 13.2 --- --- --- --- --- --- --- --- --- --- 90 100 12.5 --- --- --- --- 75 100 70 100 --- --- --- --- 10.0 --- --- 90 100 60 90 50 80 15 40 --- --- 9.5 --- --- --- --- --- --- --- --- --- --- 25 65 6.3 20 30 40 55 --- --- --- --- --- --- --- --- 5.00 --- --- --- --- 32 50 15 30 5 20 --- --- 4.75 --- --- --- --- --- --- --- --- --- --- 10 15 3.35 8 14 22 30 --- --- --- --- --- --- --- --- 2.50 --- --- --- --- 10 18 10 22 --- --- --- --- 2.36 --- --- --- --- --- --- --- --- --- --- 8 15 2.00 --- --- --- --- --- --- --- --- 0 12 --- --- 0.63 --- --- --- --- 6 12 6 13 --- --- --- --- 0.008 --- --- --- --- 3 6 3 6 --- --- --- --- 0.075 2 7 2 7 --- --- --- --- 0 7 2 8 Asphalt Binder Grade 100 pen. + SBS 100 pen. + EVA 100 pen. 100 pen. +SBS 100 pen. + EVA 100 pen. 60/70 + SBS 60/70 + EVA 80/100 + SBS 80/100 + EVA 4.5% typical 60/70 + SBS 60/70 + EVA 80/100 + SBS 80/100 + EVA 4.5% typical 80/100 + SBS Asphalt rubber Polymer modified Content 4.5% min. 4.5% min. 4.5% typical 4.5% typical 4-6% 4.5% min. Aggregate Properties L.A. abrasion Flakiness index Sand equivalent Crushed faces (2 faces) 12% max. 25 max. --- 100% 12% max. 25 max. --- 100% 20% max. 25 max. --- 100% 20% max. 25 max. --- 100% 16% max. --- --- --- 21% max. 25 max. 45 min. 100% (high traffic) 90% (low traffic) Mixtures Properties Compaction Air voids 50 blow Marshall 20% min. 50 blow Marshall 20% min. 50 blow Marshall 20% min. 50 blow Marshall 20% min. Marshall 18-23% 50 blow Marshall >22% high volume 18-22% low volume Cantabro dry Cantabro aged Cantabro wet --- --- --- --- --- --- 25˚C/25% max. --- --- 25˚C/25% max. --- --- 25˚C/25% max. --- 20˚C/30% max. 25˚C/25% max. 25˚C/30% max 25˚C/30% max Note: L.A. = Los Angeles; SBS= styrene-butadiene-styrene; EVA= ethylene-vinyl acetate; pen.=penetration.

93 1.21.4 Construction Practices Huber described common practices for the production and laydown/compaction of OGFCs. Huber indicates that no specific plant modifications are required to produce OGFCs. However, when modified asphalt modifiers are used, mechanical agitators may be required in the asphalt storage tank. When fibers are included, the plant must be equipped with a metering system to feed the fibers into the production process. Also, the system must be able to evenly distribute the fibers. Nonuniform fiber distribution can result in portions of the mixture being dry and unworkable while other portions of the mix can be soft and susceptible to draindown problems. Once the mixture’s produced, storage time should be minimized as long storage times in a heated silo can result in an increased potential for draindown. Huber reports that agencies require maximum storage times from 1 to 12 hours. Transportation of the produced mix to the construction site can be conducted in the same trucks in which dense-graded mixes are transported. Release agents can be used to prevent the mixture from sticking to the truck bed. Tarps should also be used to maintain mix temperature during transportation and to prevent crushing of the mix. Once at the construction site, the OGFC can be delivered to the paving train in much the same manner as dense-graded mixtures. As with any HMA, good construction practices are required. When windrows are used, the length of the windrows should be carefully controlled so that the mixture will not lose temperature. Huber indicates that within favorable climatic conditions, windrows should not be over 50m (150 ft). Huber indicates that roll down may be less for OGFC mixes than for dense-graded HMA. Also, when an extendable screed is used, auger extensions should be included. Huber also states that typical rolling patterns should be used for compaction. Transverse joints are more difficult to construct with OGFC mixes than for dense-graded HMA. Handwork tends to be difficult. Construction of longitudinal joints according to Huber, are similar to dense-graded pavements. Huber states that tacking the vertical face of longitudinal joints can be done but care should be taken not to apply excess tack coat. Excessive tack coats on the longitudinal joint can limit drainage across the joint. Huber recommends the use of static steel wheel rollers for compacting OGFC layers. Rollers should be 10 Mg (11 tons) or smaller to prevent excessive aggregate breakdown. 1.21.5 Maintenance Practices Huber defines two categories of maintenance for OGFCs: surface maintenance and winter maintenance. Winter maintenance, as the title indicates, deals with the activities required to maintain the driving surface during cold and inclement weather. Huber indicates that OGFC mixes behave differently in freezing environments. The high air void contents associated with OGFC mixes act as insulation, resulting in the OGFC layer more resistant to the flow of heat through the pavement structure. Huber indicates that the heat conductivity of a porous asphalt mixture is 40 to 70 percent of a typical dense- graded mix. The surface temperature of OGFC layers is generally 2˚C (3.6˚F) cooler

94 than a dense-graded layer. This means that an OGFC surface can drop below freezing resulting in the formation of ice/frost when a nearby dense-grade surface does not freeze. Because of the thermal properties of OGFCs, they will also stay frozen longer then dense-graded mixes. Salts used on OGFCs during winter maintenance have less contact time with the icy surface. On OGFC, salt beings to melt the ice and form brine, which then disappears into the void structure of the OGFC. Surface maintenance entails activities required to restore or preserve the surface condition of the pavement. Typical surface maintenance activities include crack sealing, pot hole repair, fog sealing and striping. Of the 17 states that responded to Huber’s survey, all 17 report that potholes and delaminated areas within OGFC are repaired with dense-graded HMA. Only one state agency mentioned crack filling as a routine surface maintenance activity. Crack sealing may cut off the flow of water within an OGFC, leading to other problems. Huber described a technique used in Britain in which pavement texture depth is monitored in the spring of each year. An increase in surface texture resulting from loss of surface fines may be an indication of impending failure due to raveling. Huber states that lane markings are difficult to maintain on porous asphalt. Because of the high macrotexture and high air voids associated with OGFC, low viscosity striping materials will flow down into the voids and texture. A report from Ohio referenced in the synthesis states that 30 percent more material is required if epoxy is used and 50 percent more material is required if paint is used. Some agencies reported that high viscosity thermoplastic works well. The use of fog seals in maintenance is mixed in the US. Some agencies use them while others do not. Fog seals are reported to reduce in-place air voids and, therefore, drainage capacity. Vacuum-machines are reportedly used in some European countries to remove debris from an in-place OGFC pavement. High pressure water is sprayed on the pavement surface to dislodge debris and a high power vacuum is then used to remove the debris and water. This cleaning action is supposed to declog the pavement and extend the performance life. 1.21.6 Rehabilitation Practices Huber states that the literature recommends the removal of OGFC layers prior to replacement. Hot in-place recycling has been used to rehab OGFC layers. 1.21.7 Performance Huber defines the performance of OGFC into two categories: performance life and service life. Performance life is used to describe how long the OGFC maintains its permeability and ability to reduce noise levels. Service life deals with the ability to maintain friction and smoothness. A number of referenced papers were discussed that indicates that OGFCs maintain their sound attenuation for five years or more as long as

95 the design air voids are above about 18 percent. Another important factor in maintaining performance life is traffic speed. Huber indicates that the combination of high air voids and high traffic speeds helps maintain an open void structure. The pressures induced by fast moving tires over a wet pavement surface causes hydraulic scouring of debris from the void structure. This was illustrated through a reference that indicated an OGFC placed in a slow speed urban environment lost permeability within 2 years. In order to combat clogging tendencies in slow speed environments, a two-layer porous asphalt has been developed in Europe. The concept is to place a large aggregate layer of porous asphalt with a smaller aggregate OGFC on top (two-layer OGFC). The smaller aggregate size wearing surface will trap larger debris in order to maintain the permeability of the lower layer. The air void space in the bottom layer allows a water/jet vacuum machine to restore permeability. Huber conducted a survey of 17 agencies that were major and minor users of OGFC. Fourteen of the respondents indicated that raveling was the most common failure mechanism for OGFCs. Cracking and potholes were both reported as the cause of failure by two agencies of the respondents. Delamination was identified by three agencies as cause of failure. 1.21.8 Structural Design No specifics about inclusion of OGFC layers in structural design were given; however, Huber states that typical layer thicknesses in the U.S. are 20 to 25 mm (3/4 in. to 1 in.). Europe and South Africa have used thicker layers, 40 to 50 mm (1.5 in. to 2 in.). 1.21.9 Limitations Huber cites a number of issues that can be considered detrimental to the performance of the pavement rather than true limitations. The synthesis states that OGFC pavements tend to form ice on the pavement surface at a warmer temperature than dense-graded surface. This leads to more frequent deicing applications. Huber also indicates that raveling is the most typical distress. Raveling can occur very quickly causing the pavement to almost disintegrate within a few months time. Clogging can be the reason to discontinue the use of OGFC. Significant clogging will reduce all of the benefits related to permeability. Huber also states that clogged OGFC pavements may also accelerate moisture damage within underlying pavement layers. An unclogged OGFC layer may also result in moisture damage in underlying pavement layers. Huber states that the placement of OGFC may create a moist microenvironment at the surface of the underlying layer (bottom of the OGFC). The increased humidity caused by the moisture may retard the evaporation of moisture from the underlying layer resulting in moisture damage within the existing pavement.