Fiber Additives in Asphalt Mixtures (2015)

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.

29 CHAPTER FOUR CASE EXAMPLES have been used under environmental conditions similar to their own (wet freeze). If the county decides to use fibers, it will need guidance on how to specify the fibers and mixtures, whether the type of mix design (Marshall or Superpave) makes a difference, and whether they will have to make changes to the mix design. Other questions include the typical dosage of fibers and whether it varies depending on fiber type. Moving on to construction, concerns include how to introduce the fibers into the mixture and whether that varies for different types of fibers. The types of equipment needed to uniformly disperse the fibers, where to introduce the fibers, and impacts on production time are also issues to consider. Officials wonder if they will have to make any changes to the construction process, procedures or equipment, and about the effects of fibers on constructability. Of course, cost is also an issue. To the extent that information could be obtained on these issues, answers to these questions have been sought as this synthesis was prepared. CASE 2. AGENCY WITH VARYING FIBER USAGE On the basis of research in the early 1980s which showed that fiber-reinforced asphalt surfaces experienced greatly reduced rutting and cracking, the Indiana Department of Transportation (INDOT) made extensive use of fibers in asphalt in the 1980s and very early 1990s. At the time, cracking and seating concrete pavements, followed by an asphalt overlay, was a standard rehabilitation technique in the state. For several years, INDOT required the use of polypropylene fibers in these dense-graded overlays. Both cracking and seating and the use of fibers were intended to This chapter presents five case examples of agencies and their use of fibers. The first example is of an agency that is contemplating the use of fibers but has no previous experience. This example is intended to illustrate the types of questions an agency might have regarding the use of this technology. The second case example is of a state that has had a dramatic change in the use of fibers over time. The third case example describes the evolution of specifications for fiber mixes and the experiences of contractors. The fourth case example summarizes an ongoing research effort by a state with little previous use of fibers in asphalt. The fifth case example is of a state that has used fibers extensively in SMA and porous mixes and is now exploring the possibility of using a different type of fiber in dense- graded mixes. CASE 1. AGENCY CONSIDERING USE OF FIBERS This case example is offered as an illustration of the types of questions an agency might have when considering requiring or allowing the use of fibers in asphalt mixtures. Erie County, New York, is considering that possibility to help asphalt pavements better withstand the rigors of their environment. Erie County, where Buffalo is located, experiences substantial lake-effect snowfall and very cold winter weather. Cracking is a frequent distress in asphalt pavements, so fiber reinforcement could be a good tool to have available. However, the county lacks experience with fibers. County officials are seeking information on the types of fibers that have been used and are currently available. They also wonder which applications are most suitable for fibers, which pavement layers would benefit most, and which distress types fibers can address effectively. The performance of fiber mixes is an obvious area of interest. Officials are particularly interested in cases where fibers

30 reduce reflective cracking, which was a common problem. Fibers also helped reduce rutting, another major problem. A 1989 study by El-Sheikh and Sudol found that cracking and seating reduced reflective cracking, compared with the control, by 75% after 5 years. Sections with fibers in the overlay over cracked and seated pavements showed a reduction in transverse cracking of 85%. In addition, fibers improved pavement strength (measured by falling weight deflectometer testing) by 15% compared with sections without fibers (El-Sheikh and Sudol 1989). Reduced rutting was also observed on this project and in previous trials (Galinsky 1984; McDaniel 1985). On the basis of this and other research, INDOT began specifying the use of fibers in all dense-graded overlays over cracked and seated pavements, and in some other overlays as well. Anecdotally, it was estimated that Indiana used more fibers in asphalt mixes than the next four states combined, but this has not been documented. Reportedly, problems began to develop in the late 1980s or early 1990s when INDOT loosened its specifications to allow more types of fibers, including waste carpet fibers, with reduced controls on their content or coatings. Construction problems with clumping of fibers began to appear. Shortly after the problems began to increase, INDOT implemented the new Superpave binder and mixture specifications. Under this system, it was difficult to quantify the effects of using fibers. Attempts to test the fibers as a binder modifier were ineffective because of severe difficulties in preparing the specimens for testing; the fibers could not be uniformly dispersed in poured binder specimens. Attempts to cut binder specimens from sheets of fiber-reinforced binder were also unsuccessful, as the specimen geometry and smooth edges could not be ensured. Tests of fiber-reinforced mixtures in the Superpave shear tester and the indirect tensile (IDT) tester, according to AASHTO TP 7 and TP 9, were also unsuccessful. It was speculated at the time that the tests were not sensitive enough to detect the contributions of the fibers. Superpave provided a means for the state to use polymer-modified binders for high-volume roadways; these were expected to help reduce both rutting and cracking. In addition, changes in the mix designs and aggregate requirements helped to greatly reduce the occurrence of rutting. Given these performance benefits, it was hard to justify the added cost of using fibers in Superpave mixes; consequently, the use of fibers in the state dropped dramatically. There was an increase in fiber usage in the late 1990s when the state began using SMA surfaces widely. In these mixes, however, cellulose or mineral fibers were used instead of polymer fibers. The high cost of SMAs during hard economic times led to a decrease again in fiber use. SMAs are making a small comeback in the state, so fiber usage may see an uptick. Virginia is another state that has seen ups and downs in fiber usage. McGhee et al. (2013) reported that Virginia had used open-graded friction courses but experienced various problems so discontinued their use in the late 1980s. Draindown was frequently observed and led to the pavement being underasphalted; these pavements often suffered durability problems leading to early failure. Those OGFCs that did not experience draindown reportedly developed black ice in some conditions, leading to safety concerns. Finally, OGFCs were linked to increased moisture damage of the underlying layer, which resulted in failures deeper in the pavement that were more difficult and expensive to repair. New generation OGFCs used polymer-modified binders and fibers to prevent draindown. Higher void contents were also maintained with these mixes, so they could dissipate sound energy, making the pavements quieter. [Virginia had a legislative mandate to explore options for quieter pavements (McGhee et al. 2013).] CASE 3. CONTRACTORS’ EXPERIENCES WITH FIBERS IN ASPHALT MIXTURES In the 1980s and 1990s, the Florida Department of Transportation (FDOT) used an open-graded friction course, designated FC-2, to provide a high-friction, drainable surface for high-speed, multilane roadways to reduce hydroplaning. This open-graded mixture used a 3/8-in. nominal maximum aggregate size and, initially, an unmodified AC-30 asphalt binder. Later, 12% asphalt rubber was added to the AC-30 (Cunagin et al. 2014) to increase the binder content (and film thickness) without creating a draindown problem. This surface type tended to have a relatively short service life and typically failed because of raveling of aggregate from the surface. The raveling was attributed to the open-graded structure of the mix, a low binder content, and the resulting thin binder film. The open structure of the mix allowed oxygen to enter the pavement and accelerate oxidation and embrittlement of the binder. The use of an unmodified binder was also a contributing factor, but the use of rubber-modified asphalt did not entirely solve the problems with this mix; relatively thin lifts and low tack coat application rates also aggravated performance problems. The development of a new mixture specification implemented in 2000 was motivated by the problems with earlier friction courses and the Georgia DOT’s success with modified OGFCs. The resulting FC-5 surface course uses a ½-in. nominal maximum aggregate size and modified binder—either the same asphalt rubber or a PG 76-22 polymer

31 modified binder. In addition, cellulose or mineral fibers are required to stabilize the asphalt binder and prevent excessive draindown. The fiber rate is specified at 0.4% for mineral fibers and 0.3% for cellulose fibers (Cunagin et al. 2014). FDOT’s pavement management system data show that the implementation of the FC-5 specification increased the median service life of the surface to 15 years, compared with 12 years for the FC-2. Although fibers presumably contributed to the increase in the life of the surface, other changes made at the same time also contributed. These include a heavier tack coat, thicker lifts, higher production and placement temperatures, and more polymer-modified binder (Cunagin et al. 2014). Nonetheless, the use of this fiber-reinforced material has resulted in better durability of the surface, and its use is expected to continue. When the FC-5 specification was implemented, contractors in Florida had to adjust to adding fibers to the mixtures. There were a number of issues in the beginning. Owing to a patent issue, there was initially only one supplier for the fibers. This caused occasional supply shortages, which eventually led some contractors to stockpile fibers so they would be available when needed. The expiration of a patent on the fibers allowed other suppliers to enter the market, which increased the supply and led to lower prices through competition. Today, supply issues are a thing of the past (Jim Musselman, FDOT correspondence, Aug. 29, 2014). Another problem in the early days of implementation was with clumping of the fibers during mix production. Some contractors reported more clumping problems with certain sources of fibers; with multiple suppliers, these sources can usually be avoided. Storing the fibers properly and keeping them dry are also credited with resolving a number of the clumping problems. Although the FDOT specifications allow the use of either cellulose or mineral fibers, the majority of contractors use mineral fibers. Cellulose fibers require about 0.3% to 0.5% additional binder compared with mineral fibers and, because binder is included in the mix price, result in a higher mix cost for the contractors (Jim Musselman, FDOT correspondence, Aug. 29, 2013). The addition of fibers also meant retrofitting some asphalt plants and renting or buying the equipment to blow in the fibers. While most contractors initially rented the equipment, most of them have now purchased it; it is routinely used, so owning is more cost-effective than renting. In one case, for example, the cost to rent was $7,000 per month and the purchase price was $75,000, so the payback period was quite short (Jim Musselman, FDOT correspondence, Aug. 29, 2013). The example of the Florida DOT and its contractors shows that issues can arise when fiber use begins, but that with attention to detail and experience with the product the issues can be addressed and overcome. CASE 4. ONGOING RESEARCH ON FIBERS IN DENSE- GRADED ASPHALT As the survey results show, nearly all the current use of fibers in asphalt in the United States is in open-graded or SMA mixtures. The Idaho Transportation Department (ITD) provides an example of an agency that is currently researching the use of fibers in dense-graded asphalt. (Others include the Pennsylvania DOT, Ontario Ministry of Transportation, and some local agencies.) Fibers are being investigated to determine their effectiveness at reducing rutting and cracking, as ITD continues its efforts to reduce costs and prolong pavement life. In addition to comparing the laboratory performance of the fiber mixes, the research effort will assess the mixture properties needed as inputs to the mechanistic-empirical pavement design software. A later phase of the project is intended to monitor the field performance for a period after construction (University of Idaho 2015). Three test sections incorporating different fibers and one control section were constructed on US-30 in August 2014. This highway carries heavy truck traffic near the border with Wyoming, which has caused rutting. Cracking of the existing pavement was also observed. The existing pavement was milled 122 mm (0.4 ft) before the new overlay was placed in two 61-mm (0.2-ft) lifts. Fibers were added to both lifts in the test sections without increasing the binder content. The milled material was reused in the new mixtures (Mike Santi, ITD communication, Aug. 29, 2013; University of Idaho 2015). The mix design was performed without fibers, which were later added at the drum mix plant. The mix incorporated a high reclaimed asphalt pavement content of 47% by mass of mix with a binder replacement value of about 54%. The fibers under study include aramid (0.35 lb/ton), fiberglass (3 lb/ton), and a blend of polypropylene and aramid (1 lb/ton). The fibers were blown into the plant along with the reclaimed asphalt pavement through the vendors’ equipment, which was calibrated to inject the required amount of fibers (Mike Santi, ITD communication, Aug. 29, 2013; University of Idaho 2015). The laboratory evaluation of the mixtures will include the following: • Gyratory stability analysis using the Superpave gyratory compactor;

32 • Dynamic modulus testing in indirect mode at a range of temperatures (-20°C, -10°C, 0°C, 10°C, 20°C, and 30°C) and frequencies (0.1, 1, 5, 10, 20 Hz) to assess stiffness; • Flow number testing to evaluate resistance to permanent deformation; • Creep compliance testing to evaluate thermal cracking; • Fatigue analysis using the fracture work density concept; • Transverse cracking analysis, also based on fracture work density; • Asphalt pavement analyzer (APA) testing to evaluate permanent deformation; • Performance prediction using mechanistic-empirical pavement design software from AASHTO; and • X-ray tomography to explore the distribution of fibers through the mix. The testing and analysis will be done by the University of Idaho and Washington State University, with the exception of the APA testing, which will be conducted by ITD (University of Idaho 2015). The X-ray tomography results may be of particular interest because of ongoing and widespread concerns about the uniformity of fiber dispersion during construction. A researcher present during construction of the Idaho test sections noted that some of the fibers, which were visible through clear tubes leading into the plant, did not flow uniformly into the plant but rather agglomerated into a ball that was pushed through the tube. It is currently unknown whether the fibers were then uniformly distributed during the mixing process (Fouad Bayomy, University of Idaho correspondence, Aug. 28, 2013). Fibers have been promoted in Idaho and elsewhere as a means to reduce layer thicknesses and thereby reduce costs. The current study is not evaluating reduced thickness, though this is a possibility in the future (Mike Santi, ITD correspondence, Aug. 29, 2013). Regarding the use of fibers to reduce pavement thickness, the Asphalt Pavement Association of Oregon (APAO) has recently prepared a position paper urging agencies to treat the use of fibers as an experimental technology. APAO recommends that control sections be placed on projects incorporating fibers to expand the range of materials and conditions (e.g., traffic, climate, pavement structure) in which fibers are used to develop a clearer understanding of their effects on performance. Although APAO supports the use of proven technologies, its position is that the benefits of fibers, especially for reducing pavement thickness, have not yet been proven (APAO position paper). CASE 5. STATE WITH HIGH FIBER USAGE RESEARCHING OTHER APPLICATIONS The Texas Department of Transportation (TxDOT) uses large quantities of fiber-reinforced SMA and PFC every year—about 700,000 tons of fiber mix annually. This is about 40% more fiber mix than Florida uses and three to five times more than other high-use states such as Georgia, South Carolina, and Tennessee. Texas uses cellulose and mineral fibers in these applications. In 2013, TxDOT began placing test sections around the state to evaluate the use of fibers in dense-graded surface mixes to control cracking. There is some concern that some of the DOT mixes might be too stiff, and cracking has been observed, so the department is interested in exploring options to reduce cracking and increase the service lives of its asphalt surfaces. The test sections reflect a variety of climates and traffic conditions to determine whether fibers are beneficial in various settings. The test sections include 1 lb of blended polyolefin and aramid fibers per ton of hot mix. They also include unmodified control sections. In addition to exploring the field performance of these mixes, two other questions are being investigated: Are there differences between lab and plant mixing when fibers are used? Are the tests used in Texas—including IDT testing, the Hamburg tester, and the overlay tester—applicable to fiber mixes? There is a concern in Texas, as elsewhere, that some laboratory tests might not accurately reflect the effects of fiber reinforcement. No construction difficulties were reported during construction of the test sections. In the coming years, the results of this research will likely be of interest to other agencies that have expressed some of the same concerns about field performance and laboratory testing of fiber mixes.