Relationship Between Chemical Makeup of Binders and Engineering Performance (2017)

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.

41 CHAPTER FOUR ASPHALT ADDITIVES AND MODIFIERS INTRODUCTION As refiners’ efficiency allows them to extract more gasoline and other petroleum products from crude oil and as the source of crudes that yield quality asphalt residua decreases, the need for additives to upgrade straight-run asphalts increases. Further, the physical specifications for ACs defined by the PG system have prompted the use of more additives and modifiers to meet state DOT requirements. Highway agencies have recognized the benefits of using modified asphalts to reduce the amount and severity of pavement distresses and to increase service life. The primary benefit of using these high-performance asphalts is improved rut- ting resistance, with less thermal (cold-temperature) cracking and overall improved mixture durability being secondary benefits. Additionally, some modified binders provide improved stripping (moisture damage) resistance. Many agencies estimate that an additional 4 to 6 years of pavement life from a pavement constructed using a modified asphalt binder is a reasonable expectation. The state DOT or other agency responsible for managing construction may not know what type of modifier is used in its binders. The goal of Superpave was to make the PG system blind to the modifier that is used. As long as the asphalt meets the PG requirements it should not matter what modifier is used. The type of additive used may be proprietary and not necessarily known by the purchaser. In practice the asphalt supplier will normally provide some information about the type of modifier used but may not give specific details. Modifiers and additives being used to boost AC performance include polymers, chemical modifiers, extenders, oxidants and antioxidants, hydrocarbons, and antistripping additives. “Polymers” cover a broad range of modifiers, with elastomers (rubbers or elastics) and plastomers (plastics) being the most commonly used types. Styrene-butadiene rubber, styrene- butadiene-styrene, and crumb rubber are frequently used elastomers. SBS is the most often used modifier. These modifiers are used to reduce rutting and to improve fatigue and thermal cracking resistance. Crumb rubber is an elastomer made from ground tires. Several technologies are in place for using ground tire rubber. This material is used primarily to address rutting and fatigue. Plastomers are used to improve the high-temperature (rutting) properties of modified materials. Low-density polyethylene (LDPE) and EVA are examples of plastomers used in asphalt modification. The most commonly used chemical modifier is polyphosphoric acid. This modifier may be used in combination with poly- mers to increase the high-temperature stiffness. Other modifiers that may be used include asphalt binder extenders (primarily sulfur) and hydrocarbon materials. Hydrocarbons can produce either hardening or softening effects. The modifiers also include antioxidants, antistripping agents, softening agents, stiffening agents, wax-based additives, and recycling (rejuvenating) agents. The increase in the amount of modifiers used in asphalt cements can primarily be attributed to the following factors (Rob- erts et al. 2009): 1. An increased demand on HMA pavements. Traffic volume, traffic loads, and tire pressures have increased significantly in recent years, requiring that HMA be more resistant to rutting. 2. Superpave binder specification requires the asphalt binder to meet the stiffness requirements at high as well as low pavement service temperatures. Most neat (non-modified) asphalt cements do not meet these requirements in regions with extreme climatic conditions. 3. Environmental and economic pressure to dispose of some waste materials and industrial by-products (such as tires, roofing shingles, glass, sulfur, and ash) as additives in HMA. 4. Public agency willingness to pay a higher first cost for pavements with a longer service life or that will reduce the risk of premature failure.

42 Depending on the asphalt source and the average climatic conditions, the main reasons for including a given modifier are the following (Roberts et al. 2009): • Increase the stiffness of the mixture to minimize rutting • Soften and increase the elasticity of the mixture to minimize cracking • Improve the fatigue resistance of the mixture • Improve asphalt-aggregate binding to reduce stripping or moisture sensitivity • Improve abrasion resistance to reduce raveling • Rejuvenate aged asphalt binders • Reduce flushing or bleeding • Improve resistance to aging or oxidation • Reduce structural thickness of pavement layers • Reduce life-cycle costs of HMA pavements • Improve overall performance of HMA pavements. An ideal binder should be modified to achieve the properties illustrated in Figure 18. FIGURE 18 Stiffness characteristics of conventional binder and ideal modified binder (Source: Terrel and Epps 1989a, b). The lower service temperature region prefers lower stiffness and faster relaxation properties to reduce thermal cracking. Lower stiffness (or viscosity) is desired at high construction temperatures to facilitate pumping of the liquid asphalt binder, and mixing and compaction of HMA, but a higher stiffness should be retained at high service temperatures to reduce rutting and shoving. Adhesion between asphalt binder and aggregate should be strong in the presence of moisture to reduce stripping. Most of these criteria can be met with polymer additives. POLYMER ADDITIVES Elastomers Currently the most common polymer additives are styrene-butadiene-styrene copolymer, SBR, and EVA. Polymers are mac- romolecules made by chemically reacting many (poly) smaller molecules (monomers) with one another to form long chains or clusters. The sequence and chemical structure of the monomers determine the physical properties of the resulting polymer. Copolymers consist of a combination of two different monomers that can be in a random or block arrangement. For example,

43 polystyrene is a hard, brittle plastic whereas polybutadiene is soft and rubbery. If these two distinctly different monomers are randomly mixed and reacted together, a new polymer called a copolymer is formed, with varying properties depending on the molar ratio of monomers incorporated into the chains. Polymers can be engineered to obtain a broad range of physical properties. However, they can be divided into three general categories: fibers, plastics (plastomers), and rubbers (elastomers). The divisions are defined by the polymer thermal properties and morphology. A fiber will be predominately crystalline with a high melting point and a modest glass transition temperature. A plastic is predominately amorphous with a Tg higher than its use temperature. A rubber is amorphous with a Tg well below its use temperature; to prevent the rubber molecules from flowing at room temperature, the rubber is crosslinked. The SBR is usually crosslinked with sulfur (vulcanization); the level of crosslinking controls the final properties ranging from a soft flexible material suitable for tire treads to a hard material suitable for bowling balls. The SBR emulsions used as modifiers are lightly crosslinked rubber that does not completely melt during processing. The SBR is normally introduced as a latex emulsion and the water content is flashed from the asphalt cement. SBS triblock copolymer is a self-crosslinking elastomeric material; the polystyrene blocks have a Tg above the use tem- perature and serve as crosslinks. When the material is heated above the Tg of the polystyrene block, the material becomes a free-flowing liquid that is easy to disperse at asphalt processing temperatures. The structure of the SBS may vary from a linear polymer chain to a branched polymer chain that is designated as a radial copolymer. Radial SBS copolymers exhibit lower melt viscosities than linear SBS copolymers and thus allow lower processing temperatures. The polymers are used to increase the PG high temperature grade of the binder. The benefit of the polymer modifiers will depend on the concentration, morphology, molecular weight, chemical composition, and molecular structure of the material. The crude source, refining process, and grade of the neat asphalt binder are equally important. The residual reactivity of the polymer is useful for improving compatibility of the additive with the binder. Plastomers The utilization of plastomers in asphalt modification is limited. Polyethylene, which can be found in three forms—LDPE, high-density polyethylene (HDPE), and linear low-density polyethylene (LLDPE)—is the most common plastic. Other poly- olefins employed include polypropylene and ethylene-propylene copolymer, and EVA copolymer. Although the modification of bitumen with virgin polymers can improve the properties of asphalt mixtures, the use of recycled plastic may also show a similar result with additional environmental advantages (Garcia-Norales et al. 2006). A recent study evaluated the possible advantages of modifying the bitumen with different plastic wastes, namely polyethylene (HDPE and LDPE), EVA, acryloni- trile-butadiene-styrene, and crumb rubber. The performance of modified binders with recycled polymers was compared with that of conventional bitumen and of a commercial modified binder (Styrelf) (Costa et al. 2013). Reclaimed polyethylene (PE) is recovered from low-density domestic waste PE carry bags. An 80/100-paving grade asphalt was blended with different PE ratios (10%, 7.5%, 5.0%, and 2.5% by asphalt weight). The blends were tested using Hamburg wheel track tests, resilient modulus tests, indirect tensile tests, and unconfined dynamic creep tests. Test result analysis showed that the PE-modified asphalt mixture exhibited better performance characteristics than a conventional mixture. Including 5 weight percent PE in the asphalt mixture can reduce temperature susceptibility and rutting potential (Punith and Veeraragavan 2007). EVA is a plastomer, which is a copolymer obtained by copolymerization of ethylene and vinyl acetate. Though it is a potential modifier, problems of phase separation have been encountered attributable to the presence of two separate phases of bitumen and polymer that are incompatible with each other. In an effort to ascertain the optimum blending requirements for EVA, an 85–100 binder was modified with varying percentages of EVA from 1% to 7%. Modification was carried out at different combinations of mixing temperature, blending time, and shear rate and a total of 80 combinations were obtained. Further, the paper evaluated the optimum modifier content for obtaining a homogenous blend that could be stable at high tem- peratures. Physical and rheological properties of the modified binder were also evaluated and compared with the base binder. It was found that temperature is the most critical parameter for EVA modification. Shear rate had minimum influence over obtaining a storage-stable blend. Fluorescence microscopy showed a change in morphology as the modifier content increased, which could be used to assess the optimum modifier content for modification. The rheological response of the modified binder significantly improved. EVA modification was found to be best suited at high temperatures (Saboo 2015). Reactive elastomeric terpolymers (RETs) can be used to minimize phase separation by forming polymer networks in blends. An RET is functionalized with glycidyl methacrylate and can crosslink and/or chemically bond with asphalt mol-

44 ecules and functional groups to improve the functional performance of the asphalt. The functional groups improve the rheo- logical performance of the asphalt as demonstrated using Superpave tests. Low-temperature PG cracking resistance and mass loss before and after aging of different types of modified asphalts were assessed. The results show that an RET modifier can significantly improve high-temperature stability and low-temperature crack resistance of asphalt (Cao 2015). The viscosity functions of several polymer-modified asphalts were studied at different temperatures in steady-state rate sweep tests. The materials were obtained by mixing different base asphalts with either SBS, EVA, or RET. In the presence of SBS or EVA, at certain temperatures, the viscosity curves exhibit a Newtonian behavior at low shear rates, followed by two distinct shear- thinning phenomena. In some cases, the first shear-thinning is preceded by a small shear-thickening region. Similar phenom- ena are not present in the viscosity curves of the RET-modified asphalt and can be related to a temporary nature of the physical polymer network (Polacco et al. 2004). A Turkish bitumen was modified with RET, EVA, and SBS polymers. Penetration, penetration index, softening point, ductility, and percent elastic recovery tests were performed with the modified bitumen and raw bitumen. The samples of raw bitumen and modified bitumen with 2% RET, 1% SBS, and 1% EVA were investigated by means of IR spectroscopy (FTIR) and thermogravimetric analysis DTA (TGA/DTA). The penetration and ductility values of the modified bitumen decreased while the penetration index, softening point, and percent elastic recovery increased (Keyf 2015). The use of ground poly (ethylene terephthalate) (PET) particles in asphalt may provide an environmentally friendly solution for the disposal of large quantities of PET waste. The performance of PET as a modifier for asphalt binders was evaluated using rheological and viscosity properties. Tests were performed on the unaged and RTFO aged modified binders with recycled PET particles at contents of 5%, 10%, and 15% by weight of the binders. The addition of recycled PET increased the high-temperature performance. A higher-temperature performance grade was achieved by adding 10% PET. The viscosity and resulting work- ability of the modified binders were not adversely affected for the amounts of PET studied (Shen et al. 2016a). PET-modified HMA cements were prepared using either a dry or wet process; both wet and dry process mixtures contained 10% PET by weight of the base asphalt. Mixture performance tests were performed using an asphalt pavement analyzer (APA) and a retrofitted APA Hamburg test to determine rutting resistance. Moisture susceptibility was estimated by an indirect tensile strength test, and dynamic modulus (E*) was deteremined by asphalt mixture performance tester (AMPT). The wet process mixture exhibited better rutting resistance and a higher tensile strength ratio (TSR) than the control. The dry process mixture exhibited better resistance to permanent moisture damage in APA Hamburg testing and also exhibited a higher TSR than the control. The modified mixtures exhibited lower E* and higher phase angles than the control (Shen et al. 2016a). Crumb Rubber (GTR) With scrap tire stockpiles continually growing, the federal government pushed state agencies and private businesses to develop environmentally friendly ways to dispose of tire waste. Currently, only two states mandate the use of GTR. One way to solve this disposal problem is by grinding or breaking tire rubber into small crumb-like particles to be used in HMA pavements. This crumb rubber material, also known as crumb rubber modifier (CRM), can be blended with HMA mixtures by either a wet process or a dry process (Lo Presti and Airey 2013). In the wet process, also known as the MacDonald process, the rubber is “melted” and blended with the asphalt binder, whereas in the dry process the CRM is added as an aggregate to substitute for a small portion of the fine aggregate. Though dry addition of GTR has had only limited success, recent efforts have been employed to recycle GTR by dry addition in the HMA mixing process using additives and processing aids (Baumgardner et al. 2012). Chemical analysis of the binder–CRM blends is complicated by the insolubility of the GTR in the binder. In the MacDonald process, two factors are critical: (1) the development of performance-related properties and (2) binder compatibility or stor- age stability; the performance-related properties develop early in the process but compatibility may require a few hours to stabilize. Mixing binder with CRM at a high temperature results in a mixture of swollen rubber particles and binder matrix containing the soluble components in the CRM. The swelling and degradation (devulcanization and depolymerization) of the rubber particles leads to improvement in the properties of the binder matrix. However, the swelling and degradation is highly dependent on CRM variables, the binder source, and the mixing conditions (Shen et al. 2009). The surface area and the particle size of the GTR are critical variables in the blending process. CRM is produced by one of two processes, ambient grinding or cryogenic fracture. Ambient grinding operates at room temperature and tears the tire carcasses apart to process particles with a very porous structure. Cryogenic fracture is conducted at temperatures below the glass transition temperature of the rubber (liquid nitrogen) so that the rubber shatters like breaking glass to process particles with smooth surfaces. The difference in the microstructure influences the properties of the CRM binders; a higher PG at high temperature was observed

45 with ambient modified binder than that observed with cryogenic modified binders under the same condition (Shen et al. 2009). A significant analysis of trends reported in the literature regarding GTR reaction time and attempts to dissolve it in asphalt is available (Bahia 2011). A dissertation titled “Characterization and Implementation of Ground Tire Rubber as Post-Consumer Polymers for Asphalt Concrete” gives a comprehensive study of all the variables involved in modifying asphalt with GTR (Baumgardner 2015). BIOBINDERS Bio-based alternatives, which are being developed across the industry in various countries, could be a solution to reduce the asphalt industry’s dependence on petroleum resources. In addition to efforts in providing alternative binders, a trend toward more sustainable pavements has led the pavement industry to place more emphasis on the application of technology to reduce carbon footprints of pavements, including the use of warm-mix asphalt (WMA), half-warm-mix asphalt, and cold-mix asphalt to reduce fuel consumption and CO2 production (Fini et al. 2016). Bio-oil is derived from non-petroleum-based renewable resources such as woody biomass, waste oil, and animal manure; in addition to bio-oils, efforts have been made to convert these waste materials to liquid fuel utilizing different methods such as pyrolysis, fast pyrolysis, and gasification (Mohan et al. 2006). Among commonly used resources to produce bio-modifiers are woody biomass that is also a source for biofuel. Oasmaa et al. (2010) used different wood-based feedstocks and agri- cultural residues to produce biofuel; their study showed that although the liquid yield from the process was relatively high, the accompanying gas component was low compared with that using other agricultural residues, making wood feedstock a promising source for bio-oil production. According to the Energy Information Administration, nearly 9% of all energy con- sumed in the United States is renewable, with 49% of it being derived from biomass. It can be noted that the main focus on the conversion of biomass has been to produce fuel from various sources such as sawdust and cottonseed cake (Ozbay et al. 2006; Salehi et al. 2011; Klabunde and Shrestha 2014). Seidel and Haddock (2012) derived bio-oil from soy fatty acids (SFA); they modified four Strategic Highway Research Program binders and one recycled asphalt binder with 1% and 3% SFA to study the effect of introducing SFA on high-temperature proper- ties of asphalt. Their study concluded that the inclusion of SFA could facilitate reduction of mixing and compaction temperatures. In addition, several studies evaluated the merits of utilizing food crops to produce biofuel and bio-modifiers (bio-oils/ biobinders); however, in order to not create strains and competing demands, it is necessary to look into non-food-crop raw materials. Raouf and Williams (2010) investigated the physical and chemical properties of bio-oils derived from oakwood, corn stover, and switch grass in order to determine their applicability as a bio-modifier for pavement application. They con- cluded that bio-modifiers could be used as a full or partial replacement for asphalt binder; their study further showed that addition of their bio-oils to the asphalt binder increased the stiffness and high-temperature properties of base asphalt binders. Chailleux et al. (2012) performed research on bio-oil derived from microalgae. Their chemical and rheological character- ization showed that microalgae-based bio-oil has similar temperature dependence to asphalt binders and could be a promising candidate for use in asphalt. With the need for advancing waste management practices, it would be much more advantageous to determine the feasibility of producing bio-modifiers from certain waste products in order to further help environmental practices as well as decrease the cost of raw materials to produce biobinders (Wen et al. 2013). Wen at al. evaluated the feasibil- ity of using waste cooking oil to generate a bio-based asphalt. Binder studies utilizing 0%, 10%, 30%, and 60% waste cooking oil showed an increased susceptibility to fatigue and rutting. Mixture tests also showed a reduction in dynamic modulus and thermal cracking (Wen et al. 2013). Recently, several researchers have shown that using a bio-modifier along with the petroleum-based asphalt could produce a bio-modified binder with enhanced performance (Williams et al. 2009; Fini et al. 2012). Although there have been several studies on the effect of introduction of various bio-modifiers to asphalt binders, their variational impact on the physicochemi- cal characteristics of the base asphalt before and after oxidative aging has not been fully studied. It is important to evaluate several bio-oils derived from different raw materials in order to determine the merits of their application while conducting a comparative study. The effects of introduction of four different bio-modifiers (biobinders) on the rheological and chemi- cal properties of a selected asphalt binder (PG 64-22) before and after oxidative aging have been published (Fini et al. 2011, 2016). Overall, bio-modifiers were found to be significantly different in terms of their aging characteristics. Accordingly, their surface and rheological properties were found to be ranked differently before and after aging when compared with those of a control asphalt binder. The results showed that the BB from swine manure is less susceptible to aging relative to plant-based

46 bio-oils. This can be further attributed to the chemical structure and the high lipid contents of the BB from swine manure, making it less affected by oxidative aging (Fini et al. 2016). Fini et al. (2011) synthesized a bio-oil from swine manure. They further studied the effect of introduction of the bio-oil to a control asphalt at 2%, 5%, and 10% (by weight of base asphalt). Their rheological characterizations showed a clear trend in improving asphalt workability and low-temperature properties resulting from the introduction of the BB from swine manure. They further examined the merits of adding their BB (5%) in several asphalt mixtures along with various RAP and recycled asphalt shingles (RAS) percentages; they showed the introduction of their BB was effective in compensating the stiffening effect of RAP and RAS, resulting in more workable mixtures while improving the mixtures’ low-temperature cracking prop- erties (Fini et al. 2011). Yang et al. (2013) evaluated the use of asphalt binder modified by bio-oils generated from wood waste. Samples were blended at 5% and 10% original bio-oil, dewatered bio-oil, and polymer-modified bio-oil, by weight of the base binder. The results concluded that the addition of bio-oil can reduce the mixing temperature and improve high-temperature performance while compromising low-temperature performance. Another major concern noted by the authors is the aging susceptibility of the bio-oils. Williams and McCready (2008) characterize the rheological properties and aging mechanism of asphalt binders blended with high percentages of biobinders using FTIR. The petroleum asphalt was partially replaced by the biobinders at fractions of 30% and 70% by weight. Rotational viscometer and dynamic shear rheometer tests were conducted for the rheological properties. Loss of volatiles was obtained from the RTFO test, whereas the oxidation was investigated by an FTIR test. The rheological results showed that the bioblended asphalt binders exhibit different rheological properties as compared with the control asphalt binder before and after the RTFO aging. The mass loss test showed that biobinders had a much greater loss of volatiles than the control asphalt binder. FTIR spectra analysis confirmed that additional C=C, C-O, C=O, and OH bonds were generated during the aging. Further chemical analysis revealed that the aging of a biobinder can be attributed to three processes: the loss of volatiles, dehydrogenation/condensation that forms higher-MW molecules such as asphaltenes, and the oxidation to produce carboxylic acids, alcohols, and esters (Yang et al. 2015). Morphological features and structural characteristics of asphalt binders are strongly affected by factors such as aging, which can alter the performance of petroleum-based asphalt binders. Biobinder (BB), a newly produced amide-enriched bio-adhesive obtained from biomass, has been found to be promising for reducing the negative effects that aging can have on molecular conformation. Doping of BB into a commonly used petroleum-based asphalt binder creates a bio-modified binder that has experimentally performed especially well at low temperatures. This improvement can be attributed to the effect of modifier fragments, which contain high concentrations of amide functional groups, on the π–π stacking of asphaltenes. Disturbing the π–π interactions alters the stacking distance, the corresponding binding energy, and ultimately the extent of clustering of asphaltene units. Any change in the clustering of asphaltene affects the rheology and morphological proper- ties of asphalt, which in turn alters the asphalt’s performance, including but not limited to its resistance to fatigue and low- temperature cracking (Mousavi et al. 2016). OTHER NONBITUMINOUS MODIFIERS Low-MW modifiers are used for the specific purpose of improving the asphalt binders’ performance in a given environment. These materials are often proprietary, but they can be classified by their intended application; that is, antioxidants, hydrocar- bon supplements, antistripping agents, and stiffening agents. Antioxidants Oxidation is the primary cause of long-term aging in asphalt pavements. As a pavement oxidizes, it stiffens and can eventu- ally crack. The use of an antioxidant as a performance enhancer in an asphalt binder could delay aging and thus increase the life of an asphalt pavement. Experience with antioxidants to date is limited. Laboratory evaluations have been conducted but there is a dearth of adequate field experience and validation. Some of the most recent efforts to use antioxidants are cited here. Investigating the cracking potential in asphalt binders containing various antioxidant levels through a rheological test series suggests that there can be a significant reduction in asphalt pavement cracking through antioxidant use. Laboratory- prepared mixtures of asphalt containing antioxidants were subjected to short-term and long-term aging simulation. The

47 samples were evaluated using dynamic modulus tests and creep tests in the indirect tensile mode. Testing and analysis results indicate a significant reduction in asphalt mixture cracking resistance through asphalt aging. Predicted cracking susceptibility was substantially lower through antioxidant use (Apeagyeai et al. 2008). Lignin is a readily available, well-studied antioxidant. Four lignin-containing coproducts were combined with four asphalt binders in varying amounts to discover the optimum amount of coproduct that would provide the greatest benefit to the asphalt binders. The asphalt-coproduct blends were evaluated according to Superpave specifications and performance graded on a continuous scale. The data indicate a stiffening effect of the binder caused by the addition of coproduct; the more coproduct was added, the greater the stiffening. Binder stiffening benefits high-temperature properties, while the low-temperature binder properties are negatively affected. However, the low-temperature stiffening effects are small. Instead of acting purely as a filler and shifting the use temperature range, the lignin has an overall effect of widening the temperature range of the binders. Testing reveals that the lignin in the coproducts benefits the intermediate- and low-temperature properties of the binders. Oxidative aging products were evaluated and some antioxidant effects were noticed (Williams and McCready 2008). The oxidation mechanism in polymer-modified bitumen tends to alter physical and chemical properties. Dessouky et al. (2015) evaluated five antioxidants to examine their potential to mitigate polymer-modified bitumen oxidation. Styrene copo- lymers and hindered phenol additives were evaluated at low, intermediate, and high temperatures using aging indices. The styrene copolymers were effective at high temperatures while the hindered phenols were effective at low and intermediate temperatures. Two blends of both additives were introduced to cover a wide spectrum of temperatures. Frequency and tem- perature sweep testing suggested that the blend improved the recoverable imposed energy at a wide range of temperatures/ frequencies. The blends have also improved resistance to rutting and moisture susceptibility for asphalt mixes. A further study evaluated the effect of antioxidants on retarding of asphalt mixture aging using the GPC technique. An antioxidant, hydrated lime, and linear low-density polyethylene were used when preparing the dense-graded asphalt and the stone-mastic asphalt mixes, and the effect of aging retardation was evaluated using GPC. The asphalt mixes were aged artificially in the laboratory in two stages: short-term aging for 1 hour at 160ºC (170ºC) and long-term aging for 164 hours at 76ºC. It was found that the antioxidant and hydrated lime were effective on retarding aging of asphalt mixtures when the large molecular size ratio was compared before and after aging treatment. The estimated viscosity levels of antioxidant- and hydrated-lime-added asphalt mixtures were found be reduced by 27% of normal dense-graded asphalt mix after long-term aging. The normal and LLDPE-modified stone-mastic asphalt were found to show even further age-retarding effects (by 42%) without any antioxidant (Kwon et al. 2016). Antistripping Agents Antistripping agents are used to minimize or eliminate stripping of asphalt cement from the aggregate in HMA mixtures. Both liquid antistripping additives and lime additives are used to resist stripping. Most liquid antistripping agents are surface active agents that, when mixed with asphalt cement, reduce surface tension and, therefore, promote increased adhesion to the aggregate. The chemical composition of most commercially produced antistripping agents is proprietary. However, the major- ity of antistripping agents currently in use are chemical compounds that contain amines. Many antistripping agents have been used in asphalt mixtures in the past, including amidoamines, imidazolines, polyamines, hydrated lime, organo-metallics, and acids. These antistripping agents must be heat stable; that is, they will not lose their effectiveness when the modified asphalt cement is stored at high temperatures for a prolonged period of time. Of these products, the amines and hydrated lime have been used most commonly. In all cases, the purpose of these products is to inoculate the mixture against moisture damage, often called stripping. Many liquid antistripping compounds have an objectionable odor. New formulations are less objectionable, but are typically more expensive. Considerable research is ongo- ing among various industry groups to develop products that promote adhesion between the asphalt binder and the aggregate in the presence of moisture. Since the late 1970s much research has been done to better understand the stripping phenomenon in asphalt mixtures. As a result, there have been changes in both materials and technology over the past 30 years to improve asphalt mixtures’ resistance to moisture damage and the ability to test for performance under adverse moisture conditions. Because of changes in materials and technologies related to antistripping agents, a research study was conducted to evaluate the effectiveness of current antistripping agents used in hot-mix asphalt pavements. The objectives were to construct a field test section that used three different antistrip- ping agents in a conventional Superpave surface mixture and conduct a series of laboratory performance test comparisons using different aging periods to make long-term comparisons of the effectiveness of hydrated lime, liquid additive, and warm-mix

48 asphalt antistripping additives. Sets of zero, one, five, and 10 freeze–thaw cycles were used for a portion of the research study. The results showed that hydrated lime had the highest tensile strength and highest TSR values and was the only additive treat- ment to meet the minimum of 80% TSR for all freeze–thaw cycle combinations. Both five and 10 freeze–thaw cycles were sig- nificantly more discriminating for moisture susceptibility than one freeze–thaw cycle alone. Warm-mix asphalt treated mixtures produced low initial tensile strengths, but the strength of these mixtures improved with time (Watson et al. 2013). An important material property that influences the performance of an asphalt mixture is the surface free energy of the asphalt binder and the aggregate. Surface free energy governs the adhesive bond strength between the asphalt binder and the aggregate as well as the cohesive bond strength of the asphalt binder. These bond energies in turn influence the resistance of the asphalt mixture to distresses such as fatigue cracking and moisture-induced damage. Asphalt binders undergo several types of engineering and natural modifications that influence their chemical and mechanical properties. Three common exam- ples of modifications are the addition of polymers, addition of additives (e.g., antistripping agents), and oxidative aging of the asphalt binder. Bhasin et al. (2007) conducted a study examining the effect of different types of modifications on the surface free energy components of the asphalt binder. The change in surface free energy was used to calculate parameters related to the performance of the asphalt mixtures. Results from this study demonstrate that the magnitude and nature of change to the surface free energy and concomitant performance-related parameters varied significantly among different asphalt binders. Adding liquid antistripping agents typically reduced the surface free energy and consequently the work of cohesion of the asphalt binders. This modification can indirectly improve fracture resistance by promoting better adhesion between the fine aggregate particles and the binder during the mixing and compaction process. Use of liquid antistripping agents either improved or did not significantly change the moisture resistance of the asphalt binder with the selected aggregates (gauged using the parameter elastic recovery). The liquid antistripping agents from the two sources demonstrated different levels of changes in moisture resistance when used with the same combination of asphalt binder and aggregate. In most cases, long-term aging reduced the work of cohesion and thus indicated lower fracture resistance of the aged binder. In the case of one unmodified binder and one modified binder, the work of cohesion increased after long-term aging. At that time, asphalt binders from one source demonstrated a decrease in moisture sensitivity, while asphalt binders from the other source demonstrated an increase or no change in moisture sensitivity with the two aggregates used in this study. The difference in the behavior of the two asphalt binders is attributed to the influence of aging on the magnitudes of the polar functional groups (Bhasin et al. 2007). Stiffening Agents The modification of asphalt binders to improve performance properties has grown significantly since the implementation of the SHRP binder specifications. The use of polymers and crumb rubber modifiers has increased. To compensate for the loss of stiffness at high temperature ranges, stiffening agents such as polyphosphoric acid and gilsonite are added. Polyphosphoric Acid Several highway agencies have been concerned about the performance characteristics of PPA modification and possible nega- tive interactions with other mix components, such as lime and liquid antistripping agents. A workshop on Polyphosphoric Acid Modification of Asphalt Binders was held in April 2009 in an attempt to pull together the facts about PPA-modified asphalt and performance. The PPA workshop covered extensive laboratory and field evaluations on the use of PPA as a modifier for asphalt binders. The following points were made by various workshop participants during the discussions: 1. The stiffening effect of PPA on the binder is crude source dependent with anywhere from 0.5% to more than 3% needed to increase the binder grade. 2. PPA works as a stiffener and crosslinker when used with polymers such as SBS and ethylene terpolymers (e.g., Elvaloy). 3. PPA can significantly improve the delayed elastic response of the polymer-modified binder. 4. There is some indication that hydrated lime can somewhat reduce the stiffening effect of PPA but the increased stiffen- ing from the lime outweighs any loss. 5. Limestone aggregate could not reverse or reduce the stiffening effect of PPA on the binder (D’Angelo 2012).

49 PPA has been used in 3.5% to 14% of the asphalt placed in the United States over the past 10 years. This represents up to 400 million tons of hot mix (Fee et al. 2010). Experienced industry practitioners have found that the addition of small amounts (about 0.5%) of PPA to polymer-modified binders improved both their handling and performance. When used with styrene- butadiene-styrene polymers it enables suppliers to achieve higher Superpave PG while improving mixing and compaction characteristics. With ethylene terpolymers, PPA catalyzes the reactivity of the glycidyl methacrylate groups. Both types of polymer-modified binders have shown the addition of PPA to increase rut resistance of the binder. More recently, the increas- ing popularity of PPA has led to its use as a partial replacement for polymer modification (Arnold et al. 2012). Used in conjunction with polymers, PPA enables suppliers to achieve performance grades that can be handled, mixed, and compacted at reasonable temperatures. Both non-polymer-modified and polymer-modified plus PPA-modified binders have shown improved rutting resistance with the addition of PPA. Though successful PPA modification of asphalt binders has a long track record, its use is often debated, sometimes to the point that PPA-modified asphalt binders have been banned. Opinion has it that such actions result from misinformation as well as lack of understanding of the benefits of PPA as an available tool to improve the performance of asphalt binders. Although PPA is banned in some states, others continue its use without issue. Baumgardner (2010) discusses common asphalt binder specifications, the relationship of asphalt binder chemical composition to PG properties of asphalt binders, necessary use of PPA modification to meet specifications, the relationship of asphalt composi- tion to PPA loadings necessary to achieve desired properties, expected PG enhancements of PPA-modified asphalt binders, and effects of using PPA in non-polymer-modified and polymer-modified asphalt binders to meet current PG specifications. The stiffening effect of phosphoric acid was found to be dependent on the particular asphalt being modified. Asphalt binders from eight different sources were tested: AAD-1, AAK-1, AAM-1, ABM-1, two asphalts from Venezuela provided by Citgo (a 60% Bachequero and a 94% Bachequero), an asphalt from BP Whiting Refinery, and an asphalt from Holly Cor- poration. Of these binders, AAK-1 (Boscan) exhibited the greatest reactivity to phosphoric acid, whereas ABM-1 (California Valley) was the least reactive and showed only a very slight increase in stiffness even at high dosage levels (Arnold et al. 2012). The reactivity appears to be related to the effect of asphaltene dispersion. Asphalts with highly dispersed asphaltenes (sol asphalts) are strongly impacted; asphalts with more condensed asphaltenes (asphalt gels) show little reaction. As with all other components of the mix, testing is required to demonstrate the performance of PPA with each formula- tion of asphalt and aggregate, together with polymer, antistripping agents, and other additives that may be used. Results of the following tests are presented: dynamic shear rheometer, Hamburg, Lottman, and multiple stress creep recovery tests on a matrix of a common asphalt with aggregate, three antistripping agents, two types of polymers, and PPA. In cases supported by laboratory data for the materials tested, the performance of PPA-modified asphalt can be improved with the addition of antistripping agents such as a phosphate esters, a particular polyamine compound, and hydrated lime. These findings hold true for cases where modification includes the use of polymers: styrene-butadiene-styrene and Elvaloy (Fee et al. 2010). Gilsonite In tropical countries, roads built with asphalt layers must be made with bituminous mixtures containing asphalt that is rea- sonably stiff, to increase resistance against permanent deformations; that is, rutting. Gilsonite (10 weight percent) modified HMAs were prepared using either wet or dry processes. Gilsonite increases stiffness and improves the performance grade of a virgin binder at high temperatures of service. However, at low temperatures, the mixture could experience embrittlement, thus decreasing its resistance to low-temperature fatigue cracking (Quintana et al. 2016). Sixty to 70 penetration-grade asphalt cement and Iranian natural bitumen (gilsonite) were subjected to physical and per- formance tests. Physical and conventional tests were conducted on modified and unmodified bitumen (penetration, softening point, ductility, and viscosity). Performance tests, including Marshall stability, indirect tensile strength, moisture suscepti- bility, resilient modulus, and rutting resistance, were conducted on unmodified and modified stone-mastic asphalt mixtures. Results of conducted tests on the modified asphalt binders with 5%, 10%, and 15% gilsonite show that using gilsonite improves performance of stone-mastic asphalt mixtures (Babagoli et al. 2015). Gilsonite was added to the biobinder, and rheological characterization of each bio-modified gilsonite (BMG) sample was then conducted to investigate its performance. Results indicate that BMG samples show improved high-temperature perfor- mance as well as comparable low-temperature performance and adhesion compared with PG 64-22. In some cases, BMG samples show even better performance at low temperatures as evidenced by higher m-values and lower creep stiffness when compared with PG 64-22. Specifically, 30% BMG had the most effective low-temperature properties (−12ºC) among samples studied. Furthermore, FTIR spectra were collected to compare the chemical functional groups of PG 64-22 with those of

50 BMG. Bio-modified gilsonite can be a promising candidate for use in asphalt mixtures because of its enhanced rheological properties (Yaya et al. 2016). A detailed study of the chemical composition of gilsonite has been reported. Gilsonite, naturally occurring asphaltic bitumen, consists of a complex mixture of organic compounds. The gilsonite was characterized by elemental analysis to determine the concentrations of carbon (C), hydrogen (H), nitrogen (N), sulfur (S), and oxygen (O), by FTIR for compara- tive analysis of the chemical structures; by nuclear magnetic resonance spectroscopy of hydrogen (1H NMR) to ascertain the aliphatic and aromatic hydrogen fractions; and by thin layer chromatography-flame ionization detection (Iatroscan TLC-FID) to quantify saturated and aromatic hydrocarbons, and resin/asphaltene fractions (Nciri et al. 2014). Asphalt Recycling Asphalt recycling has become an important instrument used to minimize production costs of new pavements as well as to mitigate their effects on the environment. Furthermore, stricter environmental regulations and depleting resources have led to the exploitation of recycled materials by promoting the application of higher RAP and RAS in new mixtures. Some of the benefits of utilizing recycled materials include the conservation of nonrenewable natural resources such as virgin aggregates and asphalt binder, reduction in the amount of construction debris disposed of in landfills, decreased variability in material expenditures, and potential reduction of the overall life-cycle cost. Recycling also helps to cut greenhouse gas emis- sions by reducing the energy spent on extraction and processing of petroleum products and aggregates. Moreover, the increas- ing price of asphalt binder along with more restrictive environmental legislation has forced highway agencies and contractors to search for novel materials and construction techniques. Such efforts are aimed at fulfilling the current sustainability needs without compromising the pavement quality and performance. There is, at this time, considerable emphasis on the use of RAP as the preferred recycled material for highway construction as a result of its abundance and successful prior experiences. Collins and Ciesielski (1994), in an NCHRP Synthesis of Highway Practice, noted that highway agencies have been proactive in the recycling of reclaimed and by-product materials into construction materials, with RAP being the material most frequently used. RAS, defined by AASHTO MP 23-14, Standard Specification for Use of Reclaimed Asphalt Shingles as an Additive in Hot-Mix Asphalt (HMA), as “any type of waste roofing asphalt shingles that have been processed into a recyclable product,” have become another promising recycling candidate. The utilization of RAP and RAS in modified asphalts has been reviewed extensively (Stroup-Gardiner and Wattenberg-Komas 2013; Zhou et al. 2014; del Barco Carrión et al. 2015; Hoppe et al. 2015; Hossain et al. 2015; Lee et al. 2015; Stroup-Gardiner 2016). A detailed discussion of these modifiers is beyond the scope of this synthesis. Recycling Agents/Softening Agents With increased interest in RAP and RAS, the use of recycling agents (RAs) is considered essential for softening and/or reju- venating aged and stiff binders in RAS. Recycling agents are classified as two types: rejuvenating agents and softening agents. Softening agents lower the viscosity of the aged binder, while rejuvenating agents are intended to restore the rheological and chemical properties of the aged binder (Im et al. 2014). Most recycling agents are proprietary mixtures, but generic compounds can be identified. Examples of softening agents include asphalt flux oil, lube stock, and slurry oil. Examples of rejuvenating agents include lubricating and extender oils, which contain a high proportion of maltenes and low saturates contents that do not react with asphaltenes (Im et al. 2014). The effectiveness of the rejuvenators depends on the mixture types and engineering properties evaluated. Rheological properties may be improved, but reversing the chemical effects of extensive aging is unlikely. The hot-mix recycling operation for bituminous mixes commonly uses a recycling agent to restore aged asphalt cement to a rheological condition that resembles a virgin asphalt cement. In the laboratory this is done on the extracted asphalt cement, thus ensuring thorough mixing. In the actual recycling operation, however, the RA is added to the material during the mixing process and merely coats the salvaged asphalt concrete particles that are being recycled. A certain amount of time is required for the RA to combine with the old asphalt and redistribute the older component throughout the modified binder. Carpenter and Wolosick (1980) investigated the influence of this diffusion process on material behavior shows that the diffusion process exerts a strong influence on the material properties required for long-term performance. Immediately after sample prepara- tion, the mix may have high stiffness and excellent resistance to rutting. A week after preparation the stiffness had decreased by a factor of two, and the resistance to rutting had decreased accordingly. This behavior is explained by a predicted rate of the diffusion process. The redistribution of components is physically verified by extracting the outer and inner layers of the modified asphalt concrete prepared in a simulated recycling operation and by comparing their consistency. Understanding this phenomenon is critical for long-term performance predictions and in evaluating the effects of various laboratory conditioning procedures (Carpenter and Wolosick 1980).

51 Shen et al. (2007a) investigated the effects of an RA on properties of recycled asphalt binders and mixtures by adding varying dosages of RAs. They found that the RAs significantly affected the properties of both recycled aged binders and their mixtures. They also noted that the optimum percentages of the RAs could be obtained by satisfying SHRP specifications and that they varied according to the neat binder properties. RAP rejuvenator-containing binder blending charts were developed. Based on asphalt pavement analyzer evaluation of mixture rutting, as well as indirect tensile strength; mixtures incorporating 10% RAP; and RA mixtures were superior to HMA prepared with RAP blended with softer binders (Shen et al. 2007a, b). Mogawer et al. (2013) queried if asphalt recycling agents can offset the stiffness attributed by the hardened binder from RAP and RAS in mixtures that incorporate high RAP and RAS contents and if RAs can help the hardened binder from the RAP/RAS comingle with the virgin binder. Overall, the results showed that asphalt RAs can mitigate the stiffness of the resul- tant binder. The cracking characteristics of the mixture improved by the addition of the RAs; however, rutting and moisture susceptibility were adversely impacted at the dosage and the testing conditions used. The molecular composition of asphalt binders obtained from asphalt mixtures containing either RAP, RAS, or mixtures of the two were characterized using gel permeation chromatography, the extent of aging using FTIR, and fracture resistance of laboratory- produced mixtures using the Semi-Circular Bending test at intermediate temperature. Molecular fractionation through GPC of RAS samples confirmed the presence of associated asphaltenes in greater concentrations than RAP pavement samples. High concentra- tions of high-MW asphaltenes decrease the fracture resistance of asphalt mixtures. The use of recycling agents—Cyclogen-L (a mix- ture of naphthenes) and Hydrogreen (vegetable-derived oil)—did not reduce the concentration of the highly associated asphaltenes, and thus they failed to improve the cracking resistance of the asphalt mixtures evaluated in this study (Cooper et al. 2015). REOB Asphalt Extenders/Softening Agents Refined engine oil bottoms are obtained from the refining of recovered engine oil and have been used in the asphalt paving industry since the 1980s. To obtain the low-temperature properties required in an asphalt binder, REOBs are typically used from 3% to 10% by weight of the binder because REOB generally softens the base binder when used. When asphalt mixtures are being designed with high RAP and RAS contents, resulting in stiffer binder blends because of the available recycled asphalt binders, REOB may also be used as an RA in order to improve the cracking resistance properties of the asphalt mix- ture given the stiffer composite binder blend (DeDene and You 2014). FIGURE 19 Molecular weight (MW) distribution of molecular species of extracted 70 PG binder containing 15% refined engine oil bottoms and 5% recycled asphalt shingles (70PG5P_B-RA1) (Source: Cooper et al. 2015). The use of REOB in paving mixtures as a rejuvenating agent was examined. A series of mixtures were prepared containing increasing REOB amounts: 5%, 10%, and 15%. The distribution of molecular species in REOB is concentrated in the domain of asphalt maltenes. The higher molecular weight components of REOB contribute to the asphaltene domain, with a tail toward the species of high-MW polymers (Figure 19). It appears that REOB acts only as a diluting agent, which is effective in extracting

52 the aged binder from RAS material. For example, the binder extracted from the mixture containing 15% REOB contained all the asphalt binder present in RAS (100% RAS extraction; compared with only 36% availability for the mixture with no REOB). Less RAS binder extraction was observed when the REOB addition to the mixtures dropped to 10% and 5%. However, the REOB content did not improve the cracking resistance; Jc remained below the threshold limit of 0.5 kJ/m 2. The extracting power of REOB is reflected by the increase of the concentration of associated asphaltenes originating from RAS; that is, species with apparent size of associated asphaltenes originating from RAS; that is, species exceeding apparent MW >40K Daltons (Figure 19). High-MW-associated asphaltenes are not significantly dissociated by adding rejuvenators. Use of rejuvenators negatively impacted intermediate-temperature performance for the mixtures evaluated in this study (Cooper et al. 2015).