Published in "Environmental Degradation of Advanced and Traditional Engineering Materials", Editor(s): Lloyd H. Hihara, Ralph P.I. Adler, Ronald M. Latanision, CRC Press, 2013 (http://www.crcpress.com/product/isbn/9781439819265#!) CHAPTER 29. ASPHALT PAVEMENT DURABILITY Didier Lesueur1 and Jack Youtcheff2 1 Lhoist Recherche et Développement rue de l'Industrie, 31 B-1400 Nivelles (Belgium) 2 Turner-Fairbank Highway Research Center 6300 Georgetown Pike McLean, VA 22101 Asphalt Durability Contents 1. 2. 3. Introduction ....................................................................................................................4 Background ....................................................................................................................5 Pavement Deterioration ..................................................................................................7 3.1 Deterioration due to Material Chemical Aging ..............................................................7 3.2 Deterioration due to Climate .........................................................................................7 3.2.1 Moisture Damage and Frost ...................................................................................7 3.2.2 Thermal Cracking ..................................................................................................8 3.3 Deterioration due to Traffic...........................................................................................8 3.3.1 Rutting ...................................................................................................................8 3.3.2 Fatigue Cracking ....................................................................................................8 3.3.3 Loss of Skid Resistance..........................................................................................9 3.3.4 Other degradation modes........................................................................................9 3.4 Observed Field Durability .............................................................................................9 4. Laboratory Evaluation of Asphalt Mixture Durability ...................................................11 4.1 Deterioration due to Material Chemical Aging ............................................................11 4.1.1 Asphalt Binder Conditioning………………………………………………………11 4.1.2 Asphalt Mixture Conditioning……………………………………………………..11 4.2 Deterioration due to Climate ....................................................................................................... 11 4.2.1 Moisture Damage and Frost................................................................................................. 12 4.2.2 Thermal Cracking ................................................................................................12 4.3 Deterioration due to Traffic.........................................................................................12 4.3.1 Rutting .................................................................................................................12 4.3.2 Fatigue Cracking ..................................................................................................12 4.4 Practical use of the Lab Tests: Specifications ..............................................................13 4.4.1 Binder Specifications ...........................................................................................13 4.4.2 Aggregate Specifications......................................................................................14 4.4.3 Asphalt Mixture Specifications ............................................................................15 5. General Properties of Asphalt Mixtures ........................................................................17 5.1 Composition ...............................................................................................................17 5.1.1 Asphalt Cement: Definitions ................................................................................17 5.1.2 Asphalt Cement: Physico-Chemical Properties .....................................................17 5.1.3 Manufacturing Process .........................................................................................19 5.2 Properties....................................................................................................................20 5.2.1 Mechanical Properties ..........................................................................................20 5.2.2 Physical Properties ...............................................................................................22 6. Distress Mechanisms ....................................................................................................22 6.1 Asphalt Chemical Aging .............................................................................................22 6.1.1 Oxidative Hardening ............................................................................................23 6.1.2 Physical and Steric Hardening ..............................................................................24 6.1.3 Exudative Hardening ............................................................................................25 6.2 Microb / Vegetal Attack ..............................................................................................25 7. Enhancing Longevity Performance of Asphalt ..............................................................27 7.1 Improved Mix Design and Construction......................................................................27 7.1.1 Mix Design ..........................................................................................................27 7.1.2 Construction.........................................................................................................27 7.1.3 Warm Mixes ........................................................................................................28 2/63 Asphalt Durability 7.2 Additives ....................................................................................................................29 7.2.1 Polymer ...............................................................................................................30 7.2.2 Acid………………………………………………………………………………..35 7.2.3 Hydrated Lime .....................................................................................................31 7.2.4 Others (liquid antistrips, fibers, epoxy asphalt) .....................................................32 8. ASTM / AASHTO Tests ...............................................................................................32 9. Summary ......................................................................................................................36 10. Acknowledgement ....................................................................................................37 11. Further Reading ........................................................................................................37 13. Nomenclature and Units............................................................................................37 14. Tables .......................................................................................................................38 15. Figures .................................................................................................................... . 49 12. References ................................................................................................................58 3/63 Asphalt Durability 1. INTRODUCTION In a World concerned with environmentally friendly construction technologies and sustainability, asphalt durability is taking on an added significance. Worldwide asphalt consumption was on the order of 108 Mt in 2010 (USA 33 Mt) which accounted for 2.4% of the total oil demand. Asphalt’s use as a paving material accounts for roughly 80% of this and is the focus of this chapter. Asphalt mixtures have greatly evolved from their beginning in the late 19th century to modern times (Lay 1992). They are now becoming very technical materials obtained by careful selection of raw materials among which are asphalt binder and aggregate, and then mixed in dedicated plants whose operations are computer-controlled in order to optimize product quality and minimize energy consumption, waste and emissions (Lee and Mahboub 2006). Today roughly 90% of the roads in the USA are paved with asphalt. Due to the rising cost of asphalt binders, there is a keen interest in the concept of sustainable asphalt mixtures. However, the environmental aspects are generally covered with a somewhat narrow view, consisting essentially of (i) minimizing wastes through the recycling of Reclaimed Asphalt Pavements (RAP) in new production (Dunn 2001), (ii) supplementing the asphalt through the use of other by products such as reclaimed asphalt shingles (RAS), sulfur, and recycled motor oil, and (iii) using lower manufacturing temperatures in the manufacturing of the so-called warm or even semi-warm mixtures, thus decreasing CO2 emissions and energy consumption (Button et al. 2007, D´Angelo et al. 2008). Clearly, the basic principle of waste management, i.e. preventing and reducing waste generation, is not fully taken into account by the industry. This is all the more surprising as the European Waste Directive (Directive 2008/98/EC) or the waste hierarchy rule known as the 4 R´s (Reduce, Reuse, Recycle and Recover); all converge on the primary objective: make your mixtures more durable! Therefore, this chapter tries to fill a gap in our knowledge about asphalt binders and mixtures by addressing the main factors affecting their durability, namely, their ability to maintain satisfactory performance and structure in long-term service. We review the current understanding of relevant parameters and provide some advice on how to maximize durability. 4/63 Asphalt Durability 2. BACKGROUND Asphalt’s usage dates back 180,000 years ago to the El Kowm Basin of Syria, where it was applied to stick flint implements as handles for various tools; this application persisted up until Neolithic time (Connan 1999). In this earliest record and in many of the later uses, the adhesive and waterproofing properties of local sources of natural asphalt were generally recognized. Examples such as the waterproofing of Noah's arch, of the Babel tower, or of the cradle of Moses are given in the Bible (Abraham 1960). Medical uses were also reported, with asphalt acting as a remedy for various illnesses (trachoma, leprosy, gout, eczema, asthma, etc.) as well as a disinfectant or as an insecticide (Abraham 1960). Another well studied historical application was for the embalming of mummies by the Egyptians (Abraham 1960). The first mention of the use of asphalt in road construction dates back to Nabopolassar, King of Babylon (625-604 BC); an asphalt-containing mortar cemented both the foundation made of three or more courses of burnt bricks as well as the stone slabs that were placed on top (Abraham, 1960). However, asphalt essentially disappeared from pavement applications until the early 19th century, when a rediscovered European source of natural asphalt led to the development of the modern applications for this material (Lay 1992). The use of natural asphalt in road construction started to decline in the 1910s with the advent of vacuum distillation which made possible the manufacture of asphalt from crude oil. Nowadays, paving grade asphalt is almost exclusively obtained as the vacuum residue of petroleum distillation (Lesueur 2009). The word bitumen is mostly used in Europe as a synonym for the terms asphalt binder and asphalt cement used in the U.S. In this chapter, we will adhere to the American usage and refer to it as asphalt binder. The asphalt binder is blended with mineral aggregates to form asphalt mixtures (also called bituminous mixes, asphalt concrete or bituminous concrete). The binder serves two purposes; one is to glue the aggregate together, and the other is to protect the aggregate from environmental distresses. Hot mix asphalt (HMA), the most prevalent paving product, is generally fabricated by heating typically 5 wt.% asphalt binder up to around 160ºC in order to decrease its viscosity and then blending it with 95 wt.% of hot aggregate (Monismith 2006). The specifications for paving asphalt generally follow this application. Other techniques to manufacture asphalt mixes using asphalt emulsions or foamed asphalt also exist; these amount to less than 5 % of the total asphalt mix production but their percentages are increasing. To produce HMA that resists climate and increasing traffic demands, specifications on paving grade asphalts are becoming more stringent (Read and Whiteoak 2003). The properties that are needed to obtain a suitable asphalt binder are largely rheological. First, the asphalt has to be fluid enough at high temperature (around 160°C) to be pumped and workable to enable a homogeneous coating of the aggregates upon mixing. Second, it has to be stiff enough at the highest pavement temperature to resist rutting (around 60°C, depending on local climate). Third, it must be pliable enough and sufficiently strong at the lowest pavement temperature (down to around -20°C, depending on local climate) to resist cracking. All these properties operate at cross purposes, and it is therefore difficult to obtain an asphalt binder that performs well under all possible climates. As a consequence, different paving grades exist, the softer being generally suitable for cold climates and the harder, for hotter regions. In order to widen 5/63 Asphalt Durability the temperature range of asphalt, additives such as polymers and/or acids are finding increased usage (PIARC 1999, Lesueur 2009 and McNally 2011). 6/63 Asphalt Durability 3. PAVEMENT DETERIORATION 3.1 Deterioration due to Material Chemical Aging The chemical aging process can be separated into two stages, one during the manufacture of HMA, and the second during the service life of the pavement. The consequence of both processes is the same: there is a global hardening of the asphalt binder (Wright 1965, Bell 1989), which in turn increases its cracking propensity (Isacsson and Zeng 2003) and embrittlement of the asphalt mixture. This risk of cracking increases, whether attributed to thermal factors or traffic-induced fatigue. In some instances oxidation products are formed that may enhance the moisture sensitivity of the asphalt mix and induce stripping or rutting (as discussed below). 3.2 Deterioration due to Climate 3.2.1 Moisture Damage and Frost Moisture induced-damage and the effect of freeze-thaw cycles are common phenomena with asphalt mixtures. They generally materialize through the progressive loss of aggregate as illustrated in Figure 1. The asphalt-aggregate bonds weaken in the presence of water to the point that the cohesive and adhesive bonds are no longer strong enough to hold the aggregate in place. This results in aggregate stripping or ravelling when it is limited to the surface. This is induced by environmental factors and worsened by traffic. Flushing is another type of water damage that similarly leads to the loss of aggregate; this mechanism occurs in the bulk of the material as a consequence of the traffic-induced internal water pressures. If left untreated, such damage can deteriorate into potholes. Frost and freeze-thaw cycles tend to enhance these detrimental effects, thus a tough winter can directly generate potholes. According to a USA survey taken in the early 1990s (Hicks 1991), water-induced damage on average is evident in untreated HMA between 3 to 4 years post construction, sometimes this can be manifested in the very first year. Moisture damage arises when water forms a thermodynamically favored film at the asphaltaggregate interface thereby displacing the asphalt binder from the aggregate surface. In addition water scour can subject the pavement to both adhesive and cohesive induced damage (Kringos et al. 2008a, 2008b). Ice formation is also a contributor as the associated volume change can damage the material (Mauduit et al. 2010), and enhance the further uptake of water. While some asphalt oxidation products may contribute to the adhesiveness of the binder to the aggregate (Curtis et al 1989a, 1989b), other functional groups may be converted to hydrophilic and surface active compounds and attract moisture such as sulfones (functionalities found in soaps, etc.). The presence of such groups can adversely affect the adhesive properties and freeze thaw tolerance of the binder. The presence of surface active species in the aggregate, typically clay contaminants, has been linked to moisture damage (Aschenbrener et al. 1995) Selective adsorption of the viscosity building components of the asphalt binder (asphaltenes) by clays can lead to a softening of the binder and result in rutting of the pavement. 7/63 Asphalt Durability Thus, moisture damage has multiple effects and multiple impacts on pavement durability from cracking to rutting. Consequently, attributing and assessing damage to this distress mode is difficult. 3.2.2 Thermal Cracking Thermal cracking is largely observed in cold climatic zones. In these regions, the low temperatures experienced by the asphalt, make it essentially perform in its glassy state where it is brittle. As a consequence, the thermal shrinkage occurring during cooling leads to the development of stresses that can overcome the materials strength, hence generating transverse cracks in the pavement. Thermal cracking is not limited to cold regions. Large day-night amplitudes can also generate cracking patterns with the crack propagating from the top of the layer to the bottom. This has been observed in Southern France (GNB 1997) and elsewhere (Ferne 2006). 3.3 Deterioration due to Traffic 3.3.1 Rutting Rutting has been observed in asphalt pavements since the very beginning of the use of asphalt pavements, but became increasingly significant following World War II as traffic volumes and loads started to increase rapidly (Sousa et al. 1991). Rutting occurs when the traffic load over the asphalt mixture exceeds its visco-plastic limit, hence generating permanent plastic deformations evidenced by depressions in the wheel path as illustrated in Figure 2. The rutting attributed to visco-plastic permanent deformation is inversely proportional to the traffic speed and asphalt binder stiffness. As a result, rutting is more predominant under lowspeed, high loads, and at high temperatures (Ould-Henia et al. 2004). (Typical pavement temperatures in Europe and North America range from 40-60°C.) However, rutting remains a complex phenomenon, because the asphalt mixes deform in a visco-elasto-plastic way under these conditions (Gibson 2006). Rutting in asphalt mixes is associated with several factors such as higher than optimum asphalt content, high natural sand content, round aggregate shape (e.g., uncrushed gravel) or high binder deformability (Sousa et al. 1991). Consequently, factors favoring the stiffening of the mixtures should also increase the rutting resistance. Rutting on average accumulates for untreated mixes over the 5 years post construction, but sometimes premature rutting may occur during the very first year (Hicks 1991). 3.3.2 Fatigue Cracking Fatigue cracking of asphalt pavements is a more recently studied phenomenon, though it had been recognized back in the 1950s as a possible failure mode for asphalt mixtures (Duriez and Arrambide 1954), and later demonstrated in the highly visible American Association of State Highway Officials (AASHO) trials from 1957 to 1961 (AASHO 1962). Fatigue cracking occurs when the repeated traffic loads progressively damage the asphalt mixtures generating cracks that propagate from the bottom to the top of the layer as shown in Figure 3. As a consequence, fatigue cracking is more pronounced in thin thickness layers or 8/63 Asphalt Durability where poor adhesion exists between the successive layers; both of these conditions promote high flexural stresses at the bottom of the asphalt layers (LCPC-SETRA 1997). Significantly aged pavements will experience top down cracking whereby the top layer is significantly embrittled and the stresses overcome the strain tolerance of the pavement layer. The cracks initiating in this brittle surface can then propagate to softer layers below. 3.3.3 Loss of skid resistance The top layer of the asphalt pavement, called the wearing course, has specific requirements that go beyond the mechanical properties assuring that the pavement resists traffic-induced damage. In particular, skid resistance is a big issue, as it ensures that the tires of the riding vehicles maintain good contact with the pavement surface, hence contributing to the safety of the road users. Skid resistance is controlled largely by the aggregate sizing and properties and to a lesser extent, by the type of mixture and binder used. More precisely, microtexture, i.e., the roughness of the mixture at the micron level (Brosseaud et al. 2005, Masad et al. 2009), governs this property and is essentially related to the aggregate texture and hardness. As a result, the wear of the aggregate due to traffic decreases the skid resistance as shown in Figure 4 (Stasse 2000) and therefore makes it essential to use wear-resistant aggregates in the surface or wearing courses. 3.3.4 Other distress modes Bleeding occurs when higher than optimum asphalt binder exceeds the mix void capacity in the pavement and moves upward under traffic or with thermal expansion. This is a nonreversible, cumulative process that is a safety related issue as it creates a slippery surface. Block cracking results from the shrinkage and hardening of aged asphalt binders. This consists of a series of large interconnected rectangular cracks on the asphalt pavement surface which can range in size from 1’ to over 10’ in length. Reflective cracking occurs when cracks from the base or underlying layer propagate to the top-layer; this can be observed in asphalt mixtures overlaying rigid (cement concrete) pavements. Another significant distress mode results from the rutting or deformation of the subbase, especially when this consists of granular materials or soils. This deformation may not be limited to the wheel paths and results in a general uneven pavement. Other distress modes can also be found that are somewhat related to traffic. In such cases, these failure modes are observed on the top asphalt layer but are really due to structural problems underneath. A thorough cure for these distresses would necessitate a major treatment of the responsible layers. However, the higher processing costs of such operations makes repeated milling and replacing the modus operandi which does not eliminate the primary cause of such disorders. 3.4 Observed field durability Given all the above factors, asphalt pavements gradually deteriorate and lose their strain tolerance or develop load associated cracks and finally need to be preserved via partial milling and overlaying or ultimately full-depth replacement at the end of their service life. In general, 9/63 Asphalt Durability the wearing course is the most distressed pavement layer and the tendency is to design pavement structures that can withstand 20 years or more of cumulative traffic where only the wearing course needs to be repaired. At present, the durability of European wearing course materials has been studied and service lives were found to range from 8 to 25 years, depending on mixture type. The results of this field survey are presented in Figure 5 (EAPA 2007). Care is needed in interpreting this data, as the definition of service life was not clearly stated in this study. Some countries reported the expected service life, others provided the real service life (i.e., when distresses necessitated rehabilitation of the pavement to maintain the riding quality), and some defined this as the usual time between maintenance operations, regardless of the level of pavement deterioration. In addition, climate and traffic vary from one country to another. Nevertheless, this shows that wearing courses are expected to last 8-25 years and shorter durations are then considered premature failures. The most frequently observed distress modes in Europe are detailed in Figure 6, where it appears that rutting is the main issue, followed by the loss of skid resistance and top-down cracking (FEHRL-ELLPAG 2004). Improvements needed to restore the pavement condition will depend on type of damage that has accumulated. In the case of fatigued or moisture damaged materials, milling and replacing of the surface lift is generally the best option. 10/63 Asphalt Durability 4. LABORATORY EVALUATION OF ASPHALT MIXTURE DURABILITY 4.1 Deterioration due to Material Chemical Aging 4.1.1 Asphalt Binder Conditioning The resistance of the asphalt binder to the short term (asphalt mix production) and long term (in service life) chemical aging mentioned in 3.1.1 has been evaluated in the laboratory through the conditioning of binders at elevated temperatures and or pressures for extended periods of time. The short-term aging of asphalt binder is simulated in the laboratory by conditioning a thin asphalt film at a high temperature, for a short duration; the most prevalent specification is the Rolling Thin Film Oven Test (RTFOT - ASTM D2872 - EN 12607) which exposes the binder in 1.25-mm thick moving films to air at 163ºC for 85 min. This simulated “average processing condition” typically result in a nominal doubling of the viscosity after cooling, although the extent of hardening is asphalt-dependent, 1.5 to 4 fold increases in viscosity at 60ºC have been reported (Anderson et al. 1994; Bell 1989). Note that part of the hardening originates from the loss of volatile compounds, and asphalt specifications generally require that mass loss after short-term aging remain below 0.5%. Current European and USA specifications indirectly eliminate asphalts whose aging upon mixing is deemed too rapid. The long-term aging occurs over the service life of the pavement which can extend to several decades. It is largely dependent on the diffusion of oxygen and the pavement temperature, the latter being dependent on the local climate. The exposed layer will obviously be subjected to a greater level of oxidation, however, the mix formulation and the achievement of the desired level of compaction (pavement density) also come into play, as do asphalt film thickness and mix porosity. All these factors influence asphalt aging and make it quite complicated to accurately describe in-situ aging. One widely used conditioning procedure, the Pressure Aging Vessel Test (PAV - ASTM D6521 - EN 14769), was shown to match the increase in viscosity of approximately 4 to 8 year old asphalt binders in surface sources taken from two diverse locations, namely, Wyoming and Florida (Anderson et al. 1994). However, the PAV does not accurately simulate the in-situ aging of the binder in the pavement. 4.1.2 Asphalt Mixture Conditioning Plant produced mix samples taken from paver can be compacted into lab specimens using a gyratory or linear kneading compactor or can employ some other mode of compaction. Lab produced mixes which mimic this level of aging involve conditioning the loose mix for 2 hours at 135ºC prior to compaction when designing mixes; 4 hours conditioning are used to produce specimen for performance related testing. Long term aging of asphalt mixtures is simulated by aging the compacted cores for 5 days at 85ºC (AASHTO R 30-02). While this conditioning is to reflect 7 to 10 years of field aging, the post conditioning appearance of the compacted samples do not visually simulate the field aged samples obtained from aged pavements. 4.2 Deterioration due to Climate 11/63 Asphalt Durability 4.2.1 Moisture Damage and Frost Many test methods have been developed to evaluate the moisture resistance of asphalt mixtures. Table 1 lists those most frequently referenced in the literature. Their predictive power is still debated and there is no clear consensus on which method is best suited for predicting moisture damage in the field. The Lottman test (AASHTO T-283 – Table 1), evaluates the loss of indirect tensile strength following one freeze-thaw cycle, this was found to be one of the more effective tests in use in the USA as shown in Figure 7 ( Hicks 1991). The Hamburg Wheel Tracking (HWT) test (AASHTO 324, EN 12697-22) was not addressed in this survey and has since gained significant popularity. This employs a loaded steel wheel which reciprocates over samples submerged in a heated water bath. The magnitude of the deformation as a function of loading passes is captured. Typical experimental curves are shown in Figure 8. 4.2.2 Thermal Cracking Thermal cracking is generally studied in the lab by measuring stresses and specimen temperature in a restrained specimen subjected to a cooling cycle (10ºC/hr is typical). The dimensions of the prismatic specimens are maintained constant such that thermal stresses build up as the specimen shrinks. The temperature at which the specimen breaks is then recorded. Indirect tension testing has been used to determine creep compliance and strength of HMA in tension, which allows a computation of the thermal shrinkage, stress build up and theoretical failure temperature, phenomena that occur physically in the TSRST test. Roque and others (1999) have also used this to determine crack growth rate parameters for HMAs in intermediate temperature range for fatigue cracking analysis. 4.3 Deterioration due to Traffic 4.3.1 Rutting Several test methods are available to evaluate the rutting resistance of asphalt mixtures. Most are traffic simulators, such as wheel trackers, others are mechanical tests quantifying the permanent deformation accumulated by the material under repeated loads at high temperature (generally in the 40-60°C range). The European standard EN 12697-22 incorporates several test set-ups in testing asphalt mixtures for rutting resistance. The Asphalt Pavement Analyzer (APA), the Hamburg Rut Tester and the French Pavement Rutting Tester are examples of traffic simulators. The latter is shown in Figure 9. Mechanical tests in use impose creep or dynamic compression and measure irrecoverable deformation and stiffness, an example of the former is the Asphalt Mixture Performance Tester (AMPT); the data from this are incorporated into pavement designs. 4.3.2 Fatigue Cracking Fatigue life is generally studied in the lab by submitting a specimen to repeated loads of constant intensity. The load can be either stress or strain-controlled. In stress-controlled experiments, failure is easily detected as the breaking point of the specimen. In straincontrolled experiments, failure is conventionally defined as the point where the specimen 12/63 Asphalt Durability modulus is decreased by 50%. The number of cycles to failure is measured as a function of loading intensity. Representative fatigue curves of both modes are shown in Figure 10. European standard EN 12797-24 describes various test methods in order to evaluate the fatigue resistance of asphalt mixtures. Apart from the inherent low repeatability and reproducibility of the fatigue tests, the different set-ups used in the industry yield quite different conclusions (Di Benedetto et al. 2003) making it difficult to assess the fatigue life in sound theoretical terms. Currently models based on viscoelastic continuum damage (VECD) theories are being developed and applied. These theories take into account the development of small microcracks and their coalescence and growth into macrocracks (Lee and Kim 1998a; Lee and Kim 1998b). Cyclic axial fatigue characterization tests can be conducted on field and laboratory produced specimens to generate a damage characteristic curve which can be used to describe the damage and cracking response at any temperature and under any generalized loading (i.e., stress-control and strain-control) (Kutay et al. 2008). Healing is also another key phenomenon that affects the fatigue life of asphalt mixtures. This phenomenon has been observed in the early studies on asphalt mixture fatigue (Saunier 1968) and as a result, the occurrence of rest periods between loading sequences has been observed to increase the fatigue life of asphalt mixtures (Kim et al. 2003). The rest periods allow for the microcracks to heal and therefore decrease fatigue damage. At the moment, there is no standardized test method where healing effects can be taken into account. 4.4 Practical use of the lab tests: Specifications Specifications on asphalt mixtures are generally based on 3 complementary elements. First, specifications on the binder are largely purchase based that allow for an adequate choice with respect to foreseen climate and traffic conditions. Second, specifications on the aggregate exist to guarantee that the most important component in terms of weight will fulfil minimum requirements. Finally, specifications on the asphalt mixture constituents make sure that the proportions and qualities of the raw materials are fit for their intended use. For example, hotmix asphalt mix design specifications are used to ensure that an optimum amount of asphalt binder is used in a mix along with an appropriate gradation that ensures adequate volumetric properties. 4.4.1 Binder specifications European paving grade asphalts are still defined using empirical tests such as penetration and ring and ball softening temperature, whereas, the USA has adopted specifications that control asphalt rheology which were developed during the Strategic Highway Research program (SHRP) in the early 1990’s (AASHTO T315). The current USA specifications for bituminous binders are collectively known as the performance grade (PG) specification. The PG system relies essentially on two test methods: one characterizing the binder in the high temperature range, the other one for the low temperature range. The binder is said to be PG H-L, where H is the limiting high temperature and -L, the limiting low temperature. For example, a binder with PG 58-28 grade means that its limiting high field service temperature is 58ºC while its limiting low temperature is -28ºC. The limiting high field service temperature corresponds to the temperature at which the inverse viscous compliance (G*/sinδ) of the binder is equal to 1 kPa when measured at a 13/63 Asphalt Durability frequency of 10 rad/s (AASHTO T315). The test is typically performed on the unaged binder. Grades are defined by 6ºC steps starting from 46ºC up to 82ºC. The idea behind this criterion is that too fluid a binder generates a risk of rutting for the pavement. Thus, the limiting high temperature can be thought of as the temperature above which the risk of rut formation becomes significant for a hot mix made with the corresponding binder (Asphalt Institute 1994). The limiting low field service temperature is based on the more restrictive of two criteria. One refers to the m-value which controls 98% of the USA binders. The other corresponds to the temperature at which the flexural creep modulus of the binder is less than or equal to 300 MPa when measured at a loading time of 60 s (AASHTO T315). Grades are defined by 6ºC steps starting from -10ºC down to -46ºC. The limiting temperature stated in the paving grade is in fact 10ºC lower than that of the creep test. In other words, if the binder has a creep modulus of 300 MPa or lower at -12ºC and 60s, the limiting low temperature will be -22ºC. The idea behind this criterion is that too rigid a binder generates a risk of cracking for the pavement. Thus, the limiting low temperature can be thought of as the temperature below which the risk of cracking becomes significant for a hot mix made with the corresponding binder (Asphalt Institute 1994). Since the risk increases with binder aging, the test is typically performed on binders subjected to the short term aging (RTFOT – ASTM D2872, EN 12607) followed by the long-term conditioning (PAV - ASTM D6521, EN 14769). A refinement to the PG specification is the Multiple Stress Creep Recovery (MSCR) test procedure (AASHTO TP70) which more accurately predicts the performance of polymer modified binders as it relates to rutting. The test captures the non-recoverable creep compliance (JNR) and percentage of recovery during each loading cycle (D’Angelo et al. 2007). 4.4.2 Aggregate specifications The choice of proper aggregate is critical to the performance of the HMA. The aggregate accounts for roughly 95 wt. % of the mixture. The specifications classify aggregates in terms of particle size and shape, mechanical properties, and surface activity. The geometrical criteria are of the utmost importance. Aggregate are separated into fractions depending on their particle size by sieving. In Europe, aggregate fractions are generally designated by a d/D value, where d is the maximum sieve size (in mm) which retains all of the aggregate, and D is the smallest sieve size through which all of the aggregate passes. For example, a 6/12 aggregate passes thru a 12-mm sieve but is retained on a 6-mm one. The fraction finer than 75 microns (63 microns in Europe) is called the filler. The fraction passing the 2 mm sieve is referred to as sand, and generally includes most of the filler. Coarser fractions are called gravel. Asphalt mixtures generally do not use aggregate larger than 30mm, and more often no larger than 20mm, because of the high risk of segregation. Aggregate shape also plays a critical role. Ideally, cubic aggregate would be desirable in order to maximize aggregate aggregate interlocking which in turn maximizes the rutting resistance. In the case of rounded materials such as river gravels, the level of crushing (number of fracture faces) is often stipulated. Flat or elongated particles are not very favourable, as they tend to break into smaller pieces during compaction and trafficking. Consequently, specifications are in place to restrict the amount of such particles. 14/63 Asphalt Durability Finally, good mechanical properties are very important; these include strength and resistance to polishing. The closer the asphalt layer is to the surface, the more severe the specifications. Mechanical resistance is generally evaluated using the Los Angeles abrasion test (ASTM C131 and C535 – EN 1097-2). Abrasion resistant materials, with Los Angeles values below 20% are generally preferred for surface courses. In base or subbase courses, higher Los Angeles values, up to 40% in some countries, are accepted. Also, the wear resistance of aggregate is of primary importance for wearing courses because of its strong effect on skid resistance (Brosseaud et al. 2005, Masad et al. 2009). This is generally assessed through the Polished Stone Value test (EN 1097-8). The Polishing Stone Value (PSV) reflects the micro texture and is a measure of the resistance of an aggregate to polishing, In order to maintain a good skid resistance, the most demanding specifications require a PSV above 55. The specifications based on LA or PSV generally prohibit the use of soft aggregates, such as limestone, in the wearing courses, especially for high-traffic roads; though they have been blended with a hard aggregate to achieve a good texture due to their differential wearing (Abdallah et al. 2008). An approach that has been developed to quantify three-dimensional shape, angularity, and texture of coarse aggregate particles as well as the angularity of fine aggregate particles is the aggregate imaging system (AIMS). This has been used to identify accelerated aggregate polishing and changes in micro-texture (Masad 2005) with more objectivity than conventional techniques like Fine Aggregate Angularity (AASHTO T304). Finally, specifications on cleanliness are generally invoked. They limit the amount of very fine materials (generally clays) in the aggregate, where finer particulates have a detrimental effect on the asphalt binder and its adhesion to the aggregate. As a matter of fact, clay materials have the tendency to form layers around the aggregate such that the asphalt binder sticks to this layer instead of the aggregate. It is similar in concept to the use of wheat flour as a demoulding agent in the kitchen where the flour does not stick to the walls of the pan, so the cake can be easily removed. The same happens with asphalt binder when it adheres to the clay or fines instead of directly to the aggregate particles. Sand equivalent or methylene blue values are generally used to quantify the amount of surface active clays. 4.4.3 Asphalt mixture specifications The specifications on asphalt mixtures help the formulator choose the right components and quantity. Most specifications worldwide give the particle size distribution of the aggregate for a given mixture formula, for example, an asphalt concrete for a wearing course (Monismith 2006). The Bailey Method and gyratory locking point are used to optimize the aggregate packing and structure; the locking point is defined as the point where mix density is not affected by consecutive gyrations(Mohammad and Shamsi, 2007; Li and Gibson, 2011). Interestingly, the French devised a local method where no particle size distribution is given, only maximum aggregate size (Delorme et al. 2007). This is because the full curve is validated indirectly through its compaction behaviour, for which tight specifications are present. Once preliminary binder contents and grades, and an aggregate gradation are selected, asphalt testing is performed on mixtures to ensure their constructability and durability (Asphalt Institute 2001, Monismith 2006, Delorme et al. 2007). Depending on the local requirements, specifications are generally based on: moisture resistance, density and compaction behaviour, 15/63 Asphalt Durability and mechanical properties. Testing the latter may entail evaluating one or more of the following properties: strength, modulus, fatigue resistance, rutting resistance. Other specific properties may be considered in a specification; for example, the permeability of porous asphalt may be evaluated. While specifications are striving to become more performance-based (Asphalt Institute 2001, Monismith 2006, Delorme et al. 2007), we are far from being able to anticipate the service life of an asphalt mixture from a laboratory study. Therefore, a lot of additional research is needed to assess the field relevance of most test methods. 16/63 Asphalt Durability 5. GENERAL PROPERTIES OF ASPHALT MIXTURES 5.1 Composition HMA is generally fabricated by first heating the asphalt binder to around 160ºC in order to decrease its viscosity and then blending it with heated aggregate. The specifications on asphalt binders for pavements are mostly based on this application. Other techniques to manufacture asphalt mixes using asphalt binder emulsions, warm and cold mix technologies are also available but they amount to less than 5 % of the total asphalt mix production. 5.1.1 Asphalt Binder: Definitions As the name suggests, asphalt mixtures are usually obtained by blending aggregate and asphalt binder. Asphalt cement, generally called bitumen in Europe, is defined as a “virtually involatile, adhesive and waterproofing material derived from crude petroleum, or present in natural asphalt, which is completely or nearly completely soluble in toluene, and is very viscous or nearly solid at ambient temperatures” in the current European specifications (EN12597). A solubility value in trichloroethylene (TCE) above 99 % is required in the paving specifications (EN12591). Historically, natural sources of asphalt were initially used. In Europe, the deposits of Seyssel in France and Val de Travers in Switzerland were at first the principal asphalt sources for road applications (although not exactly as asphalt mixtures). The reference material for paving applications in the USA at the beginning of the 20th century, were mostly the Trinidad Lake asphalt and, to a lesser extent, the Bermudez Pitch Lake asphalt from Venezuela, both of which constitute some of the largest asphalt deposits in the World. Although they have very similar uses and properties, asphalt binder is not to be confused with coal tar, which is the residue of the pyrolysis of coal. Unfortunately, the name coal tar remains widely used in everyday language as a general term for a black paving material. Health issues associated with the use of coal tar have largely eliminated its use. These issues arise from the higher content of carcinogenic polynuclear aromatic hydrocarbons (PAH) found in coal tars whereas they are only present in trace amounts in asphalt binders (Burstyna et al. 2000). Currently, asphalt binder is essentially obtained by the distillation of crude oil (Corbett 1965, Lesueur 2009). Not every crude oil source yields sufficient amounts of asphalt binder. Typically, the heavier the crude oil, the higher is its yield of asphalt binder (Corbett 1965, Read and Whiteoak 2003). Native asphalt binders such as Trinidad Lake asphalt and Gilsoniteare still in use in the paving industry but accounts for small and very specific markets, generally as an additive to straight-run asphalt binder. 5.1.2 Asphalt Cement: Physico-Chemical Properties 17/63 Asphalt Durability Asphalt cement densities at room temperature typically range between 1.01 and 1.04 g/cm3, depending on the crude source and paving grade (Read and Whiteoak 2003). As a rule of thumb, the harder it is, the denser the asphalt binder. Asphalt binder exhibits a glass transition temperature around -20ºC, although it varies over a very wide range from +5ºC down to -40ºC depending essentially on the crude origin and to a lesser extent on the processing history. The transition range typically spans 30 to 45ºC where -20ºC corresponds to the typical midpoint value (Lesueur 2009). From a thermodynamic standpoint, asphalt binder is a very viscous liquid at room temperature. The complexity of asphalt binder chemistry lies in the fact that many different species are present. Typical functional groups found in asphalt binder are shown in Figure 11. As an overall descriptor, the chemical nature of the crude oil is generally described as paraffinic, naphthenic or aromatic if a majority of saturate, cyclic or aromatic structures, respectively, are present. This classification of the petroleum is sometimes applied to the corresponding asphalt binder. For example, Venezuelan asphalt binders are generally known as naphthenic asphalt binders. The elemental composition of an asphalt binder depends primarily on its crude source (Mortazavi and Mouthrop 1993). The data in Table 2 illustrate this fact. These come from the extensive research effort on asphalt binder chemistry, structure and properties undertaken the as part of the Strategic Highway Research Program. The coding (AAA-1, etc.) refers to the various crude sources and grades for the asphalts used within the SHRP (Mortazavi and Mouthrop 1993). As shown in Table 2, asphalt binder mainly consists of carbon (typically 80-88 wt.%) and hydrogen atoms (8-12 wt.%). This gives a hydrocarbon content generally above 90 wt.% with the hydrogen-to-carbon molar ratio H/C around 1.5. This H/C ratio is therefore intermediate between that of aromatic structures (benzene has H/C = 1) and that of saturate alkanes (H/C ~ 2) (Branthaver et al. 1994). In addition, heteroatoms such as sulphur (0-9 wt.%), nitrogen (0-2 wt.%) and oxygen (0-2 wt.%) are generally present. Traces of metals are also found, usually as metallic porphyrins, the most ubiquitous being vanadium, up to 2000 parts per million (ppm), and nickel (up to 200 ppm) (Mortazavi and Moulthrop 1993; Lesueur 2009). Sulfur is generally the most abundant heteroatom. In the neat asphalt, this appears in the form of sulfides, thiols and thiophenes; upon oxidation many of these groups are converted to sulfoxides and occasionally sulfones. Oxygen is typically present in the form of ketones, phenols and, to a lesser extent, carboxylic acids. Nitrogen exists typically in pyrrolic and pyridinic structures and also forms amphoteric species such as 2-quinolones (Branthaver et al. 1994) which have both an acid and base functionality on the same molecule. Given the concentration of polar atoms, functional groups generally do not amount to more than a few multiples of 0.1 mol/l for straight-run asphalt binders (Branthaver et al. 1994), however, their concentration can increase significantly upon aging. The number-average molecular weight of asphalt binder falls typically in the range 600-1,500 g/mol range (Table 2 - Traxler 1961, Lesueur 2009). The values provided in Table 2 were measured by Vapor Pressure Osmometry in toluene and pyridine at 60ºC (ASTM D2503). 18/63 Asphalt Durability Average molecular structures for asphalt binder have been proposed (Jennings et al, 1993). The molecular weight of the average structure equals the average molecular weight of the studied asphalt binder, and the atoms are distributed to account for NMR spectral data obtained for the corresponding asphalt binder (Jennings et al. 1992). Two examples are provided in Figure 12. Note that asphalt binder molecules are not macromolecules in the polymeric sense. As a consequence, care must be taken when trying to compare the properties of polymers to that of asphalt binder, especially when it comes to the modelling of the viscoelastic properties based on this averaged molecular approach. Given these molecular weights and the proportion of polar atoms, it is clear that only 1-3 polar atoms are present on average in each asphalt binder molecule, as illustrated in Figure 12. However, approaching asphalt binder chemistry on a global basis is inadequate when one tries to understand the properties of asphalt binder. Thus, the molecules are generally operationally separated into different chemical families, depending on their solubility in given aromatic or non-polar solvents. Using the separation procedure such as ASTM D-4124, asphalt is separated into asphaltenes and maltenes based on the solubility in n-heptane. Maltenes are further separated into resins (polar aromatics), aromatics and saturates. Subsequently, the composition of asphalt binder is usually given in terms of the relative quantity of its so-called SARA fractions for Saturates, Aromatics, Resins and Asphaltenes. Specific SARA values for a diverse suite of binders is provided in Figure 13 and typical ranges for their chemical properties are provided in Table 3. The polar asphaltenes are essentially the viscosity builders whereas the low polarity components comprising the maltenes appear to control the low temperature performance. More recently researchers have been fractionating the asphaltenes using high performance column chromatography (WRI 2010). A profile from such a separation is provided in Figure 14. Recent work using Fourier Transform Ion Cyclotron Resonance Mass Spectrometry indicates that the upper molecular weight range is below 2000 g/mol. (WRI 2010). Some researchers concluded that asphalt is a simple homogeneous fluid, calling this model the Dispersed Polar Fluid (DPF – Christensen and Anderson 1991). Others now claimed that asphaltenes in asphalt binder form a molecular solution based on their solubility parameters (Redelius 2006). All these models have been shown to be incomplete and the most accepted view now is that asphaltenes form micelles of radius 2-8 nm dispersed in the maltenes (Lesueur 2009). 5.1.3 Manufacturing Process HMA is manufactured in dedicated plants, sometimes continuous, sometimes batch (Lee and Mahboub 2006). In all cases, the manufacturing steps include aggregate drying and then asphalt / aggregate mixing. In order to insure a good wetting of the aggregate surface by the asphalt binder, a binder viscosity around 200 mPa.s is sought. This is generally obtained by having the asphalt binder at temperatures close to 160°C. The aggregate exits the dryer at a similar temperature whereupon it may be stored in heated and insulated silos, therefore the HMA is manufactured at temperatures close to 160°C 19/63 Asphalt Durability The HMA is then transported hot by truck to the jobsite. Compaction is performed immediately after laying; the compaction temperature is generally 30 Cº lower than manufacturing temperature. Note that additives or innovative processes are now used in order to manufacture the asphalt mix at lower temperatures. They are called warm mixes when the manufacturing temperature is lowered by 30 to 50 Cº and cold mixes when the temperature is below 100°C. In all cases, the objective is to maintain the properties of HMA while using lower manufacturing temperatures (Button et al. 2007). 5.2 Properties 5.2.1 Mechanical Properties Pavements are essentially designed in order to withstand traffic-induced damage. This is done by first calculating the stresses and strains induced by traffic and then computing the stress distribution at any point of the different pavement layers (LCPC-SETRA 1997). From this exercise, it is possible to optimize the thickness and properties of each of the layers such that the corresponding materials can withstand the anticipated stresses for the design period, usually between 10 to 40 years. As a consequence, the material parameters that are used for pavement design are mostly based on mechanical properties. More precisely, modulus and fatigue resistance are the key parameters. Construction materials are also widely characterized by strength measurements. This is also the case for asphaltic materials and considerable data are available in the literature on the subject (Taylor 2000). Modulus is a fundamental mechanical property of materials (Timoshenko 1976). It is the ratio between the stress applied to the material and the resulting deformation (or the opposite if the material is tested under a deformation-controlled mode). The Young’s or tension modulus is measured using tensional forces (compression thus is a negative tension). It is a Coulomb or shear modulus when measured using shear forces (though pure shear is never really generated). The mechanical properties of an asphalt mixture are known to be temperature sensitive and loading time dependent, this is a consequence of their viscoelastic behaviour (Francken 1998). So, the modulus is temperature and time (or frequency) dependant, and it is generally expressed in terms of a complex number, the complex modulus (Figure 15). From a mix formulation standpoint, modulus has been found to peak at optimum asphalt binder content; it also increases with the modulus of the binder and decreases as the air void content increases (Francken 1998). Modulus is of critical importance in the design of pavement layers, because it governs the stress distribution inside each pavement layer. For given load and thickness, higher moduli translate to the presence of lower stresses in each respective layer. As the modulus is an intrinsic property, that it should be essentially independent of the specific testing set-up, however, small differences are generally observed whenever that modulus is measured in compression, flexion, tension or indirect tension testing modes. Testing geometry, i.e. specimen shape and dimensions, as well as the signal type, i.e. that 20/63 Asphalt Durability controls the application of deformation or force, or if cyclic its amplitude and its sinusoidal frequency also affects somewhat the magnitude of the specific data measured. Therefore, it is necessary to be aware of the measurement conditions and parameters when evaluating those asphalt mixture moduli values. A European standard exists in order to limit the differences (EN12697-26). Strength is another important mechanical property of materials (Timoshenko 1976). Here it is specified as the maximum applied stress that causes that material to break apart or fail. Strength for asphalt mixtures is usually measured using either compressive or indirect tensile modes of deformation; these are generally conducted at controlled temperatures close to room temperature. In general, modulus and strength are somewhat related when measured under the same temperature and loading conditions, although one is an intrinsic property (modulus) and the other strongly depends on specimen shape and dimensions and is therefore not intrinsic (Timoshenko 1976). However, it is a lot easier to measure strength than modulus, hence its reporting predominance in materials engineering. Stiffer asphalt mixes or soils tend to have smaller rutting. In other words, there can be trends between modulus and strength with varying degrees of reliability. For example, a popular relationship used in practice for unbound pavement materials is the correlation between the California Bearing Ratio ( AASHTO T193) and the resilient modulus (AASHTO T 307), However, this relationship is generally used for analyses where the criticality is small and does not warrant materials testing. Because of this almost constant ratio between strength and modulus, mix variables affect the strength in the same way as the modulus. Therefore, strength is known to peak at an optimum asphalt binder content, to increase with the modulus of the binder and to decrease as the air void content increases (Francken 1998). As previously detailed, fatigue cracking occurs when the repeated traffic loads progressively damage the asphalt mixtures, generating cracks propagating from the bottom of each layer to its top. As a consequence, fatigue cracking is favoured by smaller layer thicknesses or poor adhesion between the successive adjacent layers; both factors promote high tensile strains at the bottom surfaces of each asphalt layer. From the mix formulation standpoint, fatigue resistance is known to be enhanced by higher asphalt binder contents or the use of high-performance binders (Francken 1998, Tangella et al. 1990). Depending on the method of measuring fatigue, a soft binder can increase the fatigue life (strain-controlled) or decrease it (stress-controlled). Fatigue cracking is the main failure mode considered in the design of pavement structures. More precisely, the bituminous layers are designed to be thick enough to insure that fatigue cracking does not appear until the end of design life, which can vary from 10 to 40 years in Europe (FEHRL-ELLPAG 2004). The design period has even been extended to 50 years for so-called “perpetual pavements” (Newcomb et al. 2001). Interestingly though, as the asphalt binder is fundamentally a very viscous liquid at room temperature, it can heal small cracks. Factors affecting healing rate are those favoring asphalt autodiffusion; higher pavement temperatures and lower viscosity binders will tend to heal these cracks more rapidly (Bhasin et al. 2011). However, healing is not taken into account in current pavement design. 21/63 Asphalt Durability The resistance of mixes to thermal cracking is largely binder dependent. Soft binders increase the cracking resistance; hence they are preferred in Nordic-like climates. As with fatigue cracking, this resistance can also be enhanced by formulating mixes with higher binder contents or high-performance binders (Marasteanu et al. 2004). 5.2.2 Physical Properties Apart from their mechanical properties, some other physical properties are of great importance for asphalt mixture durability. As mentioned earlier, the wearing course of a pavement must insure a good skid resistance performance. This is evaluated using appropriate methods; the simplest one uses a pendulum (Masad et al. 2009). More advanced methods use a braking wheel with controlled slip-ratio and a controlled surface water layer thickness in order to record the transversal or longitudinal friction coefficient at varying speeds (Brosseaud et al. 2005). Good correlation between skid resistance and braking distance has been found (Wallman and Astrom 2001). Second, reduced water permeability is also quite important. In fact, the wearing course generally provides waterproofing for the rest of the pavement. Exceptions are open graded friction courses (porous asphalt) that are designed to improve the safety of wet surfaces by reducing splash and spray; here the waterproofing function is done below that porous layer. Permeability is largely influenced by the aggregate size. Most asphalt mixes have a nominal maximum aggregate size (NMAS) between 9.5 mm and 12.5 mm. For asphalt mixtures with less than 4% air voids, the mixtures are generally believed to be impervious to water. When the air voids reach ~6-8%, permeability starts to increase significantly from 10-5 up to 10-4 m/s for continuous graded mixtures (Huang et al. 1999). The term Pessimum Voids was coined for this most commonly targeted range of air voids for HMA (Terrel and Al-Swailmi 1994); this is schematic shown in Figure 16. In contrast, for fine aggregate mixtures such as -4.75 mm mixtures, these are considered essentially nonporous even with 12% air voids. Newly constructed pavements of porous asphalt must have permeability above 10-4-4.10-3 m/s to be acceptable (Delorme et al. 2007). Interestingly, permeability of continuous graded mixtures tend to increase with service time; this is attributed to microcracking as well as moisture damage. However under certain conditions, the permeability of porous asphalt may decrease with service time, especially under low traffic speed conditions, because of dirt and debris accumulation (Gal 1992). 6. DISTRESS MECHANISMS 6.1 Asphalt Chemical Aging As noted in Section 5.1.3, asphalt binder is a complex mixture of mainly hydrocarbon molecules. Some of these molecules may irreversibly evolve through chemical aging. The latter process is generally thought to be a combination of oxidation and polymerisation reactions, and to a lesser extent, the evaporation of lighter components (Traxler 1961, Petersen 2009). 22/63 Asphalt Durability This aging can be separated into two conditions. First, during the manufacture of HMA, there is a rapid chemical aging with some loss of volatiles when the hot aggregate is coated with a thin film of asphalt. Second, there is an in-situ aging during the service life of the pavement whereby high temperatures and UV-exposure accentuate chemical aging (Petersen 2009). The short-term aging results in the asphaltene content of the asphalt binder typically increasing by 1-4 wt. % (Lesueur 2009) and a doubling of the asphalt binder viscosity. Most of the sulphides are converted to sulfoxides, and there is a jump in the carbonyl content. During the service life of the pavement that can extend to several decades, the asphalt composition changes (e.g. increase in asphaltenes) and the increases in carbonyl content slow though remain linear with time. It depends of course on the position of the asphalt inside the pavement, the top layers being more exposed than the base course. Consequently, pavement characteristics and environmental factors make it quite complicated to accurately describe insitu aging. 6.1.1 Oxidative Hardening In chemical terms, aging leads first to a decrease in aromatic content and subsequent increase in resin content, together with a higher asphaltene content. Therefore, it is generally accepted that the aromatics are converted into resins, and the resins in turn generate asphaltenes. The saturates remain essentially unchanged, as could be assessed from their low chemical reactivity. All these changes result in a slightly higher but almost unchanged glass transition temperature. Of note and consistent with changes in the distributions of these fractions is that the width of the glass transition range increases with aging (WRI 2011). The rate of asphaltenes formation was found to be essentially linear with time in RTFOT laboratory experiments at 163ºC with 6-7 wt. % asphaltenes formed in 340 min. This linear increase was also observed for in-situ aging of asphalt recovered from real pavement sites, with increases between 2 and 10 wt. % in a 90 months period in southern France (Lamontagne et al. 2001). Asphaltenes produced upon aging may be somewhat different than the initial ones. Increased molecular weights of the asphaltenes upon air-blowing were reported, suggesting the presence of polymerisation reactions (Lesueur 2009). Aging results first in the formation of sulfoxides, followed by carbonyls although at a somewhat slower rate. Some of the carbonyls formed may end up as anhydrides and carboxylic acids (Petersen 2009). The sulfoxides are rapidly formed but are thermally unstable. Therefore, they reach a steady-state level that not only depends on the initial sulphur content of the asphalt, but also on the oxygen diffusion into the asphalt. Ketones and carboxylic acids are more stable and do not reach an asymptotic value in laboratory aging experiments. Anhydrides may form once a significant amount of ketones have been generated. They are thought to derive from the oxidation of benzylic carbons at the 1,8 bridgehead position of a naphthalene ring. The in-situ aging seems to yield a steady-state level not only of sulfoxides but also of carboxylic acids, as observed after 2 years of service life in Southern France. As a result, the amount of functional groups in asphalt after aging may increase by more than 1 mol/l. 23/63 Asphalt Durability In the pavement, the aging rate will depend both on mixture parameters such as porosity and environment as well as depth and local climate (temperature, UV-exposure, etc.). Asphalt aging rates are slower with lower air void contents and at colder climates (Petersen 2009). In one set of field monitoring trials, the aging extent as quantified by viscosity ratio (aging index) was almost doubled when air void contents rose from 3-5% to 7-9% (Petersen 2009). Note that this last value is generally believed to be at the threshold where interconnected voids are formed, in agreement with the data on permeability. Site location can have a much bigger effect with aging indices increasing from 5 to 10 when going from milder locations (Lake Tahoe with yearly average temperature of 42ºF) to hotter USA desert conditions (Indio with yearly average temperature of 73ºF) ( Petersen 2009). For low-air-voids dense mixtures, aging is seen to predominantly occur in the top 1 cm (half inch) of the layer (Petersen 2009). The intensity of aging in this top section is a lot more pronounced than what is generally observed when thick cores are taken from the pavement and asphalt from the topmost section mixes with that from the bottom during the recovery procedure. This effect explains much of the top-down cracking phenomenon in hot regions (see 3.2.2 on thermal cracking). In addition, ultraviolet light is known to increase the oxidation process by activating photooxidation reactions. Photo-oxidation is believed to generate polymerisation reactions, not only among the asphaltene molecules, but also for the less polar fractions. Photo-oxidation is strongly radiation intensity dependent and is almost temperature-independent whereas thermal oxidation is highly temperature dependent. Finally, and as a consequence of its black color, the mean temperature of newly placed asphalt pavements are generally about 10°C higher than the mean air temperature (Han et al. 2011). Therefore, oxygen diffusion, UV-exposure and thermal effects all combine to increase the aging intensity of the upper part of the asphalt layer. 6.1.2 Physical and Steric Hardening When monitoring the viscosity or any mechanical property of asphalt versus time at negative or room temperature, a slow heat-reversible hardening is indeed observed and generally referred to as physical hardening for low temperatures (below 0ºC) and steric hardening for higher temperatures. At low temperature, depending on the crude source and paving grade, the stiffness modulus can double after 3 days at -15°C, while some asphalts do not experience any significant hardening (Anderson et al. 1994). At room temperature, the viscosity increase follows a power law as a function of time with a slope between 0.017 and 0.183, corresponding to a viscosity increase of between 10 and 200 % after 2,000 hrs at 25ºC,where the magnitude depends on bitumen type (Lesueur 2009). Physical hardening was first reported by SHRP researchers (Anderson et al. 1994) but steric hardening was observed by Traxler and Schweyer as early as 1936 when they showed that it is not a consequence of chemical aging as described above, and especially due to its reversible nature (Traxler and Schweyer 1936). Physical hardening was initially thought to be largely a consequence of free volume contraction (Anderson et al. 1994), but this interpretation is not consistent with the observation that it even occurs at temperatures above the glass transition temperature and that its extent is highly asphalt-dependent whereas all asphalts exhibit quite similar features in 24/63 Asphalt Durability terms of mechanical behavior in the vicinity of the glass transition temperature. A more acceptable explanation was proposed relating it to the crystallization of the waxy components of the asphalt (Claudy et al. 1992). Still, the relationship between wax content and physical hardening is not straight forward (Edwards et al. 2006). Similarly, the proposed interpretation for steric hardening generally relies on the sol-gel effect related to the build-up of a stronger asphaltenes network with time (Traxler and Schweyer 1936); however this effect might indeed be consistent with crystallisation kinetics showing that both steric and physical hardening are in fact due to the slow kinetics of wax crystallization (Lesueur 2009). 6.1.3 Exudative Hardening Exudative hardening is one possible mechanism for the accelerated hardening of the asphalt binder. This occurs when mixes are made with porous aggregate. In such cases, some of the lighter compounds slowly diffuse into the aggregate, leaving a somewhat stiffer binder with reduced film thickness on the aggregate. Both factors contribute to increasing the cracking probability of the mixture. Lime-treatment of porous aggregates can diminish this effect by allowing calcium carbonate to precipitate within the aggregate porosity. Moreover, porous aggregate are generally harder to fully dry at the asphalt plant, thus the presence of lime, especially quicklime, can reduce the influence of water that would otherwise be trapped inside the aggregate causing a “soup” phenomenon that prevents a proper laying process. 6.2 Microb / Vegetal Attack Asphalt binder is generally considered as being a quite stable material with respect to biodegradability. In fact, asphalt is itself the result of the slow degradation under high pressure and temperature of organic compounds (Durand 2003). Still, it is not chemically inert as previously noted in the discussion on chemical aging. As a hydrocarbon it is not surprising to find that biological agents can have an effect on it. This effect is very limited from practical standpoint and, to our knowledge, no road failures due to asphalt biodegradability have been recorded so far. Asphalt-containing materials from the archaelogical ages confirm that the probability of biodegradation under normal outside conditions is quite low (Connan 1999). Even though biodegradation is probably not so relevant for asphalt durability in pavements, it has been studied carefully. It was found that aerobic degradation occurs 100 times faster than anaerobic degradation under optimum laboratory conditions, where measured degradation rates were of the order of 27-55 g of asphalt/m2/yr (Wolf and Bachofen 1991). Pseudomonas aeroginosa and species of streptomyces and alcaligenes were identified as the main microorganisms responsible for the aerobic degradation. Still, only the light components of asphalt are biodegradable and a large part, consisting mostly of resins and asphaltenes is resistant to biodegradation (Potter and Duval 2001). In more representative situations, but still highly favorable for microbial growth, (burial under active compost), only the first mm of an asphalt lining was found to degrade after 3 years in a canal (Read and Whiteoaks 2003). 25/63 Asphalt Durability Fungi, moulds and yeasts have been observed to occasionally develop on asphalt roofs under hot and humid climates (Read and Whiteoaks 2003). However, their growth remains largely unreported in asphalt pavements. Plants can grow in contact with bituminous materials (Read and Whiteoaks 2003). Under normal pavement conditions, the low permeability of asphalt mixtures makes it unlikely that plants will develop inside the paved area. However, for porous asphalt mixes and highly damaged pavements, with numerous cracks and potholes, water and dirt can accumulate and make a suitable environment for plant growth. 26/63 Asphalt Durability 7. ENHANCING LONGEVITY PERFORMANCE OF ASPHALT 7.1 Improved mix design and construction 7.1.1 Mix design Mix design consists of defining the following: the aggregate type and relative quantity, asphalt binder type and relative quantity, additive(s) type and content, when applicable, manufacturing temperature, final relative amount of air voids of the in-place mixture (density). Adequate selection of these parameters helps ensure that the final properties of the mixtures are the ones sought. Here, specifications come into play, and depending on the local practice and on the final use of the material (base course, wearing course, etc. for high / low traffic road), specific specification metrics are generally chosen. These are based on lab scale tests to measure: mechanical properties (modulus and/or rutting resistance and/or fatigue and/or strength), moisture damage and/or frost resistance, and density. Other important properties such as permeability, noise emission, skid resistance, etc. are generally not measured at the lab scale and are therefore left out of the formulation process. Still, an appropriate choice of adequate mix design and components ensures that the desired properties will be obtained in the field, due in part to a thorough experience linking mixture type to in-situ properties. In all cases, the first way to increase durability is to ensure that the specifications take into account minimization of the relevant failure modes for the in-service conditions and use situations at stake (e.g., environmental factors, traffic load). Then, provided that the specifications are sound (This means that the testing methods are representative of the field behaviour, which is not always the case.), ensure that tight quality controls are in place to assure that the manufactured and placed product meet these specifications. Specific asphalt plant controls include means to validate the quality of the raw materials, ensure proper and reproducible operational conditions (raw materials content, manufacturing temperature, etc.), and validate asphalt mixture composition. In terms of mix design, it is generally accepted that the higher the asphalt binder content, the higher the durability. However for contents above the optimal level, excessive amounts can lead to conditions inducing rutting and thus must be avoided. But for a given set of performance metrics, including rutting, increasing the asphalt binder content, generally increases the durability. In fact, this helps explains the very high durability performance shown by mixtures such as SMA or Mastic Asphalts. 7.1.2 Construction 27/63 Asphalt Durability On the jobsite, additional controls further assure that the mixture complies with the requirements of the formulation study. It generally suffices to base them on control of: asphalt mixture composition, processing temperatures (especially during compaction), thicknesses of the layers, and in place density. If all these steps are performed correctly, together with adequate material selection and pavement design, then premature failures should be avoidable. Thickness control is very important in pavement design. As a matter of fact, too thin a layer, will lead to premature fatigue failure; whereas, too thick a layer is also not desirable, as very thick layers can be challenging to compact. Density is critical, because the desired mixture properties are only obtained when the design density is attained. A lower density than expected generally results in a lower moisture resistance, lower modulus and lower fatigue resistance; whereas an over-compacted mixture is generally more sensitive to rutting. Note that pavement design generally implies that the overlaying layers of asphaltic materials are glued together, so that stresses propagate through all the layers. This is obtained through the tack coat, usually made of a bituminous emulsion with application of 300-500 g/m2 of residual asphalt binder. However, if the tack coat is missing or has been poorly applied, the top layers will experience much higher stresses at their bottom surface, generating a premature fatigue failure in the very first years following construction. New solutions such as the use of harder grades of asphalt in the tack coat or the spreading of a protecting layer of diluted milk of lime prevents the tack coat from being tracked by traffic off the job site. Finally, it is well known in the industry that the transverse joints formed between the ends of adjacent laid asphalt layers, form weak spots that can easily degrade into potholes. One simple way to limit their occurrence is to place extension devices along the full width of the pavement. Note that this needs a constant asphalt mix supply in order to avoid unevenness issues. Then, while joints still exist, the application of a tack coat emulsion on the cold surface can be used to strengthen the bond. For longitudinal joints, upon laying a new layer, an excess 3-4cm of asphalt mixture on the old existing surface is recommended. This excess material is then compacted from the cold surface, with the roller mostly on the old surface with only a few centimeters on the new one, leaving a nicely closed joint. A last recommendation regarding transverse joints, is to avoid superimposing joints in subsequent layers. A minimum separation of 20 cm between joints guarantees that the possible degradation of one joint will not contaminate other nearby joints. 7.1.3 Warm Mix Asphalts (WMA) As previously mentioned, warm mixes (WMA) are increasingly used today. Because of their very recent development, with few sections more than 6-years old, the impact of these new technologies on durability remains to be observed, as there are both positive and negative aspects that make it difficult to predict whether or not WMA durability will be similar to that of HMA. 28/63 Asphalt Durability Firstly, the current WMA technologies facilitate the compaction of the mix during laydown. Secondly, the decreased manufacturing temperature means that the asphalt binder experiences less aging during fabrication. With aging proceeding in an autocatalytic way, this means that WMA would be less sensitive to aging. Consequently, both factors favor higher WMA durability. On the negative side, decreasing the manufacturing temperature tends to decrease the moisture resistance (Traxler 1961, Aschenbrener 1995). This is particularly the case when using porous aggregates which may not be completely dried. Additives such as surfactants or hydrated lime can be used to counteract this effect, but only time will tell whether or not the problem is definitely solved through their use. 7.1.4 Recycling The asphalt mixtures that are milled or removed during resurfacing, rehabilitation and reconstruction operations are readily recyclable. With increasing binder costs and aggregate availability issues, the inclusion of this reclaimed asphalt pavement (RAP) in new asphalt mixtures is becoming the norm. Currently most State highway agencies in the USA allow the use of greater than 30 percent RAP, though the actual utilization of RAP at this level is considerably less (Copeland et al. 2011). Utilization of RAP sometimes includes a crushing step in order to limit the size of the particle agglomerations. In most cases, addition of 10 to 20 wt. % RAP has little effect on the overall mixture formula and only slightly changes the final properties (Dunn 2001). This can be performed with limited equipment modifications at the plant. High rates of RAP can also be used, even attainments close to 100% recycling, but this requires special equipment and very tight control of the RAP source (Dunn 2001); applications for these high RAP mixes have generally been limited to lower volume roads, lower pavement lifts, or shoulders. The durability of high RAP mixes is largely determined by the ability to control the aggregate gradation, and address a number of binder related issues. Firstly, the RAP binder is considerably stiffer than the original binder. The compatibility or ability of the new and old binders to blend during manufacturing of the mix may affect the cohesive strength of the binder. This is likely to be more of an issue for WMA-RAP which are produced at lower temperatures. Secondly, determining the optimum binder content of these mixes is not straight forward. A portion of the RAP binder is considered in this determination. The amount of RAP binder performing as a black rock (not blending) but merely acting as a precoating of the aggregate is a subject of investigation. Ascribing too large an amount will result in binder lean, less durable pavements whereas an excessive amount of binder will increase the propensity for rutting. 7.2 Additives Increasing demands on the pavment and implementation of the Superpave system (discussed in Section 4.4.1) have increased the use of unconventional binder additives, air blowing, blending, and chemical modification. PG operating range spreads (limiting high temperature 29/63 Asphalt Durability minus limiting low temperature) greater than 95ºC are not achievable using most crudes (Youtcheff and Jones 1994) and require some form of asphalt binder modification. 7.2.1 Polymer Polymer additives are now commonly used to modify asphalt binders (PIARC 1999, McNally 2011). The idea of mixing asphalt binder and natural rubber dates back to 1843 (Thomson 1964). There the original objective was more to find a substitute for rubber than to modify asphalt binder. At the turn of the 20th century, field trials with rubber-modified asphalt binders were performed and continued for several decades until polymer-modified asphalt binders gained significant commercial interest in the late 1970s (Thomson 1964). The typical amount used today ranges between 3 and 6 wt. %. Most of the currently used polymers include elastomers, plastomers and reactive polymers. Elastomers include natural or synthetic rubbers such as styrene-butadiene (SB) copolymers (random SBR, diblock SB, or triblock SBS) and others (e.g., styreneisoprene), Plastomers include ethylene-vinylacetate random copolymers (EVA) or related molecules (e.g., ethylene-methacrylate, ethylene-butylacrylate), polyolefins such as polyethylenes (PE) and polypropylenes (PP), and Random terpolymers comprising ethylene, glycidyl methacrylate (GMA) and an ester group (usually methyl, ethyl, or butylacrylate) represent reactive polymers. These are generally referred to as Reactive Ethylene Terpolymers (RET) because of the chemical reactions that are thought to occur between functional groups on the asphaltenes and the polymer (Polacco et al. 2004). Many grades of each of these polymers are available (PIARC 1999, McNally 2011) but it is generally sufficient for the asphalt binder industry to characterize a polymer by its monomer composition and molar mass or in the case of reactive polymers, the number of reactive groups per repeat unit. Different processes are in place to produce modified bituminous mixes (PIARC 1999). Polymers may be added either directly to the asphalt binder prior to the mixing with the aggregate (wet process) or to the mix, at the same time asphalt binder is blended with the aggregates (dry process). These dry process mixes have different properties than those obtained via the wet process. While some details might apply to the dry process, the following discussion only focuses on the wet process product. Adding polymer to asphalt binder greatly expands its Performance Grade (PG) temperature range; namely, the low temperature can be established and the high temperature greatly increased. More specifically, the applicable rule of thumb states that for every 1% of added polymer, there is a 2ºC gain in the high temperature PG (PIARC 1999). On the low temperature end, the rule becomes more that 1 % of added polymer produces a 1ºC rise in low temperature PG for SB type modifiers. Since the PG classes are based on 6ºC steps, the typical 3 % polymer-content modification generally allows a one step high-temperature class increase leaving the low temperature class step sometimes unchanged. 30/63 Asphalt Durability The use of crumb rubber originating from the recycling of scrap tires is another environmentally friendly polymer modification technology with growing interest due to the large amount of old tires to dispose of (Caltrans 2003) and their low cost relative to conventional polymer modification and even the binder itself (as a low cost binder extender). The gains in high temperature PG with level of loading are not as marked as noted above for conventional polymers. This assumes that the rubber has been well dispersed within the asphalt binder and with the caveat that given the coarse particle size (at least a few 100 µm sometimes up to 1 mm) of crumb rubber, rheological testing with the tools currently used for asphalt binder evaluation might not be adequate (Kim et al. 2001). 7.2.2 Acids Polyphosphoric acid (PPA) [Hn+2PnO3n+1] is a polymer of orthophosphoric acid [H3PO4]. This has been used for over thirty years to stiffen asphalt binders (Baumgardner 2010). Recently, PPA has been used to supplement polymer modification to improve the handling and performance. While PPA is hygroscopic, the addition of small amounts (<0.75%) has not been demonstrated to adversely effect the moisture resistance of mixes (Arnold et al. 2009). 7.2.3 Hydrated Lime Hydrated lime has been incorporated in asphalt mixtures since their very beginning. At the end of the 19th century the National Vulcanite Company paved roads in Washington DC and Buffalo, New York with, a proprietary asphalt mixture called Vulcanite containing hydrated lime (Lesueur 2010). Other proprietary asphalt mixtures of the time using hydrated lime as a filler included Warrenite and Amiesite (Lesueur 2010). A few decades later, hydrated lime was still not widely used and was merely listed as a possible filler component in asphalt mixtures in the USA (Asphalt Institute 1947). In the early 1950s in France, Duriez and Arrambide recommended the use of hydrated lime as a way to improve asphalt binderaggregate adhesion (Duriez and Arrambide 1954). However, hydrated lime did not experience a renewed interest until the 1970s in the USA. Partly as a consequence of a general decrease in asphalt binder quality due to the petroleum crisis of 1973 when moisture damage and frost became some of the more pressing pavement failure modes of the time (Hicks 1991). The various additives to asphalt mixtures available to limit moisture damage were tested both in the laboratory and in the field, and hydrated lime was observed to be the most effective additive (Hicks 1991). As a consequence, hydrated lime is now specified in many States and it is estimated that 10% of the asphalt mixtures produced in the USA now contain hydrated lime (Hicks and Scholz 2003). Given its extensive use in the past 30 years in the USA, hydrated lime has been seen to be more than an additive to mitigate moisture damage (Little and Epps 2001, Lesueur 2010). Hydrated lime is known to reduce chemical aging of the asphalt binder. Furthermore, it generally stiffens the mechanical properties of the asphalt mixture which has a positive impact on the rutting resistance of the mixtures. In parallel, the resistance to cracking is also mentioned to be improved. The National Lime Association survey of 2003 gave some precise numbers on the changes in asphalt mixtures durability associated with the use of hydrated lime (Hicks and Scholz 2003). 31/63 Asphalt Durability The survey was performed by sending a questionnaire to all the agencies that are experienced in the use of hydrated lime. The full results are given in Table 4. From these data, it can be seen that the life expectancy for all types of roads is increased by 2 to 10 years when hydrated lime is added. Given that the life expectancy of untreated roads ranges from 5 to 20 years, the relative improvement goes from 20 up to 50% higher durability. The European experience is not yet as developed as in the USA, but the beneficial effect of hydrated lime on asphalt mixture durability has also been largely reported. As an example, the Sanef motorway company, managing 1,740 km of highways in Northern France, currently specifies hydrated lime in the wearing courses of its network (Raynaud 2009). Sanef observed that hydrated lime modified asphalt mixtures have a 20-25% higher durability (Raynaud 2009). Similar observations led the Netherlands to specify hydrated lime in porous asphalt, a type of mix that now covers 70% of the highways in the country. As a result, hydrated lime is being increasingly used in asphalt mixtures in most European countries, in particular Austria, France, the Netherlands, the United Kingdom and Switzerland (Lesueur 2010). 7.2.4 Others (liquid antistrips, fibers, epoxy asphalt) Other additives are also used in asphalt mixtures to improve their durability; these include liquid antistips, fibers and expoxy asphalts. Adhesion promoters, such as polyamines and polyphosphates often referred to as antistrips, are used when the moisture resistance of the mixture is not sufficient. Although each formula has a distinct dosage, a typical content is 0.5wt. % based on asphalt binder. Adhesion promoters are generally surface active agents. The polar head is typically amine based, but other chemistries are available. A key issue is the thermal stability of the molecule in the asphalt and care must be taken not to store the treated asphalt binder for a day or more at elevated temperatures. Fibers are used in some mixtures, especially in the German split mastic asphalts (SMAs) to prevent draindown of the asphalt binder. The fiber enables the use of thicker binder films (i.e., higher binder content mixes) with no risk of binder drainage. Fibers are also used as a reinforcement aid to improve the fatigue performance of the mix (Button and Hunter 1984; McDaniel and Shah 2003). In an accelerated pavement testing experiment (Gibson et al. 2012), the fiber modified section was found to be considerably more durable than that of the control. The use of epoxy asphalt has been limited to bridge decks and small critical surfaces (i.e., dangerous curves, and intersections) due to its higher cost. This is the subject of an ongoing study of Long-Life Surfaces for busy roads by the OECD (2008) where its application is as a thin asphalt mix overlay and for use in an open graded friction course (OGFC). 32/63 Asphalt Durability 8. ASTM / AASHTO/ EN METHOD SPECIFICATIONS AND TESTS Aggregate • • • • • • • Aggregate Durability Index, AASHTO 210-10. Resistance to Degradation of Small-Size Coarse Aggregate by Abrasion and Impact in the Los Angeles Machine, AASHTO 96-02; ASTM C131 and C535, EN 1097-2. Resistance of Coarse Aggregate to Degradation by Abrasion in the Micro-Deval Apparatus, AASHTO T327-09; EN1097-1. Polishing Stone Value, EN 1097-8. The Qualitative Detection of Harmful Clays of the Smectite Group in Aggregates Using Methylene Blue, AASHTO T 330-07. Sizes of Aggregate for Road and Bridge Construction, AASHTO M43-05. Uncompacted Void Content of Fine Aggregate, AASHTO T 304. Asphalt Binder • • • • • • • Performance-Graded Asphalt Binder, AASHTO M 320, EN 12591. Performance-Graded Asphalt Binder Using Multiple Stress Creep Recovery (MSCR) Test, AASHTO MP 19-10. Determining the Rheological Properties of Asphalt Binder Using a Dynamic Shear Rheometer (DSR), AASHTO T315. Determining the Flexural Creep Stiffness of Asphalt Binder Using the Bending Beam Rheometer (BBR), AASHTO T313. Determining the Rheological Properties of Asphalt Binder in Direct Tension (DT), AASHTO T314. Multiple Stress Creep Recovery (MSCR) Test of Asphalt Binder Using a Dynamic Shear Rheometer, AASHTO TP 70. Determining the Cracking Temperature of Asphalt Binder Using the Asphalt Binder Cracking Device, AASHTO TP92-11. Aging • Effect of Heat and Air on a Moving Film of Asphalt Binder (Rolling Thin-Film Oven Test), AASHTO T240-09; ASTM D2872; EN 12607. • Accelerated Aging of Asphalt Binder Using a Pressurized Aging Vessel (PAV), AASHTO R28; ASTM D 6521; EN14769. Moisture • Boiling Water Test, ASTM D-1075. Chemistry • • • • Terminology, EN 12597. Corbett Separation, D4124. Solubility of Bituminous Materials, AASHTO T44-03. Test Method for Molecular Weight of Hydrocarbons by Thermoelectric Measurement of Vapor Phase, ASTM D2503. 33/63 Asphalt Durability • SARA Separation, ASTM D4124. Asphalt Mix • • • • • • California Bearing Ratio, AASHTO T 193. Resilient Modulus Test, AASHTO T 307. Determining Dynamic Modulus of Hot mix Asphalt (HMA), AASHTO T-342-11; EN 12697-26. Determining the Creep Compliance and Strength of Hot Mix Asphalt (HMA) Using the Indirect Tensile Test Device, AASHTO 322-07. Compressive Strength of Hot Mix Asphalt, AASHTO T167-10. Dynamic modulus test, ASTM D3497. Aging • • Mixture Conditioning of Hot Mix Asphalt (HMA), AASHTO R30. Cantabro Test, EN 12697-17. Moisture • • • • • • Resistance of Compacter Hot Mix Asphalt (HMA) to Moisture-Induced Damage (Indirect Tensile Strength Ratio with one freeze-thaw cycle- Lottman), AASHTO T283. Hamburg Wheel-Track Testing of Compacted Hot Mix Asphalt, AASHTO 324; EN12697-22 method B under water. Freeze-thaw Pedestal Test (Plancher et al. 1980). Resistance of Compacted Hot Mix Asphalt (HMA) to Moisture-Induced Damage, AASHTO 28307. Retained Tensile Strength or Indirect Tensile Strength Ratio, ASTM D4867; EN 12697-12 – method A. Immersion / Compression Test; AASHTO T165; ASTM D1075; EN12697-12- method B (Duriez). Rutting • • • • Determining the Dynamic Modulus and Flow Number for Hot Mix Asphalt (HMA) Using the Asphalt Mixture Performance Tester (AMPT), AASHTO TP79-10. Developing the Dynamic Modulus Master Curves for Hot Mix Asphalt (HMA) Using the Asphalt Mixture Performance Tester (AMPT), AASHTO PP61-10. Determining the Creep Compliance and Strength of Hot Mix Asphalt (HMA) Using the Indirect Tensile Test Device, AASHTO T322; EN 12697-22. Rutting Resistance of Asphalt Mixtures, EN 12697-22. Thermal Cracking / Fatigue • • • • • • Determining the Fatigue Life of Compacted Hot mix Asphalt (HMA) Subjected to Repeated Flexural Bending, AASHTO T321-07. Resilient Modulus by Indirect Tension (IDT) Subjected to Repeated Flexural Bending, AASHTO TP31. Indirect tensile strength/Resilient Modulus Test, ASTM D4123. Thermal Stress Restrained Specimen Tensile Strength (TSRST), AASHTO TP10. Determining the Fatigue Life of Compacted Hot Mix Asphalt (HMA), AASHTO T321. Fatigue Resistance of Asphalt Mixtures, EN 12697-24. 34/63 Asphalt Durability 35/63 Asphalt Durability 9. SUMMARY In a World concerned with environmentally friendly construction technologies and sustainability, asphalt durability takes on an added significance. Asphalt mixtures are now very technical materials obtained by careful selection of raw materials among which are asphalt binder and aggregate, and then mixed in dedicated computer-controlled plants. Due to the rising cost of asphalt binders, there is a keen interest in the concept of sustainable asphalt mixtures. However, the environmental aspects are mostly covered by (i) minimizing wastes through the recycling of Reclaimed Asphalt Pavements (RAP) in new production (ii) supplementing the asphalt through the use of other by products such as reclaimed asphalt shingles (RAS), sulfur and recycled motor oil, and (iii) using lower manufacturing temperatures in the manufacturing of the so-called warm or even semi-warm mixtures, thus decreasing CO2 emissions and energy consumption. Hence, the basic principle of waste management, i.e. preventing and reducing waste generation, is not fully taken into account by the industry and more durable asphalt pavements should be a clear objective. Therefore, this chapter addresses the main factors affecting the durability of asphalt pavements, namely, their ability to maintain satisfactory performance and structure in long-term service under increasingly demanding conditions. Considering that asphalt pavement deterioration is attributed to combined effects of chemical aging, climate and traffic, some guidelines emerge in order to enhance their durability. First, a careful selection of the ingredients, especially the asphalt binder, insures a proper function (climate and loading) and resistance to chemical aging. Tests exist that simulate longterm aging and they are therefore highly recommended to be part of any binder specification. Then, the resistance to climate, in particular moisture damage, must be evaluated using appropriate test methods on the asphalt mixture and setting sound specifications. Finally, the resistance to traffic is first obtained by an appropriate pavement design and then by the validation of the mechanical properties of the mixture. Still, the construction stage is critical in guaranteeing good pavement durability, because too-thin a layer, badly executed joints or lack of tack coat are critical factors that can lead to premature failure. From a mixture formulation standpoint, several solutions are available to the designers in order to improve any of the mentioned key properties. In particular, hydrated lime and polymers have been the most widely used additives to enhance the durability of asphalt mixtures. In all cases, a tight control of the raw materials properties (asphalt binder, aggregate, recycled materials, and additives), of the manufacturing process (quantities and qualities of materials, and temperature) and of the laying operations (thickness, and density) is necessary in order to obtain long lasting roads. Improvements in the durability of asphalt are arising from multiple fronts. Improved testing and classification of asphalt binders is enhancing the screening and identification of poor performing binders for a given climatic region or traffic loading. Additives and polymer modification are increasingly being used to target potential distresses, and together with improvements in asphalt compaction, this is is leading to more durable binders and pavements. Ultimate goals are to be able to predetermine the asphalt pavement’s lifetime and utilize pavement preservation strategies to minimize or offset embrittlement of the pavement. environmental as well as economic perspective, consideration of asphalt durability is paramount to the sustainability of asphalt and ensuring that asphalt remains a cost effective paving material. Thus from 36/63 Asphalt Durability 10. ACKNOWLEDGEMENT The authors wish to acknowledge Drs. Raj Dongre and Nelson Gibson for their critical review of the manuscript. 11. FURTHER READING American Association of State Highway and Transportation Officials.2011. Standard specifications for transportation materials and methods of sampling and testing : Part 2A Tests. AASHTO, Washington, DC. Annual Book of ASTM Standards, Philadelphia : American Society Testing Materials. Asphalt Institute Manual Series 2 (MS-2) Fatigue Response of Asphalt-Aggregate Mixes, SHRP National Research Council, SHRP A404, Washington DC 1994, 309 pages 13. NOMENCLATURE AND UNITS Asphalt binder – an asphalt-based cement that is produced from petroleum residue either with or without the addition of non-particulate organic modifiers. Asphalt modifier -- any material of suitable manufacture that is used in virgin or recycled condition and that is dissolved, dispersed, or reacted in asphalt binder to enhance its performance. 37/63 est method Wheel Tracking ensile Strength SR) n Aging Tensile SATS) Asphalt Durability 14. TABLES Standard EN 12697-22 – method B under water EN 12697-12 – method A EN 12697-12 – method B EN 12697-17 UK Specification for Highway Works – Clause 953 AASHTO T283 Tex 531-C Lottman Lottman 1978 eze-Thaw Pedestal Plancher et al. 1980 ensile strength plitting ratio / nsile Strength / nicliff test) ASTM D4867 n/Compression ASTM D1075 AASHTO T165 Marshall l Test ASTM D3625 Tex 530-C Type of specimen 260mm x 300mm rectangular slabs with final thickness Testing method Wheel tracking device under water Conditioning - testing under water at 50°C Test Resu - Rut depth in m 100mm diameter (or 150 or 160 for large aggregate size) cylindrical specimen of the asphalt mixture to be tested 100mm diameter (or 80 or 120 or 150 or 160) cylindrical specimen of the asphalt mixture to be tested Indirect Tensile Strength (ITS) at 25°C and 50mm/min - specimens in vacuum (7kPa) for 30min - 70hrs in water at 40°C - 2hrs at 25°C - specimens in vacuum (47kPa) for 120min - 7days in water at 18°C ITS ratio in % (a conditioning / no conditioning) 101.6mm diameter x 63.5mm cylindrical specimen of the asphalt mixture to be tested (generally a porous asphalt) 100mm diameter x 60mm cylindrical specimen of the asphalt mixture to be tested, cored from a slab with 8% voids Mass loss after 300revolutions in the Los Angeles test (without steel balls) Indirect Tensile Stiffness Modulus measured at 20°C using the Nottingham Asphalt Tester Indirect Tensile Strength (ITS) at 25°C and 50.8mm/min Generally the same as ASTM D1075 Mass loss ratio ( conditioning / no conditioning) - specimens in vacuum (55kPa) for 30min - 65hrs at 85°C and 2.1MPa in water saturated vessel - 24hrs at 30°C and 2.1MPa - 70-80% pore saturation - 16hrs at -17.8°C - 24hrs in water at 60°C - 2hrs in water at 25°C Lottman conditioning but with consecutive freezethaw cycles (generally from 1 to 20) - briquette immersed in distilled water - 15 hrs at -12°C - 45min in water at 24°C - 9hrs at 49°C then repeat - 55-80% pore saturation - 24 hrs in water at 60°C - 1hr at 25°C Indirect Tensile ratio in % (after conditioning / no conditioning) Compressive strength at 25°C and 5mm/min - 4 days in water at 48.9°C or 1 day in water at 60°C Compressive stre ratio (after condi no conditioning) Marshall stability at 60°C – 50.8mm/min - 24 hours in water at 60°C Stability ratio (af conditioning / no conditioning) Visual (aggregate surface covered in asphalt) - Asphalt mixture in boiling water for 10min % of retained asp after boiling 101.6mm diameter x 63.5mm cylindrical specimen of the asphalt mixture to be tested compacted to 6.57.5% voids 101.6mm diameter x 63.5mm cylindrical specimen of the asphalt mixture to be tested compacted to 6.57.5% voids 41.3mm diameter x 19mm cylindrical briquette of 0.4/0.8 sand coated with asphalt at optimum + 2% compacted by static pressure of 27.58 kN for 20min 101.6mm diameter x 63.5mm cylindrical specimen of the asphalt mixture to be tested compacted by any mean (static / Marshall, etc.) to 6-8% air voids 101.6mm diameter x 101.6mm cylindrical specimen of the asphalt mixture to be tested compacted by static compaction on both sides (3,000 psi during 2min) 101.6mm diameter x 76.2mm cylindrical specimen of the asphalt mixture to be tested compacted by impact compaction (50 or 75 blows) (300g + asphalt content) of asphalt mixture to be tested or (100g + asphalt) of 4.8/9.8 aggregate Compressive strength at 18°C and 55mm/min Indirect Tensile Strength (ITS) at 25°C and 50.8mm/min Visual (crack) Indirect Tensile Strength at 25°C and 50.8mm/min Table 1: Most used testing methods for evaluating the resistance to moisture or frost damage for asphalt mixtures. 38/63 ITS ratio in % (a conditioning / no conditioning) ITS ratio (after conditioning / no conditioning) ITS ratio vs num freeze-thaw cycl Number of freez cycles to failure ITS ratio (after conditioning / no conditioning) Asphalt Durability 39/63 AAA-1 AAB-1 AAC-1 AAD-1 AAF-1 AAG-1 AAK-1 AAM-1 origin Canada USA Canada USA USA USA Venezu USA ela C wt.% 83.9 82.3 86.5 81.6 84.5 85.6 83.7 86.8 H wt.% 10.0 10.6 11.3 10.8 10.4 10.5 10.2 11.2 H+C wt.% 93.9 92.9 97.8 92.4 94.9 96.1 93.9 98.0 H/C molar 1.43 1.55 1.57 1.59 1.48 1.47 1.46 1.55 O wt.% 0.6 0.8 0.9 0.9 1.1 1.1 0.8 0.5 N wt.% 0.5 0.5 0.7 0.8 0.6 1.1 0.7 0.6 S wt.% 5.5 4.7 1.9 6.9 3.4 1.3 6.4 1.2 V ppm 174 220 146 310 87 37 1480 58 Ni ppm 86 56 63 145 35 95 142 36 Mn g/mol 790 840 870 700 840 710 860 1300 Table 2: Elemental analysis for the core SHRP asphalt binders (data from Mortazavi and Moulthrop 1993). Asphalt Durability H/C C H O N S Mn Asphalt cement Saturates Aromatics 1.5 1.9 1.5 % 80-88 78-84 80-86 % 8-12 12-14 9-13 % 0-2 < 0.1 0.2 % 0-2 < 0.1 0.4 % 0-9 < 0.1 0-4 g/mol 600-1,500 470-880 570-980 Resins Asphaltenes 1.4 1.1 67-88 78-88 9-12 7-9 0.3-2 0.3-5 0.2-1 0.6-4 0.4-5 0.3-11 780-1,400 800-3,500 Solvent in ASTM D4124 n-heptane toluene and toluene/methanol 50/50 trichloroethylene n-heptane insoluble Table 3: Chemical properties of asphalt binder and the SARA fractions: typical H/C, elemental analysis, number average molar mass and solvent used in ASTM D-4124 (data from Lesueur, 2009). 41/63 Asphalt Durability Table 4: Life expectancy (years) of hydrated lime treated and untreated mixes in the USA (from Hicks and Scholz 2003). 42/63 Asphalt Durability 15. FIGURES Figure 1: Aggregate stripping as a consequence of moisture induced-damage (from Miller and Bellinger 2003).[For scale, the pavement marking is 4” wide.] 43/63 Asphalt Durability Figure 2: Rutting in an asphalt mixture (from Miller and Bellinger 2003). [For scale, the pavement marking is 4” wide.] 44/63 Asphalt Durability Figure 3: Fatigue cracking (alligator cracking) in an asphalt mixture (from Miller and Bellinger 2003). [These interconnected cracks can be from 1” to 6” in length.] 45/63 Asphalt Durability Longitudinal Friction Coefficient at 120km/h (%) 40 35 30 VTAC neat asphalt VTAC polymer-modified VTAC asphalt-rubber VTAC fibers 25 20 15 10 10,000 100,000 1,000,000 10,000,000 Number of cumulated heavy trucks Figure 4: Longitudinal skid resistance at 120 km/h versus cumulative traffic (number of heavy trucks) for different kinds of 0/10 Very Thin Asphalt Concrete: with neat asphalt binder, with polymer-modified asphalt, with asphalt-rubber, with fibers (data from the Carat database adapted from Stasse 2000). 46/63 Asphalt Durability Observed durability in Europe (years) 30 25 20 15 10 5 0 Asphalt Concrete (AC) Thin Layer AC Very Thin Layer AC Ultra Thin Layer AC Porous Double Stone Asphalt Layer PA Mastic (PA) Asphalt Hot Rolled Asphalt Mastic Asphalt Figure 5: Durability of asphalt layers on major European roads as a function of type of wearing course. The error bar gives the 15% lowest and 85% highest observed durability, respectively (adapted from EAPA 2007). 47/63 Asphalt Durability Figure 6: Most observed degradation modes in Europe (from ELLPAG – [ELLPAG Phase 1 Report 2004] reproduced with permission of FEHRL). Each country gave a rating of 0 (irrelevant) – 5 (major issue) to quantify the importance of this damage on its road network. AT = Austria, BE = Belgium, HR = Croatia, DK = Denmark, FI = Finland, FR = France, DE = Germany, GR = Greece, HU = Hungary, IS = Iceland, IE = Ireland, IT = Italy, NL = Netherlands, NO = Norway, PL = Poland, PT = Portugal, RO = Rumania, SI = Slovenia, ES = Spain, CH = Switzerland, UK = United Kingdom 48/63 Asphalt Durability Figure 7: Comparison of the effectiveness of several test methods in order to predict moisture damage as evaluated by State agencies experience (from Hicks 1991). 49/63 Asphalt Durability 25 Rut Depth (mm) 20 15 10 5 0 0 5000 10000 15000 20000 Wheel Tracking Cycles Figure 8. Hamburg Wheel Tracker Rut depth is plotted on the vertical axis versus wheel tracking cycles plotted on the horizontal axis. Three replicate curves from a single mixture are plotted together where the typical trend is a rapid increase in rut depth (compaction) which then grows nearly linearly and then transitions into a tertiary phase where rut depth growth accelerates with loading cycles (from Gibson et al. 2012). 50/63 Asphalt Durability Figure 9. Photograph of pneumatic wheel in the French Pavement Rutting Tester and rutted test specimen (from Gibson et al. 2012). [Slab dimensions are 20 inches (500 mm) long, 7 inches (180 mm) wide, nad 4 or 2 inches (100 or 50 mm) thick.] 51/63 Asphalt Durability Figure 10 a. Dynamic modulus |E*| is plotted on the left vertical axis and phase angle is plotted on the right vertical axis versus number of fatigue cycles on the horizontal axis. Typical trends observed during a stress control fatigue test is illustrated with a gradual reduction of modulus and gradual increase in phase angle as loading cycles accumulate (from Gibson et al. 2012). Figure 10b. Dynamic modulus |E*| is plotted on the left vertical axis and phase angle is plotted on the right vertical axis versus number of fatigue cycles on the horizontal axis. Typical trends observed during a strain control fatigue test is illustrated with a gradual reduction of modulus and gradual increase and then decrease in phase angle as loading cycles accumulate (from Gibson et al. 2012). 52/63 Asphalt Durability Figure 11: Functional groups present in asphalt binder (from Branthaver et al. 1994). 53/63 Asphalt Durability A/ California Coastal (AAD-1) B/ West Texas Intermediate (AAM-1) Figure12: Average molecular structure for two asphalt binders of extreme compositions: California Coastal (AAD-1): A) and West Texas Intermediate (AAM–1): B) (after Jennings et al. 1993). 54/63 Asphalt Durability 100% weight (%) 80% Asphaltenes Resins Aromatics Saturates 60% 40% 20% 0% AAA-1 AAB-1 AAC-1 AAD-1 AAF-1 AAG-1 AAK-1 AAM-1 Figure 13: Separation into SARA fractions for the core SHRP asphalt binders (data from Mortazavi and Moulthrop 1993). 55/63 Asphalt Durability Figure 14.Asphaltene Determinator Separation Profile for SHRP core asphalt AAK-1 (Boscan) using 500 nm Absorbance Detectors (from WRI 2010). The maltenes are eluted with the heptanes, and the asphaltenes are fractionated using solvents of increasing polarity. The cyclohexane peak is a measure of the highly alkyl substituted pericondensed aromatic components of the asphaltenes; the toluene peak is a measure of the less alkyl substituted pericondensed aromatic molecules, and the methylene chloride: methanol peak is a measure of the highly pericondensed aromatic molecules. 56/63 Asphalt Durability Modulus (MPa) 100,000 -10°C 0°C 10°C 20°C 30°C 40°C 10,000 1,000 100 1 10 Frequency (Hz) Figure 15: Complex tensile modulus versus temperature and frequency for an asphalt mixture. Figure 16: Concept of Pessimum Voids stemming from relationship between strength of mixtures and air void content (from Terrel and Al-Swailmi 1994). 57/63 100 12. REFERENCES (ASTM, AASHTO & EN SPECS & TESTS LISTED SEPARATELY IN SECTION 8) AASHO (American Association of State Highway Officials). 1962. AASHO Road Test Report 5. Highway Research Board special report 61E. Publication 954. Washington, D.C.: National Research Council. Abdallah,I., E. Mahmoud, S. Nazarian, and E. Masad. 2008. Quantifying the role of coarse aggregate strength on resistance to load in HMA for blended aggregates, Research report 0-5268-3. The University of Texas El Paso: Center for Transportation Infrastructure Systems. Abraham, H. 1960. Asphalt and allied substances, 6th Ed. New York: Van Nostrand. Alexander, S.H.1973. Method of treating asphalt, USA Patent 3,751,278 filed Aug. 7, 1973. Anderson, D. A., D. W. Christensen, H. U. Bahia et al. 1994. Binder characterization and evaluation, Volume 3: Physical characterization. SHRP report A-369. Washington, DC: National Research Council. Arnold, T., S. Needham, and J.Youtcheff. 2009. The use of phosphoric acid as a modifier for hot mix asphalt, Transportation Research Board, annual meeting CD-ROM. Aschenbrener, T., R. McGennis, and R. Terrel. 1995. Comparison of several moisture susceptibility tests in pavements of known field performance. Journal of the Association of Asphalt Paving Technologists 66:163-208. Asphalt Institute. 1947. The Asphalt handbook. Lexington KY: Asphalt Institute. Asphalt Institute. 1994. Superpave: Performance graded asphalt binder specification and testing/SP-1. Lexington KY: Asphalt Institute. Asphalt Institute. 2001. Superpave mix design. Lexington KY: Asphalt Institute. Baumgardner, G.L. 2010. Why and how of polyphosphoric acid modification – an industry perspective. Journal of the Association of Asphalt Paving Technologists79:663-678. Bell, C. 1989. Summary report on aging of asphalt-aggregate systems, SHRP report A-305. Washington DC: National Research Council. Bhasin, A., R. Bommavaram, M. L. Greenfield et al. 2011. Use of molecular dynamics to investigate self-healing mechanisms in asphalt binders. Journal Materials Civil Engineering 23:485-92. Branthaver, J. F., J. C. Petersen, R. E. Robertson et al. 1994. Binder characterization and evaluation – vol. 2 Chemistry. SHRP report A-368, Washington, DC: National Research Council. Brosseaud, Y., V. Le Turdu and Y. Delanne. 2005. Adhérence des revêtements de chaussées routières. Bulletin Laboratoires Ponts Chaussées 255:71-90. Burstyna I., H. Kromhouta and P. Boffettab. 2000. Literature review of levels and determinants of exposure to potential carcinogens and other agents in the road construction industry. Journal American Industrial Hygiene Association 61:715-26. Button, J. and T. Hunter. 1984. Synthetic fibers in asphalt paving mixtures. FHWA report TX-85-73-319-1F. Button, J., C. Estakhri and A. Wimsatt. 2007. A synthesis of warm-mix asphalt. TTI report SWUTC/07/0-5597-1. College Station TX: Texas Transportation Institute. Caltrans (California Department of Transportation). 2003. Asphalt rubber usage guide. Sacramento: Caltrans. Christensen, D. W. and D. A. Anderson. 1991. Rheological evidence concerning the molecular architecture of asphalt cements. Proceedings Chemistry of Bitumen: Rome 2:568-95. Asphalt Durability Claudy, P., J.-M. Létoffé, G. N. King, et al. 1992. Characterization of asphalt cements by thermomicroscopy and differential scanning calorimetry: correlation to classic physical properties. Fuel Science Technology International 10:735-65. Connan, J. 1999. Use and trade of bitumen in antiquity and prehistory: molecular archaeology reveals secrets of past civilizations. Phil. Trans. R. Soc. Lond. B 354:33-50. Copeland, A. 2011. Reclaimed Asphalt Pavement in Asphalt Mixtures: State of the Practice. Report FHWA-HRT-11-021, Federal Highway Administration. Corbett, L.W. 1969. Composition of asphalt based on generic fractionation using solvent deasphalteneing, elution-adsorption chromatography and densiometric characterization. Analytical Chemistry 41: 576-579. Curtis, C. W., Y. W. Jeon and D. J. Clapp. 1989a. Adsorption of asphalt functionalities and oxidized asphalts on aggregate surfaces. Fuel Sci. Technol. Int. 7:1225-8. Curtis, C. W., Y. W. Jeon, D. J. Clapp, et al. 1989b. Adsorption behavior of asphalt functionalities, AC-20, and oxidized asphalts on aggregate surfaces. Transportation Research Record 1228:112-27. D’Angelo, J., E. Harm, J. Bartoszek, et al. 2008. Warm-mix asphalt: European practice. Report FHWA-PL-08-007. Arlington VA: Federal Highway Administration. D’Angelo, J., et al., 2007. Revision of the Superpave high temperature binder specification: the multiple stress creep recovery test. Journal of the Association of Asphalt Paving Technologists 76:123-162. Delorme, J.-L., De La Roche C. and L. Wendling Eds. 2007. LCPC bituminous mixtures design guide. Paris: LCPC. Di Benedetto, H., C. de La Roche, H. Baaj, et al. 2003. Fatigue of bituminous mixtures. Materials and Structures 37:202-16. Dunn, L. 2001. A basic asphalt recycling manual. Annapolis MA: Asphalt Recycling and Reclaiming Association. Durand, B. 2003. A history of organic geochemistry. Oil & Gas Sci. Technology – Rev. IFP 58:203-31. Duriez, M. and J. Arrambide. 1954. Liants Hydrocarbonés. Paris (France): Dunod. Edwards, Y., Y. Tasdemir and U. Isacsson. 2006. Influence of commercial waxes and polyphosphoric acid on bitumen and asphalt concrete performance at low and medium temperatures. Materials and Structures 39:725-37. EAPA (European Asphalt Pavement Association). 2007. Long-life asphalt pavements – technical version. Brussels (Belgium): EAPA Ed. FEHRL-ELLPAG (Forum of European National Highway Research Laboratories – European Long-life Pavement Group). 2004. Making best use of long-life pavements in Europe – Phase 1: A guide to the use of long-life fully-flexible pavements, Brussels (Belgium): FEHRL Ed. Ferne, B. 2006. Long-life pavements—A European study by ELLPAG. Int. J. Pavement Engineering 7:91–100. Francken, L. Ed. 1998. Bituminous binders and mixes. RILEM Report 17. London (UK): E & FN Spon. Gal, J.-F. 1992. Evolution de la perméabilité des enrobés drainants. Revue Générale des Routes et Aérodromes 702:118-9. Gibson, N.H. 2006. A viscoelastoplastic continuum damage model for the compressive behavior of asphalt concrete, Ph. D. Dissertation. College Park, MD: University of Maryland. Gibson, N., X. Qi, A. Shenoy, et al. In Press. Full-scale accelerated performance testing for Superpave and structural validation. Final report Transportation pooled fund study 59/63 Asphalt Durability TPF-5(019) and SPR-2(174) Accelerated pavement testing of crumb rubber modified asphalt pavements. Lanthum, MD : FHWA. GNB (Groupe National Bitume). 1997. Susceptibilité au Vieillissement des Bitumes – Expérimentation A08. Research report CR19. Paris (France): LCPC Ed. Han, R., X. Jin, and C. J. Glover. 2011. Modeling pavement temperature for use in binder oxidation models and pavement performance prediction. Journal Material Civil Engineering 23:351-9. Herrington, P.R., S. Reilly, and S. Cook. 2005. Porous asphalt durability test, Transfund New Zealand research report No. 265. 32pp. Hicks, R. G. 1991. Moisture damage in asphalt concrete. NCHRP synthesis of highway practice 175, Washington, DC: Transportation Research Board. Hicks, R. G. and T. V. Scholz. 2003. Life cycle costs for lime in hot mix asphalt, 3 vol., Arlington, VA: National Lime Association. Huang, B., L. N. Mohammad, A. Raghavendra, et al. 1999. Fundamentals of permeability in asphalt mixtures. Journal of the Association of Asphalt Paving Technologists 68:47993. Isacsson, U. and H. Zeng. 2003. Cracking of asphalt at low temperature as related to bitumen rheology. Journal Materials Science 33:2165-70. Jennings, P.W., J.A. Pribanic, M.A. Desando, et al. 1993. Binder characterization and evaluation by nuclear magnetic resonance spectroscopy. SHRP report A-335. Washington, DC: National Research Council. Jennings, P. W., M. A. Desando, M. F. Raub et al. 1992. NMR spectroscopy in the characterization of eight selected asphalts. Fuel Science Technology International 10:887-907. Kim, S., S.-W. Loh, H. Zhai et al. 2001. Advanced characterization of crumb-rubber modified asphalts, using protocols developed for complex binders. Transportation Research Record 1767:15-24. Kim, Y., D.N. Little and R.L. Lytton. 2003. Fatigue and healing characterization of asphalt mixtures. Journal Materials Civil Engineering 15:75-83. Kringos, N., A. Scarpas, C. Kasbergen, et al. 2008a. Modelling of combined physicalmechanical moisture-induced damage in asphaltic mixes. Part 1: Governing processes and formulations. International Journal Pavement Engineering 9:115-28. Kringos, N., A. Scarpas, A. Copeland, et al. 2008b. Modelling of combined physicalmechanical moisture-induced damage in asphaltic mixes. Part 2: Moisture susceptibility parameters. International Journal Pavement Engineering 9:129-51. Kutay, M.E., N.H. Gibson, J. Youtcheff. 2008. Conventional and viscoelastic continuum damage (VECD) based fatigue analysis of polymer modified asphalt pavements. Journal of the Association of Asphalt Paving Technologists 77: 395-434.Lamontagne, J., P. Dumas, V. Mouillet, et al. 2001. Comparison by fourier transform infrared (FTIR) spectroscopy of different aging techniques: application to road bitumens. Fuel 80:483– 8. Lay, M. G. 1992. Ways of the world. New Brunswick. NJ: Rutgers University Press. LCPC-SETRA (Laboratoire Central des Ponts et Chaussées – Service des Etudes Techniques des Routes et Autoroutes). 1997. French design manual for pavement structures. Paris (France): LCPC Ed. Lee, H.J. and Y.R. Kim. 1998a. Viscoelastic constitutive model for asphalt concrete under cyclic loading. Journal of Engineering Mechanics, ASCE, 124:32-40. Lee, H.J. and Y.R. Kim. 1998b. Viscoelastic continuum damage model for asphalt concrete with healing. Journal of Engineering Mechanics, ASCE, 124:1224-1232. 60/63 Asphalt Durability Lee, W. and K. Mahboub Eds. 2006. Asphalt mix design and construction: past, present, and future. Reston, VA: American Society of Civil Engineers. Lesueur, D. 2010. Hydrated lime: a proven additive for durable asphalt pavements – critical literature review. Brussels: European Lime Association. Lesueur, D. 2009. The colloidal structure of bitumen: consequences on the rheology and on the mechanisms of bitumen modification. Advances in Colloid Interface Science 145:42-82. Li, X. and N. Gibson. 2011. Mechanistic Characterization of Aggregate Packing to Assess Gyration Levels During HMA Mix Design. Asphalt Paving Technology, Journal of the Associaiton of Asphalt Paving Technologists. 80:.33-64.Little, D. N. and J. A. Epps. 2001. The benefits of hydrated lime in hot mix asphalt. Arlington, VA: National Lime Association. Lottman, R.P.1978. Predicting moisture-induced damage to asphaltic concrete. NCHRP Report 192. Washington DC: Transportation Research Board. Marasteanu, M. O., X. Li, T. R. Clyne, et al. 2004. Low temperature cracking of asphalt concrete pavements. Report MN/RC 2004-23. St-Paul, MN: MN Department of Transportation Ed. Masad, E.A. 2005. Aggregate Imaging System (AIMS): Basics and Applications. FHWA/TX-05/5-1707-01-1. College Station, TX: Texas Transportation Institute. Masad, E., A. Rezaei, A. Chowdhury, et al. 2009. Predicting asphalt mixture skid resistance based on aggregate characteristics. TTI Report 0-5627-1. College Station, TX: Texas Transportation Institute. Mauduit, C., F. Hammoum, J.-M. Piau, et al. 2010. Quantifying expansion effects induced by freeze–thaw cycles in partially water saturated bituminous mix: laboratory experiments. Road Materials Pavement Design 11:443-57. McDaniel, R. and A. Shah. 2003. Asphalt additives to control rutting and cracking. FHWA report IN-JTRP-2002-29. McNally, T. Ed. 2011. Polymer modified bitumen: properties and characterization. Cambridge, UK: Woodhead Publishing. Miller, J.S. and W.Y. Bellinger. 2003. Distress identification manual for the long-term pavement performance program. FHWA-RD-03-031. Lanthum, MD: FHWA. Mohammad, L.N. and K.A. Shamsi. 2007.A Look at the Bailey method and Locking Point Concept in Superpave Mixture Design. Transportation Research Circular E-C124, Practical Approaches to Hot-Mix Asphalt Mix Design and Quality Control Testing, Transportation Research Board, Washignton DC. Monismith, C. L. 2006. State of the art: bituminous materials mix design. In Asphalt Mix Design and Construction: Past, Present, and Future, ed. W. Lee and K. Mahboub. Reston, VA: American Society of Civil Engineers. Mortazavi M. and J. S. Moulthrop. 1993. The SHRP materials reference library, SHRP Report A-646, Washington DC: National Research Council. Newcomb, D. E., M. Buncher and I. J. Huddleston. 2001. Concepts of perpetual pavements. Transportation Research Circular 503:4-11. OECD (Organisation for Economic Co-Operation and Development). 2008. Long-life surfaces for busy roads. Paris: OECD/ITF. Ould-Henia, M., M. Rodriguez and A.-G. Dumont. 2004. Élaboration d’une méthode prédictive de l’orniérage des revêtements bitumineux. Lausanne (Switzerland): Ecole Polytechnique Fédérale de Lausanne. 61/63 Asphalt Durability Petersen, J. C. 2009. A review of the fundamentals of asphalt oxidation – chemical, physicochemical, physical property and durability relationships. Report EC140, Washington, DC: Transportation Research Board. PIARC (World Road Association). 1999. Use of modified bituminous binders, special bitumens and bitumens with additives, In road pavements, routes/roads 303 also available from LCPC Eds, Paris: PIARC. Plancher, H. G. Miyake, R.L. Venable and J.C. Petersen. 1980. A simple laboratory test to indicate the susceptibility of asphalt-aggregate mixtures to moisture damage from repeated freeze-thaw cycling. Proceedings of the Canadian Technical Asphalt Association 25: 246-262. Polacco, G., J. Stastna, D. Biondi, et al. 2004. Rheology of asphalts modified with glycidylmethacrylate functionalized polymers, Journal Colloid Interface Science 280:366-73. Potter , T. L., and B. Duval. 2001. Cerro Negro bitumen degradation by a consortium of marine benthic microorganisms. Environmental Science Technology 35:76-83. Raynaud C. 2009. L’ajout de chaux hydratée dans les enrobés bitumineux. BTP Matériaux 22:42-3. Read, J. and D. Whiteoak. 2003. The Shell Bitumen Handbook, 5th Ed., London: Thomas Telford Publishing. Redelius, P. 2006. The structure of asphaltenes in bitumen. Road Materials Pavement Desig special issue EATA:143-62. Saunier, J. 1968. Autoréparation des enrobes bitumineux. Revue Générale des Routes et Aerodromes 435:69-72. Sousa, J. B., J. Craus and C. L. Monismith. 1991. Summary report on permanent deformation in asphalt concrete. SHRP-A/IR-91-104 Report. Washington DC: National Research Council. Stasse, G. 2000. Caractérisation d´adhérence de revêtements de chaussées routières. Report Chaussées CR 25, Paris: LCPC Ed. Tangella, R., J. Craus, J. A. Deacon, et al. 1990. Summary report on fatigue response of asphalt mixtures. SHRP Report A-312, Washington, DC: National Research Council. Taylor, G.D. 2000. Materials in Construction: an Introduction. Longman Publisher. 3rd Ed. Terrel, R.L. and S. Al-Swailmi. 1994. Water sensitivity of asphalt-aggregate mixes: test selection. SHRP Report A-403. Washington, DC: National Research Council. Thomson, D. C. 1964. Rubber Modifiers, In Bituminous Materials: Asphalts, Tars and Pitches, ed, A. J. Hoiberg, Vol. 1, 375-414. New York: Interscience Publishers. Timoshenko, S. 1976. Strength of Materials. 2 vol. 3rd edition, Malabar, FL: Krieger Publishing Company Traxler, R.N. and H. E. Schweyer. 1936. Increase in viscosity of asphalts with time. Proceedings of American Society for Testing Materials 36:544–51. TRB Committee on General Issues in Asphalt Technology (A2D05) Traxler, R. N. 1961. Asphalt: Its Composition, Properties and Uses, New York: Reinhold Publishing. Wallman, C.-G. and H. Åström. 2001. Friction measurement methods and the correlation between road friction and traffic safety : a literature review. VTI report 911A, Stockholm: VTI. Waste Framework Directive (Directive 2008/98/EC) of the European Parliament and of the Council of 19 November 2008 on waste. Western Research Institute. 2010. Fundamental Properties of Asphalts and Modified Asphalts, III. Quarterly technical progress report January 1 – March 31, 2010 prepared 62/63 Asphalt Durability for Federal Highway Administration under contract no. DTFH61-07-D-00005, Laramie, WY : Western Research Institute. (www.westernresearch.org) Wolf, M. and R. Bachofen. 1991. Microbial degradation of bitumen matrix used in nuclear waste repositories. Naturwissenschaften 78:414-7. Wright, J. R. 1965. Weathering: Theoretical and Practical Aspects of Asphalt Durability, in Bituminous Materials: Asphalts, Tars and Pitches, ed. A. J. Hoiberg Ed. Vol. 2, Part 1, 249-306, New York: Interscience Publishers. Youtcheff, J.S. and D.R. Jones. 1994. Guidelines for asphalt refiners and suppliers. SHRP Report A-686. Washington, DC: National Research Council. 63/63