David A Anderson

Evaluation of fatigue criteria for asphalt binders

The original SuperPave asphalt binder specification criterion for fatigue, G* sin δ, has received considerable criticism. Recently, a time sweep using the dynamic shear rheometer (DSR) has been proposed as an alternative test method for developing load-associated fatigue information for asphalt binders. This proposed test method is examined with respect to a phenomenon called edge fracture. Edge fracture is reported in the literature for steady state and oscillatory flow in DSR, but it has not been reported for asphalt binders. The modulus, when plotted versus number of cycles generated in a time sweep test, has the appearance typical of fatigue behavior; however, the actual response of the material depends markedly on the initial modulus of the material. The development of the modulus with repeated shearing is described with respect to flow of the asphalt binder at its circumference. The data are examined with respect to their validity as a measure of fatigue, and recommendations with respect to the use of time sweep data in a binder specification are presented.

Low-Temperature Thermal Cracking of Asphalt Binders as Ranked by Strength and Fracture Properties

The original Superpave low-temperature specification for asphalt binders placed limits on the low-temperature stiffness and m-value. A recently approved alternative to the original Superpave asphalt binder low-temperature specification makes use of the measured stiffness and tensile strength of the binder to determine a critical cracking temperature. Thermal cracking temperatures are presented for 42 plain and modified asphalt binders. Thermal cracking temperatures determined by the original and recently approved alternative Superpave specification are compared. The fracture toughness, KIC, can also be used to evaluate low-temperature cracking properties of asphalt binders. Fracture properties obtained for 14 asphalt binders are compared with the thermal cracking temperatures determined by the original and recently adopted alternative Superpave specifications. The set of 14 binders was produced from a common base material but modified by different means. For the set of 14 binders, there is little difference in their ranking according to both the original and recently proposed alternative Superpave low-temperature criteria; however, their ranking is quite different on the basis of the fracture properties as measured by KIC. KIC appears to provide a much more discriminating ranking of the binders than either of the Superpave specification criteria.

Zero shear viscosity of asphalt binders

Recently there has been considerable interest, especially in Europe, in the use of zero shear viscosity (ZSV) as a specification criterion for asphalt binders. This interest is precipitated by the apparent inability of the current Superpave® criterion, G*/sin(δ), to capture the contribution to rutting resistance afforded by polymer modification. ZSV can be determined directly from long-term creep tests, but such tests are timeconsuming and are often very difficult to perform. Several alternative methods for determining the ZSV have been proposed in the literature, including extrapolating the dynamic viscosity to zero frequency; applying the Cross model to dynamic data; and superimposing multiple short-term, non-steady-state creep tests. A number of methods for determining the ZSV from both creep and dynamic data were evaluated. Laboratory test data for 10 unmodified and modified binders were obtained through a series of creep and dynamic experiments. ZSV values obtained from two of the more promising methods were compared, along with a comparison of the ZSV ranking with the Superpave grading temperature. Two of the methods provided very similar values for the ZSV when applied over a considerable range in test temperature, and the results from the two methods could be used interchangeably for the materials that were tested. The binders ranked quite differently when ranked according to their Superpave grading temperature or their ZSV.

PHYSICAL HARDENING OF ASPHALT BINDERS RELATIVE TO THEIR GLASS TRANSITION TEMPERATURES

Physical hardening (physical aging) is a process that occurs below room temperature in asphalt binders. Physical hardening causes time-dependent isothermal changes in the rheological behavior and specific volume of asphalt binders. The process is reversible: when the asphalt binder is heated to room temperature or above, the effect of physical hardening is completely removed. Physical hardening for amorphous materials is generally reported as occurring below the glass transition temperature (Tg), but this is not the case for asphalt binders, in which physical hardening is observed both above and below Tg. The glass transition temperature of asphalt binders is measured by using three different techniques: dilatometry, differential scanning calorimetry, and rheological considerations (peak in the loss modulus versus temperature). These three techniques give roughly equivalent estimates of the glass transition temperature. The behavior of physical hardening in asphalt binders is somewhat different than that reported for polymers and other organic materials. This difference is explained in terms of the presence of crystalline fractions in the asphalt binder. Techniques for modeling physical hardening are described, and possible explanations for the anomalous behavior of asphalt binders are given.

Establishing Linear Viscoelastic Conditions for Asphalt Binders

The linear viscoelastic regime is defined in terms of the constitutive relationship between the stress and the strain. The set of equations that define the fundamental linear viscoelastic material properties in the time and frequency domains and their relationship to one another is based on the validity of the linearity principle. A material must obey two simultaneous conditions to be linear viscoelastic: the homogeneity (also called proportionality) condition and the superposition principle. On the basis of these considerations a testing procedure was developed to check linear viscoelastic conditions for tests performed on asphalt binders with the dynamic shear rheometer (DSR), the bending beam rheometer (BBR), and direct tension (DT). The testing procedure for the DSR requires performing strain sweeps and multiwave single-point tests. For the BBR, tests performed using different constant loads are required. In addition, the recovery part of the specification test is recorded. For the DT, tests performed at different strain rates and relaxation tests performed at different strain levels are required. When applied to asphalt binder data, the testing procedure found no departure from viscoelastic conditions for the DSR and BBR test data. However, the DT procedure indicated a departure from linear viscoelastic conditions.

Low temperature fracture properties of polymer-modified asphalts relationships with the morphology

A methodology for studying the relationships between fracture behavior and morphology of polymer-modified asphalts used as binders was developed by using the linear elastic fracture mechanics (LEFM) method and confocal laser scanning and environmental and cryo-scanning electron microscopies. Different types of polymers were used as modifiers: (i) copolymers from ethylene and methyl acrylate (EMA), butyl acrylate (EBA) or, vinyl acetate (EVA); (ii) diblock or star-shape triblock styrene-butadiene copolymers (SB or SBS*). The 4 to 6 wt. % blends display an heterogeneous structure with a polymer-rich dispersed phase based on the initial polymer swollen by the aromatic fractions of the asphalt. The fracture toughness of the blends is higher than for the neat asphalt even if KIc of blends remains low compared to usual polymer blends due to the brittleness of the asphalt matrix. The fracture behavior which is strongly dependent on the nature of the polymer is discussed from the toughening mechanisms given for the filled polymers and the polymer blends. The EBA, SB, and SBS-based blends compared to the EMA and EVA-based ones display a higher KIc due to the elastomeric behavior of the polymer phase leading to a more efficient energy dissipation during crack propagation. The sample prepared with 4% crosslinked SB (Styrelf) and the corresponding physical blend (non-crosslinked) display the better fracture properties.

New Test Protocol to Measure Fatigue Damage in Asphalt Mixtures

ABSTRACT The measurement, analysis and interpretation of hot-mix asphalt concrete fatigue tests have been a subject of research for many years. Standard flexural tests using ectangular or trapezoidal beams are non-homogeneous in that the stress-strain fields and the temperature distribution within the test specimens are unknown, making a rigorous analysis of the phenomena that occur during these tests very difficult. A new protocol for fatigue testing is proposed using a uniaxial “push-pull” test with three loading stages on a cylindrical specimen. The first and last stages are at very low loading levels, below the endurance limit of the mixture being tested. Fatigue damage occurs during the second stage of loading when the applied stress or strain is sufficient to cause fatigue damage. The difference between the moduli at the end of the first and at the end of the third stage of loading is caused by fatigue alone. The presence of an endurance limit is verified and it is shown that the drop in modulus during a standard fatigue test is primarily caused by nonfatigue phenomena. The rate of fatigue damage obtained with the proposed protocol is approximately one tenth of that predicted by standard tests. The effects of non-fatigue phenomena such as self heating, self cooling, and nonlinearity are discussed, as well as a phenomenon that the authors, for want of a better term, refer to as thixotropy. The authors conclude that the major portion of the modulus lost during a fatigue test, as well as modulus gained during periods of rest or decrease of loading, is the result of thixotropy.

Rheological Properties of Mineral Filler-Asphalt Mastics and Its Importance to Pavement Performance

The rheological and failure properties of eight asphalt-filler mastics were characterized using new testing techniques being developed within the SHRP program for testing asphalt binders. The mastics were prepared using four SHRP core asphalts and two fillers (calcite and quartz). The rheological measurements were used to construct rheological master curves and temperature shift functions. The failure properties were measured at different temperatures and strain rates and were shifted using time- temperature superposition to construct stress and strain-to- failure master curves. A subset of the mastics was also tested after oxidative aging with the pressure aging vessel (PAV) and after isothermal aging at low temperatures. The properties of the mastics were compared with the asphalt binder properties to describe the changes resulting from adding the fillers. The fillers were found to change the shape of the rheological master curves and to significantly increase the failure stress at all combinations of temperatures and loading times. The effects of mineral fillers upon relative stiffness, oxidative aging, hardening, low temperature cracking fatigue and rutting resistance are discussed. For the covering abstract see IRRD 858225.

Field Performance of Modified Asphalt Binders Evaluated with Superpave Test Methods: I-80 Test Project

In the early 1980s heavy-duty pavements in Pennsylvania showed evidence of excessive rutting. As a consequence, the Pennsylvania Department of Transportation adopted several changes in its materials specifications and mixture design procedures. In addition, a number of modified binders were evaluated in an experimental test road that was constructed in 1989 in Clearfield County on Interstate 80. The construction was a 175-mm thick asphalt concrete overlay over an existing portland cement concrete pavement. Although the construction predated Superpave, original samples of the asphalt binder and loose asphalt mix were retained and were characterized using Superpave test methods. Field performance evaluations were performed immediately after construction and in subsequent years, giving a record of rutting, cracking, raveling, and overall visual performance. Overall, the mixtures have performed well during their 9 years of service. However, differences in the performance of the mixtures with the different binders are evident. These differences are related to the properties of the binder and the properties of the mixture as measured with the Superpave mixture and binder tests.

Fracture mechanics and asphalt binders

ABSTRACT In this paper, the applicability of linear elastic, elastic plastic, and time-dependent (viscoelastic) fracture theories to asphalt binders at lower service temperatures is examined. Linear elastic fracture mechanics was found to be applicable only at temperatures below the glass transition temperature. Above the glass transition temperature, neither elastic nor elastic plastic fracture mechanics are applicable to asphalt binders because in this temperature regime they behave as time dependent viscoelastic materials and time-dependent fracture mechanics models must be used. Fracture parameters from each of the theories provide a different low temperature ranking of asphalt binders than the PG specification used in the United States. The authors conclude that, pending further development, the direct use of fracture mechanics parameters in a purchase specification is not realistic.