AAPT
Abstracts for the March 18-20, 2002 Annual Meeting

LINEAR VISCOELASTIC LIMITS OF BITUMINOUS BINDERS

by Gordon D. Airey, Behzad Rahimzadeh, and Andrew C. Collop

Abstract

In order to predict the engineering performance of a material, it is necessary to understand its stress-strain behaviour. This is usually accomplished by means of laboratory tests under controlled stress or controlled strain conditions. As bitumen is a viscoelastic material, its performance must be characterised with test methods and analytical techniques that account for time (or rate) of loading and temperature. In addition, it is usually advisable to confine the characterisation of a bitumen to its linear viscoelastic response (small strains) to simplify the mathematical modelling of the material, as non-linear response, particularly for viscoelastic materials, is extremely difficult to characterise in the laboratory and model in practical engineering problems.
In general, conventional bituminous binders possess a relatively wide linear range. However, concern has been expressed over the linear range for modified bitumens with the assumption that highly modified bitumens, particularly high polymer content PMBs, will show a narrower linear range. For this reason a series of stress/strain sweeps have been performed on a range of process and polymer modified bitumens using a dynamic shear rheometer. Using these stress/strain sweeps, the linearity limits for the binders have then been determined as a function of a percentage reduction in initial complex modulus and a series of LVE stress and strain limits versus complex modulus and phase angle have been produced.
The linearity results showed that for sensibly engineered modified binders (softer base bitumens with higher modifier contents) there is no significant narrowing of the linear range. The process modified, plastomeric and conventional binders showed a strain dependent LVE criteria between 2% and 6% at low temperatures (high stiffness and intermediate to low phase angles) as well as a stress dependent LVE criteria between 1.5 and 7 kPa at high temperatures (low stiffness and high phase angles). The thermoplastic rubbers (SBS PMBs) showed, in addition to the low temperature strain criteria, a high temperature (low stiffness) polymeric-based strain dependent LVE criteria between 50% and 200%. The strain dependent LVE criteria for the SBS PMBs at high temperatures did translate to a narrowing of the linear range and a reduction in linearity limit over the range were the elastomeric polymer was dominant.

VALIDATION OF THE SUPERPAVE ASPHALT BINDER FATIGUE CRACKING PARAMETER USING THE FHWA'S ACCELERATED LOADING FACILITY

by Kevin D. Stuart and Walaa S. Mogawer, Ph.D., P.E.

Abstract

An Accelerated Loading Facility (ALF) was used to evaluate the Superpave asphalt binder parameter for intermediate-temperature (fatigue cracking) performance, namely, G*sin delta. The ALF is a full-scale, pavement-testing machine that applies one half of a single rear truck axle load.
The variables evaluated were Performance Grade (PG), asphalt pavement layer thickness, and temperature. Two unmodified asphalt binders were used. These binders had PG's of 58-34 and 64-22, and continuous intermediate-temperature PG's of 9 and 17, respectively. Both asphalt binders were from the same crude source. Two thicknesses were tested: 100 and 200 mm. Each pavement was tested at three temperatures: 28, 19, and 10 C. Therefore, 12 pavement sites were tested by the ALF.
As expected, each 100-mm-thick pavement failed sooner than the 200-mm-thick pavement having the same asphalt binder and test temperature. The pavements failed sooner at the middle test temperature of 19 C compare to 10 C and 28 C for a given asphalt binder and asphalt pavement layer thickness. Neither the strain-control asphalt binder parameter, G*sin delta, nor the stress-control asphalt binder parameter, G*/sin delta, can by itself explain the effect of temperature. Only one of six comparisons used to evaluate the effect of PG on fatigue cracking performance showed that the mode of loading was strain control, which is the mode of loading used by the Superpave asphalt binder specification for intermediate-temperature performance. The pavement data also showed that the mode of loading can change with temperature.

ACID/BASE CHEMISTRY FOR ASPHALT MODIFICATION

by Gayle King, S.W. Bishara, and Glenn Fager2

Abstract

The advent of SUPERPAVE heightened the need for asphalts with an extended range between the temperature at which the asphalt has sufficient modulus to resist rutting (Th) and the temperature limit for critical stiffness to prevent thermal cracking (Tl). Acidic and basic modifiers have been shown to improve asphalt stiffness, elevating Th without unduly harming Tl. The low cost of these materials relative to polymers and other modifiers that improve asphalt temperature susceptibility makes their use compelling in a competitive environment focused upon manufacturing Performance-Graded (PG) asphalt binders.
Amine antistripping agents (ASA) have long been added to asphalt mixtures to reduce deleterious effects of moisture. The antistripping activity of the amine is typically attributed to the nitrogen lone pair of electrons, which show strong affinity for the aggregate surface in the presence of water. What happens, then, when such basic compounds are added to mixtures containing acid-modified asphalt? Fundamental laws of chemistry would dictate that the acid modifier and the basic ASA may react with each other to form an amine salt. This raises several questions:
Will Th drop if the acid modifier is neutralized?
Can the ASA continue to resist moisture if the active electron pairs have been blocked by acidic protons (H+)?
Will the product salt contribute to moisture damage in the pavement, either by leaving the asphalt film more permeable to water, or by emulsifying the asphalt in the presence of heat, moisture, and traffic?
To address these questions, a research study was initiated by the Kansas Department of Transportation (KDOT). Phase 1 was a binder study, with DSR measurements of Th on asphalt binders modified with varying concentrations of phosphoric acid (H3PO4), two different modified fatty imidazoline antistripping agents, and blends of acid and imidazoline. These same binders were then dissolved in toluene and extracted with water to determine extracted pH and to monitor for possible emulsification of asphalt into the aqueous phase. Phase 2 was a mixture stripping study. A moisture sensitive aggregate was blended with PG 58-22 that had been modified with various acid/base combinations. The resulting mixtures were evaluated for moisture damage. Hamburg and Asphalt Pavement Analyzer wheel-tracking tests were run alongside KDOT's retained tensile strength test (KT-56), since it was theorized that mechanical loading might be necessary to simulate damage from in-situ asphalt emulsification.
Key findings:
Phosphoric acid additives significantly increase asphalt stiffness at high temperatures. Moisture damage may occur if acid is used at elevated concentrations.
For the mix studied here, ASAs consistently reduced moisture damage in both KT-56 and wheel-tracking tests.
Phosphoric acid reacts with ASA. When the two additives are used in the same mix, the binder Th is substantially reduced, the ASA activity is blocked, and organic salts form. KT-56 and wheel-tracking tests predict significant moisture damage.
Binder Th should be specified or confirmed following amine addition when a mix requires ASA.Wheel-tracking tests provide valuable information on the stripping potential of mixes that contain acidic and/or basic modifiers.

USE OF DYNAMIC MECHANICAL ANALYSIS (DMA) TO EVALUATE THE FATIGUE AND HEALING POTENTIAL OF ASPHALT BINDERS IN SAND ASPHALT MIXTURES

by Yong-Rak Kim, Dallas N. Little, and Robert R. Lytton

Abstract

Fatigue of asphalt mixtures is a complex subject. In phenomenological fatigue models, the shift factor between laboratory fatigue tests and field data is huge. A significant portion of the shift factor is due to healing of microdamage, and the major contributor to healing is the asphalt cement and/or mastic component of the mixture. The DMA offers an efficient and reliable means of assessing both the fatigue and healing of binders and mastics. In this study cylindrical, sand asphalt samples are tested in the DMA. Several candidate fatigue failure parameters are considered to assess fatigue life and the effect of rest periods on the extension of fatigue life through healing. The effects of rubber modification and aging are also evaluated. The DMA analysis not only allows a repeatable and accurate assessment of healing but also provides insight into the mechanism of fatigue damage.

RELATIONSHIP BETWEEN SUPERPAVE GYRATORY COMPACTION PROPERTIES AND THE RUTTING POTENTIAL OF ASPHALT MIXTURES

by R. Michael Anderson

Abstract

The NCHRP 9-16 project, Relationship Between Superpave Gyratory Compaction Properties and Permanent Deformation of Pavements in Service, was initiated to determine if there was any connection between properties determined during the Superpave gyratory compaction (SGC) process and the rutting resistance of an asphalt mixture. Some SHRP and post-SHRP research indicated that properties of the SGC densification curve could potentially be related to the expected rutting resistance of an asphalt mixture. Confirmation of the permanent deformation resistance of Superpave mixtures during the mix design and quality control processes designs is desirable. As a result, state agencies and the asphalt industry are considering the use of an additional test (or tests) during the mix design process. If research indicates that the SGC can be used to estimate the rutting resistance of asphalt mixtures, or identify gross mixture instability, then the need for an additional performance testing may be lessened.
Evaluation of the results from the NCHRP 9-16 laboratory experiment indicated that the best compaction parameter related to asphalt mixture shear stiffness and rutting potential appeared to be N-SRmax as determined using a modified Pine AFG1 SGC. The N-SRmax parameter is the number of gyrations at which the stress ratio (shear stress divided by vertical stress) reaches a maximum value.
The N-SRmax parameter appears to lend itself easily to mix design or quality control testing. Threshold values were identified separating mixtures with good and poor expected performance. Unlike other parameters, the N-SRmax value for an asphalt mixture specimen can be determined immediately following the compaction process. Initial laboratory mix design validation of two mixes confirms the utility of the N-SRmax parameter as a mixture performance-screening tool.
Based on the findings of the experiment, it appears that the N-SRmax determined during SGC compaction provides a general indication of expected mixture rutting performance.

STRESS DEPENDENT MASTER CURVE CONSTRUCTION FOR DYNAMIC (COMPLEX) MODULUS

by Terhi K. Pellinen and Matthew W. Witczak

Abstract

This paper presents a method of constructing stress dependent asphalt mix master curve using compressive dynamic (complex) modulus test data. The presented method uses a k1-k3 non-linear elasticity model to incorporate stress dependency to the master curve. The stress dependency is included in the equilibrium modulus value of the master curve, instead of incorporating it into the shift factors. The stress dependency and non-linearity are considered only at intermediate and high temperatures and the model assumes a linear visco-elastic material behavior at cold temperatures. As a practical implementation of the method, a Stress Dependent Dynamic Modulus Predictive Equation was developed analogous to the Witczak at al. Dynamic Modulus Predictive Equation.

AGGREGATE GRADATION CHARACTERIZATION FACTORS AND THEIR RELATIONSHIPS TO FRACTURE ENERGY AND FAILURE STRAIN OF ASPHALT MIXTURES

by Byron E. Ruth, Reynaldo Roque, and Bensa Nukunya

Abstract

Ten limestone mixtures with different gradations were designed according to SuperPave procedures and criteria. These mixtures were compacted to 93 percent of maximum theoretical density to prepare specimens for indirect tension testing. Tensile strength, fracture energy and failure strain parameters were evaluated for test specimens subjected to short and long term oven aging. Aggregate gradation characterization factors were used in multiple linear regression analyses to establish relationships with the fracture energy and failure strain parameters. Power law constants (aCA, aFA ) and exponents (nCA, nFA ) for the coarse and fine aggregate portions of these mixtures were established by regression analyses.
The results of these analyses identified gradation characteristics that are detrimental to mixture properties. Specifically, gap-graded or gradations with an excess amount of aggregate retained on a sieve did not yield properties equivalent to well balanced, continuously graded, aggregate blends. Greater asphalt content and percent passing the 4.75 mm sieve resulted in greater tensile strength and fracture energy (FE) for coarse graded mixtures and lower FE for fine graded, long term aged, mixtures. The failure strain of fine graded mixture improved with increase in asphalt content and percent passing the 4.75 mm sieve.
The implications of the findings from this investigation are that the gradation characterization factors relate well to mixture properties and potentially can provide guidance in the selection of aggregate gradations for improved pavement performance.

HIGH TEMPERATURE FATIGUE AND FATIGUE DAMAGE PROCESS OF AGGREGATE-ASPHALT MIXES

by Bor-Wen Tsai, John T. Harvey, and Carl L. Monismith

Abstract

The term, "High Temperature Fatigue" has long been identified in mechanical engineering as an important phenomenon that limits the life of metallic components. Asphalt pavement routinely encounters high temperatures in several California climate regions, yet there has been little study of this phenomenon. In this paper, the effect on fatigue performance of high temperatures (30-40 degrees C) has been studied by placing an aluminum plate under the conventional four-point fatigue test setup and bending the beam upward so as to prevent temperature-induced creep during the test.
For crack initiation, the Weibull stochastic process was found to successfully describe the fatigue damage process, a process of stiffness deterioration, under any temperature or test conditions. The Weibull proportional hazards model of survival analysis was used as a tool to integrate tests performed under various conditions of strain, temperature, and material variables, including the high temperature fatigue tests and conventional fatigue tests, into a single regression equation. This equation was used as a tool to assess the fatigue damage. A dynamic and recursive fatigue simulation program was developed using the Weibull approach and a layered-elastic theory subroutine to calculate the tensile strain. The Weibull accelerated failure time model was also used to estimate a correction factor to account for the inherent difference of boundary conditions between laboratory test results and in-situ fatigue performance.
Weibull dynamic approach as applied to field fatigue performance prediction: To verify the Weibull dynamic approach, the Federal Highway Administration (FHWA) sponsored the WesTrack full-scale accelerated pavement testing project provided a complete and valuable data set containing not only information on various mixes, traffic and wander, pavement temperatures, and laboratory test results, but also the condition survey of fatigue performance.
The computer simulation results were compared with the WesTrack condition survey results to validate the feasibility of the Weibull dynamic approach as applied to field fatigue performance prediction.

FIELD SHEAR TEST DEVICE FOR QUALITY CONTROL TESTING OF ASPHALT CONCRETE

by Donald W. Christensen, Jr., Ramon F. Bonaquist, and Titin Handojo

Abstract

The field shear test (FST) was originally developed during NCHRP Project 9-7 as a simple and inexpensive alternative to the Superpave shear test (SST), meant primarily for use in quality control/quality assurance (QC/QA) testing. The first FST prototype, however, did not perform as well as hoped. The primary objectives of NCHRP 9-18 were to refine the FST and evaluate its sensitivity to changes in asphalt concrete composition. The redesigned FST is very compact, rugged and extremely easy to operate. Although capable of performing a variety of tests, best results are obtained for frequency sweep data at high temperatures and frequencies of 5 to 10 Hz. Under these conditions, the FST produces dynamic modulus values about 40% higher and phase angles 5 to 15 degrees lower than those generated using the SST. FST dynamic modulus data were found to be fairly sensitive to changes in asphalt concrete composition. Data repeatability is similar for the FST and other methods for measuring asphalt concrete stiffness, including the SST. Future research efforts should initially focus on reducing the variability inherent in FST data and other asphalt concrete stiffness test results, followed by several field demonstration projects.

INVESTIGATION OF THE RESTRICTED ZONE IN THE SUPERPAVE AGGREGATE GRADATION SPECIFICATION

by Prithvi S. Kandhal and L. Allen Cooley Jr.

Abstract

The recommended aggregate specification for Superpave hot mix asphalt (HMA) mixtures includes a restricted zone which lies along the maximum density gradation between the intermediate size (either 4.75 or 2.36 mm, depending on the nominal maximum size of the aggregate) and the 300 æm size. The restricted zone forms a band through which gradations were recommended not to pass. The restricted zone requirement was adopted in Superpave to help reduce the incidence of tender or rut-prone HMA mixes.
According to many asphalt paving technologists, compliance with the restricted zone criteria may not be desirable or necessary to produce paving mixes that give good performance. Some highway agencies and suppliers can provide examples of aggregate gradations that pass through the restricted zone, but produce paving mixes that have historically performed well.
This research project was undertaken to evaluate the effect of the Superpave restricted zone on permanent deformation of dense-graded HMA mixtures. Its primary objective was to determine under what conditions, if any, compliance with the restricted zone requirement is necessary when all the other Superpave requirements such as fine aggregate angularity (FAA) and volumetric mix criteria for the specific project.
The following factors were evaluated: two coarse aggregates, ten fine aggregates, two nominal maximum aggregate size mixes (9.5 and 19.0 mm), five aggregate gradations, and three compactive efforts (Ndesign=75, 100, and 125). Of the five gradations used, three violate the restricted zone and two fall outside of the restricted zone (control). Permanent deformation characteristics of mixes meeting Superpave volumetric requirements were evaluated by two different types of tests: empirical and fundamental. For the empirical test, the Asphalt Pavement Analyzer was used. The Superpave shear tester and a repeated load confined creep test were utilized as fundamental tests. Test results from the three mechanical tests were analyzed statistically to evaluate the effect of the five gradations on permanent deformation of the HMA mixtures. Based upon the analysis of data, mixes having gradations passing through the restricted zone did not necessarily have lower rut resistance compared to mixes having gradations outside the restricted zone. It was recommended to delete the restricted zone as a guideline or requirement in Superpave mix design.

KEY WORDS: Superpave, asphalt mixtures, HMA, asphalt concrete, gradation, restricted zone, permanent deformation, rutting

INTRODUCTION

The Strategic Highway Research Program's (SHRP) asphalt research was primarily aimed at the properties of asphalt binders and HMA mixes and their effect on asphalt pavement performance. The study of aggregate properties (including gradation) was intentionally excluded from the asphalt research program. Yet, the SHRP researchers had to recommend a set of aggregate properties and an aggregate gradation specification without the benefit of experimentation so that a comprehensive Superpave mix design system could be formulated.
SHRP formed an Aggregate Expert Task Group (ETG) consisting of 14 acknowledged experts in the area of aggregates. In lieu of a formal aggregate research program, the aggregate ETG used a modified Delphi approach to develop a set of recommended aggregate properties and criteria that are now included in the Superpave volumetric mix design method (AASHTO MP2 and PP28). The Delphi process was conducted with five rounds of questionnaires. The final recommended aggregate gradation criteria included control points between which the gradation must fall as well as a restricted zone that lies along the maximum density line (MDL) between the intermediate size (either 4.75 or 2.36 mm depending on the nominal maximum size of the aggregate in the mix) and the 0.3 mm size.
Although the restricted zone was included in Superpave as a recommended guideline and not a required specification, some highway agencies interpreted it as a requirement. Many asphalt technologists believe that compliance with the restricted zone criteria may not be desirable or necessary in every case to produce asphalt mixes with good performance. If highly angular aggregates are used in the mix it is likely that the mix will not exhibit any tenderness during construction and will be rut-resistant under traffic regardless if its gradation passes through the restricted zone. The Georgia Department of Transportation has used such mixes successfully for many years. Some asphalt technologists also question the need for the restricted zone when the mix has to meet volumetric properties such as minimum voids in the mineral aggregate (VMA) and specified air void contents at Ninitial, Ndesign, and Nmaximum gyrations.

EFFECTS OF LOADING CONFIGURATION AND MATERIAL PROPERTIES ON NON-LINEAR RESPONSE OF ASPHALT MIXTURES

by Eyad Masad and Hussain Bahia

Abstract

Hot mix asphalt (HMA) stiffness is used in pavement engineering to evaluate the relative quality of mixtures and to predict the response of pavements subjected to wheel loads. It is often described using parameters such as the resilient, dynamic, and relaxation moduli. The dynamic modulus is currently under consideration as a parameter to assess the mixture resistance to permanent deformation as part of the Superpave volumetric mixture design procedure. The dynamic modulus can be measured under shear or axial loading. There has been discussion over the years about the correlation between the shear and axial moduli and the nonlinear behavior of asphalt mix stiffness at different strain levels in the shear and axial tests.
The purpose of this paper is to study the influence of loading configuration, aggregate distribution and binder nonlinear viscoelastic properties on mix stiffness. Experimental measurements of binder stiffness at different strain and frequency levels, and HMA stiffness under shear and axial loading are analyzed and presented. The results show that the binder nonlinearity, loading configuration and the aggregate structure are major factors influencing the HMA nonlinearity. The binder nonlinearity follows different trends than the mix nonlinearity due to the influence of aggregates. The loading configurations in the shear and axial tests are analyzed in terms of the rotation of principal stresses during the test, and the direction of the major principal stress with respect to aggregate orientation. The results offer a possible explanation for the discrepancies between the axial and shear moduli measured on the same HMA.

DEVELOPMENT OF A THERMO-VISCOPLASTIC CONSTITUTIVE MODEL FOR HMA MIXTURES

by B. Huang, L. Mohamad and W. Wathugala

Abstract

Hot mix asphalt (HMA) mixtures exhibit temperature and rate dependencies under monotonic and cyclic loads. Most current constitutive models used in design and prediction of the performance of asphalt pavements do not incorporate these characteristics into the equations. These limitations cause serious problems in practical applications.
This paper presents the development of a thermo-visco-plastic constitutive model that incorporated the temperature and loading rate into the Hierarchical Single Surface (HiSS) plasticity based model. This model was able to reflect the non-linear plasticity, as well as the temperature and loading rate dependencies of the asphalt mixtures. A series of triaxial and creep tests at three temperatures were performed to calibrate the material properties. A back-predicting algorithm for the HiSS thermo-visco-plastic model was developed during this study. A comparison of numerical analysis obtained from the proposed model and experimental results was conducted based on this algorithm. A good agreement was observed from the back- calculation and the experimental results. Further studies of the applicability of this model in predicting pavement performance were recommended.

DEVELOPMENT OF A SIMPLIFIED FATIGUE TEST AND ANALYSIS PROCEDURE USING A VISCOELASTIC, CONTINUUM DAMAGE MODEL

by Jo Sias Daniel and Y. Richard Kim

Abstract

This paper presents a methodology by which the material response under various uniaxial tensile testing conditions (type of loading and temperature) can be predicted from the material response obtained from a single testing condition. The methodology makes use of a uniaxial constitutive model for asphalt concrete that is based upon the elastic-viscoelastic correspondence principle and work potential theory, a continuum damage theory based on the thermodynamics of irreversible processes. Uniaxial tensile testing is performed under a controlled crosshead mode for both cyclic and constant rate to failure tests. Various strain amplitudes, frequencies, and rates are applied at several test temperatures. A single characteristic curve can be found that describes the reduction in material integrity as damage grows in the specimen, regardless of the applied loading conditions (cyclic versus monotonic, amplitude/rate, frequency). The characteristic curve at any temperature below 20C can be found by utilizing the time-temperature superposition principle and the concept of reduced time. In this study, eight WesTrack mixtures were tested and the methodology was applied to successfully predict the fatigue damage at different testing conditions from a single condition. A test and analysis procedure for the fatigue characterization of asphalt mixtures based on this methodology is proposed and potential applications are discussed.

ISSUES PERTAINING TO THE PERMEABILITY CHARACTERISTICS OF COARSE- GRADED SUPERPAVE MIXES

by L. Allen Cooley, Jr., Brian D. Prowell, and E. Ray Brown

Abstract

In order to evaluate the relationships between in-place air voids, lift thickness, and permeability, 23 on-going HMA construction projects were visited and field permeability tests conducted. Field permeability tests were conducted at 15 randomly determined sites for each project. Cores were taken at each of the 15 locations to determine pavement density using AASHTO T166. In addition, for some of the projects, the cores were tested with the Corelok device and a laboratory permeameter.
As agencies begin to include permeability specifications, mix designers need tools they can use during the mix design process to evaluate the permeability characteristics of a given aggregate structure. Two techniques were evaluated: laboratory permeability measurements on samples compacted using the Superpave gyratory compactor and water absorption determined with AASHTO T 166 or the Corelok device.
Results of testing within this study indicated a good relationship between permeability (measured in the field and lab) and pavement density. Both the gradation's nominal maximum aggregate size (NMAS) and the lift thickness placed in the field were shown to affect the permeability-density relationship. Increasing the NMAS requires higher densities to ensure an impermeable pavement. Also, as the lift thickness of a given pavement (and mixture) increases, permeability decreases at a given density level.
Some reasonable relationships were found between the permeability of samples compacted using the gyratory compactor and field samples. Reasonable relationships were found between permeability and water absorption regardless of nominal maximum aggregate size.

PROPOSED METHODOLOGY FOR PREDICTING HMA PERMEABILITY

by Aslam Al-Omari, Laith Tashman, Eyad Masad, Allen Cooley Jr., and Thomas Harman

Abstract

Permeability is an important property that influences the performance of hot mix asphalt (HMA) mixes. It is a function of compaction effort, and several properties of HMA constituents such as asphalt content, size of aggregates and shape of aggregates. All these factors manifest themselves in different air void distributions, which in turn control fluid flow and permeability of asphalt mixes. The current practice is to relate measured permeability to the percent in-place air voids of HMA mixes. This paper relates HMA mix permeability to the three dimensional distribution of air voids. An innovative approach is developed to quantify air void connectivity, flow paths irregularity (Tortuosity), effective percent air voids, and specific surface area of air voids. This was made possible through X-ray computed tomography to capture the three-dimensional internal structure of HMA mixes and developing imaging techniques to analyze fluid flow paths.
The developed methods were used to find the components of a modified expression for the Kozeny-Carman equation to predict HMA permeability depending on air void distribution properties only. The predicted permeability values had reasonable correlation with laboratory measurements. The permeability equation was further simplified to allow predicting permeability based on percent total air voids and aggregate gradation. This simplified equation was shown to have good correlation with laboratory and field measurements of permeability for HMA mixes with a wide range of properties.

USE OF SURFACE FREE ENERGY PROPERTIES OF THE ASPHALT- AGGREGATE SYSTEM TO PREDICT DAMAGE POTENTIAL

by DingXin Cheng, Dallas N. Little, Robert L. Lytton, and James C. Holste, Professor

Abstract

Moisture damage analysis based on surface free energy theory is presented in this paper. Two methods, the universal gas adsorption and the Wilhelmy plate, are used to measure surface free energies of aggregate and asphalt, respectively. The universal gas adsorption method utilizes the character of the adsorption of a particular chemical gas solvent onto the surface of an aggregate to indirectly determine the surface energy of the aggregate. This method can accommodate the peculiarity of the irregular shape, size, mineralogy, and rough surface texture of the aggregate. The Wilhelmy plate method measures both the surface free energy of wetting and the surface free energy of dewetting from the advancing angles and receding angles, respectively. Calculation of surface free energy of adhesion using this approach can help select more compatible asphalt-aggregate combinations for asphalt mixtures. More compatible combinations can improve the adhesive bond and reduce debonding potential in the presence of moisture, stripping.

TERTIARY FLOW CHARACTERISTICS OF ASPHALT MIXTURES

by K.E. Kaloush and M.W. Witczak

Abstract

In the past two years, major research work was conducted to recommend a fundamentally based laboratory Simple Performance Test (SPT) for permanent deformation evaluation to be used within the Superpave volumetric mixture design procedure. This SPT research was evaluated using mixtures and performance data from three major experimental sites. Five laboratory tests had parameters that resulted in an overall good to excellent correlation to the measured rut depths. The test methods and responses were ranked according to their advantages and disadvantages, and best correlation obtained to field rutting in the following order: 1) the dynamic modulus as measured triaxial compression test at high temperatures, 2) the flow time (tertiary flow) as measured by the triaxial creep test, 3 and 4) the flow number of repetitions (tertiary flow) and the cumulative permanent strain as measured by the triaxial repeated load test, and 5) the permanent shear strain measured at 1,000 loading cycles using the SST repeated shear load at constant height test.
This paper discuses the two tertiary flow parameters, which stood out in the SPT study as having excellent correlation with field rut depth. Several failure analysis concepts were introduced and verified throughout the testing program. These concepts included failure strains, plastic strain ratio, volume change, constant compliance at failure, and plastic to resilient strain ratio at failure as a design guide parameter. The results in both tests indicated that tertiary flow could be captured from either axial or radial strain measurements. This was a significant finding related to the simplification of testing equipment and sample instrumentation that could be realized with the use of radial measurements only. Furthermore, the rate of permanent deformation obtained from the confined repeated load tests best simulated the rate of deformation occurring in the field.

FOAMED ASPHALT PRODUCED USING WARMED AGGREGATES

by K.J. Jenkins, A.A.A. Molenaar, J.L.A. de Groot, and M.F.C. van de Ven

Abstract

Foamed asphalt mixes are produced through the mixing of foamed bitumen with aggregate under ambient conditions. However, the variations in the temperature of mineral aggregate at the time of mixing have a profound influence on the properties of the foamed asphalt produced. A feasibility study previously undertaken into the possible benefits of heating the aggregate moderately (above ambient temperature but below 100C) before foam treatment, highlighted the improvement of mix properties that is achievable. The properties that can be enhanced include particle coating, mix cohesion and tensile strength, as well as a degree of compaction when compared with the equivalent properties of conventional cold foam mixes. This has been found to be applicable to, in particular, reclaimed asphalt pavement (RAP) and densely graded crushed aggregates.
A second phase, more focussed research project was subsequently launched in the Netherlands using the mixes that show the most potential to benefit most from the "Half-warm Foamed Bitumen Treatment" process viz, STAB (Asphalt Concrete Gradation) and RAP (Reclaimed Asphalt Pavement) combinations. This paper discusses the findings of the entire investigation into half-warm foamed bitumen treatment, carried out in South Africa and the Netherlands. The three "Half-warm Foamed Asphalt Mixes" investigated in Delft are compared with the equivalent HMA (Hot Mix Asphalt) and Cold Foamed Mix materials. Initially the manufacturing technique is described in the paper. This is followed by simple characterisation of different foamed mixes with different percentages of RAP has been carried out using two monotonic tests i.e. Unconfined Compression Tests and Leutner Shear Testing (on cylindrical specimens), at different temperatures and displacement rates. Finally, fatigue and tri-axial testing was carried out on selected mixes in order to investigate comparative performance characteristics.
With up to 40% less energy consumption during manufacture than HMA, Half-warm Foamed Asphalt Mixes can provide comparative monotonic properties at higher test temperatures, similar fatigue properties and a lower phase angle at low loading frequencies. The improvement of mix properties relative to Cold Foamed Mix is significant. This indicates that the half-warm foam process holds the potential for successful implementation in pavement layers.

TOP-DOWN CRACK PROPAGATION IN BITUMINOUS PAVEMENTS AND IMPLICATIONS FOR PAVEMENT MANAGEMENT

by Leslie Ann Myers and Reynaldo Roque

Abstract

Surface-initiated longitudinal wheel path cracking has been cited as a widespread mode of failure among asphalt concrete pavements and yet, few theories have been presented that fully explain the mechanism. This study serves as complementary follow-up to the research presented on initiation of top-down cracks, as it explores and explains the mechanisms for propagation of top-down cracks and attempts to forecast the implications for pavement design. It may be concluded that designing and analyzing pavements using the current approach that employs averaged pavement conditions and ESALs will neither predict nor address longitudinal wheel path cracks that initiate and grow down from the surface. Research was performed that showed that existing design and evaluation methods that use averaged conditions are inadequate for predicting this type of surface cracking. It was proven that the propagation of surface-initiated longitudinal cracks advances only under critical conditions and that crack development is highly dependent differential pavement temperature gradients. This factor is addressed in the conventional approach to pavement analysis; therefore, an alternative approach to designing and rehabilitating pavements should be formulated, or the existing approach adjusted, to include surface-initiated longitudinal wheel path cracks. A concept formulated to identify crack length stages in terms of time, in order to better describe crack growth, may be the initial step in addressing top-down cracking using pavement management.

SUPERPAVE FOR LOW VOLUME ROADS AND BASE MIXTURES

by Brian D. Prowell, P.E. and John E. Haddock, Ph.D., P.E.

Abstract

This study evaluated theVirginia Department of Transportation's Marshall designed Subdivision surface mixture, SM-1, and base mixture, BM-2 using Superpave criteria. Twenty- four production samples were collected of 4 SM-1 mixture designs and 9 production samples of 8 BM-2 mixture designs. Samples were tested for binder content, gradation, Superpave and Marshall volumetric properties. When sufficient material was available, rut testing was performed on the SM-1 mixtures using the Asphalt Pavement Analyzer. Various SGC compaction levels were used, including stopping at both Ndesign and Nmax.
Superpave criteria for the SM-1 mixtures were generally found to be acceptable. The exceptions were the width of the gradation band and fines to effective binder content. A non- standard compaction level was recommended for SM-1 type mixtures to be compacted to Ndesign. Rut testing indicates that SM-1 mixtures are generally very rut resistant. This indicates VDOT could allow even lower compaction levels than those traditionally used for low volume Virginia roads in order to promote durability.
The BM-2 mixtures that were tested all fell on the fine side of the maximum density line for a Superpave 25.0 mm (1 in) nominal maximum size mixture. Compaction levels for mixtures with less than 1 million equivalent single axle loads were evaluated for VDOT's base mixture. Even at these low compaction levels, a decrease in design binder content is anticipated. Historically, these mixtures have suffered from moisture damage and not rutting.

TIME-TEMPERATURE SUPERPOSITION PRINCIPLE FOR ASPHALT CONCRETE MIXTURES WITH GROWING DAMAGE IN TENSION STATE

by Ghassan R. Chehab, Y. Richard Kim, Richard A. Schapery, Matthew W. Witczak and Ramon Bonaquist

Abstract

It is known that asphalt concrete in its linear viscoelastic range is thermorheologically simple. This paper presents the experimental/analytical research that demonstrates the validity of the principle of time-temperature superposition even with growing damage and viscoplastic straining by conducting constant crosshead rate tests on specimens that were pulled apart in tension until failure. Direct benefits and applications from this finding include reduction in testing program conditions, development of strength and corresponding strain mastercurves as a function of reduced time, a methodology for the prediction of stress-strain curves for any given crosshead strain test, and simplification of thermorheological analysis of pavement structures.