This article was downloaded by: [Universitaets und Landesbibliothek] On: 30 April 2013, At: 21:20 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Road Materials and Pavement Design Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/trmp20 Relationship between bond energy and total work of fracture for asphalt binder-aggregate systems Jonathan Howson a , Eyad Masad a , Dallas Little a & Emad Kassem b a Zachry Department of Civil Engineering, Texas A&M University, College Station, Texas, USA b Texas Transportation Institute, College Station, Texas, USA Published online: 01 May 2012. To cite this article: Jonathan Howson , Eyad Masad , Dallas Little & Emad Kassem (2012): Relationship between bond energy and total work of fracture for asphalt binder-aggregate systems, Road Materials and Pavement Design, 13:sup1, 281-303 To link to this article: http://dx.doi.org/10.1080/14680629.2012.657094 PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.
Transcript
This article was downloaded by: [Universitaets und Landesbibliothek]On: 30 April 2013, At: 21:20Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Road Materials and Pavement DesignPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/trmp20
Relationship between bond energyand total work of fracture for asphaltbinder-aggregate systemsJonathan Howson a , Eyad Masad a , Dallas Little a & Emad Kassemb
a Zachry Department of Civil Engineering, Texas A&M University,College Station, Texas, USAb Texas Transportation Institute, College Station, Texas, USAPublished online: 01 May 2012.
To cite this article: Jonathan Howson , Eyad Masad , Dallas Little & Emad Kassem (2012):Relationship between bond energy and total work of fracture for asphalt binder-aggregate systems,Road Materials and Pavement Design, 13:sup1, 281-303
To link to this article: http://dx.doi.org/10.1080/14680629.2012.657094
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representationthat the contents will be complete or accurate or up to date. The accuracy of anyinstructions, formulae, and drug doses should be independently verified with primarysources. The publisher shall not be liable for any loss, actions, claims, proceedings,demand, or costs or damages whatsoever or howsoever caused arising directly orindirectly in connection with or arising out of the use of this material.
Road Materials and Pavement DesignVol. 13, No. S1, June 2012, 281–303
Relationship between bond energy and total work of fracture for asphaltbinder-aggregate systems
Jonathan Howsona*, Eyad Masada, Dallas Littlea and Emad Kassemb
aZachry Department of Civil Engineering, Texas A&M University, College Station, Texas, USA;bTexas Transportation Institute, College Station, Texas, USA
Surface free energy is a thermodynamic material property representing the work required tocreate new surfaces of unit area in a vacuum. Surface free energy has been used to quantifyand screen both the cohesive bond energy of asphalt binders and the adhesive bond energy ofasphalt binder–aggregate interfaces in wet and dry conditions. The bond energy is computedbased on the surface free energies of the constituent materials. The total work of fracture isthe cumulative effect of energies applied to the sample to create two new surfaces of unit area.These energies include the bond energy, calculated from surface free energy, dissipated plasticenergy, and dissipated viscoelastic energy. This paper presents experimental results from aseries of pull-off tests using asphalt binder-aggregate samples that demonstrate the relationshipbetween bond energy and total work of fracture. In order to fully explore this relationship,temperature, loading rate, specimen geometry, and moisture content were varied.
Keywords: total work of fracture; bond energy; surface free energy; pull-off test; aggregate;asphalt binder
1. IntroductionSurface free energy has been used in several fields of engineering for several decades to deter-mine the ability of one material to adhere to another. Only recently, however, has surface freeenergy been applied to the field of asphaltic materials and used by researchers to determinethe cohesive bond energy of asphalt binders and the adhesive bond energy of asphalt-aggregatecombinations under both dry conditions and in the presence of water (Cheng, Little, Lytton, &Holste, 2002; Bhasin, 2006; Bhasin, Masad, Little, & Lytton, 2006; Little, Bhasin, & Hefer,2006; Masad, Zollinger, Bulut, Little, & Lytton, 2006; Bhasin, Howson, Masad, Little, & Lytton,2007). Cheng, Little, Lytton, and Holste (2002) and Hefer, Bhasin, and Little (2006) developeda detailed methodology for using the Wilhelmy plate to measure the surface free energy compo-nents of asphalt binder, while Bhasin and Little (2007) developed the methodology for using theUniversal Sorption Device (USD) to measure the surface free energy components of aggregateparticles. Cheng et al. (2002) also called attention to the fact that aggregates have a much greateraffinity for water than for bitumen. Therefore, given enough time and access to the aggregatesurface, water will likely replace (strip) the asphalt binder from the surface of the aggregate.Several other studies also demonstrated a good correlation between parameters determined usingthe surface free energy components of asphalt binders and aggregates and the moisture sensitivityof asphalt mixtures (Bhasin et al., 2006; Little et al., 2006). These correlations were based on theperformance of asphalt mixtures in the field and in the laboratory.
Even though the cohesive and adhesive bond energy of asphalt binder and asphalt binder–aggregate combinations, respectively, can be obtained from the individual surface free energiesof the component materials, it has been observed that important differences exist between themagnitude of these values and the total work of fracture obtained from mechanical tests. Bondenergy, WB, is based on fundamental material properties and is independent of any external orexperimental factors; however, the total work of fracture, WT, depends on various experimentalfactors such as specimen geometry and loading conditions (e.g., peel test versus pull-off test,loading rate for time dependent materials, etc). As a consequence, it can be expected that differentexperimental tests, or test conditions, will provide different data regarding the total work of fracture(Mittal, 1978).
Many studies have evaluated the differences between bond energy and total work of fracture.The main finding was that WT can be much greater than WB depending on material properties,geometry of the specimen, adhesive thickness, temperature, and loading rate for time-dependentmaterials. However, it has also been shown that plastic (WPL) and viscoelastic (WVE) componentsof energy dissipation are correlated to bond energy (WB), and that any change in WB is reflectedin a significant change in WT (Okamatsu, Yasuda, & Ochi, 2001). In other words, for a given setof experimental conditions and class of materials, bond energy is directly related to total work offracture.
2. Background2.1. Surface free energySurface free energy (SFE) is defined as the work required to create a unit area of new surfaceof a material in a vacuum. Good–van Oss–Chaudhury theory, or acid-base theory, divides SFEof a material into three components based on the origin of the intermolecular forces (Van Oss,Chaudhury, & Good, 1987). These components are Lifshitz–van der Waals, γ LW; monopolar acid,γ +; and a monopolar basic, γ −. The total SFE of any material, γ total, is defined in equation (1).
γ Total = γ LW + 2√
γ +γ − (1)
Failure in an asphalt binder-aggregate system can occur in two locations: through the bulk ofthe asphalt binder or along the interface between the asphalt binder and aggregate. The energyrequired for the crack to propagate through the bulk of a material is known as the cohesive bondenergy, and is twice the total SFE of the material (equation (2)). Failure along the interface ofthe asphalt binder-aggregate system is known as adhesive failure. Adhesive bond energy is theenergy required for a crack to propagate along the interface and is a function of the SFE of boththe asphalt binder and aggregate (equation (3)).
�Gcoh = 2γ Total = 2(γ LW + 2
√γ +γ −
)(2)
�Gadh = 2√
γ LW1 γ LW
2 + 2√
γ +1 γ −
2 + 2√
γ −1 γ +
2 (3)
where, subscript 1 represents the asphalt binder and subscript 2 represents the substrate (aggre-gate). Cohesive and adhesive bond energies are a measure of a material’s and interface’sproperties, respectively. A higher value of either bond energy indicates that a greater amountof energy is required for a crack to propagate through the bulk of a material or along the asphaltbinder-aggregate interface.
A three-phase system occurs when water is present along the asphalt binder–aggregate interface.Aggregates almost always have a greater affinity for water than asphalt binder; or, in other words,
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
Road Materials and Pavement Design 283
the interaction between asphalt binder and aggregate in the presence of water is a almost always ahydrophobic interaction. This means that the energy required for water to displace asphalt binderfrom the aggregate surface is almost always less than zero (i.e., the displacement of asphalt binderfrom the aggregate surface by water is thermodynamically favored and will occur with no outsideenergy input). However, a recent study in which surface energy was measured using the universalsorption device on 22 minerals at Texas A&M University revealed that four of the 22 mineralstested demonstrated a thermodynamically favored bond with certain asphalt binders than withwater (Miller 2010). However, this finding should be considered in the context that the mineralswith a thermodynamically stable bond with asphalt binder may make up a very small fraction ofthe total surface area of most aggregates. Therefore, it is realistic to continue to assume that totalsurface energy of most aggregate surfaces generally favor a bond with water over a bond withasphalt binder. The energy required for a crack to propagate is a function of the SFE of the threephases (asphalt binder, aggregate, and water), and is defined as follows:
�Ga123 = γ13 + γ23 − γ12 (4)
where, the subscripts 1, 2, and 3 represent the asphalt binder, substrate (aggregate), and water,respectively. More generally stated, the terms on the right-hand side of Equation 4, γij , representthe energy of the interface between any two materials i and j and are computed from their respectiveSFE components as follows:
γij = γi + γj − 2√
γ LWi γ LW
j − 2√
γ +i γ −
j − 2√
γ −i γ +
j (5)
The surface free energy of paving materials has been successfully used in previous studies todetermine the cohesive bond energy of asphalt binders and the adhesive bond energy asphaltbinder–aggregate combinations with and without the presence of water (Lytton, Masad, Zollinger,Bulut, & Little, 2005; Bhasin et al., 2006; Little et al., 2006; Masad et al. 2006)
Little et al. (2006) reported an energy parameter (ER) that was correlated to the moisture sensi-tivity of asphalt mixtures. ER was used in this study to assess the change in moisture sensitivity ofasphalt mixtures due to modifications made to the asphalt binders. ER is a function of the surfacefree energy components of the asphalt binder and the aggregate and is expressed as:
ER =∣∣∣∣�Gadh − �Gcoh
�Ga123
∣∣∣∣ (6)
In equation (6) the terms �Gadh, �Gcoh and �Ga123 represent the adhesive bond energy between
the asphalt binder and the aggregate (equation (3)), cohesive bond energy of the asphalt binder(equation (2)), and work of debonding when water displaces the asphalt binder from its interfacewith the aggregate (equation (4)), respectively. A higher value of �Gadh indicates that morework is required to break the adhesive bond between the asphalt binder and the aggregate andhence implies better resistance to moisture damage. The term, �Gadh − �Gcoh, represents theability of the asphalt binder to wet or coat the surface of the aggregate. The microtexture onan aggregate surface also affects the mechanical bond between the surface of the aggregate andthe asphalt binder. Both the thermodynamic bond and the mechanical bond make up the overallbond between asphalt binder and aggregate surface. A lower magnitude of �Ga
123 indicates alower energy potential for water to displace asphalt binder from its interface with the aggregateand hence a higher resistance to moisture damage.
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
284 J. Howson et al.
2.2. Fracture energyThe fracture energy of the interface of two viscoelastic materials or a viscoelastic material bondedto a rigid substrate was studied by Xu, Hui, and Kramer (1992). In agreement with the similarstudies conducted on this topic by Gent and Scultz (1972), Gent and Kinloch (1971), and Andrewsand Kinloch (1973), Xu et al. (1992) stated that crack propagation and energy dissipation arefunctions of loading history and loading rate. Xu et al. (1992) derived a relationship showing thatthe total work of fracture of the interface between two viscoelastic materials is related to the bondenergy:
WT = WE[1 + f (aT , d, k)] (7)
where aT is the time-temperature shift factor for the bulk viscoelastic material, a∗ is crack growthvelocity, and k is a factor that is a function of the micromechanical properties of the interface,mechanical properties of viscoelastic material, and specimen geometry. The presence of aT , andd in equation (7) represents the influence of temperature and loading rate, respectively, on theresponse of viscoelastic materials. Results by Masad, Howson, Bhasin, Caro, and Little (2010)and Okamatsu et al. (2001) showed that even small increases in the bond energy resulted in largeincreases in the measured total work of fracture.
Kim, Freitas, & Allen (2008), Marek and Herrin (1958), and Chang (1994) used pull-off teststo evaluate the effect of loading rate on the failure force of thin asphalt films. These researchersdiscovered that a slower loading rate resulted in a smaller force to failure. A slower loading ratewill allow the viscoelastic asphalt film greater time to relax, thus producing a lower failure force.
The effect of varying the sample geometry on fracture (i.e., asphalt film thickness) has beenstudied by several researchers (Masad et al., 2010; Kim et al., 2008; Marek and Herrin, 1958; andChang 1994). The researchers found that an increase in the film thickness caused a decrease in themaximum observed force to failure. Masad et al. (2010) found that an increase in film thicknessalso caused an increase in the total work of fracture due to increases in displacement at the timeof failure.
Studies by Harvey and Cebon (2003, 2005), and Marek and Herrin (1958) evaluated the effect ofchanges in temperature on the failure force of thin asphalt films. Marek and Herrin (1958) reportedthat an increase in temperature caused a decrease in tensile strength when loading rate and filmthickness were held constant. Harvey and Cebon (2003, 2005) found that under brittle fractureconditions, the peak stress was rate independent; however, the peak stress was rate dependentwhen ductile failure occurred. They also reported that the strain at failure was rate independentfor both ductile and brittle fracture.
The adhesive bond between asphalt binder and aggregate is greatly affected by moisture. Givenenough time, water will almost always strip or at least provide the thermodynamic potentialto strip asphalt from the surface of an aggregate. Studies by Cho, Bahia, and Kamel (2005),Copeland, Youtcheff, and Shenoy (2007), and Kanitpong and Bahia (2008) used a device calledthe pneumatic adhesion tensile testing instrument (PATTI) to measure how different modificationsmade to asphalt binders affected their moisture resistance. Specimens were tested dry and after 24hours of conditioning in a water–bath, and the difference in failure strength reported. The tensilestrength of the specimens decreased after moisture conditioning and the failure mode sometimestransitioned from cohesive to partially or fully adhesive as the interface weakened.
3. Objectives and scopeThe cohesive bond energy of asphalt binders and the adhesive asphalt binder-aggregate bondenergy are important material properties. The magnitude of the bond energy; however, can be
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
Road Materials and Pavement Design 285
much smaller in comparison to the total work of fracture from mechanical tests. The objective ofthis research was to explore the relationship between bond energy and total work of fracture forasphalt binder-aggregate systems to determine if the bond energy can be used to predict the per-formance of the asphalt binder-aggregate systems. A series of uniaxial pull-off experiments wereconducted on asphalt binders-aggregate specimens, with known surface free energies, to deter-mine the total work of fracture for each combination. Various specimen geometries, loading rates,temperatures, and moisture contents were tested for each asphalt binder-aggregate combinationto better understand and quantify the relationship between bond energy and total work of fracture.The total work of fracture measured under the various experimental variations was comparedwith the measured bond energies in order to determine a linear regression relationship. Withinthis relationship, variations in loading rate, testing temperature and viscoelastic properties of theasphalt binder were considered.
4. MaterialsTables 1 and 2 display the materials used to determine the total work of fracture for the pull-off tests between asphalt binder and aggregate and the surface free energy components of thesematerials. The surface free energy terms used are presented and defined for equation (1) in theprevious section.
Three asphalt binders (AAB, AAD, and ABD) were procured from the Strategic HighwayResearch Program Materials Reference Library (Table 1). They were specifically chosen becauseof the range of surface free energy values they exhibited as determined from previous research atTexas A&M University (Bhasin 2006). The three asphalt binders were tested using the WilhelmyPlate device to determine their SFE values. Details on the measurement of surface free energy ofthese materials can be found in Little et al. (2006).
A constant temperature frequency sweep was performed for each asphalt binder to obtain itsbulk viscoelastic properties at various frequencies. The tests were performed at 23 ◦C using 8 mmdiameter parallel plates with a 2 mm gap. Figure 1 displays the magnitude of the storage and lossmodulus obtained at each frequency for the three asphalt binders.
Table 1. Surface free energy values of asphalt binders.
Surface free energycomponents (mJ/m2) Standard deviation (mJ/m2)
Figure 1. (a) Storage modulus, and (b) loss modulus for asphalt binders AAB, AAD, and ABD.
Two carefully chosen aggregates were selected to act as the substrate (Table 2). The firstaggregate, limestone, is a sedimentary rock that exhibited good performance in the field in termsof resistance to moisture damage. The second aggregate, andesite, is an igneous rock that exhibitedpoor observed performance in the field. The surface free energy components of the aggregateswere measured using the Universal Sorption Device (USD) and are shown in Table 2. A detailedmethodology regarding the use of the USD to measure the surface free energy of aggregates canbe found in the publication by Bhasin and Little (2007).
Surface free energy values in Tables 1 and 2 were used in equations (2) and (3) to obtain thecohesive and adhesive bond energy values displayed in Table 3. Table 4 displays the values of theEnergy Ratio (ER), computed using equation (6), for each asphalt-aggregate combination. Thematerial combination with a higher value of ER should display a greater resistance to moisture-induced damage.
Table 3. Adhesive bond energy between asphalt binders and aggregates and cohesive bond energy ofasphalt binders.
Table 4. Energy ratio (ER) of asphalt-aggregate com-binations.
ER (Asphalt)
Aggregate AAB AAD ABD
Limestone 0.45 0.22 0.12Andesite 0.24 0.14 0.08
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
Road Materials and Pavement Design 287
5. MethodologyThe experimental setup used in this research was similar to the setup used by Masad et al.(2010) to measure the force and displacement of asphalt-stainless steel samples with different filmthicknesses. In summary, the force was recorded using a 5 kip load cell and displacement of the thinasphalt film was measured using a high-resolution digital camera with image correlation software.The following factors were varied in testing each asphalt binder-aggregate combination:
• Film thickness: The asphalt film thickness was set at one of the following levels: 5 μm,10 μm, 30 μm, 50 μm, and 100 μm. The testing temperature and loading rate were keptconstant at 23 ◦C and 0.01 mm/s, respectively, for all tests with different film thicknesses.
• Loading rate: Specimens were tested using loading rates of: 0.005 mm/s, 0.01 mm/s, and0.02 mm/s. When the loading rate was varied, asphalt film thickness was kept constant at30 μm.
• Temperature: Specimens were tested at temperatures of 10 ◦C, 23 ◦C, and 36 ◦C. Asphaltfilm thickness was kept constant at 30 μm for specimens with varied testing temperatures.
• Moisture Content: Specimens were moisture conditioned after preparation for 12 hrs, 24 hrs,and 48 hrs. Asphalt film thickness, testing temperature, and loading rate were kept constantat 30 μm, 23 ◦C, and 0.01 mm/s, respectively for specimens tested at different moisturecontents.
A brief description of the sample preparation process is discussed below as well as some ofthe challenges faced. The testing and data acquisition systems were almost identical to the setupdescribed in Masad et al. (2010), with the only change being a slight modification in the grips toaccommodate the different sample holder geometry.
5.1. Sample preparationAggregate particles with sizes between 15 cm and 30 cm in diameter were obtained from theappropriate quarry. The top and bottom of the aggregate particle were removed using a diamondsaw to produce a flat and stable surface in which a core was taken from the aggregate. A 1.91 cm.diameter diamond core bit was used to extract the cores. A low speed saw, using distilled wateras a lubricant, was used to cut the aggregate cores into 1 cm cylindrical stubs. The cylindricalaggregate stubs were then lightly polished on each flat face using 6 μm aluminum oxide polishingpowder. Following the polishing process, the aggregate stubs were thoroughly rinsed by handusing distilled water, followed by submersion in a sonic bath of high purity distilled water toremove residual fine aggregate particles or polishing powder. The finished aggregate stubs wereplaced in an oven for 12 hours at 150 ◦C to remove all moisture from the sample.
Asphalt binder samples were prepared between the aggregate substrates using an AR 2000rheometer manufactured by TA Instruments, with a gap resolution of 1 μm. Custom designedgrips were manufactured for the AR 2000 rheometer that allowed preparation of the asphalt-aggregate samples. The end of the sample holder, both upper and lower, which fits into the grip,was cylindrical. This design virtually reduces any chance of misalignments between the upperand lower sample holders.
The aggregate stubs were glued into the sample holders using a fast drying epoxy. After theepoxy had cured, a propane torch was used to heat the aggregate surfaces of the upper and lowersample holders to remove water vapor and organic matter. A small drop of asphalt binder, heatedto 130 ◦C was applied to the aggregate surface of the bottom sample holder. The gap was reducedto the desired film thickness, and the sample was allowed to cool for 15 minutes before the excessasphalt binder was removed by means of a heated razor blade. Holders were designed to allow
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
288 J. Howson et al.
the samples to rest horizontally, while maintaining perfect alignment, in order to minimize anycreep within the asphalt film induced by the weight of the upper sample holder.
The samples were conditioned for 24 hours at 10 ◦C to guarantee the formation of a full adhesivebond. After conditioning, the samples were moved to the enclosed testing chamber, which wasmaintained at a constant temperature of 23 ◦C. Three layers of flat white acrylic paint were appliedto the outside of the sample holders. Painting the samples was a necessary step to provide theconditions required for the digital camera and correlation software to compute the displacement ofthe specimen during the test. Following painting, a light speckle coating of black flat spray paintwas applied. The samples were returned to the temperature chamber and allowed to equilibrateto the test temperature for two hours.
Aggregate, as opposed to stainless steel, is a naturally occurring raw material. Its properties andcomposition can vary due to small changes in position or depth. As a result, the utmost care wasneeded when selecting aggregate particles, including sampling from the same geologic strata andposition to ensure uniform aggregate composition. In addition, aggregate particles can containcracks, voids, or inclusions that can affect the structural integrity, surface texture or roughness,or surface free energy, respectively. The micro-cracks can cause the substrate to fail in tensionprior to failure of the interface. To reduce the possibility of micro-cracks, the largest possiblesamples were obtained directly from the quarry. Care was taken to minimize the content of voidsand inclusions on the aggregate testing surface to minimize variability between replicates. Manysamples were discarded for the above reasons; however, variability between replicates was stillhigh. Possible causes for the variability include absorption of the asphalt binder into the aggregatesubstrate, differences in texture, and non-uniform surface free energy between aggregate particles.
6. ResultsThe results of this study are presented in four subsections. The first subsection discusses therelationship of total work of fracture to asphalt film thickness. The second subsection examineshow the total work of fracture changes due to moisture conditioning. The third subsection discussesthe effect of loading rate and testing temperature on the total work of fracture. The final sectionexamines the relationship between bond energy and total work of fracture. More details about theexperimental results are available in Howson (2011).
6.1. Effect of change in film thicknessFilm thickness of the asphalt binder was varied from 5 μm to 100 μm in five steps: 5 μm, 10 μm,30 μm, 50 μm, and 100 μm. All samples were tested at 23 ◦C with a loading rate of 0.01 mm/s.Figure 2(a)–2(c) displays the total work of fracture for asphalt binders AAB, AAD, and ABDwith the limestone and andesite substrate, respectively.
Figure 2 illustrates the increase in the total work of fracture due to an increase in the asphalt filmthickness. Three primary observations were made in regard to the effect of film thickness of theasphalt binder on the measured total work of fracture. As the asphalt film thickness was increased,more energy was dissipated in the bulk of the viscoelastic asphalt binder prior to failure. This wascaused by increased viscous flow or yielding of the asphalt binder and the formation of largercavitations and fibrils at the aggregate surface.
The second observation was related to the differences in the total work of fracture betweenasphalt binders. As shown in Figure 2, large differences in the magnitude of the total work offracture were measured among the three asphalt binders. All asphalt binders used in this studywere unmodified and graded as PG64-22. Asphalt binders AAB and AAD exhibited very similarmagnitudes of total work of fracture for both limestone and andesite substrates. The magnitude of
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
Road Materials and Pavement Design 289
Figure 2. Comparison of total work of fracture for asphalt binders (a) AAB, (b) AAD, and (c) ABD withboth limestone and andesite substrate.
the total work of fracture for asphalt binder ABD exceeded that of the other two asphalt bindersfor every film thickness and aggregate substrate. This was in good agreement with the measuredsurface free energy of the asphalt binders. The cohesive bond energy for asphalts AAB and AADwere similar (30.92 and 39.04 mJ/m2, respectively) and lower than the cohesive bond energy ofasphalt ABD (68.07 mJ/m2). The difference in total work of fracture between the three asphaltbinders was more pronounced with andesite substrate. The total work of fracture of asphalt ABDwas at least 200% greater than the total work of fracture of asphalts AAB and AAD. The reasonfor this difference could be caused by surface texture, porosity, or chemical composition of theaggregate.
The third observation was the difference in the magnitude of the total work of fracture due tochanges in aggregate substrate. Figure 2 displays the difference in magnitude of the total work offracture when the aggregate substrate was changed. The limestone aggregate resulted in a highermagnitude of total work of fracture when asphalt AAB (Figure 2(a)) and asphalt AAD (Figure 2(b))were used. Substrate mineralogy; however, did not have a consistent trend in affecting the totalwork of fracture for asphalt ABD (Figure 2(c)). For this binder, the limestone aggregate resultedin a higher total work of fracture for film thickness at 30 μm, while the andesite aggregate resultedin a higher total work of fracture for film thickness at 5 μm, 50 μm, and 100 μm. These findings
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
290 J. Howson et al.
were counterintuitive when the bond energy was considered (Table 3). According to the bondenergies, the magnitudes of total work of fracture should have been identical for the two aggregatesubstrates when paired with each of the three asphalt binders if purely cohesive failure occurred,and the total work of fracture should have been greater for the andesite substrate for each ofthe three asphalt binders if any adhesive failure occurred. This finding suggests that anothermechanism besides physical adhesion influenced the total work of fracture. Work by Lesueur andLittle (1999), Little and Petersen (2005), and Little and Epps (2001) found that the addition ofhydrated lime to asphalt binders has the possibility of improving the rheology of the binder athigh temperatures and increasing the fracture toughening at low temperatures. In addition, theeffectiveness of hydrated lime was found to be dependent on the chemical composition of theasphalt binder (Hoffman et al. 1998). While it was not feasible that free lime was formed fromlimestone, it was possible that calcium ions on the surface of the limestone aggregate could havedeveloped stronger and more durable bonds with carboxylic acids or other functional groups inthe asphalt binder, forming stronger and more durable bonds at the bitumen-aggregate interface.As described by Little and Petersen (2005), this interaction could be reflected through the binderby amphoteric compounds.
6.2. Effect of moisture conditioningThe asphalt binder–aggregate samples were submerged in distilled water for time periods of 0 hrs,12 hrs, 24 hrs, and 48 hrs. All samples were prepared with a film thickness of 30 μm and weretested at 23 ◦C with a loading rate of 0.01 mm/s.
Several interesting findings can be extracted from the moisture conditioning experiments. Thefirst is related to the effect of moisture conditioning on the asphalt film. The asphalt film bonded toboth limestone and andesite substrates displayed an increased ability to flow as the conditioningtime was increased. Figure 3 displays the failed surfaces of samples of asphalt binder ABD withlimestone and andesite substrate for moisture conditioning times of 0 hrs, 12 hrs, 24 hrs, and48 hrs, respectively. The surface above the dotted line was the upper sample surface, with thelower sample surface being mirrored below the dotted line. As seen in Figure 3, all samples failedcohesively, but the cavitations became much larger, and visible fibrils appeared after the sampleswere conditioned for at least 12 hrs. The fibrils began as cavitations, but the asphalt binder’sincreased ability to flow, caused the pattern seen in the 12 hr, 24 hr, and 48 hr images. The lightercolor parts of the images are the center of the cavitations and were influenced by the colorationof the underlying aggregate substrate because of the very thin film of asphalt binder at theselocations. The black (or dark color) areas had more substantial film thicknesses and marked theridges where the fibrils were formed and finally failed.
The force-displacement graph for asphalt binder ABD with both limestone and andesite sub-strate is shown in Figure 4. Observing Figure 4, it was clear that the aggregate substrate had asubstantial effect on the measured total work of fracture. The increased flow of asphalt binderABD after moisture conditioning resulted in a decrease in the measured force for both aggre-gate substrates. With the limestone substrate, the decreased failure force was countered by anincrease in displacement until failure; however, this increase in displacement was not presentwhen andesite was used as the substrate.
When considering the impact of moisture conditioning, it was apparent that the increaseddisplacement experienced by the asphalt binders bonded to the limestone substrate after moistureconditioning offset the accompanied decrease in force in most cases. In general, the total workof fracture increased upon moisture conditioning for asphalt binder AAB (Figure 5(a)), remainedrelatively constant for asphalt binder AAD (Figure 5(b)), and decreased for asphalt binder ABD(Figure 5(c)) with limestone substrate. When andesite was used as a substrate, the total work of
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
Road Materials and Pavement Design 291
Figure 3. Effect of moisture conditioning on thin asphalt film for asphalt binder ABD with (a) limestoneand (b) andesite.
Figure 4. Force and displacement for asphalt binder ABD with (a) limestone and (b) andesite substrate.
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
292 J. Howson et al.
Figure 5. Comparison in total work of fracture for asphalt binders (a) AAB, (b) AAD, and (c) ABD withboth limestone and andesite substrate.
fracture decreased for the majority of tests. Only asphalt binder AAB exhibited any increase inthe total work of fracture after moisture conditioning for the andesite substrate. These results werein good agreement with the observed field performance, which showed limestone and andesitehaving good and poor moisture resistance, respectively.
Each asphalt binder responded uniquely to moisture conditioning. As stated above, asphaltbinder ABD displayed the greatest decrease in the magnitude of the total work of fracture aftermoisture conditioning. Prior to moisture conditioning, the magnitude of the total work of fracturefor asphalt binder ABD was at least 200% greater than the next highest asphalt binder for bothsubstrates. After moisture conditioning; however, the difference in magnitude of the total workof fracture was greatly reduced.
The aggregate substrate had a substantial effect on the total work of fracture during the moisturetests. Moisture conditioning of the specimens revealed the influence of aggregate substrate onthe moisture resistance of an asphalt binder-aggregate combination. Looking at Figure 5, the
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
Road Materials and Pavement Design 293
limestone substrate resulted in a substantially higher total work of fracture compared with theandesite substrate for all three asphalt binders. The difference in the magnitude of total workof fracture between aggregate substrates was much greater for moisture conditioned specimens(Figure 5) as compared with non-moisture conditioned specimens (Figure 2).
The results of the total work of fracture were in agreement with the values given by the energyparameter (ER). Table 5 displays the percent change of each asphalt binder-aggregate combinationfor a given moisture conditioning period, and the ER for that combination. A higher value of ERindicates that an asphalt binder-aggregate combination will be less susceptible to moisture-induceddamage and will have a smaller decrease in total work of fracture. ER was highest for asphaltbinder AAB and lowest for asphalt binder ABD for both limestone and andesite aggregates. Therelationship between ER and the average percentage change in WT for these experiments is shownin Figure 6. In addition, the relative percentage changes in the total work of fracture of the asphaltbinder–aggregate combinations with similar values of ER were comparable. Asphalt binder AADwith limestone exhibited an ER very close to that of asphalt binder AAB with andesite. Theresulting percentage changes in the total work of fracture due to moisture conditioning werevery similar (Table 5 and Figure 6). The observation was the same for asphalt binder ABD withlimestone and asphalt binder AAD with andesite. There was very good agreement between ERand the average percentage change in WT , demonstrating that surface free energy can be used asa screening tool to select moisture-resistant asphalt binder–aggregate combinations.
Figure 6. Relationship between ER and average percent change in WT.
Table 5. Percent change in total work of fracture due to moisture conditioning.
6.3. Effect of change in loading rate and testing temperatureThree loading rates (0.005 mm/s, 0.01 mm/s, and 0.02 mm/s) and three temperatures (10 ◦C,23 ◦C, and 36 ◦C) were applied to samples in which the film thickness was held constant at 30 μm.Figures 7 and 8 illustrate the force and displacement graphs for asphalt binder ABD with limestoneand andesite substrates, respectively. Asphalt binder ABD was chosen as an example because itdisplayed the greatest change properties when loading rate and temperature were changed.
There was an increase in the total work of fracture with an increase in the displacement rate.The increase in the total work of fracture with an increase in displacement rate stemmed from anincrease in the maximum failure force and/or an increase in the displacement at failure. ExaminingFigures 7 and 8, the maximum force increased for asphalt binder ABD as a result of increasing thedisplacement rate for every temperature and aggregate substrate except andesite at 23 ◦C. Asphaltbinder ABD paired with andesite at 23 ◦C experienced a steady decrease in force, but a steadyincrease in displacement at failure as the displacement rate was increased.
Figure 7. Force and displacement for asphalt binder ABD with limestone substrate for various loadingrates and temperatures of (a) 10 ◦C, (b) 23 ◦C, and (c) 36 ◦C.
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
Road Materials and Pavement Design 295
Figure 8. Force and displacement for asphalt binder ABD with andesite substrate for various loading ratesand temperatures of (a) 10 ◦C, (b) 23 ◦C, and (c) 36 ◦C.
When testing temperature was increased and loading rate held constant, the total work offracture decreased for each asphalt binder and aggregate substrate. The decrease in the total workof fracture was the result of a decrease in the failure force and/or a decrease in the displacement atfailure. The decrease in the failure force with the increase in temperature was caused by softeningof the asphalt binder. The two aggregate substrates produced different force-displacement profiles(Figures 7 and 8). Limestone substrate resulted in lower failure forces and higher displacementswith asphalt binder ABD than andesite for both the 10 ◦C and 36 ◦C tests for all displacementrates; however, higher failure forces and lower displacement were observed for the 23 ◦C tests at0.01 mm/s and 0.02 mm/s displacement rates.
Figures 9 and 10 reflect the influence of asphalt binder on the total work of fracture for thevarious temperatures and displacement rates. In agreement with the previous findings, asphaltbinder ABD exhibited the largest magnitude of total work of fracture for the standard temperatureand displacement rate of 23 ◦C and 0.01 mm/s, respectively. In addition, the magnitude of thetotal work of fracture for asphalt binder ABD was the greatest at a 10 ◦C testing temperature for alldisplacement rates and aggregate substrates. When the testing temperature was increased to 36 ◦C,
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
296 J. Howson et al.
Figure 9. Effect of change in loading rate for asphalt binders AAB, AAD, and ABD for limestone substrateat temperatures of (a) 10 ◦C, (b) 23 ◦C, and (c) 36 ◦C.
the total work of fracture of asphalt binder ABD greatly decreased. The increase in temperatureresulted in a reduction in the total work of fracture by an average of 80% with limestone substrateand 74% with andesite substrate, respectively, when referenced to the 23 ◦C testing temperature.The total work of fracture reduction experienced by asphalt binder ABD was significantly morethan asphalt binders AAB and AAD. As seen in Figure 9(c), the total work of fracture for asphaltbinder ABD with limestone substrate was very comparable with those of asphalts AAB and AAD,with the total work of fracture for asphalt binder ABD with andesite substrate being only slightlyhigher than those for asphalts AAB and AAD.
The importance of the aggregate substrate was recognized by looking at the total work offracture values at the different temperatures and loading rates (Figures 9 and 10). In the previoussection, limestone aggregate was found to yield higher values of total work of fracture thanandesite before and after moisture conditioning. Investigating the magnitudes of total work offracture resulting from changing the displacement rate and/or testing temperature, it was foundthat 23 of the 27 tests run with the limestone substrate exhibited higher magnitudes of total workof fracture than those run with andesite substrate. Three of four cases in which andesite exhibited
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
Road Materials and Pavement Design 297
Figure 10. Effect of change in loading rate for asphalt binders AAB, AAD, and ABD for andesite substrateat temperatures of (a) 10 ◦C, (b) 23 ◦C, and (c) 36 ◦C.
a higher total work of fracture occurred when the temperature was 36 ◦C (the highest value tested)and the failure was cohesive. The aggregate substrate should have minimum if any effect on thetotal work of fracture in these cases.
6.4. Fracture master curvesThe matrix of testing temperatures and displacement rates to which the asphalt binders weresubjected allowed the formation of a master curve for each asphalt binder and aggregate substrate.The 10 ◦C and 36 ◦C total work of fracture curves were shifted with respect to the 23 ◦C total workof fracture curve by multiplying the displacement rates by the shift factors, aT . The shift factorsused for the limestone and andesite substrates are shown in Tables 6 and 7, respectively. Theshift factors were found using the William, Landel, and Ferry method and were necessary tocalculate the temperature modified displacement rate (Ferry, 1980). The master curves shown inFigures 11 and 12 allowed the total work of fracture data to be estimated over a large range ofdisplacement rates.
Figure 11. Master curves for asphalt binder AAB, AAD, and ABD with limestone substrate at a referencetemperature of 23 ◦C.
6.5. Relationship between WT and WB
As discussed earlier, equation (7) was developed to predict the total work of fracture for a givenviscoelastic material with known bond energy and under various testing conditions. The testingconditions that had the greatest effect on the total work of fracture were temperature, loading rate,film thickness, and the interfacial properties. The experiments conducted in this study resulted inthe following observations.
• Increases in film thickness resulted in increases in total work of fracture due to greaterdissipation of energy in the bulk of the viscoelastic asphalt binder. The effect of energydissipation in viscous deformation can be captured by substituting k in equation (7) withthe loss modulus. The loss modulus for each asphalt binder was determined using a dynamicshear rheometer (DSR) test at different frequencies (Figure 1).
• Increases in loading rate, related to a, resulted in an increase in the total work of fracture.
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
Road Materials and Pavement Design 299
Figure 12. Master curves for asphalt binder AAB, AAD, and ABD with andesite substrate at a referencetemperature of 23 ◦C.
• Changes in testing temperature were combined with the loading rate (using term aT ) toformulate master curves (Figures 11 and 12).
In addition to the results listed above, moisture and aggregate type also influenced the totalwork of fracture. These effects, however, were dependent on the chemical makeup of the asphaltbinder and aggregate. Limestone, which has a lower bond energy than andesite, resulted in ahigher total work of fracture for two of the three asphalt binders. The authors believe that thiswas a result of stronger and/or more durable bonds developed between the calcium ions on thesurface of the limestone aggregate and certain functionalities in the asphalt binder. In regardto moisture conditioning, limestone resulted in a higher total work of fracture for each asphaltbinder and for each conditioning time than andesite. In addition, asphalt binder ABD displayedthe largest decrease in total work of fracture due to moisture conditioning, despite having thehighest bond energy. These experimental results were accurately predicted using ER as describedabove.
The results obtained in this study were compared against the relationship shown in equation (7)using a linear regression analysis. The linear regression function is displayed in equation (8), withthe regression constants for the two aggregate substrates shown in Table 8.
WT
WB= A1 + A2aT + A3d + A4k (8)
where, A1 through A4 are regression constants, aT is the shift factor, a is the loading rate, and kis the frequency dependent loss modulus at low frequencies. The values of the viscous moduluswere determined at frequencies of 0.0679, 0.14, and 0.289 Hz using the DSR and were presented
Table 8. Regression constants for linear regression model.
Figure 13. Relationship between experimental WT/WB and WT/WB found through regression analysisfor limestone substrate.
Figure 14. Relationship between experimental WT/WB and WT/WB found through regression analysisfor andesite substrate.
in Figure 1. These frequencies were selected since they reflect the slow rate of loading that wasused in the pull-off tests.
Figures 13 and 14 display the relationship between the ratio of measured total work of frac-ture to measured adhesive bond energy (WT/WB) and the ratio of WT/WB found using linearregression for limestone and andesite substrate, respectively. The linear regression model shownin equation (8) proved a good fit to the data (Figures 13 and 14) and displays that there wasa strong relationship between the bond energy and the total work of fracture. The R2 valueswere found to be 0.634 and 0.864 for limestone and andesite, respectively. The residual scatterin the data was likely caused by errors in the experimental data or limitations in the regressionmodel.
7. Conclusions and recommendationsThe prime objective of this study was to investigate the relationship between the bond energyand total work of fracture for asphalt binder–aggregate systems. In addition, the loading rate,temperature, film thickness, and moisture content at the interface were varied for different asphalt-aggregate specimens to investigate their influence on the total work of fracture. The primaryfindings are as follows.
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
Road Materials and Pavement Design 301
• An increase in film thickness resulted in increases in total work of fracture. Increased filmthickness allowed more energy to be dissipated in the bulk of the viscoelastic asphalt binder.This was observed in 28 of 30 cases.
• The influence of moisture on the total work of fracture varied for each asphalt-aggregatecombination. The total work of fracture for the combination of limestone with asphaltAAB increased upon moisture conditioning while the andesite and asphalt ABD combi-nation displayed the largest decrease due to moisture conditioning. The effect of moistureconditioning on the measured total work of fracture was accurately predicted using theparameter ER.
• Both testing temperature and loading rate affected the measured total work of fractureindependently of asphalt or aggregate type. Increasing the testing temperature decreasedthe total work of fracture, and increasing the loading rate increased the total work of fracturefor every asphalt binder-aggregate combination tested. This information was combined toformulate master curves for work of fracture for each asphalt binder-aggregate system topredict the total work of fracture for a large range of loading rates.
• The chemical composition of the asphalt binder greatly affects its total work of fractureunder various conditions. Asphalt binder ABD exhibited the highest bond energy andhighest measured total work of fracture at 10 ◦C and 23 ◦C, but also showed the greatestreduction in total work of fracture due to moisture conditioning and temperature change(10 ◦C to 36 ◦C). Asphalt binder AAB exhibited the lowest bond energy, but showed anincrease in total work of fracture due to moisture conditioning and a small decrease in totalwork of fracture due to changes in temperature (10 ◦C to 36 ◦C). The effects of moisturewere accurately predicted by ER and surface free energy.
• Aggregate substrate has a substantial effect on the total work of fracture. Limestonesubstrate performed better than andesite in every test. It resulted in higher total workof fracture across the range of film thicknesses, resulted in a smaller reduction in totalwork of fracture due to moisture conditioning, and resulted in a smaller reduction in totalwork of fracture due to increases in temperature. However, limestone exhibited a loweradhesive bond energy as compared with andesite. The authors hypothesize that calciumions on the surface of the limestone aggregate may have provided the source of a moredurable bond with certain functional groups, i.e., carboxylic acids in the asphalt bindersused.
• A regression model was developed between bond energy and total work of fracture. Theregression model demonstrated a strong relationship between bond energy and total workof fracture when loading rate, time-temperature shift, and viscous deformation were takeninto account. The results support the fact that the bond energy was a very good indicatorof performance.
The authors recommend conducting more experiments in which modified asphalt binders (i.e.,different PGs, anti-strip agents, short and long-term aged) or mastics are used. The use of mod-ifications, as compared to unmodified binders, will allow one to determine the effect of bindermodification on the total work of fracture. The impact of mineral filler and chemically active fillersshould be considered as part of the composite film. The authors suggest that the interaction ofmineral filler and bitumen may impact the bond between the mastic and the aggregate surface andthat the dispersion of mineral filler may toughen the mastic and impact work of fracture. More-over, the authors recommend developing a test method to determine the effect of temperature onthe surface free energy of asphalt binders.
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
302 J. Howson et al.
ReferencesAndrews, E.H., & Kinloch, A.J. (1973). Mechanics of adhesive failure. II. Proceedings of the Royal Society
of London, Series A, 332, 401–414Bhasin, A. (2006). Development of methods to quantify bitumen-aggregate adhesion and loss of adhesion
due to water. College Station, TX: Civil Engineering, Texas A&M University.Bhasin, A. Masad, E., Little, D., & Lytton, R. (2006). Limits on adhesive bond energy for improved resistance
of hot mix asphalt to moisture damage. Transportation Research Record: Journal of the TransportationResearch Board, 1970, 3–13.
Bhasin, A., Howson, J.E., Masad, E., Little, D.N., & Lytton. R.L. (2007). Effect of modification processeson bond energy of asphalt binders. Transportation Research Record: Journal of the TransportationResearch Board, 1988, 29–37.
Bhasin, A., & Little, D.N. (2007). Characterization of aggregate surface energy using the universal sorptiondevice. Journal of Materials in Civil Engineering, 19, 634–641.
Chang, W. (1994). Application of acoustic emission to study the cohesive and adhesive strength of asphalt.Report No. SHRP-A-682, Strategic Highway Research Program, Washington, DC.
Cheng, D.X., Little, D.N., Lytton, R.L., & Holste, J.C. (2002). Surface energy measurement of asphalt andits application to predicting fatigue and healing in asphalt mixtures. Transportation Research Record:Journal of the Transportation Research Board, 1810, 44–53.
Cho, D.W., Bahia, H.U., & Kamel, N.I. (2005). Critical evaluation of use of the procedure of super-pave volumetric mixture design for modified binders. Transportation Research Record: Journal ofthe Transportation Research Board, 1929, 114–125.
Copeland, A.R., Youtcheff, J., & Shenoy, A. (2007). Moisture sensitivity of modified asphalt binders: Factorsinfluencing bond strength. Transportation Research Record: Journal of the Transportation ResearchBoard, 1998, 18–28.
Ferry, J.D. (1980). Viscoelastic properties of polymers. New York: Wiley.Gent, A.N., & Kinloch, A.J. (1971). Adhesion of viscoelastic materials to rigid substrates III. Energy criterion
for failure. Journal of Polymer Science: Part A-2, 9, 659–669.Gent, A.N., & Schultz, J. (1972). Effect of wetting liquids on the strength of adhesion of viscoelastic materials.
The Journal of Adhesion, 3, 281–294.Harvey, J.A.F., & Cebon, D. (2003). Failure mechanisms in viscoelastic films. Journal of Materials Science,
38, 1021–1032.Harvey, J.A.F., & Cebon, D. (2005). Fracture tests on bitumen films. Journal of Materials in Civil
Engineering, 17, 99–107.Hefer, A.W., Bhasin, A., & Little, D.N. (2006). Bitumen surface energy characterization using a contact
angle approach. Journal of Materials in Civil Engineering, 18, 759–767.Hoffman, P., van Veldhuizen, M., Noor, H., & Smits, R. (1998). Hydroxide in filler. Utrecht, Netherlands:
Netherlands Pavement Consultants.Howson, J.E. (2011). Relationship between surface free energy and total work of fracture of asphalt binder
Kanitpong, K., & Bahia, H.U. (2008). Evaluation of Hma moisture damage in Wisconsin as it relates topavement performance. International Journal of Pavement Engineering, 9(1), 9–17.
Kim, Y.R., Freitas, A.C., & Allen, D.H. (2008). Experimental characterization of ductile fracture-damageproperties of asphalt binders and mastics. Presented at the 87th Annual Meeting of TransportationResearch Board, Washington, DC.
Lesueur, D., & Little, D. (1999). Effect of hydrated lime on rheology, fracture, and aging of bitu-men. Transportation Research Record: Journal of the Transportation Research Board, 1661,93–105.
Little, D., & Epps, J. (2001). The benefits of hydrated lime in hot mix asphalt. Report for National LimeAssociation, Arlington, VA.
Little, D., & Petersen, J. (2005). Unique effects of hydrated lime filler on the performance-related properties ofasphalt cements: Physical and chemical interactions revisited. Journal of Materials in Civil Engineering,17, 207–218.
Little, D.N., Bhasin, A., & Hefer, A.W. (2006). Using surface energy measurements to selectmaterials for asphalt pavements. Final Report 09-37, National Cooperative Highway ResearchProgram.
Dow
nloa
ded
by [
Uni
vers
itaet
s un
d L
ande
sbib
lioth
ek]
at 2
1:20
30
Apr
il 20
13
Road Materials and Pavement Design 303
Lytton, R.L., Masad, E., Zollinger, C., Bulut, R., & Little, D. (2005). Measurements of surface energyand its relationship with moisture damage. Publication FHWA/TX-05/0-4524-2, Texas TransportationInstitute, College Station, TX.
Marek, C.R., & Herrin, M. (1958). Tensile behavior and failure characteristics of asphalt cements in thinfilms. Proceedings of the Association of Asphalt Paving Technologists, 386–421.
Masad, E., Howson, J., Bhasin, A., Caro, S., & Little, D. (2010). Relationship of ideal work of fracture topractical work of fracture: Background and experimental results. Journal of the Association of AsphaltPaving Technologists, 79, 81–118.
Masad, E., Zollinger, C., Bulut, R., Little, D.N., & Lytton, R. (2006). Characterization of HMA moisturedamage using surface energy and fracture properties. Journal of the Association of Asphalt PavingTechnologists, 75, 713–754.
Miller, C. (2010). Adhesion and the surface energy components of natural minerals and aggregates. CollegeStation, TX: Geology and Geophysics, Texas A&M University.
Mittal, K. (1978). Adhesion measurement: Recent progress, unsolved problems, and prospects. ASTM SpecialTechnical Publication, 640, 5–17.
Okamatsu, T., Yasuda, Y., & Ochi, M. (2001). Thermodynamic work of adhesion and peel adhesion energy ofdimethoxysilyl-terminated polypropylene oxide/epoxy resin system jointed with polymeric substrates.Journal of Applied Polymer Science, 80, 1920–1930.
Van Oss, C., Chaudhury, M., & Good, R. (1987). Monopolar surfaces. Advances in Colloid and InterfaceScience, 28, 35–64.
Xu, D.B., Hui, C.Y., & Kramer, E.J. (1992). Interface fracture and viscoelastic deformation in finite sizespecimens. Journal of Applied Physics, 72, 3305–3316.
![ASPHALT CONCRETE [Types] - KSU ?· ASPHALT CONCRETE [Types] Introduction ... Sand Asphalt Mix Sand asphalt…](https://static.documents.pub/doc/80x56/5b72d5437f8b9a674d8d5d0c/asphalt-concrete-types-ksu-asphalt-concrete-types-introduction-sand.jpg)
