Embed Size (px)
Citation preview
Selective Absorption of Asphalt Binder by LimestoneAggregates in Asphalt Mixtures
Rong Luo, Ph.D., P.E., M.ASCE1; and Robert L. Lytton, Ph.D., P.E., F.ASCE2
Abstract: It has been found that aggregates used in paving asphalt mixtures absorb the asphalt binder into the porous structure of theaggregates. National test methods are available to measure the aggregate absorption. However, most research and measurements on aggregateabsorption are limited to the accessible voids at the aggregate surface. This paper presents recent findings on the selective absorption ofasphalt binder by aggregate particles in asphalt mixtures. The selective absorption of the binder is visualized on the aggregate surface undernatural light and ultraviolet light. Rings with different colors are identified on the cross sections of the aggregates in both hot-mix asphalt andwarm-mix asphalt, which indicates different asphalt components at different radial distances from the center of the aggregates. The asphaltcomponents penetrating into aggregates are verified using the laser desorption ionization–ion mobility–mass spectrometer. A significantlyhigher concentration of asphalt components is identified at the edge of the limestone sample than in its center after it is soaked in a PG 58-28asphalt binder (labeled AAD) for 32 h. Creep tests are also conducted on fresh limestone samples and limestone samples soaked in the AADbinder. The fresh limestone sample behaves elastically and is approximately twice as stiff as the sample soaked in the binder, which exhibitsnonnegligible viscoelastic properties in the creep test. DOI: 10.1061/(ASCE)MT.1943-5533.0000578. © 2013 American Society of CivilEngineers.
CE Database subject headings: Asphalts; Mixtures; Absorption; Limestone; Binders (material); Aggregates.
Author keywords: Asphalt mixture; Selective absorption; Limestone; Asphalt binder.
Introduction
It has been found that aggregates used in paving asphalt mixturesabsorb the asphalt binder into the porous structure of the aggregates(Lee et al. 1990; Curtis et al. 1993; Roberts et al. 1996). Nationalstandard test procedures are also available to measure aggregateabsorption (AASHTO 2010; ASTM 2007). These research obser-vations and measurements on aggregate absorption are limited tothe accessible voids at the aggregate surface. Little literature hasreported any finding on the distribution of the molecules of theasphalt binder inside of the aggregates in an asphalt mixture.
This paper reports very recent findings on molecules in the as-phalt binder traveling into the inside of aggregates in an asphaltmixture, which has nonnegligible effects on the mechanical proper-ties of the aggregates. Because the typical asphalt binders forpaving purposes have complex chemical compositions, moleculesare found to travel different distances into the aggregates. Specifi-cally, a typical asphalt binder has four broad component groups,including asphaltenes, resins, aromatics, and saturates (Readand Whiteoak 2003). Molecules in these component groups havevarious molecular weights and chemical characteristics, whichdetermine their traveling speed and distance inside the aggregate
particles. With selective absorption, the aggregate particle changesfrom a homogeneous material to a heterogeneous material withasphalt molecules scattered inside the aggregate particle. The selec-tive absorption not only changes the composition of the aggregateparticle but also its mechanical properties, such as the modulus.
This paper is organized in five sections. The next sectionpresents the visualization of asphalt components in aggregateparticles from both lab-mixed-lab-compacted (LMLC) mixturesand field cores. The following section details the verification ofasphalt components inside aggregates using the laser desorptionionization–ion mobility–mass spectrometer. The subsequentsection describes the creep tests that are conducted on fresh lime-stone samples and limestone soaked in an asphalt binder for aperiod of time in order to quantify the variation in mechanical prop-erties of the aggregate particles. The last section summarizes themajor findings of this paper.
Visualization of Aggregates in Asphalt Mixtures
In order to investigate the selective absorption of the asphalt binderby aggregates, asphalt mixtures are first examined visually undernatural light and ultraviolet (UV) light (black light), respectively.The selected asphalt mixtures include both hot-mix asphalt (HMA)mixtures and warm-mix asphalt (WMA) mixtures. These mixturesare either LMLC mixtures or field cores taken from asphalt pave-ments. In every selected asphalt mixture, the aggregate is acommon Texas limestone that is considered to be porous.
In order to examine the asphalt mixtures under UV light, a hand-held multiband UV lamp is used to generate UV light in a closeddark room without any other source of light. After the asphaltmixture specimen is placed on a brown paper board, UV light isswitched on to focus on the specimen. A digital camera is then used
1Associate Research Engineer, Texas Transportation Institute, TexasA&M Univ. System, 3135 TAMU, CE/TTI Building 503C, CollegeStation, TX 77843 (corresponding author). E-mail: [email protected]
2Professor, Fred J. Benson Chair, Zachry Dept. of Civil Engineering,Texas A&M Univ., 3136 TAMU, CE/TTI Building 503A, College Station,TX 77843. E-mail: [email protected]
Note. This manuscript was submitted on July 27, 2011; approved onMay 11, 2012; published online on May 15, 2012. Discussion period openuntil July 1, 2013; separate discussions must be submitted for individualpapers. This paper is part of the Journal of Materials in Civil Engineering,Vol. 25, No. 2, February 1, 2013. © ASCE, ISSN 0899-1561/2013/2-219-226/$25.00.
JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2013 / 219
J. Mater. Civ. Eng. 2013.25:219-226.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Uni
vers
ity o
f C
alif
orni
a, S
an D
iego
on
04/1
1/13
. Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
to take pictures of the specimen under UV light. More details ofthese pictures are presented later in this section.
Lab-Mixed-Lab-Compacted Mixtures
When investigating the LMLC HMA, the limestone aggregates arefirst sieved into a Type C dense aggregate gradation specified by theTexas DOT (2004). The aggregate matrix is then mixed with anunmodified asphalt binder labeled as AAD in the Strategic High-way Research Program (SHRP) Material Reference Library (MRL)(Jones 1993). The AAD binder has a PG 58-28 grade. The mixingtemperature is 135°C according to the Texas DOT specification forHMA. Immediately after mixing, a coarse aggregate coated withthe binder is picked up and then is sawed into halves, one halfof which under natural light is shown in Fig. 1(a). This figureclearly demonstrates a distinctive difference between the outerlayer and the inner part of the aggregate particle. Because of theabsorption of the asphalt binder, the outer layer of the aggregatehas a black color, while the inner part remains the white colorof the original limestone rock. When the same aggregate particleis placed under UV light, as illustrated in Fig. 1(b), a black layerwith a thickness of approximately 2 mm is shown on the peripheryof the cross-sectional area of the aggregate. Awhite (fluorescence)layer is shown next to the black layer and inwards to the center ofthe particle. Within the white layer, little fluorescence is shown inthe picture.
After examining the individual aggregate particle after mixing,the asphalt mixtures are compacted using the Superpave gyratorycompactor at a compaction temperature of 121°C to produce acylindrical specimen 152 mm in diameter and 178 mm in height.
The specimen is cored and cut to 102 mm in diameter by 102 mm inheight to achieve approximately uniform air-void distribution.When visualizing the cross section of this specimen under naturallight, similar observations are made on the aggregates in the HMAspecimen. As shown in Fig. 2(a), the outer layers of most aggregateparticles in the specimen have a black or brown color while thecenter of most aggregates remains the white color. A good exampleis the aggregate particle in the red ellipse shown in Fig. 2(a). Thisparticle has a dark brown outer layer with a thickness of approx-imately 2 mm, and its center part has the white color close to thatof the original limestone rock. Subsequently, the same HMA speci-men is examined under UV light, as presented in Fig. 2(b). Fluo-rescence is identified in most aggregate particles under UV light.Particularly, the particle in the red ellipse in Fig. 2(b) has a darkouter layer, an inner fluorescence layer, and the center part withlittle fluorescence. These observations match what is identifiedon the individual aggregate immediately after mixing.
Figs. 1 and 2 indicate that the asphalt binder not only is absorbedinto the accessible pores at the aggregate surface but also travelsinto the interior of the limestone aggregate. The layers with differ-ent levels of fluorescence shown on the aggregates under UV lightdemonstrate that different asphalt components travel different dis-tances in the aggregates. The travel distance of a specific asphaltcomponent may depend on its molecular weight and chemicalproperties as well as the mixing and compaction temperature.With different asphalt components at different locations in the crosssection of an aggregate particle, the cross section shows a numberof rings with different colors.
Because of the polar nature of the minerals in the aggregates, thenonpolar asphalt components with lighter weights may penetrate
Fig. 1. Limestone aggregate particle after laboratory mixing (a) undernatural light and (b) under UV light
Fig. 2. LMLC HMA specimen with AAD binder and limestone aggre-gates (a) under natural light and (b) under UV light
220 / JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2013
J. Mater. Civ. Eng. 2013.25:219-226.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Uni
vers
ity o
f C
alif
orni
a, S
an D
iego
on
04/1
1/13
. Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
into the center of the aggregates and the polar components withheavy weights may stay in the outer layer of the aggregate particlesshowing the black color. The white (fluorescent) layer on the ag-gregates under UV light is a good indication of aromatic hydrocar-bons because they strongly fluoresce under UV light.
Field Cores of Asphalt Pavements
In addition to LMLC asphalt mixtures, field cores of asphalt pave-ments are also investigated to evaluate the absorption of binder byaggregates in the field. The field cores selected in this study weretaken from an HMA section and a WMA section, respectively, ofLoop 368 in San Antonio, Texas. Both cores were taken when thesections had been in service for 1 year. The asphalt binder forthe HMA section is a Valero PG 76-22; the base asphalt binderfor the WMA section was a Valero PG 64-22 prior to modification,which was then modified using a warm-mix additive Evotherm tomeet the specifications of PG 76-22 (Button et al. 2007). Theaggregates of both HMA and WMA sections are typical Texaslimestone and field sand. Both HMA and WMA sections havethe same aggregate gradation that meets the requirement of a TexasDOT Type C dense-graded HMA. The HMA section was com-pacted at a temperature of 149°C; the WMA section was compactedat a temperature of 116°C.
Figs. 3(a and b) present the HMA field core under natural lightand UV light, respectively. A black or brown outer layer is easilyidentified on most aggregate particles under natural light, such asthe one in the red solid ellipse and the one in the yellow dashedellipse. When the same HMA field core is placed under UV light,
the aggregate in the red solid ellipse clearly shows a black outerlayer, a fluorescence layer and little fluorescence in its centerportion. This finding agrees with what is observed on the LMLCmixtures. In contrast, the aggregate particle in the yellow dashedellipse has a dark brown outer layer and strong fluorescence inits entire center portion when it is under UV light. This observationindicates that the aromatic hydrocarbons may have alreadypenetrated into the center of some aggregates in the HMA layerof an asphalt pavement in the field. A longer contacting periodbetween the asphalt binder and the aggregates leads to moreabsorption of asphalt components by limestone aggregates.
Compared to the aggregates in the HMA field core, the aggre-gate particles in the WMA field core do not have clear black orbrown outer layers when they are under natural light, as shownin Fig. 4(a). When the WMA field core is placed under UV light,most aggregates have a clearly fluorescent outer layer at the periph-ery of the aggregate particles, while a black/brown outer layer can-not be identified on the aggregate images, as illustrated in Fig. 4(b).Fig. 4 indicates that the polar components with heavy molecularweight may not be absorbed by the aggregate due to the signifi-cantly lower compaction temperature. In contrast, the limestoneaggregates still absorb the aromatic hydrocarbons, which stronglyfluoresce under UV light.
The comparison between the HMA and WMA field cores showthat the selective absorption of asphalt binder depends on thecompaction temperature. The limestone aggregates in an asphaltmixture compacted at a higher temperature tend to absorb moreasphalt components. In other words, a higher temperature facilitatesthe absorption of asphalt components.
Fig. 3. HMA field core taken from Loop 368, San Antonio, Texas,(a) under natural light and (b) under UV light
Fig. 4. WMA field core taken from Loop 368, San Antonio, Texas,(a) under natural light and (b) under UV light
JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2013 / 221
J. Mater. Civ. Eng. 2013.25:219-226.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Uni
vers
ity o
f C
alif
orni
a, S
an D
iego
on
04/1
1/13
. Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
Verification of Aggregate Absorption Using MassSpectrometer
Sample Preparation
The visual observation of the selective absorption of binder byaggregates is verified using the laser desorption ionization–ionmobility–mass spectrometer (LDI-IM-MS). This equipment isused to identify possible asphalt components at different locationswithin a limestone specimen that is soaked in an asphalt binder for aperiod of time. Instead of a regular aggregate particle in an asphaltmixture, a thin circular slice of limestone is preferred for theconvenience of the LDI-IM-MS test. This section will present thedetails of the limestone collection, sample preparation, and LDI-IM-MS testing.
Limestone rocks are first collected from a quarry in San Marcos,Texas, which is the same quarry as the one from which the lime-stone aggregates are obtained to make the LMLC HMA specimenspresented in last section. Each limestone rock is placed in a cylin-drical plastic container, as shown in Fig. 5(a). Cement mortar isthen cast in the container to firmly position the limestone rock,as illustrated Fig. 5(b). Every limestone rock is cored into a cylin-drical limestone specimen approximately 51 mm in diameter and76 mm in height. The actual height depends on the size of theindividual rock. A couple of cylindrical fresh limestone specimensare shown in Fig. 6(a).
Subsequently, a number of cylindrical limestone specimens aresoaked in the AAD binder at the compaction temperature 121°C for32 h. Fig. 6(b) shows an example of the limestone specimen aftersoaking in the asphalt binder for 32 h. The soaking period (32 h) is
calculated based on the following algorithm. Suppose that a roundaggregate particle has a diameter of D and that the radial distancebetween the aggregate surface and the wetting front is C, as shownin Fig. 7. Then the percentage of the completed absorption is
u ¼ CD
ð1Þ
It is assumed that u is proportional to the square root of thesoaking time t (Bear 1972; Holtz and Kovacs 1981; Lambe andWhitman 1979), as shown in Eq. (2).
u ∝ ffiffit
p ð2Þ
Fig. 5. Limestone rocks collected from San Marcos, Texas, (a) incylindrical plastic containers and (b) with cast cement concrete
Fig. 6. (a) fresh cylindrical limestone specimens cored from limestonerocks; (b) limestone specimen soaked in AAD ssphalt binder for 32 h
D
Aggregate
Wetting Front C
Fig. 7. Round aggregate with absorbed asphalt binder
222 / JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2013
J. Mater. Civ. Eng. 2013.25:219-226.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Uni
vers
ity o
f C
alif
orni
a, S
an D
iego
on
04/1
1/13
. Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
Eqs. (1) and (2) indicate that
u · D ¼ C ∝ ffiffit
p ð3Þ
Eqs. (4) and (5) are then inferred for aggregates with differentdiameters as follows:
u1 · D1 ∝ ffiffiffiffit1
p ð4Þ
u2 · D2 ∝ ffiffiffiffit2
p ð5Þ
When u1 ¼ u2, the following equations are derived:
D2
D1
¼ffiffiffiffit2t1
rð6Þ
Fig. 8. Cross section of limestone specimen after soaking in AADbinder for 32 h (a) under natural light and (b) under UV light
Sample
Laser
Bath Gas
DC Drift CellIon Optics
TOF
Fig. 9. Schematic illustration of LDI-IM-MS
200(a)
(b)
400 600 800 1000 1200m/z
1200
1000
800
600
400
200
0
Mob
ility
dri
ft t
ime
(µs)
200 400 600 800 1000 1200m/z
1200
1000
800
600
400
200
0
Mob
ility
dri
ft ti
me
(µs)
Fig. 10. IM-MS spectrums of (a) point A (edge spot) and (b) point B(center spot) on the limestone sample
JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2013 / 223
J. Mater. Civ. Eng. 2013.25:219-226.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Uni
vers
ity o
f C
alif
orni
a, S
an D
iego
on
04/1
1/13
. Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
t2 ¼ t1
�D2
D1
�2
ð7Þ
Eqs. (6) and (7) indicate the relationship between the soakingperiods of two aggregate particles with different sizes when thesame percentage of binder absorption is completed. When anasphalt pavement is constructed, it takes approximately 2 h to pro-duce the asphalt mixture and to ship the mixture to the constructionsite. Therefore, the soaking period of the aggregates in the asphaltbinder approximates 2 h. If taking an aggregate with 13 mm indiameter in the mixture as an example, then D1 ¼ 0.5 andt1 ¼ 2. Therefore, t2 can be calculated for the limestone rock speci-men with 51 mm in diameter using Eq. (7), provided that theaggregate with 13 mm in diameter has the same percentage ofbinder absorption completed as the limestone specimen with51 mm in diameter. The value of t2 is computed to be 32 h. In sum-mary, the binder absorption in the limestone specimen with 51 mmin diameter after soaking for 32 h is equivalent to that in theaggregate particle with 13 mm in diameter after 2 h of mixingand shipping to the construction site.
After soaking the limestone specimens in the AAD binder for32 h, a specimen is cut into halves. Fig. 8 presents two pictures ofthe cross section of one half of this specimen, one of which is the
specimen under natural light, and the other is the specimen underUV light. Layers with different colors are easily identified on thecross section of the specimen. This fact once again confirms theobservations on the LMLC specimens and field cores that are pre-sented in the previous section. Subsequently, the specimen is cutinto thin slices with a thickness of approximately 2 mm for theLDI-IM-MS testing.
LDI-IM-MS Testing
The circular limestone slice is tested using the LDI-IM-MS. Fig. 9is a schematic illustration of the LDI-IM-MS. The limestone slicesample is placed on a stainless steel plate inside the LDI-IM-MS, asshown in Fig. 9. A laser beam is then focused on a specific point onthe specimen to desorb and ionize that point on the specimen. Thisprocess is defined as laser desorption ionization (LDI). After theLDI process, the ions from the specific point on the sample enterthe drift cell with a length of 150 mm and filled with an inert gas. Inthe drift cell, there is a periodic electrical voltage gradient field thatseparates the ions according to their mass-to-charge ratio (m=z).Once the separation process is finished, ions are focused in theion optics and are mass analyzed using a reflection time-of-flight (TOF) mass spectrometer (Becker et al. 2009).
Fig. 11. Detailed IM-MS spectrums
224 / JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2013
J. Mater. Civ. Eng. 2013.25:219-226.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Uni
vers
ity o
f C
alif
orni
a, S
an D
iego
on
04/1
1/13
. Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
This testing process is applied to two specific points on the lime-stone sample, point A and point B as shown in Fig. 8(a). The reasonfor selecting an edge spot and a center spot is to compare the chemi-cal components at the two points so as to verify the selective ab-sorption of the limestone rocks visualized in previous sections.Fig. 10 shows the IM-MS spectrums of the edge spot and the centerspot. A significant difference in asphalt component concentration isidentified between the edge spot and the center spot when compar-ing Figs. 10(a to b). Specifically, the heteroatom hydrocarbonscorrespond to a nominal ion mass level of approximately 310on the IM-MS spectrum. The edge spot has a considerably higherdensity in heteroatom hydrocarbons than the center spot. Whenmagnifying the spectrum in the m=z range between 300 and320, more than one peak at a nominal ion mass level is identifiedon the spectrum, as presented in Fig. 11. This fact indicates thatthe complexity of the sample is significant, which is commonlyobserved in crude oils and asphaltene samples. In summary, theIM-MS spectrums plotted based on the LDI-IM-MS test resultsverify the visual observations on the nonuniform selective absorp-tion of asphalt binders by limestone rocks.
Quantification of Absorption Effect
It is commonly understood that a limestone rock can be taken as anelastic material, whereas an asphalt binder is viscoelastic. Theselective absorption of asphalt binder by the limestone rock defi-nitely changes the mechanical properties of the limestone rock withabsorbed asphalt binder. In order to quantify the absorption effect interms of the change of mechanical properties of the limestone rock,creep tests are conducted at room temperature on both fresh lime-stone rocks and limestone samples soaked in the AAD binderfor 32 h.
The material test system (MTS) is used to conduct the creeptests at a constant temperature of 20°C. As shown in Fig. 12, threepairs of linear variable differential transformer (LVDT) holders arefirst glued to each test sample. Then three vertical LVDTs aremounted on the surface of the sample when it is set up in theenvironmental chamber of the MTS. The three vertical LVDTsare placed 120° apart from each other around the sample surfaceto measure the vertical deformation of the test sample. The averagevalue of the measured deformations from the three LVDTs is usedin this study. After setting up the sample in the MTS, a staticloading of 17.8 kN is applied to the test sample for 2 h (7,200 s).
The vertical strains of the test samples are calculated using thevertical deformations measured by the LVDTs. Fig. 13 presentsthe vertical creep strains of a fresh limestone sample and alimestone sample soaked in the AAD binder for 32 h. The verticalstrain of the fresh limestone sample stays approximately constantwith the elapse of time. In contrast, the vertical strain of the lime-stone sample soaked in the binder has an increasing absolute valueof the strain as the testing time increases. Specifically, this lime-stone sample has a slow continuous deformation under constantload. The time-dependent behavior of this limestone sampleindicates its nonnegligible viscoelastic characteristics introducedby the absorbed asphalt binder. As a result, the selective absorptionof the asphalt binder by limestone aggregates is shown to changethe mechanical properties of the limestone aggregates, which turninto viscoelastic materials from elastic materials. This findingverifies the theoretical modeling results of the self-consistentmicromechanics models that were developed for asphalt mixtures(Luo and Lytton 2011). In addition, the magnitude of the compres-sive strain of the fresh limestone is approximately half of that of thelimestone soaked in the AAD binder when they are subjected to the
same compressive load. This fact indicates that the fresh limestonesample is approximately twice as stiff as the limestone samplesoaked in the binder. In other words, the selective absorption ofthe asphalt binder by the aggregates makes the aggregate particlessignificantly more compliant.
Conclusions
This paper presents the findings on the selective absorption ofasphalt binder by aggregate particles in asphalt mixtures. The se-lective absorption of the binder is visualized on the aggregatesurface under natural light and UV light. Rings with different colorsare identified on the cross sections of the aggregates in both HMAand WMA, which indicates different asphalt components at differ-ent radial distances from the center of the aggregates. The asphaltcomponents penetrating into aggregates are verified using the laserdesorption ionization–ion mobility–mass spectrometer. A signifi-cantly higher concentration of asphalt components is identified
Fig. 12. Configuration of creep test on (a) fresh limestone sample and(b) limestone sample soaked in AAD binder for 32 h
y = -0.0015x - 264.16
-300
-250
-200
-150
-100
-50
00 2000 4000 6000 8000
Cre
ep S
trai
n (µ
)
Time (s)
Limestone soaked inAAD binder
Fresh Limestone
Linear (Limestonesoaked in AADbinder)
Fig. 13. Measured axial strain of limestone samples in creep tests
JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2013 / 225
J. Mater. Civ. Eng. 2013.25:219-226.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Uni
vers
ity o
f C
alif
orni
a, S
an D
iego
on
04/1
1/13
. Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
at the edge of the limestone sample than in its center after it issoaked in the AAD asphalt binder for 32 h. Creep tests are alsoconducted on fresh limestone samples and limestone samplessoaked in the AAD binder. The fresh limestone sample behaveselastically and is approximately twice as stiff as the sample soakedin the binder, which exhibits nonnegligible viscoelastic propertiesin the creep test.
The findings of this study have significant impact on the fatigueand moisture damage of asphalt mixtures in terms of the bondingstrength between the asphalt binder and aggregates. Because of theselective absorption of binder by aggregates, the bonding strengthbetween the binder and the aggregates is in fact the bondingstrength between specific asphalt components remaining on theaggregate surface and the aggregates with absorbed asphalt com-ponents. Therefore, these results suggest that the adhesive bond en-ergy between the asphalt binder and the aggregates may need tobe calculated using the surface energies of the asphalt componentsremaining on the aggregate surface (instead of the entire asphaltbinder) and the aggregates with absorbed asphalt components(instead of the original limestone aggregates). In addition, whenusing micromechanics models to study an asphalt mixture as acomposite material, aggregates in the asphalt mixture may haveto be considered as viscoelastic materials due to the selective ab-sorption of the asphalt binder.
Identification and quantification of the asphalt components lefton the aggregate surface are the authors’ ongoing work in order tomore accurately estimate the adhesive bond energy between theasphalt binder and the aggregates.
Acknowledgments
The authors acknowledge Ms. Cindy Estakhri for providing fieldcores for testing in this study and Dr. Francisco Fernandez-Lima forconducting the laser desorption ionization–ion mobility–massspectrometer on the limestone samples.
References
AASHTO. (2010). “Standard method of test for specific gravity andabsorption of coarse aggregate.” AASHTO T85, Washington, DC.
ASTM. (2007). “Standard test method for density, relative density (specificgravity), and absorption of coarse aggregates.” ASTM C127-07,West Conshohocken, PA.
Bear, J. (1972). Dynamics of fluids in porous media, Dover, New York.Becker, C., Fernandez-Lima, F. A., and Russell, D. H. (2009). “Ion mobil-
ity-mass spectrometry: A tool for characterizing the petroleome.”Spectroscopy, 24(4), 38–42.
Button, J. W., Estakhri, C., and Wimsatt, A. J. (2007). “A synthesis ofwarm-mix asphalt.” Rep. No. FHWA/TX-07/0-5597-1, Texas Transpor-tation Institute, Texas A&M Univ. System, College Station, TX.
Curtis, C. W., Ensley, K., and Epps, J. (1993). “Fundamental properties ofasphalt-aggregate interactions including adhesion and absorption.”Strategic Highway Research Program (SHRP) Rep. No. SHRP-A-341,National Research Council, Washington, DC.
Holtz, R. D., and Kovacs, W. D. (1981). An introduction to geotechnicalengineering, Prentice Hall, Englewood Cliffs, NJ.
Jones, D. R. (1993). “SHRP materials reference library: Asphalt cements:A concise data compilation.” Strategic Highway Research ProgramRep. No. SHRP-A-645, National Research Council, Washington, DC.
Lambe, T. W., and Whitman, R. V. (1979). Soil mechanics, SI version,Wiley, New York.
Lee, D. Y., Guinn, J. A., Khandhal, P. S., and Dunning, R. L. (1990).“Absorption of asphalt into porous aggregates.” Strategic HighwayResearch Program (SHRP) Rep. No. SHRP-A/UIR-90-009, NationalResearch Council, Washington, DC.
Luo, R., and Lytton, R. L. (2011). “Self-consistent micromechanics modelsof an asphalt mixture.” J. Mater. Civ. Eng., 23(1), 49–55.
Read, J., and Whiteoak, D. (2003). The shell bitumen handbook, ThomasTelford, London.
Roberts, F. L., Kandhal, P. S., Brown, E. R., Lee, D. Y., and Kennedy, T. W.(1996). Hot mix asphalt materials, mixture design, and construction,NAPA Research and Education Foundation, Lanham, MD.
Texas Dept. of Transportation (DOT). (2004). “Standard specifications forconstruction and maintenance of highways, streets, and bridges.” Aus-tin, TX ⟨ftp://ftp.dot.state.tx.us/pub/txdot-info/des/specs/specbook.pdf⟩(May 16, 2008).
226 / JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2013
J. Mater. Civ. Eng. 2013.25:219-226.
Dow
nloa
ded
from
asc
elib
rary
.org
by
Uni
vers
ity o
f C
alif
orni
a, S
an D
iego
on
04/1
1/13
. Cop
yrig
ht A
SCE
. For
per
sona
l use
onl
y; a
ll ri
ghts
res
erve
d.
