当前位置：搜档网 › 1-s2.0-S0950061814010216-main

1-s2.0-S0950061814010216-main

Application of density gradient column to ?exible pavement materials:Aggregate characteristics and asphalt

absorption

Guanlan Liu a ,Xin Jin b ,Avery Rose a ,Yuanchen Cui a ,Charles J.Glover a ,?

a Artie McFerrin Department of Chemical Engineering,Texas A&M University,College Station,TX 77843,United States b

Shanghai Research Institute of Petrochemical Technology,Shanghai,China

h i g h l i g h t s

Density gradient column can measure volumetric properties and asphalt absorption. Individual aggregate pieces possess different volumetric properties.

Relations between aggregate properties and asphalt absorption are investigated. Asphalt absorption correlates very well with air void volume. Density gradient column provides precise results.

a r t i c l e i n f o Article history:

Received 10July 2014

Received in revised form 2September 2014Accepted 5September 2014

Available online 28September 2014Keywords:

Density gradient column Aggregate characteristics Bulk speci?c gravity

Apparent speci?c gravity Asphalt absorption Binder absorption Asphalt pavement Asphalt performance

a b s t r a c t

Accurate measurements of aggregate volumetric properties are essential to a satisfactory mix design and successful production of pavement.Moreover,determination of aggregate asphalt absorption is required to correctly estimate air voids and optimal binder content.The objective of this research was to apply the density gradient column technique to ?exible pavement materials and to investigate aggregate volumet-ric properties,asphalt absorption and their possible correlation.Results indicated that the density gradi-ent column can accurately measure piece-by-piece volumetric properties and asphalt absorption,and revealed a very good correlation between asphalt absorption and aggregate void volume (as re?ected by the aggregate bulk density).

1.Introduction

Accurate measurements of aggregate volumetric properties and absorption are essential to the development of satisfactory mix design and production [1].A key to the calculation of mixture vol-umetric properties are aggregate speci?c gravities,which are numerically equal to densities (for a water density of 1.0g/cm 3).Aggregate bulk density and apparent density allow conversion between aggregate mass and volume,necessary for calculating mixture volumetric properties such as air void percent,void ?lled by asphalt (VFA),voids in the mineral aggregate (VMA)and asphalt absorption [2].Therefore,precise and reliable measurements of aggregate densities are fundamental to the quality of hot mix asphalt or warm mix asphalt pavements.For several decades,various laboratory tests have been used to evaluate asphalt absorption at different conditions,which is neces-sary for successful mix designs [3–7].In paving mixtures,asphalt is absorbed by the porous structure of aggregates.Asphalt absorption is important because incorrect estimates translate into erroneous calculations of air void percent,VFA and VMA,all important param-eters used in mixture design to control pavement durability and stability.It has been reported that asphalt absorption is a process controlled by capillary force that depends both on asphalt proper-ties (including asphalt composition,viscosity,and surface tension)and aggregate properties (such as porosity,pore size distribution and surface chemical composition)[8,9].Particularly,comparison of water absorption and asphalt absorption shows they are correlated [3].The driving force for asphalt absorption is mainly determined by capillary action and absorption is reported to be a nonlinear function of time [9].It also has been reported that aggre-gate tends to selectively absorb some asphalt components over

?Corresponding author.Tel.:+1(979)8453389.

E-mail address:c-glover@https://www.sodocs.net/doc/704163717.html, (C.J.Glover).

others[10].Recent studies of aggregate volumetric properties and asphalt absorption in asphalt pavement can also be found [11–15].

Although standard testing procedures are available to measure volumetric properties and asphalt absorption in aggregates [16–18],these standard methods have some limitations[19].The ?rst problem is precision.Measuring aggregate volumetric proper-ties requires achieving a saturated surface dry(SSD)condition of the aggregate,a rather subjective assessment and therefore dif?-cult to reproduce precisely or accurately among investigators.Also, precision is limited by the inherent dif?culty of measuring volume accurately and of subtracting two large numbers that differ by a small amount.Second,standard methods are inadequate to under-stand the fundamentals of asphalt absorption.The standard meth-ods are based on measurements of samples comprised of large numbers of aggregate pieces having variable properties.However, asphalt absorption varies according to the properties of each spe-ci?c aggregate piece.Reporting average results on large samples veils individual differences between aggregate pieces.

The density gradient column(DGC),often used for determining the densities of small samples of polymers and other materials,is based on the preparation of a stable column of single phase liquid of variable density along its length[20–22].A specimen introduced into the column settles at the vertical position where it is in hydro-static equilibrium with the?uid in the column.In the DGC,the density of small specimens can be measured quite precisely.A key feature that improves precision is that density is measured directly rather than mass and volume separately.Although the DGC method has been used for polymer and other materials test-ing,it has not been applied previously to measurements of aggre-gate density and asphalt absorption.

A principal objective of this study was to adapt the DGC method to?exible pavement materials in order to more accurately mea-sure aggregate bulk and apparent densities and water and asphalt absorption.Part of achieving improved precision was to eliminate the subjective determination of the saturated surface dry(SSD) condition.In addition,this approach provides a method of measur-ing aggregate bulk and apparent densities and the ability to com-pare asphalt absorption directly to water absorption,i.e.to the aggregate void volume.An additional objective was to compare the precision of the DGC method for asphalt absorption to the pre-cision of using the standard methods and practice for measuring asphalt absorption.

2.Materials and methods

2.1.Materials

Six types of pavement aggregates(sandstone,granite,manufactured light-weight,quartzite,limestone,and gravel)and one asphalt binder(PG64-22)were selected for study.The size of the aggregates ranged from0.1cm3to0.5cm3.Lith-ium metatungstate heavy liquid was used to generate a liquid density gradient suit-able for pavement aggregate materials,and precision glass beads were used to calibrate the density gradient.Paraf?n wax was used as a coating for determining aggregate bulk density.

2.2.Density gradient column apparatus and methodology

The density gradient method is based on the direct measurement of particle density in a column of?uid that has a linear density gradient.The gradient is gen-erated by continuously blending two completely miscible make-up?uids of differ-ent densities and feeding them slowly to the bottom of the column.For asphalt binder measurements the two make-up?uids were water and a brine solution of appropriate density.For aggregate measurements,the two make-up?uids were water and a heavy liquid(lithium metatungstate,density2.95g/cm3).The heavy liquid was regenerated by evaporating the water.

The DGC method is direct and ef?cient because it does not require measuring particle volume,a dif?cult measurement to do precisely.Instead,the accuracy of the method depends upon an accurate calibration of the density variation with position in the column.Two sets of glass beads,traceable to NIST weights and measures,were used for calibrations;one set(for asphalt density determinations) provides density calibrations at23°C from0.94to1.10g/cm3±0.0002g/cm3and the other set(for aggregate density determinations)provides density calibrations from2.0to2.8g/cm3±0.0005g/cm3.The thermal expansion coef?cient of the beads was given to be0.000037g/cm3/°C.An example calibration of a column prepared for aggregate measurements is presented in the Results and Discussion section, below.The measurement requires generating a linear density?uid in an appropri-ate graduate cylinder containing the calibration beads.The beads,each of different density,settle at the point in the column equal to their density.Particles of asphalt or aggregate are dropped into a column of appropriate density range and also settle according to their density.The position of each asphalt bead or aggregate particle was read with a cathetometer of0.001cm precision.Asphalt or aggregate densities were determined by linear interpolation of the vertical calibration.

Fig.1shows a schematic of the DGC;aggregate particles of varying states were dropped into the DGC to measure their density.These states included wax coated, water saturated,and asphalt treated,and are described below.

2.3.Asphalt absorption calculation

The DGC methodology requires immersing the aggregate with asphalt binder at 121°C for15s on a hot plate followed by curing the coated aggregate at143°C for 2h in the oven,to approximate TxDOT speci?cations on laboratory preparation of asphalt mixture[23].The mass of asphalt absorbed by aggregate m abs-b is deter-mined by subtracting the aggregate mass from the mass of aggregate plus absorbed and coated asphalt and then correcting for the mass of asphalt coating:

m absàb?fem ab;m a;q b;q ab;q a vT?m abàm aà

m ab q b

m a q b

a v

e1T

where m ab is the mass of aggregate coated with asphalt(including both absorbed asphalt and excess asphalt coating),m a is the mass of aggregate(including air voids of negligible mass),q b is the asphalt density,q ab is density of asphalt-treated aggre-gate(including both absorbed asphalt and excess asphalt coating),q av is bulk density of the aggregate,i.e.,the density of the aggregate including its air voids.The masses were measured using a precision balance,and densities were measured using the DGC.While the?rst four properties can be measured directly,q av is measured using the wax-coating method described below.

2.4.Bulk density determination

The bulk density of a single aggregate particle is the density(mass/volume)of the aggregate including its accessible air voids volume.The bulk density of a sample of many aggregate particles is the density(mass/volume)including all of the parti-cle accessible internal air voids but excluding the void spaces in the interstices between particles.Thus,the bulk density of an aggregate sample of many particles is an average bulk density of those particles.In the standard method,a saturated surface dry(SSD)condition is necessary to determine aggregate bulk density.Under this SSD condition,the pores of aggregate particles are?lled with water and no excess water is on the particle surfaces(i.e.,in the spaces between particles).The SSD condition is experimentally achieved by removing this excess water from the aggregate by absorbing surface moisture with towels(for large aggregate)or allow-ing?ner aggregate to dry to‘‘the appropriate state’’.This method is subjective to

G.Liu et al./Construction and Building Materials72(2014)182–188183

operator’s judgment and also dif?cult to achieve uniformly.Thus,achieving the SSD condition is likely one of the main limits to the precision of aggregate bulk density measurements.

In order to achieve better precision of bulk density measurements(for single particles),a wax coating method was developed.This method is designed to coat the aggregate with paraf?n wax,without absorption into the pores.The aggregate is chilled in a refrigerator and molten wax,near its melting point is coated onto the surface of aggregate with a?ne brush,taking care to?ll the surface contours.In this procedure,the cold aggregate freezes the wax upon contact and prevents it from penetrating into the aggregate pores.The bulk density is calculated using Eq.(2), with the amount of absorption(in this case of wax)assumed equal to zero:

0?m acàm aàm ac q c

m a q c

a v

e2T

from which the aggregate bulk density q av is calculated as:

q a v ?

m a q c

m atm ac q c q

àm ac

e3T

where m ac is the mass of the wax-coated aggregate,m a is the mass of the uncoated aggregate,q c is the density of wax,q ac is the density of the wax-coated aggregate.

2.5.Apparent density measurement

The aggregate apparent density is the aggregate mass per volume excluding the accessible air void volume(but including trapped void volume inside the aggregate particles).Vacuum water saturation for45min,followed by24h additional water immersion are applied before measuring the density of the water-saturated aggre-gate in the DGC.The apparent density of the aggregate is calculated using Eq.(4):

q a ?

a v

à1q

a v

à1

e4T

where q a and q av are as de?ned above and q aw is the density of water-saturated aggregate and q w is the density of water at the laboratory temperature.

2.6.Accessible void volume calculation

With the aggregate mass and densities determined,the(accessible)void vol-ume(V void)of each aggregate was calculated as its bulk volume minus the aggregate apparent volume:

V v oid?m a

a v

m a

e5T

3.Results and discussion

3.1.DGC calibration

An example DGC linear calibration,with a range designed for aggregate density measurements,is shown in Fig.2.The density of the aggregate was determined by measuring each aggregate’s vertical position in the column using the telescopic cathetometer having a precision of0.001cm.Additionally,the paraf?n wax density(0.917g/cm3)and the PG64-22asphalt density(1.036g/ cm3)were measured.The two column?uids used to create the DGC for measuring the wax density were water and isopropyl alco-hol,a very poor solvent for the wax.

3.2.Aggregate characteristics

From DGC measurements,piece-by-piece aggregate mass,bulk density,apparent density,void volume,and void fraction were measured for?ve natural aggregate materials and one lightweight, manufactured material.Detailed piece-by-piece data for the lime-stone and quartzite are shown in Tables1and2.

Summary data for all six aggregate types(sandstone,granite, limestone,gravel,quartzite,and the manufactured lightweight material)are reported in Table3.It is clear that even for the same type and source of aggregate,intrinsic aggregate properties show signi?cant differences between individual pieces;air void frac-tions,for example,are not distributed evenly among the separate aggregate pieces.Also notable is the variation in average void frac-tion among the six aggregates,ranging from0.06for the limestone to0.33for the manufactured lightweight material.

It should be noted that the measurement uncertainty is much less than the piece-to-piece variability,which is dominated by inherent aggregate properties.Based on the aggregate calibration bead uncertainty of0.0005g/cm3,a void fraction uncertainty(esti-mated by error propagation and expressed as a fraction of the aggregate accessible voids)is estimated to range from0.001(for the sandstone)to0.004(for the limestone).It is expected that the actual density uncertainty is greater than the calibration beads precision(due to uncertainty of the location of the irregularly shaped aggregate centers of mass,e.g.);thus,if the uncertainty is a factor of10greater(i.e.±0.005g/cm3),similar to the ASTM reported precision,then these uncertainty ranges in void fraction would be from0.01to0.04respectively,i.e.,from one to four percent of the calculated aggregate void volumes.These values would translate directly to water absorption uncertainty(i.e.,an uncertainty of from one to four percent of the reported

absorption

Table1

Aggregate characteristics,limestone(n=10).

Aggregate Mass(g)Bulk

density

(g/cm3)

Apparent

density

(g/cm3)

Void

volume

(cm3)

Void

fraction(%)

Limestone0.9145 2.491 2.7980.04011

Limestone0.9289 2.560 2.7870.0298

Limestone0.5146 2.510 2.7390.0178

Limestone 1.2332 2.635 2.7180.0143

Limestone0.4096 2.624 2.7120.0053

Limestone0.4866 2.587 2.6730.0063

Limestone0.7165 2.513 2.9030.03813

Limestone0.5355 2.612 2.6970.0063

Limestone0.8588 2.597 2.7530.0196

Limestone 1.0510 2.645 2.6880.006 1.6

Table2

Aggregate characteristics,quartzite(n=7).

Aggregate Mass(g)Bulk

density

(g/cm3)

Apparent

density

(g/cm3)

Void

volume

(cm3)

Void

fraction(%)

Quartzite0.5587 2.369 2.8020.03615

Quartzite0.2840 2.536 2.8060.01110

Quartzite0.3256 2.492 3.0410.02418

Quartzite0.6626 2.412 3.0980.06122

Quartzite0.2766 2.550 2.9080.01312

Quartzite0.2705 2.579 2.8310.0099

Quartzite0.3184 2.519 2.8880.01613

184G.Liu et al./Construction and Building Materials72(2014)182–188

number)and compare to ASTM?ne and coarse aggregate precision ranges of from approximately3%to20%of the average water absorption value[16–18].

3.3.Asphalt absorption

The DGC method can be used to measure asphalt absorption on an aggregate piece-by-piece basis,thus providing the ability to relate absorption directly to the other aggregate characteristics shown in Tables1–3.Typically,asphalt absorption is expressed per mass of aggregate,suggesting that for a given type of aggre-gate,absorption is proportional to mass.However,data in Fig.3 show that this presumption is not necessarily true.There may be a linear correlation on a per mass basis,as for the quartzite aggre-gate,or,there may be no correlation,as for the limestone.3.4.Factors important to asphalt absorption

To evaluate factors important to asphalt absorption,several aggregate properties provided by DGC measurements were evalu-ated.Fig.4shows the relationship of void volume to aggregate mass for individual pieces of limestone and quartzite.For the lime-stone,there is no correlation seen whereas for the quartzite,the void volume follows a clear linear correlation with aggregate mass. These two different relations are consistent with those shown in Fig.3and suggest that the volume of absorbed asphalt relates to the aggregate void volume,relationships that are shown in Fig.5. The origin for each data set is not considered an a priori data point in these plots,but certainly,it is very consistent with the results.It is clear that there are very good linear relationships between asphalt absorption volume and void volume.Apparently,void

Table3

Summary characteristics for samples of the six aggregate types.

Aggregate Mass(g)Bulk density(g/cm3)Apparent density(g/cm3)Void volume(cm3)Void fraction(%)

Sandstone(n=9)AVG0.9773 2.346 3.0620.10123

SD0.38070.1920.1470.0809 Granite(n=7)AVG0.5016 2.504 2.8500.02412

SD0.16430.0600.1210.0115 Limestone(n=10)AVG0.7649 2.577 2.7470.0186

SD0.28630.0500.0700.0124 Gravel(n=14)AVG0.5756 2.399 2.8350.03815

SD0.21570.0600.1270.0215 Quartzite(n=7)AVG0.3852 2.494 2.9110.02414

SD0.15830.0760.1170.0195 Lightweight(n=7)AVG0.3167 1.504 2.2690.06933

SD0.17130.0440.1660.0436

Note:AVG:average;SD:standard deviation.

G.Liu et al./Construction and Building Materials72(2014)182–188185

volume is a more meaningful property for describing and correlat-ing asphalt absorption than aggregate mass.

3.5.Importance of void volume to asphalt absorption

In order to verify the importance of void volume in absorption research,DGC absorption data were obtained for additional types of aggregate;sandstone,granite,gravel,and lightweight aggregate also were evaluated.All of these experiments used the PG64-22 asphalt binder with15s mixing(immersion)time at143°C fol-lowed by2h curing time at121°C.Fig.6shows absorbed asphalt volume versus aggregate void volume for all six types of aggre-gates.Again,the piece-by-piece aggregate asphalt absorption cor-relates very well with the aggregate air voids.Moreover,this type of correlation is found for all six types of aggregates(sand-stone,granite,limestone,quartzite,gravel,and lightweight manu-factured aggregate),indicating that the aggregate void volume plays a key role in determining asphalt absorption.As noted below, with the exception of manufactured lightweight aggregate,the slopes of the correlations for the?ve natural aggregate materials are0.4(to within95%con?dence intervals),indicating that for the same asphalt and identical mixing/curing condition protocol, the percentages of void volume ultimately occupied by the asphalt is similar for these?ve different aggregate types.

3.6.Statistical analysis

A statistical analysis on the DGC regression equations is sum-marized in Table4and compared to data of Lee[3].Lee presented a careful and thorough study of absorption using standard absorp-tion methods(Rice,immersion,and bulk impregnated speci?c gravity)on a large number of aggregate types and binders of differ-ent penetration grade.The number of absorption determinations in each case provides a unique data set for evaluating the precision of these methods.The DGC method provides data of greater precision than methods reported by Lee,as re?ected by a greater R2and more precise estimates of regression coef?cients(smaller standard error in both slope and intercept),for the same or smaller sample size.Also,it is notable that standard errors of the intercepts for the DGC data are dramatically less than for Lee’s data.

In order to further quantify the regression parameter precision, 18points from the DGC data for the?ve natural aggregates(47 aggregate pieces)were randomly selected to match the number of regression points reported by Lee[3]on his asphalt absorption data set,Rice127-1.For these18data points,the95%con?dence intervals for the slope is0.037for the DGC method versus0.084 for the Rice127-1data;the intercept con?dence interval is0.002 for DGC versus0.413for the Rice method.

Interestingly,this DGC linear correlation is strengthened con-siderably when statistically analyzing the regression parameters based on the47data points for all?ve natural aggregates studied (excluding the manufactured lightweight material).The95%con?-dence intervals for the overall regression parameters are0.012for the slope and0.001for the intercept.

The Rice127-1data and the DGC results are summarized in Fig.7.The difference in slopes(average absorption level)for the two methods is likely due to differences in the immersion and cur-ing protocols.The interesting feature appears to be that a consis-tent protocol provides a consistent absorption level,when viewed as a function of void volume.Furthermore,while the DGC method provides more precise measurements than the standard methods and thus can be used as a tool to explore the fundamen-tals of aggregate absorption,it is limited to characterizing absorp-tion on individual aggregate pieces of a single aggregate size range rather than aggregate samples large enough to represent mixture aggregate coarse or?ne samples.

186G.Liu et al./Construction and Building Materials72(2014)182–188

4.Conclusions

In order to better understand the fundamentals of ?exible pave-ment materials,the density gradient column (DGC),was applied to this research.Characteristics of individual aggregate pieces includ-ing volumetric properties and asphalt absorption were investi-gated.Also,the relationship between these aggregate properties and asphalt absorption was studied.The key ?ndings of this study are:

The DGC provides a convenient method for measuring aggre-gate bulk density and apparent density,for individual aggregate pieces.From these data,other volumetric properties of aggre-gate such as air void fraction can be calculated.Although this procedure would require many measurements to representa-tively characterize an entire aggregate source,the method has the advantage of providing speci?c aggregate characteristics and as such,can provide an improved understanding of the importance of these fundamentals to pavement materials.

Analysis of sets of aggregate pieces showed (unsurprisingly)that different aggregate types may possess inherently different volumetric properties.Interestingly,even for the same aggre-gate type,there can be signi?cant variability among individual

aggregate pieces in their void volume and void fraction.For example,the void fraction (void volume per bulk volume)for limestone varied from 1.6%to 13%for a sample of 10pieces of aggregate,with the range in average void fraction for the ?ve natural aggregate samples studied varying from 6%to 23%and the manufactured lightweight aggregate material at an average void fraction of 33%.

The density gradient column provides the capability to deter-mine asphalt absorption for individual aggregate pieces,thus allowing correlations between the amount of asphalt absorp-tion and other aggregate-speci?c properties to be explored.Experimental results indicate that asphalt absorption generally correlates well with aggregate void volume and may not corre-late at all with aggregate mass.This conclusion holds for all six aggregate types that have been characterized in this work,indi-cating that a volumetric based correlation is fundamental to aggregate absorption.

Statistical analysis of the linear regression coef?cients relating asphalt absorption to void volume show that the DGC pro-vides signi?cantly more precise results than the standard methods and practice allow,with much smaller scale and fewer replicates (around 10pieces of aggregates,ranging from 0.5to 2g each piece).Thus,the DGC method can be a valuable tool to study small amounts of samples at different conditions,which is vital to reveal the fundamental process of asphalt absorption.Acknowledgements

The author thanks Texas Department of Transportation (TxDOT)and Federal Highway Administration (FHWA),project 0-6613,for their ?nancial support and all those who helped in this research work.Particularly,special thanks to those followings for their assistance:Mr.Jerry Peterson (TxDOT),Dr.Robert Lytton (TTI),Ms.Cindy Estakhri (TTI),and Dr.Edith Arambula (TTI).References

[1]Huang YH.Pavement analysis and design.Englewood Cliffs,N.J.:Prentice Hall;

1993.

[2]Buchanan MS.An evaluation of selected methods for measuring the bulk

speci?c gravity of compacted hot mix asphalt (HMA)mixes.Association of Asphalt Paving Technologists Proc,2000.p.69.

[3]Lee DY.The relationship between physical and chemical properties of

aggregates and their asphalt absorption.Assoc Asphalt Paving Technol Proc 1969;38:242–75.

Table 4

Regression equations and statistical analysis:asphalt absorption versus void volume.Methodology Materials Regression equation R 2n SE a SE b

Rice a

127-1Y =0.5162X à0.15840.9137180.03970.1947127-2Y =0.5067X à0.41060.9059160.04360.2264Immersion a 127-1Y =0.7760X +0.01370.8103160.10030.4616127-2Y =0.6749X +0.50410.8221140.09060.4380BISG a

127-1Y =0.6604X +0.11550.8367240.06220.2806127-2Y =0.5822X à0.38760.7626270.06500.2812Density gradient column

Sandstone Y =0.3905X +0.00070.993390.01210.0015Granite Y =0.3908X à0.00130.901870.05770.0015Limestone Y =0.3892X à0.000040.8922100.04780.0011Gravel Y =0.4030X +0.00040.9581140.02430.0010Quartzite Y =0.3639X à0.00070.992470.01430.0004Lightweight Y =0.3190X à0.00720.975870.02250.0018Five natural b

Y =0.3969X à0.0003

0.9894

0.0061

0.0004

Note:Y :dependent variable,volume of asphalt absorption;X :independent variable,void volume;SE a :standard error of slope;SE b :standard error of intercepts;n :points in linear regression.a

Literature data.b

Five natural aggregate types,this study (excludes manufactured lightweight).

0.39n=9

n=10

n=14

0.36n=7

0.40n=47

0.38n=18

R i c e 127-1

S a n d s t o n e

G r a n i t e

L i m e s t o n e

G r a v e l

Q u a r t z i t e

A l l F i v e , A l l d a t a p t s

A l l F i v e , 18 r a n d o m d a t a p t s

Lee

Density Gradient Column

Summary of asphalt absorption:Rice 127-1data of Lee [3]and DGC method G.Liu et al./Construction and Building Materials 72(2014)182–188

187

[4]Kandhal PS,Lee D.Asphalt absorption as related to pore characteristics of

aggregates.Highway Res Rec1972;404:97–111.

[5]Curtis CW,Jeon YW,Clapp DJ.Adsorption of asphalt functionalities and

oxidized asphalts on aggregate surfaces.Fuel Sci Technol Int1989;7(9): 1225–68.

[6]Curtis CW,Ensley K,Epps J.Fundamental properties of asphalt-aggregate

interactions including adhesion and absorption.Washington,DC:Strategic Highway Research Program,National Research Council;1993.

[7]Brown ER,Prithvi S.Kandhal,Freddy L.Roberts,Y.Richard Kim,Dah-Yinn

Lee,Thomas William Kennedy.Hot Mix Asphalt Materials,Mixture Design, and https://www.sodocs.net/doc/704163717.html,nham,Md:NAPA Research and Education Foundation, 2009.

[8]Tan Y,Guo https://www.sodocs.net/doc/704163717.html,ing surface free energy method to study the cohesion and

adhesion of asphalt mastic.Constr Build Mater2013;47:254–60.

[9]Kandhal PS,Khatri MA.Relating asphalt absorption to properties of asphalt

cement and aggregate.Transp Res Rec1992:76–84.

[10]Luo R,Lytton R.Selective absorption of asphalt binder by limestone aggregates

in asphalt mixtures.J Mater Civ Eng2012;25(2):219–26.

[11]Doyle J,Howard I,Robinson W.Prediction of absorbed,inert,and effective

bituminous quantities in reclaimed asphalt pavement.J Mater Civ Eng 2012;24(1):102–12.

[12]Doyle JD,Howard IL.Factors affecting the measurement of effective speci?c

gravity for reclaimed asphalt pavement.Int J Pavement Res Technol2012;5(2): 128–31.

[13]Xu T,Wang H,Li Z,Zhao Y.Evaluation of permanent deformation of asphalt

mixtures using different laboratory performance tests.Constr Build Mater 2014;53:561–7.[14]Hou Y,Ji X,Su X,Zhang W,Liu https://www.sodocs.net/doc/704163717.html,boratory investigations of activated

recycled concrete aggregate for asphalt treated base.Constr Build Mater 2014;65:535–42.

[15]Pasandín AR,Pérez https://www.sodocs.net/doc/704163717.html,boratory evaluation of hot-mix asphalt containing

construction and demolition waste.Constr Build Mater2013;43: 497–505.

[16]ASTM C127,Standard Test Method for Density,Relative Density(Speci?c

Gravity),and Absorption of Coarse Aggregate,2012.

[17]ASTM C128.Standard Test Method for Density,Relative Density(Speci?c

Gravity),and Absorption of Fine Aggregate,2012.

[18]ASTM D4469,Standard practice for calculating percent asphalt absorption by

the aggregate in an asphalt pavement mixture,2011.

[19]West R,Dukatz E,Haddock J,Hall K,Kliewer J,Marek C,et al.A review of

aggregate and asphalt mixture speci?c gravity measurements and their impacts on asphalt mix design properties and mix acceptance.FHWA-HIF-11-033,Washington,DC:Of?ce of Pavement Technology;2010.

[20]Coombs SH.A density-gradient column for determining the speci?c gravity of

?sh eggs,with particular reference to eggs of the mackerel Scomber scombrus.

Mar Biol1981;63(1):101–6.

[21]Fortuin JMH.Theory and application of two supplementary methods of

constructing density gradient columns.J Polym Sci1960;44(144): 505–15.

[22]Hammonds RL,Stephens CP,Wills A,Hallman RL,Benson RS.Behavior of

polyethylene in a variable temperature density gradient column.Polym Test 2013;32(7):1209–19.

[23]TxDOT.Standard speci?cations for construction and maintenance of highways,

streets,and bridges.Austin,TX,2004.

188G.Liu et al./Construction and Building Materials72(2014)182–188

1-s2.0-S0950061814010216-main

相关文档

最新文档