Construction and Building Materials 48 (2013) 1009–1016

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier .com/locate /conbui ldmat

Laboratory evaluation of the effect of nano-organosilane anti-strippingadditive on the moisture susceptibility of HMA mixtures underfreeze–thaw cycles

0950-0618/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.conbuildmat.2013.07.030

⇑ Corresponding author. Address: Iran University of Science & Technology (IUST),Hengam Ave., Narmak 16844, Iran. Tel.: +98 21 77240399; fax: +98 21 77240398.

E-mail address: ameri@iust.ac.ir (M. Ameri).

Mahmoud Ameri a,⇑, Sareh Kouchaki a, Hossein Roshani b

a School of Civil Engineering, Iran University of Science and Technology (IUST), Tehran, Iranb Department of Civil Engineering, University of Guilan, Rasht, Iran

h i g h l i g h t s

� Addition of both anti-stripping additives significantly increased the ITS and TSR values of all asphalt mixtures.� Effects of anti-stripping additives on mixtures made with siliceous aggregate are more noticeable than limestone mixtures.� The fracture energy area under the load–deflection curve, for the Zycosoil-modified mixtures is higher than other mixtures.� Generally, influence of freeze–thaw cycling on resilient modulus and fracture energy was in accordance with TSR results.

a r t i c l e i n f o

Article history:Received 25 April 2013Received in revised form 14 July 2013Accepted 18 July 2013Available online 24 August 2013

Keywords:Moisture damageNano-organosilaneFourier Transform Infrared SpectroscopyResilient modulusIndirect tensile strengthTensile strength ratioFracture energy

a b s t r a c t

The objective of this research study was to evaluate moisture susceptibility of hot mix asphalt (HMA)with and without Zycosoil as a nano-organosilane anti-stripping additive and hydrated lime in the formof slurry. Moisture resistance was evaluated on mixtures prepared with two sources of aggregate: lime-stone and siliceous. The base bitumen used was AC 60–70 grade, and two anti-stripping additives. Thefunctional characterization of the base bitumen with and without Zycosoil was analyzed by means ofFourier Transform Infrared Spectroscopy (FTIR). The performance of HMA mixtures under multiplefreeze–thaw cycles was evaluated through the following tests: resilient modulus; indirect tensilestrength; tensile strength ratio and fracture energy. The findings of this research indicate that use of bothadditives enhanced the resilient modulus ratio of the mixtures. It was also observed that the effects ofanti-stripping additives on specimens made by siliceous aggregate are more pronounced than those pre-pared with limestone aggregates. Fracture energy results also proved that use of Zycosoil additive willincrease adhesion bond between the aggregates and asphalt binders, and in turn influences the moistureresistance of the mixture to moisture damage.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

The moisture damage of asphalt mixtures is defined as the pro-gressive loss of functionality of the material due to loss of theadhesive bond between the asphalt binder and the aggregate sur-face [1]. Penetration of moisture in asphalt mixtures reducesstrength and stiffness of asphalt mixtures and prone the mixturesto develop various forms of premature pavement distress. The pre-mature deterioration of asphalt pavements caused by moisture inasphalt mixtures includes stripping, raveling, and hydraulic scour.Additional distresses provoked by moisture in asphalt mixturesare: rutting, alligator cracking, and potholes [2]. The presence of

water in pavements can be detrimental if combined with otherenvironmental factors such as freeze–thaw cycling [3].

The chemical and physical interactions that occur betweenbitumen and aggregate at the interface are influenced by freeze–thaw cycling over extended periods of time. Among these interac-tions affected by freeze–thaw cycling are those influencing adhe-sive strength [4].

To alleviate or to control the deformations due to water dam-age, various researches were performed leading to the utilizationof anti-stripping additives. Anti-stripping additives are used toincrease physico-chemical bond between the bitumen and aggre-gate and also to improve wetting by lowering the surface tensionof the bitumen. The additives that are used in practice or testedin the laboratory include: (i) traditional liquid additives, (ii) metalion surfactants, (iii) hydrated lime and quick lime, (iv) silanecoupling agents, and (v) silicone. Among them, hydrated lime

1010 M. Ameri et al. / Construction and Building Materials 48 (2013) 1009–1016

and quicklime are the most commonly used solid type anti-strip-ping agents [5]. When lime is added to HMA, it reacts with aggre-gate and strengthens the bond between the bitumen and theaggregate interface. The ability of hydrated lime to make an as-phalt mix stiffer, tougher, and resistant to rutting, is a reflectionof its superior performance as active mineral filler [5].

It should be noted that an effective anti-stripping agent mustimprove both the unconditioned and moisture conditioned proper-ties of asphalt mixtures in order to ensure long term performanceof asphalt pavements [6].

Even though engineers are interested in the material propertiesat the macro- and meso-scales, the phenomena at nano- and mi-cro-scales provide fundamental insight for the underlying interac-tions defining the physico-chemical behavior of materials [7].

In the last decade or so, nanotechnology has emerged as the po-tential solution to greatly enhance the performance and durabilityof construction materials. Nanomaterials are defined as materialswith at least one dimension that falls in the length scale of 1–100 nm. Due to the small size and high surface area, the propertyof nanomaterials is much different from normal size materials[8]. Therefore, research engineers tried to apply the nanomaterialsinto the asphalt technology. Some research showed that fatiguecracking, rutting and moisture resistance of asphalt binders andmixtures are improved with the addition of nanomaterials [6–9].

Arabani in 2012 focused on evaluating the moisture susceptibil-ity of warm mix asphalt (WMA) with and without Zycosoil as anano-organosilane additive by determining the micromechanismsthat affect the adhesion bond between the aggregates and asphaltbinders and the cohesion strength of the asphalt binder. The resultsof the surface free energy (SFE) method indicate that addition ofZycosoil into asphalt binders of WMA mixtures increase the sur-face energy of adhesion between aggregate and asphalt in dryand wet condition. Thus, the energy released in mixes modifiedby Zycosoil is lower than other mixtures, and therefore their strip-ping resistance is higher [9]. In another study conducted byMoghadas Nejad et al. it was shown that addition of Zycosoil inhot mix asphalt containing limestone and granite aggregates willincrease tensile strength ratio (TSR) as compared to the controlmixtures [6].

In recent years, several studies focusing on use of nano-scaleorganosilane as antistrip additives in asphalt mixtures have beenconducted; however, the effects of these additives on moisturesusceptibility of asphalt mixture under freeze–thaw cycles are stillnot fully understood [9].

2. Materials

2.1. Aggregate

In this research study, two types of widely used aggregates in the countrynamely siliceous and limestone were used for evaluation of asphalt mixturesagainst moisture damages. These aggregates represent a substantial range in min-eralogy and associated degree of stripping. The chemical compositions are deter-mined with an electron microprobe analyzer (EMPA). The results of chemicalcompositions are given in Table 1. The physical and engineering properties of the

Table 1Chemical composition of two types of aggregates with electron microprobe analyzer.

Properties Siliceous Limestone

Calcium oxide, CaO (%) 4.2 51.3Aluminum oxide, Al2O3 (%) 13.6 1Ferric oxide, Fe2O3 (%) 1.5 0.4Magnesium oxide, MgO (%) 1.7 1.2R2O3 (Al2O3 + Fe2O3) (%) 17.4 18Silicon dioxide, SiO2 (%) 61.2 3.8PH 6.9 8.8

aggregates are shown in Table 2. The gradation of the aggregates employed in thestudy (mean limits of ASTM specifications for dense aggregate grading) is presentedin Table 3. The nominal size of this grading was 19.0 mm.

2.2. Asphalt binder

In this experimental investigation, an asphalt binder of 60/70 penetration gradefrom Isfahan mineral oil refinery was used. To characterize the properties of the as-phalt binder, conventional tests including the penetration, softening point and duc-tility were carried out. Table 4 summarizes the basic properties of the asphaltbinder.

2.3. Additives

Zycosoil is a water soluble reactive organo-silicon compound that is speciallydesigned to improve the adhesion between bitumen and aggregates [9]. In thisstudy, as recommended by the manufacturer, Zycosoil was added in quantity equalto 0.5% by weight of the asphalt binder. The properties of the Zycosoil are presentedin Table 5.

2.4. Hydrated lime

Lime slurry was added to aggregates in an attempt to improve asphalt mixtureresistance to moisture damage, whereby optimum content of hydrated lime addedto aggregate was obtained after preliminary tests. The percent of hydrated lime thatshowed the best strength against moisture damage was 1% by weight of aggregatethat was selected as optimum content of hydrated lime.

2.5. Sample preparation and test methods

The asphalt mixtures were designed by using the standard marshal mix designprocedure with 75 blows on each side of cylindrical samples in accordance withASTM D1559 standard [10]. Marshall tests were carried out to find the optimum as-phalt binder content of mixtures. The optimum asphalt binder content of asphaltmixture with siliceous and limestone aggregate was 5.3% and 4.8%, respectively.

2.6. Fourier Transformed Infrared Spectroscopy (FTIR)

From the infrared spectra of test, the material information about chemicalbonding and materials structure would be obtained [11]. Zycosoil were applied tomodify the base bitumen, and the FTIR test were employed to determine the func-tional characteristics of bitumens in wave numbers ranging from 4000 to 400 cm�1

[12]. The FTIR test was implemented in the central Chemical Engineering ResearchLaboratory of the country.

2.7. Resilient modulus test

The resilient modulus is directly affected by the loss of adhesion and cohesion[13]. It is believed that resilient modulus is more sensitive to changes in asphaltbinder properties and a mixture’s sensitivity to damage by water than the tensilestrength of the mixture [14]. In this study, the ratio of resilient modulus test inwet-to-dry conditions was used to assess the effect of additives on the rate of strip-ping in HMA. This test is a non-destructive test, and in this research, it is undertakenin accordance with the ASTM D 4123 [10] using an assumed Poisson’s ratio of 0.35.The samples were tested with Universal Testing Machine (UTM) at 25 �C by apply-ing a haversine load consists of 0.1-s load application followed by a 0.9-s rest per-iod. The sample preconditioning of resilient modulus test was similar to ITS test.The load and deformation were continuously recorded and resilient modulus werecalculated using the following equation:

Mr ¼Pðv þ 0:27Þðt � DhÞ ð1Þ

where P is the peak value of the applied vertical load (N), Dh is the mean amplitudeof the horizontal deformation obtained from five applications of the load pulse(mm), t is the mean thickness of the test specimen (mm), and v is the Poisson’s ratio(assumed 0.35) [15–17].

The ratio of resilient modulus is used for evaluating of the moisture susceptibil-ity of asphalt mixtures. The resilient modulus ratio is defined by the followingequation:

Resilient Modulus Ratio ðMRRÞ ¼ MR of conditioned specimenMR of dryðcontrolÞspecimen

ð2Þ

where MR is the resilient modulus of mixtures [13]. A resilient modulus ratio (MRR)of 70% or above has typically been utilized as a minimum acceptable value for HMAmixtures [1,13].

Table 2Physical and engineering properties of the aggregate.

Test Standard Siliceous Limestone Specification

Specific gravity (coarse agg.) ASTM C 127Bulk 2.649 2.612 –SSD 2.671 2.643 –Apparent 2.689 2.659 –

Specific gravity (fine agg.) ASTM C 128Bulk 2.669 2.618 –SSD 2.652 2.633 –Apparent 2.692 2.651 –

Specific gravity (filler) ASTM D854 2.663 2.640 –Los Angeles abrasion (%) ASTM C 131 20.1 25.6 Max 45Flat and elongated particles (%) ASTM D 4791 6.8 9.2 Max 10Sodium sulfate soundness (%) ASTM C 88 1.7 2.56 Max 10–20Fine aggregate angularity ASTM C 1252 57.5 46.65 Min 40

Table 3Gradation of the aggregates used in the study.

Sieve (mm) 19 12.5 4.75 2.36 0.3 0.25 0.106 0.075Lower–upper limits 100 90–100 44–74 28–58 5–21 2–10 7–18 4–10Passing (%) 100 88 75 56 40 20 9 6

Table 4Results of the experiments conducted on 60/70 penetration grade asphalt binder.

Test Standard Result

Penetration (100 g, 5 s, 25 �C), 0.1 mm ASTM D5-73 64Penetration (200 g, 60 s, 4 �C), 0.1 mm ASTM D5-73 23Penetration ratio ASTM D5-73 0.36Ductility (25 �C, 5 cm/min), cm ASTM D113-79 112Solubility in trichloroethylene (%) ASTM D2042-76 98.9Softening point (�C) ASTM D36-76 51Flash point (�C) ASTM D92-78 262Loss of heating (%) ASTM D1754-78 0.75Properties of the TFOT residuePenetration (100 g, 5 s, 25 �C), 0.1 mm ASTM D5-73 60Specific gravity at 25 �C (g/cm3) ASTM D70-76 1.020Viscosity at 135 �C, cSt ASTM D2170-85 158.5

Table 5Properties of the Zycosoil.

Properties Zycosoil

Form SolidColor Colorless–pale yellow yellowFlash point 80 �CViscosity (at 25 �C) 0.2–0.8 Pa s

Fig. 1. Typical load–deformation curve for calculation of fracture energy at onset offailure [15].

M. Ameri et al. / Construction and Building Materials 48 (2013) 1009–1016 1011

2.8. Indirect tensile strength test

The tensile strength of an HMA mixture is generated by the cohesive strength ofthe asphalt binder and the bond strength at the binder–aggregate interface [6]. Theinfluence of water in the reduction of the strength of HMA mixtures has been re-ported as one of the main issues that need to be evaluated for moisture damage[6,18]. Thus, in this study the tensile strength of five subset of specimens com-pacted with 7.0 ± 0.5% air voids were obtained using the indirect tensile strength(ITS) test according to AASHTO T283[19]. The purpose of this test is to determinethe effect of saturation and accelerated water conditioning on the indirect tensilestrength (ITS) of cylindrical specimen [20].

The first subset was tested in a dry condition, and The other four subsets subjectto vacuum saturation (55–80% of the air voids volume), followed by one freeze–thaw (F–T) cycle, two F–T cycles and up to four F–T cycles, respectively, prior tobeing tested. In the ITS test, a compressive load applied to the cylindrical speci-mens, which act parallel to the vertical diametral plane by using the Marshall load-ing equipment. This type of loading produces a relatively uniform tensile stress,which acts perpendicular to the applied load plane. The maximum load that thesamples can sustain prior to cracking and also the deformations of them were con-tinuously recorded.

Parameters from the indirect tensile strength test that have been correlated toactual cracking values include indirect tensile strength, tensile strength ratio (TSR),and fracture energy to failure. These indirect tensile strength parameters are de-fined as follows:

1. The indirect tensile strength (St) is the maximum stress developed at the centerof the specimen in the radial direction during the loading by the followingequation:

St ¼2� Pmax

p� d� tð3Þ

where St, tensile strength of conditioned or unconditioned specimens, kPa; Pmax,maximum applied load, KN; t, thickness of the specimens, meter; d, diameter ofthe specimens, m [21].1. A tensile strength ratio (TSR) which is the ratio of the average tensile strength of

the conditioned specimens (1 F–T, 2 F–T, 3 F–T and 4 F–T) to the average tensilestrength of the unconditioned specimens was then calculated. The TSR repre-sents a reduction in the mixture integrity due to moisture damage. A minimumratio of 80% has been typically used for TSR as a failure criterion [22].

2. The energy at onset of failure is calculated from the results of the ITS test asshown in Fig. 1 [15].

3. A higher ITS and TSR values typically indicate that the mixture will performwell with a good resistance to moisture damage. At the same time, mixturesthat are able to tolerate higher deformation prior to failure are more likely toresist cracking than those unable to tolerate high deformation [23].

1012 M. Ameri et al. / Construction and Building Materials 48 (2013) 1009–1016

3. Test results and discussion

3.1. Fourier Transformed Infrared Spectroscopy (FTIR) analysis

Figs. 2 and 3 provide Infrared spectroscopy analysis mastercurves of the Zycosoil modified and original base asphalt binderat 25 �C.

According to the report by Bukka, the bands at 2923 and1376 cm�1 are characteristic of C–H asymmetrical stretching andC–H deformation in –CH3, respectively; the two bands at 2852and 1460 cm�1 are associated with C–H asymmetrical stretchingand C–H deformation in –CH2, respectively [24]. From the infraredspectroscopy data, peaks in the region of 1710–1690 cm�1 corre-sponding to carbonyl stretch were used to characterize bitumen.This region is characteristic of functional groups like carboxylicacids, 2-quinolones, anhydrides and ketones, some of which couldbe crucial as regards to moisture damage. Plancher in 1977 foundthat carboxylic acids in the bitumen were strongly adsorbed bysiliceous aggregates, but were present in very small amounts. Atthe same time, carboxylic acids tend to be displaced first fromthe aggregate in presence of moisture [25]. The positions of bandsin the curves of Zycosoil modified binder and original base binderare nearly the same. The differences is the stretching vibrationband of Si–OH and Si–O appears at 3360, 1038 cm�1.

3.2. Resilient modulus result

Figs. 4 and 5 present the resilient modulus results for the dryand conditioned HMA mixtures from the two aggregates sourcesat 25 �C. In order to more accurately assess the effect of lime slurryand Zycosoil on the resistance of asphalt mixtures to moisturedamages, four freeze–thaw cycles are applied to the samples. Asseen in Figs. 4 and 5, generally, resilient modulus decreased afterfreeze–thaw (F–T) conditioning for all mixtures. Siliceous controlmixtures were deteriorated at the third F–T cycle but limestonecontrol mixtures were lasted until the fourth F–T cycle.

The results show that mixtures with siliceous aggregate havehigher MR values than mixtures with limestone aggregate in drycondition. While mixtures with limestone aggregate show higherMR values than mixtures with siliceous aggregate after F–T cycles.Therefore, when siliceous mixtures are not under wet conditions,proper bond between asphalt and siliceous aggregate will be estab-lished. However when they are subjected to F–T cycles they expe-rience a sharp reduction in resilient modulus value. In addition,Figs. 4 and 5 show that the MR after moisture conditioning is signif-icantly improved with the addition of lime slurry and Zycosoil forboth aggregate types. Mixtures with anti-stripping additivesexhibited a less decrease in MR value than specimens withoutanti-stripping additive after F–T conditioning.

Fig. 2. FTIR test result of

Siliceous aggregate surfaces have high concentrations of hydro-xyl groups that have high affinity for both carboxylic acids andwater [26]. In the situations where siliceous asphalt mixtures areexposed to moisture, water forms a stronger hydrogen bond withthe silanol group than the hydrogen bond between the carboxylicacids of the binder and the silanol group. As a result, water will mi-grate to the interface between the asphalt and aggregate weaken-ing the adhesion bond between them leading to the separation ofbitumen from aggregates.

As depicted in Fig. 5, the resilient modulus of the siliceous mix-ture containing Zycosoil relative to control mixture was greaterthan that containing lime slurry.

The results of the FTIR test performed on asphalt binder modi-fied with Zycosoil are shown in Fig. 3, indicates the points relatedto the Si–O bonds in organosilane compounds of the Zycosoil. Inthe presence of water, this bonds turn into Si–OH silanoles, andthrough reacting with silanols on aggregate surface creates theSi–O–Si film structure over the surface. Since the covalent bondsare stronger than hydrogen bonds, the Si–O–Si film structure actsas a hydrophobic layer and covers the aggregates surface [27]. Fi-nally, this hydrophobic layer avoids water penetration and pre-vents the establishment of hydrogen bonds between water andaggregate surface at binder–aggregate interfaces.

3.3. Resilient modulus ratio (MRR)

Fig. 6 shows the change of resilient modulus ratio versus thenumber of F–T cycles. The MRR was used to evaluate the propen-sity of the mixtures to stripping potential. Mixtures with resilientmodulus ratios of less than 70% are considered prone to strippingpotentials [1,13]. The siliceous control mixtures did not performas adequately as the limestone control mixtures did in thefreeze–thaw cycles. The MRR values of all of the siliceous controlmixtures were less than 70% and thus these mixtures are moresensitive to moisture damage. Fig. 6b shows that Zycosoil and limeslurry have enhanced the MRR of siliceous mixtures in all freeze–thaw processes. As can be seen in Fig. 6a and b both siliceousand limestone mixtures with Zycosoil additives have better resis-tance during freeze–thaw cycles than the control mixtures and alsothe mixtures containing lime slurry.

3.4. Indirect tensile strength (ITS) result

The ITS and TSR test are often used to evaluate the moisturesusceptibility of an asphalt mixture [23]. A higher ITS and TSR val-ues typically indicate that the mixture will have better resistanceto moisture damage. At the same time, mixtures that are able totolerate higher strain prior to failure are more likely to have betterresistance to environmental cracking [23,28].

original base binder.

Fig. 3. FTIR test result for Zycosoil modified binder.

Fig. 4. Resilient modulus for mixtures with limestone aggregate in dry conditionand under different freeze–thaw cycles.

Fig. 5. Resilient modulus for mixtures with siliceous aggregate in dry condition andunder different freeze–thaw cycles.

Fig. 6. Impact of freeze–thaw cycles on the resilient modulus ratio (a) Mixtureswith limestone aggregate and (b) Mixtures with siliceous aggregate.

Fig. 7. ITS results for mixtures with limestone aggregate in dry condition and underdifferent freeze–thaw cycles.

M. Ameri et al. / Construction and Building Materials 48 (2013) 1009–1016 1013

Figs. 7 and 8 show the indirect tensile strength of specimens. Asshown in these figures, F–T cycles have reduced the indirect tensilestrengths of the mixtures, specifically the mixture containing sili-ceous aggregate. Siliceous control mixtures were deteriorated atthe third F–T cycle but limestone control mixtures could resistthe detrimental F–T effects, indicating limestone aggregate mix-ture has better resistant to moisture damage relative to siliceousaggregate.

According to Figs. 2 and 3, the FTIR test shows the existence ofcarboxylic acids in the virgin base bitumen and also modified bin-der. During the mixing of binder with aggregate, carboxylic acid isquickly absorbed by aggregates [27]. At binder–aggregate inter-face, the bonds between carboxylic acids and Si–OH compoundson the aggregate surface are unstable against water [27]. The

Fig. 8. ITS results for mixtures with siliceous aggregate in dry condition and underdifferent freeze–thaw cycles.

Fig. 10. Impact of freeze–thaw cycles on the TSR values of mixtures with siliceousaggregate.

1014 M. Ameri et al. / Construction and Building Materials 48 (2013) 1009–1016

reduction in the indirect tensile strengths of the samples byincreasing number of F–T cycles can be attributed to the loss ofadhesion bond between the aggregates and the asphalt bindersand also the loss of cohesive strength of the asphalt binder dueto F–T cycles effects.

Based on the South Carolina Department of Transportation(SCDOT) criterion, moisture resistant mixtures should have wetITS value more than 448 kPa. All of the control siliceous and lime-stone aggregates mixtures failed the SCDOT requirement under theF–T conditioning with the exception of the limestone samples inthe first F–T cycle.

As can be seen from Figs. 7 and 8, addition of both anti-strippingadditives significantly increased the indirect tensile strengths val-ues of all asphalt mixtures. It is noteworthy that the effects of anti-stripping additives on specimens made by siliceous aggregate aremore pronounced than those on mixture made by limestoneaggregate.

From Figs. 7 and 8, it is evident that addition of Zycosoil to thelimestone mixtures improves the indirect tensile strengths of themix more than the lime slurry. However, test results demonstratedthat the influence of lime slurry on increasing the indirect tensilestrengths of siliceous mixtures was higher than Zycosoil until thesecond F–T cycle. In terms of ITS, the Zycosoil showed better effectthan lime slurry at the third and fourth F–T cycle.

3.5. Tensile strength ratio (TSR)

TSR of wet group to dry group specimens was computed fromthe results of the indirect tensile strength test at 25 �C. The higherthe TSR value, the higher the resistance of the mixture to moisturedamage.

According to Fig. 9, in the case of control limestone mixtures inF–T conditioning, it was observed that just in the first F–T cycle theTSR values of the mixtures were above 80%. While for all controlsiliceous mixtures in F–T conditioning, the TSR values which areshown in Fig. 10 were extremely below 80%. From these findings,

Fig. 9. Impact of freeze–thaw cycles on the TSR values of mixtures with limestoneaggregate.

it can be concluded that the siliceous aggregates are more suscep-tible to moisture damage than the limestone aggregates.

According to the research conducted by Little and Jones in 2003[29], hydrated lime ties with carboxylic acids in the bitumen, withthe formation of insoluble calcium organic salts, which preventthese functionalities from reacting with a siliceous surface to formwater sensitive bonds. This leaves important active sites on thesiliceous surface to form strong water resistant bonds with nitro-gen groups in bitumen.

Besides Zycosoil produces a hydrophobic nano-layer on aggre-gates since it converts the hydrophilic silanol groups to hydropho-bic siloxane groups. So Zycosoil can eliminate water sensitivesurface permanently and makes it oil-loving [6].

The test results of the mixtures studied in this research indi-cated that the effect of both anti-stripping additives on increasingthe TSR values of specimens was similar. By taking both ITS andTSR results into account, although, the differences between the ef-fects of Zycosoil and lime slurry are not significant, but the resultsconfirm the tendency of the Zycosoil additive to increase the mois-ture damage resistant of limestone mixtures.

3.6. Fracture energy result

This stage of analysis focused on computing the fracture resis-tance of modified and unmodified HMA mixtures exposed to accel-erate moisture damage process based on the load–displacementcurves obtained from ITS test which presented in Fig. 1. Figs. 11and 12 show the fracture energy of HMA specimens fabricatedwith two aggregate types and two kinds of anti-stripping additives.

3.6.1. Limestone mixturesAccording to ITS test results presented in Fig. 11, it was found

that the average mean dry fracture energy values of limestone con-trol mixtures were not higher than the modified limestone mix-tures. Comparison of wet fracture energy values indicated that

Fig. 11. Fracture energy for mixtures with limestone aggregate before and aftermoisture conditioning.

Fig. 12. Fracture energy for mixtures with siliceous aggregate before and aftermoisture conditioning.

Fig. 13. load–displacement curve for mixtures with limestone aggregate after firstF–T cycle.

M. Ameri et al. / Construction and Building Materials 48 (2013) 1009–1016 1015

the mean wet fracture energy values of all Zycosoil modified lime-stone mixtures were greater than the control limestone mixturesand mixtures containing lime slurry.

The area under the load–deflection curve for the Zycosoil-mod-ified mixtures is also higher than control mixture and mixturescontaining lime slurry.

As can be seen in Table 1, the limestone aggregate has smallpercentage of Silica (SiO2). In the case of control limestone mix-tures, a small percentage of bitumen and aggregate bonding func-tional groups are sensitive to moisture during F–T conditions [26].As a result, the fracture energy values of these specimens in wetconditions are somewhat lower than the fracture energy valuesof these specimens in dry condition. In addition, according to theresults, it appears that the adhesion between binder and aggregatein the mixtures containing additives, particularly the specimensmodified by Zycosoil, is higher than the mixtures without anti-stripping additives. Thus, when a mixture is modified with Zyco-soil, more energy is needed to break the bonds between aggregateand bitumen in both dry and wet conditions. Therefore it can beconcluded that if an asphalt mixture sustains more load and alsomore deformation at the onset of failure it will have more resis-tance to moisture damage.

Fig. 14. load–displacement curve for mixtures with siliceous aggregate after first F–T cycle.

3.6.2. Siliceous mixturesFrom Figs. 11 and 12, it is clear that the mixtures prepared with

the siliceous aggregate have higher fracture energy than the mix-tures prepared with the limestone aggregate. The siliceous controlmixtures had lower fracture energy values in the wet conditionthan modified siliceous mixtures. Thus it is deduced that use ofadditives has increased fracture energy values of siliceous mixturesfollowed by F–T conditions. Although the indirect tensile strengthsof asphalt mixtures containing Zycosoil additive in the first andsecond cycles of F–T were somewhat lower than asphalt mixturescontaining lime slurry, but conversely, the fracture energy of mix-tures containing Zycosoil were higher than those mixtures contain-ing lime slurry.

Siliceous aggregates predominantly comprised of SiO2 and pro-duce unstable bonds with bitumen against water. Upon exposureto moisture conditions, water is rapidly absorbed by the siliceousaggregate. Due to existence of the high percentage of hydroxylson the surface of siliceous aggregate, the Zycosoil has a very goodeffect in combination with siliceous aggregate.

As can be seen from Figs. 11 and 12, the fracture energy valuesof both modified and unmodified mixtures decrease as the numberof F–T cycles increases. However, according to Figs. 13 and 14, atthe same F–T cycle, the Zycosoil-modified mixture always hashigher fracture energy as compared to the mixtures that are mod-ified with or without slurry lime. Also from Figs. 11 and 12 it is de-duced that use of Zycosoil will produce a more cohesive binder

with a better adhesion to aggregate surface, this is evident fromthe higher residual strength and also higher displacement valueat maximum load for specimens containing Zycosoil.

The fracture energy obtained from the area under the load–dis-placement curve is the amount of work required for breaking thesample. In fact, the failure of the specimen in the indirect tensilestrength test depends on two parameters namely: friction, andadhesion of mixture. Since the fracture energy of the Zycosoil mix-tures is higher than those mixtures modified with hydrated limeslurry, it can be deduced that the Zycosoil additive improves theadhesion between binder and aggregate and thus enhanced theresistance of the mixture to the moisture damage.

4. Conclusion

To investigate the moisture damage susceptibility of asphaltmixtures under freeze–thaw cycles, two anti-stripping additivesnamely Zycosoil and hydrated lime slurry with two different kindsof aggregate and one grade of bitumen were used. Control mixtureswere produced with AC60–70 using the same aggregate sources andgradation, for comparison purposes. Various laboratory tests wereconducted to evaluate the characteristics of using Zycosoil as anano-organosilane additive, and hydrated lime slurry in the hotmix asphalt under freeze–thaw cycles. The following conclusions

1016 M. Ameri et al. / Construction and Building Materials 48 (2013) 1009–1016

were drawn based upon the experimental results obtained from thisresearch study:

� In the resilient modulus test, although mixtures with siliceousaggregate have higher MR values than mixtures with limestoneaggregate in dry condition, but siliceous control mixtures werefailed in the fourth F–T cycle whereas limestone control mix-tures lasted until the fourth cycle.� Use of anti-stripping additives improves the MRR of the mix-

tures. Both Zycosoil and lime slurry have obvious effects onMRR for siliceous mixtures in all freeze–thaw cycles.� Addition of both anti-stripping additives significantly increased

the indirect tensile strength values and tensile strength ratio ofall asphalt mixtures. It is worthy to note that the effects of anti-stripping additives on mixtures made with siliceous aggregateare more noticeable than those mixtures made with limestoneaggregates.� Regardless of the aggregate source, the mean wet ITS values of

all the Zycosoil modified mixtures were higher than the controlmixtures and mixtures containing hydrated lime slurry. Alsothe deflection of modified mixtures increased due to the addi-tion of Zycosoil. Consequently, the area under the load–deflec-tion curve (fracture energy) for the Zycosoil-modifiedmixtures is also higher than other mixtures.� Based on fracture energy results, the fracture energy of the

Zycosoil mixtures is higher than hydrated lime slurry mixtures.� In the light of current research findings from laboratory investi-

gations conducted in this research study, it is found that addi-tion of Zycosoil additive improves HMA mixture resistance tomoisture damage.� Generally, influence of freeze–thaw cycling on resilient modu-

lus and fracture energy was in accordance with TSR results. Itcan be concluded that these tests are useful for evaluation ofmoisture sensitivity of hot mix asphalt.� It is recommended to conduct a field study to investigate the

performance of HMA mixtures modified by Zycosoil.

References

[1] Kiggundu BM, Roberts FL. Stripping in HMA mixtures: state-of-the-art andcritical review of test methods. National Center for Asphalt Technology, ReportNo. 88–02, Auburn Univ., Auburn, GA; 1988.

[2] Caro S, Masad E, Airey G, Bhasin A, Little D. Probabilistic analysis of fracture inasphalt mixtures caused by moisture damage. Transp Res Rec, J Transport ResBoard 2008;2057:28–36.

[3] Breakah TM, Bausano JP, Williams RC. Integration of moisture sensitivitytesting with gyratory mix design and mechanistic-empirical pavement design.J Transport Eng 2009;135(11):852–7.

[4] Huang S, Robertson RE, Branthaver JF. Physico-chemical characterization ofasphalt–aggregate interactions under the influence of freeze–thaw cycles.Annual Meeting of the TRB; 2003.

[5] Gorkem C, Sengoz B. Predicting stripping and moisture induced damage ofAsphalt concrete prepared with polymer modified bitumen and hydrated lime.Constr Build Mater 2009;23(6):2227–36.

[6] Moghadas Nejad F, Azarhoosh AR, Hamedi GH, Azarhoosh MJ. Influence ofusing nonmaterial to reduce the moisture susceptibility of hot mix asphalt. JConstr Build Mater 2012;31:384–8.

[7] Goh SW, Akin M, You Z, Shi X. Effect of deicing solutions on the tensile strengthof micro-or nano-modified asphalt mixture. J Constr Build Mater2011;25(1):195–200.

[8] Yao H, You Z, Li L, Goh SW, Lee CH, Yap YK, et al. Rheological properties andchemical analysis of nanoclay and carbon microfiber modified asphalt withFourier transform infrared spectroscopy. J Constr Build Mater2013;38:327–37.

[9] Arabani M, Roshani H, Hamedi GH. Estimating moisture sensitivity of warmmix asphalt modified with Zycosoil as an antistrip agent using surface freeenergy method. ASCE. J Mater Civil Eng 2012;24(7):889–97.

[10] ASTM D1559-89, Standard test method for resistance to plastic flow ofbituminous mixtures using Marshall apparatus. American Society for Testingand Materials; 1994.

[11] Griffiths P, De Haseth JA. Fourier transform infrared spectrometry. John Wileyand Sons; 2007.

[12] Zhang F, Yu J. The research for high-performance SBR compound modifiedasphalt. J Constr Build Mater 2010;24(3):410–8.

[13] Heinicke JJ, Vinson TS. Effect of test condition parameters on IRMr. J TransportEng 1988;114:153–72.

[14] Epps JA, Sebaaly PE, Penaranda J, Mather MR, Mccann MB, Hand AJ.Compatibility of a test for moisture-induced damage with superpavevolumetric mix design. NCHRP report no.508, Transportation ResearchBoard, National Research Council; 2000.

[15] Khodary FMH. Evaluation of fatigue resistance for modified asphalt concretemixtures based on dissipated energy concept. Department of Civil Engineeringand Geodesy Technische Universität Darmstadt; 2010.

[16] VuralKok B, Yilmaz M. The effects of using lime and styrene–butadiene–styrene on moisture sensitivity resistance of hot mix asphalt. J Constr BuildMater 2009;23(5):1999–2006.

[17] Xie J, Wu Sh, Pang L, Lin J, Zhu Z. Influence of surface treated fly ash withcoupling agent on asphalt mixture moisture damage. J Constr Build Mater2012;30:340–6.

[18] Silva HMRD, Oliveira JRM, Peralta J, Zoorob SE. Optimization of warm mixasphalts using different blends of binders and synthetic paraffin wax contents.J Constr Build Mater 2010;24(9):1621–31.

[19] AASHTO:T283-03, Standard method of test for resistance of compacted asphaltmixture to moisture induced damage. American Association of State Highwayand Transportation Officials.

[20] Poulikakos LD, Partl MN. Evaluation of moisture susceptibility of porousasphalt concrete using water submersion fatigue tests. J Constr Build Mater2009;23(12):3475–84.

[21] Aksoy A, Samlioglu K, Tayfur S, Ozen H. Effects of various additives on themoisture damage sensitivity of asphalt mixtures. J Constr Build Mater2005;19(1):11–8.

[22] Boyes AJ. Reducing moisture damage in asphalt mixes using recycled wasteadditives. Master of Science thesis, California Polytechnic State University, SanLuis Obispo; 2011.

[23] Xiao F, Amirkhanian SN. Laboratory investigation of moisture damage inrubberized asphalt mixtures containing reclaimed asphalt pavement. Int JPavement Eng 2009;10(5):319–28.

[24] Bukka K, Miller JD, Oblad AG. Fractionation and characterization of Utah tarsands bitumens: influence of chemical com position on bitumen viscosity.Energy Fuels 1991;5:333–40.

[25] Bagampadde U, Isacsson U, Kiggundu BM. Influence of aggregate chemical andmineralogical composition on stripping in bituminous mixtures. J PavementEng 2005;6(4):229–39.

[26] Huang S, Robertson RE, Branthaver JF, Petersen JC. Impact of lime modificationof asphalt and freeze–thaw cycling on the asphalt–aggregate interaction andmoisture resistance to moisture damage. ASCE J Mater Civil Eng2005;17(6):711–8.

[27] Hefer A, Little D. Adhesion in bitumen–aggregate systems and quantificationof the effects of water on the adhesive bond. Report No. ICAR/505-1, TexasTransportation Institute, The Texas A&M University, System; 2005.

[28] Xiao F, Amirkhanian SN. Effects of liquid antistrip additives on rheology andmoisture susceptibility of water bearing warm mixtures. J Constr Build Mater2010;24(9):1649–55.

[29] Little D, Jones D. Chemical and mechanical mechanisms of moisture damage inhot mix asphalt pavements. National Seminar in Moisture Sensitivity, SanDiego, California; 2003.