Performance of Glass Wool Fibers in Asphalt Concrete Mixtures

materials

Article Performance of Glass Wool Fibers in Asphalt Concrete Mixtures

Agathon Honest Mrema 1 , Si-Hyeon Noh 1, Oh-Sun Kwon 2 and Jae-Jun Lee 1,*

1 Department of Civil Engineering, Jeonbuk National University, Jeonju-si, Jeollabuk-do 54896, Korea; [email protected] (A.H.M.); [email protected] (S.-H.N.) 2 Principal Researcher, Korea Expressway Corporation Research Center, 208–96 Dongbu-daero 922beon-gil, Dongtan-myeon, Hwaseong-si, Gyeonggi-do 39660, Korea; [email protected] * Correspondence: [email protected]

Received: 23 September 2020; Accepted: 19 October 2020; Published: 22 October 2020

Abstract: Nowadays, in order to improve asphalt pavement performance and durability and reduce environmental pollution caused by hydrocarbon materials, many researchers are studying different ways of modifying asphalt concrete (AC) and finding alternative paving materials to extend the service life of pavements. One of the successful materials used in the modification of AC is fibers. Different types of fibers have been reinforced in AC mixtures and improvements have been observed. This research studies the performance of glass wool fibers reinforced in a dense-graded asphalt mixture. Generally, glass fibers are known to have excellent mechanical properties such as high tensile modulus, 100% elastic recovery and a very high tolerance to heat. Glass wool fibers are commonly used as a thermal insulation material. In this research, to evaluate the performance of glass wool fibers in AC, laboratory tests, the Marshall mix design test, indirect tensile strength (IDT), tensile strength ratio (TSR) and the Kim test were conducted to determine a proper mix design, tensile properties, moisture susceptibility, rutting and fatigue behaviors. Results show that the addition of glass wool fibers does affect the properties of AC mixtures. The use of glass wool fibers shows a positive consistence result, in which it improved the moisture susceptibility and rutting resistance of the AC. Additionally, results show that the addition of fiber increased tensile strength and toughness which indicates that fibers have a potential to resist distresses that occur on a surface of the road as a result of heavy traffic loading. The overall results show that the addition of glass wool fibers in AC mixtures is beneficial in improving properties of AC pavements.

Keywords: asphalt concrete; glass wool ﬁbers; indirect tensile strength; Kim test; Marshall test; tensile strength ratio

1. Introduction Asphalt concrete (AC), also known as hot-mix asphalt (HMA), is a combination of asphalt binder and aggregates mixed together at a high temperature. The AC is then placed and compacted on the road while still hot [1]. Asphalt pavement is the predominant pavement type in the world, and it is used for all types of applications from residential streets to expressways, from parking lots to harbor facilities and from bike paths to airport runways [1,2]. AC can be designed to form different types of mixtures depending on what the designer wants to achieve in terms of satisfactory performance over traffic and climate conditions of the designated region and durability of a pavement structure throughout the whole designed life expectancy [3,4]. After being applied to a road, AC mixtures tend to have many distresses such as cracking, rutting (permanent deformation), stripping (separation of aggregates from the AC mixture) and potholes. These distresses are caused by heavy traffic loading and harsh environment or weather conditions.

Materials 2020, 13, 4699; doi:10.3390/ma13214699 www.mdpi.com/journal/materials Materials 2020, 13, 4699 2 of 14

The occurrence of distresses on road pavements may lead to a total failure of the pavement structure. Road engineers first and foremost have to ensure that they are avoiding all distresses at most, and to do so for the past few decades, engineers and material scientists have been involved in the process of modifying the traditional AC by adding other materials in it so as to improve its performance and durability and reduce costs while also trying to reduce environment pollution caused by the asphalt binder [5,6]. Different materials have been mixed together with asphalt binder and aggregates, but the most common ones are fibers and polymers [7]. Fibers, when added in asphalt mixture, have two main purposes which are to prevent drain down for gap- and open-graded asphalt mixtures and to increase strength and stability (reduce rutting) and improve resistance to cracking for dense-graded asphalt mixtures [7,8]. Cellulose (plant-based) and mineral fibers are used in AC to prevent drain down since they have high absorption capabilities [7,9,10]. Polypropylene, aramid, polyester and glass fibers have high tensile strength; therefore, they are used to add more strength to AC [7,11]. The process of adding fibers in either asphalt or cement concretes can be categorized into two categories: direct random inclusion (matrix), of which the outcome in AC is known as fiber-reinforced asphalt concrete, and the geo-synthetics family [11]. Polypropylene, cellulose and polyester fibers are the most prevalent fibers used in AC mixtures. Polypropylene fibers are preferred due to their low cost [12], strong bonding with asphalt, reduced cracking and shoving [7,13] and improved fatigue life [14]. Cellulose fibers are relatively inexpensive, widely available and stabilize the binder in gap- and open-graded stone matrix asphalt mixtures [15]. Polyester fibers resist cracking, rutting and potholes [7] and increase the stability and strength of the asphalt mixture [16]. This research studies a glass wool fiber reinforced in AC and its applicability. Glass wool is an insulating material made from fibers of glass arranged using a binder into a texture similar to wool [17]. Glass wool fibers have good flexibility, are easy to install, have high thermal resistance with low thermal conductivity (working temperature at about 450 ◦C) and are a fire-safe and environmentally friendly material [18]. Glass wool fibers are not commonly used in AC mixtures. Glass fibers, on the other hand, have been reinforced in AC mixture for years now [19]. Glass fibers have a high tensile modulus of about 60 GPa, low elongation of 3–4%, high elastic recovery, do not absorb water and have a high softening point (815 ◦C) [7]. When reinforced in AC mixture or cement concretes, glass fibers tend to increase the tensile strength and their compressive strength improves elastic recovery, and they increase the softening point of the asphalt binder [19–22]. The objectives of this research are to determine the effects of adding glass wool fibers in dense-graded AC in terms of strength properties, moisture susceptibility and fatigue properties, and determine the optimum fiber content to be used.

2. Materials and Experiment Method

2.1. Materials The materials used in this research include virgin asphalt binder with PG 64-20 produced by S-OIL company from Seoul Korea which was used in all AC samples and experiments. The properties of the binder used are given in Table1 below. The dense-graded AC mixture used is known as Wearing Coarse-2 (WC-2), and it has a maximum aggregate size of 13 mm. WC-2 is commonly used in Korea as a surface asphalt pavement and it is highly water-resistant. Table2 shows aggregate gradation, the required gradation limits and the aggregates’ sieve passing percentages used in this study, and as it is shown, aggregates used were all within the required limits. Figure1 shows the normal appearance and an SEM micrograph of glass wool fibers used in this research. As shown in Figure1, the glass wool fiber is white color. The thickness of the glass wool fiber is about 5~8 µm and it has a similar look to cotton candy. Table3 shows the chemical composition of glass wool fiber. Materials 2020, 13, x FOR PEER REVIEW 3 of 13

Table 1. Properties of virgin asphalt.

Property Test Value Standards Penetration (25°C, 100g, 5s) 0.1mm 63 ASTM D 5

Ductility (15°C, 5 cm/min) (cm) 150 ASTM D 113

Materials 2020, 13, 4699 Softening (°C) 47.5 ASTM D 36 3 of 14

Flash PointTable (°C) 1. Properties of virgin asphalt. 354 ASTM D 92

DensityProperty (15°C) (Kg/m3) Test Value 1.0410 Standards ASTM D 70

Penetration (25 ◦C, 100 g, 5 s) 0.1mm 63 ASTM D 5 Ductility(15 ◦C, 5 cm/min) (cm) 150 ASTM D 113 SofteningTable (◦C) 2. Gradation of asphalt 47.5 mixture. ASTM D 36 Flash Point (◦C) 354 ASTM D 92 3 Sieve Size Density (15◦C) (Kg/m ) 1.0410 ASTM D 70 25 20 13 10 5 2.5 0.6 0.3 0.15 0.08 (mm) Table 2. Gradation of asphalt mixture. Limits Sieve Size100 100 100–95 84–92 55–70 35–50 18–30 10–21 6–16 4–8 (%) 25 20 13 10 5 2.5 0.6 0.3 0.15 0.08 (mm) Passing Limits (%)100 100 10096 100–95 85 84–92 55–7056 35–5036 18–3019 10–21 126–16 4–87 5 (%) Passing (%) 100 100 96 85 56 36 19 12 7 5

(a) (b)

Figure 1.Figure (a) Normal 1. (a) Normal appearance appearance of glass of glass wool wool fibers ﬁbers and and (b ()b SEM) SEM micrograph micrograph of glass of wool glass ﬁbers. wool fibers.

Table 3. Chemical composition of glass wool fiber. Table 3. Chemical composition of glass wool fiber. SiO2 Al2O3 Fe2O3 CaO MgO Na2OK2O SO3 B2O3 TiO2 SiO66.9%2 Al2 1.2%O3 Fe 0.2%2O3 CaO 9.0% MgO 1.9% 15.2%Na2O 0.3%K2O 0.06%SO3 5.0%B2O 0.0453 TiO2 66.9% 1.2% 0.2% 9.0% 1.9% 15.2% 0.3% 0.06% 5.0% 0.045 2.2. Sample Preparations 2.2. Sample PreparationsIn this research, glass wool fibers with a thickness of 5–8 µm were mixed in a dense-graded AC mixture together with asphalt binder and aggregates. One control sample with no fiber and In thisanother research, four samples glass withwool di fibersfferent with percentages a thickness of glass of wool 5–8 fibersµm were content mixed (0.2%, in 0.3%, a dense-graded 0.4% and AC mixture 0.5%together by weight with of asphalt aggregates) binder were and used. aggregat Duringes. mixing, One fiberscontrol were sample mixed withwith aggregates no fiber and first another four samplesat a temperature with different of 150 percentages◦C for 1 min, andof glass then asphaltwool fibers binder content was added. (0.2%, This 0.3%, method 0.4% is known and 0.5% by weight ofas aaggregates) dry mixing method, were andused. it isDuring preferred mixing, over the wetfibers mixing were method mixed because with it allowsaggregates the best first at a temperaturefibers of distribution 150 ℃ for in the1 min, mixture and [19 then]. The asphalt already binder mixed hot was AC added. samples This were method placed in is molds known and as a dry compacted, then cured for 24 h prior to testing. mixing method, and it is preferred over the wet mixing method because it allows the best fibers distribution2.3. Marshall in the Mixmixture Design Method[19]. The already mixed hot AC samples were placed in molds and compacted, thenThe Marshall cured for mix 24 design h prior is a commonto testing. method used for designing dense-graded AC mixtures. It has two major features, namely density voids analysis and stability flow test [23]. In this research, 2.3. Marshallthe Marshall Mix Design mix design Method test was done in accordance with the Korean standard for construction sector KS F 2337 [24]. Marshall specimens with a weight of 1200 g, width of 63.5 1.27 mm and a diameter ± Theof Marshall 101.6 mm mix were design compacted is a 75common times on method both sides used and for then designing cured at room dense-graded temperature AC for 24 mixtures. h. It has two major features, namely density voids analysis and stability flow test [23]. In this research, the Marshall mix design test was done in accordance with the Korean standard for construction sector

Materials 2020, 13, x FOR PEER REVIEW 4 of 13

KS F 2337 [24]. Marshall specimens with a weight of 1200 g, width of 63.5 ± 1.27 mm and a diameter of 101.6 mm were compacted 75 times on both sides and then cured at room temperature for 24 h. For each set of fiber contents, there were at least 3 specimens. The Marshall mix design method was used to determine voids in mineral aggregates, air voids, voids filled with asphalt, bulk specific gravity Marshall stability and flow of AC mixtures with different dosages of glass wool fibers. Materials 2020, 13, 4699 4 of 14 2.4. Indirect Tensile Strength

ForThe each indirect set of tensile fiber contents, strength there (IDT) were test at is least used 3 specimens. to determine The the Marshall tensile mix properties design method of the was asphalt mixture;used totensile determine properties voids in include mineral aggregates,tensile strength, air voids, toughness voids filled and with strain asphalt, (displacement). bulk specific gravity Tensile propertiesMarshall play stability an important and flow ofrole AC in mixtures the performa with diffnceerent of dosagesan AC ofmixture glass wool under fibers. fatigue, rutting and moisture susceptibility [25]. The IDT tests were conducted according to the Korean standard KS F 2.4. Indirect Tensile Strength 2382 [24], where a Marshall specimen with a weight of 1150 g and diameter of 101.6 mm, compacted 75 times Theon both indirect sides, tensile was strength loaded (IDT) with test compressive is used to determine load at a the constant tensile propertiesrate of 51 of mm the⁄ asphalt min acting parallelmixture; to and tensile along properties the diametric include plane tensile of the strength, specimen. toughness The compressive and strain (displacement). load indirectly Tensile creates a tensileproperties load on play the anhorizontal important direction role in the of performance the sample ofand an tensile AC mixture failure under occurs fatigue, in a rutting sample and rather moisture susceptibility [25]. The IDT tests were conducted according to the Korean standard KS F than a compressive failure [18]. Tensile strength values were obtained using the mathematical 2382 [24], where a Marshall specimen with a weight of 1150 g and diameter of 101.6 mm, compacted 75 equation given bellow and the toughness was obtained from the dissipated energy during the times on both sides, was loaded with compressive load at a constant rate of 51 mm/min acting parallel to material’sand along fracture the diametric process. plane of the specimen. The compressive load indirectly creates a tensile load on the horizontal direction of the sample and tensile2000 failure × occurs𝑃 in a sample rather than a compressive 𝑆= (1) failure [18]. Tensile strength values were obtained using𝜋𝑡𝐷 the mathematical equation given bellow and the toughness was obtained from the dissipated energy during the material’s fracture process. where 𝑆 = tensile strength, kPa; 𝑃 = maximum load, N; 𝑡= specimen height before testing, mm; 2000 P 𝐷= specimen diameter, mm. S = × (1) πtD 2.5. Tensilewhere SStrength= tensile Ratio strength, (TSR) kPa; P = maximum load, N; t = specimen height before testing, mm; D = specimen diameter, mm. Tensile strength ratio (TSR) test is used to determine the moisture susceptibility of asphalt mixtures.2.5. Tensile It measures Strength the Ratio workability (TSR) of asphalt pavements in terms of performance and durability in the presenceTensile of strength moisture. ratio (TSR)The TSR test istest used wa tos determinecarried out the in moisture accordance susceptibility with the of Korean asphalt mixtures.standard KS F 2398It measures[24] as shown the workability in Figure of asphalt2. The pavementsstandard inrequ termsires of the performance TSR test and to be durability done at in a the volume presence of air void ofof moisture. 7 ± 0.5% The. Each TSR set test had was 6 carried Marshall out specimens, in accordance and with after the curing, Korean standard3 specimens KS F were 2398 [24soaked] as in watershown at a temperature in Figure2. The of standard60 ℃ for requires 24 h and the the TSR other test to remaining be done at aspecimens volume of airwere void kept of 7 dry0.5%. at room ± temperature.Each set had TSR 6 Marshallresults are specimens, obtained and as after a ratio curing, of the 3 specimens tensile strength were soaked values in water of conditioned at a temperature samples whichof were 60 ◦C soaked for 24 h andin water the other to the remaining tensile strength specimens values were kept of unconditioned dry at room temperature. dry samples TSR of results the same mixture.are obtained as a ratio of the tensile strength values of conditioned samples which were soaked in water to the tensile strength values of unconditioned dry samples of the same mixture.

Figure 2. Tensile strength ratio test. Figure 2. Tensile strength ratio test. 2.6. Kim Test for Deformation Strength 2.6. Kim Test for Deformation Strength This is a new test recently developed in Korea for determining the rutting properties of AC mixtures.This is a Anew round-edged test recently loading developed head shown in in Korea Figure 3for, which determining simulates almostthe rutting the exact properties contact area of AC mixtures.between A round-edged the tire and road loading surface, head was appliedshown insteadin Figure of applying3, which a simulates load to the almost whole cross-sectional the exact contact area. The resistance against the formation of a dimple (not ﬂat) on the surface of the specimen is known

Materials 2020, 13, x FOR PEER REVIEW 5 of 13 area between the tire and road surface, was applied instead of applying a load to the whole cross- sectional area. The resistance against the formation of a dimple (not flat) on the surface of the specimen is known as the deformation strength (SD). The deformation strength is a combination of Materials 2020, 13, 4699 5 of 14 compressive stress and shear stress. The Kim test showed good correlations with well-known rutting tests for dense-graded AC mixtures like wheel tracking [26,27]. The Kim test requires a normal as the deformation strength (SD). The deformation strength is a combination of compressive stress and Marshallshear specimen stress. Theto Kimbe submerged test showed goodin a correlations water bath with for well-known 30 min at rutting 60 ℃ tests, and for dense-gradedthen after a load is AC mixtures like wheel tracking [26,27]. The Kim test requires a normal Marshall specimen to be applied at a speed of 30 mm/min. The SD value is calculated as submerged in a water bath for 30 min at 60 C, and then after a load is applied at a speed of 30 mm/min. ◦ 0.32𝑃 The SD value is calculated as 𝑆 = 0.32P S = (2) (2) D 10h + p20𝑦 −i 𝑦 2 10 + 20y y2 − where 𝑆 = deformation strength; 𝑃= maximum load; 𝑦= deformation. where SD = deformation strength; P = maximum load; y = deformation.

FigureFigure 3. Kim 3. Kim test test for for deformation strength. strength.

2.7. Scanning Electron Microscopy (SEM) 2.7. Scanning Electron Microscopy (SEM) In order to investigate the distribution of glass fiber wool in the asphalt mixture, scanning electron In ordermicroscopy to investigate (SEM) was adoptedthe distribution in this study. of Theglas SEMs fiber device wool used in was the Hitachi, asphalt model mixture, SU8230 scanning electron frommicroscopy Tokyo, Japan. (SEM) This was method adopted was usedin this to evaluate study. theThe dispersion SEM device of glass used fiber was wool Hitachi, and to model SU8230 fromensure Tokyo, uniformity Japan. of the This dispersion method within was theused matrix to evaluate of the asphalt the dispersion mixture. Specimens of glass were fiber cut wool and into small pieces of approximately (10 10 5) mm for testing, and the samples were then coated × × to ensurewith uniformity a gold/palladium of the alloydispersion to increase within the sti theffness matrix at the surfaceof the andasphalt to obtain mixture. conductivity Specimens without were cut into smallaff ectingpieces the of observed approximately surface morphology, (10 × 10 × in 5) order mm to for make testing, them relatively and the more samples stable were during then the coated with a gold/palladiumtesting [28,29]. alloy to increase the stiffness at the surface and to obtain conductivity without affecting the observed surface morphology, in order to make them relatively more stable during the 3. Results and Discussion testing [28,29]. 3.1. Optimum Asphalt Content (OAC) 3. Results andThe Discussion optimum asphalt binder content for the fiber-reinforced AC mixture was determined using the Marshall mix design method. OAC is the amount of asphalt binder in terms of volume percentage, 3.1. Optimumrequired Asphalt to achieve Content the 4 (OAC)0.5% volume of air voids. The values of OAC were obtained through the ± interpolation of volume of air voids from three different asphalt contents at an increment of 0.5%. TheThe optimum OAC increases asphalt with binder the increasecontent in for fiber the content, fiber-reinforced as shown in AC Figure mixture4. Generally, was determined the fiber using the Marshall mix design method. OAC is the amount of asphalt binder in terms of volume percentage, required to achieve the 4 ± 0.5% volume of air voids. The values of OAC were obtained through the interpolation of volume of air voids from three different asphalt contents at an increment of 0.5%. The OAC increases with the increase in fiber content, as shown in Figure 4. Generally, the fiber absorbed the asphalt binder in asphalt mixtures [30]. The glass wool fibers can absorb the asphalt binder, which leads to the addition of more binder. Figure 5 shows a trend of the asphalt binder

Materials 2020, 13, x FOR PEER REVIEW 6 of 13 Materials 2020, 13, 4699 6 of 14 absorption capabilities of glass wool fibers as a function of fiber content. The asphalt binder Materials 2020, 13, x FOR PEER REVIEW 6 of 13 absorptionabsorbed ratio the increased asphalt binder rapidly in asphalt until 0.4% mixtures fiber [ 30co].ntent. The glassIt may wool be fibersconsidered can absorb that the asphaltaddition of fibersbinder,absorption in AC which is acapabilities complex leads to process, theof additionglass and wool ofa more morefibers reasonable binder.as a function Figure explanation5 ofshows fiber a content.needs trend ofto The the be asphaltinvestigatedasphalt binderbinder in the future.absorption capabilitiesratio increased of glass rapidly wool until fibers 0.4% as a fiber function content. of fiber It may content. be considered The asphalt that binder the addition absorption of ratiofibers increased in AC is a rapidly complex until process, 0.4% fiber and content.a more reasonable It may be consideredexplanation that needs the to addition be investigated of fibers inin ACthe isfuture. a complex5.3 process, and a more reasonable explanation needs to be investigated in the future.

5.2 5.3

5.2 5.1

5.1 5 5 4.9 4.9 4.8 4.8

Optimum Asphalt content (%) 4.7

4.6 4.6

4.5 4.5 00.200.20.3 0.40.4 0.50.5 GlassGlass wool wool fibers contentcontent (%) (%)

FigureFigure 4. 4. Optimum Optimum asphalt asphalt contentcontent (OAC). (OAC).

2.5 2.5

2 2 ) ) % ( % ( 1.5 1.5

tion ratio 1 p

Absor 0.5

Absor 0.5 0 00.20.3 0.4 0.5 0 Glass wool fibers content (%) 00.20.3 0.4 0.5 Figure 5. Glass wool fibers’ asphalt absorption ratio. Glass wool fibers content (%)

3.2. Bulk Specific Gravity FigureFigure 5. 5.GlassGlass wool wool fibers’ ﬁbers’ asphalt absorptionabsorption ratio. ratio. During the Marshall mix design test, the bulk specific gravity of each dosage was measured and 3.2. Bulkthe results Specific are Gravity given in Figure 6 below. The bulk specific gravity slightly decreases from 2.525 for the ACDuring mixture the withMarshall no fiber mix to design 2.50 for test, the ACthe mixture bulk specific with 0.5% gravity of fiber of eachcontent. dosage The additionwas measured of fiber and the resultsand the are increase given in in optimum Figure 6 asphaltbelow. content The bulk led specific to a decrease gravity in the slightly aggregates decreases volume from used, 2.525 which for the AC mixturehas the higher with no specific fiber gravity,to 2.50 forand the hence AC the mixture decrease with in bulk 0.5% specific of fiber gravity content. of the The AC addition mixture. of fiber and the increase in optimum asphalt content led to a decrease in the aggregates volume used, which has the higher specific gravity, and hence the decrease in bulk specific gravity of the AC mixture.

Materials 2020, 13, 4699 7 of 14

3.2. Bulk Specific Gravity During the Marshall mix design test, the bulk specific gravity of each dosage was measured and the results are given in Figure6 below. The bulk specific gravity slightly decreases from 2.525 for the AC mixture with no fiber to 2.50 for the AC mixture with 0.5% of fiber content. The addition of fiber and the increase in optimum asphalt content led to a decrease in the aggregates volume used, which has the higher specific gravity, and hence the decrease in bulk specific gravity of the AC mixture. Materials 2020, 13, x FOR PEER REVIEW 7 of 13

2.525

2.52

2.515

2.51

2.505

2.5 Bulk Specific gravity 2.495

2.49

2.485 00.2 0.3 0.4 0.5

Figure 6. Bulk specific speciﬁc gravity gravity..

3.3. Marshall Stability Figure7 7 shows shows thethe MarshallMarshall stabilitystability results,results, andand thethe measuredmeasured resultsresults explainexplain thethe resistanceresistance capabilities of a specimenspecimen related toto distortion,distortion, displacement,displacement, rutting and shearing stresses under maximum loading. The The results results of each set were obtained from the Marshall stability of an asphaltasphalt content which correlated correlated with with the the 4 4± 0.50.5%% volume volume of of the the air air void. void. Upon Upon the theaddition addition of fibers, of fibers, the ± theMarshall Marshall stability stability increased increased and andreached reached a peak a peak when when 0.3% 0.3% of fiber of fiber content content was wasadded added in the in AC, the and AC, andthenthen it started it started to decrease. to decrease. Glass Glass fibers fibers have have high high tensile tensile strength strength which which is is transferred transferred to to the AC mixture and increasesincreases thethe MarshallMarshall stability.stability.

3.4. Indirect18,000 Tensile Strength Figures8 and9 show the indirect tensile strength (IDT) results of asphalt mixtures containing different dosages of glass wool fibers. The results show that when fibers are added into the mixture, 17,000 both the tensile strength and toughness of the AC mixture increase to a maximum value and then start to decrease. In asphalt pavements, tensile strength (stiffness) relates to the cracking properties of the )

pavement,N with16,000 cracking being one of the primary asphalt pavement distress types along with rutting (

and fatigue.y The resistance of AC to fatigue cracking which is caused by repeated or ﬂuctuated stresses is dependent upon its tensile strength and extensibility characteristics. A higher tensile strength corresponds15,000 to a stronger crack resistance.

14,000 Marshall stabilit 13,000

12,000 0 0.2 0.3 0.4 0.5

Glass wool fibers content (%) Figure 7. Marshall stability.

Materials 2020, 13, x FOR PEER REVIEW 7 of 13

2.525

2.52

2.515

2.51

2.505

2.5 Bulk Specific gravity 2.495

2.49

2.485 00.2 0.3 0.4 0.5

Figure 6. Bulk specific gravity.

3.3. Marshall Stability Figure 7 shows the Marshall stability results, and the measured results explain the resistance capabilities of a specimen related to distortion, displacement, rutting and shearing stresses under maximum loading. The results of each set were obtained from the Marshall stability of an asphalt content which correlated with the 4 0.5% volume of the air void. Upon the addition of fibers, the Marshall stability increased and reached a peak when 0.3% of fiber content was added in the AC, and then Materialsit started2020 , to13, 4699decrease. Glass fibers have high tensile strength which is transferred to8 ofthe 14 AC mixture and increases the Marshall stability.

18,000

17,000 )

N 16,000 (

15,000

Materials 2020, 13, x FOR PEER REVIEW 8 of 13 14,000 3.4. Indirect Tensile Strength Marshall stabilit Figures13,000 8 and 9 show the indirect tensile strength (IDT) results of asphalt mixtures containing different dosages of glass wool fibers. The results show that when fibers are added into the mixture, both the tensile strength and toughness of the AC mixture increase to a maximum value and then start to 12,000decrease. In asphalt pavements, tensile strength (stiffness) relates to the cracking properties of the pavement, with cracking0 being0.2 one of the primary0.3 asphalt pavement0.4 distress0.5 types along with rutting and fatigue. The resistance of ACGlass to fatigue wool cracking fibers whichcontent is caused (%) by repeated or fluctuated stresses is dependent upon its tensile strength and extensibility characteristics. A higher tensile strength corresponds to a stronger crackFigureFigure resistance. 7. 7. MarshallMarshall stability.stability.

1.6

1.4 ) 2

1.2

1 Tensile strength (N/mm Tensile strength

0.8 0 0.2 0.3 0.4 0.5 Glass wool fibers content (%)

FigureFigure 8.8. Tensile strengthstrength results.results.

13,400

13,200

13,000 mm) ∙

12,800

12,600 Toughness (N

12,400

12,200 00.20.3 0.4 0.5

Glass wool fibers content (%) Figure 9. Toughness results.

Materials 2020, 13, x FOR PEER REVIEW 8 of 13

3.4. Indirect Tensile Strength Figures 8 and 9 show the indirect tensile strength (IDT) results of asphalt mixtures containing different dosages of glass wool fibers. The results show that when fibers are added into the mixture, both the tensile strength and toughness of the AC mixture increase to a maximum value and then start to decrease. In asphalt pavements, tensile strength (stiffness) relates to the cracking properties of the pavement, with cracking being one of the primary asphalt pavement distress types along with rutting and fatigue. The resistance of AC to fatigue cracking which is caused by repeated or fluctuated stresses is dependent upon its tensile strength and extensibility characteristics. A higher tensile strength corresponds to a stronger crack resistance.

1.6

1.4 ) 2

1.2

1 Tensile strength (N/mm Tensile strength

0.8 0 0.2 0.3 0.4 0.5 Glass wool fibers content (%) Materials 2020, 13, 4699 9 of 14 Figure 8. Tensile strength results.

13,400

13,200

13,000 mm) ∙

12,800

12,600 Toughness (N

12,400

12,200 00.20.3 0.4 0.5

Glass wool fibers content (%)

FigureFigure 9. 9. ToughnessToughness results.results.

For the IDT test, each fiber content set had three specimens to be tested and results were obtained as an average. The Korean Asphalt standards guide for the AC mixture WC-2 given in Table4 requires that tensile strength and toughness should be over 0.8 N/mm2 and 8000 Nmm, respectively. The tensile strength results show that tensile strength increased from 1.28 for a mixture without fibers to 1.44 N/mm2 for a mixture with 0.3% dosage of glass wool fibers, then for the higher dosage of fiber, the values of tensile strength decreased. Toughness results have the same trend as the tensile strength, where the maximum toughness value was observed at a mixture with 0.3% dosage of fibers, and for higher dosages, the toughness decreases. The decrease in tensile strength and toughness for higher dosages is caused by the large amount of fibers in a mixture which replace aggregates, while aggregates have higher tensile strength than fibers. Glass wool fibers, when added in an asphalt mixture, increase the strength of a bond between the asphalt and aggregates by firmly binding aggregate particles inside the matrix and preventing them from moving, which makes the mix stiffer [15].

Table 4. Korean Asphalt standard for WC-2.

Test WC-2 Etc Marshall Stability (N) Over 7500 Air Void (%) 3–6 Compaction 75 both sides Saturation Degree (%) 65–80 TSR Over 0.8 Air void 7 0.5% ± Tensile Strength (N/mm2) Over 0.8 Compaction 75 both sides Toughness (N mm) Over 8000 × 2 Kim Test SD (N/mm ) 3.20 Air void 3–5% Materials 2020, 13, x FOR PEER REVIEW 9 of 13

For the IDT test, each fiber content set had three specimens to be tested and results were obtained as an average. The Korean Asphalt standards guide for the AC mixture WC-2 given in Table 4 requires that tensile strength and toughness should be over 0.8 N/mm2 and 8000 Nmm, respectively. The tensile strength results show that tensile strength increased from 1.28 for a mixture without fibers to 1.44 N/mm2 for a mixture with 0.3% dosage of glass wool fibers, then for the higher dosage of fiber, the values of tensile strength decreased. Toughness results have the same trend as the tensile strength, where the maximum toughness value was observed at a mixture with 0.3% dosage of fibers, and for higher dosages, the toughness decreases. The decrease in tensile strength and toughness for higher dosages is caused by the large amount of fibers in a mixture which replace aggregates, while aggregates have higher tensile strength than fibers. Glass wool fibers, when added in an asphalt mixture, increase the strength of a bond between the asphalt and aggregates by firmly binding aggregate particles inside the matrix and preventing them from moving, which makes the mix stiffer [15].

Table 4. Korean Asphalt standard for WC-2.

Test WC-2 Etc Marshall Stability (N) Over 7500 Air Void (%) 3–6 Compaction 75 both sides Saturation Degree (%) 65–80 TSR Over 0.8 Air void 7 ± 0.5% Tensile Strength (N/mm) Over 0.8 Compaction 75 both sides Toughness (N×mm) Over 8000 Kim Test SD (N/mm ) 3.20 Air void 3–5% Materials 2020, 13, 4699 10 of 14 3.5. Tensile Strength Ratio (TSR) 3.5. Tensile Strength Ratio (TSR) In order to estimate the moisture susceptibility of AC, the tensile strength ratio (TSR) test was conducted.In Figure order to10 estimate shows thethe moisture results susceptibilityof the tensile of AC,strength the tensile ratio strength (TSR) ratioof asphalt (TSR) test mixtures was for differentconducted. dosages Figureof fiber 10 contents.shows the As results the ofamount the tensile of fiber strength dosage ratio increases, (TSR) of asphalt the percentage mixtures for of TSR different dosages of fiber contents. As the amount of fiber dosage increases, the percentage of TSR also increases. The TSR increases up to a maximum value at a 0.3% dosage of fiber and then starts to also increases. The TSR increases up to a maximum value at a 0.3% dosage of fiber and then starts to decrease.decrease. A higher A higher TSR TSRvalue value indicates indicates that that a amixture mixture will havehave good good resistance resistance to moisture to moisture damages. damages. The additionThe addition of fiber of improves fiber improves the moisture the moisture susceptibility susceptibility of ofthe the AC AC mixture. mixture. The The main main reason reason behind this phenomenonbehind this phenomenon is that when is that glass when wool glass woolfibers fibers interact interact with with an an asphaltasphalt mixture, mixture, this notthis only not only increasesincreases the bonding the bonding strength strength but but also also increases increases the thicknessthickness of of the the asphalt asphalt film film which which is beneficial is beneficial for preventingfor preventing moisture moisture from from entering entering the the interface interface betweenbetween the the asphalt asphalt and and aggregates. aggregates.

Tensile strength ratioTensile strength (%) 79

75 0 0.2 0.3 0.4 0.5

Glass wool fibers content (%) FigureFigure 10. 10. TensileTensile strength strength ratio ratio results. results.

3.6. Kim Test Figure 11 shows the results of the Kim test for the deformation strength of AC mixtures with different dosages of glass wool fibers. The results show that the addition of fibers improves the high-temperature rutting properties of the AC. The SD values increased with the increased dosage of glass wool fibers, and they reached a maximum value at 7.32 N/mm2 when a dosage of 0.4% was added to the AC mixture, which is 25% more than the AC mixture without fibers. The SD values increased because glass wool fibers absorbed the asphalt binder and increased its viscosity, and therefore, due to the increased viscosity and bridging effects, the bonding strength between the asphalt and aggregates was increased.

3.7. Scanning Electronic Microscopy (SEM) The SEM images of the fibers in an asphalt mixture are commonly used to show the dispersion of fibers in the mixture so as to understand the factors influencing the mechanical properties of the AC mixture. Images obtained by the use of scanning electron microscopy (SEM) device used Hitachi, model SU8230 from Tokyo, Japan are shown in the figures below. As shown in Figure 12a, the glass wool fibers were well immersed in the asphalt binder. The large portion of glass wool fibers was distributed in the asphalt mixture, and on the edges, there were spike-like fibers that emerged. The distribution of fibers in AC is mostly influenced by the size of fibers, thus when small-size fibers are well distributed Materials 2020, 13, x FOR PEER REVIEW 10 of 13

3.6. Kim Test Figure 11 shows the results of the Kim test for the deformation strength of AC mixtures with differentMaterials dosages2020, 13 of, 4699 glass wool fibers. The results show that the addition of fibers improves11 of the 14 high- temperature rutting properties of the AC. The SD values increased with the increased dosage of glass N/mm wool fibers,in a mixture, and they this reached leads to ana maximum increase in thevalue strength at 7.32 and stiffness whenof the mixture.a dosage In of AC 0.4% mixtures, was added to the ACglass mixture, wool fibers which interact is 25% with more the aggregates than the and AC binder, mixture as shown without in Figurefibers. 12 Theb. Figure SD values 13 shows increased becausea microstructureglass wool fibers with absorbed a resolution the of asphalt 50 µm of bind a destroyeder and partincreased of an IDT its sampleviscosity, after and the therefore, test was due to the conducted.increased Fibersviscosity interact and with bridging the aggregates effects, and the binder bonding in the AC,strength and when between tensile the strength asphalt is and aggregatesapplied was to theincreased. AC, more tensile strength is needed to break the fibers.

8 ) 2

5 Deformation Strength (N/mm Deformation Strength

4 00.20.3 0.4 0.5 Glass wool fibers content (%) Materials 2020, 13, x FOR PEER REVIEW 11 of 13 FigureFigure 11. 11. KimKim test results.results.

3.7. Scanning Electronic Microscopy (SEM) The SEM images of the fibers in an asphalt mixture are commonly used to show the dispersion of fibers in the mixture so as to understand the factors influencing the mechanical properties of the AC mixture. Images obtained by the use of scanning electron microscopy (SEM) device used Hitachi, model SU8230 from Tokyo, Japan are shown in the figures below. As shown in Figure 12a, the glass wool fibers were well immersed in the asphalt binder. The large portion of glass wool fibers was distributed in the asphalt mixture, and on the edges, there were spike-like fibers that emerged. The distribution of fibers in AC is mostly influenced by the size of fibers, thus when small-size fibers are well distributed in a mixture, this leads to an increa se in the strength and stiffness of the mixture. In AC mixtures, glass wool fibers interact with the aggregates and binder, as shown in Figure 12b. (a) (b) Figure 13 shows a microstructure with a resolution of 50 μm of a destroyed part of an IDT sample Figure 12. (a) Glass wool ﬁbers mixed with asphalt binder and (b) asphalt concrete (AC) mixture with after theFigure test 12.was (a) Gconducted.lass wool fibers Fibers mixed interact with asphalt with binder the andaggregates (b) asphalt and concrete binder (AC )in mixture the AC, with and when reinforced ﬁbers. tensile strengthreinforced isfibers applied. to the AC, more tensile strength is needed to break the fibers.

Figure 13. SEM micrograph of an AC mixture with fiber after indirect tensile strength (IDT) testing.

4. Conclusions This paper studied the performance of glass wool fibers in a dense-graded AC mixture. Glass wool fibers were added into the AC mixture together with an asphalt binder and aggregates. Laboratory tests, Marshall mix design, indirect tensile strength, tensile strength ratio and the Kim test were conducted to evaluate the effect of adding fibers to the AC. The following results were obtained based on laboratory investigations. 1. The addition of glass wool fibers to the AC mixture improves the performance of the AC in terms of tensile strength, toughness, rutting resistance, moisture susceptibility and fatigue. 2. The AC mixture with 0.3% of fiber content by weight of aggregates showed superior results over the other mixtures, where the results of indirect tensile strength, toughness, deformation strength and tensile strength ratio increased by 13%, 6%, 25% and 8%, respectively. 3. It was also observed that the addition of fibers leads to the addition of more asphalt binder. The addition of more binder enhances the durability of asphalt pavements.

Lastly, the authors suggest that more investigations should be done on the physical properties of glass wool fibers, fibers’ asphalt binder absorption capabilities and energy dispersive X-ray (EDX) analysis of the glass wool fibers. Additionally, since the glass wool fibers improved the tensile

Materials 2020, 13, x FOR PEER REVIEW 11 of 13

(a) (b)

MaterialsFigure2020 ,12.13, ( 4699a) Glass wool fibers mixed with asphalt binder and (b) asphalt concrete (AC) mixture with12 of 14 reinforced fibers.

Figure 13. SEMSEM micrograph micrograph of an AC mixture with fiber ﬁber after indirect tensile strength (IDT) testing.testing.

4. Conclusions This paperpaper studiedstudied the the performance performanc ofe of glass glass wool wool fibers fibers in a in dense-graded a dense-graded AC mixture. AC mixture. Glass Glass wool fiberswool fibers were addedwere intoadded the into AC mixturethe AC togethermixture withtogether an asphalt with an binder asphalt and binder aggregates. and aggregates. Laboratory tests,Laboratory Marshall tests, mix Marshall design, mix indirect design, tensile indirect strength, tensile strength, tensile strength tensile strength ratio and ratio the and Kim the test Kim were test conductedwere conducted to evaluate to evaluate the eff theect effect of adding of adding fibers fi tobers the to AC. the The AC. following The following results results were obtainedwere obtained based onbased laboratory on laboratory investigations. investigations. 1. TheThe addition addition of of glass glass wool wool fibers fibers to to the the AC AC mixture mixture improves improves the performance of the AC in terms ofof tensile tensile strength, strength, toughness, toughness, rutting rutting resistance, moisture susceptibility and fatigue. 2. TheThe AC AC mixture mixture with with 0.3% 0.3% of of fiber fiber content content by by weig weightht of of aggregates aggregates showed showed superior superior results over thethe otherother mixtures,mixtures, where where the the results results of indirect of indire tensilect tensile strength, strength, toughness, toughness, deformation deformation strength strengthand tensile and strength tensile strength ratio increased ratio increased by 13%, 6%,by 13%, 25% 6%, and 25% 8%, respectively.and 8%, respectively. 3. ItIt was was also also observed observed that that the the addition addition of offibers fibers leads leads to the to theaddition addition of more of more asphalt asphalt binder. binder. The additionThe addition of more of more binder binder enhances enhances the durability the durability of asphalt of asphalt pavements. pavements. Lastly, the authors suggest that more investigations should be done on the physical properties of glassLastly, wool the fibers, authors fibers’ suggest asphalt that binder more absorptioninvestigations capabilities should be and done energy on the dispersive physical X-ray properties (EDX) analysisof glass wool of the fibers, glass fibers’ wool asph fibers.alt binder Additionally, absorption since capabilities the glass and wool energy fibers dispersive improved X-ray the tensile(EDX) propertiesanalysis of of the AC glass and havewool high fibers. absorption Additionally, capabilities, since investigationsthe glass wool should fibers beimproved done on reinforcingthe tensile glass wool fibers in stone mastic asphalt mixtures.

Author Contributions: Conceptualization, A.H.M., and J.-J.L.; methodology, A.H.M. and S.-H.N.; writing–review and editing, A.H.M.; formal analysis, J.-J.L.; funding acquisition, O.-S.K. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Korea Agency for Infrastructure Technology Advancement grant number 20TSRD-B151356-02. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Mamlouk, M.S.; Zaniewski, J.P. Materials for Civil and Construction Engineers, 3rd ed.; Pearson Education: New Jersey, NJ, USA, 2011; ISBN-13: 978-0-13-800956-4. 2. Asphalt Institute. MS-2 Asphalt Mix Design Methods, 7th ed.; Asphalt Institute: Lexington, KY, USA, 2014; ISBN 978-1-934154-70-0. 3. National Cooperative Highway Research Program (NCHRP). A Manual for Design of Hot Mix Asphalt with Commentary; National Academy of Science: Sterling, WV, USA, 2011; ISBN 978-0-309-15564-9. Materials 2020, 13, 4699 13 of 14

4. Department of Transport and Main Roads (State of Queensland). MRTS30 Dense Graded and Open Graded Asphalt Mixture Specifications. Available online: https://www.tmr.qld.gov.au/-/media/busind/techstdpubs/ Specifications-and-drawings/Specifications/Superseded-Specifications/5-Pavements-Subgrade-and- Surfacing/MRTS30/2013_06/MRTS30_Superseded.pdf?la=en (accessed on 24 May 2020). 5. Abiola, O.S.; Kupolati, W.K.; Sadiku, R.; Ndambuki, J. Utilisation of Natural Fibre as Modifier in Bituminous Mixes: A Review. Constr. Build. Mater. 2014, 54, 305–312. [CrossRef] 6. Xu, Q.; Chen, H.; Prozzi, J.A. Performance of Fiber Reinforced Asphalt Concrete under Environmental Temperature and water Effects. Constr. Build. Mater. 2010, 24, 2003–2010. [CrossRef] 7. National Cooperative Highway Research Program (NCHRP). Fiber Additives in Asphalt Mixtures, A Synthesis of Highway Practice; National Academy of Science: Washington, DC, USA, 2015; ISBN 978-0-309-27177-6. 8. Musa, N.F.; Aman, M.F.; Shahadan, Z.; Taher, M.N.; Noranai, Z. Utilization of Synthetic Reinforced Fiber in Asphalt Concrete—A Review. Int. J. Civ. Eng. Technol. 2019, 10, 678–694. 9. Decoene, Y. Contribution of cellulose fibers to the performance of porous asphalts. Transp. Res. Rec. 1990, 1265, 82–86. 10. Irfan, M.; Ali, Y.; Ahmed, S.; Iqbal, S.; Wang, H. Rutting and Fatigue Properties of Cellulose Fiber-Added Stone Mastic Asphalt Concrete Mixtures. Adv Mater. Sci. Eng. 2019, 2019, 5604197. [CrossRef] 11. Slebi-Acevedo, C.J.; Lastra-Gonzalez, P.; Pascual-Munoz, P.; Castro-Fresno, D. Mechanical Performance of Fibers in Hot Mix Asphalt: A review. Constr. Build. Mater. 2019, 200, 756–769. [CrossRef] 12. Bindu, C.S.; Beena, K.S. Comparison of shear strength characteristics of stone matrix asphalt mixture with waste plastics and polypropylene. Int. J. Struct. Civ. Eng. Res. 2013, 2, 13–21. 13. Yi, J.; McDaniel, R.S. Application of Cracking and Seating and Use of Fibers to Control Reflection Cracking. Transp. Res. Rec. 1993, 1388, 150–159. 14. Serkan, T.; Abdulkadir, C.; Ün, U. Accumulated Strain Prediction of Polypropylene Modified Marshall Specimens in Repeated Creep Test Using Artificial Neural Networks. Expert Syst. Appl. 2009, 36, 11186–11197. [CrossRef] 15. Sharear, K.M.; King, W.; Abadie, C.; Icenogle, P.; Cooper, S.B. Louisiana’s Experience with Open-Graded Friction Course Mixtures. Transp. Res. Rec. 2012, 2295, 63–71. [CrossRef] 16. Shaopeng, W.; Qunshan, Y.; Ning, L. Investigation of rheological and fatigue properties of asphalt mixtures containing polyester fibers. Constr. Build. Mater. 2008, 22, 2111–2115. [CrossRef] 17. Hollandt, J.; Hartmann, J.; Strus, O.; Gartner, R. Chapter 1 Industrial Applications of Radiation Thermometry. Exp. Methods Phys. Sci. 2010, 43, 1–56. [CrossRef] 18. Marhoon, I.I.; Rasheed, A.K. Mechanical and Physical Properties of Glass Wool-Rigid Polyurethane Foam Composite. Al-Nahrain Univ. Coll. Eng. J. 2015, 18, 41–49. 19. Najd, A.; Chao, Z.; Ying, G. Experiments of Fracture Behavior of glass fiber reinforced asphalt concrete. J. Chang. Univ. (Nat. Sci. Ed.) 2005, 25, 28–32. 20. Abtahi, S.M.; Sheikhzadeh, M.; Hejazi, S.M. Fiber Reinforced Asphalt Concrete—A review. Constr. Build. Mater. 2010, 24, 871–877. [CrossRef] 21. Alidadi, M.; Khabiri, M.M. Experimental study on the Effect of Glass and Carbon Fibers on Physical an Micro-Structure Behavior of Asphalt. Int. J. Integr. Eng. 2016, 8, 1–8. 22. Abtahi, S.M.; Esfandiarpour, S.; Kunt, M.; Hejazi, S.M.; Ebrahimi, M.G. Hybrid Reinforcement of Asphalt-Concrete Mixture using Glass and Polyproppylene Fibers. J. Eng. Fibers Fabr. 2013, 8, 25–35. 23. Mathew, T.V.; Krisna, R.K. Chapter 1 Introduction to Transport Engineering. Available online: https: //nptel.ac.in/content/storage2/courses/105101087/downloads/Lec-1.pdf (accessed on 5 June 2020). 24. Ascon Industry Cooperative Federation of Korea. Available online: https://www.standard.go.kr/KSCI/ct/ ptl/download.do%3Bjsessionid%3DRZJBu8ySl9YAP1OC4FSvPKQH.node01%3FfileSn%3D91172+&cd=4& hl=en&ct=clnk&gl=kr&client=opera (accessed on 19 May 2020). 25. Maleka, A.M.; Alkali, I.A.; Jaya, R.P. The Indirect Tensile Strength of Palm Oil Fuel Ash (POFA) Modified Asphalt Concrete. Appl. Mech. Mater. 2014, 587–589, 1270–1275. [CrossRef] 26. School of Engineering, Cochin University of science and Technology. Available online: https://shodhganga. inflibnet.ac.in/bitstream/10603/6226/10/10_chapter%205.pdf (accessed on 20 June 2020).

27. Kim, K.W. Development of Deformation Strength (SD) for Estimation of Asphalt Concrete Performance and Its Applicability. In Proceedings of the Advanced Technologies In Asphalt Pavements (ATAP) International Symposium, Chuncheon, Korea, 26–27 February 2020; pp. 1–24. Materials 2020, 13, 4699 14 of 14

28. Zhao, H.; Guan, B.; Xiong, R.; Zhang, A. Investigation of the Performance of Basalt Reinforced Asphalt Mixture. Appl. Sci. 2020, 10, 1561. [CrossRef] 29. Mubaraki, M. The Effect of modified asphalt binders by Fourier Transform Infrared Spectroscopy, X-Ray Diffraction, and Scanning Electron Microscopy. J. Mater. Eng. Struct. 2019, 6, 5–14. 30. Guo, F.; Li, R.; Lu, S.; Bi, Y.; He, H. Evaluation of the Effect of Fiber Type, Length, and Content on Asphalt Properties and Asphalt Mixture Performance. Materials 2020, 13, 1556. [CrossRef][PubMed]

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional aﬃliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).