Construction and Building MATERIALS Construction and Building Materials 19 (2005) 11–18 www.elsevier.com/locate/conbuildmat Effects of various additives on the moisture damage sensitivity of asphalt mixtures Atakan Aksoy a a,* € , Kurtulusß S ureyya Tayfur c, Halit Ozen ß amlioglu b, S€ d Department of Civil Engineering, Karadeniz Technical University, Trabzon, Turkey b General Directorate of Highways, Ankara, Turkey c ISFALT Asphalt Company, Istanbul, Turkey d Department of Civil Engineering, Yıldız Technical University, Istanbul, Turkey Received 15 November 2003; received in revised form 27 April 2004; accepted 5 May 2004 Available online 31 July 2004 Abstract Effects of four additives, namely two fatty amine (Wetfix I, Lilamin VP 75P), one catalyst (Chemcrete) and a polymer (rubber), on the moisture damage of asphalt mixtures were studied. Rheological characteristics of the binders were measured using conventional methods both original and thin-film oven aged. Mechanical characteristics of the mixtures were evaluated with Marshall, indirect tensile and Lottman treatment tests. The additives used in this study reduced the level of damage due to moisture in asphalt mixtures. Minimum acceptable indirect tensile strength ratio (0.70) is achieved when Chemcrete and 0.2% of Wetfix I, and 0.4–0.6% of Lilamin VP 75P are used in asphalt mixtures. Indirect tensile strength ratio may decrease due to the relatively higher strength obtained in dry specimens with respect to the conditioned ones. Indirect tensile strength ratios of asphalt paving specimens were found to be less than the Marshall Stability ratios. Ó 2004 Elsevier Ltd. All rights reserved. Keywords: Additives; Stripping; Marshall Stability and flow; Indirect tensile strength; Marshall Stability ratio; Marshall Stability–flow ratio; Indirect tensile strength ratio 1. Introduction In bituminous mixtures, many problems are due to stripping of the binder from the aggregate. To overcome this phenomenon, adhesion-improving agents are often used. Lately, the use of such additives has gained acceptance by engineers. Some of these are hydrated lime, sulfur, anti-oxidants, anti-stripping agents, rubber, carbon black and a variety of polymers. In order to improve adhesion and reduce moisture sensitivity in asphalt mixtures, two different approaches become apparent. The first, approach suggests the aggregate surface to be coated by a suitable agent that will reverse the predominant electrical charges at the surface and thus reduce the surface energy of the aggregate. The second approach is to reduce the surface energy of the * Corresponding author. E-mail address: aaksoy@risc01.ktu.edu.tr (A. Aksoy). 0950-0618/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.conbuildmat.2004.05.003 binder and give an electrical charge opposite to that of the aggregate surface. To achieve this surface active agents, the so-called ‘‘surfactants’’ are used. Surfactants, when used as anti-stripping agents, affect physiochemical properties of both the asphalt and the aggregate. Usually, asphalt pavements are subjected to extreme damage because of the adverse effect of moisture. Stripping occurs when the bond between the asphalt and the aggregate is broken by water. The water may be sent on or in the aggregate because of incomplete drying or it may come from some other source after construction. Water can cause stripping in different ways, such as spontaneous emulsification, displacement, detachment, pore pressure, hydraulic scouring, and osmosis [1]. Stripping of asphalt concrete has been defined as the weakening or eventual loss of adhesive bond usually in the presence of moisture between the aggregate surface and the asphalt cement in a bituminous mixture. Unfortunately, there has been no universally accepted 12 A. Aksoy et al. / Construction and Building Materials 19 (2005) 11–18 method to evaluate the proposed aggregate–asphalt combinations and to determine their water susceptibility potential or the effectiveness of various anti-stripping agents [2]. Tests for the evaluating stripping potential may be divided into two types. There are tests to which visually estimate stripping which measure the time-to-disruption of mix specimens stressed in some manner in the presence of water and tests which measure the change in mechanical properties of mix specimens exposed to water in some type of conditioning scheme [3]. The main objective of this research is to study the effects of four additives (Wetfix I, Lilamin VP 75P, Chemcrete and rubber) on stripping of asphalt mixtures. By adding additives to the asphalt and aggregate combinations, it will be determined whether these additives are improving the performance of asphalt pavements or not. Asphalt without an additive was ‘‘mixed’’ using the same procedure as a reference test. Eleven asphalt mix groups were prepared and tested, each having 30 specimens. Thus engineering properties evaluated on the 330 specimens containing additives regarded: Marshall Stability and flow, Marshall Stability ratio, Marshall Stability–flow ratio, indirect tensile strength, and effects of moisture by vacuum-saturation and Lottman treatments. 1.1. Literature review Shuler and Douglas [4] investigated stripping problem in open graded asphalt mixtures. Three different conventional and polymer-modified asphalt cements were used with both anti-stripping agents and hydrated lime. Tests were apted in optimal bitumen. It was concluded that additives decrease stripping. Ramaswamy and Low [5] indicated that amino additives develop stripping resistance and Marshall Stability values were found higher than in the control mixtures. Law [6] pointed out that amino addition to the bituminous mixtures increases up to 25% the service life of the road. Ramanathan et al. [7] determined the adhesion strength between eight asphalt cements and two aggregates and claimed that while there is no important difference in one aggregate–eight asphalts mixtures, it is concerned statistically big difference when using different type of aggregate. Parker [8] investigated stripping problem with siliceous gravels and dolomitic limestones and cleared that siliceous mixtures showed less tension strength like expected. Unfortunately dolomitic limestones did not showed the same trend. Tension strength of the dolomitic limestone mixtures increased unlike the expecting result. Kennedy indicated that [9] indirect tensile test could be used for determining water sensitivity of the asphalt mixtures and showed good results. Gharaybeh [10] concluded from his study that there is a parallel interaction between the visual stripping test and indirect tensile test results. Maupin [11] founded misleading results using tensile strength ratio for fatty amine mixed asphalt pavements. 2. Materials and methods 2.1. Aggregate and asphalt cement Aggregate coming from Alacatlı was chosen due to its low stripping resistance. Chemical composition was listed in Table 1. 60–70 penetration asphalt cement was used. Engineering properties of the asphalt cement was presented in Table 2. 2.2. Additives Wetfix I and Lilamin VP 75P are liquid heat stable anti-stripping agents specially designed to improve the adhesion between bitumen and aggregate in hot-mixed asphalt. The heat stability makes the Wetfix I and Lilamin VP 75P possible to store in tank for up to one week. Also they can be injected directly into the bitumen storage tank. The dosage of these anti-stripping agents is normally between 0.1% and 0.6% by weight of bitumen, depending on the aggregate and the bitumen used. Wetfix I is a mixture of alkyl and alkaline amines. These agents are manufactured by Scan Road of Nobel Industries Sweden and their physical properties were given in Table 3. Although these anti-stripping agents are generally added to the asphalt mix, this study was done by adding them in the bitumen. Chemcrete modifier is a catalyst which changes the molecular structure of bitumen through a series of chemical reactions which are dependent on the temperature and oxygen. Usually, the modifier is added to bitumen at a rate of 2% by weight. Since the Chemcrete polymerization reaction with bitumen involves air, no essential changes in the bitumen properties occur until it has been spread in a thin film over a large surface. Therefore, all normal procedures and practices can be Table 1 Chemical composition of Alacatlı aggregate Properties Value pH Silicon dioxide, SiO2 (%) R2 O3 (Al2 O3 + Fe2 O3 ) (%) Ferric oxide, Fe2 O3 (%) Sulphur trioxide, SO3 (%) Aluminum oxide, Al2 O3 (%) Calcium oxide, CaO (%) Magnesium oxide, MgO (%) Insoluble residue (%) 8.80 2.60 0.15 0.00 0.04 0.15 50.95 5.00 3.00 13 A. Aksoy et al. / Construction and Building Materials 19 (2005) 11–18 Table 2 Conventional rheological properties of asphalt cement Test Standard AC 60–70 Penetration (100 g, 5 s, 25 °C), 0.1 mm Penetration (200 g, 60 s, 4 °C), 0.1 mm Penetration ratio Ductility (25 °C, 5 cm/min), cm Ductility after loss of heating test, cm Solubility in trichloroethylene, % Softening point, °C Flash point, °C Loss of heating, % ASTM ASTM ASTM ASTM ASTM ASTM ASTM ASTM ASTM 75 25 33.3 100+ 100+ 99.7 48.5 260+ 0.02 Properties of the TFOT Residue Spot test Penetration (100 g, 5 s, 25 °C), 0.1 mm Ductility (25 °C, 5 cm/min), cm Specific gravity, g/cm3 Viscosity at 135 °C, cSt AASHTO T102-83 ASTM D5-73 ASTM D113-79 ASTM D70-76 ASTM D2170-85 used for the mixing, placing and compaction of asphalt materials produced with Chemcrete-modified bitumen. After mixing and placing, the properties of the Chemcrete-modified bitumen will be changed gradually until the catalytic process stops. The primary effect of this process is that some of the weak electrostatic forces, which link conventional asphalt molecules, are replaced with strong irreversible chemical bonds. Different tests performed by the manufacturer have shown that the modification of the rheological properties induced by the Chemcrete modifier produces on asphalt with decreased temperature susceptibility, thus improved high temperature strength, and improved deformation resistance, greater adhesion [12]. Therefore Chemcrete modified bitumen will improve the performance of the pavement. The Transport and Research Laboratory stated that Chemcrete binder has a significant effect on the predicted life of the road structure increasing it by a factor of 4–6 over the same thickness of conventional bituminous construction. So for a given design life the instruction of Chemcrete would lead to a reduction in the design thickness of bound materials of between 15% and 30% [13]. Rubber is added to the asphalt during mixing. The rubber ingredient (1.2, 0.6 and 0.4 mm of size) was mixed the asphalt at 165–200 °C temperature, 0.25% by total weight of blend. The major purpose of using the rubber was to decrease the temperature susceptibility Table 3 Some properties of the Wetfix I and Lilamin VP 75P Properties Wetfix I Lilamin VP 75P Visual appearance at 20 °C Pour point, °C Flash point, °C Density at 20 °C, g/cm3 Brown liquid Brown liquid <0 >160 0.975 )15 120 1.010 D5-73 D5-73 D5-73 D113-79 D113-79 D2042-76 D36-76 D92-78 D1754-78 – 61 100+ 1.038 166.97 and increase the resilient modulus. That is, the rubber asphalt improves the resistance against the deformation due to its high viscosity, and alleviate the initiation and propagation of the reflection crack and the temperature crack due its resiliance modulus. The rubber asphalt also increases the oxidation resistance of the binder by the antioxidants and enhances the lasting quality of the asphalt mix by building a thick film around the aggregate particles. The rubber is added to the mix either in the plant or in a premix manner [14]. 2.3. Effects on the properties of asphalt mixtures Eleven series of asphalt mixtures were prepared with the combination of AC 60–70 pen. Asphalt and four types of additives, namely anti-stripping agents (Wetfix I and Lilamin VP 75P), Chemcrete-modified asphalt, the rubber. The rubber ingredient (1.2, 0.6 and 0.4 mm of size) was mixed the asphalt at 165–200 °C temperature, 0.25% by total weight of blend. 2.3.1. Moisture sensitivity on loose mixtures ASTM D1664 Test Method for uncompacted (loose) mixtures was used. In this test, coarse aggregate (9.5– 6.35 mm) coated with asphalt cement is immersed in water for 24 h and the degree of stripping is determined by visual inspection after a condition time. Test results were given in Table 4. 2.3.2. Mixture design Alacatlı aggregate was used with the AC 60–70 (with and without additives) brought from Alia
ga refinery. To evaluate the characteristics of each material, standard tests were performed and results for each material were presented (see Fig. 1). Los Angeles Abrasion test result of the aggregate was found to be as 20.1%. Sodium sulphate soundness was 14 A. Aksoy et al. / Construction and Building Materials 19 (2005) 11–18 Table 4 Visual stripping resistance of aggregate with the two different asphalt and additives Additive choice Visual stripping resistance (for loose mixtures) retaining (%) Control With Wetfix I (0.2%) With Wetfix I (0.4%) With Wetfix I (0.6%) With Lilamin VP 75P (0.2) With Lilamin VP 75P (0.4) With Lilamin VP 75P (0.6) With rubber (1.2 mm) With rubber (0.6 mm) With rubber (0.4 mm) With Chemcrete AC 60–70 AC 120–150 25–30 90–95 95–100 95–100 60–65 80–85 80–85 75–80 80–85 80–85 85–90 30–35 70–75 80–85 85–90 60–65 70–75 85–90 65–70 70–75 70–75 80–85 Specific gravities (g/cm3 ) Fraction Coarse aggregate Fine aggregate Filler aggregate Effective specific grade of blended aggregate Bulk specific grade of blended aggregate Apparent specific grade of blended aggregate Apparent Bulk Standard 2.707 2.697 2.716 2.690 2.680 2.667 – – ASTM ASTM ASTM ASTM C127 C128 C128 D2041 2.675 2.702 Determining optimal asphalt cement content both conventional and modified asphalt mixtures Marshall Method (ASTM D1559) was applied. Three identical samples were prepared for all alternatives (same asphalt cement content) and seven different addition ratios were used. The results of the Marshall Test results were summarized in Table 7. Eleven series of Marshall specimens were prepared and tested, including 30 reference specimens containing no additives. For each aggregate-binder combination, 24 specimens were selected according to their bulk specific gravities. The other six specimens were disregarded. All specimens were prepared using at 5.0% asphalt content which is the optimum asphalt content of Alacatlı aggregate and Alia
ga asphalt (see Table 8). 100,00 90,00 Percentage Passing, % Table 6 Specific gravities of aggregate 80,00 70,00 60,00 50,00 40,00 30,00 20,00 10,00 Table 7 Summary of Marshall design results 0,00 0,01 0,10 1,00 10,00 100,00 Sieve Size, mm Design parameters Values Board specification in Turkey Minimum Maximum Fig. 1. Aggregate gradation chart. performed and it was found as 1.21%. The chemical properties of aggregate are listed in Table 1. The properties of the AC 60–70 penetration asphalt cement had been given in Table 2. The combined gradation of aggregates used in the study is given in Table 5. Aggregate-specific gravities were presented in Table 6. Bulk specific gravity, Gmb Marshall Stability, kg Air voids, Pa , % Void filled with asphalt, Vf , % Flow, F, 1/100 in. Filler/bitumen Asphalt cement, Wa (by weight of agg.) 2.408 – 1160 3.6 77 900 3 75 3.20 1.24 5.0 2 – – – 5 85 4 1.5 Table 5 Design gradation of aggregates Sieve Sieve (mm) Passing (%) Lower–upper limits 3/4 in. 1/2 in. 3/8 in. No. 4 No. 10 No. 40 No. 80 No. 200 19.1 12.7 9.52 4.76 2 0.42 0.177 0.074 100 86 75 58 35 13.8 9.1 6.2 100 77–100 66–84 46–66 30–50 12–28 7–18 4–10 Table 8 Samples identification Type of additive Additive content Control Low Wetfix I c Lilamin VP 75P Rubber Chemcrete Medium High wl (0.2%) wm (0.4%) wh (0.6%) ll (0.2%) lm (0.4%) lh (0.6%) rs (0.4 mm) rm (0.6 mm) rh (1.2 mm) cc (2%) A. Aksoy et al. / Construction and Building Materials 19 (2005) 11–18 The Marshall specimens were prepared from each batch. Compaction was accomplished by a mechanical Marshall compactor with 75 blows per side at 135 °C for all specimens. Specimens with Chemcrete-modified asphalt were placed in the oven for 2 h before compaction to maintain its chemical reaction. 2.3.3. Moisture damage on compacted mixtures Resistance to moisture and effect of additives on moisture-induced damage of asphalt concrete mixtures were evaluated by using Marshall conditioning (24 h at 60 °C), and retained tensile strength ratio after vacuum saturation and after Lottman-accelerated moisture conditioning (vacuum saturation followed by freezing and warm water soaking). The Lottman wet-to-dry tensile strength ratio should be used in conjunction with the ACMODAS (Asphalt Concrete Moisture Damage Analysis System) computer program developed by Lottman and Leonard to predict changes in fatigue life because of additives and to predict field life benefit-to-cost ratios for different additives [15]. Initially, 15 Marshall specimens were prepared at the same bitumen content (5%) for the asphalt mixes with and without an additive. Of these 15 specimens, three specimens yielded different bulk specific gravities as with respect to the remaining 12 specimens. The remaining specimens are then divided in two groups; the average specific gravity of the specimens of the each group shall be equal. First group specimens were placed in water bath at 60 °C for 35 min. And then loaded at a ratio of 2 in./min, and the stability and flow values were recorded (which is named unconditioned specimens). The second group of specimens was placed in water bath at 60 °C for 24 h. And then the same loading as described above was applied. The Marshall Stability ratio (MSR) was then found by using the average stability of each group using the following formula MSR ¼ 100
(MScond: / MSuncond: ), where MSR: Marshall Stability ratio, MScond: Average Marshall Stability for unconditioned specimens (kg) and MScond: Average Marshall Stability for conditioned specimens (kg). An index of retained stability can be used to measure the moisture susceptibility of the mix being tested. A ratio of stabilities for ‘‘conditioned’’ specimens to ‘‘unconditioned’’ specimens is the criterion to identify a moisture susceptibility of a mix [3]. The ratio of retained Marshall Stability to flow (MSFR) and the ratio of average Marshall Stability to flow for each group of specimens were determined using the following formula MSFR ¼ 100
((MS/F)cond: /(MS/ F)uncond: ) where MSFR, Marshall Stability flow ratio; (MS/F)cond: Ratio of average Marshall Stability to flow for conditioned specimens (kg/mm) and (MS/F)uncond:; Ratio of average Marshall Stability to flow for unconditioned specimens (kg/mm). 15 Indirect tension test involves loading a cylindrical specimen with vertical compressive loads. This generates a relatively uniform tensile stress along the vertical diametral plane. Failure usually occurs by splitting along this loaded plane. Fifteen specimens of each mixture were prepared to determine the tensile strength values. Of these 15 specimens, three were rejected which have different bulk specific gravity. The remaining 12 specimens were divided into two groups (six specimens each). The two set of cylindrical specimens 2.5 in.
4 in. diameter was compacted to the expected pavement density. First set was preconditioned by vacuum saturation. That is, 55– 80% of the air voids were filled with water. Specimens showing above 80% saturation after the vacuum soaking were discharged since they were accepted as severely saturated; this process was repeated with a new specimen. If saturation has not reached 55% in a conditioned specimen after the initial vacuum soaking, then the specimen was returned for additional vacuum soaking until a minimum saturation level of 55% is reached. These specimens were weighted in water and saturated surface-dry conditioned in air. Results related with saturation levels are given in Table 9 for only control specimens. Specimens were wrapped in plastic bags and put in a freezer for 16 h at )18 °C. After the specimens were put into a water bath for 24 h at 60 °C, finally they were placed in a water bath for 2 h at 25 °C. Second set is tested at 25 °C in indirect tension at 2 in./min deformation rate. The load at failure was determined. The tensile strength of specimens was found by the following formula; St ¼ ð2
Pult Þ= ðp
d
tÞ, where St , tensile strength of conditioned or unconditioned specimens, psi; Pult , ultimate applied load required to fail specimens, lb; t, thickness of the specimens, inches; d, diameter of the specimens, in. The indirect tensile strength ratio (ITSR), which is calculated as the ratio of preconditioned indirect tensile strength to dry indirect tensile strength) that is ITSR ¼ (Stcond: /Stdry )
100, where Stcond: Average tensile strength of conditioned specimens, psi and Stdry average tensile strength of unconditioned specimens, psi is used to predict the stripping susceptibility of the mixtures. The minimum ITSR necessary to ensure good pavement performance has not been identified in T€ urkiye, however 0.70 is generally considered to be reasonable minimum value. Mixtures with tensile strength ratios less than 0.70 are moisture susceptible and mixtures with ratios greater than 0.70 are relatively resistant to moisture damage [16]. Marshall Stability test results were summarized in Table 10 and indirect tensile test results were given in Table 11. In Fig. 2, the relationships between Marshall Stability (MS) and type of asphalt mixes, Flow, F and type of asphalt mixes for both conditioned and unconditioned 16 A. Aksoy et al. / Construction and Building Materials 19 (2005) 11–18 Table 9 Degree of saturation in Marshall specimens for control specimens (Gmm ¼ 2.500) Sample number Weight in air (A) Weight in water Sat. Surf. Dry. (C) Weig. (B) Volume of specimen (D ¼ B ) C) Bulk specific grav- Air voids ity (Gmb ¼ A=D) Pa ¼ ðGmm  Gmb Þ=Gmm Before saturation C6 1201.0 C8 1202.9 C13 1204.1 C15 1189.5 C24 1200.5 C25 1192.6 701.5 701.6 701.3 693.9 700.5 691.4 1202.3 1203.5 1204.6 1189.6 1201.8 1193.6 500.8 501.9 503.3 495.7 501.3 502.2 2.398 2.397 2.393 2.400 2.397 2.377 4.1 4.1 4.3 4.0 4.1 4.9 After saturation Weight in water (E) Sat. Surf. Dry. Weig. (F) Volume of specimen (G ¼ F ) E) Bulk specific gravity (Gbw ¼ A=G) Air voids Paw ¼ ðGmm Gbw Þ=Gmm Degree of saturation (S*) S ¼ ðF  AÞ= ðPaw GÞ10; 000 C6 C8 C13 C15 C24 C25 1216.2 1216.6 1218.4 1203.0 1214.4 1205.1 500.2 499.8 502.0 495.2 499.3 497.1 2.401 2.407 2.399 2.402 2.404 2.399 4.0 3.7 4.1 3.9 3.8 4.0 76.8 73.5 70.2 69.6 72.8 62.3 716.0 716.8 716.4 707.8 715.1 708.0 Table 10 Summary of effects of additives on mixture properties for Marshall Test Property Control Wetfix I (0.2%) Wetfix I (0.4%) Wetfix I (0.6%) L.VP 75P (0.2%) L.VP 75P (0.4%) L.VP 75P (0.6%) Rubber (1.2 mm) Rubber (0.6 mm) Rubber (0.4 mm) Chemcrete (2%) Marshall test results Bulk specific gravity (Gmb ) Air voids, % (Pa ) Stability 35 min Flow F1 Stability 24 h at 60 °C MS1 (mm) at 60 °C MS2 (kg) (kg) Flow F2 MSFR, % (mm) 2.408 2.398 2.412 2.411 2.412 2.420 2.417 2.420 2.414 2.419 2.394 3.7 4.1 3.5 3.6 3.5 3.2 3.3 3.2 3.4 3.2 4.2 1290 1371 1264 1221 1330 1376 1442 1224 1410 1439 2232 3.16 3.37 3.13 3.23 3.39 3.50 3.98 3.39 3.22 3.11 3.44 2.78 2.86 2.88 3.07 3.00 3.42 3.66 3.06 2.95 2.88 3.21 1178 1351 1202 1158 1274 1366 1416 1181 1309 1412 1662 ðMS2 =F2 Þ ðMS1 =F1 Þ 80.4 84.3 87.5 90.1 85.0 96.9 90.1 86.8 85.6 91.1 69.3 MSR, % (MS2 =MS1 ) 91.3 98.5 95.2 94.8 95.8 99.3 98.2 96.5 92.8 98.1 74.5 Table 11 Summary of effects of additives on mixture properties for indirect tensile test Property Control Wetfix I (0.2%) Wetfix I (0.4%) Wetfix I (0.6%) L.VP 75P (0.2%) L.VP 75P (0.4%) L.VP 75P (0.6%) Rubber (1.2 mm) Rubber (0.6 mm) Rubber (0.4 mm) Chemcrete (2%) Indirect tensile test results Bulk specific gravity (Gmb ) Air voids, % (Pa ) 2.397 2.417 2.394 2.392 2.390 2.418 2.408 2.415 2.414 2.419 2.423 4.1 3.3 4.2 4.3 4.4 3.3 3.7 3.4 3.4 3.2 3.1 Maximum load (kg) P(dry) 525 424 610 658 740 868 1088 1036 941 929 1971 St (psi) (dry) 74.3 60.8 86.2 93.2 105.1 125.5 158.3 149.9 134.8 133.5 283.5 Maximum load (kg) P(cond.) 255 337 409 358 402 772 800 325 379 505 1604 St (psi) (cond.) ITSR St (cond.)/ St (dry) 36.0 48.3 57.6 50.7 57.2 111.8 116.6 47.1 54.5 72.3 230.4 0.485 0.794 0.668 0.544 0.544 0.891 0.737 0.314 0.404 0.542 0.813 17 A. Aksoy et al. / Construction and Building Materials 19 (2005) 11–18 100 2500 Marshall Stability Ratio,% MS2 (24 hrs. stability) 2000 o Marshall Stability at 60 C, kg MS1 (35 min. stability) 1500 1000 80 60 40 20 500 0 c wl wm wh ll 0 c wl wm wh ll lm lh rs rm rh cc rm rh cc increased the Marshall Stability ratio as compared to control mixes except for Chemcrete. Fig. 5 shows the relationship between Marshall Stability–flow ratio (MSFR) and types of asphalt mixes. All additives increased the MSFR except for Chemcrete. However, Chemcrete increased Marshall Stability and flow, the ratio of stability over flow was decreased. Indirect tensile strength versus asphalt mixes type in Marshall Specimens was shown in Fig. 6 for Stcond: and 100 Marshall Stability Flow Ratio,% situation are illustrated, respectively. There were significant increases in both conditioned and unconditioned stability due to the addition of Chemcrete (CC) as compared to control mixes (C). Slight increase in stability for Wetfix I at 0.2% (WL), Lilamin VP 75P at all percent (LL, LM, LH) and rubber at 0.6 and 0.4 mm of size (RM and RS) were achieved. One of the reasons for increased stability for Chemcrete was that Chemcrete-modified asphalt had high viscosity. Increases in the percent of Wetfix I yielded decrease in Marshall Stability, thus Wetfix I should be used at low percents. By increasing the amount of Lilamin VP 75P, higher Marshall Stabilities are obtained. On the other hand, better performances were observed with smaller sizes of rubber additives. As shown in Fig. 3 all additives increased slightly Marshall Flow for both conditioned and unconditioned specimens compared to control mixes. In Fig. 4, the relationship between Marshall Stability ratio (MSR) versus type of asphalt mixes was shown. All additives rs Fig. 4. Effects of additives on Marshall Stability ratio. Mixture Fig. 2. Effects of additives on Marshall Stability. lm lh Mixture 80 60 40 20 0 c wl wm wh ll lm lh Mixture rs rm rh cc Fig. 5. Effects of additives on Marshall Stability flow ratio. 300 dry mixtures Indirect Tensile Strength, psi 6 F1 (35 min. flow) 4 o Flow at 60 C, mm F2 (24 hrs. flow) 2 0 250 conditioned mixtures 200 150 100 50 0 c wl wm wh ll lm lh rs rm Mixture Fig. 3. Effects of additives on Marshall flow. rh cc c wl wm wh ll lm lh rs rm rh Mixture Fig. 6. Effects of additives on indirect tensile strength. cc 18 A. Aksoy et al. / Construction and Building Materials 19 (2005) 11–18 The relative performance of different additives seems to be different. The choice for any field application should therefore be made on the basis of field trials or at least by conducting a simulation model study to assess their relative performance because long term effectiveness of the amino additives is still controversial. Indirect Tensile Strength Ratio, % 100 80 60 40 References 20 0 c wl wm wh ll lm lh rs rm rh cc Mixture Fig. 7. Effects of additives on indirect tensile strength ratio. Stuncond: Chemcrete, Lilamin VP 75P, and rubber increased tensile strength significantly; Wetfix I increased the strength slightly. Tensile strengths of dry specimens and moisture-conditioned specimens generally increased when additives were used. The increase in conditioned strength was less than the increase in dry strength. The relationship between ITSR and different type of asphalt mixtures was illustrated in Fig. 7. The specimens with 0.4% of Lilamin VP 75P have the maximum indirect tensile strength ratio (89.1). There were little differences between Wetfix I-modified asphalts and control mixes in terms of stability and tensile strengths. Rubbermodified asphalts increased the stability and indirect tensile strengths but decreased the indirect tensile strength ratio. Chemcrete significantly increased the Marshall Stability and tensile strength values of all mixes, so Chemcrete increased the moisture resistance of mixes as determined by the Lottman procedure. 3. Conclusion The additives used in this study reduced the level of moisture-induced damages in asphalt mixtures. Minimum acceptable indirect tensile strength ratio (0.70) is achieved when Chemcrete and 0.2% of Wetfix I, and 0.4–0.6% of Lilamin VP 75P are used in asphalt mixtures. Indirect tensile strength ratio may decrease due to the relatively higher strength obtained in dry specimens with respect to the conditioned ones. Indirect tensile strength ratios of asphalt paving specimens were found to be less than the Marshall Stability ratios. [1] Cawsey DC, Raymond-Williams RK. Stripping of macadams performance tests with different aggregates. Highways and Transportation 1990;(July):16–21. [2] McGennis RB, Machemehl RB, Kennedy TW. Stripping And Moisture Damage In Asphalt Mixtures. Research Report 253-1, Center For Transportation Research, Bureau Of Engineering Research, The University Of Texas, Austin; 1981. [3] Asphalt Institute, Cause and prevention of stripping in asphalt pavements, educational series – 10. 2nd ed. Maryland: College Park. [4] Shuler S, Douglas I. Improving durability of open-graded friction courses. Transp Res Rec 1990;1259:35–41. [5] Ramaswamy SD, Low EW. The effects of amino antistrip additives on stripping of bituminous mixes. High Transp 1990:9–13. [6] Law EW. Anti-stripping agent for roads compound for bituminous pavements. Department of Civil Engineering, National University of Singapore; 1995. [7] Ramanathan K, Stallings RL, Newsome JR. Ultrasonic technique for the measurement of adhesion of asphalt to aggregate. J Adhesion Sci Technol 1991;5:181–90. [8] Parker, F. A field study of stripping potential of asphalt concrete mixtures. Final Report, Alabama Highway Department Project No. ST 2019-6; 1989. [9] Anagnos JN, Roberts FL, Kennedy TW. Evaluation of the effect of moisture conditioning on blackbase mixtures. Research Report 183-13, Center For Transportation Research, The University of Texas at Austin, Texas. [10] Gharaybeh FA. Evaluation of tests to assess stripping potential for asphalt concrete mixtures. PhD, Auburn University; 1988. [11] Maupin, GW. Effectiveness of anti-stripping in the field. Virginia Transportation Research Council, VA 22903-0817 Charlottesville, Edgemont; 1995. [12] Extracts of projects/specifications in various countries using chemcrete modifier. Chemcrete International, California; 1987. [13] Nunn, ME, Powell, WP, Colwill, DM. Theoretical assesment of the contribution of chemcrete binder to the performance of road pavements, TRRL, No. 131; 1986. [14] Estakhri, CK, et al. Use availability and cost effectiviness of asphalt rubber in Texas, TTI:2-8-90-1902-1F, Texas Transportation Institute, Texas, September; 1990. [15] Lee, DY, Demirel, T. Beneficial effects of selected additives on asphalt cement mixes. Iowa Department Of Transportation PROJECT HR-278, Iowa, August; 1987. [16] Kenedy, TW, Anagnos, JN. Wet-dry indirect tensile test for evaluating moisture susceptibility of asphalt mixtures. Research report 253-8, Center For Transportation Research, University Of Texas, Texas, November; 1984.