J. Cent. South Univ. Technol. (2008) 15(s1): 135-139

DOI: 10.1007/s11771-008-332-0

Experimental evaluation on high temperature rheological properties of various fiber modified asphalt binders

CHEN Zheng(陈 筝), WU Shao-peng(吴少鹏), ZHU Zu-huang(朱祖煌), LIU Jie-sheng(刘杰胜)

(Key Laboratory of Silicate Materials Science and Engineering of Ministry of Education,

Wuhan University of Technology, Wuhan 430070, China)

Abstract: High temperature rheological properties of fiber modified asphalt binders and impact of the type and content on such properties were studied. Three types of fiber, including polyester (PET), polyacrylonitrile (PAN) and cellulose (CEL), a control content (0%) and four levels of fiber content (2%, 4%, 6% and 8% by total asphalt binder mass) were used with asphalt binders. The high temperature rheological properties, consisting of complex modulus (G*) and phase angle δ, were measured using SHRP’s dynamic shear rheometer (DSR) between 46-82 ℃. Experimental results indicate that the changes of G* and tan δ of fiber modified asphalt binders with the increase of test temperature tend to slow down, and the temperature susceptibility is improved obviously compared to that of original asphalt binder. Fiber modification results in the increase of rutting parameter (G*/sin δ) at high temperatures, the decrease of temperature susceptibility, and further improved high temperature performance of asphalt binder. An excellent correlation exhibits between fiber content and high temperature performance of asphalt binder. Moreover, fiber type also has different influences on the improvement of G*/sin δ, G*/sin δ of PET and PAN fiber asphalt binders are both higher than that of CEL fiber, but G*/sin δ of CEL fiber is still higher than that of original asphalt. However, there is a critical fiber content when fibers start to interact with each other. Therefore, based on the critical fiber content and economic consideration, the optimum fiber contents for various fiber-modified asphalt binders are obtained.

Key words: fiber modification; asphalt binder; rheological properties; high temperature

1 Introduction

As the modern highway transportation has high speed, high traffic density, heavy load and channelized traffic, asphalt concrete continues to reveal various types of distress such as reflective and load associated cracking, rutting and raveling. To solve these problems, technologies of asphalt pavement have changed greatly and many methods have been developed to improve the performance of asphalt pavement, such as the technology of modified asphalt, the stone matrix asphalt (SMA), etc. Among them, adding fibers is a relatively new concept in asphalt pavement technology, although many earlier applications with fibers have been reported. Recently as a new additive, the fiber is a successful alternative by reinforcing asphalt pavement^[1].

Various types of fibers, i.e., mineral, cellulose and polyester fibers, are mixed with asphalt, and this mixture is called fiber modified asphalt binder. Studies show that different fibers offer different sets of toughening mechanisms to strengthen the asphalt composite and slow crack growth. Fibers also increase the available

wetting surface area and behave as binder thickener which reduces asphalt bleeding and adds fiber toughening mechanisms^[2].

The mechanism of fibers affecting asphalt is complex, and its impact upon asphalt binder properties is profound. However, so far there is seldom research on the effect of fiber on asphalt binders, and fiber modified asphalt binder is also poorly characterized scientifically. The effect is an important issue that needs to be investigated. It helps the understanding of the properties of fiber modified asphalt binder^[3].

Commonly, fibers reinforce the binder phase and optimal content depends on types of fibers used strongly. Furthermore, considering economic factors, it is also imperative to determine the optimal fiber content that improves bitumen properties to a satisfactory level at an effective cost. Therefore, the objectives of this study are as follows: to characterize the behavior of fiber modified asphalt binder, to evaluate the effect of fibers on rheological properties of asphalt binder, and to decide the optimum content of fibers mixed with asphalt binder.

2 Raw materials

2.1 Asphalt

AH-70 paving asphalt was used, which was produced by Koch Asphalt Co. Ltd of Ezhou, Hubei Province, China. The physical properties of this asphalt are shown in Table 1.

Table 1 Physical properties of AH-70 paving asphalt

2.2 Fiber

Three fibers including polyester (PET), polyacrylonitrile (PAN) and cellulose (CEL) fibers were used as an additive to mix with asphalt respectively in the process of experiment. And the product’s physical chemistry performances of three fibers are as follows.

1) Polyester fiber: Diameter (16±2.5) ?m, length 6 mm, relative density 1.32-1.40, white, fusion temperature 248 ℃, fire point 538 ℃, tensile strength (517±34.5) MPa, maximal tension strain (33±9)%.

2) Polyacrylonitrile fiber: Diameter 13 ?m, length 6 mm, relative density 1.18, light yellow, fusion temperature greater than 200 ℃, fire point 515 ℃, tensile strength larger than 910 MPa, maximal tension strain (10±2)%.

3) Cellulose fiber: Value of pH (7.5±1), volume density 25-30 kg/m³, maximal length 5 mm, average length 1 mm, and average diameter 45 μm.

3 Preparation of specimens

Fibers were first placed into a 165 ℃ oven for 24 h to ensure moisture-free fiber surfaces, and asphalt stored in a one-quart can was preheated in the can for 2 h in a 165 ℃ oven to make asphalt liquid and ready to mix. An experimental protocol was developed to obtain homogeneous fiber modified asphalt binder. The minitype blender was controlled at a constant mixing speed to ensure the binder homogeneous adequately. In order to investigate the effects of fibers on bitumen, a full range of concentrations, by referring to the optimum content of fiber and asphalt among asphalt mixture, ranged from 0 to 8% by total mass of asphalt binder. At the same time, considering the performances of different fiber-asphalt binders compared with the same proportion of asphalt and fiber, it is necessary to prepare an array of the fiber-asphalt binder samples according to the predefined proportion of asphalt and fiber. And the concentration of 0% represents original asphalt, and is used as the control sample.

During preparation, upon reaching 165 ℃, a pre-weighed amount of fibers was slowly added to the asphalt, while the mechanical stirring was continued under high speed for 60 min to prevent the fibers from possible agglomeration, and the blend became essentially homogenous. After completion, the fiber modified asphalt binder was put into small containers. The blend was cooled to room temperature, sealed with aluminum foil and stored for further testing.

4 Methodology

4.1 Scanning electron microscopy

Scanning electron microscopy was used to observe bitumen-fiber mastics^{[2, 4]}. According to SEM, fiber modified asphalt binders are visible and the main structure can be observed. The dispersed fiber phase is bright while the continuous asphaltene-rich phase remains dark.

4.2 Dynamic shear rheometer

The linear rheological, viscoelastic properties of the asphalt binders were determined by means of dynamic mechanical analysis (DMA) using oscillatory-type dynamic shear rheometer (DSR) testing^[5]. The principle used with DSR is to apply sinusoidal, oscillatory stresses and strains over a range of temperatures and loading frequencies to a thin disc of asphalt, which is sandwiched between the two parallel plates with 25 mm diameter of the DSR. A sample of about 1.0 g was put onto the bottom plate, and the upper plate of the DSR is then gradually lowered to contact tightly with the sample until the required nominal testing gap is 50 μm. The asphalt that has been squeezed out between the plates is then trimmed. Reaching to the required temperature, the bottom plate is fixed while the upper place oscillates at a frequency of 10 rad/s to simulate the loading rate of traffic traveling at highway speeds.

The principal viscoelastic parameters usually measured are the complex shear modulus (G*) and the phase angle (δ). G* is defined as the ratio of maximum (shear) stress to maximum strain and provides a measure results of the total resistance to deformation when the binder is subjected to shear loading. It contains elastic and viscous components, which are designated as the storage modulus (G′) and loss modulus (G″), respectively. These two components are related to the complex (shear) modulus and to each other through the phase (or loss) angle (δ) which is the phase, or time, lag between the applied shear stress and shear strain responses during  tests^[6].

5 Results and discussion

5.1 Scanning electron micrograph

A scanning electron micrograph of three fibers, which was magnified 500 times, is shown in Fig.1. For CEL fiber, Fig.1(a) shows that it presents flocculent, porous and with relatively rough surface. Moreover, the cross section is relatively flat, and the surface area is obviously greater than that of PAN and PET fiber, as mentioned above. These surface characteristics show the efficiency of cellulose fibers to adsorb more asphalt. However, according to PAN and PET fiber, the microstructures of this two fibers differs from those of cellulose fibers since their cross-sections are quite round, with a smooth surface, a smaller surface area and protuberant antennas in the end as shown in Figs.1(b) and (c). It benefits fiber give birth to bridge function and reinforcement, and forms a three-dimensional network in asphalt. This three dimensional structure of fibers can assist in the formation of a thicker structure asphalt film over the fiber and improve the adhesion.

5.2 Dynamic shear properties

Asphalt mixtures are more susceptible to rutting and shoving at high service temperature when the mixture has a lower viscosity and is easier to creep under heavy traffic loading. An asphalt binder with a harder stiffness or higher viscosity should give better rutting resistance to the asphalt mixture. Consequently，increased complex modulus (G*) and elastic modulus (G′) and low tangent phase angle (tan δ) are favorable. The higher the G* value, the stiffer and thus the more resistant to rutting the binder will be. The lower the tan δ, the more elastic the binder is^[7-8].

5.2.1 Effect of fiber types

Fig.2 shows the comparison results of rheological properties of three fiber modified asphalt binders under the same fiber content and original asphalt binder (control sample). It can be seen that G* of the original asphalt decreases sharply with the increase of temperature and is very low at high temperature, while

Fig.1 Microstructures of three fibers (500 times): (a) Cellulose fiber; (b) Polyacrylonitrile fiber; (c) Polyester fiber

tan δ of the original asphalt dramatically increases at the same time. The varying trend of tan δ and G* is slowed down when fiber is added to the asphalt. A similar phenomenon almost appears in all three fiber modified asphalt binders, and the trend of adding PET fiber is most obvious. Nevertheless, the addition of fiber leads to the increase of G* more significantly at elevated temperatures, and the tan δ curve becomes flatter over a wide range of tested temperatures. It indicates that the elasticity of the modified binder is improved effectively with the addition of fiber due to the formation of a crosslinked network in the modified binders. And the performance grade of PET fiber modified asphalt binder is greatly promoted, which appears much more obvious when compared with other fiber modified asphalt binders. Simultaneously, experimental data indicates that the addition of fiber results in the improvement of asphalt binder’s high temperature stability. The value of G* decreases rapidly with the increase of test temperature under the same fiber content. It indicates that asphalt binder characterizes with obvious temperature susceptibility.

Fig.2 G* and tan δ versus temperature for original and three fiber modified asphalt binders under same fiber content

5.2.2 Effect of fiber content

Fig.3 shows the dynamic mechanical properties of three fiber modified asphalt binders with various fiber contents ranged from 0 to 8% by total weight of asphalt binder. For PAN, increasing fiber contents generally led to the increase of complex modulus (G*). At the concentration level of 4%, adding fiber results in the marked increase of G*, as illustrated in Fig.5. PAN fiber contents higher than 4% causes a fiber-fiber entanglement that can result in the reduction of complex modulus (G*). The decrease of complex modulus (G*) is also observed for other fibers when the concentration level is higher than a certain value. Furthermore, a similar phenomenon almost appears in tan δ. The most significant reduction on tan δ occurs with the addition of 4% PAN fibers, and this observation is in good agreement with content determined by complex modulus (G*). Therefore, adding more than 4% might not be economically feasible because of the limited benefits from the increase of G* and the reduce of tan δ. A similar phenomenon almost appears in other two fibers modified asphalt binders, 8% and 6% by asphalt mass for CEL and PET, respectively.

Fig.3 Effects of temperature on rheological properties under different fiber contents: (a) CEL; (b) PAN; (c) PET

6 Conclusions

1) The addition of fiber can improve the high temperature performance of asphalt binder remarkably. And the type and content of fiber have important effects on the high temperature performance of asphalt binder.

2) The stabilizing effect of asphalt mixed with fibers can be explained on the basis of the electron microscopy of the fibers. The three-dimensional network thickens structure asphalt film over the fiber and improves the adhesion between fibers and asphalt, which enhances the performance of fiber modified asphalt binders.

3) The improvement of asphalt binder’s high temperature performance should be emphasized on the synthesis function of stability and reinforcement. The reinforcement of polyester fiber is more obvious, but the stability of cellulose fiber remains prominence.

4) On the basis of viscoelastic properties, the criteria used to select an optimum fiber concentration can be obtained. For fibers tested in this study, the optimum content is found to be 8%, 4% and 6% by asphalt mass for CEL, PAN, and PET, respectively.

References

[1] SHEN Jin-an. Pavement performance of asphalt and asphalt mixture[M]. Beijing: China Communications Press, 2001. (in Chinese)

[2] CHEN J S, LIN K Y. Mechanism and behavior of bitumen strength reinforcement using fibers[J]. Journal of Materials Science, 2005, 40(1): 87-95.

[3] ZOU Gui-lian, ZHANG Xiao-ning, HAN Chuan-dai. Utilization of DSR for evaluation of pavement performance[J]. Journal of Harbin University of Civil Engineering & Architecture, 2001, 34(3): 112-115.

[4] WEN Gui-an, ZHANG Yong, ZHANG Yin-xi, et al. Rheological characterization of storage-stable SBS-modified asphalt[J]. Polymer Testing, 2002, 21(3): 295-302.

[5] JIN Hai-long, GAO Guang-tao, ZHANG Yong, et al. Improved properties of polystyrene-modified asphalt through dynamic vulcanization[J]. Polymer Testing, 2002, 21(6): 633-640.

[6] AIRAY G D, RAHIMZADEH B. Combined bituminous binder and mixture linear rheological properties[J]. Construction and Building Materials, 2004, 18(7): 535-548.

[7] RUAN Yong-hong, DAVISON R R, GLOVER C J. The effect of long-term oxidation on the rheological properties of polymer modified asphalts[J]. Fuel, 2003, 82(14): 1763-1773.

[8] WU Shao-peng, CHEN Zheng, YE Qun-shan, et al. Effects of fiber additive on the high temperature property of asphalt binder[J]. Journal of Wuhan University of Technology: Mater Sci Ed, 2006, 21(1): 118-120.

(Edited by CHEN Can-hua)

Foundation item: Project(2004243) supported by the Science and Technology Key Project of Hubei Province, China

Received date: 2008-06-25; Accepted date: 2008-08-05

Corresponding author: CHEN Zheng, Doctoral Candidate; Tel: +86-27-87162595; E-mail: chenz@whut.edu.cn