J. Cent. South Univ. (2013) 20: 256−266 DOI: 10.1007/s11771-013-1483-1
Long term performance of warm mix asphalt versus hot mix asphalt Ziari Hasan, Behbahani Hamid, Izadi Amir, Nasr Danial School of Civil Engineering, Iran University of Science and Technology, Tehran, 1684613114, Iran © Central South University Press and Springer-Verlag Berlin Heidelberg 2013 Abstract: The fatigue behavior, indirect tensile strength (ITS) and resilient modulus test results for warm mix asphalt (WMA) as well as hot mix asphalt (HMA) at different ageing levels were evaluated. Laboratory-prepared samples were aged artificially in the oven to simulate short-term and long term ageing in accordance with AASHTO R30 and then compared with unaged specimens. Beam fatigue testing was performed using beam specimens at 25 °C based on AASHTO T321 standard. Fatigue life, bending stiffness and dissipated energy for both unaged and aged mixtures were calculated using four-point beam fatigue test results. Three-point bending tests were performed using semi-circular bend (SCB) specimens at –10 °C and the critical mode I stress intensity factor KI was then calculated using the peak load obtained from the load–displacement curve. It is observed that Sasobit and Rheofalt warm mix asphalt additives have a significant effect on indirect tensile strength, resilient modulus, fatigue behavior and stress intensity factor of aged and unaged mixtures. Key words: warm mix asphalt; hot mix asphalt; fatigue behavior; resilient modulus; tensile strength; stress intensity factor; ageing
1 Introduction Ageing in asphalt pavements occurs due to bitumen oxidation and volatilization when it is exposed to air especially at high temperatures. This phenomenon causes hardening and material embrittlement with time, which eventually leads to fatigue and thermal cracking as common types of distresses in asphalt pavements. The amount of ageing (ageing rate) has been found to vary significantly depending upon crude source, additives, climate, and characteristics of the mixture. Furthermore, aged mixtures might be less durable than original mixtures in terms of wear resistance and moisture susceptibility [1–3]. In recent years, several technologies have been developed to produce asphalt pavements. Warm mix asphalt (WMA) is one of the newest ones which have recently gained wide popularity. This technology was proposed to address the Kyoto protocol agreement in lowering emission in construction phase. Employing WMA technique also improves working conditions for production and paving workers due to reduced fumes, emissions and odors. Furthermore, the ability to transport it over longer distances to pave at lower temperatures, and a longer paving season are the additional benefits of this technology. The most promising advantage of this technology is reducing asphalt binder ageing due to lowering the mixing and paving temperatures which in
turn results in less cracking, etc [4]. All in all, few investigations are performed on the ageing property of the mixtures. Fatigue cracking is one of the most important damages in asphalt concrete pavements and causes structural damages. In Ref. [5], a four-point beam fatigue test from IPC global company was utilized to conduct the fatigue test in accordance with AASHTO T321 standard. Ageing of asphalt mixtures is the controlling factor leading to the failure in asphalt layers. In this work, the simulation of asphalt mixtures at different ageing levels namely short term and long term was run in a forced-draft oven in accordance with AASHTO R30 standard [5–6]. During the course of short term ageing, asphalt mixture was short-term aged, compacted and then kept in an oven at 85 ºC for 120 h. In this work, the effect of ageing on indirect tensile strength, resilient modulus, moisture susceptibility, fatigue performance and fracture toughness of WMA and hot mix asphalt (HMA) mixtures with two different aggregate gradations was evaluated.
2 Experimental 2.1 Materials The mineral aggregate used in this work was obtained from a limestone source. Table 1 demonstrates engineering properties of the aggregate used. Two different aggregate gradations (A and B) with nominal
Received date: 2012–02–06; Accepted date: 2012–04–04 Corresponding author: Izadi Amir; Tel: +98–9111279814; E-mail address:
[email protected]
J. Cent. South Univ. (2013) 20: 256−266
257
Table 1 Engineering properties of aggregate sources Property
Aggregate A
Aggregate B
Test method
2.493 2.2 4.2 22.3 94
2.512 2.2 4.2 22.3 94
ASTM C127 ASTM C127 ASTM C128 AASHTO T96 ASTM D5821
−3
Bulk density/(g·cm ) Absorption coarse aggregate/% Absorption fine aggregate/% Los Angeles abrasion loss/% Two fractured faces/%
maximum aggregate size (NMAS) of 9.5 and 12.5 mm were used in this work, as shown in Fig. 1. It can be seen that the gradations fall within the upper and lower limits. The asphalt binder utilized to prepare both HMA and WMA mixtures was a PG 64–22.
Table 2 Asphalt binder property Property
Bitumen 60/70
Test method
Density at 25 °C/ (g·cm−3)
1.03
ASTM D-70
Penetration at 25 °C
64
ASTM D-5
Softening point/°C
52
ASTM D-36
Ductility at 25 °C
102
ASTM D-113
Flash point/°C
305
ASTM D-92
Fire point/°C
317
ASTM D-70
different wax additives, namely Sasobit and Rheofalt Sasolwax of Ventraco Company were used. Sasobit® is a crystalline, long chain aliphatic polymethylene hydrocarbon produced from natural gas using the Fischer-Tropsch (FT) process. RheoFalt® product is classified as organic FT waxes. Formerly, these waxes were frequently used in the reclaimed asphalt pavements industry. They are easily melted in the binder at temperatures higher than 110 ºC. Table 3 gives the properties of the wax additives used in this work. Table 3 Properties of WMA additives Property
Rheofalt L T70
Sasobit
Ingredients
Paraffin waxes, Hydrocarbon waxes
Aliphatic polyethylene hydrocarbon
Physical state
Pastille
Pastille, Prill
Color
White
White
Odorless
Odorless
Bulk density/(kg·m )
770
590–622
Flash point/ ºC
210
290
Solubility in water
Insoluble
Insoluble
Dosage/%
2–4
1–3
Odor Fig. 1 Gradations of designated aggregate: (a) Aggregate A; (b) Aggregate B
Table 2 gives the properties of the binder source used in this work. Two types of warm mix asphalt additives, Sasobit and Rheofalt, were utilized. The additives were melted in the asphalt binder at 110 °C, causing reduction in viscosity of the binder. In preparing warm mix asphalt samples, respectively equal to 2.5% and 4% of the weight of the binder, Sasobit and Rheofalt were used. 2.2 Wax additives To prepare warm mix asphalt mixtures, two
–3
2.3 Mix Design The first step in laboratory process was to determine the optimum asphalt content for the asphalt mixtures. To that end, the marshal mix design was used by means of gyratory compactor. The optimum asphalt content values for aggregate A and B were 4.8% and 5.1%, respectively. These values were used to prepare both hot mix asphalt and warm mix asphalt specimens. Table 4 gives
J. Cent. South Univ. (2013) 20: 256−266
258 Table 4 Mixing and compaction temprature Mixture type
Mixing temperature/ºC
Compaction temperature/ºC
Control mix
165
135
Rheofalt mix
135
110
Sasobit mix
135
110
the mixing and compaction temperature for preparing asphalt mixes. 2.4 Samples preparation and test procedures In this work, a linear kneading compactor was applied for compacting fatigue beam specimens in the laboratory environment [7]. All the samples were made with 4% air void at optimum asphalt content. In order to prepare aged samples, prepared loose mixtures were placed in a forced-draft oven at 135 ºC for 4 h, and then they were compacted. Finally, the compacted samples were positioned in a forced-draft oven at 85 ºC for 5 d to simulate long term ageing in service. To avoid a high air void at the sample surfaces, 10 mm from each side of the sample was cut. The final dimensions of prepared beams were 38.5 mm×63.5 mm×50 mm according to AASHTO T321 standard. Four point fatigue beam from the IPC global company was used in this work (Fig. 2). All samples tested at constant strain of 600 micro-strain and haversine mode of loading. All tests were conducted in an environmentally-controlled chamber at a temperature of (25±0.5) °C. Specimens were pre-conditioned at 25 °C for a minimum period of 2 h. For this purpose, samples were placed for at least 2 h at this temperature. The frequency of loading was 10 Hz and the initial stiffness of the specimens is determined in the 50th cycle after the beginning of the test. The test continued until the stiffness dropped to 50% of the initial stiffness. The number of loading cycles was reported as specimens fatigue life. Other variables calculated according to AASHTO T321 are as follows [5]: S / (aP (3l 2 4a 2 )) (4b h3 )
(1)
where S is flexural stiffness (Pa), is tensile stress (Pa), is tensile strain, P is applied load (N), a is space between inside clamps (m), b is beam width (m), is maximum deflection at center of beam (m), h is beam height (m), and l is length of the beam between outside clamps (m). D π sin(2πf )
(2)
where D is dissipated energy (J/m3), f is loading frequency (Hz) and is time lag (s). i n
W
D
i
i 1
(3)
Fig. 2 Fatigue beam samples (a) and four-point beam fatigue apparatus (b)
where W is cumulative dissipated energy (J/m3), and Di is dissipated energy for the i-th load cycle. Cylindrical samples of 100 mm in diameter were prepared using gyratory compactor. Half of the prepared samples were aged artificially in accordance with AASHTO R30-02. Resilient modulus test was conducted for both aged and unaged cylindrical samples at 25 °C [8]. To evaluate the moisture susceptibility, samples were submerged in water bath at 60 °C for 24 h followed by submerging in water bath at 25 °C for 2 h. ITS test was applied on both wet and dry samples to evaluate TSR. The resilient modulus and ITS tests were performed on unaged as well as oven aged samples. To calculate stress intensity factor, cylindrical samples (130 mm in height, and 150 mm in diameter) were prepared using superpave gyratory compactor (SGC) machine. These samples were sliced into three disks with 32 mm in thickness by means of a water-cooled masonry sawing machine. They were then halved to obtain six semi-circular bend specimens. At the next stage, an artificial crack (20 mm in length) was
J. Cent. South Univ. (2013) 20: 256−266
259
generated in the middle of the specimen utilizing a water-cooled cutting machine with a very thin blade. Figure 3 shows a typical generated crack within the SCB specimen and fracture test set-up.
Fig. 4 Resilient modulus of unaged (a) and long term aged (b) samples at 25 ºC Fig. 3 Fracture test set-up (a) and typical generated crack within SCB specimen (b)
3 Results and discussion 3.1 Resilient modulus Resilient modulus is the most common parameter for measuring the stiffness of both field core and laboratory prepared asphalt concrete samples [9]. The resilient modulus at low temperatures is somehow related to thermal cracking. It has been shown that the stiffer mixtures at lower temperatures are more prone to thermal cracking [10]. Figure 4 shows the resilient modulus of the unaged and aged mixtures. The results indicate that mixes containing Sasobit have higher resilient modulus values as opposed to the control mix at 25 °C. 3.2 Indirect tensile strength (ITS) The indirect tensile test applies a constant rate of vertical displacement by applying a compressive load across the diametrical axis of a cylindrical specimen until it fails. It is commonly used to evaluate the potential of stripping and fracture properties of asphalt in the laboratory. To evaluate the effect of ageing on ITS values,
samples were prepared at four different ageing levels. Figure 5 shows the indirect tensile strength changes during different terms of ageing at 20 °C. It can be seen that the more aged the mixture, the higher the ITS value. The increase in ITS value is initially considerable going from unaged to long term aged then reaching a plateau. Fracture energy and tensile strength are known as failure limits. To determine these parameters, indirect tensile strength test is performed. To evaluate cracking performance using indirect tensile test, few parameters such as indirect tensile strength, total fracture energy and fracture energy to failure should be considered. The fracture energy is defined as the work to be done to create the unit area of crack in the specimen, and is equal to the area under the load–deflection curve up to the maximum failure load. The total fracture energy is calculated as the total area under the load–deformation curve [11]. Figures 6 and 7 represent the definition of total and failure energy and fracture energy for unaged and aged mixtures. The fracture energy can be calculated according to the following equation: FE
max
0
P ( )d ( HD)
(4)
where FE is total fracture energy to failure (J/m2), P is
260
Fig. 5 ITS test results of samples: (a) Aggregate A; (b) Aggregate B
J. Cent. South Univ. (2013) 20: 256−266
Fig. 7 Fracture energy calculated for unaged and aged mixtures: (a) Aggregate A; (b) Aggregate B
with control mixes. In the case of aged and unaged mixes, mixtures containing Rheofalt have lower TSR values compared with the other mixes.
Fig. 6 Total fracture work and fracture work to failure
load (N), is deformation (mm), H is thickness of the specimen (mm) and D is diameter of the specimen (mm). In this work, tensile strength ratio (TSR) was used to measure moisture susceptibility of different mixtures [12]. According to superpave specification, asphalt mixtures shall have a minimum tensile strength ratio of 80% during mix design [13]. Figure 8 shows the TSR values for the unaged and aged samples. It can be seen that TSR values for all types of asphalt mixtures in this work are higher than 80% and warm additives do not have a significant influence on TSR values compared
3.3 Fatigue test In this work, in order to determine fatigue behavior of asphalt samples, four point beam test from IPC global company was applied. Fatigue lives of different mixtures are exhibited in Fig. 9. It can be seen that unaged samples have a longer fatigue life than aged samples for both mixtures. Also samples with Aggregate B have higher fatigue life value than samples prepared with Aggregate A. It is observed that mixtures with Sasobit additives have higher fatigue life cycles than control mixtures for both unaged and aged mixtures. In contrast, mixtures with Rheofalt additives have lower fatigue life values than control mixtures. These results show that warm mix asphalt additives have a significant effect on mixture fatigue life due to changing the binder rheology and creating a lattice structure in asphalt mixture. This trend repeats for both aggregates. Results of initial flexural stiffness for different mixtures are shown in Fig. 10. It can be seen that warm mix asphalt samples have a significantly higher flexural
J. Cent. South Univ. (2013) 20: 256−266
261
Fig. 8 TSR values for unaged (a) and aged (b) mixtures
Fig. 10 Flexural stiffness of unaged (a) and aged (b) mixtures
Fig. 9 Fatigue life for unaged (a) and aged (b) mixtures
stiffness than hot mix asphalt samples for both aggregate sizes for unaged samples. However, for aged samples, there is a different trend. Stiffness values for all the mixtures are about the same. It can be inferred that mixtures with warm mix asphalt additives have been less aged as opposed to control mixtures. This suggests that these two additives have a significant effect on asphalt property and also on binder property of sample mixtures due to the changing behavior of the binder property. Figure 11 presents results for cumulative dissipated energy. It can be seen that the values for mixtures containing Sasobit additives are higher than those of control mixtures. This difference is related to the stiffness and fatigue life of the mixtures and Sasobit mixes have higher stiffness and fatigue life than control mixtures for unaged samples. For aged samples, it can be seen that these values are about the same. The slight difference in the cumulative dissipated energy is related to differences in fatigue life of the mixtures. The correlation between cumulative dissipated energy (W) and the number of cycles to failure indicates that parameter W for Sasobit mixture is higher compared with the control mixtures. This shows that the potential of the Sasobit mixtures to dissipate energy before 50% reduction of flexural stiffness (fatigue failure) is more than control mixtures.
J. Cent. South Univ. (2013) 20: 256−266
262
Fig. 11 Cumulative dissipated energy of unaged (a) and long ferm aged (b) mixture
Fig. 12 Stiffness of unaged sample: (a) Aggregate A; (b) Aggregate B
In this work, a natural logarithm model is used to show the relationship between loading cycles versus stiffness and dissipated energy. This model is best fitted with data compared with other models such as exponential and power models. This model is as follows: S ln n
(5)
where S is the sample stiffness at loading cycles, n is loading cycle, and and are regression constants. The decreasing trend of stiffness through repeated loading is plotted in Fig. 12. It can be seen that for all samples, the stiffness value decreases rapidly before 5 000 load cycles. In the next phase, this trend decreases with a slight slope before the end of the test. For aged samples, this modeling is shown in Fig. 13. Dissipated energy reduces as the number of cycles increases. Therefore, cyclic energy dissipation capacity can be employed as a convenient measure in differentiating between non-deteriorating and deteriorating systems. Figures 14 and 15 show the dissipated energy reduction and equation for unaged and aged samples with increasing the number of cycles. Damping influence on energy dissipation for different mixtures is a considerable point. In viscoelastic materials in the case of facing external loading, part of
Fig. 13 Stiffness of long term aged samples: (a) Aggregate A; (b) Aggregate B
J. Cent. South Univ. (2013) 20: 256−266
263
Fig. 14 Dissipated energy for unaged samples: (a)Aggregate A; (b) Aggregate B
Fig. 15 Dissipated energy for long term aged samples: (a) Aggregate A; (b) Aggregate B
dissipated mechanical energy can be converted to thermal energy through viscoelastic damping. This phenomenon leads to reduced material fatigue damage. This part of dissipated energy will not result in fatigue crack extension and should be removed from the total dissipated energy calculation in fatigue failure prediction process [14].
The same result is also true for the resilient modulus except that the interaction effect of three main factors is not significant here. This means that the variability in resilient modulus in different treatments can be referred to three main factors and all their level two interactions but not the level three interactions. It can be deduced from the above three ANOVA tables that the three factors, aggregation, mixture type and mixture condition, undoubtedly have a significant effect on the fatigue life, dissipated energy and resilient modulus.
3.4 Analytical data by variance method The analytical data were obtained using variance method to evaluate the effect of three important factors including gradation, mixture type and condition (aged or unaged) on tests results. Tables 5, 6, 7 present the results of variance method for different test parameters. The first ANOVA table is for fatigue life variables. Obviously, all three factors (gradation, mixture type and condition) and their interactions have a significant effect on the fatigue life of the samples. This can be deduced from the Sig. Column which shows the p-value of each effect. Since all the Sig. numbers are less than 0.01, all the effects are significant with 99% confidence (or some may call it at the level of 0.01). The same result is true for the dissipated energy (Table 6). That is to say, the variability in dissipated energy data can be attributed to all the main effects and their interactions.
3.5 Fracture test The critical stress intensity factor was calculated by replacing fracture load, as presented in Table 8, into Eq. (6). The results, critical stress intensity factor and its average value, for both mixtures have also been presented in Table 8. It is clear that fracture toughness (KIC) for the HMA mixture is higher than WMA mixtures. This phenomenon is associated with the fact that the WMA mixture at the low temperature presents a better elastic behavior than HMA mixture and thus is more brittle. K I YI
Pcr πa 2 Rt
(6)
J. Cent. South Univ. (2013) 20: 256−266
264 Table 5 Tests of between-subjects effects (dependent variable: fatigue life) Source
Mean square
F
Sig.
Sum of squares
df
Corrected model
30 038 647 500
11
2 730 786 136
25 310.48
0.000
Intercept
70 410 622 500
1
70 410 622 500
652 605.84
0.000
Agg
1 249 622 500
1
1 249 622 500
11 582.21
0.000
Mix
1 900 395 000
2
950 197 500
8 806.97
0.000
Age
25 488 122 500
1
25 488 122 500
236 238.47
0.000
Agg * Mix
51 695 000
2
25 847 500
239.57
0.000
Agg * Age
657 922 500
1
657 922 500
6 098.00
0.000
Mix * Age
673 595 000
2
336 797 500
3 121.63
0.000
Agg * Mix * Age
17 295 000
2
8 647 500
80.15
0.000
Error
2 589 396
24
107 891.5
Total
1.004 52×1011
36
Corrected total
30 041 236 896
35 R2=1.000 (Adjusted R2 = 1.000)
A
Table 6 Tests of between-subjects effects (dependent variable: dissipated energy) Source
Sum of squares
df
Mean square
F
Sig.
Corrected Model
2 154.941 4
11
195.903 763 6
1 444.45
0.000
Intercept
6 945.555 6
1
6 945.555 6
51 211.47
0.000
Agg
99.201 6
1
99.201 6
731.44
0.000
Mix
268.627 65
2
134.313 825
990.33
0.000
Age
1 564.993 6
1
1 564.993 6
11 539.12
0.000
Agg * Mix
31.156 65
2
15.578 325
114.86
0.000
Agg * Age
50.979 6
1
50.979 6
375.89
0.000
Mix * Age
118.683 65
2
59.341 825
437.54
0.000
Agg * Mix * Age
21.298 65
2
10.649 325
78.52
0.000
Error
3.255
24
0.135 625
Total
9 103.752
36
Corrected Total
2 158.196 4
F
Sig.
35 R2 =1.000 (Adjusted R2 = 0.998)
A
Table 7 Tests of between-subjects effects (dependent variable: resilient modulus) Source
Sum of squares
df
Mean square
Corrected Model
55 685 639
11
5 062 330.818
1 236.17
0.000
Intercept
811 965 025
1
811 965 025
198 273.99
0.000
Agg
2 640 625
1
2 640 625
644.82
0.000
Mix
4 641 698
2
2 320 849
566.73
0.000
Age
48 066 489
1
48 066 489
11 737.37
0.000
Agg * Mix
143 678
2
71 839
17.54
0.000
Agg * Age
36 481
1
36 481
8.91
0.006
Mix * Age
154 134
2
77 067
18.82
0.000
Agg * Mix * Age
2 534
2
1 267
0.31
0.737
Error
98 284
24
4 095.166 667
Total
867 748 948
36
Corrected total
55 783 923
35
A
R2=0.998 (Adjusted R2 =0.997)
J. Cent. South Univ. (2013) 20: 256−266
265
Table 8 Fracture test results for mixtures prepared from different technologies Mixture type
HMA
HMA (long term aged)
WMA (Sasobit)
WMA (Sasobit) (Long term aged)
WMA (Rheofalt)
WMA (Rheofalt) (Long term aged)
Fcrav /kN
FIf /(MPa· m )
Replicate No.
Fcr/kN
1
4.11
2
4.09
3
4.10
0.8
1
4.47
0.87
2
4.19
3
4.54
0.8 4.1
4.4
0.8
0.82
1
3.28 3.03
3
3.01
0.59
1
3.6
0.7
2
3.74
3
3.42
0.67
1
2.70
0.53
2
2.81
3
2.63
0.51
1
3.20
0.62
2
3.15
3
3.34
4 Conclusions 1) Ageing has an increasingly significant effect on MR values of both HMA and WMA mixtures but this influence is higher for HMA mixtures. This can be due to the greater effect of ageing on HMA mixtures compared with WMA mixtures. Also mixtures with gradation B (NMAS: 9.5 mm) have lower MR values as opposed to mixtures with gradation A (NMAS: 12.5 mm) and warm mixtures containing Sasobit and Rheofalt have higher MR values than control mixtures. 2) The ITS values increase as the ageing level of the mixture increases from unaged to long term aged. Warm mixes show higher tensile strength as opposed to control mixtures. ITS values do not have significant changes from long term aged to extra aged. 3) The warm asphalt additives do not have a significant effect on moisture susceptibility of the mixes compared with the control mixes. 4) Although warm mixtures have greater values of ITS, there is no significant difference between fracture
0.8
0.86
0.88
2
where YI is the geometry factor using the J-integral technique in ABAQUS code, and YI is obtained for the tested SCB specimen, as 3.734. Pcr is the peak load at the load–load line displacement curve which is obtained from the test results. R, t and a are the radius, thickness and crack length, respectively.
FIfav /(MPa· m )
0.64 3.1
3.6
2.71
3.23
0.59
0.73
0.55
0.61
0.61
0.7
0.53
0.63
0.65
energy to failure and total fracture energy for different asphalt mixtures. 5) The critical stress intensity factors for HMA and WMA mixtures are calculated from the fracture loads. It is observed that the critical stress intensity factor of HMA mixtures is higher than that of WMA mixtures. Since cracking in asphalt mixture is inevitable and there has been no way so far to completely prevent this type of pavement deterioration, employing HMA mixtures in very cold climate can be a possible way to retard crack propagation. 6) By comparing the results of fatigue performance of warm mix asphalt versus hot mix asphalt, it will be concluded that wax additives have a significant effect on stiffness values of the mixtures. However, for long term aged samples, there is a different scenario as the stiffness does not increase at the same rate as in control mixtures. In fatigue life, wax additives have a different impact on these mixtures than hot mix asphalt. Mixtures containing Sasobit additives have higher fatigue cycles, up to 50% of the initial stiffness, than hot mixtures and Rheofalt mixtures. This trend repeats for aged mixtures, indicating that ageing has less effect on fatigue life and stiffness values than hot mix asphalt specimens. In the case of cumulative dissipated energy, it can be inferred that Sasobit mixtures can dissipate more energy before failure compared with hot mixtures as control specimens. All in all, although mixture containing Sasobit additives
J. Cent. South Univ. (2013) 20: 256−266
266
increase mixtures stiffness values, it does not have a significant effect on fatigue performance of these mixtures. It seems that like the long term performance of these mixtures, warm mix asphalt specimens are less affected by air in the ageing process.
References
Zheng. Effect of ageing on fatigue properties of asphalt [J]. Journal of Central South University of Technology, 2008, 15(s1): 111−114. [7]
Colorado Department of Transportation (CDOT). Standard method of test for linear kneading compaction of bituminous mixture [S]. Colorado Procedure Laboratory 5116, 2009.
[8]
ASTM. Annual Book Of Standards [M]. Vol 04.03, 2005.
[9]
WANG Lin-bing. Mechanics of asphalt microstructure and
[10]
ROBERTS F L. Hot mix asphalt materials, mixture design, and
micromechanics [M]. New York: McGraw-Hill, 2011: 227−230. [1]
MOUSTAPHA T. Influence of bitumen ageing on asphalt quality, comparison between bitumen and asphalt ageing [D]. Belgium,
construction [M]. Lanham, MD: National Asphalt Pavement
Universiteit Gent Vrije Universiteit Brussel, 2009. [2]
AIREY G D. State of the art report on ageing test methods for
Association Education Foundation, 1996: 173−181. [11]
bituminous pavement materials [J]. International Journal of
resistance for modified asphalt concrete mixtures based on dissipated
Pavement Engineering, 2003, 4(3): 165–176. [3]
BROWN S F, SCHOLZ T V. Development of laboratory protocols for the ageing of asphalt mixtures [C]// Proc 2nd Eurasphalt &
energy concept [D]. Technische Universität Darmstadt, 2010. [12]
International Journal of Pavement Engineering, 2010, 11(2):
Specifications for Binder and Mixtures. Barcelona, 2000, 83–90. PROWELL G C, HURLEY B D. Evaluation of Aspha-Min zeolite for use in warm mix asphalt [R]. Report 05–04. Auburn: National
133–142. [13]
[6]
American Association of State Highway and Transportation Officials
Asphalt Institute. Superpave mix design [S]. Superpave Series No. 2 (SP–02). Asphalt Institute, Lexington, KY, 2001.
Center for Asphalt Technology, 2005. [5]
WILLIAM R, AMIRKHANIAN S, GANDHI T. Laboratory evaluation of warm mix asphalt ageing characteristics [J].
Eurobitume Congress. Session 1: Performance Testing and [4]
FARAG KHODARY MOALLA HAMED. Evaluation of fatigue
[14]
YI H H, HILTON S. Analytical formulation of optimum material
(AASHTO). Standard specification for transportation materials and
properties for viscoelastic damping [J]. Smart Material Structure,
methods of sampling and testing [S]. 2006.
1992, 1: 113–122.
WANG Ji, PANG Ling, WU Shao-peng, LIU Quan-tao, CHEN
(Edited by YANG Bing)