Fatigue Behavior of SMA and HMA Mixtures
Fatigue Behavior of SMA and HMA Mixtures
Fatigue Behavior of SMA and HMA Mixtures
a r t i c l e
i n f o
Article history:
Received 1 June 2009
Received in revised form 5 November 2009
Accepted 18 December 2009
Available online 12 January 2010
Keywords:
HMA
SMA
Fatigue
Indirect tensile test
Stiffness modulus
a b s t r a c t
Fatigue crack is a main form of structural damage in exible pavements. Under the action of repeated
vehicular loading, deterioration of the asphalt concrete materials in pavements caused by the accumulation and growth of the micro and macro cracks gradually takes place. The indirect tensile tests was carried out on hot mix asphalt HMA and stone matrix asphalt SMA mixtures comprising different nominal
maximum aggregate sizes NMAS in three temperatures of 5, 25 and 40 C. Stiffness modulus, fatigue lives
and fatigue prediction equation of the mixtures were developed and characterized in terms of aggregate
gradation type, coarseness and neness of gradation, temperature and asphalt content.
2009 Elsevier Ltd. All rights reserved.
1. Introduction
Asphalt pavements in general are facing serious distress problems world-wide. So much has been done to improve the quality
of the mix through research and innovations. However, the problem of fatigue cracking is still persistent on asphalt roads.
The fatigue resistance of a bituminous mixture is its ability to
withstand repeated bending without fracture. Fatigue manifests itself in the form of cracking from repeated trafc loading. It has
been shown that low asphalt content, high air void mixes are prone
to show fatigue cracking, and high asphalt content mixes tend to
rut rst [1]. Previous researches have been reported that there
are several factors affecting the fatigue lives of bituminous mixtures. According to the NCHRP APPENDIX II-I and SHRP-A-404
[2,3], pavements with different mix stiffness, thickness, subgrade
modulus, air void and asphalt content exhibit different fatigue
behaviors.
Numerous models of varying sophistication have been developed to predict the fatigue behavior of asphalt concrete including
elastic, viscoelastic, elastoplastic, viscoplastic, and crack models
that are developed by using different laboratory testing modes.
The most common methods for characterizing the fatigue behavior
of asphalt mixtures are the Wohler approach, dissipated energy
method and fracture mechanics approach [4]. Various kinds of asphalt mixes with different properties show different fatigue behav-
1159
mixture was measured and the fatigue life prediction equation was
obtained using the regression analysis. Fatigue behavior of the two
mixtures was compared to each other and the effect of each mixture parameter was investigated.
3. Research approach
To satisfy the objectives of this study, HMA and SMA specimens
were prepared at optimum asphalt content. Limestone aggregate,
asphalt with 60/70 penetration, and two aggregate gradations
were used in preparing HMA specimens and the same materials
where used in preparing SMA specimens with three aggregate gradations. Fatigue tests were performed on the prepared Marshall
specimens with a diameter of 10 cm and height of 7 cm to evaluate
the effect of aggregate gradation type, temperature, asphalt content, neness and coarseness of gradation on fatigue life and to obtain the fatigue life prediction equations of mixtures on specic
temperature.
3.1. Materials used
This study included two types of wearing course materials, conventional HMA and SMA. The particle size distribution of the
aggregates is shown in Fig. 1 for HMA and Fig. 2 for SMA. SMA mixture types include type A, B and C with a nominal maximum aggregate size of 19, 12.5 and 9.5 mm, respectively and HMA mixture
types include type H1 and H2 with a NMAS of 9.5 and 12.5 mm,
respectively. The limestone was used in the mixture for both
coarse and ne aggregates. The properties of coarse and ne aggregates are given in Table 1. The bitumen used in SMA and HMA is
the 60/70 penetration bitumen. Results of its properties are listed
in Table 2.
Fig. 2. Aggregate gradation curve of SMA mixtures types C, B, and A by the NMAS of
9.5, 12.5 and 19 mm, respectively.
Table 1
Properties of coarse and ne aggregates.
Properties
Value
Coarse aggregates
Los angeles abrasion value (%)
Water absorption (%)
Particle density (g/cm3)
25
1.2
2.622
Fine aggregates
Water absorption (%)
Particle density (g/cm3)
1.8
2.597
Table 2
Physical properties measured of bitumen.
Parameter measured
Value
T228
T49
T53
T201
T201
T201
T51
1.014
60
52
882
407
135
>10
Table 3
Summary of design results.
Parameter
HMA
type H1
HMA
type H2
SMA
type A
SMA
type B
SMA
type C
NMAS (mm)
Asphalt content (%)
Stability
Bulk specic
gravity (g/cm3)
Air void (%)
VMA (%)
VCAmix (%)
VCAmix (%)
Theoretical maximum
gravity (g/cm3)
9.5
5.07
849
2.511
12.5
4.95
800
2.579
19
7.8
12.5
5.5
9.5
6.5
2.305
2.316
2.319
3.5
14.7
N.A
N.A
2.618
3.5
14.4
3.52
17.1
35.55
37.44
2.492
3.56
17.53
35.58
38.46
2.364
3.86
17.85
38.31
38.72
2.381
2.663
3.3. Methodology
Fig. 1. Aggregate gradation curve of HMA mixtures types H1 and H2 by the NMAS
of 9.5 mm and 12.5 mm respectively.
Two laboratory tests that were used in this research are the
indirect tensile stiffness modulus test and the indirect tensile fati-
1160
Table 4
Results of indirect tensile stiffness modulus test and indirect tensile fatigue test for HMA mixtures.
Temperature (C)
Gradation
HMA type H1 (NMAS:9.5 mm)
25
40
Sm (MPa)
Nf
Sm (MPa)
Nf
800
13,623
493,531
800
11,794
422,081
850
13,413
335,471
850
11,824
300,261
900
13,162
172,792
900
11,344
135,224
950
13,412
129,363
950
10,871
98,686
1000
13,529
111,764
1000
10,607
67,576
400
2398
2744
400
2107
2328
500
2200
1299
500
1906
1224
600
2202
444
600
1798
603
700
1957
206
700
1653
404
800
1768
168
800
1452
170
900
1517
80
900
1291
50
100
492
2814
100
462
2099
150
477
1224
150
451
914
200
467
443
200
422
512
250
440
235
250
415
158
300
409
104
300
396
92
1161
Gradation
SMA type A (NMAS:19 mm)
25
40
Sm (MPa)
Nf
Sm (MPa)
Nf
Sm (MPa)
Nf
600
9738
177,781
600
10,127
243,813
800
11,741
423,746
700
9630
147,838
700
10,590
182,639
850
11,482
265,175
800
9533
78,819
800
10,887
143,423
900
11,346
201,493
900
8853
48,303
900
10,597
79,159
950
11,379
125,411
1000
8762
46,412
1000
10,347
58,261
1000
11,431
104,448
50
1970
37,687
100
2190
55,429
200
2485
26,140
100
1565
20,609
200
2155
4969
250
1796
9044
200
1329
2285
300
1937
1999
300
2169
4364
300
1087
733
400
1760
544
400
1988
1109
400
691
163
500
1606
178
500
1812
474
600
1381
42
600
1630
235
10
288
16,242
30
510
36,323
30
618
40,470
30
208
3828
50
496
6598
50
560
9008
50
182
985
75
448
1335
75
500
2440
75
165
323
100
391
644
100
473
1133
100
147
49
150
408
196
150
437
217
200
370
47
200
384
169
Using the regression analysis, the fatigue equations were developed which are in the form of Wohlers fatigue prediction model
(Eq. (1)). For every type of the gradation in every three temperature, the fatigue equations are shown in Table 6.
Nf k1
k2
1
Table 6
Fatigue prediction equations of HMA and SMA mixtures.
Asphalt type
R2
Fatigue equation
SMA type A
SMA type B
SMA type C
HMA type H1
HMA type H2
0.944
0.958
0.987
0.926
0.978
Nf = 4 1013r2.989
Nf = 3 1013r2.869
Nf = 1 1024r6.378
Nf = 2 1026r7.071
Nf = 3 1030r8.578
25
SMA type A
SMA type B
SMA type C
HMA type H1
HMA type H2
0.944
0.970
0.971
0.981
0.961
Nf = 2 109r2.614
Nf = 3 1012r3.787
Nf = 4 1012r3.647
Nf = 5 1015r4.684
Nf = 2 1013r3.823
SMA type A
SMA type B
SMA type C
HMA type H1
HMA type H2
0.910
0.995
0.991
0.983
0.960
Nf = 6 106r2.351
Nf = 4 109r3.405
Nf = 1 109r3.021
Nf = 3 109r2.987989
Nf = 1 109r2.879
Temperature (C)
40
1162
Fig. 3. Coefcient of k1 in fatigue prediction equation for HMA and SMA mixtures in test temperatures.
Fig. 4. Coefcient of k2 in fatigue prediction equation for HMA and SMA mixtures in test temperatures.
1163
tent have smaller fatigue lives comparing with the HMA mixture
by the same NMAS.
Considering the effect of asphalt content on fatigue life of SMA
mixtures, type C by NMAS of 9.5 mm with the asphalt content percentage of 6.5% should have a fatigue life less than type B by NMAS
of 12.5 mm and asphalt content percentage of 5.5%. But the results
do not conrm this conclusion. So it can be inferred from the results that the effect of coarseness and neness of aggregate grada-
1164
5. Conclusions
Fig. 11. Fatigue diagrams of HMA types H1, H2 and SMA type A in 5 C.
Fig. 12. Fatigue diagrams of HMA types H1, H2 and SMA type A in 25 C.
References
Fig. 13. Fatigue diagrams of HMA types H1, H2 and SMA type A in 40 C.
1165