Construction and Building Materials 70 (2014) 379–387

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Experimental investigation on the effect of hydrated lime on mechanical

properties of SMA
A. Shafiei ⇑, M. Latifi Namin
Tehran University, College of Engineering, School of Civil Engineering, Iran

h i g h l i g h t s

 Partial substitution of the filler with HL significantly improves the resistance indices of SMA.
 There is an optimum amount of hydrated lime in SMA mixture.
 Using HL as additive leads to superior performance comparing with partial substitution of HL.
 The HL modified SMA with 9.5 NMAS is much stronger against rutting compared to the SMA of 12.5 NMAS.

a r t i c l e i n f o a b s t r a c t

Article history: One of the most important and costly damages to the pavements, especially in high traffic areas is rutting.
Received 28 October 2013 An optimal use of lime is significantly important with regard to expensive cost of the lime and its improv-
Received in revised form 10 July 2014 ing properties in stone matrix asphalt (SMA). In this paper, simple performance tests (i.e. dynamic mod-
Accepted 23 July 2014
ulus, dynamic creep, and static creep) were used based on NCHRP-465 report. In which the effect of
Available online 24 August 2014
directly added different percentages of hydrated lime filler to the mixture in dry state as a partial substi-
tute of the aggregate materials in comparison with introducing it as an additive to the mixture on per-
Keywords:
formance and mechanical properties of SMA was assessed. For this purpose, the SMA samples with
Asphalt
SMA
two different aggregate gradation and different hydrated lime contents were prepared as a part of filler.
Filler The obtained results indicate that partial substitution of the filler with hydrated lime will cause an
Hydrated lime improvement in the resistance indices like flow number, flow time as well as rutting and fatigue cracking
Rutting factors of SMA, while the permanent deformations of the asphalt decrease. Obviously, it indicates exis-
Dynamic modulus tence of an optimum amount of hydrated lime in mixture. The results obtained from this contribution
Creep can be used to design and implement asphalt mixtures in accordance with the requirements of the con-
tractors, engineers and researchers.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction money. Some European countries, as well as US and Canada have

The road network is usually described as the essential need of
any country, so that this valuable asset consumes a noticeable
amount of the national budget every year for development and
maintenance. Therefore, taking into consideration the limitations
of financial and technological resources, current situation of the
roads must be maintained and even improved by spending much
less money. Implementation of asphaltic pavements with continu-
ous distribution of grain size would possibly lead to damages like
rutting of wheel raceway and bleeding. These damages require
maintenance and rehabilitation in a relatively short time after
the road comes into service. This will involve spending of a lot of
⇑ Corresponding author.
decided to use stone matrix asphalt (SMA) in order to meet some
special expectations of paving such as resistance against rutting,
preventing expansion of reflective cracks. The SMA contains cer-
tain ingredients and changing material, mixing method and mix
design of each component will alter mechanical properties of the
mixture. In asphalt mixture with continuous graded aggregate,
using filler, increase the contact points and the loading capacity,
improve the compressive and shear strengths, and reduce the
deflection. Filler in stone matrix asphalt plays an important role
which is increasing stiffness of the mixture [1]. Different types of
fillers including crushed stone, cement and lime can be used in
the asphalt mixtures. The multifunctional effects of hydrated lime
in open graded aggregate need to be evaluated when added as
additive or part of filler aggregate in mixture.

http://dx.doi.org/10.1016/j.conbuildmat.2014.07.084
0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.
380 A. Shafiei, M. Latifi Namin / Construction and Building Materials 70 (2014) 379–387
2. Objective
Stone matrix asphalts (SMA) need a kind of open grain distribu-
tion for production and enhancement of bitumen content and con-
tact between aggregates at the same time. In fact, a direct contact
between the aggregates will cause stability. Application and devel-
opment of the SMA was initially started within 1980s in some of
European countries. After 1990 when the first SMA projects were
started in Canada and later within some states of America in
1991, application of the SMA mixture became popular in these
regions [2].
Different fillers are used in this mixture. Craus et al. showed in
1978 that the physiochemical properties are dependent on inten-
sity of interfacial adsorption between the bitumen and the filler.
Thus, an active surface of the filler was expected to create a strong
composition between them [3]. Previous studies conducted by Kim
in 1990, 1994, 1995 and 1999 as well as an earlier contribution of
Bahia in 1999 indicates that the fatigue damage and the restorabil-
ity depend heavily on the properties of the bitumen, properties of
bitumen additives, interaction between bitumen and its additives,
and ingredients which affect growth of the micro-cracks in the
mastic [4]. According to the results obtained from the performance
and volumetric tests, and also by establishing a relationship
between the parameters extracted from these tests, finer filler will
make the asphalt mixture stiffer and will eventually reduce the
rutting phenomenon [5]. The crushed stone is a common filler
material. Moreover, the hydrated lime is more effective filler in
comparison with other mineral fillers because accumulation of
greater micro-cracks in the sample from the initial loading until
the final failure will increase fatigue life of the asphalt. The physio-
chemical interactions between the filler and the bitumen in terms
of the fine particles and their surface features affect characteristics
of the fatigue failure [6]. It was 2005 that the effect of hydrate lime
on dynamic modulus and stiffness of the HMA mixture was
reported in some studies [7], while some researches were con-
ducted in 2008 and 2009 on the effect of hydrated lime on strip-
ping and moisture damages due to the bitumen-lime filler
interaction [8,9]. Meanwhile, some research works were directed
in 2010 using hydrated lime in the HMA asphalt to examine the
effect it has on the rutting phenomenon in addition to predict
the cracking behavior in asphalt [10,11].
Also the hydrated lime is added by different techniques to the
hot mix asphalt. The commonest methods for adding the lime to
the mixture include dry, wet and slurry method [12]. Lime can
be proportioned and mixed in HMA in both batch and drum mixers
in the plant. Dry lime can be added to dry aggregate and to wet
aggregate. Moisture levels in wet aggregate are typically about
two to three percent above the saturated surface dried condition
of the aggregate. Lime slurries made from hydrated lime or quick-
lime have also been used. Lime-slurried aggregates are conveyed
directly to the drying and mixing portion of the HMA facility or
placed into stockpiles for marination. Adding dry lime to the
asphalt binder and storing the lime-modified binder prior to mix-
ing with the aggregate has not been practiced widely in the field
[13]. The effect of hydrated lime and mixing approach on perfor-
mance properties of the HMA was evaluated in the same year
[14–16]. One other study investigated the effect of optimal content
of hydrated lime on aging and workability at high temperatures
[17]. In 2011, Lee et al. employed modern experimental techniques
and new models to study the effect of modified HMA with
hydrated lime on rutting and fatigues cracks [11].
The hydrated lime will reduce the cracking as a result of aging,
fatigues or low temperature. The cracking usually occurs after for-
mation of micro-cracks. Fine particles of the hydrated lime are
expected to prevent formation of these micro-cracks [7]. The
hydrated lime as filler improves stiffness and reduces rutting.
Moreover, adding the hydrated lime will not allow the asphalt to
stiffen at low temperatures because the lime will act as inactive
mineral filler at these low temperatures. The hydrated lime will
also reduce the oxidation and ageing effects [8]. As the HMA ages
due to oxidation, hydrated lime reduces not only the rate of oxida-
tion but also the harm created by the products of oxidation. This
effect keeps the asphalt from hardening excessively and from
becoming highly susceptible to cracking (through fatigue and low
temperature (thermal) cracking). Synergistically, the filler effect
of the hydrated lime dispersed in the asphalt improves fracture
resistance and further improves cracking resistance [13]. There
are many observations which show the hydrated lime modified
HMA has better performance of mixture against rutting, fatigue
and thermal cracking [18]. In 1987, Petersen et al. evaluated two
asphalt binders modified with limestone and hydrated lime at
20% by weight of binder. The results of this research indicate that
lime treatment would improve the resistance of the aged pave-
ment to thermal cracking through the reduced aging index. Since
the behavior of HMA mixtures at low temperatures is mainly con-
trolled by the properties of the aged binder, lime treatment would
produce HMA pavements that are highly resistant to thermal
cracking [19]. The lime-modified bitumen demonstrates a greater
potential for dissipating energy through deformation (at low tem-
perature) than the unmodified bitumen. This is a positive effect at
low temperatures because it reduces fracture potential. Although
the filler effect increases low temperature stiffness, fracture tough-
ness is substantially increased. Fracture toughness is the energy
expended in fracturing a material [20]. Nevertheless, base on some
studies, the influence of filler on low temperature cracking was
hypothesized to be independent of filler type. Unlike rutting and
fatigue cracking, adding different amount of hydrated lime as filler
in mixture is not so effective on the thermal cracking [21,22].
Some studies have been conducted recently in 2013 toward fail-
ure mechanics and advanced suggested models, the effect of
hydrated lime on failure performance, and the mechanical proper-
ties of mixtures [23,24].
Permanent deformation is one of the most critical parameters
for pavement design, which is believed to increase with traffic
and tiers pressure. Most of the permanent deformations occur at
the upper layers rather than the subgrade. There are different
approaches to determine the plastic deformation, out of which
repeated load test is used more than the others [1]. In this study
simple performance tests were used in order to evaluate perfor-
mance of asphalt mixtures. Based on NCHRP-465 report, simple
performance test (SPT) method(s) accurately and reliably measures
a mixture response characteristic or parameter that is highly corre-
lated to the occurrence of pavement distress (e.g., cracking and rut-
ting) over a diverse range of traffic and climatic conditions.
Numerous techniques have been suggested to assess pavement
performance and distresses. One of the best available solutions
for this purpose is to perform three simple performance tests,
namely complex modulus (E*), dynamic creep – flow number test
(Fn), and static creep – flow time test (Ft). They yield the best rela-
tionship between the experimental results and the field perfor-
mance [25].
The complex modulus is a way of addressing the relationship
between stress and strain of the viscoelastic material. The complex
modulus test is often conducted on the cylindrical samples which
are exposed to a haversine compressive loading [26]. The complex
modules changes by variation in the loading frequency. A Loading
capacity which models the traffic load much closer than the other
frequencies must be chosen for this test. The dynamic modulus
obtained in this regard is equivalent to resilient modulus which
is used for design purposes [27].
 A. Shafiei, M. Latifi Namin / Construction and Building Materials 70 (2014) 379–387 381

The main objective of the dynamic creep test is to study asphalt Table 3
concrete performance against rutting phenomenon. In other Physical properties of the used hydrated lime.

words, using the repeated dynamic load test in several thousand Property Value
runs and recording the accumulative strain as a function of the cyc- Specific gravity 2.343
lic loading during the test is a method to determine the permanent Dry brightness, G.E. 92%
deformation of asphalt concrete [28]. Mean particle size 2 lm
The static creep test is not only useful for finding the plastic PH 12.6
BET surface area 22.0 m2/g
properties of the material, but also can be used to measure the per- 100 Mesh 94%
manent deformations of the mixture. 200 Mesh 90%
There are many researches investigating the effect of different Apparent dry bulk density (loose) 20 lbs./ft.3
filler on dense grade mixtures. But there are a few studies contain- Apparent dry bulk density (packed) 30 lbs./ft.3
ing the investigation on different content of hydrated lime as par-
tial substitution of filler in SMA in which total amount of filler has
considerable effect on open graded structure of stone matrix
asphalts mixtures. In this paper the effect of adding different con- Table 4
Chemical properties of the used hydrated lime.
tents of hydrated lime as the effective element of open-graded
asphalts is investigated on the mechanical properties of SMA Property Value
through SPT examinations. Ca(OH)2 0.91
CaO-equivalent 70
CaO total 72
3. Methodology CaCO3 2
Insoluble in acid 2
Mechanical properties of the lime-containing SMA mixtures are evaluated in CaSO4 0.6
this research. For this purpose, quality and type of the needed materials are Free H2O 0.8
described first and then, grain distribution and preparation of the samples are MgO 0.5
explained. AT last, method of performing the tests is introduced along with the rea- SiO2 3
son to use the initial parameters for measurement of the performance indexes and Al2O3 0.3
the mixture properties. Meanwhile, some control samples are also made without Fe2O3 0.3
hydrated lime addition to the SMA samples to examine change in the mixture prop- SO3 0.2
erties. Two groups of samples were used based on two types of fillers and two types P2O5 0.05
of grain distributions. The optimal bitumen content was extracted for each group by Na2O 0.04
making the test samples and running the Rice and Marshal Tests. The main samples MnO 0.0045
were made by using the optimal content of bitumen in each of the groups and were
further checked via dynamic modulus, dynamic creep and static creep tests.

3.1. Materials
For making the asphalt samples, a pea gravel of definite specifications (Table 1)
was used with the required quality tests being performed (Table 2) and the 60/70
bitumen as an adhesive for the gravel aggregates. Some fibers may be used to avoid
drain down of the bitumen in the SMA mixtures due to its great content. VIATOP
cellulose fibers were used in this study which was composed of 90% ARBOCEL ZZ
and 10% bitumen. For better mixing of them with the mixture, it is recommended
to powder them first and mix them for 1 min with hot gravel of the mixture and
add the bitumen at the end.
The crushed stone is used as the filler for some samples which are made just for
comparison with the lime-containing samples. Tables 3 and 4 list chemical and
physical properties of the hydrated lime filler, respectively.
3.2. Aggregate gradation
The suggested grain size distribution of NCHRP 9-8 report was used to address
this parameter in making the SMA samples [2].
Taking into account the relatively large range of the applied SMA asphalt and its
being practical, the nominal sizes of 9.5 and 12.5 mm were selected for the first and
second gradations, respectively (Table 5).

3.3. Sample preparation

Table 1
Aggregate specific gravities (kg/m3). With respect to the previous studies, the SMA samples were decided to be made
from 5 different contents of lime and to be tested in 5 different temperatures. In
Grain-size fraction Specific gravity Water
each group and after defining the optimal bitumen content, this value has been kept
absorption (%)
Apparent specific Bulk specific constant and just the hydrated lime content was changed. The lime contents were
gravity gravity chosen from 0 to 7.5 wt% of the aggregate materials increasing in ranges of 2.5 wt%,
in which the mixtures with zero (0%) percentage of hydrated lime were considered
Coarse aggregate 2699 2499 2.5
as control mix to compare the results and make conclusion in each case (Fig. 1).
(AASHTO T85)
The hydrated lime, crushed stone filler and the other aggregates were initially
Fine aggregate 2716 2529 2.7
heated at 180 °C in the oven and they were later mixed with the bitumen at
(AASHTO T84)
155 °C. This method is used in the field as was described as adding dry hydrated
Filler aggregate 2653 – –
lime to the aggregates before blending with the bitumen that prepares aggregate
(AASHTO T100)
surface for better bonding condition. In order to evaluate the simultaneous effect
Total bulk specific 2514 –
of total filler content and lime content of the mixture, a sample of 10% crushed
gravity of aggregate
stone filler and 2% hydrated lime as an additive (in terms of total mixture weight,
and 2.1% in terms of aggregate weight) was prepared. Therefore, after heating of

Table 2
The results of tests performed on aggregate.

Properties Test method Value (%) Allowable value

L.A. Abrasion AASHTO T96 16 Max 30%
Coarse aggregate with one fractured faces ASTM D5821 95 100%
Coarse aggregate with two fractured faces ASTM D5821 91 Min 90%
Flat and elongated particles of coarse aggregates with ratio of 1:3 ASTM: D4791 14 Max 20%
Flat and elongated particles of coarse aggregates with ratio of 1:5 ASTM: D4791 2 Max 5%
382 A. Shafiei, M. Latifi Namin / Construction and Building Materials 70 (2014) 379–387

Table 5 3.5. Dynamic creep test

Used gradation and limits for SMA.
The dynamic creep test was done by UTM5 according to NCHRP 9-19 in uncon-
Sieve size (mm) Passing (%) fined condition at 50 °C. A haversine loading of 138 kPa was applied with 0.1 s and
9.5 mm NMAS 12.5 mm NMAS 0.9 s as loading time and rest time, respectively.

Used Lower Upper Used Lower Upper

3.6. Dynamic modulus test
19 – – – 100 100 100
12.5 100 100 100 95 90 100
The dynamic modulus test was performed on the cylindrical unconfined sam-
9.5 95 90 100 67 26 78
ples under a haversine compressive loading and the loading pulse according to
4.75 45 26 60 28 20 28
the NCHRP has a haversine shape. The applied frequencies are 25, 10, 1, 0.5 and
2.36 25 20 28 22 16 24
0.1 Hz, while the 25 Hz frequency is considered as the pre-loading. Number of
1.18 18 13 21 19 13 21
the loading cycles associated with each frequency are 206, 106, 56, 25, 6 and 6,
0.6 16 12 18 17 12 18
respectively. A rest time of 1 min long is considered between every two frequencies,
0.3 14 12 15 14 12 15
so that the samples could have their recoverable strains. The static and dynamic
0.075 10⁄ 8 10 10⁄ 8 10
stresses are selected for each sample regarding the loading temperature at 5, 15,
⁄
10% Filler for both gradation (the equal amount in gram). 25, 35 and 45 °C, respectively and finally, the master curves was shown for dynamic
modulus and phase angle.

%filler by percentage of aggregate weight 4. Results and discussion

HL CS HL as addive
4.1. Results of complex modulus test
2.0%
2.5%
The complex modulus testing was used to address viscoelastic
5.0% and stiffness properties of the mixture. The master curve of
dynamic modulus was derived here by using the equation pro-
7.5% posed in NCHRP-614 report [29] (Figs. 2 and 3). The master curve
10%
of phase angle was also extracted using the equation suggested
10.0%
by Zeng et al. in 2001 [30] for all the frequencies at 25 °C.
7.5%

5.0%
Temperature shi factor
2.5% 3
temperature degree
Regression Line
Control Samples A2, A7 A3, A8 A4, A9 A5, A10
2
Fig. 1. Increase in percentage of HL (hydrated lime) filler along with CS (crushed
stone) filler in mixtures.
Log shi factor a (T)

1
the gravel aggregates and mixing of the bitumen, 2% dry hydrated lime was added
to the mixture just after fully coating aggregates with bitumen. This method was
used in order to easier mixing procedure, better coating aggregate surfaces and also,
prevention of enveloped hydrated lime in asphalt binder. Just one sample was 0
0 5 10 15 20 25 30 35 40 45 50
enough for each of the 5 testing temperatures considering the non-destructive nat-
ure of the dynamic modulus test [29].
The SMA samples were prepared for this research with two different grain size -1
distributions and also different contents of the hydrated lime filler according to
Table 6. The Rice test was done to determine the optimum binder percent to satisfy
4% air void in the mixture, while the drain down test was implemented on the sam-
ples. Samples were compacted using gyratory compactor. At last, samples of -2
y = 0.0004x2 - 0.1378x + 3.1964
100 mm diameter and 150 mm height were made. R² = 1

3.4. Static creep test -3

Temperature, °c
NCHRP 9-19 Superpave Models were used for unconfined samples at 50 °C. A
deviator stress of 138 kPa was lasted until failure and thereafter, the sample Fig. 2. Shift factor versus temperature for SMA with9.5 mm NMAS and 2% lime as
entered the third phase. additive.

Table 6
Methodology for combined mixtures for all conditioned samples.

Sample NMAS (mm) Crushed stone filler (%) Substituted hydrated lime (%) Additive% (by mixture weight) Bitumen content (%)
A1 9.5 10 0 0.03% Fiber 6.5
A2 9.5 7.5 2.5 0.03% Fiber 6.5
A3 9.5 5 5 0.03% Fiber 6.5
A4 9.5 2.5 7.5 0.03% Fiber 6.5
A5 9.5 10 0 0.03% Fiber 2% lime 6.5
A6 12.5 10 0 0.03% Fiber 7
A7 12.5 7.5 2.5 0.03% Fiber 7
A8 12.5 5 5 0.03% Fiber 7
A9 12.5 2.5 7.5 0.03% Fiber 7
A10 12.5 10 0 0.03% Fiber 7
2% Lime

Three samples were made for each combination and average of results were used.
 A. Shafiei, M. Latifi Namin / Construction and Building Materials 70 (2014) 379–387
Fig. 3. Master curve with shifted E* data for SMA with 9.5 mm NMAS and 2% lime as
additive.
Fig. 4. Master curves of SMA mixtures with 9.5 mm NMAS and different hydrated
lime contents at reference temperature of 25 °C.
Fig. 5. Master curves of SMA mixtures with NMAS 12.5 mm and different hydrated
lime contents at reference temperature of 25 °C.
Fig. 6. Phase angle master curves for SMA mixture with NMAS 9.5 mm and different lime contents at reference temperature of 25 °C.
383
4.1.1. Dynamic modulus
HL has a multifunctional effect on mixture. It can be seen from
the dynamic modulus values in different temperatures and fre-
quencies for mixtures of different contents that the effect of
hydrated lime on dynamic modulus of the SMA mixture and stiff-
ness index of the asphalt is really small at high temperatures and
low frequencies. Comparison of the dynamic modulus master
curves at low frequencies for mixtures of 9.5 mm nominal size
(Fig. 4) indicates just a negligible rise of dynamic modulus by
increasing the hydrated lime content up to 2.5% and decreasing it
by further addition of the hydrated lime. This effect is rather differ-
ent in the mixtures of 12.5 mm nominal size (Fig. 5) somehow; by
increasing the hydrated lime up to 2.5%, the dynamic modulus
reaches to its maximum value at high temperatures or low fre-
quencies, and increasing the hydrated lime at low temperatures
or high frequencies reduces the dynamic modulus. To be more
exact, Presence of HL as active filler at high temperature and as
inactive filler at low temperature causes physiochemical combina-
tion with bitumen which reduces the interval between maximum
and minimum values of the dynamic modulus, and improve simul-
taneously its stiffness at high and low temperatures. This also may
attributed to the fact that the specific surface area and surface
energy of hydrated lime is much larger than mineral powder,
resulting in the selection absorption of light molecular weight
compounds of asphalt [17].
4.1.2. Phase angle
Phase angle is indicative of viscous characteristics of the mix-
ture due to mastic properties in dense graded structures. Compar-
ing HL contained mixtures with control samples, by replacement of
HL in the mixture a slight reduction has occurred in the viscosity
and phase angle of SMA (Figs. 6 and 7). Addition of the lime to
the mixture as a part of aggregate or using as additive in the mix-
ture seem to reduce slightly the degree of phase angle and viscosity
of SMA. This may be due to the fact that the structure of SMA is
based on direct contact of aggregate while the mastic only fills
the voids between coarse aggregates and hold the structure. There-
fore, when the mastic stiffens elastic property of coarse aggregate
influences more effectively on viscosity of the SMA. However,
increasing the temperature will increase elastic effect of gravel in
the mixture, while reduce the degree of phase angle which implies
reduced viscous effects of the mastic. In the SMA mixtures this
effect is intensified due to the direct contact of the gravels espe-
cially at high temperatures [28]. That is why the degree of phase
angle in the SMA mixtures at high temperatures is simultaneously
affected by the presence of hydrated lime and interaction of the
gravels on each other. Those indices involving the phase angle
are thus unable to evaluate the presence of hydrated lime in the
mixture properly at high temperature.
384 A. Shafiei, M. Latifi Namin / Construction and Building Materials 70 (2014) 379–387

4.1.3. Evaluation of rutting factor

Fig. 7. Phase angle master curves for SMA mixture with NMAS 12.5 mm and
different lime contents at reference temperature of 25 °C.
E*/sinf is the parameter obtained from the dynamic modulus
test and can be used in the frequencies of 5 and 10 Hz as the index
of rutting resistance. This parameter should not be smaller than
the minimum requirements of Superpave. Loading at the rate of
0.1 s (10 Hz) in the laboratory, is usually indicative of the traffic
speed in field [31].
The results of comparison between the rutting resistance which
are obtained from the dynamic modulus test and rut depth estima-
tion experiments like Hamburg wheel tracking device show that
the parameter of E*/sinf represent correctly rutting resistance in
the HMA mixture only at high temperatures [32].
This parameter was used as a criterion for evaluating rutting
resistance of SMA in this research (Table 7). The results of this

Table 7
Superpave factors for rutting and fatigue cracking in two frequencies.

% HL Control 2.5% HL 5% HL 7.5% HL 2% HL additive

NMAS (mm) 9.5 12.5 9.5 12.5 9.5 12.5 9.5 12.5 9.5 12.5
|E*|/sin(f) at 45 °C 10 Hz 411 354 458 441 443 435 420 432 420 366
5 Hz 356 321 391 386 398 388 370 367 352 346
|E*|
sin(f) at 25 °C 10 Hz 251 309 307 240 246 244 177 137 276 217
5 Hz 160 292 270 235 213 189 201 134 224 185

Fig. 8. Rutting factor for SMA mixtures with 9.5 mm NMAS and different hydrated lime contents at 45 °C.

Fig. 9. Rutting factor for SMA mixtures with 12.5 mm NMAS and different hydrated lime contents at 45 °C.
 A. Shafiei, M. Latifi Namin / Construction and Building Materials 70 (2014) 379–387 385

Nom 9.5 mm effect of interaction of the gravels on the phase angle at 45 °C. Nev-
350
ertheless, this effect becomes significant at highly elevated
temperatures.
300

4.1.4. Evaluation of fatigue crack factor

250 For fatigue cracking, a performance factor in the Superpave bin-
der specification is |G⁄|sind. Thus, the equivalent performance fac-
tor for the mix is |E⁄|sinf. [28]. Reduction of the fatigue factor with
|E*|.sinf

200
increasing the lime content indicates that hydrated lime will
150
enhance the resistance against fatigue cracking (Table 7). Mean-
while, the range of fatigue factor values for the mixtures with nom-
inal size of 9.5 mm at 10 Hz frequency implies improvement of the
100
fatigue cracking by substituting the hydrated lime for more than
5% (Fig. 10). Fig. 11 shows that the results for samples of
50 12.5 mm nominal size are almost the same, except that there is a
0% lime 5% lime 2% lime as addive
2.5% lime 7.5% lime
growth in the resistance against fatigue cracking by increasing
0 the hydrated lime from 0 to 7.5%. A comparison between fatigue
0 2 4 6 8 10 12 cracking factor of two mixtures of different nominal sizes reveals
Frequency, HZ that the values of |E*|
sinf for the samples of 12.5 mm is smaller
Fig. 10. Fatigue cracking factor for SMA mixtures with 9.5 mm NMAS and different than those of the 9.5 mm samples. It can be explained that, the
hydrated lime contents at 25 °C. internal pores of the HL particles are filled with bitumen, and this
filled particles are seen as hard spheres in the bitumen matrix,
therefore prevents growth of micro crack in mastic [33]. It seems
that the more content of Hl the more fatigue cracking resistance
will be observed in SMA which does not show an optimum content
of HL in SMA.

4.2. Creep test

Dynamic and static creep tests were done to explore resistance

Fig. 11. Fatigue cracking factor for SMA mixtures with 12.5 mm NMAS and
different hydrated lime contents at 25 °C.
evaluation at 45 °C confirm improvement of the rutting resistance
with increasing substituted hydrated lime filler content up to 5%.
In both aggregate gradation (Figs. 8 and 9), the mixtures containing
5% lime at the frequency of 5 Hz provide a superior resistance in
comparison with the 2.5% lime sample, likewise the results of the
creep test. However, this value shows just a small different at the
frequency of 10 Hz. Unlike the results of the creep test, adding
2% of the lime filler to the mixture as an additive has not much
effect in raising the rutting resistance. It seems that much reason-
able results will be generated by creep test due to destructive nat-
ure of this test which allows no recovery during the test to the
sample [32].
The close similarity between the evaluation of rutting factor
results (E*/sinf) and the indices of creep test implies a negligible
of the samples against the common phenomena of rutting and per-
manent deformations. Two indices of flow number and flow time
are the main parameters for evaluation of the rutting phenomenon.
As was expected before, the results which were obtained from the
both experiments represent the same trend for the effect of lime on
rutting. Due to physical features of HL particles the distribution of
HL throughout the binder is better, as the HL particles are much
finer than most common fillers. Finer particles have more surface
area for a given unit volume, thus providing more surface interac-
tions with asphalt. Therefore, the finer the filler, the more it poten-
tially stiffens the asphalt. The stiffening effect of filler is dependent
on Rigden voids of the filler. The voids in the compacted filler com-
bine the effects of particle density, particle shape, and particle size
distribution in the filler fraction [34].
It can be seen from Table 8 that increasing the hydrated lime fil-
ler content of the SMA samples has reinforced the resistance indi-
ces of rutting Fn and Ft, also decreased the permanent deformation
of these samples against the cyclic dynamic load. Increasing the fil-
ler content from 0% to 5% enhanced rutting resistance of the mix-
ture. Stiffer asphalt binder provides more holding ability to
maintain the aggregate structure leading to less permanent defor-
mation. As is demonstrated in Figs. 12 and 13, control mixes
reached a stable slope faster; this is likely attributable to less stiff
asphalt mastic that allows faster densification and a steep slope
during secondary phase where the rate of accumulation of perma-
nent deformation remains constant.

Table 8
Dynamic and static creep test results at 50 °C.

% HL Control 2.5% HL 5% HL 7.5% HL 2% HL additive

NMAS (mm) 9.5 12.5 9.5 12.5 9.5 12.5 9.5 12.5 9.5 12.5
Flow number 425 349 1075 427 2983 2295 2375 1123 2135 1527
Flow time 334 312 631 400 5012 1259 1933 626 1996 798
Total permanent deformation (mm) 1.81 1.66 2.10 2.34 1.94 2.48 2.40 2.10 2.29 3.09
386 A. Shafiei, M. Latifi Namin / Construction and Building Materials 70 (2014) 379–387

Fig. 12. Dynamic creep curves for 9.5 mm NMAS mixtures at 50 °C.

Fig. 13. Dynamic creep curves for 12.5 mm NMAS mixtures at 50 °C.

Based on experimental observations adding HL more than 5% in

the mixture changed the trend of permanent deformation resis- 3500
nom 9.5
tance. It is possible that when more HL is added along constant nom 12.5
amount of bitumen the filler surface absorption increases and the 3000
absorbed filler particles gather into a group that may leads to dete-
rioration of interface bonding between aggregates and lower cohe- 2500
sion. On the other hand, excessive amount of HL reduces coat of
coarse aggregates which affects reorientation of aggregate particles
flow num

2000
in compaction process and structure of mixtures that results in less
cohesiveness and less rutting resistance. It is also clearly observed
1500
that the grain distribution of 9.5 nominal size gives a greater flow
number in comparison with the grain distribution of 12.5 nominal
size (Figs. 14 and 15) because Uniform coat of coarse aggregate and 1000
film thickness lead to better aggregate interaction and strong
structure in SMA. 500
By comparing the results of 2.5% HL content mixes with the
mixes in which 2% HL has been added as additive, it can be con- 0
cluded that the amount of total filler in mixture and the method Control (2.5,7.5) (5,5) (7.5,2.5) (2,10)
of adding HL filler considerably affect SMA performance. This (HL (%), CS (%))
may attributed to the fact that in SMA with open graded structure,
Fig. 14. Flow number for mixtures with different percentage of HL and CS filler.
10% crushed stone filler along 2% HL additive enhances filling up ⁄
HL: hydrated lime filler, CS: crush stone filler.
the voids between coarse aggregate and improves aggregate con-
tacts. However, introduction of 2% dry HL to the mixture after full
aggregate coating by bitumen mostly stiffens the asphalt binder, up the air void between aggregate. As a consequence, lower inter-
While when 2.5% HL filler is replaced in the mixture, after mixing, face bonding strength occurs and mastic stiffness decreases which
not all of the hydrated lime is in ‘‘direct’’ contact with the surface leads to less rutting resistance compared with 2% HL additive in
of the aggregate, but some becomes part of the binder itself or fills mix.
 A. Shafiei, M. Latifi Namin / Construction and Building Materials 70 (2014) 379–387
6000
nom 9.5
nom 12.5
5000
4000
Flow me
3000
2000
1000
0 pavements. Constr Build Mater 2010;24(4):538–44.
Control (2.5,7.5) (5,5) (7.5,2.5) (2,10)
(HL (%), CS (%))
Fig. 15. Flow time for mixtures with different percentage of HL and CS filler. ⁄HL:
hydrated lime filler, CS: crush stone filler.
5. Conclusions
Mechanical properties of SMA with regard to different content
of HL (Hydrated Lime) were questioned. Simple performance tests
such as creep tests and dynamic modules were conducted in order
to investigate how the different percentage of hydrated lime in dif-
ferent HL mixing method affect performance of mixtures with
open grain size distribution. The following results may be drawn.
According to obtained results of creep tests, presence of
hydrated lime reduced rutting and rut depth much more than con-
trol mixtures. In both aggregate gradations (9.5 NMAS, 12.5 NMAS)
Rutting resistance increases as the HL percent rise up to optimum
amount of HL. It was demonstrated that in SMA, increasing HL
more than 5% changes the trend and reduces rutting indices.
In both gradation (9.5 NMAS, 12.5 NMAS), presence of HL effect
slightly on |E*| and phase angle. It is hypothesized that in SMA with
direct contact of aggregate HL has negligible effect on viscosity and
stiffness of SMA mixtures.
The results of this research indicate that total percentage of fil-
ler and mixing method considerably impact on performance of
SMA. As a result, using HL as additive leads to superior perfor-
mance comparing with partial substitution of HL for filler
aggregate.
The obtained measures from E*/sinf do not establish properly a
relationship with the results of creep test in order to evaluate the
rutting potential of the SMA mixtures at high temperatures below
50 °C. It is thought that, E*/sinf and E*
sinf do not represent the
performance of SMA mixture as well as creep tests and related fati-
gue tests.
References
[1] Huang YH. Pavement analysis and design. 2nd ed. Upper Saddle River,
NJ: Prentice Hall; 2003.
[2] Brown ER, Cooley LA. Designing stone matrix asphalt mixtures for rut-resistant
pavements. Transport Res Board 1999;425.
[3] Craus J, Ishai I, Sides A. Some physico-chemical aspects of the effect and the
role of the filler in bituminous paving mixtures. Assoc Asphalt Paving Technol
Proc 1978;47.
387
[4] Kim YR, Little DN, et al. Development of specification-type tests to assess the
impact of fine aggregate and mineral filler on fatigue damage. Texas
Transportation Institute, Texas A&M University System; 2005.
[5] Kandhal PS, Lynn CY, Parker F. Characterization tests for mineral fillers related
to performance of asphalt paving mixtures. Transport Res Rec J Transport Res
Board 1998;1638( 1):101–10.
[6] Kim Y-R, Little D, Song I. Effect of mineral fillers on fatigue resistance and
fundamental material characteristics: mechanistic evaluation. Transport Res
Rec J Transport Res Board January 2003;1832( 1):1–8.
[7] Bari J, Witczak MW. Evaluation of the effect of lime modification on the
dynamic modulus stiffness of hot-mix asphalt: use with the new mechanistic-
empirical pavement design guide. Transport Res Rec J Transport Res Board
2005;1929( 1):10–9.
[8] Kok BV, Yilmaz M. The effects of using lime and styrene–butadiene–styrene on
moisture sensitivity resistance of hot mix asphalt. Constr Build Mater May
2009;23(5):1999–2006.
[9] Airey GD, Collop AC, Zoorob SE, Elliott RC. The influence of aggregate, filler and
bitumen on asphalt mixture moisture damage. Constr Build Mater September
2008;22(9):2015–24.
[10] Aragão FT, Lee J, Kim YR, Karki P. Material-specific effects of hydrated lime on
the properties and performance behavior of asphalt mixtures and asphaltic
[11] Lee S, Seo Y, Kim YR. Effect of hydrated lime on dynamic modulus of asphalt–
aggregate mixtures in the state of North Carolina. KSCE J Civ Eng November
2010;14(6):829–37.
[12] Button JW, Epps JA. Evaluation of methods of mixing lime in asphalt paving
mixtures. Texas Transportation Institute; 1983.
[13] Little DN, Epps JA, Sebaaly PE. The benefits of hydrated lime in hot mix asphalt.
Natl Lime Assoc 2001.
[14] Lee S, Mun S. Performance based evaluation of lime addition methods in hot
mix asphalt. Can J Civ Eng February 2012;39(2):172–9.
[15] Diab A, You Z, Othman A, Ahmed H. Effect of hydrated lime application method
on mechanical and fatigue properties of HMA. In: CICTP, American Society of
Civil Engineers; 2012. p. 3327–34.
[16] Albayati AH. Mechanistic evaluation of lime-modified asphalt concrete
mixtures. In: Scarpas A, Kringos N, Al-Qadi I, Loizos A, editors. 7th RILEM
international conference on cracking in pavements. Netherlands: Springer;
2012. p. 921–40.
[17] Chen J, Zhou L, Yan ZH. Hydrated lime content in asphalt mastic based on high
temperature performances. Appl Mech Mater November 2010;40–41:669–74.
[18] Huang JF, Wu SP, Ma LX, Liu ZF. Material selection and design for moisture
damage of HMA pavement. Mater Sci Forum 2009;614:269–74.
[19] Petersen JC, Plancher H, Harnsberger PM. Lime treatment of asphalt to reduce
age hardening and improve flow properties. In: Association of asphalt paving
technologists proceedings technical sessions, vol. 56, 1987, Reno, Nevada, USA,
1987.
[20] Johansson LS. Bitumen ageing and hydrated lime. 1998.
[21] Meininger RC. Effects of aggregates and mineral fillers on asphalt mixture
performance. ASTM International; 1992.
[22] Aschenbrener T. Investigation of low temperature thermal cracking in hot mix
asphalt. No. CDOT-DTD-R-95–7, February 1995.
[23] Vansteenkiste S, De Visscher J, Vanelstraete A. Impact of hydrated lime on the
durability of SMA mixtures: laboratory and field evaluation. Road Mater
Pavement Des 2013;14(Suppl.1):162–74.
[24] Zou J, Isola M, Roque R, Chun S, Koh C, Lopp G. Effect of hydrated lime on
fracture performance of asphalt mixture. Constr Build Mater July
2013;44:302–8.
[25] Witzcak MW. Simple performance test for superpave mix design. Transport
Res Board 2002;465.
[26] Papazian HS. The response of linear viscoelastic materials in the frequency
domain with emphasis on asphaltic concrete. In: International conference on
the structural design of asphalt pavements, vol. 203, 1962.
[27] Clyne TR, Li X, Marasteanu MO, Skok EL. Dynamic and resilient modulus of Mn/
DOT asphalt mixtures, 2003.
[28] Kim Y. Modeling of asphalt concrete. McGraw Hill Professional; 2008.
[29] Bonaquist RF. Refining the simple performance tester for use in routine
practice. Transport Res Board 2008;614.
[30] Zeng M, Bahia HU, Zhai H, Anderson MR, Turner P. Rheological modeling of
modified asphalt binders and mixtures (with discussion). J Assoc Asphalt
Paving Technol 2001;70.
[31] Feng X, Hao P. Summarization on mixture design method of bituminous
stabilized broken-stone base mixture. Highway 2007;6:036.
[32] Chen Z, Zhang X, Cong L, Lu H, Yang J. Rutting resistance performance
evaluation of superpave mixes with antirutting additives, In: Transportation
Research Board 89th Annual Meeting, 2010.
[33] Lesueur D, Petit J, Ritter H-J. The mechanisms of hydrated lime modification of
asphalt mixtures: a state-of-the-art review. Road Mater Pavement Des
2013;14(1):1–16.
[34] Zulkati A, Diew WY, Delai DS. Effects of fillers on properties of asphalt-
concrete mixture. J Transport Eng 2011;138(7):902–10.