Construction and Building Materials 173 (2018) 228–237

Contents lists available at ScienceDirect

Construction and Building Materials

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

Performance analysis of nano modified bitumen and hot mix asphalt

Mehmet Saltan, Serdal Terzi, Sebnem Karahancer ⇑
Suleyman Demirel University, Engineering Faculty, Civil Engineering Department, 32260, Turkey

h i g h l i g h t s

Energy saving is improved by modification.

Optimum binder contents are decreased by modification.

Performance of HMA mixtures modified with nano materials are improved.

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

Article history: The aim of the research was to evaluate the performance of Hot Mix Asphalt (HMA) and bitumen by mod-
Received 13 November 2017 ifying bitumen with nano materials according to SuperpaveTM mix design procedure. In this research, bitu-
Received in revised form 30 March 2018 men was modified by Multi-Walled Carbon Nanotube (MWCNT) and Carbon Nanotubes Doped with 50
Accepted 2 April 2018
wt% SiO2 Nanopowder (SiO2dopedCNT). Nano materials were mixed with base bitumen at contents of 1, 3
and 5% by weight. Rutting and fatigue performance of bitumen was determined. HMA is prepared with
nano modified bitumen at optimum binder content (OBC). Moisture susceptibility of modified hot mix
Keywords:
asphalt was evaluated according to Modified Lottman test procedure. Nano modified bitumens were also
Carbon nanotube
Silica
evaluated by scanning electron microscopy whether the modification was uniform or not.
Nano-modified bitumen As a result, better performance was obtained by SiO2dopedCNT modification. Performance Grade (PG)
Rutting and fatigue performance of all modified bitumen was determined as PG 64-22 except MWCNT modified bitumen at the content of
1% (which was PG 58-22). OBCs were decreased by modification. All nano modified hot mix asphalt sam-
ples were determined as high resistive to moisture compared to reference sample except samples mod-
ified with SiO2dopedCNT at the content of 5%. Homogenous mixture was obtained according to scanning
electron microscopy results. It was found that the agglomeration of nano materials is mostly lower than
4 lm in the mixtures.
Ó 2018 Elsevier Ltd. All rights reserved.

1. Introduction
The performance of HMA is affected by characteristics of com-
ponents and behaviour of these components in the mixture. This
effect depends on the subjected environmental factors of compo-
nents. When the asphaltic materials do not satisfy the require-
ments for constructing a well-performing bituminous structure,
modification usually is adopted as one of the best and most attrac-
tive strategies for meeting the desired properties of used materials.
Better engineering of complex materials such as asphalt at the
nano level will result in a range of newly introduced smart charac-
teristics. The service life of pavements varies according to struc-
tural design conditions, characteristics of materials, layer
thickness, maintenance and accepted damage criterion.
⇑ Corresponding author.
E-mail addresses: mehmetsaltan@sdu.edu.tr (M. Saltan), serdalterzi@sdu.edu.tr
(S. Terzi), sebnemsargin@sdu.edu.tr (S. Karahancer).
Damages that occur during the useful life of pavement mainly
are rutting, permanent deformation and fatigue cracking. Since
the recovery and reconstruction of defects will be costly, therefore,
the prevention of such cases would be more economical [1].
It is expected that modification of bitumen with nano-materials
improve the mechanical properties of asphalt mixtures including
an increase of stiffness modulus, increase of strength against strip-
ping, increase of strength against moisture damage, prevention of
cracks and increase of resistance against creep. Several nanomate-
rials have the possibility to be utilized to modify bitumen, as for
examples, nano-sized hydrated lime, nano-sized plastic powders
or polymerized powders, nanoclay, nanosilica, nanotubes and
nanofibres.
Nano materials have unique properties; they have been devel-
oped rapidly as useful additives for asphalt mixture. These proper-
ties include a large fraction of surface atoms, high surface work,
quantum effects, structural features, and spatial confinement.

https://doi.org/10.1016/j.conbuildmat.2018.04.014
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
 M. Saltan et al. / Construction and Building Materials 173 (2018) 228–237
Benefits of nano modified bitumen include improvement of resis-
tant to rutting, cracking, and fatigue [2,3].
Nanosilica and carbon nanotube has been extensively used as
an inorganic additive to improve the properties of bituminous
materials [4–7]. Shafabakhsh and Ani aim to analyze the results
of laboratory tests that were performed on modified steel slag
asphalt mixture (SSAM) containing TiO2 and nano SiO2 particles,
the materials that are added to improve the rutting resistance of
nano modified asphalt mixtures. Results showed that adding nano
TiO2 and nano SiO2 boost the bitumen’s rheological characteristics
and improve toughness and viscosity by an average of 30% and
109%, respectively while reducing penetration grade. In addition,
the asphalt’s rutting resistance and fatigue life were improved
[2]. Tanzadeh et al. studied the effect of nano-TiO2 on rutting per-
formance of asphalt pavements. The purpose of this study was
research on the effect of nano-TiO2 in improving bitumen property
and rutting resistance in asphalt pavement under dynamic loading.
The results of their study illustrated that using nano-TiO2 in bitu-
men samples caused to an improvement in rutting depth in com-
parison with the ordinary mixtures [8].
Yusoff et. al. investigated the performance characteristics of
polymer-modified asphalt mixture (PMA) with the addition of
nano-silica particles. The addition of 4% nano-silica with PMA
appeared to have the greatest potential for beneficial modification
of the binder [9]. Motlagh et. al. investigated that promote techni-
cal characteristics of bitumen and asphalt mixtures using carbon
nanotubes as an additive material for bitumen. As a result, mixture
with minimal mechanical changes in the characteristics of asphalt
mixtures was obtained that not only in terms of physical proper-
ties is not much different, but also improves the quality of its
chemical properties [10]. Tan et. al. in order to improve the perfor-
mance of composite modified asphalt, the central composite
design-response surface methodology was used to determine the
optimum content of raw materials. The rheological behaviour
including accumulation of deformation, viscoelastic ratio, creep
strength, creep strain rate ‘‘m” was analyzed by testing in central
composite model. Response surface of the overall desirability was
established. The test results indicated that the optimum raw mate-
rials ratio was asphalt: rubber powder: additive: SBS: SBR =
100:25:7:5:3 (by weight) [11].
Also, the characterization of modified bitumen is important for
better understanding the dispersion of nano materials into bitu-
men. Guo et. al. were conducted an adsorption-separation test of
asphalt binder on surface of mineral fillers to separate the struc-
ture asphalt binder and free asphalt binder [12].
forbidden area
100
90
80
Passing Aggregate, %
70
60
50
40
30
20
10
0
0 0.5
229
In this research, modified carbon nanotubes with SiO2
nanopowder by weight of 50% to improve the performance of car-
bon nanotube were chosen. MWCNT was also chosen to compare
the effect of modified carbon nanotube with SiO2 nano powder
by weight of 50%. These nano materials were added to bitumen
to improve the rutting and fatigue performance of bitumen and
moisture susceptibility of asphalt mixtures. Nano materials were
mixed with base bitumen at contents of 1, 3 and 5% by weight.
Rheological properties of nano-modified bitumen were analyzed
by use of bitumen tests such as Rotational Viscosity (RV), Dynamic
Shear Rheometer (DSR) and Bending Beam Rheometer (BBR). In
addition, the short- and long-term ageing properties of nano-
modified bitumen were analyzed, with the ageing process
simulated by Rolling Thin Film Oven Test (RTFOT) and the Pressure
Ageing Vessel (PAV). Furthermore, rutting and fatigue performance
of bitumen was determined by G⁄/sind and G⁄.sind parameters
obtained by DSR after ageing process. Nano modified asphalt
mixtures were prepared and compacted by Superpave Gyratory
Compactor (SGC). OBC of all nano modified mixtures were also
determined and compared to the non-modified mixture. Indirect
Tensile (IDT) strength and moisture susceptibility of modified hot
mix asphalt was determined according to Modified Lottman test
procedure (AASHTO T283). Nano modified bitumens were also
evaluated by Scanning Electron Microscopy (SEM) whether the
modification was uniform or not.
2. Materials
2.1. Aggregate
Limestone aggregate was used in hot mix asphalt. HMA sam-
ples were prepared for wearing course. The nominal maximum
aggregate size was selected as 12.5 mm. Gradation curve was
selected in accordance with SuperpaveTM for the dense graded
HMA (Fig. 1).
2.2. Modification materials
The chemical and physical properties of MWCNT and SiO2-
dopedCNT were given in Tables 1 and 2. Because SiO2dopedCNT
is obtained from modification of CNT by SiO2 it is given only
MWCNT and SiO2 seperately. The same component is used in this
study to compare the effect of modified carbon nanotube with
SiO2 nano powder by weight of 50%.
gradation curve control points max gradation
1 1.5 2 2.5 3 3.5 4
Sieve Size (mm, X0.45)
Fig. 1. Gradation curve for HMA.
230 M. Saltan et al. / Construction and Building Materials 173 (2018) 228–237

Table 1 Bitumen was modified by nano materials at 4000 rpm, 160 °C

Properties of MWCNT. for two hours. SiO2dopedCNT and MWCNT were added to the mix-
Properties ture by weight of base bitumen at contents of 1%, 3% and 5%.
Purity (%) >%95 (carbon nanotube)>%97
(carbon content)
3.2. Ageing bitumen
–OH content %3.6 by weight
Outer diameter 10–20 nm
Inner diameter 5–10 nm Ageing process was performed with RTFOT and PAV. These tests
Length 10–30 lm were conducted according to present performance grade (AASHTO-
Surface area >200 m2/g MP1).
Color Black
Ash <%0.5 by weight
RTFOT ageing procedure is used to provide ageing while mixing
Electric conductivity >100 S/cm and placement in the land. RTFOT is undertaken according to the
Density (Tap) 0.22 g/cm3 AASHTO T 240. According to AASHTO MP1, the mass loss should
Density (True) 2.1 g/cm3 be maximum 1.0%.
PAV ageing procedure is used to provide long term ageing of
bitumen. PAV is performed after RTFOT ageing process. PAV is
Table 2
undertaken according to AASHTO R 28.
Properties of sio2.

Properties
3.3. Performance of bitumen and HMA
Chemical formula SiO2
Molar mass 60.08 gmol1
Appearance Transparent solid (amorphous), white (powder) 3.3.1. Dynamic shear rheometer
Density 2.648 (a-quartz), 2.196 (amorphous) gcm3 DSR test is used to describe the viscous and elastic behaviour of
Melting point 1713 °C (3115 °F; 1986 K) (amorphous) bitumen at medium and high temperatures. Both shear modulus
Boiling point 2950 °C (5340 °F; 3220 K)
(G⁄) and phase angle (d) are measured by DSR. The test is per-
Thermal conductivity 12 (|| c-axis), 6.8 (\ c-axis), 1.4 (am.) W/(mK)
Refractive index (nD) 1.544 (o), 1.553 (e) formed on unaged, RTFOT aged and PAV aged bitumen. The test
is largely software controlled. The DSR test is undertaken according
to AASHTO T 315.
2.3. Bitumen

PG 64-22 performance grade bitumen was used in this research.
All properties of nano modified and non-modified bitumen were
determined according to SuperpaveTM bitumen tests which are
DSR, BBR, RTFOT and PAV (Table 3). Penetration, softening point,
ductility and rotational viscometer tests were also performed on
modified and non-modified bitumen.
3.3.2. Bending beam rheometer
BBR test describes a measure of low-temperature stiffness and
relaxation properties of bitumen. These parameters give an indica-
tion of a bitumen’s ability to resist low temperature cracking. BBR
test is performed on PAV aged bitumen. BBR is undertaken accord-
ing to AASHTO T 313.

3. Method 3.3.3. Rotational viscometer

3.1. Modification of bitumen
The high shear mixer was used for modifying bitumen with
nano materials. The mixer can be set at 8000 rpm and 210 °C and
the capacity of the mixer is 1.5 L.
RV test can determine the workability during the mixing and
compaction processes of a bitumen. The rotational viscometer val-
ues of the bitumen are determined according to AASHTO TP48
standard. Brookfield DV-III rotational viscometer is used to per-
forming the viscosity of bitumen. The test is undertaken at 135
°C, 165 °C and 185 °C.

Table 3
Properties of nano modified and non-modified bitumen.

Test Unit Base bitumen SiO2dopedCNT MWCNT

1% 3% 5% 1% 3% 5%
*
Penetration 0.1 mm 62.2 57 51 48.5 57.6 50.9 48.6
Softening point* °C 49.9 51.35 50.7 53.55 48.95 49.3 50.3
Ductility* cm >100 >100 78 40 >100 >100 85.75
RV1* Pa.s 0.475 0.55 0.6 0.65 0.487 0.537 0.7
RV2* Pa.s 0.15 0.175 0.175 0.237 0.137 0.15 0.237
DSR1* Fail temp. °C 67.9 66.9 68.3 69.4 66.1 67 68.9
Grade °C 64 64 64 64 64 64 64
Mass loss % 0 0 +0.06 0.04 0.11 0.07 +0.04
Permanent penetration % 70.4 72 69 69 69 68 68
Change in softening point °C +3.2 +1.9 +3.1 +0.4 +4.2 +4.7 +5.2
DSR2* Fail temp. °C 67 64.3 64.5 65.6 63.9 65 65
Grade °C 64 64 64 64 58 64 64
DSR3* fail temp. °C 28.6 26.9 28.5 25.7 26.6 27.5 23.5
Grade °C 22 22 22 22 22 22 22
BBR* °C 12 12 12 12 12 12 12
m-value 0.325 0.314 0.331 0.313 0.329 0.323 0.305
Stiffness MPa 213 219 195 239 228 209 222
Performance grade PG 64-22 PG 64-22 PG 64-22 PG 64-22 PG 58-22 PG 64-22 PG 64-22
*
Penetration – @25 °C, 100 g, 5 s; Softening point – Ring&Ball; Ductility – @25°C, 5 cm/dk; RV1 – Rotational Viscometer @135 °C, 3Pa.s; RV2 – Rotational Viscometer
@165 °C; DSR1 – Dynamic Shear Rheometer G*/sind>1 kPa @10 rad/s; DSR2 – DSR G*/sind>2.2 kPa @10 rad/s; DSR3 – DSR G*.sind<5.000 kPa @10 rad/s; BBR – BBR S300 MPa,
m0.300 @60 s.
 M. Saltan et al. / Construction and Building Materials 173 (2018) 228–237 231

3.3.4. Modified Lottman test Samples are also exposed to the freeze-thaw cycle. Indirect ten-
Modified Lottman test is conducted to determine the moisture sile strength test is run on each sample. Tensile strength values are
susceptibility of HMA samples by means of IDT strength. Six sam- recorded and Tensile Strength Retained (TSR) is calculated. The
ples are prepared and divided into two as three subsets. One subset resistance to moisture damage is expressed as a ratio of the uncon-
is ‘‘unconditioned (IDTu Þ ” (tested in a dry state) and the other is ditioned sample tensile strength that is retained after conditioning.
‘‘conditioned (IDTc Þ ” (tested in a saturated state). Each sample is
2P
placed in a vacuum container and vacuum saturation is applied IDT ¼ ð5Þ
for conditioning. The saturation level is calculated as follows: pdh
msurf :dry  ma IDT c
SL ¼ 10:000 ð1Þ TSR ¼ 100 ð6Þ
Va  Vs IDT u
ðGmax  Gbulk Þ where IDT is the indirect tensile strength, P is the max load, d is the
Va ¼ ð2Þ
Gmax diameter of the specimens, h is the height of the specimens. IDT c is
the average strength value of the conditioned specimens and IDT u is
V s ¼ msurf :dry  mw ð3Þ the average strength value of the unconditioned specimens.

ma
Gbulk ¼ ð4Þ 4. Results and discussion
Vs
where SL is saturation level (%),msurf :dry is saturated surface dry 4.1. Optimum binder content
weight (g), ma is weight in air (g), V a is air voids (%), V s is volume
of the specimen, Gmax is theoretical gravity, Gbulk is bulk specific OBC was determined according to SuperpaveTM. SGC was used
gravity and mw is the weight of the specimen in water. for compaction at 4% air void content and in four percentages

SiO2dopedCNT1% SiO2dopedCNT3% SiO2dopedCNT1% SiO2dopedCNT3%

SiO2dopedCNT5% Reference SiO2dopedCNT5% Reference
Specification Limit Minimum VMA
9 17.5
R² = 0.996
8
7 16.5 R² = 0.9655
6
Air Void, %

R² = 0.9805
R² = 0.9984
VMA, %

15.5
5
4
R² = 1 14.5 R² = 0.8382
3 R² = 0.9997
2 13.5
1 R² = 0.9999
0 12.5
3.5 3.7 3.9 4.1 4.3 4.5 4.7 4.9 3.5 4 4.5 5
Bitumen Content, % Bitumen Content, %

SiO2dopedCNT1% SiO2dopedCNT3% SiO2dopedCNT1% SiO2dopedCNT3%

SiO2dopedCNT5% Reference
Specification Limit SiO2dopedCNT5% Reference
100 2.45
R² = 1
2.43
90
Specific Gravity, gr/cm3

R² = 0.942
2.41
R² = 0.2679
80 2.39
VFA, %

R² = 0.9916 R² = 0.9979 R² = 1
2.37
70 2.35
R² = 0.9876
2.33
60
2.31
R² = 0.9996
50 2.29
3.5 4 4.5 5 3.5 4 4.5 5
Bitumen Content, % Bitumen Content, %
Fig. 2. Optimum binder content for SiO2dopedCNT.
232 M. Saltan et al. / Construction and Building Materials 173 (2018) 228–237

4.5%, 5%, 5.5% and 6%. OBC was determined for base HMA as 4.5%
by weight. Optimum binder graphs were drawn by air void content
obtained from SGC. OBC (4.5%) ensured the specification limit for
both Voids in Mineral Aggregate (VMA) and Voids Filled with
Asphalt (VFA).
OBC was also determined for all three percentages of both SiO2-
dopedCNT and MWCNT, separately. Compaction effort was con-
ducted for 4% air void content and in four percentages 3.5%, 4%,
4.5% and 5% because 4% air void was obtained for these percent-
ages. OBC was determined for 1%, 3% and 5% SiO2dopedCNT mod-
ified HMA as 4.46%, 3.97% and 4.59% by weight, respectively.
Optimum binder graphs were drawn in accordance with air void
content (Fig. 2) for all percentages of modifications together. OBC
was also determined for 1%, 3% and 5% MWCNT modified HMA
as 4.05%, 4.6% and 4.51% by weight, respectively (Fig. 3). OBCs of
1% and 3% SiO2dopedCNT and 1% MWCNT modified HMA were
decreased by modification compared to base bitumen. OBC of SiO2-
dopedCNT at the content of 3% was 11% lower than base bitumen.
4.2. DSR results
In this research, SiO2dopedCNT and MWCNT were utilized in
three percentages of 1%, 3% and 5% to improve the rutting and fati-
gue performance of bitumen. For this aim, SiO2dopedCNT and
MWCNT modified and non-modified bitumen were exposed to
short term ageing by RTFOT and long term oxidative ageing by
PAV. After that unaged and aged bitumen were tested by DSR.
Complex shear modulus (G⁄) and phase angle (d) were com-
pared for unaged bitumens (Fig. 4a and 4b). When compared to
base bitumen, phase angle was decreased by SiO2dopedCNT and
MWCNT modified with 5% and also complex shear modulus was
increased by these modifications at 64 °C (Fig. 4a). But the phase
angle was decreased for all modifications except MWCNT 5% when
compared to base bitumen at 70 °C (Fig. 4b). It means that the
decrease in phase angle (at 64 °C) bitumen became more elastic.
But the increase in phase angle (at 70 °C) means that the bitumen
became more viscous.
The G⁄/sind parameters were determined for unaged bitumen
(Fig. 5). G⁄/sind 1.00 kPa limit value for unaged bitumens was
ensured by all modifications @64 °C. G⁄/sind parameters of SiO2-
dopedCNT 3% and 5% and MWCNT 5% modified bitumens were
higher than base bitumen. These modified bitumens had more
deformation resistance when compared to base bitumen for the
unaged process.
The G⁄/sind parameter was determined for RTFOT aged bitu-
mens for 2.2 kPa limit value (Fig. 6). The rutting performance of

MWCNT1% MWCNT3% MWCNT1% MWCNT3%

MWCNT5% Reference MWCNT5% Reference
artname Limit De eri Minimum VMA De eri
9 17
16.5
R² = 0.9888 R² = 0.9649
7 16
R² = 0.9997
15.5
Air Void, %

5
VMA, %

R² = 0.9956 R² = 0.8853
15
R² = 0.9991 14.5
3
14

1 13.5
R² = 0.9999
13
R² = 0.9881
-1 12.5
3.5 4 4.5 5 3.5 4 4.5 5
Bitumen Content, % Bitumen Content, %

MWCNT1% MWCNT3%
MWCNT1% MWCNT3% MWCNT5% Reference
MWCNT5% Reference 2.48
Specification Limit
110
2.46
100 R² = 0.9898 R² = 0.9789
Specific Gravity, gr/cm3

2.44
90
2.42
R² = 0.9996 R² = 0.9984
VFA, %

80 2.4

70 2.38 R² = 0.9569
R² = 0.9996
60 2.36
R² = 0.2679
50 R² = 0.9875 2.34

40 2.32
3.5 4 4.5 5 3.5 4 4.5 5
Bitumen Content, % Bitumen Content, %

Fig. 3. Optimum binder content for MWCNT.

 M. Saltan et al. / Construction and Building Materials 173 (2018) 228–237 233

(a) (b)

Complex Shear Modulus, kPa

Complex Shear Modulus, kPa
2.50 1.00

0.90
2.00
0.80

1.50
0.70

1.00 0.60
87 87.2 87.4 87.6 87.8 88 88.2 87.5 88 88.5 89
Phase Angle,
Phase Angle,
SiO2dopedCNT%1 SiO2dopedCNT%3 SiO2dopedCNT%5 SiO2dopedCNT1% SiO2dopedCNT3% SiO2dopedCNT5%
MWCNT%1 MWCNT%3 MWCNT%5 MWCNT1% MWCNT3% MWCNT5%
Reference Reference

Fig. 4. Complex shear modulus vs. phase angle (a) @64 °C (b) @70 °C.

2.1

1.9

1.7
G*/sin , kPa

1.5

1.3

1.1

0.9

0.7

0.5
62 64 66 68 70 72
Temperature, °C
SiO2dopedCNT1% SiO2dopedCNT3% SiO2dopedCNT5%
MWCNT1% MWCNT3% MWCNT5%
Reference limit

Fig. 5. G*/sind 1.00 kPa for unaged bitumens.

4.8
4.6
4.4
4.2
4
3.8
3.6
3.4
G*/sin , kPa

3.2
3
2.8
2.6
2.4
2.2
2
1.8
1.6
1.4
1.2
1
57 59 61 63 65 67 69 71
Temperature, °C
SiO2dopedCNT1% SiO2dopedCNT3% SiO2dopedCNT5% MWCNT1%
MWCNT3% MWCNT5% Reference limit
*
Fig. 6. G /sin d 2.2 kPa for RTFOT aged bitumens.
234 M. Saltan et al. / Construction and Building Materials 173 (2018) 228–237

bitumen could be evaluated with G⁄/sind 2.2 kPa parameter. The dopedCNT at the content of 3% was found that the optimum per-
rutting performance of SiO2dopedCNT 5% and MWCNT 3% and 5% centage for this research.
was higher than other modified bitumens and base bitumen. The
G⁄/sind 2.2 kPa parameter was also ensured at 58 °C for MWCNT 4.3. BBR results
1%. Performance grade of MWCNT 1% is determined as PG58-22.
The G⁄.sind parameter was determined for RTFOT + PAV aged Creep stiffness of modified bitumens was determined at 60 s.
bitumens (Fig. 7). Fatigue performance of bitumens could be eval- SiO2dopedCNT and MWCNT modified bitumens showed lowest
uated with G⁄.sind 5000 kPa parameter. The fatigue performance creep stiffness for 3% modifications. Thermal cracking performance
of SiO2dopedCNT 1% and 3% and MWCNT 1% and 3% was higher could be evaluated with creep stiffness. So that the thermal crack-
than other modified bitumens and base bitumen. ing resistance was increased by SiO2dopedCNT and MWCNT 3%
SiO2dopedCNT showed the highest performance when all DSR modification when compared to base bitumen and the other mod-
results evaluated. For the modification of base bitumen with SiO2- ified bitumens.
Creep stiffness and creep ratio of all bitumens were determined
9000 and compared (Fig. 8). The lowest creep stiffness was obtained for
8000 SiO2dopedCNT 3% modified bitumen. Creep stiffness should be
7000 maximum 300 MPa according to specification. Specification limit
6000 was ensured by all modifications. Creep ratio (m-value) could also
G*.Sin , kPa

5000 be obtained from BBR test results. Creep ratio (m-value) should be
4000 minimum 0.300 according to specification. Specification limit was
3000 ensured by all modifications. Creep stiffness results were also sup-
2000 ported by m-value results.
1000
0
20 22 24 26 28 30 32
4.4. RV results
Temperature, °C
Rotational viscometer test was performed on modified and
SiO2dopedCNT1% SiO2dopedCNT3% SiO2dopedCNT5%
unmodified bitumens. Rotational viscometer results were
MWCNT1% MWCNT3% MWCNT5%
Reference limit
determined for all bitumens (Fig. 9). The viscosity of SiO2dopedCNT
1% and MWCNT 1% and 3% modified bitumens were lower than
Fig. 7. G*.sind 5000 kPa for RTFOT + PAV aged bitumens. base bitumen. This means that the mixing and compaction

limit value (a) limit value (b)

350 0.34
Creep Stiffness, MPa

300 0.33
250 0.32
m-value

200
0.31
150
0.3
100
50 0.29
0 0.28
Referans

Referans
MWCNT%1

MWCNT%3

MWCNT%5

MWCNT%1

MWCNT%3

MWCNT%5
SiO2dopedCNT%1

SiO2dopedCNT%3

SiO2dopedCNT%5

SiO2dopedCNT%1

SiO2dopedCNT%3

SiO2dopedCNT%5

Fig. 8. BBR results (a) creep stiffness (b) m-value of all modified bitumens.

(a) (b)

0.5 0.5
Viscosity, Pa.s
Viscosity, Pa.s

0.05 0.05
130 135 140 145 150 155 160 165 170 175 180 185 190 130 135 140 145 150 155 160 165 170 175 180 185 190
Temperature, C Temperature, C

MWCNT1% MWCNT3% SiO2dopedCNT1% SiO2dopedCNT3%

MWCNT5% Reference SiO2dopedCNT5% Reference
mixing interval compaction interval mixing interval compaction interval

Fig. 9. Rotational viscometer test results (a) MWCNT (b) SiO2dopedCNT.

 M. Saltan et al. / Construction and Building Materials 173 (2018) 228–237 235

temperatures were lower than the base bitumen and so that these temperatures were also determined as 148.4 °C and 147.8 °C for
modified bitumens had more workability. MWCNT 1% and SiO2dopedCNT 3%, respectively while base bitu-
Mixing and compaction intervals and temperatures of bitumens men was 149 °C.
were determined by the cut point of 0.17 ± 2 Pa.s and 0.28 ± 3 Pa.s
intervals. According to mixing and compaction temperatures 4.5. Indirect tensile strength test results
MWCNT 1% and SiO2dopedCNT 3% modified bitumens were most
workable modified bitumens. Mixing temperatures were deter- IDT Strength of HMA samples was determined (Fig. 10a). IDT u
mined as 160.2 °C and 160.5 °C for MWCNT 1% and SiO2dopedCNT values were increased with the increase of SiO2dopedCNT and
3%, respectively, while base bitumen was 162.7 °C. Compaction MWCNT percentage. Only MWCNT 1% modified HMA sample had

unconditioned conditioned (a) Specification Limit (b)

Indirect Tensile Strength, kPa

1400 100
90
1200 80
1000 70

TSR, %
60
800 50
600 40
30
400 20
200 10
0
0

Reference

MWCNT1%

MWCNT3%

MWCNT5%

SiO2dopedCNT1

SiO2dopedCNT3

SiO2dopedCNT5
Reference

MWCNT1%

MWCNT3%

MWCNT5%

SiO2dopedCNT1

SiO2dopedCNT3

SiO2dopedCNT5

%
%

Fig. 10. (a) Indirect tensile strength (conditioned and unconditioned) (b) moisture susceptibility (TSR).

Fig. 11. SEM images (a) non-modified bitumen (b) MWCNT 1% (c) MWCNT 3% (d) MWCNT 5% (e) SiO2dopedCNT 1% (f) SiO2dopedCNT 3% (g) SiO2dopedCNT 5%.
236 M. Saltan et al. / Construction and Building Materials 173 (2018) 228–237

Table 4
Comparing all results as% reclamation.

Test Penetration Softening point Stiffness Rutting performance Fatigue performance IDTU TSR OBC
(% reclamation)
SiO2dopedCNT
1% ;8.36 "2.90 "2.81 ;8.44 "9.83 "15.63 "5 ;0.88
3% ;18.00 "1.60 ;8.45 ;5.85 "37.72 "27.00 "17.5 ;11.77
5% ;22.02 "7.31 "12.20 "10.12 ;7.43 "38.05 ;8.75 "2
MWCNT
1% ;7.39 ;1.90 "7.04 ;14.22 "4.18 ;6.51 "7.5 ;10
3% ;18.16 ;1.20 ;1.87 "0.47 "16.42 "22.53 "8.75 "2.22
5% ;21.86 "0.80 "4.22 "0.44 ;28.02 "39.76 "8.75 "0.22
*
" Higher% when compared to reference sample.
*
; Lower% when compared to reference sample.

less performance when compared to reference HMA sample. So all
modified HMA samples showed higher strength compared to refer-
ence HMA sample. IDT u values of SiO2dopedCNT 1%, 3% and 5%
modified HMA samples were increased in the ratios of 15.63%,
27.0% and 38.05%, respectively. IDT u values of MWCNT 3% and 5%
modified HMA samples were increased in the ratios of 22.53%
and 39.76%, respectively. So that the high-performance HMA sam-
ples were obtained by SiO2dopedCNT and MWCNT modifications.
TSR of HMA samples was determined (Fig. 10b) for moisture
susceptibility. TSR should be minimum 80% according to the spec-
ification. Moisture susceptibility of HMA samples was evaluated by
TSR. All modifications ensured the specification limit. TSR was
higher for all modifications than the reference HMA sample except
SiO2dopedCNT 5%. As a result, highest moisture resistance was
obtained by SiO2dopedCNT 3% modified HMA sample.
4.6. SEM results
SEM analysis was conducted on modified bitumens and com-
pared to non-modified bitumen, to determine the dispersion of
SiO2dopedCNT and MWCNT in bitumen matrix (Fig. 11a). Accord-
ing to SEM analysis, MWCNT nanotube dispersion was found in
bitumen as 1–5 lm conglomerates. SiO2dopedCNT dispersion
was found as 1–10 lm conglomerates (Fig. 11b, 11c, 12d, 12e,
12f and 12 g). So that, MWCNTs and SiO2dopedCNTs were uni-
formly dispersed in bitumen up to 4 lm. Still, there were agglom-
erations in bitumen up to 20–40 lm. It was hard to find MWCNTs
in bitumen because nanotubes have same chemical formulas with
bitumen. But there were different chemical bond structures with
bitumen. Tube structure of MWCNT could also be seen with
MWCNT5% (Fig. 11d). The structure of nanotube was saved in
bitumen.

SiO2dopedCNT 5% modified bitumen had highest rutting perfor-
mance with 10.12% increase compared to base bitumen after
RTFOT ageing. Elastic behaviour and rutting performance of
SiO2dopedCNT 5% modified bitumen were increased.

SiO2dopedCNT 3% modified bitumen had highest fatigue perfor-
mance with 37.72% increase compared to base bitumen after
PAV ageing.

The mixing and compaction temperatures of SiO2dopedCNT 3%
and MWCNT 1% modified bitumens were decreased compared
to base bitumen. Energy saving was improved by modification.

OBC of non-modified HMA was determined as 4.5%. SiO2-
dopedCNT 3% modified HMA had lowest OBC with 11.77%
decrease compared to non-modified HMA.

IDT u values of all modified HMA samples were increased except
MWCNT 1% modified HMA. So that the strength of HMA sam-
ples were improved.

SiO2dopedCNT 3% modified HMA had highest moisture resis-
tance with 17.5% increase compared to non-modified HMA.

SEM images showed that the dispersion of nano materials in
bitumen matrix was homogenous. There were 1–40 mm sized
nano material conglomerates in bitumen rarely but generally,
the size of the nanotubes was in micrometre and nanometre
size.

When comparing all results, SiO2dopedCNT at the content of 3%
modification showed highest performance.
Conflict of interest
There is no conflict of interest.
Ethical statement

5. Conclusions I confirm that we have given due consideration to the protec-

All results were summarized by reclamation percent (Table 4).
Reclamation percent was expressed as good (") for higher perfor-
mance and bad (;) for lower performance. But, lower values were
good for penetration, creep stiffness and OBC results of modifica-
tions. This means that having the lowest penetration, creep stiff-
ness and OBC of modified bitumens have good performance. So
that, according to results, conclusions were drawn as follows:
tion of intellectual property associated with this work and that
there are no impediments to publication, including the timing of
publication, with respect to intellectual property. In so doing I con-
firm that I have followed the regulations of my institutions con-
cerning intellectual property.
Funding body

Suleyman Demirel University with the project number: 4211-

Penetration was decreased minimum 7% by all modifications.
D2-14.
Softening point was increased by SiO2dopedCNT modification.
The consistency of modified bitumen was improved maximum
22.02% by SiO2dopedCNT modification. Acknowledgement

SiO2dopedCNT 3% modified bitumen had lowest creep stiffness
with 8.45% decrease compared to base bitumen. SiO2dopedCNT The authors would like to acknowledge the financial support for
3% modified bitumen had more resistance against thermal this research provided by Suleyman Demirel University with the
ageing. project number: 4211-D2-14.
 M. Saltan et al. / Construction and Building Materials 173 (2018) 228–237 237

References [7] S.S. Galooyak, M. Palassi, A. Goli, H.Z. Farahani, Performance Evaluation of
Nano-silica Modified Bitumen, Int. J. Transp. Eng. 3 (1) (2015) 55–66.
[8] J. Tanzadeh, F. Vahedi, P. Kheiry, R. Tanzadeh, Laboratory study on the effect of
[1] Gh. Shafabaksh, S.M. Mirabdolazimi, M. Sadeghnejad, Evaluation the effect of
nano-TiO2 on rutting performance of asphalt pavements, Adv. Mater. Res. 622–
nano-TiO2 on the rutting and fatigue behaviour of asphalt mixtures, Constr.
623 (2013) 990–994.
Build. Mater. 54 (2014) 566–571, https://doi.org/10.1016/
[9] N.I.Md. Yusoff, A.A.S. Breem, H.N.M. Alattug, A. Hamim, J. Ahmad, The effects of
j.conbuildmat.2013.12.064.
moisture susceptibility and ageing conditions on nano-silica/polymer-
[2] G.H. Shafabaksh, O.J. Ani, Experimental investigation of effect of Nano TiO2/
modified asphalt mixtures, Constr. Build. Mater. 72 (2014) 139–147, https://
SiO2 modified bitumen on the rutting and fatigue performance of asphalt
doi.org/10.1016/j.conbuildmat.2014.09.014.
mixtures containing steel slag aggregates, Constr. Build. Mater. 98 (2015) 692–
[10] A.A. Motlagh, A. Kiasat, E. Mirzaei, F.O. Birgani, Bitumen modification using
702, https://doi.org/10.1016/j.conbuildmat.2015.08.083.
carbon nanotubes, World Appl. Sci. J. 18 (4) (2012) 594–599, https://doi.org/
[3] A.E.A. Mostafa, Examining the performance of hot mix asphalt using nano-
10.5829/idosi.wasj.2012.18.04.1443.
materials, IOSR J. Eng. (IOSRJEN) 6 (2) (2016) 25–34. V2.
[11] Y. Tan, M. Guo, L. Cao, L. Zhang, Performance optimization of composite
[4] E. Mahmood, A. Diab, Characteristics of asphalt binder and mixture containing
modified asphalt sealant based on rheological behavior, Constr. Build. Mater.
nanosilica, Int. J. Pavement Res. Technol. 10 (2017) 148–157, https://doi.org/
47 (2013) 799–805, https://doi.org/10.1016/j.conbuildmat.2013.05.015.
10.1016/j.ijprt.2016.11.009.
[12] M. Guo, Y. Tan, J. Yu, Y. Hou, L. Wang, A direct characterization of interfacial
[5] H. Zalnezhad, S.S. Galooyak, H. Farahani, A. Goli, Investigating The Effect of
interaction between asphalt binder and mineral fillers by atomic force
Nano-Silica on the Specification of the Sasobit Warm Mix Asphalt, Pet. Coal 57
microscopy, Mater. Struct. 50 (141) (2017), https://doi.org/10.1617/s11527-
(5) (2015) 509–515.
017-1015.
[6] F. Zafari, M. Rahi, N. Moshtagh, H. Nazockdast, The improvement of bitumen
properties by adding nanosilica, Study Civil Eng. Arch. (SCEA) 3 (2014) 62–69.