Content uploaded by Mahmood Al-Obaidi
Author content
All content in this area was uploaded by Mahmood Al-Obaidi on Jun 25, 2018
Content may be subject to copyright.
International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2016): 79.57 | Impact Factor (2015): 6.391
Volume 7 Issue 2, February 2018
www.ijsr.net
Licensed Under Creative Commons Attribution CC BY
Effect of Incorporating Processes of Fumed Silica
on Marshall Properties of HMA Concrete
Mahmood Khalid Jumaah Al-Obaidi1
1Assistant Lecturer, Civil Engineering Department, Al-Farabi University College
Al-Masafi Road, Al-Daora District, Baghdad, Iraq
eng_mahmood_k[at]yahoo.com
Abstract: Highways play an important role in the economic and social development of societies; therefore, many studies are directed
towards modifying pavement properties. In Iraq as well as other countries, pavement surface cracking and rutting are considered as
major problems in roads and airports. The main objective of this work is to evaluate the effect of incorporating method of additives on
Marshall properties of hot mix asphalt (HMA) concrete (wearing surface layer) using different percentages of additive. OAC of (4.7)%
for control mixtures was determined using Marshall mix design. All modified HMA consist of AC of (40-50) penetration grade brought
from Al-Duarah refinery, aggregates obtained from AL-Nibaae quarry with NMAS of (12.5) mm, and mineral filler (Portland cement)
obtained from Tazloja factory, Additive (Fumed Silica (F.S)) its chemical compositions were tested in the laboratories of General
Directorate of Geological survey and Mining. Modified mixtures with (3, 7, 10)% of F.S were prepared based on two methods of
incorporating F.S into the HMA concrete are Stirring and Addition methods. They subdivided into the two groups, the first group refers
to the addition of F.S as a percentage of the total weight of AC (the weight of modified AC stayed constant), The second group refers to
the addition of F.S as a percentage of the total weight of AC (as additional amount) to the weight of AC. Results show that modified
specimens prepared based on stirring method and containing (3% + AC) are the best among all mixtures.
Keywords: Marshall Mix Design, Modified Asphalt Cement, Fumed Silica, Stirring and Addition Processes
1. Introduction
According to previous studies, two methods of incorporating
additives into HMA mixtures as will be described. Many
researchers was used stirring process of incorporating the
additive with hot A.C. It was continued for 45 minutes as a
constant blending time. Three percentage of the Silica Fumes
(1% - 5%) with a constant increment of 1% by weight of
asphalt cement (40-50) have been introduced based on
previous work. The produced modified asphalt cement was
poured inside the testing mold for furth testing [1]. Asphalt
cement of (40-50) penetration grade from Al-Dura refinery
was adopted in this study and blended with 2% of Silica
Fumes, which was obtained from local market, it is an ultra
fine powder consisting of nearly spherical particles around
100 time small than a grain of cement .Other percentages of
Silica Fumes were also tried by testing the blend for
(penetration, Softening point and Ductility) and 2% of Silica
Fume was selected. Soft Asphalt was heated to nearly 25°C,
and the Silica Fume was added to the asphalt cement with
stirring until homogenous blend was achieved, mechanical
blender continued the mixing and stirring for 30 minutes [2].
For the Addition process, addition of additives powder to dry
aggregates is the simplest method of incorporating the
additive into asphalt mixes. It was first adopted by the State
of Georgia in early 1980's. In this method, the additive and
mineral filler is incorporated in a drum mixer just after the
point at which asphalt is introduced. The additive thus
introduced comes in contact with aggregates and directly
results in improved bond between aggregate and asphalt.
Some portion of additive that fails to come in contact with
aggregate will get mixed with asphalt. This result in additive
reacting with highly polar molecules in asphalt to form
insoluble salts that no longer attract water thus reducing
stripping and oxidation potential [3].
2. Framework, Materials and Methods
2.1 Operational and Theoretical Framework
Two steps required to achieve the study objective as shown
in Figure 1. The first step is the tests done on mixture
components; while second step is the tests done on prepared
specimens:
Figure 1: Flow Diagram for Laboratory Analysis Process
2.2 Materials
2.2.1 Asphalt Cement
Asphalt cement of penetration grade (40-50) was used as a
binder; it was brought from Al-Dura refinery. Table 1
presents the physical properties of asphalt cement.
Paper ID: ART20179902
DOI: 10.21275/ART20179902
127
International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2016): 79.57 | Impact Factor (2015): 6.391
Volume 7 Issue 2, February 2018
www.ijsr.net
Licensed Under Creative Commons Attribution CC BY
Table 1: Physical Properties of Asphalt Cement
Property Dura Asphalt Cement
Penetration (0.01mm) 44
Softening point (°C) 50
Ductility (Cm) >100
2.2.2 Coarse and Fine aggregate
Crushed coarse aggregate (retained on sieve No.4) was
obtained from AL-Nibaae quarry. Crushed sand and natural
sand were used as fine aggregate (passing sieve No.4 and
retained on sieve No.200), brought from the same source. It
consists of hard, tough grains, free from deleterious
substances. Table 2 presents the physical properties of
aggregate.
Table 2: Physical Properties of Coarse and Fine Aggregate
Property Coarse Aggregate Fine Aggregate
Bulk specific gravity 2.610 2.640
Water absorption (%) 0.448 0.720
Los Angeles Abrasion (%) 22.2 -
The selected gradation in this study followed by [4]
specification for wearing course with 12.5 (mm) nominal
maximum aggregate size. Table 3 shows the selected
aggregate gradation.
Table 3: Physical Properties of Asphalt Cement
Sieve size (mm) 19 12.5 9.5 4.75 2.36 0.3 0.075
Finer by weight (%) 100 95 83 59 43 13 7
(SCRB 2003) Specification 100 90 - 100 76 – 90 44 - 74 28 – 58 5 – 21 4 – 10
2.2.3 Mineral Filler
Ordinary Portland cement has been used as mineral filler in
this study, which is obtained from Tasluga cement factory.
The physical properties are shown in table 4.
Table 4: Physical Properties of Mineral filler
Property Test Results
Specific Gravity 2.794
Passing sieve No.200 (0.075mm) (%) 94
2.2.4 Additive
Fumed silica (F.S) as shown in Figure 2 was used for this
study. It is produced by a vapor phase hydrolysis process
using chlorosilanes such as: silicon tetrachloride in a flame of
hydrogen and oxygen. It is supplied as a black, fluffy
powder, Chemical compositions and physical properties were
tested in the laboratories of General Directorate of
Geological survey and Mining and given in Table 5 and 6,
respectively.
Table 5: Chemical Components of Fumed Silica
Component SiO2 Fe2O3 Al2O3 TiO2 CaO MgO SO3 L.O.I
Test Result 99.1 35 ppm < 0.035 < 0.006 0.03 52 ppm < 0.07 0.7
Table 6: Physical Properties of Fumed Silica
Property Test Results
surface area (m²/g) 450
Density (kg/m³) 170
Loss of weight when drying at 1000˚c for 2hrs (%) < 2
Loss of weight (%) when drying at 105˚c for 2 hrs < 1.5
pH 4.3
Retained on 40 µm sieve (%) < 0.04
Moisture (%) 0.82
Figure 2: Sample of Fumed Silica
2.3 Methods of Preparing Marshall Specimens and Tests
2.3.1 Preparation of Marshall Specimens
The specimens were prepared in accordance with [5],
aggregates and filler were put in the pan, then, the pan was
put in the oven and being heated to (160) °C, the pan was
charged with the heated aggregates and dry mix thoroughly.
Asphalt was heated up to (150) 0C prior to mixing, and it was
added to the hot aggregate in the pan. The aggregates and
asphalt cement were rapidly mixed using automatic mixer
until thoroughly coated and viscosity was (170 20) cSt, and
lastly, the mixture was removed from the pan and was ready
for compaction process. The procedure begins with recording
the mixture temperature and observing until it reaches the
desirable compaction temperature. The mold is 4˝(10.16 cm)
in diameter and 2.5˝ ± 0.05 (6.35 cm) in heighten assembly
and the face of compaction hammer was cleaned and heated
Paper ID: ART20179902
DOI: 10.21275/ART20179902
128
International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2016): 79.57 | Impact Factor (2015): 6.391
Volume 7 Issue 2, February 2018
www.ijsr.net
Licensed Under Creative Commons Attribution CC BY
in oven at (120) to (150) 0C, filter paper that was cut into
pieces was placed in the bottom of the mold before the
mixture is introduced, the mixture that has been prepared is
then placed in the mold, and stirred by the spatula or trowel
for (15) times around the perimeter and (10) times over the
interior, the collar is removed and the surface will be
smoothed with the trowel to slightly rounded shape, next, the
compaction temperature immediately prior to compaction
temperature was (140) ºC and viscosity (280 30) cSt, the
collar then will be assembled to the compaction pedestal in
the mold holder, the (75) blows of compaction hammer are
applied with a free fall of 4.536 kg (10 lb) sliding weight and
a free fall of (457.2) mm from the mold base, and the
compaction hammer is assured to be perpendicular to the
base of the mold assembly, after compaction, the base plate is
removed and the same blows are applied to the bottom of the
specimen that has been turned around, after that, the collar is
lifted from the specimen carefully. Next, the specimen was
transferred to smooth surface at room temperature for over-
night [6]. The procedure of mixing and compaction are
shown in Figure 3
Figure 3: Procedure of Preparing Marshall Specimen
In this study, modified Marshall Specimens were prepared
based on the two methods of incorporating the additive
(fumed silica) into the HMA concrete as shown later.
Three percentage of the fumed silica were added, these are
(3, 7, 10)% by weight of asphalt cement, but also the
methods of addition subdivided into the two groups:
The first group refers to the addition of F.S as a percentage
of the total weight of AC (the weight of modified AC stayed
constant) as shown in Table 7. While, the second group
refers to the addition of F.S as a percentage of the total
weight of AC (as additional amount) to the weight of A.C
shown in Table 8.
Table 7: Weights of AC and F.S in the Mixes of First Group
Percentage of F.S (%) Asphalt Cement (g) F.S (g) Total Weight of Modified A.C
3 55.3 1.7 57
7 53 4 57
10 51.3 5.7 57
Table 8: Weights of AC and F.S in the Mixes of Second Group
Percentage of F.S (%) Asphalt Cement (g) F.S (g) Total Weight of Modified A.C
3 57 1.7 58.7
7 57 4 61
10 57 5.7 62.7
Paper ID: ART20179902
DOI: 10.21275/ART20179902
129
International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2016): 79.57 | Impact Factor (2015): 6.391
Volume 7 Issue 2, February 2018
www.ijsr.net
Licensed Under Creative Commons Attribution CC BY
2.3.1.1 Preparation of Specimens using Stirring Process
This process was used by many researchers. [1], used this
process where the fumed silica was added to the asphalt
cement with stirring on the hot plate until homogenous blend
was achieved, fabricated mechanical blender continued the
mixing and stirring for 30-45 minutes.
In this study, two types of specimens were prepared based on
the stirring process. The fumed silica added and mixed on the
hot plate with a hot AC for 45 minutes as a constant blending
time using manufactured mechanical blender operates with a
2400-2900 rpm on 0.6/0.8A (220V and 50/60Hz) as shown
in Figure 4.
Figure 4: Manufactured mechanical blender
18 modified specimens with (3, 7, 10)% of F.S were prepared
for this method. 9 of them based on addition of F.S as in first
group which mentioned previously represented by (F.S% by
wt. of AC), as shown in Figure 5.
Figure 5: Modified Marshall Specimens with (F.S% by wt.
of AC)
Another 9 specimens based on addition of F.S as in second
group which mentioned previously represented by (F.S% +
AC), as shown in Figure 6.
Figure 6: Modified Marshall Specimens with (F.S% + AC)
2.3.1.2 Preparation of Specimens using Addition Process
In this method, the additive and mineral filler is incorporated
in a drum mixer just after the point at which asphalt is
introduced as used by [3]. 9 modified specimens with (3, 7,
10)% of F.S were prepared for this method based on addition
of F.S as in second group which mentioned previously
represented by (F.S% AC), as shown in Figure 7.
Figure 7: Modified Marshall specimens with (F.S% AC)
2.3.2 Marshall Test
Procedure of preparing and testing specimens according to
this method as shown in Figure 8 is described in [5]. This
method covers the measure of the resistance to plastic flow of
cylindrical specimens (2.5 in. height and 4.0 in. diameter) of
asphalt paving mix loaded on the lateral surface of specimen
holder by means of Automatic Marshall apparatus shown
below, at a constant rate of (50.8) mm/min until the
maximum load was reached. The maximum load resistance
and the corresponding strain values were recorded as
Marshall stability and flow respectively, at test temperature
of (60) °C for (30 to 45) minute in water bath. The entire test
was performed within (30) sec after the specimen was
removed from water bath. Three specimens for each
combination were prepared and average results are reported.
Paper ID: ART20179902
DOI: 10.21275/ART20179902
130
International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2016): 79.57 | Impact Factor (2015): 6.391
Volume 7 Issue 2, February 2018
www.ijsr.net
Licensed Under Creative Commons Attribution CC BY
Figure 8: Marshall Test Procedure
3. Results and Discussions
3.1 Optimum Asphalt Content (OAC)
The primary objective of Marshall mix design was to
determine the OAC of the designed mixes, with 75-blows
compaction using Marshall hammer.
Mixtures with four different asphalt contents (4, 4.5, 5, 5.5)%
were prepared and tested. OAC for control mixture of (4.7)%
by weight of mixture was obtained by averaging the three
values of AC at maximum stability, bulk density and (4)% air
voids as shown in Figure 9.
Figure 9 Marshall Test Results for Control Mixture
Paper ID: ART20179902
DOI: 10.21275/ART20179902
131
International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2016): 79.57 | Impact Factor (2015): 6.391
Volume 7 Issue 2, February 2018
www.ijsr.net
Licensed Under Creative Commons Attribution CC BY
3.2 Marshall Test
Specimens of Marshall test were prepared by adopting OAC
of (4.7)% with (3, 7, 10)% of F.S based on stirring and
addition processes of incorporating F.S with AC as
mentioned previously, then Marshall properties of these
specimens compared with those prepared by OAC alone
(control mixes). Data of Marshall Test are mentioned in
Table 9.
Table 9: Results of Marshall Test
Asphalt Cement (%) F.S (%) Stability (kN) Flow (mm) Bulk Density (g/cc) V.T.M (%) V.M.A (%) V.F.A (%)
Control Mix - 10.600 2.720 2.350 3.850 14.725 73.900
F.S% by wt. of AC 3 11.49 2.70 2.330 4.800 15.560 70.260
7 7.231 0.463 2.266 7.320 17.795 58.917
10 5.770 0.737 2.252 7.902 18.311 56.855
F.S% + AC 3 14.410 1.213 2.347 4.001 14.851 73.722
7 6.927 0.613 2.250 7.965 18.368 56.669
10 6.252 0.933 2.245 8.172 18.551 56.042
F.S% AC 3 10.017 3.303 2.339 4.333 15.146 71.402
7 8.822 2.870 2.254 7.817 18.236 57.181
10 10.895 3.590 2.250 7.989 18.389 56.574
SCRB Specification - ≥ 8 2 – 4 - 3 – 5 ≥ 14 70 - 85
3.2.1 Marshall Stability
Results of Marshall Test as mentioned in Table 8 indicated
that the stability of mixtures containing (3% F.S by wt. of
AC), (3% F.S + AC), and (10% F.S AC) had higher stability
values than that of control mixtures by (8.66, 36.31, and
3.06)%, increasing in stability show that stirring method (3%
F.S + AC) gives good values, this may be due to the
homogenous blend was achieved between F.S and AC. While
specimens containing (7, 10% F.S by wt. of AC), (7, 10%
F.S + AC), and (3, 7% F.S AC) had lower stability values
than those of control mixtures by (31.6, 45.42, 34.43, 40. 86,
5.24, and 16.54)%. Decrease in stability values in the mixture
containing additives, may be due to the amount of additives
within asphalt cement in the mixture which create loose
contact points between the aggregates therefore resulting in
lower values of stability.
3.2.2 Marshall Flow
Results of Marshall Test as mentioned in Table 8 indicated
that the flow of mixtures containing (3, 7, 10% F.S AC) had
significantly higher values than that of control mixtures by
(19.69, 3.99, and 30.07)%, results show that addition process
gives values more than stirring process. While specimens
containing (3, 7, 10% F.S by wt. of AC), and (3, 7, 10% F.S
+ AC) had lower values than those of control mixtures by
(2.17, 83.21, 73.31, 56.04, 77.78, and 66.18)%. Regardless
of finding the same amount of asphalt content for all
mixtures, but the increase in flow values in the mixture
prepared by addition process, may be due to the present of
F.S molecules between aggregates which prevent AC to
penetrate the spaces easily resulting in higher values of flow.
3.2.3 Bulk Density
Results of Marshall Test as mentioned in Table 8 indicated
that the bulk density of mixtures containing (3, 7, 10% F.S by
wt. of AC), (3, 7, 10% F.S AC), and (3, 7, 10% F.S + AC)
had lower values than those of control mixtures by (0.99,
3.61, 4.22, 0.16, 4.29, 4.50, 0.51, 4.13, and 4.31)%. Mixtures
containing (3% + AC) which prepared using stirring process
of incorporating F.S with AC, shows that the values of bulk
density is lower than control mixtures by (0.16%), which
means that this amount of F.S is not effecting significantly
the bulk density of modified mixtures, that may be used to
improve another property of Marshall properties. While
another percentages are significantly influencing the bulk
density, this is related to some additives are dispersed less
homogeneously when mixed.
3.2.4 Voids in Total Mix (V.T.M)
Results of Marshall Test as mentioned in Table 8 indicated
that the VTM of mixtures containing (3, 7, 10% F.S by wt. of
AC), (3, 7, 10% F.S + AC), and (3, 7, 10% F.S AC) had
higher values than those of control mixtures by (24.74, 90.17,
105.30, 3.96, 106.95, 112.31, 12.58, 103.10, and 107.56)%.
Also mixtures containing (3% + AC) which prepared using
stirring process of incorporating F.S with AC, shows that the
values of VTM is higher than control mixtures by (3.91%),
which means that this amount of F.S is not effecting
significantly the VTM of modified mixtures, that may be
used to improve another property of Marshall properties.
While another percentages are significantly influencing the
VTM with a high values, this is probably due to the greater
surface areas that need to be wetted by binder failing which
would lead to an increase in VTM.
3.2.5 Voids in Mineral Aggregate (V.M.A)
Results of Marshall Test as mentioned in Table 8 indicated
that the VMA of mixtures containing (3, 7, 10% F.S by wt. of
AC), (3, 7, 10% F.S + AC), and (3, 7, 10% F.S AC) had
higher values than those of control mixtures by (5.68, 20.86,
24.36, 0.87, 24.75, 25.99, 2.87, 23.85, and 24.89)%. Also
mixtures containing (3% + AC) which prepared using stirring
process of incorporating F.S with AC, shows that the values
of VMA is higher than control mixtures by (0.87%), which
means that this amount of F.S is not effecting significantly
the VMA of modified mixtures, that may be used to improve
another property of Marshall properties. While another
Paper ID: ART20179902
DOI: 10.21275/ART20179902
132
International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2016): 79.57 | Impact Factor (2015): 6.391
Volume 7 Issue 2, February 2018
www.ijsr.net
Licensed Under Creative Commons Attribution CC BY
percentages are significantly influencing the VMA with a
high values. This property is significant in so far as the
pavements of hot regions are concerned because asphalt may
be prone to bleeding and amounting void ratio could prevent
bleeding by providing more spaces for the binder to move
into. This was probably due to greater surface areas to be
coated.
3.2.6 Voids Filled with Asphalt (V.F.A)
Results of Marshall Test as mentioned in Table 8 indicated
that the VFA of mixtures containing (3, 7, 10% F.S by wt. of
AC), (3, 7, 10% F.S + AC), and (3, 7, 10% F.S AC) had
higher values than those of control mixtures by (5.68, 20.86,
24.36, 0.87, 24.75, 25.99, 2.87, 23.85, and 24.89)%. Also
mixtures containing (3% + AC) which prepared using stirring
process of incorporating F.S with AC, shows that the values
of VFA is higher than control mixtures by (0.87%), which
means that this amount of F.S is not effecting significantly
the VFA of modified mixtures, that may be used to improve
another property of Marshall properties. While another
percentages are significantly influencing the VFA with a high
values. The increase of VFA indicates an increase of
effective asphalt film thickness between aggregates, which
will results in decreasing cracking.
4. Conclusions and Recommendations
Results show that modified specimens prepared based on
stirring process and containing (3% + AC) give a good
resistance to plastic flow when compared to other types
of mixtures.
It is recommended to evaluate the moisture and
temperature susceptibility when using stirring process
due to the large variation in temperature range usually
practiced in Iraq.
It is recommended to modify asphalt concrete by using
another types of additives such as crumb rubber, fly ash,
and etc.
It is recommended to study the laboratory performance of
modified asphalt concrete using paving materials
conform to the Superpave mix design requirements.
5. Acknowledgment
The author would like to express his gratitude to Al-Farabi
University College for supporting this research.
References
[1] Sarsam S. Issa 2015. Impact of Nano Materials on
Rheological and Physical Properties of Asphalt Cement.
International Journal of Advanced Materials Research,
Vol. 1, No. 1, 2015, pp. 8-14.
[2] Sarsam S. Issa, Israa L. 2014. Assessing Tensile and
Shear Properties of Aged and Recycled Sustainable
Pavement. International Journal of Scientific Research
in Knowledge, 2(9), pp. 444-452, 2014.
[3] Petersen, J.C., H. Plancher, and P.M. Harnsbergen
1987. Lime Treatment of Asphalt to Reduce Age
Hardening and Improve Flow Properties. Proceedings,
AAPT, Vol. 56.
[4] SCRB 2003. Standard Specification for Roads and
Bridges. Section R9, Ministry of Housing and
Construction, Iraq.
[5] ASTM D-1559. 2002. Resistance to Plastic Flow of
Bituminous Mixtures Using Marshall Apparatus.
American Society of Testing and Materials.
[6] Memon 2006. Comparison between Superpave Gyratory
and Marshall Laboratory Compaction Methods. Faculty
of Civil Engineering. Universiti Teknologi Malaysia.
Author Profile
Al-Obaidi was born in Baghdad, Iraq since
1987; received the B.S. in Civil Engineering and
M.S. degrees in the field of Roads and
Transportation from the Department of Civil
Engineering, University of Baghdad in 2009 and
2013, respectively; worked as supervisor engineer with a
Turkish Staff (Dec. 2009 - Jan. 2011); Project manager for
Dana Al-Rafedain Company (Feb. 2011- June 2011);
Supervisor engineer for Tefirom Group Company (July 2011
– Jan. 2012); Project manager for ABT Company (June 2012
– Jan. 2013); Supervising engineer for KAR-PET Company
(Jan 2013 – June 2014). He now a faculty member in Al-
Farabi University College, Civil Engineering Department,
Baghdad, Iraq, since (September 2014). He appointed as a
coordinator of the Department for three years ago and now he
is a head of the Quality and Accreditation of the Scientific
Laboratories of the Civil Engineering Department.
Paper ID: ART20179902
DOI: 10.21275/ART20179902
133