Road Materials and Pavement Design

ISSN: 1468-0629 (Print) 2164-7402 (Online) Journal homepage: https://www.tandfonline.com/loi/trmp20

Mechanical properties of open graded friction

course mixtures with different contents of electric
arc furnace steel slag as an alternative aggregate
from steel industries

Madhu Lisha Pattanaik, Rajan Choudhary, Bimlesh Kumar & Abhinay Kumar

To cite this article: Madhu Lisha Pattanaik, Rajan Choudhary, Bimlesh Kumar & Abhinay Kumar
(2019): Mechanical properties of open graded friction course mixtures with different contents of
electric arc furnace steel slag as an alternative aggregate from steel industries, Road Materials and
Pavement Design, DOI: 10.1080/14680629.2019.1620120

To link to this article: https://doi.org/10.1080/14680629.2019.1620120

Published online: 24 May 2019.

Submit your article to this journal

Article views: 72

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=trmp20
Road Materials and Pavement Design, 2019
https://doi.org/10.1080/14680629.2019.1620120

Mechanical properties of open graded friction course mixtures with

different contents of electric arc furnace steel slag as an alternative
aggregate from steel industries
∗
Madhu Lisha Pattanaik , Rajan Choudhary , Bimlesh Kumar and Abhinay Kumar

Department of Civil Engineering, Indian Institute of Technology Guwahati, Guwahati, Assam, India

(Received 26 December 2018; accepted 10 May 2019 )

With a production of 101.4 million tonnes (MT) in 2017, India is the third largest steel pro-
ducer in the world. Electric arc furnace (EAF) steel slag is the waste/by-product generated
from steel-making industries that utilise electric arc furnace process of steel production. About
12 MT of steel slag is generated annually in India, but merely 20% is put to applications and
the rest is indiscriminately dumped in nearby landfills. The primary objective of this study is to
evaluate performance characteristics of open-graded friction course (OGFC) asphalt mixtures
with varying proportions of EAF steel slag as the coarse aggregate ( > 2.36 mm). OGFCs are
gaining popularity in the Indian highway sector owing to their superior hydraulic (high perme-
ability), acoustic (noise-reduction) and frictional properties that contribute towards improved
road safety. The level of research on the use of EAF steel slag in OGFC is still very limited,
especially in developing countries like India. Ten OGFC mix designs were investigated con-
taining five percentage substitutions of coarse natural aggregates (0%, 25%, 50%, 75% and
100%) with the EAF steel slag and two modified asphalt binders (polymer and crumb rubber
modified). Static creep test, dynamic creep test, Hamburg wheel tracking test, indirect tensile
stiffness modulus test, and indirect tensile fatigue test were performed to evaluate rutting and
fatigue performance of steel slag-OGFC mixtures. Results of the study reveal that the EAF
steel slag-OGFC mixtures showed superior performance compared to control mixtures. The
steel slag-OGFC mixtures up to 75% substitution presented the best results with both binder
types.
Keywords: electric arc furnace steel slag; open-graded friction course; modified asphalt
binders; performance evaluation

Introduction
The Indian road network exceeds 5.5 million kilometres in length, making it second largest in
the world. Quality-wise, however, just about 60 percent of the roads are paved and the National
Highways, which are the multilane high-speed corridors connecting length and breadth of the
country, only constitute about 2 percent of the road network (MoRTH, 2018). Massive road
construction schemes have therefore been launched to expand, widen and upgrade the highway
network. Prodigious road construction has created a very high pressure on the demand of con-
ventional natural stone aggregates. As per recent estimates, India has consumed 3330 million
tonnes (MT) of total aggregates (in all types of constructional activities) in 2015 and will fur-
ther require about 5075 MT of aggregates by 2020 (JSW Steel, 2018). Mining and quarrying

*Corresponding author. Email: rajandce@iitg.ac.in

© 2019 Informa UK Limited, trading as Taylor & Francis Group

2 M. L. Pattanaik et al.

of natural aggregates poses severe environmental concerns such as deforestation, loss of natural
landscape and dust generation. In many regions of the country, good quality natural aggregates
are not available and even quarrying of natural aggregates has been restricted, which has caused
paucity of aggregates and an overall increase in the cost of highway projects. This has com-
pelled highway and asphalt researchers to look for alternative highway construction materials,
especially the aggregates, as they constitute about 95 percent weight of the asphalt mix.
India is the third largest steel producer in the world with a steel production of 101.4 MT in 2017
(World Steel, 2018). Steel slags are generated as by-products during the steel making and refin-
ing operations. Based on the steel manufacturing process, steel slags are generally categorised
in two main types: electric arc furnace (EAF) steel slag and basic oxygen furnace (BOF) steel
slag. Around 12 MT of steel slag is generated every year from steel industries in India (FICCI,
2018). A major portion of steel slag generated in India is dumped as a waste in nearby landfills
and remains unutilised. This study aims to assess the use of EAF steel slag as replacement of
coarse natural aggregates in open-graded friction course (OGFC) mixtures. OGFCs are special
purpose asphalt mixtures designed to achieve a high percentage of air voids to obtain a perme-
able mix. An open structure is achieved by the use of narrowly graded coarse aggregates with less
proportion of fine aggregates. The foremost advantages of OGFCs include quicker drainage of
rainwater, enhanced visibility, enhanced skid resistance, reduced splash and spray, reduced risk
of hydroplaning, and reduced tire-pavement interaction noise (Kandhal, 2002; Kumar, Choud-
hary, & Nirmal, 2018; Pattanaik, Choudhary, & Kumar, 2017; Pattanaik, Choudhary, & Kumar,
2018a). The main component in any asphalt mixture in terms of weight and volume is the aggre-
gate. More so for OGFCs, the coarse aggregates ( > 2.36 mm) usually comprise more than 90%
of the total aggregates. The coarse aggregate skeleton in OGFC mixtures provides the stone-
on-stone contact necessary for the distribution of traffic loads and hence it is desirable that the
aggregates are angular and possess high resistance to abrasion, crushing, and polishing.
On the use of EAF steel slag in dense-graded bituminous mixtures, it was observed that com-
pared to the control mix, the slag-modified mixes showed better mechanical properties in terms
of rutting resistance, fatigue life and resilient modulus (Asi, Qasrawi, & Shalabi, 2007). The
improved results were justified by the angularity properties of the EAF steel slag aggregates. For
steel slag asphalt mixtures, Arabani, Mirabdolazimi, and Sasani (2010) reported higher fatigue
life from indirect tensile fatigue tests and higher resistance to permanent deformation compared
to control mixtures. The improvements were attributed to higher angularity, shear strength and
resistance to polishing, of steel slag than the natural aggregates. Various other studies have
reported better fatigue and cracking resistance (Kavussi & Qazizadeh, 2014; Pasetto & Baldo,
2010; Wen, Wu, & Bhusal, 2015; Ziari, Nowbakht, Rezaei, & Mahboob, 2015), and better rut-
ting resistance (Ahmedzade & Sengoz, 2009; Behnood & Ameri, 2012; Wu, Xue, Ye, & Chen,
2007) of dense graded asphalt mixtures with steel slags. Also, in few studies on OGFC mixes
containing EAF steel slag, it was reported that they showed lower rut depth compared to control
mixtures (without steel slag) (Hainin, Rusbintardjo, Hameed, Hassan, & Yusoff, 2014; Wang &
Wang, 2011). However, only a single replacement content of steel slag was employed in these
studies.
The main objective of this study is to evaluate the mechanical properties of OGFC mixtures
with different percentages of EAF steel slag as a substitution for natural coarse aggregates. Five
replacement percentages are considered: 0%, 25%, 50%, 75% and 100% replacement (by weight)
of natural coarse aggregates by EAF steel slag, along with two modified asphalt binders (poly-
mer modified binder, PMB and crumb rubber modified binder, CRMB). The OGFC mix with
0% EAF slag replacement is used as the control. The different tests conducted in the study are
static creep test, dynamic creep test, Hamburg wheel tracking device test, indirect tensile stiffness
modulus (ITSM) test, and indirect tensile fatigue test (ITFT). The results are statistically analysed
 Road Materials and Pavement Design 3

by analysis of variance (ANOVA) conducted at a level of significance (α) of 0.05, followed by

Tukey’s multiple comparison procedure (Ott & Longnecker, 2015). Steel slag content and binder
type were considered as the factors for ANOVA. This is followed by a multiple linear regres-
sion analysis using a bidirectional variable selection technique to model OGFC performance
parameters as a function of OGFC mix design parameters.

Materials and experimental programme

Materials
The conventional (natural source) aggregates were collected from a stone crusher plant near
Shillong city in the state of Meghalaya (India). EAF steel slag was collected from steel produc-
tion unit of Jindal Steel Power Limited, Odisha (India). Physical properties of both aggregates
(natural and steel slag) are presented in Table 1. The surface morphology of both natural aggre-
gate and EAF steel slag was evaluated through scanning electron microscopy (SEM). The SEM
micrographs of natural and steel slag aggregates are shown in Figure 1. The SEM image of EAF
slag shows the presence of many surface pores indicating a porous morphology as compared to
the natural aggregates. On the other hand, natural aggregates are smoother and rounded than the
EAF slag aggregates. The chemical composition of both natural and steel slag aggregates was
analysed through X-ray fluorescence (XRF) and the results are presented in Table 2. EAF steel
slag has higher percentages of iron oxides, which explains the higher specific gravity of EAF
steel slag aggregates than the natural aggregates. It is also observed that EAF steel slag has a
CaO/SiO2 ratio of 2.410, which is much higher than that for the natural aggregates (0.162). The
CaO/SiO2 ratio is considered as a measure of the level of alkalinity of aggregates and a higher
CaO/SiO2 ratio is expected to improve the bond with asphalt binder and thus the resistance to
moisture-induced damages (Chen et al., 2015; Xie et al., 2012). The EAF steel slag used in this
study had undergone air aging for about one month in the manufacturer’s landfill facility, as per
the information gathered from the steel plant engineers. Prior to its use in OGFC mixes, the slag
was also tested for volumetric expansive tendency for 7 days under water as per the guidelines
of ASTM D4792 (2013). A maximum expansion of 1% after 7 days has been recommended for
steel slag (Wu et al., 2007). The results indicated that the expansion was much below 1% after 7
days.
Toxicity and leaching characteristics of EAF steel slag are of prime concern before its reuse for
civil engineering purposes. The leaching characteristics of EAF steel slag aggregate and asphalt
mixtures containing the steel slag were evaluated through TCLP (toxicity characteristic leaching
procedure) following the US-EPA Method-1311 (USEPA, 2008). This test is important to ascer-
tain the suitability of industrial by-product materials for potential use as aggregates (Onuaguluchi

Table 1. Physical properties of natural and steel slag aggregates.

Property Natural aggregates EAF steel slag Test method

Bulk specific gravity 2.96 3.26 ASTM C127-15

Water absorption (%) 0.45 0.51 ASTM C127-15
Angularity number 9.00 10.86 IS 2386 P1-16
VCAdrc (%) 44.3 45.1 ASTM C29-17
Aggregate impact value 15.76 14.56 IS 2386 P4-16
Los Angles abrasion value (%) 19.47 13.3 ASTM C131-14
Aggregate crushing value (%) 19.78 14.67 IS 2386 P4-16
Flat and elongated particles (%) 18.50 9.30 IS 2386 P1-16
4 M. L. Pattanaik et al.

(a) (b)
Figure 1. Scanning electron micrographs (SEM) of (a) EAF steel slag and (b) natural stone aggregates.

Table 2. Chemical composition of aggregates (by XRF analysis).

Oxide content (%)

Type of oxide Natural aggregate EAF steel slag

Calcium – CaO 8.10 30.70

Iron – FeO 11.63 31.05
Silica – SiO2 50.01 12.77
Aluminum – Al2 O3 16.03 12.0
Magnesium – MgO 2.35 7.72
Manganese – MnO 0.16 0.48
Titanium Dioxide-TiO2 1.72 -
Phosphorous – P2 O5 0.76 0.42
Potassium – K2 O 4.25 0.08
Sodium – Na2 O 2.69 0.09

& Eren, 2012). The procedure measures the mobility of both organic and inorganic substances
present in the form of solid, liquid and multiphasic conditions. The results of TCLP analysis
are presented in Table 3 for EAF steel slag powder and EAF slag-OGFC mixture powder. The
concentrations of all heavy metals is well below the permissible IS limits (IS: 10500, 2012). The
concentrations of heavy metals further reduces for the EAF slag-OGFC mixtures. This shows
that the usage of EAF steel slag is safe as a partial replacement of natural aggregate in OGFC
mixtures, and it would cause no instant or long-term vulnerability to the environment.
Due to high air voids content, OGFC mixtures demand the use of stiffer binder and stabilising
additives (like natural or synthetic fibres) to achieve a thicker binder film on aggregates and a low
potential for binder draindown. Two modified asphalt binders: polymer modified binder (PMB)
and crumb rubber modified binder (CRMB) were used for the design and preparation of OGFC
mixtures. These binders were obtained from Tikitar Industries, Gujarat (India). Physical prop-
erties of modified binders are shown in Table 4. In the present study, cellulose fibres were used
as stabilising additive at the dosage of 0.3% by weight of the mixture. The fibres were obtained
from Organo Chemical Industries, Mumbai (India). The naked-eye view and SEM images of
fibres are presented in Figure 2. The SEM image shows that cellulose fibre has rougher surface
and branched (thread type) structure that helps to reduce binder draindown by absorbing and
stiffening the binder.
 Road Materials and Pavement Design 5

Table 3. Results of TCLP analysis for EAF steel slag.

TCLP leaching concentration (mg/L)

Element Permissible limit (mg/L)* EAF-Powder EAF-Mix

Magnesium (Mg) 100 13.5 1.7

Aluminium (Al) 0.2 0.015 0
Silica (Si) – 0.89 0.23
Calcium (Ca) 200 93.2 13.4
Vanadium (V) 0.25 0.093 0
Chromium – total (Cr) 0.05 0.009 0
Manganese (Mn) 0.3 0.207 0.157
Iron (Fe) 0.3 0 0
Cobalt (Co) 0.05 0 0
Nickel (Ni) 0.02 0.007 0
Copper (Cu) 1.5 0.082 0
Zinc (Zn) 15 0.168 0.034
Selenium (Se) 0.01 0 0
Cadmium (Cd) 0.003 0.001 0
Mercury (Hg) 0.001 0 0
Lead (Pb) 0.01 0.008 0.001
*Permissible limits as per IS 10500 (2012).

Table 4. Physical properties of asphalt binders.

Requirements as
per IS
15462-2004 Results
Tests Test methods PMB CRMB PMB CRMB

Penetration at 25°C, dmm IS 1203 30–50 < 50 42 30.88

Softening point, (R&B), IS 1205 Min. 60 Min. 60 70.95 70.65
(°C)
Flash point, by COC, (°C) IS 1209 Min. 220 Min. 220 320 340
Elastic recovery of half IS15462 Annex A Min. 70 Min.50 76.5 72.5
thread in ductilometer
at 15°C, (%)
Separation difference in IS 15462 Annex B Max. 3 Max. 4 1.9 2.5
Softening point, R&B,
(°C)
Viscosity at 150°C, Poise IS 1206 Part1 3–9 3–9 7.78 8.12
Thin Film Oven Test (TFOT) residue
Loss in weight, (%) IS 9382 Max. 1 Max. 1 0.2 0.4
Increase in softening IS 1205 Max. 5 Max. 5 2.8 3.65
point, (°C)
Reduction in penetration IS 1203 Max. 40 Max. 40 34 31.32
of residue, at 25°C, (%)
Elastic recovery of half IS 15462 Annex A Min. 50 Min. 35 74.5 68.5
thread in ductilometer
at 25°C (%)

Design of OGFC mixtures

Aggregate gradation as per ASTM D7064 (2008) guidelines was chosen for the design of OGFC
mixtures (Figure 3). In the present study, mix design was performed as per design procedure
6 M. L. Pattanaik et al.

(a) (b)
Figure 2. Naked-eye view and scanning electron micrographs (SEM) of cellulose fibres.

Figure 3. Aggregate gradation for OGFC mixtures.

specified in ASTM D7064 (2008). The OGFC design procedure involves fabrication of OGFC
mix specimens at various binder contents followed by their evaluation for: stone-on-stone con-
tact, air voids content, binder draindown, and abrasion loss measurement through Cantabro test
under both unaged and aged conditions.
OGFC specimens were compacted by means of a Marshall compactor by imparting 50 blows
to each specimen face. The ASTM D7064 (2008) standard recommends to use ‘50 gyrations
of the Superpave gyratory compactor (SGC) or other compactor providing equivalent compacted
density’ for compaction of OGFC mix specimens. Several studies on OGFC have reported a good
agreement between compaction achieved through 50 gyrations of gyratory compactor and that
achieved through 50 blows of the Marshall compactor (Alvarez et al., 2006; Mallick, Kandhal,
Cooley, & Watson, 2000; Watson, Ann Moore, Williams, & Allen Cooley, 2003). In this study
also 50 blows of the Marshall compactor were chosen for compaction of OGFC specimens. The
 Road Materials and Pavement Design 7

Table 5. Mix design parameters of OGFC mixtures at the OBC.

Unaged Aged
Steel slag Air void % Draindown abrasion loss abrasion loss
Binder type content (Min 18) % (Max 0.3) % (Max 20) % (Max30)

PMB 0 23.93 0.040 18.15 20.69

25 23.22 0.110 13.43 16.56
50 22.91 0.126 10.58 12.97
75 22.13 0.153 7.90 9.97
100 22.96 0.162 13.92 16.64
CRMB 0 24.33 0.060 18.68 21.72
25 23.39 0.075 14.78 17.15
50 22.95 0.110 11.39 13.73
75 22.75 0.127 8.99 10.31
100 23.29 0.142 14.82 17.44

mix design trials used three asphalt binder contents of 5.5%, 6.0% and 6.5% by mix weight
(excluding weight of the fibre). Twelve OGFC specimens were prepared at each binder content.
Three compacted specimens were used for unaged abrasion loss test and three for aged abrasion
loss test. Remaining six loose (uncompacted) specimens were used to determine the theoretical
maximum density and binder draindown. The minimum binder content at which the OGFC mixes
were able to meet the design requirements specified by ASTM D7064 (2008) was selected as the
optimum binder content (OBC).
Results of mix design parameters showed that the maximum abrasion loss criterion (20% for
unaged abrasion loss and 30% for aged abrasion loss) was not met at the lower binder content
of 5.5% for some mixtures. On the other hand, at 6.5% binder content the binder draindown
was higher than the maximum specified value (0.3%). All design requirements were met at 6%
binder content, and therefore 6% was selected as OBC for the OGFC mixtures with all slag con-
tents (0%, 25%, 50%, 75% and 100%) and for both binders (PMB and CRMB). The detailed
procedures adopted for evaluating mix design parameters of OGFC mixtures and the test results
of design parameters for all combination of mixtures can be found elsewhere (Pattanaik, Choud-
hary, & Kumar, 2018b). The mix design parameters at the obtained OBC are presented in Table 5.
For the performance tests, specimens were prepared at the OBC of the mixtures. Three replicates
were used for each performance test on OGFC mixtures with all combinations of EAF slag and
natural aggregates.

Performance properties of OGFC mixtures

Static creep test, dynamic creep test, Hamburg wheel tracking device test, stiffness modulus test
and indirect tensile fatigue test were performed to evaluate the performance of EAF slag modified
OGFC mixtures. Each of these tests is discussed further.

Static creep test

The static creep test was performed as per BS 598-111 (1995) on a universal testing machine
(UTM) to determine the permanent deformation characteristics of asphalt mixtures subjected
to a constant load. The test set assembly for the test is shown in Figure 4. As the permanent
deformation distress is mobilised under the high service temperatures, the test was carried out at
40°C. Prior to the test, the OGFC specimen and test set-up with all accessories were kept inside a
temperature-controlled chamber at the test temperature for 4 h. During the test, the specimen was
8 M. L. Pattanaik et al.

Figure 4. Test setup for static and dynamic creep tests (no confinement).

subjected to a pre-conditioning load of 10 kPa for 10 min and then to a constant axial stress
of 100 ± 2 kPa for a period of 1 h. The deformation of the specimen was noted every 60 s.
Linear variable displacement transducers (LVDTs) were mounted on both sides of the specimen
to measure the deformation. After completion of the loading time, the stress was removed and
strain was recorded for another 1 h. The axial strain was measured as per Equation (1):

h(t)
ε(t) = (1)
h
where ε(t) = axial strain of the specimen at time ‘t’; h(t) = axial deformation (change in the
height of the specimen) at time ‘t’; h = specimen height.

Dynamic creep test

Dynamic creep test was performed using the guidelines of BS DD 226 (1996) on a UTM to eval-
uate the permanent deformation characteristics of asphalt mixtures under dynamic loading condi-
tions. Specimens were pre-conditioned in a temperature-controlled chamber at the test tempera-
ture (40°C) for 4 h. A load of 10 kPa was applied for a period of 600 ± 6 s to confirm that loading
platen and samples surface were entirely in contact with each other. After the pre-conditioning
period, the stress was removed and a stress of 100 kPa was applied on the horizontal diametral
 Road Materials and Pavement Design 9

Figure 5. (a) Hamburg wheel tracking test device, (b) Specimens prepared for the HWTD test.

plane of the cylindrical specimen in a square waveform with 0.5 Hz frequency. The BS DD 226
(1996) standard recommends the test to be performed for 1800 cycles; however, the test was con-
tinued up to 10,000 cycles in this study. Each cycle had a loading period of 1 s and a rest period
of 1 s. During the test, the axial deformation of the samples was recorded by LVDTs. The axial
strain of the test specimen was determined from the measured deformation at the end of the rest
period.

Hamburg wheel tracking device test

The Hamburg wheel tracking device (HWTD) was developed in Hamburg, Germany, to eval-
uate the combined effect of rutting and moisture susceptibility of bituminous mixtures. HWTD
test was performed as per AASHTO T324 (2017) test method. The device rolls a steel wheel
of diameter 203 mm and width 47 mm across the surface of a slab specimen or two SGC
specimens. The reciprocating wheel carries a load of 705 N and travels 230 mm in a single
pass before reversing its direction. Two SGC OGFC mixtures of 150 mm diameter and 60 mm
height constituted one specimen for the HWTD test. The SGC specimens prepared at the OBC
(6%) were sawed along a secant line and then joined together for the test. Each set of spec-
imens was submerged in a water bath (in the HWTD) maintained at 40°C and subjected to
20,000 passes (52 passes per minute) of the loaded wheel. Each set of the specimen was
preconditioned for 30 min at the test temperature before the commencement of the test. Two
replicates were tested at each combination of EAF steel slag content (0%, 25%, 50%, 75%
and 100%) and binder type (PMB and CRMB). The test assembly for the test is shown in
Figure 5.
10 M. L. Pattanaik et al.

Figure 6. Indirect tensile stiffness modulus (ITSM) test assembly.

Stiffness modulus
Stiffness modulus test is a non-destructive test and is used to measure the stiffness properties
of asphalt mixtures. The indirect tensile stiffness modulus (ITSM) test was performed in accor-
dance with EN 12697-Part 26 (2012) at an intermediate pavement service temperature of 20°C.
Prior to the stiffness modulus testing, the OGFC specimen and ITSM test accessories were
pre-conditioned in temperature-controlled chamber at the test temperature for 4 h. The test set
assembly for the test is shown in Figure 6. During the test, a compressive repetitive load pulse
was applied by means of the load actuator on vertical diameter of the cylindrical specimen. The
load actuator applied a repeated haversine load pulse with a rise time of 124 ms (i.e. the time for
the application of load from zero to the peak). The resultant peak transient deformations in both
horizontal and vertical directions were measured. During the test, the specimen was subjected
to 10 repetitive load pulses for conditioning and 5 repetitive load pulses for calculating the stiff-
ness modulus. After completion of the 15 pulses, the sample was axially rotated by 90° and then
allowed to stabilise at the test temperature. The average stiffness modulus obtained from the two
specimen axis orientations was reported as the stiffness modulus of the specimen. The stiffness
modulus was determined from Equation (2):

F × (μ + 0.27)
Sm = (2)
z×h
 Road Materials and Pavement Design 11

Figure 7. Test setup and failure of a samples during indirect tensile fatigue test (ITFT).

where Sm = the stiffness modulus of the specimen, MPa; F = peak value of the applied vertical
load, N ; z = amplitude of the horizontal deformation for the duration of the load cycle, mm;
h = specimen height, mm; μ = Poisson’s ratio (assumed 0.35).

Indirect tensile fatigue test (ITFT)

The ITFT was carried to evaluate the fatigue characteristics of OGFC mixes. The test was con-
ducted as per EN 12697-Part 24 (2012) at the test temperature of 20°C. ITFT is a simple and
economical test for evaluating fatigue performance of asphalt mixtures and can utilise cylindri-
cal specimens prepared in the laboratory or cored samples from a pavement. A test specimen was
subjected to a repetitive compressive load of 200 kPa with a haversine load signal through the
vertical diametrical plane to achieve a strain in the range of 100–400 με. In this test procedure,
the loading progresses as a relatively uniform tensile stress normal to the direction of the applied
load and along the vertical diametral plane, which causes the specimen to fail by splitting along
the central part of the vertical diameter. The repetitive haversine load was applied at a frequency
of 2 Hz with a loading time of 0.1 s and rest time of 0.4 s. LVDTs were mounted horizontally
to the test specimen to measure the deformations. Fatigue life/fracture life, defined as the total
number of load applications before a complete fracture of the specimen, was used as the measure
of fatigue resistance. The test setup and the sample failure at the end of the test are shown in
Figure 7.

Results and discussion

Static creep
The static creep test was carried for evaluating the permanent deformation properties of OGFC
mixes. Five important strain components can be identified from the strain vs. time graph
12 M. L. Pattanaik et al.

Figure 8. Strain components measured in a static creep test.

Figure 9. Total strain results from static creep test.

generated from the test: initial strain, final strain, total strain, reversible strain, and permanent
strain. Figure 8 presents a schematic diagram showing these strains. The strain generated after
15 s from the start of loading is referred to as initial strain. The strain generated from 15 to 3600 s
after completion of loading period is referred as final strain. The sum of the initial strain and
final strain is the total strain. After loading period, the recovered strain from the starting of the
unloading till 1 h (at the end of unloading) is referred as reversible strain. Permanent strain is
the residual strain remaining after the sample is allowed to recover. Figure 9 shows the total
strain values for each OGFC mix with both PMB and CRMB binders. It is seen that the total
 Road Materials and Pavement Design 13

Figure 10. Ratio of recovered (R) to permanent (P) strain (R to P ratio) from static creep test.

strain reduces with an increase in percentage substitution of coarse natural aggregates with EAF
steel slag up to 75% and then increases further at 100% replacement. Steel slag-OGFC mixtures
present lower total strain values as compared to the control mixtures, varying from 46–76% to
26–77% respectively, for PMB-OGFC and CRMB-OGFC mixtures. The decrease in strain with
increase in EAF steel slag content is attributed to rougher surface texture, higher angularity, bet-
ter interlocking and higher specific gravity of the EAF steel slag in steel slag-OGFC mixtures
compared to control mixtures with no EAF slag (Pasetto & Baldo, 2010).
The behaviour of a deformed material after removal of the load is termed as recovery. During
the loading process, the mixture undergoes deforming process consisting of elastic and plas-
tic strain components. Since the specimen is subjected to creep loading, the accumulation of
both viscoelastic and viscoplastic strains occurs. After elimination of the load, elastic strain is
instantly recovered and during the unloading period the viscoelastic strain recovers (depend-
ing on the length of recovery period). At end of the cycle, the remaining strain consists of the
plastic and viscoplastic components plus the remainder of viscoelastic strain that has not been
recovered, which may result in cracking and permanent deformation in the sample (Carvalho,
2012). Development of cracks and accumulation of creep results in alligator cracking and rutting
respectively, which are the two foremost distresses in bituminous pavements. In consequence,
asphalt mixture recovery contributes to the deformation characterisation and prediction as well
(Valkering, Lancon, Dehilster, & Stoker, 1990).
Figure 10 presents the results of the ratio of recovered strain to permanent strain (R-P ratio) of
OGFC mixtures. The ratio indicates the relative distribution of recovered and permanent strains
in the mixtures. A higher ratio is desirable as it corresponds to better strain recovery properties.
The R-P ratio increases with increase in percentage substitution of coarse natural aggregates with
EAF steel slag up to 75% and then it falls at 100% replacement. It is further observed that PMB
mixtures have higher R-P ratio than the mixtures with CRMB binder. The residual binder content
in the EAF steel slag-OGFC mixtures increases with an increment of slag percentage due to the
higher specific gravity of the steel slag. The increased residual binder (at 100% replacement) may
lead to the loads being carried by the asphalt binder instead of the aggregate structure. Therefore,
mixtures with 100% coarse steel slag show slightly higher strain.
14 M. L. Pattanaik et al.

Table 6. ANOVA results for total strain from static creep test.

Factor DF F value P value Significance

Binder type 1 8.41 0.015 Yes

Slag content 4 51.10 < 0.01 Yes
Interaction 4 1.19 0.37 No
Multiple comparisons based on Tukey’s procedure
PMB CRMB
0% 25% 50% 75% 0% 25% 50% 75%
25% S S
50% S S S S
75% S S S S S S
100% S S NS S S S NS S
Notes: ‘S’: Significant; ‘NS’: Not Significant; ‘DF’: Degree of Freedom.

Table 6 shows the results of ANOVA performed on total strain results at 5% significance
level. Steel slag content and binder type are found to have a statistically significant effect on
the total strain values. Further, a higher F-value for slag content indicates that it has a higher
influence on the total strain as compared to binder type. Non-significant interaction between the
two factors (binder type and slag content) indicates that the total strain values follow a similar
trend for varying slag contents for both binder types. Multiple comparison results show that the
substitution of natural aggregate by steel slag leads to significant changes in the total strain values
(in comparison to control mixes).

Dynamic creep
Dynamic creep test measured the accumulated axial strain in the specimen as a function of num-
ber of cycles. The test was run for 10,000 cycles. The accumulated strain plots of OGFC mixtures
with all steel slag contents and PMB and CRMB binders are presented in Figures 11 and 12. It
is observed that with increase in steel slag content up to 75%, the axial strain of the mixtures
reduces.
Steel slag-OGFC mixtures present lower accumulated strain values as compared to the control
mixtures, varying from 45–77% to 31–79% respectively, for PMB-OGFC and CRMB-OGFC
mixtures. This shows that the steel slag-OGFC mixtures demonstrate higher rutting resistance
than the control mixtures. Beyond 75% steel slag content there is slight increase in the accumu-
lated strain values; however, they are still lower than those for control mixtures. The results could
be ascribed to high bearing strength, angularity and hardness of EAF steel slag aggregate. As the
steel slag is rougher than the natural aggregate, a better mechanical interlock (cohesive bond)
between the binder and steel slag aggregate is developed, leading to high permanent deformation
resistance. Figure 13 shows the accumulated strain values after 1800 cycles in the dynamic creep
test. It is seen that the strain decreases with increase in percentage substitution with EAF steel
slag up to 75% and then increases with further increment in steel slag content. It is to be noted
that similar results were also seen from static creep test results. The resistance to permanent
deformation increases with increase in percentage substitution with steel slag and then decreases
at 100% substitution.
Table 7 shows the results of ANOVA performed on accumulated strain after 1800 cycles at 5%
significance level. Steel slag content and binder type are found to have a statistically significant
 Road Materials and Pavement Design 15

Figure 11. Accumulated strain results from the dynamic creep test for PMB-OGFC mixes.

Figure 12. Accumulated strain results from the dynamic creep test for CRMB-OGFC mixes.

effect on the accumulated strain values. Further, a higher F-value for slag content indicates that
it has a higher influence on the accumulated strain as compared to binder type. Non-significant
interaction between the two factors (binder type and slag content) indicates that the accumulated
strain values follow a similar trend for varying slag contents for both binder types. Multiple
comparison results show that the substitution of natural aggregate by steel slag leads to significant
changes in the accumulated strain values in comparison to control mixes.
16 M. L. Pattanaik et al.

Figure 13. Accumulated strain after 1800 cycles from dynamic creep test.

Table 7. ANOVA results for accumulated strain after 1800 cycles from dynamic creep test.

Factor DF F value P value Significance

Binder type 1 71.37 < 0.01 Yes

Slag content 4 93.10 < 0.01 Yes
Interaction 4 3.70 0.08 No
Multiple comparisons based on Tukey’s procedure
PMB CRMB
0% 25% 50% 75% 0% 25% 50% 75%
25% S S
50% S S S S
75% S S S S S S
100% S S NS S S S NS S

Hamburg Wheel Tracking Device Test

The HWTD measured the accumulated rut depth in the specimen as a function of number of
wheel passes. The accumulated rut depth plots of OGFC mixtures at all steel slag contents with
PMB and CRMB binders are presented in Figures 14 and 15. In the initial stages of the test, the rut
depth increases rapidly due to secondary compaction of the test specimen, after which the rate of
increase in rut depth stabilises and becomes steady. Steel slag-OGFC mixtures present lower rut
depth values as compared to the control mixtures, varying from 15–60% to 14–61% respectively,
for PMB-OGFC and CRMB-OGFC mixtures compared to their control. This shows that the steel
slag-OGFC mixtures demonstrate higher rutting resistance than the control mix. It is observed
that with increase in EAF steel slag content beyond 75% there is slight increase in the rut depth;
however, it remains lower than the control mixtures. The high strength, angularity and rough
 Road Materials and Pavement Design 17

Figure 14. Hamburg wheel tracking device (HWTD) test results for PMB-OGFC mixtures.

Figure 15. Hamburg wheel tracking device (HWTD) test results for CRMB-OGFC mixtures.

surface texture of EAF steel slag provides an effective interlock between the aggregates in OGFC
mixtures. This helps to reduce changes in the orientation of aggregates through rearrangement
under the passes of HWTD, and therefore enhance the permanent deformation resistance. The
HWTD test results are also in agreement with the results of dynamic creep and static creep tests.
18 M. L. Pattanaik et al.

Table 8. ANOVA results for HWTD rut depth after 20,000 passes.

Factor DF F value P value Significance

Binder type 1 50.00 < 0.01 Yes

Slag content 4 82.17 < 0.01 Yes
Interaction 4 1.82 0.18 No
Multiple comparisons based on Tukey’s procedure
PMB CRMB
0% 25% 50% 75% 0% 25% 50% 75%
25% S NS
50% S S S S
75% S S NS S S S
100% S NS S S S NS NS S

The decrease in rutting resistance at 100% substitution is likely due to the existence of higher
residual binder as discussed in the case of static creep test. It is further observed that up to the
maximum number of passes used in the test, there is no stripping in all OGFC mixtures (all slag
contents and both binders) as there is no change in the slope of the rut depth versus number of
passes curves beyond the initial change due to secondary compaction. This indicates superior
resistance of all OGFC mixtures to the effects of moisture which may be attributed to the better
asphalt bonding with rough and open-textured EAF steel slag aggregates having higher calcium
oxide content (EAF steel slag has a CaO/SiO2 ratio of 2.410, which is much higher than that for
the natural aggregates which is 0.162).
Table 8 shows the results of ANOVA performed on rut depth results obtained after 20,000
passes in HWTD test. Steel slag content and binder type are found to have a significant effect on
the rut depth. Further, a higher F-value for slag content indicates that it has a higher influence
on rut depth as compared to binder type. Non-significant interaction between the two factors
(binder type and slag content) indicates that rut depth values follow a similar trend for varying
slag contents for both binder types. Multiple comparison results show that the substitution of
natural aggregate by steel slag leads to significant changes in the rut depth values in comparison
to control mixtures.

Stiffness modulus
Stiffness is also considered as a measure of the ability of the bituminous layers to distribute a
load. The indirect tensile stiffness modulus (ITSM) was measured at the test temperature of 20°C.
The stiffness modulus results of OGFC mixtures with different percentage substitution of EAF
steel slag and both modified binders (PMB and CRMB) are presented in Figure 16. The stiffness
modulus values of the OGFC mixtures increase with increase in percentage substitution of natural
aggregates with EAF steel slag up to 75% and then show a slight decrease with further increment
to 100%. Steel slag-OGFC mixtures presented higher values of stiffness modulus as compared
to the control mixtures, varying from 30–71% to 41–80% for PMB-OGFC and CRMB-OGFC
mixtures, respectively.
Table 9 shows the results of ANOVA performed on stiffness modulus results. Steel slag content
and binder type are found to have a significant effect on the stiffness modulus. Further, a higher
F-value for slag content indicates that it has a higher influence on stiffness modulus as compared
to binder type. Non-significant interaction between the two factors (binder type and slag content)
 Road Materials and Pavement Design 19

Figure 16. Stiffness modulus results of OGFC mixtures.

Table 9. ANOVA results for stiffness modulus.

Factor DF F value P value Significance

Binder type 1 32.42 < 0.01 Yes

Slag content 4 309.14 < 0.01 Yes
Interaction 4 0.891 0.49 No
Multiple comparisons based on Tukey’s procedure
PMB CRMB
0% 25% 50% 75% 0% 25% 50% 75%

25% S S
50% S S S S
75% S S S S S S
100% S S NS S S S NS S

indicates that stiffness modulus values follow a similar trend for varying slag contents for both
binder types. Multiple comparison results show that the substitution of natural aggregate by steel
slag leads to significant changes in the stiffness modulus values in comparison to control mixes.

Fatigue resistance
Fatigue cracking is one of the key distresses in asphalt pavements. Repeated traffic loading is
the main factor responsible for fatigue cracking and it can result in a significant drop in the
serviceability of the pavement. Fatigue performance of OGFC mixtures was evaluated through
indirect tensile fatigue test (ITFT) at 20°C in terms of fracture life (i.e. the number of load cycles
until the complete failure of the specimen). Fracture life of OGFC mixtures for all percentage
substitution of steel slag with both PMB and CRMB binders are presented in Figure 17. It can
20 M. L. Pattanaik et al.

Figure 17. Fracture life results of OGFC mixes.

be observed that all steel slag-OGFC mixtures have higher fracture lives as compared to mixes
with natural aggregates.
Steel slag-OGFC mixtures presented higher fracture lives varying from 82–234% to 78–214%
respectively, for PMB-OGFC and CRMB-OGFC mixtures compared to their control mixtures.
Higher fracture lives of steel slag-OGFC mixtures may be attributed to the angularity and rough
texture of the coarse EAF steel slag aggregates resulting in increased internal friction angle of
mixtures and improvement in the particle interlocking (Kavussi & Qazizadeh, 2014). Fracture
life of OGFC mixtures increases with increase in EAF steel slag up to 75% then slightly decreases
with further increment. The fatigue life improvement up to 75% EAF steel slag was observed
because EAF steel slag improved the interlocking of the aggregate structure, whereas in case of
mixtures by 100% EAF steel slag substitution the decrease in fatigue life was due to increased
amount of residual binder due to high specific gravity of EAF steel slag.
Table 10 presents the results of ANOVA performed on fracture life results. Based on the
ANOVA, EAF steel slag content and binder type are found to have a significant effect on fracture
life. A higher F-value for slag content indicates that it has a higher influence on fracture life as
compared to binder type. Non-significant interaction between the two factors (binder type and
slag content) indicates that fracture life values follow a similar trend for varying slag contents for
both binder types. Multiple comparison results show that the substitution of natural aggregate by
steel slag leads to significant changes in the fracture life values in comparison to control mixtures.

Statistical modelling
Multiple linear regression analysis was performed to model OGFC performance parameters with
OGFC design parameters. The regression analysis was conducted to determine the best OGFC
design variables that are related to the performance parameters. Seven design variables, including
bulk density, air voids content, unaged abrasion loss, aged abrasion loss, VCAmix , VCAdrc , and
VCAmix to VCAdrc ratio, were considered as independent variables, whereas the six dependent
 Road Materials and Pavement Design 21

Table 10. ANOVA results for fracture life.

Factor DF F value P value Significance

Binder type 1 39.53 < 0.01 Yes

Slag content 4 1015.02 < 0.01 Yes
Interaction 4 2.604 0.07 No
Multiple comparisons based on Tukey’s procedure
PMB CRMB
0% 25% 50% 75% 0% 25% 50% 75%
25% S S
50% S S S S
75% S S S S S S
100% S S S S S S S S

Table 11. Developed regression models for OGFC performance parameters.

Performance Regression model Adjusted R2 R2 P-value

parameter

Axial strain in AS = − 211944.4 0.834 0.871 < 0.001

dynamic creep + 1740.2*UAL
test + 4468.9*VCAdrc
Fracture life FL = 684741.8 0.995 0.996 < 0.001
− 17477.2*VCAmix
− 2562.5*UAL
+ 5316.1*AV
Total strain in static TS = − 0.199877120 0.942 0.955 < 0.001
creep test + 0.001502171*UAL
+ 0.004246463*VCAdrc
Permanent strain in PS = − 0.0047861816 0.955 0.960 < 0.001
static creep test + 0.0006797346*UAL
Stiffness modulus SM = 18172.0 0.995 0.997 < 0.001
− 148.3*AAL
− 319.3*AV
− 2462.2*BD
Rut depth RD = 51.2 0.936 0.941 < 0.001
+ 0.351*AAL
− 1.77*AV
− 6.05*BD
Notes: UAL: unaged abrasion loss; AAL: aged abrasion loss; BD: bulk density; AV: air voids; VCAmix : voids in coarse
aggregate (mix); VCAdrc : voids in coarse aggregate (dry rodded condition); AS: axial strain after 1800 cycles; FL:
fracture life, TS: total strain, PS: permanent strain; SM: stiffness modulus; RD: rut depth after 20,000 passes of Hamburg
wheel tracking device test.

variables were stiffness modulus, axial strain after 1800 cycles in dynamic creep test, HWTD
rut depth after 20,000 passes, total strain in static creep test, permanent strain in static creep
test, and fracture life in ITFT. The multiple linear regression analysis was performed using the
open-source statistical software package ‘R’. Bidirectional elimination approach was used for
model development based on Akaike information criterion (AIC) as the fit criterion (Konishi
& Kitagawa, 2008). AIC calculates the relative quality of candidate regression models and is
based on log likelihood (likelihood is a technique for parameter estimation and to draw statistical
22 M. L. Pattanaik et al.

Figure 18. Predicted and observed performance parameters.

inferences in various situations) and the number of model parameters (Wang, Wright, Buswell,
& Brownlee, 2013). For a set of candidate models, the selected model was the one with the
minimum AIC value. Under the bidirectional elimination approach, model development starts
with no variables initially (only the constant term). A variable is added if its inclusion provides
statistical improvement in the fit, and this process it iterated until none of the variable’s addition
provides statistically significant improvement based on the AIC criterion. Additionally, in each
step that adds a new variable there is a possibility to delete a previously selected variable.
Table 11 shows the regression models obtained for each OGFC performance parameter. The
table also shows three model performance indicators: coefficient of determination (R2 ), adjusted
 Road Materials and Pavement Design 23

coefficient of determination (R2adj ), and p-value. It is to be noted that the models in Table 11
include the variable found to be statistically significant based on the AIC criterion. Figure 18
shows the predicted and observed performance parameters along with the line of equality. The
predicted and observed values lie in proximity to the line of equality indicating that the model
predictions are close to the actually observed data. It is observed from the obtained models
that unaged abrasion loss is a significant variable for four of the five modelled performance
parameters (AS, TS, PS, and FL), whereas for SM and RD, the aged abrasion loss is signifi-
cant. VCA-based parameters (VCAmix , VCAdrc , and VCA ratio) are also significant for OGFC
performance parameters. The developed models indicate the importance and significance of
abrasion loss and proper stone-on-stone contact towards achieving improved rutting and fatigue
performance for the different combination of OGFC mixtures studied.

Conclusions
The present study investigated mechanical properties of OGFC mixtures with different contents
of electric arc furnace (EAF) steel slag – a waste from steel-making industries. OGFC mixtures
were fabricated with five percent replacements of natural aggregates with EAF steel slag and two
modified asphalt binders (polymer and crumb rubber modified binders). Mechanical properties
were evaluated through static creep test, dynamic creep test, indirect tensile stiffness modulus
test, Hamburg wheel tracking test and indirect tensile fatigue tests. Based on the analysis of the
test results, the following conclusions are drawn:

• Both PMB- and CRMB-OGFC mixtures with the substitution of coarse natural aggregate
with EAF steel slag showed improved performance in terms of results of static creep,
dynamic creep, Hamburg wheel tracking, stiffness modulus and fracture life tests.
• OGFC mixtures with 75% substitution of coarse natural aggregates with EAF steel slag
presented the best performance in terms of permanent deformation, stiffness modulus and
fracture life among all the combinations studied.
• Based on statistical analyses, EAF steel slag content was found to have a higher influence
on the performance of OGFC mixtures compared to the binder type. Regression modelling
results showed that abrasion loss and stone-on-contact criteria were significant design vari-
ables affecting the mechanical properties of OGFC mixtures with and without EAF steel
slag.
• Favourable physical and chemical characteristics of EAF steel slag aggregates in terms
of higher angularity, rough surface texture, high specific gravity and high calcium oxide
content, enabled them to perform as a better coarse aggregate in OGFC mixtures.

Based on the findings of this laboratory-based study, incorporation of EAF steel slag as the coarse
aggregate showed improvements in the rutting and fatigue performance of OGFC mixtures. Use
of EAF steel slag will also help to reduce the current consumption of natural aggregates in high-
way construction industries. It will further help in conservation of the landfill space and will be
a step forward towards achieving sustainability in road construction practices.

Disclosure statement
No potential conflict of interest was reported by the authors.
24 M. L. Pattanaik et al.

Funding
The authors gratefully acknowledge the financial support by Department of Science and Technology (DST),
Govt. of India under the [grant number DST/TSG/WM/2015/525].

ORCID
Madhu Lisha Pattanaik http://orcid.org/0000-0003-0835-9694
Rajan Choudhary http://orcid.org/0000-0003-0901-5076
Bimlesh Kumar http://orcid.org/0000-0001-6001-8411
Abhinay Kumar http://orcid.org/0000-0002-7740-9029

References
AASHTO T324. (2017). Standard method of test for Hamburg wheel-track testing of compacted hot mix
asphalt (HMA). Washington, DC: American Association of State Highway and Transportation Officials.
Ahmedzade, P., & Sengoz, B. (2009). Evaluation of steel slag coarse aggregate in hot mix asphalt concrete.
Journal of Hazardous Materials, 165(1–3), 300–305.
Alvarez, A. E., Martin, A. E., Estakhri, C. K., Button, J. W., Glover, C. J., & Jung, S. H. (2006). Synthe-
sis of current practice on the design, construction, and maintenance of porous friction courses (No.
FHWA/TX-06/0-5262-1).
Arabani, M., Mirabdolazimi, S. M., & Sasani, A. R. (2010). The effect of waste tire thread mesh on the
dynamic behaviour of asphalt mixtures. Construction and Building Materials, 24(6), 1060–1068.
Asi, I. M., Qasrawi, H. Y., & Shalabi, F. I. (2007). Use of steel slag aggregate in asphalt concrete mixes.
Canadian Journal of Civil Engineering, 34(8), 902–911.
ASTM D4792. (2013). Standard test method for potential expansion of aggregates from hydration
reactions. West Conshohocken, PA: American Society for Testing and Materials.
ASTM D7064. (2008). Standard practice for open graded friction[Q4] course (OGFC) mix design. West
Conshohocken, PA: American Society for Testing and Materials.
Behnood, A., & Ameri, M. (2012). Experimental investigation of stone matrix asphalt mixtures containing
steel slag. Scientia Iranica, 19(5), 1214–1219.
BS 598-111. (1995). Method for determining resistance to permanent deformation of bituminous mixtures
subject to unconfined uniaxial loading. Brussels: European Committee for Standardization (CEN).
BS DD 226. (1996). Method for determining resistance to permanent deformation of bituminous mixtures
subject to unconfined dynamic loading. Brussels: European Committee for Standardization (CEN).
Bureau of Indian Standards. (2012). Indian standard specifications for drinking water IS: 10500.
Carvalho, R. L. (2012). Prediction of permanent deformation in asphalt concrete (Doctoral dissertation).
University of Maryland.
Chen, Z., Wu, S., Wen, J., Zhao, M., Yi, M., & Wan, J. (2015). Utilization of gneiss coarse aggregate and
steel slag fine aggregate in asphalt mixture. Construction and Building Materials, 93, 911–918.
EN 12697-24. (2012). Bituminous mixture-test methods for hot mix asphalt-part 24: Resistance to fatigue.
Brussels: European committee for standardization (CEN).
EN 12697-26. (2012). Bituminous mixture-test methods for hot mix asphalt-part 26: Stiffness. Brussels:
European committee for standardization (CEN).
FICCI. (2018). Using steel slag in infrastructure development. Retrieved from http://blog.ficci.com/steel-
slag/5291
Hainin, M. R., Rusbintardjo, G., Hameed, M. A. S., Hassan, N. A., & Yusoff, N. I. M. (2014). Utilisation
of steel slag as an aggregate replacement in porous asphalt mixtures. Jurnal Teknologi (Sciences and
Engineering), 69(1), 67–73.
IS 10500. (2012). Drinking water – specification. New Delhi: Bureau of Indian Standards.
JSW Steel. (2018). Innovation of the year. Retrieved from www.worldsteel.org/en/dam/jcr:6646634d-3d65
-4051-9736-13d54dff4cdd/Steelie + 2016_vfinal.pdf
Kandhal, P. S. (2002). Design, construction, and maintenance of open-graded asphalt friction courses,
information series 115. Lanham, MD: National Asphalt Pavement Association.
Kavussi, A., & Qazizadeh, M. J. (2014). Fatigue characterization of asphalt mixes containing electric arc
furnace (EAF) steel slag subjected to long term aging. Construction and Building Materials, 72, 158–
166.
 Road Materials and Pavement Design 25

Konishi, S., & Kitagawa, G. (2008). Information criteria and statistical modeling, springer series in
statistics. New York, NY: Springer Publications.
Kumar, A., Choudhary, R., & Nirmal, S. K. (2018). Salient attributes of open graded friction courses
towards enhanced road safety in hilly and high rainfall regions of India. Journal of the Indian Roads
Congress, 79(2), 11–19.
Mallick, R., Kandhal, P., Cooley, L. A., & Watson, D. (2000). Design construction and performance of
new-generation open-graded friction courses. Asphalt Paving Technology, 69, 391–423.
MoRTH. (2018). Annual report 2017–18. Ministry of road transport and highways. New Delhi: Govern-
ment of India.
Onuaguluchi, O., & Eren, Ö. (2012). Copper tailings as a potential additive in concrete: Consistency,
strength and toxic metal immobilization properties. Indian Journal of Engineering & Materials
Sciences, 19, 79–86.
Ott, R. L., & Longnecker, M. (2015). An introduction to statistical methods and data analysis (7th ed.).
Boston: Cengage Learning.
Pasetto, M., & Baldo, N. (2010). Experimental evaluation of high performance base course and road base
asphalt concrete with electric arc furnace steel slags. Journal of Hazardous Materials, 181(1–3), 938–
948.
Pattanaik, M. L., Choudhary, R., & Kumar, B. (2017). Evaluation of frictional pavement resistance as a
function of aggregate physical properties. Journal of Transportation Engineering, Part B: Pavements,
143(2), 220–233.
Pattanaik, M. L., Choudhary, R., & Kumar, B. (2018a). Clogging evaluation of open graded friction course
mixes with EAF steel slag and modified binders. Construction and Building Materials, 159, 220–233.
Pattanaik, M. L., Choudhary, R., & Kumar, B. (2018b). Laboratory evaluation of mix design parameters of
open-graded friction course mixes with electric arc furnace steel slag. Advances in Civil Engineering
Materials, 7(1), 616–632.
USEPA. (2008). Test methods for evaluating solid waste, physical/chemical methods. Washington, DC:
Office of Solid Waste and Emergency Response, USEPA.
Valkering, C. P., Lancon, D. J. L., Dehilster, E., & Stoker, D. A. (1990). Rutting resistance of asphalt mixes
containing non-conventional and polymer-modified binders (with discussion and closure). Journal of
the Association of Asphalt Paving Technologists, 59, 590–609.
Wang, Y., & Wang, G. (2011). Improvement of porous pavement. In Final report to US green building coun-
cil. Greenville, NC: East Carolina University. Retrieved from http://www.usgbc.org/Docs/Archive/
General/Docs9915.pdf
Wang, M., Wright, J., Buswell, R., & Brownlee, A. (2013). A comparison of approaches to stepwise
regression for global sensitivity analysis used with evolutionary optimization. Proceedings of the 13th
conference of international building performance simulation association, Chambery, France, 26–28
August 2013.
Watson, D. E., Ann Moore, K., Williams, K., & Allen CooleyJrL. (2003). Refinement of new-generation
open-graded friction course mix design. Transportation Research Record, 1832(1), 78–85.
Wen, H., Wu, S., & Bhusal, S. (2015). Performance evaluation of asphalt mixes containing steel slag
aggregate as a measure to resist studded tire wear. Journal of Materials in Civil Engineering, 28(5),
04015191-7.
World Steel. (2018). World steel in figures. Retrieved from https://www.worldsteel.org/en/dam/jcr:f9359dff
-9546-4d6b-bed096201185b12/World%2520Steel%2520in%2520Figures%25202018.pdf
Wu, S., Xue, Y., Ye, Q., & Chen, Y. (2007). Utilization of steel slag as aggregates for stone mastic asphalt
(SMA) mixtures. Building and Environment, 42(7), 2580–2585.
Xie, J., Wu, S., Lin, J., Cai, J., Chen, Z., & Wei, W. (2012). Recycling of basic oxygen furnace slag in
asphalt mixture: Material characterization & moisture damage investigation. Construction and Building
Materials, 36, 467–474.
Ziari, H., Nowbakht, S., Rezaei, S., & Mahboob, A. (2015). Laboratory investigation of fatigue character-
istics of asphalt mixtures with steel slag aggregates. Advances in Materials Science and Engineering,
2015, 1–5.