Steel Aggregate Swelling Potential in

Layers of Road Pavements

Victor H. S. Oliveira, C.Eng. 1; Nairo D. T. Buitrago, C.Eng. 2; Luís F. M. Ribeiro, Ph.D. 3; and
André L. B. Cavalcante, Ph.D. 4

Abstract: Steel slag, one of the by-products from the production of steel, is crushed, sized properly, and treated in relation to its volume
Downloaded from ascelibrary.org by East Carolina University on 07/16/18. Copyright ASCE. For personal use only; all rights reserved.

swelling potential to be used as a highly resistant aggregate called steel aggregate or steel gravel. This study analyzes the swelling mech-
anisms of steel aggregate. By laboratory testing, the volumetric swelling potential in samples of steel aggregate and samples of lateritic soil
and steel aggregate mixtures was determined in two different proportions: 90% soil þ10% steel aggregate (M9010); and 80% soil þ20% steel
aggregate, by weight (M8020). This analysis is part of a project that assesses the feasibility of the utilization of this coproduct in the con-
struction industry in Brazil. Average swelling values of 0.92, 0.73, and 0.61% were obtained from the steel aggregate samples and from the
M9010 and M8020 mixture samples, respectively. According to Brazilian and Japanese standard volumetric swelling limits, all tested ma-
terials are suitable for the use of paving as an alternative to the use of natural aggregates. DOI: 10.1061/(ASCE)EE.1943-7870.0001435.
© 2018 American Society of Civil Engineers.

Introduction
In the United States, approximately 21 million tons of steel industry
slag are produced each year. Also, there are many productive com-
mercial uses of slag that have been developed, such as road bed and
fill material (Proctor et al. 2000; USGS 1997). In Germany and the
other European countries, the steel industry is concentrated on in-
creasing the utilization rate of industrial coproducts and recycling
materials (Motz and Geiseler 2001; Krass et al. 1999). Slag repre-
sents 29% of the total coproducts and wastes generated by the steel
industry in Brazil (IABR 2014). It is a material that presents fun-
damental functions in the steel production process to ensure good
quality of the final product. In the recent past, slag was considered a
waste of little or no economic value. That point of view has
changed, and slag is already considered a by-product capable of
conferring important added value to the materials in which it is ap-
plied (Graffitti 2002).
Many studies have been performed using steel slag in pave-
ments (Timothy 1992; Xue et al. 2006; Wu et al. 2007; Maharaj
et al. 2017). In Brazil, steel slag has been used as small and large
aggregate in road paving, being particularly suitable for its hard-
ness, durability, free drainage, and high resistance to crushing.
1
Researcher, Dept. of Civil and Environmental Engineering, Univ. of
Brasilia, CEP 70910-900 Brasília, Brazil (corresponding author). Email:
victorh.souzaoliveira@gmail.com
2 swelling of the steel slag is the allotropic transformation of the di-
Adjunct Professor, Dept. of Civil and Environmental Engineering,
Universidad de la Costa, Calle 58, #55-66, 080002 Barranquilla, Atlántico,
Colombia; formerly, Researcher, Dept. of Civil and Environmental
Engineering, Univ. of Brasilia, CEP 70910-900 Brasília, Brazil. Email:
ntarazon2@cuc.edu.co
3
Associate Professor, Dept. of Civil and Environmental Engineering,
Univ. of Brasilia, CEP 70910-900 Brasília, Brazil. Email: lmartins@
unb.br
4
Associate Professor, Dept. of Civil and Environmental Engineering,
Univ. of Brasilia, CEP 70910-900 Brasília, Brazil. Email: abrasil@unb.br
Note. This manuscript was submitted on December 9, 2016; approved
on March 30, 2018; published online on June 25, 2018. Discussion period
open until November 25, 2018; separate discussions must be submitted
for individual papers. This paper is part of the Journal of Environmental
Engineering, © ASCE, ISSN 0733-9372.
All of those characteristics allow its use for extended periods with
minimal maintenance (Rohde 2002).
One of the most significant difficulties, however, in the use of
steel slag as aggregate in layers of pavements is controlling its ex-
pansive potential, which can reach 10.0% (Silva 2001; Castelo
Branco 2004). That swelling generates internal stresses that cause
cracks in the structural layers of the roads and decrease their
strength and durability. Also, steel slag is rich in calcium oxide
(CaO) content, whose uncontrolled seepage has been reported to
impair the quality of receiving surface waters and aquifers
(Koryak et al. 2002; Roadcap et al. 2005; Mayes et al. 2008),
but there are methods to mitigate this problem (Özkök et al.
2015), and there is a concern regarding the leaching of heavy met-
als present in steel slag (arsenic, barium, cadmium, chromium, and
lead) into the groundwater, however, their concentration levels may
be tolerable (Yildirim and Prezzi 2015).
Expansive reactions involve specific compounds present in steel
slag. Among those volumetrically unstable compounds, CaO and
magnesium oxide (MgO) are the main compounds responsible for
the weakening due to the molar volume difference in their reactions
(Machado 2000). Steel slag contains a significant amount of steel or
iron particles that get incorporated into the slag when oxygen is
introduced. Those particles can contribute to the volumetric insta-
bility of the slag through the generation of corrosion and oxidation
products (Machado 2000). Another reaction that determines the
calcium silicate or calcium orthosilicate (2CaO · SiO2 or Ca2 SiO4 )
which favor the worsening of the effects of the expansibility caused
by CaO and MgO. Despite the necessity of adopting curing tech-
niques for the use of this material, the possibility of the execution of
thinner structural layers for the same load resistance and durability
contributes to the economic advantages of using slag as aggregates
in the construction of pavements. Table 1 shows the chemical char-
acteristics of steel slag used in this research obtained by X-ray dif-
fraction (XRD) analyses.
Crushing, granulometric stabilization, and reduction of the
swelling potential of the steel slag result in the stabilized and stand-
ardized coproduct known as steel gravel or steel aggregate. In this
context, this study aims to evaluate the feasibility of using this

© ASCE 04018077-1 J. Environ. Eng.

J. Environ. Eng., 2018, 144(9): 04018077

 Table 1. Chemical characteristics of steel slag by XRD analyses Table 2. Textural classification of the soil
Chemical compound Formula Percent by weight
Calcite CaCO3 With chemical No chemical
Portlandite CaðOHÞ2 ABNT classification (mm) breakdown breakdown
Gypsum CaSO4 · 2H2 O
20.0 < large gravel < 60.0 0.0 0.0
Gehlenite Ca2 Al2 SiO7
6.0 < medium gravel < 20.0 2.0 2.0
Srebrodolskite Ca2 Fe2 O5
2.0 < fine gravel < 6.0 2.4 2.4
Hematite Fe2 O3
0.6 < large sand < 2.0 2.1 0.6
Siderite FeCO3
0.2 < medium sand < 0.6 3.9 3.4
Wustite FeO
0.06 < fine sand < 0.2 8.9 0.5
Quartz SiO2
0.002 < silt < 0.06 36.5 84.0
Spinel MgðAl · FeÞ2 O4
Clay < 0.002 44.2 7.1
Strengite ðMg; Fe; AlÞPO4 · 2H2 O
Todorokite NaMn6 O12 · 3H2 O
Downloaded from ascelibrary.org by East Carolina University on 07/16/18. Copyright ASCE. For personal use only; all rights reserved.

Rutile TiO2
Sulfuric (Sulfur) S
Table 3. Physical properties of the soil
Property Value
wL (%) 68.50
coproduct based on the technical specifications related to the pro- wP (%) 47.16
cess of aggregate swelling in layers of road pavements. Thus, the Plasticity index (%) 21.34
objective of this paper is to evaluate the different applicability of Unified soil classification system (USCS) Inorganic Silts
steel reinforcement in pavements replacing the natural aggregate by γ s (kN=m3 ) 29.51
means of the comparative analysis between the volumetric swelling wot (%) 31.0 (SCE)
of the steel gravel and of the mixtures composed of steel gravel and 26.5 (MCE)
soil in different proportions. γ dmax (kN=m3 ) 14.00 (SCE)
15.50 (MCE)
Note: MCE = modified compaction energy; and SCE = standard
Materials and Methods compaction energy.

In order to experimentally evaluate the swelling potential of the

steel gravel and its mixture with lateritic soil, the NBR 16364 stan-
dard (ABNT 2015) and the ME 113 methodology (DNIT 2009),
which is based on the PTM 130 standard developed by the Penn-
sylvania Department of Transportation (PennDoT 1978), were
used. Also, ASTM D2940-98 (ASTM 1998) and JIS A 5015
(JSA 2013) were used in the discussion of the results. The steel
slag was crushed and stabilized in relation to granulometry. The
reduction of the swelling potential of the material occurred by open
air storage.
The soil utilized in the research had a plasticity index of 21.34%,
belonging to the MH-OH group, composed of high-plasticity
clay-silt soils according to the Unified Soil Classification System.
According to the AASHTO Soil Classification System, the soil is
classified as A-7: clays and silts (more than 75% of the material
passes in the 0.074-mm mesh) with high liquidity index and high
volume changes in dry and wet conditions (Fig. 1 and Tables 2
and 3).
100.0
Percent Finer by Weight (%)
In order to evaluate the volumetric swelling potential of steel
gravel and soil mixtures when used in road pavement structures,
the samples were prepared with the following proportions: 80%
steel gravel þ20% soil, by weight (referred to as M8020); and
90% steel gravel þ10% soil, by weight (denominated M9010),
in addition to samples constituted by pure steel gravel. The samples
were prepared according to the methodology prescribed by NBR
6457-86 (ABNT 1986) and following the same criteria established
by ME 172 (DNIT 2016). Also, the used equipment was basically
the same as that recommended by the Brazilian standard for the
determination of the California Bearing Ratio (ABNT 1987). The
optimum compaction moisture content and the maximum density
were determined according to the specifications of ME 164 (DNIT
2013), noting, however, that the materials should be compacted in
three equal layers with 56 blows per layer.
Three samples were molded for each material: one with 2%
moisture content below the optimum moisture content (OMC),
one with 2% moisture content above the OMC, and the third with
the OMC.
The molds were immersed in preheated water in a heated cham-
ber at 38°C for 30 min. Then the measurement of the values ex-

80.0
pressed by the extensometers with the submerged samples at a
temperature of 69  3°C were taken. The measurements were re-
60.0
corded daily for 7 days, 2 h after the addition of water, which
should be sufficient for the sample to remain completely sub-
40.0 merged. After that period, the water was partially removed from
the vessel at the beginning of the second stage of the test, which
20.0 with chemical breakdown consists of maintaining the samples in a saturated, not submerged,
no chemical breakdown
condition for another 7 days. To guarantee the saturation of the
0.0 samples, 500 mL of water were added daily. After the 2 weeks
0.001 0.010 0.100 1.000 10.000 100.000 of swelling evaluation, the samples were removed from the mold
Particle Size (mm) and dismantled. The volumetric swelling is expressed as a percent-
age, given by the reading after 14 days, over the initial height of the
Fig. 1. Particle-size distribution curves.
sample, according to Eq. (1)

© ASCE 04018077-2 J. Environ. Eng.

J. Environ. Eng., 2018, 144(9): 04018077

 1.00
ðLf − Li Þ
%Exp: ¼ × 100 ð1Þ

Cumulative Expansion Values (%)

Hi
0.80
where %Exp. = volumetric swelling percentage of the steel gravel
(%); Lf = reading of the extensometer at the end of the test (mm); 0.60
Li = reading of the extensometer at the beginning of the test (mm);
and H i = initial height of the sample (mm).
SG-01 - 6.7%
0.40

SG-02 - 8.7% (OMC)

Results and Discussion 0.20
SG-03 - 10.7%

Steel Gravel Swelling Potential 0.00

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Downloaded from ascelibrary.org by East Carolina University on 07/16/18. Copyright ASCE. For personal use only; all rights reserved.

Table 4 presents the main values of the volumetric swelling of the Time (Days)
pure steel gravel samples obtained from the results of the test pro-
posed by ME 113 (DNIT 2009). Fig. 2. Cumulative expansion values of the steel gravel over the test
Day 1 represents the reading 24 h after the start of the test in the time.
submerged condition of the samples at a temperature of 69°C.
Day 7 corresponds to half the test time, when the samples were
removed from the submerged condition, and their behavior was
evaluated in the saturated, not submerged, condition. Day 14 rep-
resents the final reading, with the maintenance of the samples in the
saturated, not submerged, condition.
NBR 16364 (ABNT 2015), JIS A 5015 (JSA 2013), and ASTM
2940-98 (ASTM 1998) establish the maximum limit of swelling of
the average of the three samples at 3, 2, and 0.5%, respectively. At
the end of 14 days of the test, the maximum value of 0.96% was
obtained in the compacted sample with moisture content 2% above
the OMC. The OMC sample presented the smallest swelling among
the samples analyzed, with a value of 0.87%. The average obtained
from the three results was 0.92%, considerably below the limits
established by the Brazilian and Japanese standards. Fig. 2 shows
the swelling variation described in Table 4 as a function of the
test time.
The final swelling presented by SG-02 corresponds to the opti-
mal moisture condition, and it was the smallest of the three samples Fig. 3. Steel gravel sample after expansion test.
tested. Its rate of swelling (cumulative swelling value divided by
time) was the highest during the first 24 h of the test, gradually
decreasing until the end of the immersion stage. In the saturation
slope decreased until the end of the submersion stage. The transi-
stage, not the submerged condition, that behavior was continued
tion between stages, submerged to nonsubmerged condition, re-
and a second-degree polynomial could approximate it until the
sulted in the contraction of the sample, represented in the
12th day, when the rate became constant.
inflection observed between Days 7 and 8. From Day 8, the swell-
The sample with moisture content 2% above the OMC (SG-03)
ing rate remained constant until the end of the test time.
presented the highest accumulated swelling among the three analy-
Despite the close swelling values presented by the samples at
ses. Its rate of swelling at the beginning of the trial was the lowest,
the end of the 14 days, there are differences in the behavior of their
and it turned out to be almost constant during both stages of the
curves over the analyzed period. That fact can be explained by the
experiment.
saturation process that occurs inside each sample. Since the SG-03
The material compacted in the sample with moisture content 2%
sample was more saturated, its swelling rate showed the highest
below the OMC (SG-01) had the lowest swelling rate during the
constancy among them. Similarly, the less-saturated specimen
first 24 h of the test. That value increased between the first and
(SG-01) showed higher sensitivity to saturation conditions, with
second days and became constant until the fourth day, when this
more pronounced swelling rate variations.
At the end of the swelling test, the samples were carefully de-
molded and dismantled as indicated in the ME 113 test procedures
Table 4. Cumulative expansion values of the steel gravel over the test time (DNIT 2009). The absence of crystals on the surface of the steel
Sample gravel was observed, as shown in Fig. 3.
SG-01 SG-02 SG-03
Moisture content (%) M8020 Mixture
6.7 8.7 (OMC) 10.7 Table 5 shows the volumetric swelling values of the samples of the
Condition Day Cumulative expansion (%)
M8020 mixture obtained from the results of the test proposed by
ME 113 (DNIT 2009).
Immersed 1 0.03 0.21 0.07 Day 1 represents the reading 24 h after the start of the test in the
7 0.77 0.74 0.52 submerged condition at a temperature of 69°C. Day 7 corresponds
Saturated 14 0.92 0.87 0.96
to half of the test time, at which time the samples were removed

© ASCE 04018077-3 J. Environ. Eng.

J. Environ. Eng., 2018, 144(9): 04018077

 Table 5. Cumulative expansion values over the test time of the M8020
mixture
Sample
M80-01 M80-02 M80-03
Moisture content (%)
9.49 11.49 (OMC) 13.49
Condition Day Cumulative expansion (%)
Immersed 1 0.02 0.07 0.02
7 0.26 0.61 0.22
Saturated 14 0.58 0.84 0.42
Downloaded from ascelibrary.org by East Carolina University on 07/16/18. Copyright ASCE. For personal use only; all rights reserved.

Fig. 5. Sample of the M8020 mixture after the expansion test.

from the submerged condition, and their behavior was evaluated in
the saturated, not submerged, condition. Day 14 represents the final
reading, with the maintenance of the samples in the saturated, not
submerged, condition. The compacted material in sample with moisture content 2%
At the end of the 14 days of the test, the maximum value of below OMC (M80-01) had the lowest swelling rate during the first
0.84% was obtained from the samples compacted with optimum 48 h of the test. The value showed a considerable increase between
moisture content. The average obtained from the three results the fourth and sixth days, when that slope decreased until the end of
was 0.61%, considerably below the limits of 3% established by the saturated not submerged stage. The behavior of the M8020 mix-
NBR 16364 (ABNT 2015) and 2% established by JIS A 5015 ture was similar to the one presented by the steel gravel, with a
(2013). It was also lower than the swelling average presented by tendency characterized by a third-degree polynomial.
the pure steel gravel sample (0.92%). The lower values obtained In the three analyses performed, the curves presented the same
from the mixture are due to the swelling fraction absorbed by trends obtained from the analysis of the steel gravel due to the sat-
the natural soil fines. Those results were corroborated by the studies uration process that occurs inside each sample. Since the M80-03
carried out by da Silva (1991). Fig. 4 shows the cumulative swell- sample was more saturated, its swelling rate showed the highest
ing values of the M8020 mixture over the test time. constancy among the three samples. Similarly, the less-saturated
The swelling presented by M80-02 corresponds to the optimum sample (M80-01) showed higher sensitivity to saturation condi-
moisture content, and it turned out to be the highest of the three tions, with more pronounced variations in swelling rate.
samples tested. Its rate of swelling was the highest during the first At the end of the swelling test, the samples were carefully de-
48 h of the test, gradually decreasing until the end of the immersion molded and dismantled as indicated in the ME 113 test procedures
stage. In the saturation, not submerged, stage, that behavior was (DNIT 2009). The absence of any crystalline formation on the sur-
maintained, and a second-degree polynomial could describe its ten- face of the grains of the mixture was observed, as shown in Fig. 5.
dency. The compacted sample with moisture content 2% greater
than OMC (M80-03) presented the smallest accumulated swelling M9010 Mixture
among the three analyses. The transition between submerged and
not submerged saturated conditions resulted in the contraction of Table 6 presents the main volumetric swelling values of the
the sample, represented in the inflection observed between Days M9010 mixture samples obtained from the results of the test pro-
7 and 8. Similar to the behavior of the cumulative swelling curve posed by ME 113 (DNIT 2009).
of the steel gravel, the graph presented a linear trend in sample with At the end of the test, an average swelling value of 0.73% was
moisture content 2% greater than OMC. obtained, with a maximum value of 0.78% occurring in the com-
pacted sample in the sample with moisture content 2% below
OMC. The lowest swelling was observed in the compacted sample
1.00 with optimum moisture content, with a value of 0.64%. Consider-
M80-01 - 9.49% ing the 3% limit established by NBR 16364 (ABNT 2015) and 2%
established by JIS A 5015 (JSA 2013), the M9010 mixture is suit-
Cumulative Expansion Values (%)

0.80 able for the use in pavement structures from the point of view of
M80-02 - 11.49% (OMC)
volumetric stability. Fig. 6 shows the cumulative swelling values
0.60 M80-03 - 13.49% described in Table 6 as a function of the test time. As expected

0.40 Table 6. Cumulative expansion values of the M9010 mixture

Sample
0.20 M90-01 M90-02 M90-03
Moisture content (%)
0.00 7.87 9.87 (OMC) 11.87
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Time (Days)
Condition Day Cumulative expansion (%)
Immersed 1 0.03 0.02 0.03
Fig. 4. Cumulative expansion values of the M8020 mixture over the 7 0.56 0.51 0.41
test time. Saturated 14 0.78 0.64 0.76

© ASCE 04018077-4 J. Environ. Eng.

J. Environ. Eng., 2018, 144(9): 04018077

 1.00
Cumulative Expansion Values (%)
0.80
0.60
0.40
0.20
0.00
0 1 2 3 4 5 6 7 8 9
Time (Days)
Downloaded from ascelibrary.org by East Carolina University on 07/16/18. Copyright ASCE. For personal use only; all rights reserved.
Fig. 6. Cumulative expansion values of the M9010 mixture over the
test time.
and in line with the results obtained from the tests of the steel gravel
samples and the M8020 mixture, the saturation conditions inside
each sample of the M9010 mixture determine the behavior of its
accumulated swelling curves.
Up to the fourth day, the compacted sample with moisture con-
tent 2% below OMC (M90-01) presented the lowest cumulative
swelling rate, which increased abruptly, resulting in the inflection
observed on the fifth day. From this point, the saturation level of the
sample results in a cumulative swelling rate similar to the one pre-
sented by the compacted sample in the optimum moisture condi-
tion, which became practically zero after Day 11. Therefore, the
curve shows a behavior governed by a third-degree polynomial.
The cumulative swelling rate observed in the compacted sample
in the optimum moisture condition (M90-02) was higher until the
fifth day, decreasing gradually until the 10th day, at which point it
became zero. Similar to that observed in the steel gravel and
M8020 mixture analysis, the predominant behavior of the curve
could be approximated by a second-degree polynomial. The sample
compacted with moisture content 2% greater than OMC (M90-03)
showed a constant swelling rate during the entire period of the test,
resulting in a graph with a linear trend, possibly explained by the
saturation level of the sample, which was the highest of the ana-
lyzed samples.
After the samples were demolded and dismantled, the absence
of crystals on the surface of the grains was observed, as shown
in Fig. 7.
Fig. 7. Sample of the M9010 mixture after expansion test.
© ASCE
Conclusion
The levels of saturation of the tested samples are determinant for
the behavior of the swelling rates, expressed by the inclination of
the curve obtained over the 14 days of the test. In the three analyzed
samples, the swelling curves presented third-degree polynomial,
second-degree polynomial, and linear tendencies in the compacted
samples with moisture content 2% lower than OMC, with the op-
M90-01 - 7.97%
timal moisture content, and with moisture content 2% greater than
M90-02 - 9.87% (OMC) OMC, respectively. In all samples, volumetric stabilization was
also observed from Day 11, with low swelling rates.
M90-03 - 11.87%
When comparing the average swelling values of 0.92, 0.73, and
10 11 12 13 14 15 0.61% obtained from the steel gravel tests and the M9010 and
M8020 mixtures, respectively, the decrease of the swelling values
due to the increase of the soil content in the mixtures was observed.
That fact results from the absorption of the swelling due to the in-
crease of the presence of the fines, which are part the granulometry
of the natural soil.
All the samples fit the limit of 3% of the volumetric swelling for
use in structural layers of road pavements established by NBR
16364 (ABNT 2015) and 2% established by JIS A 5015 (JSA
2013), although it was a little bigger than the limit of 0.5% estab-
lished by ASTM D2940-98 (ASTM 1998). The fact denotes effi-
ciency in the steel slag process of volumetric stabilization for its
transformation into the steel gravel by-product. All tested materials
are suitable for the use of paving as an alternative to the use of
natural aggregates from the point of view of volumetric stability.
The determination of the ideal proportion of steel gravel and soil
depends, therefore, on the availability of the components of the
mixture at the construction site of the highway.
Acknowledgments
The authors acknowledge the support of the following agencies: the
Brazilian Council for Scientific and Technological Development
(CNPq) (Grant No. 304721/2017-4), the Coordination for the
Improvement of Higher Level Personnel (CAPES), the Support
Research of the Federal District Foundation (FAP-DF), the
Brazilian Transport Infrastructure Department (DNIT), the Brazil
Steel Institute (BSI), and the University of Brasilia (UnB) for fund-
ing this research.
References
ABNT (Brazilian Association of Technical Standards). 1986. Soil samples:
Preparation for compaction tests and characterization tests. NBR
6457-86. Rio de Janeiro, Brazil: ABNT.
ABNT (Brazilian Association of Technical Standards). 1987.
“Soil—California bearing ratio.” NBR 9895-87. Rio de Janeiro, Brazil:
ABNT.
ABNT (Brazilian Association of Technical Standards). 2015. Execution of
sub-base and base stabilized granulometrically with steel aggregate for
road paving. NBR 16364. Rio de Janeiro, Brazil: ABNT.
ASTM. 1998. Standard specification for graded aggregate material
for bases or subbases for highways or airports. ASTM D2940-98.
West Conshohocken, PA: ASTM.
Castelo Branco, V. T. 2004. “Characterization of asphalt mixtures with
the use of steel slag as aggregate.” Master’s dissertation, Dept. of Civil
Engineering, Federal Univ. of Rio de Janeiro.
da Silva, E. A. 1991. “Use of steel slag in road paving.” In 28th Annual
Paving Meeting. Rio de Janeiro, Brazil: Brazilian Paving Association.
DNIT (National Department of Transport Infrastructure). 2009. Road
paving: Artificial aggregate—Evaluation of the expansion potential
of steel slag. ME 113. Rio de Janeiro, Brazil: DNIT.
04018077-5 J. Environ. Eng.
J. Environ. Eng., 2018, 144(9): 04018077
 DNIT (National Department of Transport Infrastructure). 2016. Soils—
Determination of the California Support Index using unworked
samples—Test method. ME 172. Rio de Janeiro, Brazil: DNIT.
DNIT (National Department of Transport Infrastructure). 2013. Soils—
Compaction using unworked samples. ME 164. Rio de Janeiro, Brazil:
DNIT.
Graffitti, D. F. 2002. “Evaluation of free lime content in electric steel slag.”
Master’s dissertation, Dept. of Civil Engineering, Federal Univ. of Rio
Grande do Sul.
IABR (Brazilian Steel Institution). 2014. Sustainability report 2014. Rio de
Janeiro, Brazil: IABR.
JSA (Japanese Standards Association). 2013. Iron and steel slag for road
construction. JIS A 5015. Tokyo: JSA.
Koryak, M., L. J. Stafford, R. J. Reilly, and M. P. Magnuson. 2002.
“Impacts of steel mill slag leachate on the water quality of a small
Downloaded from ascelibrary.org by East Carolina University on 07/16/18. Copyright ASCE. For personal use only; all rights reserved.
Pennsylvania stream.” J. Freshwater Ecol. 17 (3): 461–465. https://doi
.org/10.1080/02705060.2002.9663921.
Krass, K., C. Kellermann, and M. Rohleder. 1999. “Abfall, Aufbereitung
und Verwertung von Recycling-Baustoffen und industriellen Nebenpro-
dukten im Wirtschaftsjahr 1997-Teil 2: Industrielle Nebenprodukte.”
Strasse und Autobahn 50 (9): 473–478.
Machado, A. T. 2000. “Comparative study of test methods for evaluation
of steel slag expandability.” Master’s dissertation, Dept. of Civil
Engineering, Univ. of São Paulo Polytechnic School.
Maharaj, C., D. White, R. Maharaj, and C. Morin. 2017. “Re-use of steel
slag as an aggregate to asphaltic road pavement surface.” Cogent Eng.
4 (1): 1416889. https://doi.org/10.1080/23311916.2017.1416889.
Mayes, W. M., P. L. Younger, and J. Aumônier. 2008. “Hydrogeochemistry
of alkaline steel slag leachates in the U.K.” Water Air Soil Pollut.
195 (1–4): 35–50. https://doi.org/10.1007/s11270-008-9725-9.
Motz, H., and J. Geiseler. 2001. “Products of steel slags an opportunity to
save natural resources.” Waste Manage. 21 (3): 285–293. https://doi.org
/10.1016/S0956-053X(00)00102-1.
Özkök, E., A. P. Davis, and A. H. Aydilek. 2015. “Treatment methods for
mitigation of high alkalinity in leachates of aged steel slag.” J. Environ.
© ASCE 04018077-6
J. Environ. Eng., 2018, 144(9): 04018077
Eng. 142 (2): 04015063. https://doi.org/10.1061/(ASCE)EE.1943-7870
.0001024.
PennDOT (Pennsylvania Department of Transportation). 1978. Test
method for evaluation of the steel slag expansion potential. PTM
130. Harrisburg, PA: PennDOT.
Proctor, D. M., et al. 2000. “Physical and chemical characteristics of blast
furnace, basic oxygen furnace, and electric arc furnace steel industry
slags.” Environ. Sci. Technol. 34 (8): 1576–1582. https://doi.org/10
.1021/es9906002.
Roadcap, G. S., W. R. Kelly, and C. M. Bethke. 2005. “Geochemistry of
extremely alkaline (pH > 12) groundwater in slag-fill aquifers.”
Ground Water 43 (6): 806–816. https://doi.org/10.1111/j.1745-6584
.2005.00060.x.
Rohde, L. 2002. “Electric steel slag in granular layers of pavements:
Laboratory study.” Master’s dissertation, Dept. of Civil Engineering,
Federal Univ. of Rio Grande do Sul.
Silva, P. 2001. “Concrete reinforcement study of portland cement at the
experimental circular road of the institute of highway research.”
Ph.D. thesis, Dept. of Civil Engineering, Federal Univ. of Rio de
Janeiro.
Timothy, R. 1992. Steel slag aggregates in Bituminous mixtures. Final
Rep. Springfield, VA: Bureau of Bridge and Roadway Technology,
Pennsylvania Dept. of Transportation.
USGS. 1997. Mineral industry surveys: Slag-iron and steel, annual
review—1997. Reston, VA: USGS.
Wu, S., Y. Xue, Q. Ye, and Y. Chen. 2007. “Utilization of steel slag as
aggregates for stone mastic asphalt (SMA) mixtures.” Build Environ.
42 (7): 2580–2585. https://doi.org/10.1016/j.buildenv.2006.06.008.
Xue, Y., S. Wu, H. Hou, and J. Zha. 2006. “Experimental investigation
of basic oxygen furnace slag used as aggregate in asphalt mixture.”
J. Hazard. Mater. 138 (2): 261–268. https://doi.org/10.1016/j
.jhazmat.2006.02.073.
Yildirim, I. Z., and M. Prezzi. 2015. “Geotechnical properties of fresh and
aged basic oxygen furnace steel slag.” J. Mater. Civ. Eng. 27 (12):
04015046. https://doi.org/10.1061/(ASCE)MT.1943-5533.0001310.
J. Environ. Eng.