Technical Report

Mechanical and cementitious characteristics of ground granulated blast

furnace slag and basic oxygen furnace slag blended mortar
Chia-Jung Tsai
a
, Ran Huang
b
, Wei-Ting Lin
c,d,
, His-Ning Wang
a
a
Institute of Materials Engineering, National Taiwan Ocean University, Keelung 20224, Taiwan
b
Dept. of Harbor and River Engineering, National Taiwan Ocean University, Keelung 20224, Taiwan
c
Dept. of Civil Engineering, National Ilan University, Ilan 26047, Taiwan
d
Institute of Nuclear Energy Research, Atomic Energy Council, Taoyuan 32546, Taiwan
a r t i c l e i n f o
Article history:
Received 29 January 2014
Accepted 1 April 2014
Available online 12 April 2014
a b s t r a c t
Reusing waste materials and reducing carbon emissions are crucial environmental concerns. Ground
granulated basic oxygen furnace slag (GGBOS) and ground granulated blast furnace slag (GGBS) are
the by-products of the steel industry and has positive effects on the environment because it reduces
the problems associated waste disposal. This study reused these two products to completely replace
cementitious materials, thus contributing to waste recycling, reducing the production demand for
cement, and mitigating carbon emissions. Twelve mixture proportions were designed in this study,
including an ordinary Portland mortar (OPM) as the control group and 11 steel/iron slag blended mortar
(SISBM) experimental groups for the mechanical and cementitious characteristic experiments. The
optimal mixing ratio for SISBM compressive strength ranged from GGBOS (steel slag): GGBS (iron
slag) = 3:7 to 5:5 (by weight). At the age of 91 days, the compressive strength of SISBM reached 80
90% compared with that of the control group. Based on the pH values, free-CaO, and microanalysis results,
the cementitious characteristics were mainly generated because the GGBOS increased the free-CaO or
Ca(OH)
2
concentrations in the SISBM curing water and provided alkaline environments for Ca(OH)
2
to
engage in the pozzolanic reaction with the SiO
2
and Al
2
O
3
in GGBS, forming crystals such as calcium
aluminum silicate hydrate, (CASH), calcium silicate hydrate (CSH), and calciummagnesiumalu-
minasilicate (CMAS), which generated strength and strengthened microstructure.
2014 Elsevier Ltd. All rights reserved.
1. Introduction
The concrete and steel industries are crucial industries of
human ecology. However, when producing the primary cementi-
tious material (binder) for concrete (i.e., Portland cement), large
amounts of hazardous substances such as carbon dioxide (CO
2
)
and dust are emitted. CO
2
is one of the main causes of the green-
house effect. Statistical data indicated that on average, 0.71.0 tons
of CO
2
is generated when 1 ton of Portland cement is produced [1].
The report generated by the World Business Council in 2009 for
Sustainable Development (WBCSD) stated that the global con-
sumption of concrete has exceeded 25 billion tons [2,3], resulting
in CO
2
emissions amounting to 57% of the global CO
2
emissions
[1,4,5]. Hence, if an appropriate substitute can be found to replace
Portland cement, CO
2
emissions can be reduced by reducing Port-
land cement usage. Additionally, according to the statistics of the
World Steel Association, the global iron and steel production in
2011 was approximately 1 billion and 1.5 billion tons, respectively
[6]. The production of each ton of steel generates 150200 kg of
waste such as slag [7,8]. If not appropriately treated or reused,
the slag may accumulate and immensely affect the global environ-
ment. The blast furnace process is typically employed in iron pro-
duction. The commonly used methods for steel production include
the converter, electric furnace, open hearth, and crucible processes
[9]. Various types of slag with different attributes are generated
from these processes. Thus, the specic physical or chemical char-
acteristics of slag must be considered to reuse them in appropriate
applications. The types of slag that are currently used as cementi-
tious materials are ground granulated blast furnace slag (GGBS)
and basic oxygen furnace slag (BOFS) [1014], which were there-
fore used as the main study materials in this research.
Blast furnace slag (BFS) is an amorphous glassy material with
primary chemical components such as SiO
2
, Al
2
O
3
, and CaO. GGBS
can react with the alkali product of hydrated Portland cement
Ca(OH)
2
in alkali activations or pozzolanic reactions. Therefore,
GGBS can partially replace Portland cement as a binder. Research
http://dx.doi.org/10.1016/j.matdes.2014.04.002
0261-3069/ 2014 Elsevier Ltd. All rights reserved.

Corresponding author at: Dept. of Civil Engineering, National Ilan University,

Ilan 26047, Taiwan. Tel.: +886 2 24622192x6420x7557; fax: +886 2 24632375.
E-mail address: wtlin@niu.edu.tw (W.-T. Lin).
Materials and Design 60 (2014) 267273
Contents lists available at ScienceDirect
Materials and Design
j our nal homepage: www. el sevi er . com/ l ocat e/ mat des
reports from numerous scholars have proven that GGBS can
replace approximately 70 wt.% or even over 80 wt.% of Portland
cement [12,14]. The report also indicated that replacing cement
with GGBS can reduce cement production as well as mitigate car-
bon emissions [15]. Regarding the lifecycle of Portland cement, the
production process not only necessitates the replacement of raw
material extraction, but also involves calcination and grinding,
which generate approximately 7.6 tons of CO
2
for each ton of Port-
land cement produced. Comparatively, furnace slag requires only
grinding. Hence, for every 10% of furnace slag added in Portland
cement, 910% of CO
2
emission can be reduced from concrete
manufacturing [15].
BOFS is typically used as the coarse aggregate and cement clin-
ker in concrete and comprised various compounds (e.g., FeO, C
4
AF,
C
3
A, C
3
S, C
2
S, and RO (a CaOFeOMnOMgO solid solution) in the
crystal phase) and traces of free-CaO (f-CaO) and free-MgO (f-MgO)
[1618], which easily and drastically react with water to form
Ca(OH)
2
and Mg(OH)
2
[17,19], respectively. When used as concrete
coarse aggregate, the volume expansion can easily occur in such
mixtures [10,20]. Therefore, GGBS is occasionally employed as a
stabilizer to enhance the stabilization of concrete volume mainly
because the BOFS-precipitated Ca(OH)
2
reacts with GGBS in pozzo-
lanic reactions [8], which resolves the concrete volume expansion
problem caused by f-CaO. Accordingly, if furnace slag is grounded
Fig. 1. System outlines of GGBS and GGBOS manufacturing.
Fig. 2. Binder appearance of (a) BOFS, (b) GGBOS, (c) GGBS.
Table 2
Mixture proportions of OPM and SISBM (kg).
Mix no. Water Cement GGBOS GGBS Aggregate
OPM 256 512 N/A N/A 1410
Iron (I) 256 N/A N/A 512 1410
Steel (S) 256 N/A 512 N/A 1410
S1I9 256 N/A 51 461 1410
S2I8 256 N/A 102 410 1410
S3I7 256 N/A 154 358 1410
S4I6 256 N/A 205 307 1410
S5I5 256 N/A 256 256 1410
S6I4 256 N/A 307 205 1410
S7I3 256 N/A 358 154 1410
S8I2 256 N/A 410 102 1410
S9I1 256 N/A 461 51 1410
Table 1
Chemical composition and physical analysis of cement and slag.
Chemical composition (wt.%) GGBS GGBOS Cement
SiO
2
33.42 12.2 21.04
Al
2
O
3
13.35 4.76 5.46
Fe
2
O
3
0.21 30.2 2.98
CaO 41.16 40.4 63.56
MgO 7.76 7.26 2.52
SO
3
N/A 0.18 N/A
MnO N/A 2.39 N/A
Others 4.10 3.11 4.44
Free-CaO 0.1 2.8 0.72
Physical characteristics
Specic gravity (g/cm
3
) 2.89 3.59 3.15
Specic surface (cm
2
/g) 5892 12,315 3713
268 C.-J. Tsai et al. / Materials and Design 60 (2014) 267273
into granulars (ground granulated basic oxygen furnace slag;
GGBOS), it can be mixed with GGBS and replace the remaining
20% or 30% of the Portland cement as cementitious material.
Consequently, an increased amount of wasted furnace slag can
be processed while further reducing cement usage.
In this study, GGBOS and GGBS were mixed in various propor-
tions, completely replacing Portland cement as the binder. In addi-
tion, compressive strength experiments were conducted to
determine whether the mixture of these two materials had
hydraulicity. Subsequently, the optimal mixing ratio was explored
based on the various mixture proportions. Meanwhile, the changes
in pH values and f-CaO in the curing water of the various propor-
tions were analyzed to examine the hydration mechanism, which
was veried by using X-ray diffraction (XRD) and scanning electron
microscopy (SEM). This study is aimed to completely replace
Portland cement with a combination of GGBOS and GGBS and
improved the environment by using the proposed methods in
various concrete production projects.
2. Experimental procedure
2.1. Materials and mixture proportions
The waterbinder ratio in this study was 0.5. Three types of
materials were used as binders: (1) BOFS, the by-product gener-
ated in the steel-making process, was grounded and granulated
to form GGBOS (represented as S). The specic gravity of GGBOS
was 3.59 and the specic surface area was 12,315 cm
2
/g. The
GGBOS production ow is shown in Fig. 1 and its appearance is
shown in Fig. 2a and b. (2) GGBS (represented as I), the by-product
generated fromthe blast furnace process of iron, was grounded and
granulated. The specic gravity of GGBS was approximately 2.89
and the specic surface area was 5892 cm
2
/g. The GGBS production
Table 3
Compressive strength test results.
Age (days) mix no. (MPa) 7 14 28 56 91
OPM 14.30 34.24 48.12 54.27 56.87
Iron (I) N/A N/A N/A N/A N/A
Steel (S) N/A N/A 3.42 3.77 4.26
S1I9 N/A N/A N/A N/A N/A
S2I8 N/A 6.71 11.64 15.23 20.84
S3I7 7.72 18.12 24.64 45.45 47.54
S4I6 9.27 20.61 26.92 48.21 51.63
S5I5 6.69 16.47 21.28 44.48 45.71
S6I4 5.71 11.57 17.16 21.64 22.29
S7I3 N/A N/A 3.83 5.15 7.39
S8I2 N/A N/A 3.71 4.89 5.23
S9I1 N/A N/A 3.67 3.89 4.55
Fig. 3. Compressive strength of SISBM.
Fig. 4. Relationship between the compressive strength and precuring duration.
Table 4
Free CaO or Ca(OH)
2
in curing water (%).
GGBOS (S) GGBS (I) S1I9 S3I7 S5I5 S7I3
7 0.0458 0.0283 0.0292 0.0300 0.0308 0.0383
14 0.0525 0.0292 0.0308 0.0367 0.0383 0.0417
28 0.0583 0.0308 0.0325 0.0300 0.0375 0.0425
91 0.0592 0.0317 0.0333 0.0108 0.0167 0.0333
Fig. 5. Curve of the different mixing ratios and f-CaO or Ca(OH)
2
.
Fig. 6. Curves of the f-CaO/Ca(OH)
2
and precuring duration.
C.-J. Tsai et al. / Materials and Design 60 (2014) 267273 269
owis shown in Fig. 1 and its appearance is presented in Fig. 2c. (3)
Portland Type I cement, which has a specic gravity of approxi-
mately 3.15 and specic surface area of 3713 cm
2
/g. The composi-
tion and physical attributes of the test binders are shown in
Table 1. Natural river sand with a neness modulus of 2.30 and
specic gravity of 2.55 was adopted as the aggregate. The main
variables were the changes in the mixture proportion of S and I,
respectively. These specimens were divided into 11 groups accord-
ing to the weight percentages of the materials: 10:0 (S), 1:9 (S1I9),
2:8 (S2I8),. . ., and 0:10 (I). These 11 groups were collectively
referred to as the steel/iron slag blended mortars (SISBMs). The
control group was ordinary Portland cement mortar (OPM). The
proportions of mortars were based on the requirements of
the ASTM: C109-13 standard (binder/aggregate = 1/2.75), as shown
in Table 2.
2.2. Experiments
2.2.1. Compressive strength test
The compressive strength tests of the specimens were
conducted according to ASTM: C109 standard (Using 50 mm cube
specimens). For each mixture, 50 50 50 mm cubes were
prepared and three specimens of each mixture were tested at the
age of 7, 14, 28, 56, and 91 days to determine the average compres-
sive strength.
2.2.2. Curing water of pH and free-CaO analysis
The specimens for pH and f-CaO analysis were prepared follow-
ing the guidelines in ASTM: C114-13 Standard. The specimens
were conserved in water for 7, 14, 28, and 91 days and the pH value
was recorded. Simultaneously, 60 mL of the curing water was
titrated. The weight percentages of f-CaO or f-Ca(OH)
2
were calcu-
lated by using Eq. (1). The main specimens for this test, samples I,
S1I9, S3I7, S5I5, S7I3, S9I1, and S were prepared in / 2 1 mm
cylinders.
f-CaO % EV 100=W 1
This method does not distinguish between f-CaO and f-CaOH
2

where E = CaO is the equivalent of perchloric acid, in g/mL; V is the

amount of perchloric acid solution required by a sample, in mL; and
W is the weight of the sample, in g.
2.2.3. Microstructure analysis
The phase of the specimens was analyzed by using XRD. The
specimens were in powder form (approximately 1 g) and XRD
was used with a Cu Ka radiation at room temperature, which were
scanned from 2h = 1080. To conduct SEM (Hitachi S-3400) anal-
yses, representative samples with dimensions of 10 mm 10 mm
3 mm were used. The SEM was equipped with energy dispersive
spectroscopy.
3. Results and discussion
3.1. Compressive strength
Table 3 shows the compressive strength results of the mortars
with various proportions, and the results are illustrated in Figs. 3
and 4. The x-axis at the bottom and top of Fig. 3 represent the
amounts of GGBOS and GGBS used, respectively. The y-axis repre-
sents the compressive strength. According to this gure, the mor-
tars composed of 0% and 10% GGBOS yielded no compressive
strength. However, when GGBOS comprised 20% of the mortar,
the compressive strength increased rapidly until the proportion
was 40%, then the compressive strength decreased as the GGBOS
amount increased further. When the proportion of GGOBS was
increased to 80100%, this decreasing trend gradually slowed.
The results indicated that after being hydrated, GGBOS had a
Table 5
pH of curing water.
GGBOS (S) GGBS (I) S1I9 S3I7 S5I5 S7I3
7 12.2 10.4 11.4 11.7 12.0 12.0
14 12.2 10.7 11.6 11.9 12.1 12.1
28 12.4 11.2 11.4 11.9 12.3 12.3
91 12.6 11.2 11.3 10.2 11.0 11.0
Fig. 7. Curve of pH with f-CaO or Ca(OH)
2
.
Fig. 8. X-ray diffraction results of GGBS with hydration.
Fig. 9. X-ray diffraction results of GGBOS with hydration.
270 C.-J. Tsai et al. / Materials and Design 60 (2014) 267273
compressive strength of only 4.26 MPa. Accordingly, GGBS must be
mixed with GGBOS to generate compressive strength. The optimal
mixture proportion for yielding compressive strength ranged from
GGBOS: GGBS (by weight) 3:4 to 6:5.
Fig. 4 shows the relationship between the curing duration and
compressive strength. Based on this gure, ve proportions (i.e.,
S2I8, S3I7, S4I6, S5I5, and S6I3) that had comparatively substantial
compressive strength development were explored and compared
with the control group. The SISBM proportions S3I7, S4I6, and
S5I5 had slower compressive strength development than that of
the control group during the age from 7 to 28 days. A considerable
development increase was observed for these three proportions
starting at the age of 56 days. At 91 days, the compressive strength
of these three proportions increased to 8090% compared with that
of the control group. Regarding proportions S2I8 and S6I4, the
compressive strength also increased with time, but when the
strength increased to approximately 20 MPa (at the age of 28
56 days), the rate of increase decelerated. At 91 days, the compres-
sive strength of these two proportioning groups only increased to
approximately 3739% of that of the control group.
According to relevant literature, GGBS may have cementitious
properties, but it has slow reaction velocities; thus, using this bin-
der does not yield strength [21]. To achieve alkaline activations and
pozzolanic reactions using GGBS, alkaline environments must be
guaranteed. Additionally, sufcient Ca(OH)
2
must be available to
react with the GGBS components of SiO
2
and Al
2
O
3
[14]. BOFS con-
tains substantial amounts of CaO, which drastically reacts with
water, precipitating a considerable amount of f-CaO to form
Ca(OH)
2
[8]. Accordingly, the SISBMs also generated compressive
strengths because the GGBOS component CaO reacted with water,
creating alkaline environments and allowing Ca(OH)
2
to react with
the BOFS components SiO
2
and Al
2
O
3
. Consequently, alkaline acti-
vations and pozzolanic reactions occurred which is correspond
with the previous results of pozzolanic or supplementary cementi-
tious materials [2226].
3.2. The development mechanisms of SISBM strength
According to the experimental results explained in Section 3.1,
the compressive strength of SISBM changed with the various pro-
portions of GGBOS and GGBS and with different precuring dura-
tions. Based on relevant literature, this phenomenon occurred
because the f-CaO in GGBOS form Ca(OH)
2
, which reacts with
SiO
2
and Al
2
O
3
in alkaline activations and pozzolanic reactions as
proposed by the previous study [27]. To verify, six proportions
(i.e., I, S1I9, S3I7, S5I5, S7I3, and S) were used to measure the
f-CaO and pH value changes in various proportions and different
precuring durations.
Table 4 shows the f-CaO or Ca(OH)
2
analysis results for the
aforementioned six proportions. Fig. 5 shows the relationship
between various mixture proportions of mortars and f-CaO or
Ca(OH)
2
concentrations. At the age of 728 days, the f-CaO or
Ca(OH)
2
concentrations of various proportions generally increased
with the amount of GGBOS. Only I and S1I9 (both contained 10% of
GGBS and GGBOS) had comparatively high f-CaO or Ca(OH)
2
con-
centrations at 91 days. However, for S3I7, S515, S713, and S, the
f-CaO or Ca(OH)
2
concentrations increased as the amount of
GGBOS increased at the age of 91 days.
Fig. 6 is a curve presenting the relationship between precuring
durations and f-CaO or Ca(OH)
2
concentrations. The f-CaO or
Ca(OH)
2
precipitation amount of GGBS (I) and GGBOS (S) increased
as the precuring duration increased. However, t the f-CaO or
Ca(OH)
2
precipitation amount of GGBS (S) was 1.87 times that of
GGBS (I). Hence, when GGBOS increases, the precipitation amounts
of Ca(OH)
2
also increases with increasing curing duration. How-
ever, Fig. 6 shows that only the f-CaO or Ca(OH)
2
precipitation
amounts in S1I9 increased with increasing curing duration.
Regarding S3I7, S5I5, and S7I3, the f-CaO or Ca(OH)
2
precipitation
amounts increased only at 728 days, after which a decreasing
trend was recorded.
According to the aforementioned results, in SISBM proportions,
similar to that in GGBS, the mortars with 10% GGBOS precipitated
only trace amounts of f-CaO or Ca(OH)
2
. Thus, this study inferred
that insufcient GGBOS amounts may not provide enough Ca(OH)
2
to effectively simulate the pozzolanic reactions of GGBS. Conse-
quently, no compressive strength was generated in S1I9. The
GGBOS amount must exceed 10% to yield substantial results. How-
ever, excessive amount of GGBOS may result in insufcient GGBS
in later stages of the experiment for reacting with excessive
amount of Ca(OH)
2
, thus increasing f-CaO or Ca(OH)
2
concentra-
tions. Therefore, when the amount of steel slag exceeded 40%,
the compressive strength decreased (Figs. 3 and 5).
Regarding proportions S3I7, S5I5, and S7I3, the f-CaO or
Ca(OH)
2
concentrations increased during the early stages. A possi-
ble reason for this is that f-CaO or Ca(OH)
2
were constantly being
precipitated and only a small portion of Ca(OH)
2
reacted with
GGBS. Therefore, in the early stages, the f-CaO or Ca(OH)
2
concen-
trations increased with time and the compressive strength
increased gradually. Twenty-eight days later, the f-CaO or Ca(OH)
2
concentrations exhibited a decreasing trend, probably because the
f-CaO or Ca(OH)
2
precipitation ratios approximated the maximal
value of the mixed mortars. Hence, the precipitation velocities in
later stages were comparatively slow. Because hydration was an
ongoing process, part of the Ca(OH)
2
was consumed. Thus, the
f-CaO or Ca(OH)
2
concentrations exhibited a decreasing trend,
Table 6
GGBOS and SISBM compounds, and XRD symbols.
Symbols This paper Xie (2012) Wang (2012) This paper
GGBOS GGBOS GGBOS SISBM
1 C
2
S C
2
S C
2
S NA
2 C
3
S C
3
S C
3
S NA
3 Ca
2
(Fe, Al)
3
(SiO
4
)
3
(OH) CaO(Al
2
O
3
)
2
(SiO
2
)
2
Ca
2
Al
2
Si
3
O
12
NA
4 CaCO
3
NA NA NA
5 C
3
A C
3
A C
3
A NA
6 C
4
AF = (C
4
AF or C
2
F) C
4
AF C
4
AF C
4
AF
7 Fe
0.974
O Fe
0.98
O Fe
3
O
4
Fe
0.974
O
8 Ca(OH)
2
Ca(OH)
2
NA Ca(OH)
2
9 CaO NA CaO NA
10 RO RO RO RO
11 NA NA NA CSH
12 NA NA NA CASH
13 NA NA NA CMAS
Remark: C = CaO, S = SiO
2
, A = Al
2
O
3
, H = H
2
O, F = Fe
2
O
3
and M = MgO.
Fig. 10. X-ray diffraction of different mixing ratios of SISBM.
C.-J. Tsai et al. / Materials and Design 60 (2014) 267273 271
and the compressive strengths continued to increase (Figs. 4
and 6).
Table 5 shows the pH values of the mixes at each ages and Fig. 7
shows that the pH values and the f-CaO or Ca(OH)
2
concentrations
in the curing water had a linear relationship. Increased f-CaO or
Ca(OH)
2
concentrations accompanied increased pH values. This
study result showed that adding GGBOS can increase the f-CaO
or Ca(OH)
2
concentrations in SISBM curing water, thus providing
Ca(OH)
2
products and satisfactory alkaline environments required
for the corresponding reactions.
3.3. Microstructure
Relevant studies have indicated that in the pozzolanic reactions,
Ca(OH)
2
reacted with the SiO
2
and Al
2
O
3
in GGBS and formed the
calcium silicate hydrate (CSH) and calcium aluminum silicate
hydrate (CASH) crystals, which are the main materials that
generate strength in hardened cementitious materials [14,28].
CaO (C) reacts with H
2
O (H) to form Ca(OH)
2
(Eq. (2)), which pro-
duces CSH or CASH system colloids by reacting with SiO
2
(S)
and Al
2
O
3
(A) [29]. The corresponding chemical reaction formulas
are shown as Eqs. (3) and (4).
CaO H
2
O !CaOH
2
2
CaOH
2
SiO
2
!CSH 3
C
2
S or C
3
S AH
3
5H !CASH 4
Fig. 8 shows the XRD analysis results of the GGBS before and
after hydration. According to this gure, the unhydrated GGBS
was the noncrystal structure of SiO
2
. This nding corresponded
with the experimental report of Samet and Chaabouni [29]. GGBS
did not substantially change during the hydration process. The
main change was that the noncrystal structure was supplemented
Fig. 11. SEM micrographs of SISBM: (a) GGBS, (b) S1I9, (c) S3I7, (d) S5I5, (e) S7I3, (f) S9I1 and (g) GGBOS.
272 C.-J. Tsai et al. / Materials and Design 60 (2014) 267273
with a CaCO
3
crystalline phase. Fig. 9 presents the XRD results of
GGBOS before and after the hydration. The unhydrated GGBOS
was the crystalline phase of the C
2
S, C
3
S, CaCO
3
, C
3
A, C
4
AF,
Fe
0.974
O, Ca(OH)
2
, CaO, and RO phases (a CaOFeOMnOMgO
solid solution). This nding was similar to the study results of
other scholars [8,9,17,3032] (Table 6). After hydration, the crys-
talline phase of Ca
2
(Fe, Al)
3
(SiO
4
)
3
(OH) increased. Concurrently,
the signal strengths of CaCO
3
and CaO signicantly changed, indi-
cating that GGBOS reacted with water and the primary reactant
was CaO. In addition, the crystal morphology of the hardened
mixed mortars changed considerably. Additional crystals such as
CASH and CSH were produced in the proportion S1I9. How-
ever, when the temperature was between 25 and 35, S1I9
remained exhibiting a glassy noncrystal structure. When the
amount of GGBOS increased, the presence of CASH and CS
H was enhanced, while the noncrystal structures disappeared
and the calciummagnesiumaluminasilicate (CMAS) crystals
were formed. This phenomenon indicated that the main hydration
products of SISMB were CASH, CSH, and CMAS (Fig. 10).
On the other hand, Monshi and Asgarani [11] also proposed the
XRD results of Portland cement from iron and steel slags and lime-
stone which are similar to this study.
By using SEM, the images were magnied 1000 folds (Fig. 11).
The surface morphology of GGBS was primarily granular. Based
on the images, the differences between S1I9 and GGBS were not
substantial (Fig. 11a and b). Numerous hexagonal-net, and nee-
dle-shaped structures can be easily observed on the surfaces of
the S3I7 and S5I5 specimens. Based on the EDS analysis results,
these shapes are the images of CASH and CSH crystals
(Fig. 11c and d). When the amount of GGBOS increased to over
70%, CASH began to decrease on the surfaces of S7I3, S9I1,
and GGBOS. Additionally, a considerable amount of Ca(OH)
2
and
CaCO
3
were observed (Fig. 11e and g). Overall, these results were
consistent with those of XRD. The XRD results and SEM images
showed that the hydration of GGBS and GGBOS mixed mortars
triggered the pozzolanic reactions, thereby producing product
crystals such as CASH, CSH, and CMAS.
4. Conclusions
Reusing waste materials and reducing CO
2
emissions are crucial
measures in developing new materials. In this study, the two
by-products of the steel industry, GGBS and GGBOS, were mixed
to develop a novel cementitious material, yielding these ndings:
1. The optimal SISBM proportions for compressive strength ran-
ged from GGBOS: GGBS (by weight) 3:4 to 6:5. At the age of
91 days, the compressive strength of the experiment groups
increased to 8090% compared with that of the control group.
2. The study results indicated that increasing the amount of
GGBOS can increase the f-CaO or Ca(OH)
2
concentrations in
the SISBM curing water, thus providing Ca(OH)
2
and satisfac-
tory alkaline environments required for the corresponding
reactions.
3. Pozzolanic reactions occurred after hydration of the GGBS and
GGBOS mixed mortars, yielding product crystals such as CA
SH, CSH, and CMAS.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.matdes.2014.04.
002.
References
[1] Huntzinger DN, Eatmon TD. A life-cycle assessment of Portland cement
manufacturing: comparing the traditional process with alternative
technologies. J Cleaner Prod 2009;17:66875.
[2] Klee H. The cement sustainability initiative. In: World business council for
sustainable development (WBCSD). Washington; 2009.
[3] Fairbairn EMR, Americano BB, Cordeiro GC, Paula TP, Toledo Filho RD, Silvoso
MM. Cement replacement by sugar cane bagasse ash: CO
2
emissions reduction
and potential for carbon credits. J Environ Manage 2010;91:186471.
[4] Wang Q, Yan P, Feng J. A discussion on improving hydration activity of steel
slag by altering its mineral compositions. J Hazard Mater 2011;186:10705.
[5] Iacobescu RI, Koumpouri D, Pontikes Y, Saban R, Angelopoulos GN. Valorisation
of electric arc furnace steel slag as raw material for low energy belite cements.
J Hazard Mater 2011;196:28794.
[6] World steel association (worldsteel) [Internet]. In: World steel association.
Belgium; 2012. <https://www.worldsteel.org/>. [cited 2014 March 24].
[7] Reddy AS, Pradhan RK, Chandra S. Utilization of basic oxygen furnace (BOF)
slag in the production of a hydraulic cement binder. Int J Miner Process
2006;79:98105.
[8] Mahieux PY, Aubert JE, Escadeillas G. Utilization of weathered basic oxygen
furnace slag in the production of hydraulic road binders. Constr Build Mater
2009;23:7427.
[9] Bonenfant D, Kharoune L, Sauv S, Hausler R, Niquette P, Mimeault M, et al.
Molecular analysis of carbon dioxide adsorption processes on steel slag oxides.
Int J Greenhouse Gas Control 2009;3:208.
[10] Zhang T, Yu Q, Wei J, Li J, Zhang P. Preparation of high performance blended
cements and reclamation of iron concentrate from basic oxygen furnace steel
slag. Resour Conserv Recycl 2011;56:4855.
[11] Monshi A, Asgarani MK. Producing Portland cement from iron and steel slags
and limestone. Cem Concr Res 1999;29:13737.
[12] Kumar S, Kumar R, Bandopadhyay A, Alex TC, Ravi Kumar B, Das SK, et al.
Mechanical activation of granulated blast furnace slag and its effect on the
properties and structure of Portland slag cement. Cem Concr Compos
2008;30:67985.
[13] Beushausen H, Alexander M, Ballim Y. Early-age properties, strength
development and heat of hydration of concrete containing various South
African slags at different replacement ratios. Constr Build Mater
2012;29:53340.
[14] Osborne GJ. Durability of Portland blast-furnace slag cement concrete. Cem
Concr Compos 1999;21:1121.
[15] McLellan BC, Williams RP, Lay J, van Riessen A, Corder GD. Costs and carbon
emissions for geopolymer pastes in comparison to ordinary portland cement. J
Cleaner Prod 2011;19:108090.
[16] Xuequan W, Hong Z, Xinkai H, Husen L. Study on steel slag and y ash
composite Portland cement. Cem Concr Res 1999;29:11036.
[17] Li J, Yu Q, Wei J, Zhang T. Structural characteristics and hydration kinetics of
modied steel slag. Cem Concr Res 2011;41:3249.
[18] Xie J, Wu S, Lin J, Cai J, Chen Z, Wei W. Recycling of basic oxygen furnace slag in
asphalt mixture: material characterization and moisture damage
investigation. Constr Build Mater 2012;36:46774.
[19] Wang Q, Yan P. Hydration properties of basic oxygen furnace steel slag. Constr
Build Mater 2010;24:113440.
[20] George W. Determination of the expansion force of coarse steel slag aggregate.
Constr Build Mater 2010;24:19616.
[21] Siddique R, Bennacer R. Use of iron and steel industry by-product (GGBS) in
cement paste and mortar. Resour Conserv Recycl 2012;69:2934.
[22] Cheng A, Lin WT, Huang R. Application of rock wool waste in cement-based
composites. Mater Des 2011;32:63642.
[23] Han TY, Lin WT, Cheng A, Huang R, Huang CC. Inuence of polyolen bers on
the engineering properties of cement-based composites containing silica
fume. Mater Des 2012;37:56976.
[24] Lee CL, Huang R, Lin WT, Weng TL. Establishment of the durability indices for
cement-based composite containing supplementary cementitious materials.
Mater Des 2012;37:2839.
[25] Dinakar P, Sethy KP, Sahoo UC. Design of self-compacting concrete with
ground granulated blast furnace slag. Mater Des 2013;43:1619.
[26] Islam A, Alengaram UJ, Jumaat MZ, Bashar II. The development of compressive
strength of ground granulated blast furnace slagpalm oil fuel ash-y ash
based geopolymer mortar. Mater Des 2014;56:83341.
[27] Jawahar JG, Sashidhar C, Reddy IVR, Peter JA. Micro and macrolevel properties
of y ash blended self compacting concrete. Mater Des 2013;46:696705.
[28] ParhamRabbani YD, Mohammad Reza Atrechian, Masoud Karimi, Ali Tolooiyan.
The potential of lime and grand granulated blast furnace slag (GGBFS) mixture
for stabilisation of desert silty sands. J Civil Eng Res 2012;2:10819.
[29] Samet B, Chaabouni M. Characterization of the Tunisian blast-furnace slag and
its application in the formulation of a cement. CemConcr Res 2004;34:11539.
[30] Wang Q, Yan P, Mi G. Effect of blended steel slag GBFS mineral admixture on
hydration and strength of cement. Constr Build Mater 2012;35:814.
[31] Wee TH, Suryavanshi AK, Tin SS. Inuence of aggregate fraction in the mix on
the reliability of the rapid chloride permeability test. Cem Concr Compos
1999;21:5972.
[32] Kourounis S, Tsivilis S, Tsakiridis PE, Papadimitriou GD, Tsibouki Z. Properties
and hydration of blended cements with steelmaking slag. Cem Concr Res
2007;37:81522.
C.-J. Tsai et al. / Materials and Design 60 (2014) 267273 273