Construction and Building Materials 113 (2016) 764–772

Contents lists available at ScienceDirect

Construction and Building Materials

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

Preparation of calcium sulfoaluminate-belite cement from marble sludge

waste
E.A. El-Alfi, R.A. Gado ⇑
National Research Centre, Ceramics and Building Materials Department, 12311 Dokki, Cairo, Egypt

h i g h l i g h t s g r a p h i c a l a b s t r a c t

The recycling of waste material was In this study, the feasibility of recycling marble sludge waste in the production of calcium
successfully carried out leading to the sulfoaluminate-belite cement was investigated.
production of less CO2 emission
cement known as sulfoaluminate- 1: C4A3Š
2: β-C2S
belite cement. 3: CH
1

The influences of the cement raw mix 4: CaSO4

5: SiO2 2 1250 °C
composition and the different 1
1 1
burning temperature were 1 2 1

investigated to evaluate the reuse

feasibility of marble sludge waste 1

material in cement production.

Intensity (a.u)

Calcium sulfoaluminate–belite 2 1200 °C

1
cement, environmentally cement, can 1
1 2 2

be produced by raw mix contains in

weight percent 25% kaolin, 20% 1
2
gypsum and 55% marble sludge waste
at firing temperature 1200 °C. 3 4
1150 °C
5 3 13
1 2 1 2 1

10 20 30 40 50
2-Theta-scale

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

Article history: Marble sludge waste was used as a major cement raw material in sintering sulfoaluminate cement clinker
Received 12 November 2015 successfully in the laboratory. The influences of raw mix composition as well as different burning
Received in revised form 7 March 2016 temperatures were investigated. Starting materials and prepared cement were characterized through
Accepted 20 March 2016
different techniques including; Fourier transform infrared spectroscopy (FTIR); X-ray diffraction (XRD)
and scanning electron microscopy (SEM). The results reveal that calcium sulfoaluminate–belite cement
can be produced by burning a raw mixture contains in weight percent (25% kaolin, 20% gypsum and
Keywords:
55% marble sludge waste) at firing temperature ranging between 1200 and 1250 °C.
Marble sludge waste
Environmentally cement
Ó 2016 Elsevier Ltd. All rights reserved.
Low-energy cement
Sulfoaluminate–belite cement
Ettringite

1. Introduction

Calcium sulfoaluminate–belite (CSAB) cement is promoted as a

Abbreviations: C, CaO; S, SiO2; Š, SO3; A, Al2O3; F, Fe2O3; H, H2O.
⇑ Corresponding author. sustainable alternative to Portland cement (PC) because of lower
E-mail addresses: redagado@gmail.com, redagado@hotmail.com (R.A. Gado). energy consumption as well as reduction in CO2 emission during

http://dx.doi.org/10.1016/j.conbuildmat.2016.03.103
0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
 E.A. El-Alfi, R.A. Gado / Construction and Building Materials 113 (2016) 764–772
production. The calcium sulfoaluminate–belite cement phases can
form and are stable at a temperature of approximately 1250 °C,
which is lower than the temperature used for Portland cement
production. Further, calcium sulfoaluminate–belite clinker is more
friable and soft than Portland cement clinker, which reduces the
energy needed for grinding process [1]. Calcium sulfoaluminate–
belite cement principally utilizes Ye’elimite phase (C4A3Š), instead
of tricalcium silicate (Alite, C3S) phase, as the primary early-age
strength gaining phase and utilizes dicalcium silicate (Belite,
b-C2S) phase to develop additional long-term strength.
Ye’elimite phase (C4A3Š) has lower calcium oxide content than
Portland cement phases, making it an attractive option for devel-
oping sustainable cement product [2]. Calcium sulfoaluminate–
belite cement has shown rapid setting, high early strength, and
shrinkage compensating properties due to the fast reacting of
Ye’elimite phase (C4A3Š) and the expansive nature of ettringite
[3]. In field practices, this kind of cement have been used mainly
in precast concrete applications in cold environments and have
shown good dimensional stability, low permeability, low alkalin-
ity, good durability, and comparable compressive strength to Port-
land cement [4–7].
Raw mixes for calcium sulfoaluminate cement differ from those
for Portland cement in that they contain a significant amount of
sulfate. Therefore, the reactions and products are quite different
from those normally found in Portland cement production [8].
The raw mix composition of this cement can be based on conven-
tional raw materials (limestone, clay, bauxite and iron ores) but
industrial by-products or wastes can be used as well [9,10] (for
example fly ash, pyrite ash, galvanic sludge, metallurgical slags,
phosphogypsum, etc. . ..). Furthermore, this new class of cement
can be produced in existing installations using a conventional
cement kiln system.
Several types of industrial wastes can be used as raw materials
to obtain the calcium sulfoaluminate (CSA) cement clinker. Wu
et al. obtained a clinker of this cement from a mixture of municipal
solid incineration wastes-limestone-bauxite-gypsum at 1250 °C,
reaching more than 73.2 Mpa after 28 days of curing [11]. Li et al.
[12] obtained cement by firing a mixture of fly ash, bauxite and cal-
cium carbonate at 1300 °C. Li et al. [13] obtained the calcium sulfoa-
luminate (CSA) cement at a temperature as low as 1150 °C using fly
ash and sludge as raw materials. Singh et al. [14,15] reported the
formation of a ferric calcium sulfoaluminate cement from a mixture
of calcium oxide, red mud and bauxite at 1250 °C, as well as using
waste from a fertilizer industry, bauxite and iron mineral ore at
1250 °C. In general, these works showed the feasibility to obtain
calcium sulfoaluminate (CSA) cement using many industrial wastes
as raw materials. However, there is a great variety of industrial
wastes that have the potential to be used as a source of the main
components to fabricate this kind of cement [16].
Marble processing lines in the Egyptian factories produce huge
amounts of marble wastes either solid or liquid (slurry, resulted
from sawing the blocks to slabs and grinding and polishing pro-
cesses). The random disposal of these wastes, estimated as one
million tons yearly, in the areas near the factories (irrespective of
their possible economic values) and with the increasing production
cause severe environmental problems, such as land degradation,
increase wastage of minerals, air pollution, water pollution, dam-
age to flora and fauna as well as human resources displacement
[17]. Recycling process of industrial by-products and waste mate-
rials coming from industrial manufacture activates have become
an urgent problem for the near future. In the light of environmen-
tal protection standards which aiming to limit the accumulated of
the industrial dump, the development of recycling techniques cap-
able of exploiting the industrial wastes into new marketable prod-
ucts acquires an increasing importance for the industrial and
environmental sectors.
765
In this study, the feasibility of reusing the marble sludge waste
as starting materials in the preparation of sulfoaluminate-belite
cement was evaluated. This work shows the feasibility of
producing calcium sulfoaluminate-belite cement from a mixture
of kaolin; marble sludge waste as a source of calcium carbonate
(CaCO3) and gypsum (CaSO41/2H2O). The sintering temperature
in the laboratory muffle furnace could be controlled by the temper-
ature of 1150, 1200 and 1250 °C. The cement quality was investi-
gated to evaluate the reuse feasibility of marble sludge waste
material in cement production.
2. Materials and methods
2.1. Starting raw materials
The raw materials used in this study were kaolin, marble sludge and gypsum.
Kaolin material was supplied from the general company for ceramic & porcelain
products (Sheeni); Marble sludge was collected from Shaq El Thoaban marble
industrial zone; Egypt. The surface area of marble sludge waste was determined
by using the method of N2 adsorption (BET) using automated gas sorption; model
NOVA, Version 1.12 from the quantachrome. The measured surface area and speci-
fic gravity of marble sludge waste were 0.6695 m2/g and 2.67 respectively. Gypsum
(hemihydrate, CaSO4 1/2H2O) was supplied from Gypsina company; Egypt. The
main oxide compositions of starting raw materials chemically analyzed by X-ray
fluorescence technique (Wavelength dispersive XRF; PANalytical) were reported
in Table 1. The mineral composition phases of starting raw materials were identi-
fied by X-ray diffraction analysis technique using (XRD; X’Pert-PANalytical) diffrac-
tometer with Ni filter, with Cu Ka (k = 1.5406 Å) radiation at 40 kV, 30 mA at a
scanning speed of 0.020°/s over the 2h range of 4–60°. FTIR spectra of starting
raw materials were acquired using a JASCO FT/IR-6100. The IR spectra were
recorded between 400 and 4000 cm1 with a resolution of 4 cm1 at room
temperature.
For preparing cement mixes, appropriate amounts of starting raw materials in
predetermined wt.% proportions, as shown in Table 2, were taken and ball milled
for 30 min using a top planetary ball mill (Fritsch planetary mono mill Pulverisette
6) for homogenization. The resultant powder samples with chemical oxides compo-
sition based on the XRF data of started materials, as shown in Table 3, were made
into a thick paste using a low amount of water (5%) and moulded under a pressure
of 50 Mpa into 5  5 cm cubes [18]. The cubes were dried overnight in a hot air
oven at 100 °C and then fired in an electric laboratory muffle furnace (Lenton) at
different firing temperatures (1150, 1200 and 1250 °C by a heating rate of 10 °C/
min) with sintering duration at the maximum temperature for 1 h in furnace for
all cement clinker mixes.
The theoretically expected calcium sulfoaluminate-belite cement phases were
determined by adapting the Bogue method, shown in Table 4. The Bogue method
is a technique used in the cement industry to estimate phase composition in
Portland cement clinker from its chemical oxide composition. It was adapted for
calcium sulfoaluminate-belite cement by assuming a phase assemblage of C4AF,
C4A3Š, C2S, CŠ and C, as shown in the following equations (1)–(5) [1].
%C4 AF ¼ 3:043ð% Fe2 O3 Þ ð1Þ
 ¼ 1:995ð%Al2 O3 Þ  1:273ð% Fe2 O3 Þ
%C4 A3 S ð2Þ
%C2 S ¼ 2:867ð% SiO2 Þ ð3Þ
 ¼ 1:700ð% SO3 Þ  0:445ð% Al2 O3 Þ þ 0:284ð% Fe2 O3 Þ
%CS ð4Þ
%C ¼ 1:000ð% CaOÞ  1:867ð% SiO2 Þ  1:054ð% Fe2 O3 Þ  0:550ð% Al2 O3 Þ
 0:700ð% SO3 Þ ð5Þ
The synthesized clinkers were ground using a top planetary ball mill (Fritsch
planetary mono mill Pulverisette 6) for 20 min to reach a Blaine fineness of 4500
(±100) m2/kg [19]. All the cement samples were submitted to characterization by
different techniques in order to assess the reactants conversion and evaluate the
phases formed on clinkering process at different temperatures [20].
3. Results and discussions
3.1. Characterization of starting raw materials
The results of X-ray diffraction (XRD) analysis of supplied start-
ing raw materials as shown in Fig. 1 indicates that the kaolin shows
the typical diffraction patterns of a well-crystallized layer lattice
mineral of kaolinite with muscovite impurities as Illite mineral.
766 E.A. El-Alfi, R.A. Gado / Construction and Building Materials 113 (2016) 764–772

Table 1
Chemical oxide composition of starting materials, wt.%.

SiO2 Al2O3 Fe2O3 CaO Na2O K2O TiO2 MnO MgO SO3 BaO LOIa
Kaolin 46.68 35.75 1.49 0.10 0.05 2.96 0.07 0.01 0.19 0.03 0.02 12.18
Gypsum 0.46 0.120 0.050 37.61 0.030 0.00 0.00 0.00 0.120 53.66 0.00 7.76
Marble sludge 0.150 0.010 0.040 55.32 0.700 0.010 0.010 0.010 0.110 0.130 0.00 43.30
a
Loss on Ignition at 950 °C.

Table 2 Table 4
Prepared cement mixes from starting raw materials, wt.%. The theoretically expected calcium sulfoaluminate-belite cement phases adapting by
the Bogue method [1].
Mixes Kaolin Gypsum Marble sludge waste
Phases Phase calculated (wt.%)
A 25 20 55
B 20 20 60 A B C
C 15 20 65
C4A3Š 22.30 18.05 13.71
C2S 43.60 35.44 27.11
C4AF 1.58 1.31 1.03
Characteristic peaks for kaolinite (2h = 12.34, 20.34 and 24.87°), CŠ 18.62 19.85 21.13
while peaks assigned to Illite mineral (2h = 9.11, 17.50 and C 3.76 14.41 25.31
26.90°). The X-ray diffraction pattern of marble sludge waste as
seen in Fig. 1 indicates that the marble sludge waste sample was
mainly composed of calcite mineral (CaCO3) as confirmed by XRF K: Kaolin I: Illite C: Calcite G: Gypsum
analysis of marble sludge in Table 1. The XRD pattern also indicates G
that the presence of pure crystalline phase of gypsum (calcium G
G
sulfate hemihydrate, CaSO41/2H2O) without any impurity phases
[21].
Infrared spectra (FTIR) of the starting raw materials (kaolin,
marble sludge waste and gypsum) are given in Fig. 2. A typical G GG
G Gypsum
spectrum of kaolin show bands, at 3640, 3455, 1610, 1035 and
Intensity (a.u)

700 cm1, the band observed at around 3640 cm1 has been C
ascribed to the inner hydroxyl while the bands observed at around
the other characteristic bands in this region are generally ascribed
to vibrations of the external hydroxyl. The absorption bands
observed at 3455 cm1 and 1610 cm1 could be assigned to the
C C C C Marble sludge

OH vibrational mode of the hydroxyl molecule in water, which is K K

observed in almost all the natural hydrous silicates. In the
1035 cm1 and 476 cm1 regions, main functional groups were
Si–O and Al–OH. The doublet at 700 cm1 is due to Si–O–Si inter- I
I
K, I

K, I

nal tetrahedral bridging bonds in SiO2 and OH deformation band I

I K K Kaolin
[22].
The infrared spectrum of the marble sludge waste was charac-
10 20 30 40 50 60
terized by three prominent absorption bands for calcite mineral
2-Theta-scale
at 1400, 850 and 700 cm1 and two minor peaks at 2500 and
2000 cm1, as shown in Fig. 2. The major absorption bands of Fig. 1. X-ray diffraction pattern of starting raw materials.
carbonate spectra in the wavelength region have been attributed
to the fundamental vibrations of the carbonate radical, CO-2 3 , and
various bands have been assigned to correspond to the vibrations bending vibrations in the range 1620 cml, and the bending vibra-
of the carbon and oxygen atoms along a crystallographic axial tion of the SO4 tetrahedron in the range 660–550 cml as well as
direction [23,24]. The broad bands at 3450 and 1610 cm1 corre- S–O stretching vibration for sulfate group in the range 1140–1080
sponding to the vibrational mode of the hydroxyl molecule in [25,26].
water as well as peaks in range 2920 and 2850 cm1, which are
attributed to the (tCH2) antisymmetric stretching vibration and 3.2. Characterization of prepared cement samples
symmetric stretching vibration (tCH2) of organic impurities in
marble sludge waste [22]. 3.2.1. Fourier transform infrared spectroscopy (FTIR)
The FTIR spectrum of the gypsum was shown in Fig. 2. There are The FTIR spectra of prepared cement samples (A, B and C) at
main peaks in the gypsum FTIR spectrum for O–H bending vibra- different burning temperatures (1150, 1200 and 1250 °C) are illus-
tion modes in the range 3600–3000 cm1, the in-plane O–H  O trated in Figs. 3–5, respectively. Each prepared cement sample has

Table 3
Chemical oxide composition of prepared cement mixes.

Mixes/oxides SiO2 Al2O3 Fe2O3 CaO Na2O K2O TiO2 P2O3 MgO SO3 F LOIa
A 11.84 8.96 0.40 37.97 0.40 0.74 0.02 0.07 0.13 10.81 0.03 28.41
B 9.51 7.18 0.33 40.73 0.43 0.59 0.02 0.06 0.120 10.81 0.02 29.96
C 7.19 5.39 0.25 43.49 0.46 0.45 0.01 0.06 0.12 10.81 0.02 31.52
a
Loss on Ignition at 950 °C.
 E.A. El-Alfi, R.A. Gado / Construction and Building Materials 113 (2016) 764–772 767

3000 cm-1

1630,1615 cm-1
2360 cm-1
Gypsum

1620 cm-1

1417cm-1
660 cm-1-1
3455 cm-1
3600 cm-1

550 cm

1000 cm-1
1140 cm-1

640 cm-1
1118 cm-1

880 cm-1
1250°C

505 cm-1
3415 cm-1
Marble sludge
Transmittance, T%

2850 cm-1

2500 cm-1

700 cm-1
2920 cm-1
3455 cm-1

-1
-1

850 cm
1400 cm

1630,1615 cm-1
2360 cm-1
Transmittance, T%

1157-1122 cm-1
1417cm-1
1610 cm-1
Kaolin

3640 cm-1
700 cm-1
3455 cm-1

1200°C
3640 cm-1

-1
3415 cm-1

650 cm-1
921-833 cm

490 cm-1
675cm-1
-1
476 cm
-1
1035 cm

1630,1615 cm-1
2360 cm-1

1160-1122 cm-1
1417cm-1
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber, cm-1

3640 cm-1
1150°C

-1
925-833 cm
3415 cm-1

-1
630 cm
490 cm-1
Fig. 2. FTIR spectrum of starting raw materials.
2360 cm-1

1617 cm-1
1417 cm-1

4000 3500 3000 2500 2000 1500 1000 500

1122 cm-1
3415 cm-1

Wavenumber, cm-1
688 cm-1
520 cm-1

1250°C
995-836 cm-1
1157 cm-1

Fig. 4. FTIR spectrum of cement samples (B) burned at different temperatures.

-1

800–1000 cm1, the third band at 620 cm1 region is due to

2360 cm

1617 cm-1

vibrational modes of [AlO4]5 tetrahedral as well as belite phase

Transmittance, T%

1417 cm-1

(b-C2S) the another calcium silicate phase, exhibits strong bands

1122 cm-1

in the area 1000–800 cm1 with maximums at 990 and

520 cm-1
690 cm-1
3420 cm-1

1200°C 840 cm1 due the stretching Si–O bond of the silicon tetrahedron
920-835 cm-1
1160 cm-1

and the bending vibration absorption band appear at lower fre-

quencies 520 cm1 and a shoulder at 538 cm1. A broadened signal
appears between 900 and 800 cm1, centered at 857 cm1. This
2360 cm-1

feature is strongly asymmetrical which probably due to the result

1617 cm-1

of the convolution of the two bands of belite phase (b-C2S), as a

1417 cm-1

1122 cm-1

consequence of lower crystal perfection caused mainly by the pres-

3420 cm-1

520 cm-1

1150°C
690 cm-1

ence of foreign ions in the lattice or because of the small particle

414 cm-1
1160 cm-1

size of minerals of calcium sulfoaluminate-belite cement. But it

925-835 cm-1

is also possible to highlight the presence of the three anhydrite

4000 3500 3000
Wavenumber, cm-1
Fig. 3. FTIR spectrum of cement samples (A) burned at different temperatures.
its own unique infrared spectrum bands. The infrared spectra of all
prepared cement samples, shows the main absorption band and
the identification of the functional group signals for calcium
sulfoaluminate-belite cement which can be described as follow:
ye’elimite phase (C4A3Š) has two absorption bands due to vibra-
tional modes of sulfate [SO4]2 groups at 1100–1200 cm1 and
a very intense absorption band due to silicate groups near to
bands at 677, 615, and 600 cm1, respectively [27]. Also, the pres-
ence of calcite (CaCO3) impurities was identified with IR absorption
bands at 874 and 1424 cm1 [28].
In general, from the results of FTIR spectra of prepared cement
2500 2000 1500 1000 500
samples (Figs. 3–5), as decreasing of kaolin percent in samples
from A to C, the following differential notes are observed i) the
intensity of calcium sulfoaluminate-belite cement main bands that
appear in the range 400–1200 cm1 were remarkably decreased.
ii) The intensity of the bands at 3640 cm1 and 1630 cm1 were
increases indicating liberation of more Ca(OH)2 as a result of free
unreacted lime hydration. iii) The strong asymmetric stretching
Si–O band at 925 cm1 is decreased to low intensity indicates
the low formation of belite phase (b-C2S). iv) the carbonate bands
at 2360, 1420 and 875 cm1 are observed mainly due to the reac-
tions of atmospheric CO2 with calcium hydroxide [29]. Further-
more, unreacted silica gives rise to its absorption band at 925
(Si–O–Si asymmetric stretching vibration) and unreacted lime
768 E.A. El-Alfi, R.A. Gado / Construction and Building Materials 113 (2016) 764–772

1637,1617 cm-1
1: C4A3Š

2360 cm-1
2: β-C2S

1427 cm-1
3: CH 1

1000 cm-1
4: CaSO4

1120 cm-1
3640 cm-1

880 cm-1
1250°C

675 cm-1
3415 cm-1

5: SiO2
2
1
1 1 1 2 1 1250°C

1637,1617 cm-1
2360 cm-1
Transmittance, T%

1
1469 cm-1
3640 cm-1

1159-1122 cm-1
3419 cm-1

1200°C

-1
943-833 cm
2

675 cm-1

Intensity (a.u)
1
1 12 2 1200°C
1630,1617 cm-1
2360 cm-1

1
-1

2
1427 cm

1150°C
3640 cm-1

1160-1122 cm-1
3415 cm-1

-1
925-835 cm

3 4
-1

5 3 13 1150°C
675 cm

1 2 12 1

4000 3500 3000 2500 2000 1500 1000 500 10 20 30 40 50 60

Wavenumber, cm-1 2-Theta-scale

Fig. 5. FTIR spectrum of cement samples (C) burned at different temperatures. Fig. 6. XRD pattern of cement samples (A) at different temperatures.

gives rise to its absorption bands at 3640, 1630 and 1450– 1: C4A3Š
1500 cm1 (vibrations of OH group and Ca–O stretching vibration). 2: β-C2S
Calcite arising from partial carbonation of unreacted lime gives its
3: CH 2
absorption bands at 2360, 1420, 840, and 700 cm1. The belite 1
4
4: CaSO4
phase (b-C2S) gives rise to the following absorption bands: three 4 2 1
maxima at 950, and 840 cm1 (Si–O asymmetric stretching 5: SiO2
3
2, 3 2
4
modes), medium intensity band at 430 cm1 (Si–O bending mode), 1250°C
5
and a strong band at 530 cm1 (Si–O–Si out of plane bending 2
mode) [30]. On the other hand, as clinker temperatures increase
from 1150 to 1250 °C leads to i) shifting the main bands of calcium
sulfoaluminate-belite cement that appear in the range of
400–1200 cm1 to a little higher frequencies as a result of more 1
polymerization [31]; ii) decreasing in the intensity of the bands 2
corresponding to Portlandite Ca (OH)2 at 3630 cm1 and
1630 cm1. The band assignments are in good agreement with 2, 3 2
Intensity (a.u)

3 °
those reported in the previous studies [27–31].
45 3 4 1200 C
3 12
3.2.2. X-ray diffraction analysis (XRD)
The XRD pattern of prepared cement samples (A, B and C) were
shown in Figs. 6–8, respectively. The XRD pattern of prepared
cement sample (A) as illustrated in Fig. 6, shows that the ye’elimite 2
2, 3
phase (C4A3Š, 2h = 23.66, 33.82 and 41.69°) was the major phase in 14
the clinker product with belite phase (b-C2S, 2h = 32.16, 34.20 and 4
3 3
41.17°), unreacted silica (SiO2, 2h = 20.86, 26.64 and 50.15°) as 24
1 3 1150°C
minor phases in clinker samples burned at 1150 °C. 5 2
1 5
The appearance of the main diffraction intensity for the belite 5
phase (b-C2S, 2h = 32.16°) in the cement sample (A) which fired
at 1150 °C confirming that the formation of this phase starts at
lower firing temperature within the range of 1000 °C [32]. On 10 20 30 40 50 60
the contrary, the intensity of the ye’elimite phase (C4A3Š) enhanced 2-Theta-scale
as firing temperature increasing to 1250 °C. XRD pattern shows
that the characteristic peak of unreacted silica (SiO2) is gradually Fig. 7. XRD pattern of cement samples (B) at different temperatures.
 E.A. El-Alfi, R.A. Gado / Construction and Building Materials 113 (2016) 764–772
1: C4A3Š
2: β-C2S
3: CH 4 ing in kaolin percent. Thus, the crystalline phase C4A3Š in the
4: CaSO4
5: SiO2
1 2, 3
4, 2 4
2, 3
4, 2
Intensity (a.u)
3 4
1
3 4
1 4, 2
4
3 2, 3
3 1
4 sized between 1200 °C and 1250 °C. The high amounts of C4A3Š
10 20 30
2-Theta-scale
Fig. 8. XRD pattern of cement samples (C) at different temperatures.
weakened in prepared cement samples as the firing temperature
increases, which indicates the more consuming of acidic oxides
from raw kaolin for more formation of calcium sulfoaluminate-
belite cement. The XRD show that the appearance of Portlandite
(CH, 2h = 18.04, 28.69 and 34.11°) in the sample (A) burned at
1150 °C which mainly due to the presence of uncombined calcium
oxide in the sample.
The XRD pattern of prepared cement sample (B) fired at differ-
ent temperatures (1150, 1200 and 1250 °C) are shown in Fig. 7. The
data show that, the appearance of Portlandite (CH, 2h = 18.04,
28.69 and 34.11°), unreacted silica (SiO2, 2h = 20.86, 26.64 and
50.15°) and anhydrite calcium sulfate (CaSO4, 2h = 25.44, 31.34
and 40.79°) with belite (b-C2S) and ye’elimite phase (C4A3Š). It
can be also noticed that the cement sample (B) fired at 1150 and
1250 °C shows the higher intensity of Portlandite and anhydrite
calcium sulfate (CaSO4) than those fired at 1200 °C. This probably
due to decomposition of calcium sulfoaluminate phases at a higher
temperature (1250 °C) and incomplete reaction at 1150 °C which
reflects on calcium sulfoaluminate-belite cement main phases as
illustrated in Fig. 7. The XRD results of the cement sample (B) show
that the optimum temperature was 1200 °C due to less Portlandite
(CH) phase peaks as well as less anhydrite calcium sulfate (CaSO4)
phase peaks.
The XRD pattern of the cement sample (C) burned at different
firing temperatures (1150, 1200 and 1250 °C) are shown in Fig. 8.
The results of the cement sample (C) fired at 1150 °C show the
essentially following mineralogical composition phases C4A3Š,
b-C2S, CH and anhydrite calcium sulfate (CaSO4). It can be also
noticed that cement sample (C) fired at 1250 °C shows the higher
intensity of Portlandite phase (CH) and anhydrite calcium sulfate
phase (CaSO4) as well as the lower intensity of C4A3Š and b-C2S
phases than cement samples (B and C) fired at the same
769
temperature. A decrease in the intensity of the following phases
C4A3Š and b-C2S content was observed with an increase in the per-
cent of CaO sourced from marble sludge waste as well as decreas-
doped samples became less abundant compared to the cement
samples A and B burned at the sample temperatures. The anhydrite
calcium sulfate (CaSO4) was detected in cement samples B and C
1250°C until burning temperature at 1250 °C, which indicated that the
12
reaction to form C4A3Š was still not complete at these high temper-
atures, which mainly due to a deficiency in acidic oxides content in
their raw mix.
Finally, the X-ray diffraction analysis results of different cement
samples (A, B and C) in this study indicates that the characteristic
peaks of the ye’elimite phase (C4A3Š) are gradually weakened in
the cement samples A, B and C, respectively. This is mainly due
1 4 1200°C to the reduction in kaolin percent in their mixes. According to X-
2
2 ray diffraction analysis results, the optimum burning temperature
for the formation of calcium sulfoaluminate-belite cement was
found to be (1250 °C) for the raw mix sample containing in weight
percent 25% kaolin, 20% gypsum and 55% marble sludge waste. The
formation of the ye’elimite phase (C4A3Š) in this study is in agree-
ment with previous reports which indicated the formation of the
ye’elimite phase at 1200–1250 °C by using other industrial by-
product wastes [33]. From the above experiments, it was shown
2 24
1150°C that the calcium sulfoaluminate-belite cement could be synthe-
and b-C2S in cement indicates the high durability and rapid hard-
ening of calcium sulfoaluminate cement [34]. The XRD results con-
firm all results obtain from different used techniques in this study.
40 50 60
4. Characterization of hardened cement samples
The specimens of prepared sulfoaluminate-belite cement paste
(w/c ratio = 0.50) were cured for 28 days, then they were charac-
terized by determining their physico-mechanical properties
[35,36]. The results were based on the average of three cube sam-
ples. The physico-mechanical properties of the hardened cement
specimens cured at room temperature are illustrated in Table 5.
As indicated from results, the highest compressive strength values
(within the range of 36 Mpa) were recorded to cement samples (A)
which burned at 1200 and 1250 °C, respectively. It was observed
that hardened cubes of cement samples (C) which burned at
1200 and 1250 °C respectively were self-cracked and deteriorated
during the curing period. The dimensional instability mainly has
been linked to the high free lime content in hardened samples.
The large amounts of free lime may cause expansion, strength loss
and crack of the hardened paste, due to a delayed hydration of free
calcium oxide (hard burned) to produce calcium hydroxide, which
takes place topo-chemically and usually associated with an
increase in matrix volume. Thus, excessive amounts of free calcium
oxide in clinker must be avoided [37]. After the mechanical testing,
a fractured hardened paste of highly compressive strength sample
(A-1200 °C) was chosen, crushed by hand with a metallic mortar.
The ground powder was characterized by different used techniques
as well as morphology investigation by scanning electron micro-
scope (SEM) attached with EDX Unit (Energy dispersive X-ray
analyses), with accelerating voltage 30 K.V., (JEOL Ltd., Japan).
4.1. X-ray diffraction analysis (XRD)
The XRD pattern of hardened sample cured for 28 days in com-
parison with its anhydrous sample (A-1200 °C) is illustrated in
Fig. 9. The main hydration products were ettringite phase
(C3A3CŠ32H2O, 2h = 11.49, 19.92 and 28.91°), low sulfate calcium
sulfoaluminate hydrate (monosulfate C3ACŠ12H2O, 2h = 9.92,
770 E.A. El-Alfi, R.A. Gado / Construction and Building Materials 113 (2016) 764–772

Table 5
The physico-mechanical properties of the hardened cement samples cured up to 28 days.

Properties Mixes
A B C
1150 °C 1200 °C 1250 °C 1150 °C 1200 °C 1250 °C 1150 °C 1200 °C 1250 °C
Bulk density [g/cm3] 1.88 1.90 1.90 1.83 1.84 1.86 1.80 – –
Apparent porosity [%] 16.70 15.20 14.85 19.54 17.95 17.03 24.53 – –
Compressive strength [Mpa] 27 36 35 15.7 18.45 22.11 9.86 – –

3: Ettringite Hydrated sample (Age : 28 Days) 4.2. Fourier transform infrared spectroscopy (FTIR)
4: Monosulphate
Spectroscopic methods are commonly used to study the chem-
5: CSH
4, 5 3 4 istry of cement hydration. The chemical environment of the vibra-
6: CH
43 tions band of the dry cement bonds changes as hydration occurs.
The sample will absorb some of the light at wavelengths that are
3 3 6 characteristic of its chemical composition. In the present work,
6 4 35 6 5 the hydration of calcium sulfoaluminate-belite cement has been
3 43 observed by means of infrared spectroscopy.
Infrared spectra (FTIR) of the hydrated sample in a comparison
with its anhydrous sample are given in Fig. 10. This spectrum cor-
Intensity (a.u)

responds to the difference between that acquired after 28 days of

hydration, and the spectrum of dry cement. From the results in
Fig. 10, we observe significant changes in the spectrum of dry
cement sample due to hydration process. As hydration progresses
at 28 days, there is a broad feature forming with its center at
1: C4A3Š 3450 and 1630 cm1 caused by the bending vibration of irregu-
1
2: β-C2S larly bound water [45–47]. The broadband at 3450 cm1 is intensi-
fied with hydration, indicating that the increase of hydrated
Dry cement (A-1200°C) products associated with water. The peaks at 3640 and
1630 cm1 correspond to the inner hydroxyl group in Portlandite
2 Ca (OH)2 which is formed as silicate phases in the cement dissolve

1
12
1 2 Hydrated sample (Age : 28 Days)
2360 cm-1

1630 cm-1

10 20 30 40 50
2-Theta-scale
1420 cm-1

Fig. 9. XRD pattern of hardened cement sample (A).

875 cm-1
3640 cm-1

1150 cm-1
3455-3420 cm-1
Transmittance, T%

995cm-1

19.90 and 22.28°) calcium silicate hydrate phase (CSH, 2h = 28.58,

29.06 and 31.58°) and Portlandite Ca (OH)2 (CH, 2h = 18.10, 28.72
and 34.17°).
The ettringite deficiency in the hardened sample mainly due to
the absence of additional gypsum added upon hydration process.
The ye’elimite phase is used or activated with an addition of cal- Dry cement (A-1200°C)
cium sulfate, either as gypsum or anhydrite [38]. Furthermore, it
is possible to modify the hydration process of calcium sulfoalumi-
nate cement not only by its composition but also by the selection
of different quantities or sources of calcium sulfate [39]. Hence,
the amount of added calcium sulfate strongly modifies the ettrin-
gite formation and the water demand to complete full hydration
[40,41]. In addition, the amount of ettringite is affected by the reac-
tivity of calcium sulfate (solubility and dissolution rate) at early
ages [40]. Hence, the selection of the sulfate source is a key issue
to achieve the desired properties. Furthermore, depending on the
other phases present in calcium sulfoaluminate cement, other 4000 3500 3000 2500 2000 1500 1000 500
hydration products may occur such as CSH gels, stratlingite,
Wavenumber, cm-1
katoite, monocarboaluminate or hydrogarnet [42–44]. Hence, fur-
ther studies on the chemistry of hydration for prepared calcium Fig. 10. FTIR spectrum of hydrated cement sample (A) in comparison with its
sulfoaluminate-belite cement will be required. anhydrous sample burned at 1200 °C.
 E.A. El-Alfi, R.A. Gado / Construction and Building Materials 113 (2016) 764–772
for forming calcium silicate hydrate (CSH) phases upon hydration
process. Characteristic sulfate absorption bands are found in the
range 1150 cm1 due to the vibration of the SO2 4 group in sul-
fates [48]. The asymmetric stretching Si–O band is shifted to high
frequencies centered at 995 cm1 with hydration indicates that
the formation of calcium silicate hydrate (CSH) [49]. These dip-
hump features are taken to reflect the dissolution of belite phase
(b-C2S) and simultaneously the polymerization of silica [50–52]
to form calcium silicate hydrate (CSH). The decrease and increase
in intensities of the out-of-plane and in-plane Si-O bending vibra-
tions are occurring in significant changes with hydration. The car-
bonates peaks at 2360, 1420 and 875 cm1 are observed due to the
reactions of atmospheric CO2 with Portlandite Ca (OH)2 [39].
4.3. Scanning electron microscopy (SEM)
Scanning electron microscope was used to analyze the mor-
phology of the fracture surface of the hardened sample specimens
cured at 28 days. SEM investigations can help us to get a better
understanding of the morphologies of the sample.
The microstructural examinations of the phase distribution of
the hardened sample specimens (A, w/c ratio = 0.50) are shown
in Fig. 11. The selected micrographs of the fracture transversal sec-
tion of hardened sample (A) after curing time show the presence of
a large amount of short rod-like crystals in the hardening paste as
illustrated in Fig. 11. EDX measurements data show the presence of
Ca, S, O, Si, Al and this result combined with XRD measurements
reveal that the ettringite phase (C3A3CŠ32H2O) presence in the
hydrated sample as the main phase of hydration process [53].
The SEM micrograph shows the presence of sponge-like sulfoalu-
minate phases in the whole matrix and needle-like ettringite
Fig. 11. SEM of hydrated cement sample (A) with EDX analysis.
771
phases in the voids after 28 days of hydration. The belite phase is
observed to remain much less hydrated during this period. After
28 days of hydration, amounts of Portlandite Ca (OH)2 and calcium
silicate hydrate (CSH) were observed on SEM micrographs, indicat-
ing b-C2S hydration has taken place which already confirmed from
XRD results. It is expected that the ye’elimite phase (C4A3Š) would
be completely hydrated much before 28 days of hydration. The
combination of XRD and SEM analysis indicates that calcium
sulfoaluminate phase hydrates produce ettringite phase
(C3A3CŠ32H2O) as well as low sulfate calcium sulfoaluminate
hydrate (monosulfate C3ACŠ12H2O) while belite (b-C2S)
phase hydrates to produce the calcium silicate hydrate (CSH) that
fills porosity of specimen, and results in a dense structure.
5. Conclusion
In this study, the feasibility of recycling the marble sludge
waste as starting materials in the production of calcium
sulfoaluminate-belite cement was investigated. The prepared
cement samples were subjected to furnace sintering at tempera-
tures of 1150, 1200 and 1250 °C for the same holding time (1 h).
Different techniques were performed for mineralogical examina-
tion of the obtained samples. The following conclusions were
reached under the experimental observations of this study:
1. It is possible to recycle the marble sludge waste in the synthesis
of calcium sulfoaluminate-belite cement.
2. Calcium sulfoaluminate–belite cement, environmentally
cement, can be produced by raw mix contains in weight percent
25% kaolin, 20% gypsum and 55% marble sludge waste.
772 E.A. El-Alfi, R.A. Gado / Construction and Building Materials 113 (2016) 764–772
3. As the firing temperature of the prepared cement samples
increases between 1150 and 1250 °C, the calcium
sulfoaluminate-belite cement phases enhanced.
4. The prepared cement sample (A) gives the best burn ability
with the maximum sulfoaluminate-belite phases at 1250 °C,
as well as good mechanical properties upon hydration process
than that other prepared cement samples (B and C).
Acknowledgments
The authors would like to thanks, Dr. M. Rashwan for providing
marble sludge waste material as well as their valuable help. The
authors wish to gratefully acknowledge the help and support of
National research centre (NRC) – Egypt; Science and Technology
Development Fund (STDF) – Egypt; Institut Français d’Egypte
(IFE) and Laboratoire national de métrologie et d’essais (LNE),
France.
References
[1] Irvin A. Chen, Maria C.G. Juenger, Incorporation of coal combustion residuals
into calcium sulfoaluminate-belite cement clinkers, Cement Concr. Compos. 34
(2012) 893–902.
[2] Irvin A. Chen, Craig W. Hargis, Maria C.G. Juenger, Understanding expansion in
calcium sulfoaluminate–belite cements, Cem. Concr. Res. 42 (2012) 51–60.
[3] F.P. Glasser, L. Zhang, High-performance cement matrices based on calcium
sulfoaluminate–belite compositions, Cem. Concr. Res. 31 (2001) 1881–1886.
[4] L. Zhang, F.P. Glasser, New concretes based on calcium sulfoaluminate cement,
in: R.K. Dhir, T.D. Dyer (Eds.), Modern Concrete Materials: Binders, Additions
and Admixtures, Thomas Telford, London, 1999, pp. 261–274.
[5] K. Quillin, Performance of belite–sulfoaluminate cements, Cem. Concr. Res. 31
(2001) 1341–1349.
[6] V. Zivica, I. Janotka, Chemical resistance of sulfoaluminate belite cement based
materials, Build. Res. J. 47 (1999) 117–134.
[7] N. Sherman, J. Beretka, L. Santoro, G.L. Valenti, Long-term behaviour of
hydraulic binders based on calcium sulfoaluminate and calcium sulfosilicate,
Cem. Concr. Res. 25 (1995) 113–126.
[8] Ivan Janotka, Udovit Krajei, Subhash C. Mojumdar, Performance of
sulphoaluminate-belite cement with high C4A3Š content, in: Ceramics –
Silikáty 51 (2) (2007) 74–81.
[9] J. Berekta et al., Hydraulic behaviour of calcium sulfoaluminate-based cement
derived from industrial process waste, Cem. Concr. Res. 26 (1993) 1673–1682.
[10] S. Sahu, J. Majling, Preparation of sulfoaluminate belite cement from fly ash,
Cem. Concr. Res. 24 (1994) 1065–1072.
[11] K. Wu, H. Shi, X. Guo, Utilization of municipal solid waste incineration fly ash
for sulfoaluminate cement clinker production, in: Waste Manage. 31 (9–10)
(2011) 2001–2008.
[12] J. Li, H. Ma, H. Zhao, Preparation of sulphoaluminate-alite composite
mineralogical phase cement clinker from high alumina fly ash, Key Eng.
Mater. 334–335 (2007) 421–424.
[13] H. Li, D.K. Agrawal, J. Cheng, M.R. Silsbee, Microwave sintering of
sulphoaluminate cement with utility wastes, Cem. Concr. Res. 31 (9) (2001)
1257–1261.
[14] M. Singh, S.N. Upadhayay, P.M. Prasad, Preparation of iron rich cements using
red mud, Cem. Concr. Res. 27 (7) (1997) 1037–1046.
[15] M. Singh, P.C. Kapur, Pradip, Preparation of calcium sulphoaluminate cement
using fertiliser plant wastes, in: J. Hazard. Mater. 157 (1) (2008) 106–113.
[16] M. Gallardo, J.M. Almanza, D.A. Cortés, J.C. Escobedo, J.I. Escalante-García,
Synthesis and mechanical properties of a calcium sulphoaluminate cement
made of industrial wastes, in: Materiales de Construcción 64 (315) (2014).
[17] A.O. Mashaly, B.N. Shalaby, M.A. El-Hefnawi, Characterization of the marble
sludge of the Shaq El Thoaban Industrial Zone, Egypt and its compatibility for
various recycling applications, Aust. J. Basic Appl. Sci. 6 (3) (2012) 153–161.
[18] Maneesh Singh, P.C. Kapur, Pradip, Preparation of calcium sulphoaluminate
cement using fertiliser plant wastes, in: J. Hazard. Mater. 157 (2008) 106–113.
[19] Frank Bullerjahn, Dirk Schmitt, Mohsen Ben Haha, Effect of raw mix design and
of clinkering process on the formation and mineralogical composition of
(ternesite) belite calcium sulphoaluminate ferrite clinker, Cem. Concr. Res. 59
(2014) 87–95.
[20] Milena Marroccoli, Maria Lucia Pace, Antonio Telesca, Gian Lorenzo Valenti,
Synthesis of calcium sulfoaluminate cements from Al2O3-rich by-products
from aluminium manufacture, 2010. ISBN: 9781450714907.
[21] Joint Commission on Powder Diffraction Standards – International Centre for
Diffraction Data, http://www.icdd.com, 2003.
[22] M.F. Zawrah, R.M. Khattab, E.M. Saad, R.A. Gado, Effect of surfactant types and
their concentration on the structural characteristics of nanoclay, in:
Spectrochim. Acta Part A: Mol. Biomol. Spectrosc. 122 (2014) 616–623.
[23] C.K. Huang, Paul F. Kerr, Infrared study of the carbonate minerals, in: Am.
Mineral. 45 (1960).
[24] Hans H. Adler, Paul F. Kerr, Infrared study of the aragonite and calcite, in: Am.
Mineral. 47 (1962).
[25] A. Putnis, B. Wnkler, In situ IR spectroscopic and thermogravimetric study of
the dehydration of gypsum, Mineral. Mag. 54 (1990) 123–128.
[26] Sara Palacio, Matt Aitkenhead, Adrián Escudero, Gabriel Montserrat-Martı´,
Melchor Maestro, A.H. Jean Robertson, Gypsophile chemistry unveiled: Fourier
Transform Infrared (FTIR) spectroscopy provides new insight into plant
adaptations to gypsum soils, in: PLOS ONE J. 9 (9) (2014). e107285.
[27] Cheng Xin, Chang Jun, Lu Lingchao, Liu Futian, Teng Bing, Study of Ba-bearing
calcium sulphoaluminate minerals and cement, Cem. Concr. Res. 30 (2000)
77–81.
[28] K. Baltakys, R. Jauberthie, R. Siauciunas, R. Kaminskas, Influence of
modification of SiO2 on the formation of calcium silicate hydrate, in: Mater.
Sci.-Poland 25 (3) (2007).
[29] D. Govindarajan, R. Gopalakrishnan, Spectroscopic studies on Indian Portland
cement hydrated with distilled water and sea water, Front. Sci. 1 (2011) 21–
27.
[30] M.A. Tantawy, M.R. Shatat, A.M. El-Roudi, M.A. Taher, M. Abd-El-Hamed, Low
temperature synthesis of belite cement based on silica fume and lime, Int.
Scholarly Res. Not. (2014) (2014) 10. Article ID 873215.
[31] K. Baltaks, R. Jauberthie, R. Siauciunas, R. Kaminskas, Influence of modification
of SiO2 on the formation of calcium silicate hydrate, in: Mater. Sci.-Poland 25
(3) (2007).
[32] T.M. EL-Sokkary, Kh.A. Khalil, I.A. Ahmed, Preparation of b-dicalcium silicate
(b-C2S) and calcium sulfoaluminate (C4A3Š) phases using non-traditional
nano-materials, in: HBRC J. 8 (2012) 91–98.
[33] E. Gartner, Cem. Concr. Res. 34 (2004) 1489.
[34] K. Madan Mohan Reddy, M. Srimurali, M. Bhaskar, Characterization of calcium
sulfoaluminate cement, in: Int. J. ChemTech Res. 6 (3) (2014) 1696–1698.
[35] ASTM C20-2000, Standard Test Methods for Apparent Porosity, Water
Absorption, Apparent Specific Gravity, and Bulk Density of Burned Refractory
Brick and Shapes by Boiling Water.
[36] ASTM: C109/C109M, Standard Test Method for Compressive Strength of
Hydraulic Cement Mortars.
[37] Fabiano Raupp-Pereira, Richard James Ball, Joao Rocha, Joao A. Labrincha,
Geoffrey C. Allen, New waste based clinkers: belite and lime formulations, in:
Cem. Concr. Res. 38 (2008) 511–521.
[38] Q. Zhoua, N.B. Milestone, M. Hayes, An alternative to Portland cement for
waste encapsulation—the calcium sulfoaluminate cement system, J. Hazard.
Mater. 136 (2006) 120–129.
[39] Marta García-Maté, Angeles G. De la Torre, Laura León-Reina, Enrique R.
Losilla, Miguel A.G. Aranda, Isabel Santacruz, Effect of calcium sulfate source
on the hydration of calcium sulfoaluminate eco-cement, Cement Concr.
Compos. 55 (2015) 53–61.
[40] F. Winnefeld, S. Barlag, Calorimetric and thermogravimetric study on the
influence of calcium sulfate on the hydration of ye’elimite, J. Therm. Anal.
Calorim. 101 (3) (2009) 949–957.
[41] F. Winnefeld, S. Barlag, Influence of calcium sulfate and calcium hydroxide on
the hydration of calcium sulfoaluminate clinker, ZKG Int. 62 (9) (2009) 42–53.
[42] M.A.G. Aranda, A.G. De la Torre, Sulfoaluminate cement, in: F. Pacheco-Torgal,
S. Jalali, J. Labrincha, V.M. John (Eds.), Eco-efficient Concrete, Woodhead
Publishing Limited, Cambridge, 2013.
[43] F.P. Glasser, L. Zhang, High-performance cement matrices based on calcium
sulphoaluminate-belite compositions, Cem. Concr. Res. 31 (12) (2001) 1881–
1886.
[44] G.S. Li, G. Walenta, E.M. Gartner, Formation and hydration of low-CO2 cements
based on belite, calcium sulfoaluminate and calcium aluminoferrite, in:
Proceedings of the 12th ICCC, Montreal, Canada, 2007.
[45] M.Y.A. Mollah, M. Kesmez, D.L. Cocke, An X-ray diffraction (XRD) and Fourier
transform infrared spectroscopic (FTIR) investigation of the long-term effect
on the solidification/stabilization (S/S) of arsenic(V) in Portland cement type-
V, Sci. Total Environ. 325 (2003) 255–262.
[46] M.A. Trezza, A.E. Lavat, Analysis of the system 3CaOAl2O3–CaSO42H2O–
CaCO3–H2O by FT-IR spectroscopy, Cem. Concr. Res. 31 (2001) 869–872.
[47] T. Richard et al., Diffuse reflectance infrared Fourier transform spectroscopy as
a tool to characterise water in adsorption/confinement situations, J. Colloid
Interface Sci. 304 (2006) 125–136.
[48] Rikard Ylmén, Ulf Jäglid, Britt-Marie Steenari, Itai Panas, Early hydration and
setting of Portland cement monitored by IR, SEM and Vicat techniques, Cem.
Concr. Res. 39 (2009) 433–439.
[49] K.L. Scrivener, Backscattered electron imaging of cementitious
microstructures: understanding and quantification, Cement Concr. Compos.
26 (2004) 935–945.
[50] P. Yu et al., Structure of calcium silicate hydrate (C-S-H): near-, mid-, and
farinfrared spectroscopy, J. Am. Ceram. Soc. 82 (3) (1999) 742–748.
[51] E.T. Stepkowska et al., Hydration products in two aged cement pastes, J.
Therm. Anal. Calorim. 82 (2005) 731–739.
[52] D. Govindarajan, R. Gopalakrishnan, Spectroscopic studies on Indian Portland
cement hydrated with distilled water and sea water, in: Front. Sci. 1 (1) (2011)
21–27.
[53] P. Arjunan, Michael R. Silsbee, Della M. Roy, Sulfoaluminate-belite cement
from low-calcium fly ash and sulfur-rich and other industrial by-products,
Cem. Concr. Res. 29 (1999) 1305–1311.