Embed Size (px)
Citation preview
University of South Florida University of South Florida
Scholar Commons Scholar Commons
Graduate Theses and Dissertations Graduate School
4-2-2021
Sulfate Optimization in the Cement-Slag Blended System Based Sulfate Optimization in the Cement-Slag Blended System Based
on Calorimetry and Strength Studies on Calorimetry and Strength Studies
Mustafa Fincan University of South Florida
Follow this and additional works at: https://scholarcommons.usf.edu/etd
Part of the Materials Science and Engineering Commons
Scholar Commons Citation Scholar Commons Citation Fincan, Mustafa, "Sulfate Optimization in the Cement-Slag Blended System Based on Calorimetry and Strength Studies" (2021). Graduate Theses and Dissertations. https://scholarcommons.usf.edu/etd/8770
This Dissertation is brought to you for free and open access by the Graduate School at Scholar Commons. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Scholar Commons. For more information, please contact [email protected].
Sulfate Optimization in the Cement-Slag Blended System Based on Calorimetry and Strength
Studies
by
Mustafa Fincan
A dissertation submitted in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Department of Mechanical Engineering
College of Engineering
University of South Florida
Major Professor: Alex A. Volinsky, Ph.D.
Rasim Guldiken, Ph.D
Gulfem Ipek Yucelen, Ph.D.
Aydin K. Sunol, Ph.D.
Abdul Malik, Ph.D.
Date of Approval
April 2, 2021
Keywords: SO3 Optimization, Supplementary Cementitious Materials (SCMs), Heat of
Hydration, Strength, Isothermal Calorimetry
Copyright © 2021, Mustafa Fincan
Dedication
I would like to dedicate my dissertation to my wife and my parents who have always loved,
supported, and encouraged me unconditionally. I am very grateful to have you in my life. Thank
you for all your support over the years.
Acknowledgments
I would like to express my special gratitude and thanks to my thesis advisor Dr. Alex A.
Volinsky. I would also like to thank the members of my committee, Dr. Rasim Guldiken, Dr.
Gulfem Ipek Yucelen, Dr. Aydin K. Sunol, and Dr. Abdul Malik, for their valuable comments,
feedback, and suggestions. I would like to thank all of my friends and colleagues, Dr. Ahmet
Colak, Abdullah Ozkan, Dr. Mohammad Khawaja, Dr. Ferhat Karakas, Dr. Burak Sarsilmaz, Dr.
Abdulkadir Fincan, Dr. Kasim Fincan, H. Mustafa Fincan, Mahmut Erkam Fincan, Mustafa Gecit,
Mohammed Al Yaarubi, and members of the USF Construction Materials Research Lab, for their
meaningful guidance and encouragement on this dissertation, which has inspired me to improve
my work. Finally, I am grateful to everyone for their support, both directly and indirectly, in
completing my dissertation.
i
Table of Contents
List of Tables ................................................................................................................................. iii
List of Figures ................................................................................................................................ iv
Abstract .......................................................................................................................................... vi
Chapter 1: Introduction ................................................................................................................... 1
Chapter 2: Literature Review .......................................................................................................... 5
2.1 General Overview of Portland Cement ......................................................................... 5
2.2 Ordinary Portland Cement Hydration ........................................................................... 7
2.2.1 Calcium Silicates Hydration .......................................................................... 8
2.2.2 Tricalcium Aluminate Hydration ................................................................. 12
2.2.3 Tetracalcium Aluminoferrite Hydration ...................................................... 13
2.2.4 Portland Cement Hydration ......................................................................... 14
2.2.4.1 Heat of Hydration ......................................................................... 15
2.3 Supplementary Cementitious Materials ...................................................................... 17
2.3.1 Silica Fume .................................................................................................. 20
2.3.2 Metakaolin ................................................................................................... 22
2.3.3 Fly Ash ......................................................................................................... 24
2.4 Ground Granulated Blast Furnace Slag ...................................................................... 28
2.4.1 Physical, Mineralogical and Chemical Characteristics of Slag ................... 29
2.4.1.1 Structure and Reactivity of Slags .................................................. 30
2.4.2 Blended Slag– Cement Hydration ............................................................... 32
2.4.3 Effects of Slag on the Properties of Concrete .............................................. 33
2.4.3.1 Bleeding and Segregation ............................................................. 33
2.4.3.2 Workability ................................................................................... 34
2.4.3.3 Setting Time .................................................................................. 35
2.4.3.4 Compressive Strength ................................................................... 36
2.4.3.5 Durability ...................................................................................... 38
2.5 Sulfate Optimization ................................................................................................... 41
2.5.1 Sulfates ......................................................................................................... 42
2.5.2 Alkalis .......................................................................................................... 45
2.5.3 Slag and Supplementary Cementitious Materials ........................................ 46
2.5.4 Fineness and Age ......................................................................................... 48
ii
Chapter 3: Materials and Methodology ........................................................................................ 49
3.1 Materials Characterization .......................................................................................... 49
3.1.1 Chemical oxide composition........................................................................ 49
3.1.2 Mineralogical composition .......................................................................... 51
3.1.3 Fineness........................................................................................................ 53
3.2 Methodology ............................................................................................................... 54
3.2.1 Compressive strength ................................................................................... 54
3.2.2 Isothermal calorimetry ................................................................................. 55
Chapter 4: Results and Discussion ................................................................................................ 57
4.1 Determination of Optimum SO3 content..................................................................... 57
4.2 Optimization of SO3 Content in the Cement............................................................... 59
4.3 Optimization of SO3 Content in the Slag-Cement Blended System ........................... 61
4.4 Factors Affecting the Sulfate Demand in the OPC-Slag System ................................ 69
4.4.1 The Effect of Alumina (Al2O3) in the Slag .................................................. 69
4.4.2 The Effect of Cement Particle Size and Fineness ........................................ 70
4.4.3 The Effect of the Alumina with Mean Particle Size Distribution
and Slag Activity............................................................................................. 70
Chapter 5: Conclusions ................................................................................................................. 75
5.1 Summary and Conclusions ......................................................................................... 75
5.2 Future Work ................................................................................................................ 77
References ..................................................................................................................................... 78
iii
List of Tables
Table 2.1 Typical ordinary Portland cement oxide composition .................................................... 6
Table 2.2 Typical ordinary Portland cement composition .............................................................. 7
Table 2.3 The heat of hydration of pure cement clinker minerals ................................................ 16
Table 3.1 Oxide chemical compositions for cements and calcium sulfate hemihydrate .............. 50
Table 3.2 Oxide chemical composition of slags ........................................................................... 51
Table 3.3 Mineralogical composition of cements ......................................................................... 52
Table 3.4 Mineralogical composition of the slags ........................................................................ 53
Table 3.5 Particle size analysis for the slags and cements ............................................................ 53
Table 3.6 Mixing procedure for compressive test ........................................................................ 55
Table 4.1 Optimum SO3 level estimates for Cement A-D ............................................................ 60
Table 4.2 Optimum SO3 level estimates on compressive strength and calorimetry test .............. 67
iv
List of Figures
Figure 2.1 The heat evolution rate of C3S hydration. ................................................................... 10
Figure 2.2 The heat evolution rate of Portland cement hydration.. ............................................. 16
Figure 2.3 Ternary diagram of the chemical composition of common SCMs and Portland
cement ......................................................................................................................... 19
Figure 4.1 Gaussian and 2nd-order polynomial fit for three-day Cement B-Slag1
compressive strength results versus various sulfate levels. ........................................ 59
Figure 4.2 The heat of hydration flow for 100% Cement A, B, C, and D .................................... 61
Figure 4.3 Gaussian and 2nd-order polynomial fit for Cement D-Slag1 individual
specimens compressive strength results versus sulfate level at three-day. ................. 62
Figure 4.4 Gaussian and 2nd-order polynomial fit for Cement D-Slag1 average of
specimens compressive strength results versus sulfate level at three-day. ................. 62
Figure 4.5 The Gaussian versus 2nd order polynomial estimate the average strength of
the specimens' optimum sulfate level at 1-day and 3-days ......................................... 63
Figure 4.6 Heat flow of Cement C and Slag 4 in different sulfate level....................................... 64
Figure 4.7 Comparison of compressive strength and total heat of hydration of one-day
Cement A-Slag 1 on the different sulfate levels ......................................................... 65
Figure 4.8 Comparison of compressive strength and total heat of hydration of three-day
Cement A-Slag 1 on the different sulfate levels ......................................................... 65
Figure 4.9 Optimum sulfate level for the compression strength versus optimum sulfate
level for total heat of hydration (HOH). ..................................................................... 69
Figure 4.10 The ratio of the cement alumina content to mean particle size of cement plus
slag alumina content to mean particle of slag size versus optimum SO3 level. ........ 72
v
Figure 4.11 Total amount of aluminate and ferrite to mean particle size of cement and
the ratio of alumina content of slag to the mean particle size of slags versus
optimum SO3 level .................................................................................................... 73
vi
Abstract
Optimization of sulfur trioxide (SO3) content is typically conducted for plain cements
(OPC-based) paste or mortar systems using compressive strength testing and isothermal
calorimetry for quality control. The objective of these procedures is to ensure adequate strength
and set properties by controlling the time of occurrence of the aluminate peak by adjusting the SO3
content by incremental addition of sulfate in the form of hemihydrate at three or more levels, at
the same time attempting to maximize either strength or heat of hydration at 1 and 3 days. The
inclusion of supplementary cementitious materials such as blast furnace slag at high replacement
levels in concrete mixtures is common. The incorporation of slag, typically with higher alumina
contents than OPC, can additionally impact the sulfate consumption as they are expected to
influence the reaction of the aluminate phase at an early age. Additionally, slags have variable
alumina content ranging between 7% and 18%, which can further influence this tendency.
This study investigates the influence of fineness, C3A, C4AF content of cement, and Al2O3
content of slag based on compressive strength and isothermal calorimetry testing at one and three
days. A matrix consisting of four cements with variable fineness, C3A, C4AF content of cement,
and two slags with low and high alumina contents was investigated for optimum SO3 content. A
single replacement level of 50% by mass of slag was used. The results suggest the strong influence
of C3A, C4AF content of cement, and Al2O3 content of slag and mean particle size and fineness of
both cement and slag on optimum SO3 content based on the heat of hydration and compressive
vii
strength testing. An expression incorporating these parameters is suggested for the optimization of
SO3 content for use in slag-blended systems.
1
Chapter 1: Introduction
Supplementary cementitious materials (SCMs), which are natural materials or industrial
byproducts, have been intensively used in the construction industry for decades since
sustainability, performance, and carbon footprint have been gaining global attention [1]–[4]. SCMs
are used as a partial replacement of Portland cement to improve both fresh and hardened concrete
properties [5], [6]. One of the commonly used SCMs around the world is ground granulated blast
furnace slag (GGBS). Slag manufacture and utilization in the concrete industry have been
consistently increasing [7], [8]. Since slag substitutes ordinary Portland cement (OPC), it can
provide increased durability, workability, higher ultimate strength, cost efficiency, and reduced
CO2 emissions [9]–[14]. However, slag reacts relatively slowly than OPC, and it demonstrates
lower early age strength development. Even though the levels of replacement are restricted,
contributions of slag into concrete and construction are expressed in many research reports [10],
[13]–[19].
From a historical perspective, the determination of sulfate content in cements and blended
cements have become a crucial problem for researchers. Lerch [20] conducted the first most
comprehensive research on sulfate optimization of cement paste and mortar. Researcher optimized
sulfate content of cement based on compressive strength, heat of hydration (HOH), and also with
length change of mortar bars stored in water. The researcher demonstrated how compressive
strength and HOH present virtually identical results on sulfate optimization.
2
Therefore, Lerch [20] has been a pioneer and inspired by many studies on sulfate
optimization of cement and cementitious systems. Sandberg and Bishnoi [21] displayed sulfate
optimization of calcined based LC3 cement using isothermal calorimetry. Researchers steadily
increased sulfate amount in LC3 cement, measured the heat of hydration (HOH) via calorimetry,
and selected the maximum HOH at the desired age.
Mei-Zue and et al. [22] focused on the optimization of sulfate in different blended
cementitious systems. Researchers' experimental results showed that different content of blended
cementitious systems needed different amounts of sulfate content for sulfate optimization.
Therefore, different gypsum contents such as calcined gypsum, which have different solubility and
activity, were influenced by sulfate optimization.
Another inspirited research from Lerch's work was "on optimization of sulfate level on fly
ash-OPC system" studied by Niemuth [23]. Even though there was no specific standard to
determine sulfate optimization on blended cement systems, the researcher also used compressive
strength test and isothermal calorimetry to determine sulfate demand on fly ash-OPC systems in
his dissertation and paper [23], [24]. The researcher demonstrated that fly ash demand increased
in optimum sulfate in the OPC-fly ash system. The researcher examined in detail the effect of
sulfate optimization on the OPC-fly ash system, which not only depends on the physical properties
of fly ash, like particle size but also depends on the chemical properties of fly ash.
Mohammed and Safiullah [25] studied and optimized sulfate content of Algerian Portland
cement. The authors' conclusions demonstrated that qualities of cement paste and mortar such as
setting time, strength, drying shrinkage, degree of hydration, and so on, were unfavorably affected
when calcium sulfate dihydrate was added below or above the optimum of sulfate content in
cement.
3
As detected by Lerch [20] and other researchers [22]–[25], there are many certain factors
affecting sulfate demand on cement and cementitious systems. Tricalcium aluminate (C3A),
fineness, and alkalis are the most well-known factors affecting the sulfate demand for cement. On
the other hand, the incorporation of slag, typically with higher alumina contents than ordinary
Portland cement (OPC), can additionally impact the sulfate consumption as they are expected to
influence the reaction of the aluminate phase at early ages. Besides, higher alumina slag content
can induce more rapid slag hydration in a cement-slag system [26]. Furthermore, Kocaba's research
[10] proved that slags favored the hydration of the ferrite (C4AF) phase in cements. Therefore,
C4AF in cement and alumina content in slag also have been taken into consideration on sulfate
optimization of slag-cement systems in this research.
Isothermal calorimetry, which is equipped with unique temperature control systems usually
has high repeatability compared to compressive strength tests. The test results can be obtained
without stopping the experiments at any time, in other words, it provides continuous information
[20], [23], [27]. Nevertheless, making specimens for compressive strength tests needs intensively
physical labor, and specimens have to be kept in ambient, temperature-controlled lab room or
cabinet [28]. On the other hand, from past to present compressive strength test has been commonly
used in the concrete industry. Therefore, both compressive strength tests and isothermal
calorimetry were used to determine sulfate optimization in cement-slag systems at one day and
three days in this research.
The objective of this investigation is to illustrate that the substitution of cement with slag
will alter the ideal sulfate level by expanding the sulfate demand. Isothermal calorimetry can be
utilized to foresee the ideal sulfate substance for cement–slag systems, and optimization of the
sulfate level with respect to the cement and slag can develop the early age strength. This study
4
investigated the influence of fineness, C3A, C4AF content of cement, and Al2O3 content of slag
based on compressive strength and isothermal calorimetry testing at 1 and 3 days. A matrix
consisting of 4 cements with variable fineness, C3A, C4AF content of cement, and two slags with
low and high alumina contents was investigated for optimum SO3 content. A single replacement
level of 50% by mass of slag was used. The results suggest strong influence of C3A, C4AF content
of cement, and Al2O3 content of slag and mean particle size and fineness of both cement and slag
on optimum SO3 content based on the heat of hydration and compressive strength testing. An
expression incorporating these parameters is suggested for the optimization of SO3 content for use
in slag-blended systems.
5
Chapter 2: Literature Review
The fundamental principle of ordinary Portland cement (OPC) hydration must be known
to understand how the ideal sulfate level of Portland cement is determined by slag. There is also a
need to understand slag properties, how slag reacts, and how slag can interact with cement
hydration. Reviews on the optimization of sulfate in cement and cement with other supplementary
materials are given. All this knowledge is then used in the rest of the paper to build the experiments
and the hypotheses.
2.1 General Overview of Portland Cement
Portland cement (PC) clinker is the process of burning raw materials containing a sufficient
amount of lime (CaO), silica (SiO2), alumina (Al2O3), ferric oxide (Fe2O3), and lower amounts of
oxides (MgO, Na2O, K2O and so on) at 2600-2900 °F. Limestone (CaCO3) in the raw mixture's
calcareous portion exposes lime to disintegration at about 900 °C. The lime reacts with silica,
alumina, and ferric oxide from the raw mixture's argillaceous element to produce clinker minerals
[2], [19], [29]. The clinker consists of four main minerals, namely alite (3CaO•SiO2), belite
(2CaO•SiO2), tricalcium aluminate (3CaO•Al2O3), and ferrite phases (4CaO•Al2O3•Fe2O3).
Typically, the industrial Portland cement types do not contain such minerals as pure compounds.
However, they also have varying concentrations of extraneous ions in their crystal structure. The
clinker is typically blended with about 5% gypsum to get the ordinary Portland cement [2], [19],
[29], [30].
6
The compounds found in cements are usually classified as the total amount of their oxides
despite not being present in the blends. In general, the cement's chemical compositions are
indicated by the shortened formulas in which the oxides are represented as single-letter shorthand
notations, as shown in Table 2.1 [2], [29]. Therefore, tricalcium silicate can be identified as C3S
by using abbreviations of cement oxides. Various other components are produced in the cement
hydration, which contains more complex standard chemical formulations. The use of shorthand
notation easily expresses them [2], [29]–[31]. For instance, Ca6Al2S3O50H64 is a typical chemical
formula of ettringite. It is a hydration product of aluminate and gypsum. When the conventional
chemical formulation of ettringite is divided into oxides, the ettringite can be expressed as
6CaO•Al2O3•3SO3•32H2O, which can be shortened as C6AS̄3H32 [19], [31].
Table 2.1 Typical ordinary Portland cement oxide composition
Common Name Oxide Notation
Lime CaO C
Silica SiO2 S
Alumina Al2O3 A
Ferric Oxide Fe2O3 F
Sulfur Trioxide SO3 S̄
Water H2O H
Carbon Dioxide CO2 C̄
Magnesia MgO M
Alkalis Na2O N
K2O K
7
Table 2.2 indicates the main components of the typical OPC. The total quantity of calcium
silicates (C3S and C2S) is about 75% as OPC's mass fraction. It is necessary to note that the main
compounds of the OPC do not constitute entire cement compositions. These missing components
contain a minor quantity of impurities [29], [32].
Table 2.2 Typical ordinary Portland cement composition (Adapted from [2], [29])
Chemical Name Chemical Formula Notation Weight (%)
Tricalcium Silicate (Alite) 3CaO•SiO2 C3S 55
Dicalcium Silicate (Belite) 2CaO•SiO2 C2S 21
Tricalcium Aluminate (Aluminate) 3CaO•Al2O3 C3A 11
Tetracalcium Aluminoferrite (Ferrite) 4CaO•Al2O3•Fe2O3 C4AF 8
Calcium Sulfate Dihydrate (Gypsum) CaSO4•2H2O CS̄H2 5
2.2 Ordinary Portland Cement Hydration
Once OPC is blended with water, several chemical reactions occur. The entire reaction is
termed as "hydration". Besides, "hydration products" are typically referred to as the new material
formations consisting of hydration. It is important to remember that cement is a multi-component
system. Thus, hydration of cement is a very complicated operation involving several successive
and concurrent reactions. In addition to the four main cement clinker compounds, free lime,
CaSO4, Na2SO4, and K2SO4 are also involved in the cement hydration [2], [19], [29], [33], [34].
The description of cement hydration is typically based on the individual hydrations of the
main cement compositions. Each compound of cement was considered to be hydrated
independently of other cement phases. Even though these cement compounds' interactions with
each other, as well as minor cement phases, influence both the process and the kinetics of
8
hydration, it helps to understand and provides much valuable knowledge about how OPC hydrates
[2], [29].
2.2.1 Calcium Silicates Hydration
The two calcium silicates, which are the cement's main components, have almost identical
stoichiometric hydration reactions. Except for the amount of calcium hydroxide produced (aka
portlandite or CH). The stoichiometric equations for both hydrations of calcium silicate are shown
as [2], [29] :
2C3S+ 11H (Water) → Calcium Silicate Hydrates (C-S-H) + 3CH Eq. (2.1)
2C2S+ 9H (Water) → Calcium Silicate Hydrates (C-S-H) + CH Eq. (1.2)
The main hydration product is the "Calcium Silicate Hydrates" (C3S2H3); in other words,
C-S-H. The C3S2H3 is an estimation of C-S-H composition with complete alite hydration. The
composition of C– S–H considerably variable. This change depends on the water/cement ratio,
ambient conditions, and degree of hydration. The ratio of CaO/SiO2 is “1.50” when hydration is
complete; however, it never reaches “2.00” at any point of hydration. For the full hydration of
OPC, approximately 65% of hydration products of alite is C–S–H gel [2], [29], [35].
Furthermore, C-S-H is the most crucial part of the hydrated paste since more than 50% of
the hydrated paste is in the form of C-S-H gel. However, it is hard to study due to its amorphous
features, compositional heterogeneity, and indefinite morphology. It has uncrystallized, ultra-
small, and non-regular particles. These small particles are mainly studied through electron-optical
methods (e.g., SEM and TEM), but they cannot be resolved fully. On the other side, calcium
hydroxide (CH) particles occupy roughly 20% of the cement paste volume. Unlike C-S-H, CH is
9
a well-crystallized material, and it can be easily examined by simple optical microscopy [2], [19],
[35]–[37].
The C3S hydration kinetics have been evaluated by XRD, which measures the quantity of
the unhydrated C3S over time. Conversely, this approach is not reliable and accurate enough to
ensure the early age of hydration, especially the degree of hydration below 0.1 [29], [38], [39].
Besides, the two reactions of calcium silicates hydration, as well as all cement composition
hydration, are exothermic, so that heat evolution occurs during reactions. The heat of the mortar
and the concrete increase unless it is rapidly released into the ambient. A specialized calorimeter
can follow cement hydration under stable laboratory conditions. It measures the thermal
conductivity rate required to maintain a constant temperature [2], [29]. Figure 2.1 shows the
calorimetric curve that defines the typical heat evolution rate of C3S hydration with time. Several
phases of C3S hydration can be more easily distinguished from this calorimetric curve. These steps
are I: pre-induction (initial hydrolysis) period; II: induction period; III: acceleration period, IV:
deceleration period, and V: steady-state.
10
Figure 2.1 The heat evolution rate of C3S hydration. (Adapted from [2], [29])
An incredibly high C3S hydration rate arises during the "initial hydrolysis" stage, which
only takes a few minutes (about 15-20 min) and slows down to almost null when the "induction
period" begins. O2- ions in the C3S are released to the surrounding liquid phase during interactions
with water to form OH- ions due to protonation. Likewise, SiO44- ions of C3S also join the solution
for hydrogen silicate ions production. The Ca2+ ions are incorporated into the precipitating C–S–
H. The C-S-H gel layer is then positioned on the C3S particle's surface that serves as a block to the
further water interaction with the C3S underlying and further release of OH-, SiO44-, and Ca2+ ions
from the surrounding liquid during the induction phase. Another idea about producing this
immediate C-S-H is that releasing of Ca2+ ions from the C3S particles leading to a silica-rich
surface layer that then absorbs the calcium ions in the surrounding fluid, thereby the thin, nearly
impermeable C-S-H layer covers the C3S particles [2], [29], [33], [39], [40].
0
5
10
15
20
25
30
0.1 1 10 100
Hea
t F
low
(m
W/g
)
Time (h)
C3S Hydration
I II III IV V
11
"The induction period" is the inactivity stage of C3S hydration. That is why PC concrete
stays in the plastic condition for a couple of hours [2], [39]. The duration of the "induction period"
is several hours. There are no specific explanations for the end of the "induction period" and the
beginning of the "acceleration period." Several ideas have been established to describe this
phenomenon. Without moving through deeply into these ideas, a number of compositional and
morphological changes are made in the initially developed C-S-H layer, making it more permeable
or decomposable due to the osmotic pressure produced by the fluid at the interface with the C-S-
H layer and the underlying C3S layer. Therefore, additional hydration is also possible [2], [19],
[29], [33], [39], [41], [42].
The "acceleration period" hydration rate rises quickly and reaches a peak in 5–10 hours.
During this stage, CH's concentration in the solution becomes the highest, and thus the
precipitation begins. At the end of the acceleration period, the C3S reaches the highest hydration
rate, which corresponds to the ultimate heat evolution rate. In the meanwhile, the final set was
completed, and the early hardening just started [29], [39], [40], [43], [44]. Later, as the hydration
occurs, the hydrate layer thickness grows and acts as a barrier to which the water has to flow to
the unhydrated C3S and from which the ions have to disperse to reach the growing crystals.
Ultimately, mass transfer through the C-S-H layer regulates the hydration rate of C3S, and
diffusion controls hydration (deceleration period). Diffusion-controlled reactions are relatively
slow and slower with increased diffusion barrier thickness. As a result, hydration shows a tendency
to reach asymptotically 100% completion (steady-state period) [2], [29], [33], [40], [45].
Even though the C2S hydration mechanism matches up with C3S, it is much less reactive
than C3S. Besides, the heat released by C2S hydration is less than with C3S hydration. The second
12
strong observed peak in the C3S calorimetric hydration curve is scarcely detectable for C2S [2],
[19], [29], [39].
2.2.2 Tricalcium Aluminate Hydration
Tricalcium aluminate's (C3A) hydration in the OPC involves reactions with sulfate ions
(SO42−) typically produced by gypsum dissolution. The main initial C3A reaction is [2]
C3A + 3CS̄H2 (Gypsum) + 26H → C6AS̄3H32 (Ettringite) Eq. (2.3)
Ettringite, a hydration product of aluminate, is stable when adequate amounts of sulfate are
available. If all the sulfate is used before the aluminate has been fully hydrated, in that case,
ettringite converts into the other less sulfate-containing calcium sulfoaluminate hydrate [2], [29].
2C3A + C6AS̄3H32 + 4H → 3C4AS̄H12 (Monosulfoaluminate) Eq. (2.4)
For the ettringite formation, a specific concentration of sulfate ions is needed. For this
reason, sometimes, monosulfoaluminate can be produced even before ettringite if the hydrating
C3A uses the sulfate ions more quickly [39], [46].
These two sets of C3A hydration are exothermic. Ettringite formation delays the C3A
hydration by forming a diffusion barrier all-around aluminate, like the C-S-H actions during the
calcium silicates hydration. The C3A calorimeter hydration curve, therefore, appears qualitatively
similar to the C3S hydration curve, even though the underlying reactions are dissimilar, and the
heat flow is much higher. Its very first heat flow peak is roughly completed in 15 minutes.
However, the second peak time on the calorimeter curve is dependent on the amount of sulfate.
Ettringite stays stable for longer periods with more gypsum or calcium sulfates during hydration.
The conversion process to monosulfoaluminate happens between 12 and 36 hours during most
OPC hydration after all gypsum is used in the ettringite processing [29], [47], [48].
13
Monosulfoaluminate formation occurs because gypsum is not adequate in most cement to
the formation of ettringite from all of the aluminum ions. If monosulfoaluminate interacts with
additional sulfate ions, ettringite can again form. The principal reason for the OPC sulfate attack
is this transforming ettringite from exposed to external sulfate ions source [2], [29].
C4AS̄H12 + 2CS̄H2+ 16H → C6AS̄3H32 Eq. (2.5)
Gypsum (calcium sulfate dihydrate) is added to minimize the initial vigorous C3A reaction
to water, triggering the flash set because of the swift calcium aluminate hydrates formation. These
calcium aluminate hydrates are unstable and eventually transformed into a hydrogarnet (C3AH6)
[2], [33], [49].
2C3A+ 21H → C4AH13 + C2AH8 Eq. (2.6)
C4AH13 + C2AH8 → 2C3AH6 + 9H Eq. (2.7)
This transformation is identical in "high-alumina" cement. However, the reactions happen
so quickly (simultaneously, a high heat liberation rate leads to a high temperature in the paste), the
cement paste cannot gain considerable strength. This lack of the gypsum (or any other form of
calcium sulfates) causes the flash setting. Although gypsum is present in the cement, the sulfate-
free hydrates form might not be eliminated, even it may lead to a flash set when aluminate is highly
reactive [2], [19], [29], [39], [49].
2.2.3 Tetracalcium Aluminoferrite Hydration
The tetracalcium aluminoferrite (ferrite) hydration forms are similar to those of aluminate
hydration, even without gypsum. However, the ferrite reactions are slower, and the heat flow rate
is lower than that of aluminate. Ferrite rarely hydrates fast enough to induce flash setting, and
gypsum slows down ferrite hydration much more dramatically than aluminate does [2], [29], [50].
14
The variations in ferrite composition affect only the hydration rate; as the ratio of Al2O3/Fe2O3
(A/F) decreases, the hydration rate is slower. Iron oxide or ferric oxide tends to have a similar
function as alumina (Al2O3) during hydration in that iron oxide can be used to replace alumina in
the hydration products [2], [29], [42], [50].
C4AF + 3CS̄H2+ 21H → C6(A, F)S̄3H32 +(F,A)H3 Eq. (2.8)
C4AF + C6(A, F)S̄3H32 + 7H → 3C4(A,F)S̄H12 + (F,A)H3 Eq. (2.9)
Practically, the cements which are low in aluminate but high in ferrite content have been
shown to be far more sulfate resistant. This indicates that there is no ettringite formation by
monosulfoaluminate, as in Eq 2.5. The explanation has not been explicitly stated; " it may be that
the iron-substituted monosulfoaluminate cannot react to the ettringite formation" [2], [19], [48],
[50].
2.2.4 Portland Cement Hydration
As the anhydrous cement is blended with water, it begins to dissolve form hydrated
compounds of significantly lower solubility, precipitating the solution to be deposited into areas
initially filled with water. The approximate degree of hydration for the early days is
aluminate>alite>ferrite>belite [2], [29], [35], [48]. Nevertheless, it should be noted that these
cement compounds cannot hydrate entirely at a similar rate in the different cements or under other
conditions. The hydration of ordinary portland cement (OPC) is influenced by a variety of factors,
such as; chemical oxide and mineralogical compositions of the cement, fineness and particle size
distribution (PSD) of the cement, role and volume of chemical admixtures and gypsum, the ratio
of water to cement (w/c) in the mixture, ambient conditions (i.e., temperature, humidity), and the
number of foreign ions [2], [19], [29], [33], [39].
15
It must be considered that the hydration of Portland cement is far more complicated than
that of the individual phases because interactions with one another and with other minor oxides
will modify both the hydration mechanism and kinetics. The presumption that cement compounds
are independently hydrated is plausible but not necessarily correct for most purposes [2], [19],
[29], [33]. For instance, both ferrite and aluminate react with sulfate ions. Aluminate is, however,
more reactive than ferrite and consumes much more sulfate than ferrite. On the other hand, gypsum
raises the hydration rate of calcium silicates (especially alite), which supplies sulfate during
hydration [2], [51], [52]. The presence of optimal calcium sulfate contents in a cement paste is also
a perfect example for developing the highest strength [20], [22], [53]. One of the theories is that
an excessive level of gypsum content leads to too much ettringite forming after the paste is
hardened, provoking uncontrolled expansion and destruction of the paste's microstructure. On the
other hand, a low amount of gypsum may not react with all aluminate in the cement, and aluminate
reacts with calcium hydroxide, which supplies free lime to reaction, and monosulfoaluminate
solution. Lime consumption causes delay alite hydration because of preventing the nucleation of
alite hydration. Additionally, low amounts of gypsum may cause a "flash set" due to the sudden
aluminate hydration in the paste [2], [20], [29], [33], [35], [52], [54].
2.2.4.1 Heat of Hydration
The heat of hydration (HOH) is the enthalpy of the reaction. All cement hydration steps,
which are exothermic reactions, are followed by the heat release. The HOH is a measure of heat
generated by each reacting unit mass of the anhydrous compound. In other words, HOH indicates
the system's remaining energy after the energy redistribution has happened when chemical bonds
have broken and formed during hydration [19], [34], [55], [56]. It is typically denoted as kJ/mole,
16
although the term mole cannot be used because OPC is a multi-component material. HOH can
either be estimated by calculation (see Table 2.3) or determined experimentally using an
isothermal calorimeter. The Portland cement hydration consists of five main steps that are the same
as those of alite hydration. The standard progress of OPC hydration in terms of the heat evolution
rate is demonstrated in Figure 2.2 [2], [29], [33], [35], [57].
Table 2.3 The heat of hydration of pure cement clinker minerals (Adapted From [2], [29], [33])
Cement
Compounds Reaction Product
Heat of Hydration (J/g)
3 days 1 year Full Hydration
C3S C-S-H + CH 240 490 520
C2S C-S-H + CH 50 225 260
C3A C6AS̄3H32 880 1160 1670
C3A C4AS̄H12 N/A N/A 1140
C4AF C3(A, F)H6 290 375 420
Figure 2.2 The heat evolution rate of Portland cement hydration. (Adapted from [2], [29], [33]).
0
0.5
1
1.5
2
2.5
3
3.5
0.1 1 10 100 1000
Hea
t F
low
(m
W/g
of
cem
ent)
Time (hours)
Cement Heat Flow Rate
Stage1 Stage 2 Stage 3 & Stage 4 Stage 5
C3A
Hydration
C3S
Hydration
17
The HOH of cement is roughly the measured total amount of heat evolved by its
compounds. Table 2.3 demonstrates the amount of heat liberation in individual cement compounds
hydration at specific ages. Approximately 50% of the overall HOH is generated in the first three
days, and 70% is liberated within seven days [29], [35]. Intense heat release occurs within a couple
of minutes due to the quick hydration of C3A and C3S during the pre-induction stage. On the other
hand, the heat flow rate declines to a minimal level during the induction period. Later, the second
peak becomes visible within 8 to 17 hours of hydration owing to the C3S hydration, which produces
CH and C-S-H [29], [55], [58], [59]. Then the heat evolution slows down, and within a few days,
heat liberation drops to a relatively low level. For some cements, the shoulder after the second
peak, which is typically due to the renewed formation of ettringite (C3A hydration), can appear
during the deceleration period of the rate of heat evolution curve. These characteristics are shown
in Figure 2.2 [2], [29], [35], [42].
2.3 Supplementary Cementitious Materials
Around the world, scientists, engineers, and construction specialists are striving to develop
robust and durable concrete. Their main goal is to obtain better concrete, which has high strength,
workability, durability, and performance. Moreover, they aim to protect concrete from the adverse
effects of aggressive environmental conditions. Supplementary cementitious materials (SCMs),
such as fly ash, silica fume, metakaolin, and blast furnace slag, are incorporated purposely to help
obtain particular properties of concrete and mortar cement. Another objective of the SCMs
addition in the cement and concrete systems includes the economic benefits and environmental
concerns [13], [14], [29], [60]–[63]. Additionally, the use of SCMs saves energy and has ecological
impact due to reduced greenhouse gas emissions as a result of decreased ordinary Portland cement
18
manufacturing. For concrete mixes, SCMs are also used to minimize cement content, enhance
workability, increase the sustainability of the concrete over a long period, maximize strength,
facilitating cost reduction of construction, and increase performance by hydraulic or pozzolanic
reaction with cement. The use in the concrete of these by-products not only prevents them from
being land-filled but also improves the properties of concrete in the fresh and hardened states [2],
[13], [14], [19], [23], [29], [35], [64], [65].
According to ACI Committee 116 and ASTM C125 the definition of pozzolan is "A
siliceous or siliceous and aluminous material, which in itself possesses little or no cementitious
value but will, in finely divided form and in the presence of moisture, chemically reacts with
calcium hydroxide at ordinary temperatures to form compounds possessing cementitious
properties" [29], [56], [64], [66], [67]. In other words, pozzolans are materials that, when mixed
with the calcium hydroxide components, yield cementitious properties. Most of the pozzolans
quickly react with calcium hydroxide (CH). They could be individually mixed and react with CH,
but when they are blended with cement, they react with the CH produced by the cement hydration
[2], [29], [35], [67]. Amorphous silicate, which is the main component of pozzolanic material, and
alumina react with CH. This pozzolan reaction with calcium hydroxide is called a "pozzolanic
reaction" and can be written as simple as [2], [29] :
Ca(OH)2 + SiO2 + H2O → Calcium Silicate Hydrates (C-S-H) Eq. (2.10)
Ca(OH)2 + Al2O3 + H2O → Calcium Aluminate Hydrates (C-A-H) Eq. (2.11)
Plus, aluminum can be integrated into the existing C-S-H. The C-A-S-H is a type of C-S-
H which has been replaced by aluminum (Al) and develops after the presence of aluminum ions
(Al3+) from OPC. Reactive Al3+ ions arise when additional cementing materials, such as slag and
fly ash, have been dissolved into the water [2], [19], [35], [68].
19
Hydraulic cement or hydraulic cementitious materials harden and set by chemical reaction
with water, for instance, ordinary Portland cement (OPC), high-lime fly ashes (HLFA), and slag
are hydraulic cements [56], [66]. Nevertheless, the reactivity of hydraulic cementitious material is
comparatively lower than OPC, and they generally require activators to trigger and accelerate the
hydration [14], [19], [29]. They may be chemically activated by the addition of alkali hydroxides
(like calcium hydroxide, sodium hydroxide, and potassium hydroxide), by adding the sulfates (e.g.,
gypsum, hemihydrate, anhydrite), or more frequently by the additive of lime-producing materials
(such as lime or OPC). It could be noted that hydraulic cementitious materials can substitute OPC
up to a considerably greater extent than those exhibiting pozzolanic behavior [39], [64].
The Figure 2.3 represents the typical variability and composition for most widely used
categories of supplementary cementitious materials and OPC in the CaO–Al2O3–SiO2 ternary
diagram [10], [29], [64], [65].
Figure 2.3 Ternary diagram of the chemical composition of common SCMs and Portland cement
(Adapted from [10], [64], [69], [70])
20
2.3.1 Silica Fume
Silica fume that also referred to as "condensed silica fume" and "micro silica" is a
byproduct of the production of elemental silicon and ferrosilicon alloys [56], [71]. Particles in the
silica fume are exceedingly small, with more than 95 percent of particles smaller than 1 micrometer
[13], [29]. Since its particles are remarkably fine, the surface area of the silica fume is vast. The
approximate surface area of particles from silica fume ranges between 15000 m2/kg and 30000
m2/kg [14], [29], [64], [72]. According to ASTM C1240, it has to contain more than 85% SiO2.
The chemical composition of silica fume is straightforward and typically contains more than 90%
of SiO2. The amount of other components, such as Fe2O3, K2O, Na2O, Al2O3, MgO, and CaO, are
usually less than 1%, so it practically has not so many variations in its chemical composition [2],
[19], [29], [35], [64], [71]. Since silica fume has extreme fineness, high surface area, and it has a
large amount of noncrystalline silica content, silica fume is a highly reactive pozzolanic material
[13], [14], [67].
Because of the silica fume's extreme fineness and high surface area, it has a filler effect.
Two main factors tend to contribute to the filler effect. First, if the material does not give a reaction,
the w/cm ratio is getting an increase, and it creates extra space for the hydration products of the
cement phases. Second, especially fine materials like silica fume, has been promoting hydration
rates due to their high surface area act as "nucleation sites" for the hydration products of the cement
compounds [13], [19], [63].
Silica fume influences cement hydration and the hardening process. The availability of
silica fume in OPC increases the hydration of calcium silicates [13], [14], [19], [67], [73]. It has
been observed that silica fume enhances tricalcium aluminate (C3A) hydration rate in some cases,
but generally, it delays C3A hydration due to "the electrostatic interaction" [74]. Furthermore, as
21
the OPC starts to react chemically, it begins to release calcium hydroxide as a result of calcium
silicates hydration. The silica fume reacts with this calcium hydroxide to form additional binder
material called calcium silicate hydrate (C-S-H), which is quite similar to the form of OPC [13].
Moreover, it is observed that the calcium-ions concentration decreases rapidly when silica fume is
examined in pore solution concentration. At the same time, the hydroxide ions reduce too. Given
that silica fume lacks calcium content, it is highly efficient when it comes to the consumption of
calcium hydroxide (CH), which is a product of cement hydration [63], [67], [75], [76]. Similarly,
using the same levels of replacement, the calcium hydroxide amounts in the cement paste that
contains silica fume contents are relatively low compared to the paste contents in fly ash or slag.
When silica fume is used in concrete, it inclines to reduce bleeding and segregation [13],
[67]. Likewise, it increases the stickiness and cohesiveness of the fresh concrete mixture. Other
benefits associated with the utilization of silica fume in concrete include enhancing the resistance
of the concrete against corrosion, reducing the alkali-silica reactivity, lowering permeability to
chloride and water intrusion, and thus improving the concrete durability [14], [29], [64], [67],
[77]–[80]. Therefore, concrete's physical and chemical characteristics can be developed by silica
fume. Furthermore, silica fume contributes to the developing mechanical properties of mortar and
concrete. Mechanical features of hardened concrete, which are tensile strength, compressive
strength, flexural strength, torsional strength, modulus of elasticity, and toughness improve when
silica fume utilizes in the concrete and mortar [29], [63], [64], [72], [81]–[86]. It also develops the
early strength gain of concretes.
22
2.3.2 Metakaolin
The term metakaolin is a combination of two words. The first term is "meta", which is a
prefix to denote change. The second term is "kaolin", which is a type of clay. In the production of
metakaolin, kaolin acts as the primary raw material. Metakaolin (Al2O3.2SiO2 or AS2) is a
dehydroxylated form of the clay mineral kaolinite [13], [14], [87]. Kaolin (Al2Si2O5(OH)4 or
Al2O3.2SiO2.2H2O) is a clay mineral that is phyllosilicate [14]. The production of metakaolin
involves the implementation of heat to dihydroxylation. This thermal activation of a mineral is
also referred to as "calcination". Heating dehydroxylation is an ongoing process and occurs in
many stages. In essence, dehydration is a reaction that results from the decomposition of kaolinite
to a disordered structure [14], [77]. The clay loses most of its water when the temperature is
between 100 ℃ and 200 ℃. After that, when the temperature is between 500 ℃ and 800 ℃,
heating kaolinite to such temperatures has an effect on its chemical structure and component [14],
[88]. Beyond the dehydroxylation temperature, kaolinite keeps the crystal structure in two-
dimensional order [13], [14]. This product is referred to as metakaolin (MK). The essential way
for producting MK for use as pozzolans or SCM is to reach complete dehydroxylation as close as
possible, without overheating [14], [89].
As cement is partly substituted with MK, it reacts with CH. This reaction result is extra C-
S-H gel and in conjunction with crystalline materials, which involve alumosilicate hydrates and
calcium aluminate hydrates (C2ASH8, C4AH13, and C3AH6). The metakaolin hydration reaction
depends on the type of cement, the Metakaolin/CH ratio, the amount of free water, and the
temperature at which the reaction will occur [14], [67], [90]–[94]. The chemical reactions are given
below, taken from the Murat's paper [91];
23
AS 2 + 6CH + 9H C4AH13 + 2C-S-H (AS2 /CH = 0.5) Eq. (2.12)
AS 2 + 5CH + 3H C3AH6 + 2C-S-H (AS2 / CH = 0.6) Eq. (2.13)
AS 2 + 3CH + 6H C2ASH8 + C-S-H (AS2 / CH =1) Eq. (2.14)
The use of metakaolin in construction activities eliminates the problems that arise from the
use of Portland cement. Besides, the use of metakaolin results in high efficiency of cement
production. Metakaolin improves the compressive strength of mortar and concrete due to its filler
effect and pozzolanic reaction. Metakaolin particles are remarkably finer than OPC particles. It
creates extra space for the hydration products of the cement phases, so it accelerates cement
hydration [14]. Besides, metakaolin is a highly pozzolanic substance that has contributed to the
development of higher early strength concrete. On the other hand, several variables depend on the
strength of metakaolin concretes. The fineness of metakaolin particle size, quality of MK (time
and temperature influence of kaolin activation), and the approach of mixing design are some of
the well-known effects on the compressive strength of metakaolin concrete [13], [77], [93], [95].
Brooks and Johari investigated the effect of metakaolin on concrete. They found that
autogenous shrinkage decreased at an early age with the rise of metakaolin content; however, it
increased in the long-term with the raise of metakaolin quality [14], [96]. Moreover, according to
Wild et al., chemical shrinkage increased for additional MK content in concrete, and it reached a
maximum rate of around 15% MK. Then, it declined dramatically for higher metakaolin content.
Furthermore, MK inclusion minimized both the overall and the main creep of concrete, as well as
the drying shrinkage. Also, creep decreased at a higher substantial rate of MK in the concrete [13],
[89], [92]. On the other hand, the incorporation of metakaolin in cement paste led to the refinement
of the pore structure. Metakaolin content in the cement paste causes a reduction in the size of pores.
The admixture reduces the pore diameter and increases the number of micro-pores in the concrete
24
material [97], [98]. Besides, the ability of metakaolin to reduce the expansion of concrete is
attributed to alkali reaction and external attacks from the sulfate [79], [99]. Moreover, on the
mortar and concrete samples researcher also tested the chloride ion migration, which gives
information about the resistivity of concrete against chloride penetration. From the experiment, it
is clear that the hydration phase reacts with the chloride ions from anodic solutions to form
"Friedel's salt" (Ca2Al(OH)6(Cl, OH)·2 H2O ). The salt has a chloride ion binding ability that can
block the chloride ions during the migration test of chloride [14], [98], [100], [101]. Thus,
metakaolin is more durable, has increased resistance to chemical attack, and reduces permeability
[102].
2.3.3 Fly Ash
Fly ash or pulverized fuel ash is one of the most commonly used pozzolans in the concrete
industry. Fly ash (FA), which is a by-product of thermal power generating stations, is obtained by
burning coal. The component is finely divided into the residue [56], [66], [103]. The physical and
chemical characteristics of FA and the type of fly ashes depend on many factors. Some of those
critical factors are type and mineralogical composition of coal and conditions of combustion [13],
[14], [29].
The particle size of fly ash is smaller than OPC. The fly ash particles are ordinarily
spherical and range from less than 1 µm to larger than 100 µm is size [29]. Fineness is one of the
most significant properties that contribute to the FA's pozzolanic reactivity. When the fineness of
FA decreases and the surface area increases, the pozzolanic reactivity rises, as they are regulated
by surface diffusion [13], [14], [104].
25
Fly ash is classified according to ASTM C618 [103] into two classes as Class C and Class
F fly ashes. Fly ash majorly consists of silica, alumina, calcium content (such as lime), magnesia,
and various iron oxides. If the composition of FA contains at least 70% "silicon dioxide +
aluminum oxide + iron oxide", it is called Class F fly ash. On the other hand, compositions of
Class C ash contain more than 50 percent of these phases. Class F ashes are usually produced from
anthracite or bituminous coal; however, class C ashes are generally derived from lignite or
subbituminous coal [13], [14], [19], [29], [35], [67], [103]–[105]. Class C fly ashes almost always
have higher CaO content than class F ashes [64]. Furthermore, fly ashes consist of a mixture of
identifiable phases and a mixture of glassy particles. Generally, fly ashes have 50–85% amorphous
phases due to rapid cooling [13], [29]. Class C fly ashes contain more substantial amounts of
crystalline phases. Examples of major identifiable crystalline phases include mullite, quartz,
magnetite, and various iron oxides (such as hematite). Besides, class C fly ash may contain
anhydrite, alkali sulfates, periclase, melilite, belite, aluminate, and so on [14], [23], [67], [106].
Notably, high calcium FA has self-hardening properties due to some FA crystalline phases
reacting with water [14], [19], [35]. These hydration products, which are monosulfoaluminate,
ettringite, and calcium silicate hydrate (C-S-H), provide the hardening of FA [64]. On the other
hand, low calcium FA barely has self-cementing characteristics. Amorphous phases of FA give
reaction with CH, which is a by-product of the hydration of OPC, and alkalis in the OPC-FA
mixture [13], [14], [29]. Mainly, the degree of FA hydration enhances with more CH contribution
in the mix [107]. CH reacts with the noncrystalline alumina and silica in FA [14], [108]. These
reactions are fundamentally pozzolanic. The reaction of FA with OPC is influenced by the intrinsic
and external features of the FA, such as mineralogical and chemical structure, particle size, and
volume of amorphous content. Moreover, heat treatments and chemical admixtures influence the
26
reactivity of pozzolans [14]. The pozzolanic reaction of FA with CH produces further C-S-H [64].
They are time-dependent reactions; however, they are the same properties as the products of OPC
hydration [13], [14]. Thus, this extra C-S-H improves concrete strength [23], [107].
Fly ash has many benefits when it is used in cement and concrete industries. First of all,
segregation and bleeding are greatly minimized by using FA in the concrete; thereby the
pumpability of concrete improves [13], [14], [29]. Fine and spherical particles of FA impact the
pastes and mortars rheological properties [13], [109]. These properties of FA considerably improve
workability, which reduces the amount of water needed in the mixture. For this reason, fly ash in
the fresh concrete mixture can be more appropriately placed, finished, and compacted than the
regular concrete mixture [13], [14], [29], [109]–[111].
Furthermore, hydration of OPC paste is followed by heat release, which increases the
concrete's temperature. Due to slower pozzolanic reactions, the partial substitution of cement by
fly ash results in heat release for a prolonged time, and the mixture temperature stays lower
gradually [23], [64], [67], [112]. This slow heat of hydration rate is particularly significant in the
mass concrete, where massive temperature changes will result in cracking [113]. This was
primarily used to regulate the increase in temperature in the mass concrete construction [14], [35],
[113]. In general, Class F fly ash is in the tendency to lower the rate of temperature change rather
than Class C fly ash [13], [29].
Moreover, the permeability of concrete is critical for the determination of the levels of
mass transport associated with the disruptive chemical activity [13], [29]. Besides, this has a vital
role in assessing the rate of improvement of the concrete degradation caused by the chemical attack
caused by sulfates, acids, chlorides, harsh environmental conditions, and organic agents [14]. The
concrete's permeability mainly depends on the dimension, distribution, and consistency of the
27
concrete's hydrated paste pores. The degree of hydration and water to cement ratio essentially
influences the cement paste pore structure [13], [14]. Because of the development of pozzolanic
reaction products accompanying the cement hydration, it converts larger pores into smaller ones
[13]. Therefore, FA-OPC mixture decreases the permeability, so durability and resistance against
different modes of degradation are improved by fly ash [35], [114]. Thus, FA in the mix provides
more excellent protection to sulfate attack and corrosion [63], [115].
Fly ash's pozzolanic reactions have many features which influence of the concrete strength.
The pozzolanic reaction consumes lime, and the production of reaction may effectively fill and
subdivide pores [13]. However, the pozzolanic reaction is relatively slower than OPC hydration,
so the rate of concrete's strength improvement is slow [14], [29]. Type of FA and cement, the
composition of FA, the particle size of FA, replacement level of OPC with FA, curing conditions,
and temperature affect the strength improvement of fly ash concrete [13], [14], [29], [67], [114].
Notably, high calcium FA has self-hardening properties due to some FA crystalline phases give
reaction with water [67]. This hydration provides the strength on the FA concrete [13]. On the
other hand, low calcium FA barely has self-cementing characteristics. Amorphous phases of FA
give reaction with CH, which is a by-product of the hydration of OPC [13], [14]. Therefore, Class
F fly ashes do not display any pozzolanic meaningful reaction at early ages. Such fly ashes have
demonstrated petty slow levels of strength improvement [13], [29]. This may cause a decrease in
strength at an early ages; however, fly ash in concrete gains strength later ages [14]. On the other
hand, Class C fly ash will begin to contribute to strength improvement as early as after mixing,
due to its self-hardening and pozzolanic properties. It also helps to improve early strength
development in the fly ash concrete [13], [14], [35], [67].
28
2.4 Ground Granulated Blast Furnace Slag
Ground granulated blast slag (GGBS) or ground granulated blast furnace slag (GGBFS)
comes from the manufacture of pig iron. GGBS is a secondary product of iron production [19],
[116]–[118]. The component forms of the agents including coke or fuel ash, iron ore, and limestone
added to the blast furnace [13]. The furnace was run at approximately 1500 ℃ [119]–[121]. When
components melt in the furnace, molten iron and slag are produced—molten slag forms in the
process of reducing the iron ore to iron [2], [120], [122]. This slag generally contains silica and
alumina, which come from main iron ore, and also it contains magnesia and lime that are from
limestone [2], [10], [14], [123]. Besides, there is a low amount of sulfur and manganese in the slag.
Its exact content varies, but silica (SiO2) and lime (CaO) are the most dominant phases [13], [121].
Furthermore, the molten slag becomes lighter, and it floats on the surface. After it forms,
the next step is to separate molten slag from liquid metal and cool it [13]. Ground granulated blast
furnace slag results from the molten slag quenched in water [66], [123]. Therefore, this method
produces a large amount of noncrystalline content, depending on the cooling levels [122]. The fast
cooling provides the small crystals, and granular slag contains 90% amorphous calcium-
aluminosilicates [121], [124]. Next, it is ground to obtain powder with the same fineness as
cement. The end product is named GGBS or simply slag [10], [13], [65]. In this case, it is classified
as inert hydraulic material. This means that GGBS has cementitious and pozzolanic properties
[10], [14], [64].
Slag can be used as weight-by-weight direct substitute for OPC. Slag substitution rates
range between 30% and 85% [13], [36]. Typically, slag is widely used 50% in many situations
[13]. Higher replacement rates are used in unique and advanced systems such as under the
aggressive environmental conditions and to minimize hydration heat [2], [10], [13], [29], [108].
29
Since slag substitutes OPC, it can increase durability, workability, higher ultimate strength, and
provide cost-efficiency [2], [10], [13], [14], [35]. Slag-cement blended systems reduce CO2
emissions, save energy, and preserve natural sources. Thus, slag-cement blended systems improve
the sustainability and performance of concrete and reduce the carbon footprint [9]–[14], [29],
[121].
2.4.1 Physical, Mineralogical and Chemical Characteristics of Slag
Slag is a fine amorphous material. OPC specific gravity is higher than GGBS, but slag is
finer. The fineness and surface area of slag are significant factors impacting the degree of reaction
and concrete strength. GGBS particles, which size varies from 10 µm to 45 µm, lead to increased
strength of the mortar and concrete. Larger particles of GGBS have less to no activity. Therefore,
ASTM C989 [125] limits a particle size by more than 45 µm. Besides, the Blaine surface area of
GGBS ranges from 400 m2/kg to 600 m2/kg to ensure sufficient production of the mortar and
concrete strength [10], [13], [14], [29], [35], [36].
The GGBS has nearly whole glassy phases; for this reason, a small part of the slag has
crystalline phases. The crystalline phase of the slag consists of low-active minerals and the medium
of hydraulic. These small parts of the slag's crystalline phases are melilite, which is a solid solution
of gehlenite and akermanite, merwinite, calcite, quartz, rankinite, diopside, oldhamite, pseudo-
wollastonite, and monticellite [10], [14], [29], [36], [121].
The slag consists of crystalline and glassy phases, which contain almost 90% amorphous
calcium-aluminosilicates. This glassy-amorphous phase controls the cementitious properties of
slag [36], [117]. Furthermore, GGBS is a nonmetallic material, and there is a large varying
chemical composition of the GGBS. Slag generally contains silica (25-40%) and alumina (6-25%),
30
and also it contains magnesia (<18%) and lime (28-43%) [10], [19], [64], [125]. Besides, there is
a low amount of many components and oxide in the slag, such as sulfate or sulfide (<2.5%),
Mn2O3, TiO2, Na2O, K2O, etc. Meanwhile, the chemical components of slag are the same as OPC
but vary in proportions; however, silica (SiO2) and lime (CaO) are the most dominant phases [2],
[10], [13], [14], [29], [123], [125], [126].
2.4.1.1 Structure and Reactivity of Slags
The composition of glass slag as a silicate glass is addressed by taking into account vitreous
silica in which several Si-O-Si bonds are broken and neutralized with metal cations known as
structure modifiers. Tetrahedral silica (SiO4) is polymerized or isolated by the bridging of oxygen
(O) atoms. Besides, several positions in a slag glass of silicon (Si) are filled with other elements'
atoms, particularly aluminum (Al) [10], [14], [19], [116]. The principle of reactivity indicators is
focused on the role of the various oxides in forming a noncrystalline glass structure. It states that
the more simple composition extensively favors hydraulic activity [10], [19].
The slag's reactivity variations are primarily related to the chemical composition of slag,
the concentration of alkali in the blended cement-slag mixture system. Moreover, the fineness of
the cement and the slag, variations of temperature during the early age hydration process, and
temperature of ambient affect the reactivity of slag [13]. Last but not least, the slag's glass content,
which is related to the beginning of cooling temperature, and cooling procedure during slag
manufacturing, determine slag's reactivity. By the way, high glassy phases in slag's chemical
composition are microscopically more homogeneous than high crystalline phases in slag's
structures. In fact, slag's crystalline phases are in the tendency to more MgO and less Al2O3 content
than the amorphous phase [10], [14], [36], [116], [122], [127].
31
Many hydraulic chemical modules have been described as indices of slag activity. The
reactivity of slag is usually assessed by empirical formulas (known as basicity ratios), hydraulic
moduli or indexes. Hydraulic moduli are typically associated with GGBS compressive strength.
The major oxides CaO (C), Al2O3 (A), SiO2 (S), and MgO (M) are part of the common modules
used for the hydraulic reactivity estimation for slags, or compressive strength in blended mortars
[10], [14], [117], [128], [129]. Ca, Mg, and Al are called networking-modifying elements, while
Si is a network-forming component [117]. The content of alumina, lime, and magnesia affect
strength positively, while silica has a negative impact [10].
Some of the most important well-known basicity index or hydraulic index formulae are
𝑀1 =𝐶
𝑆, 𝑀2 =
𝐶+𝑀
𝑆, and 𝑀3 =
𝐶+𝑀+𝐴
𝑆 [10], [128], [129]. The solubility and, therefore, the
reactivity of GGBS increase when M1 increases [10]. Moreover, EN 197-1 [130], [131] suggests
that M2 should be higher than 1 [129]. Finally, the index M3 is the most widely used [14]. As
specified in the slag's standard regulations, M3 must be more than 1 in Germany and it is more
than 1.4 in Japan [14], [117], [129].
All modules are convenient instruments for quickly assessing slag quality. However, the
following hydraulic index or moduli has been shown to define the hydraulic activity of one slag
form but does not necessarily describe the hydraulic index of each slag in general [10], [117].
Moreover, certain studies have shown that the module ratios in practice do not always match the
strength, especially at an early age. Since the slag is not limited to a particular composition, the
general validity of such hydraulic moduli was questioned. It is widely agreed that studies of the
slag glass structure will address the problem of slag reactivity [10], [117], [128], [129].
32
2.4.2 Blended Slag– Cement Hydration
The reaction of the slag when it is mixed with cement is more complicated than simply
cement hydration. When only slag mixes with water, many major hydration compounds may not
be detected [19], [29], [116]. A thin layer of silica on the slag surface hydrates with a small amount
of lime (CaO) when water penetrates the slag. This hydrated thin layer avoids further interaction
between the slag with water and slag ionic dissolution. As a consequence, the hydration of the slag
terminates [29]. For sustained slag hydration, the slag is supplemented by alkaline or sulfate
activators. Thus, sulfates and alkalis activate the slags to produce hydrations. Alkali-activators,
which are alkali hydroxides and silicates such as potassium hydroxide, calcium hydroxide
(portlandite), sodium hydroxide, and sodium silicate, raise the alkalinity on the mixture (pH
exceeds 12) [29], [52]. Therefore, it avoids the development of non-permeable levels and promotes
the dissolution of ions, which contribute to the product of hydration [13], [14], [29]. Moreover,
calcium sulfates such as hemihydrate, anhydrite, and especially gypsum, attract Al(OH)4− and Ca2+
ions released from the slag into the pore solution. It results in the formation of ettringite rather than
C–S–H [14], [29], [52], [116].
Slag contains both sulfate and alkaline activators when blended with cement. Slag
hydration activators are CH from the hydration process of C3S (alite), NaOH and KOH which
derive from alkalis in the OPC, and hemihydrate and gypsum, which is found in the cement as a
retardant [13], [14], [29], [52], [116]. Therefore, slag reacts with the cement hydration products
and the cement components. These reactions also produce more pore-blocking hydrates. The
consequence is a blended cement-slag paste with more but smaller gel pores and less capillary
pores for the same average amount of the pore volume [13], [29], [116]. This better pore structure
33
allows significantly less permeability of slag-cement concrete, which creates a huge improvement
in the durability of the whole structure [13].
The product of the hydration when slag is used with the cement is primarily the same
compared with those that had been obtained during the Portland cement hydration of the paste.
However, it is essential to note that there will be an additional hydrotalcite-like phase present
[117], [132]. The main product of the hydration is C-S-H. This particular product is, however,
under the complete influence of the slag used. This may show a slight variation for the cases of
the paste from Portland cement. The product has a relatively lower ratio of C/S as well as a high
content of the Al2O3 (A), which functions as a replacement of SiO2 (S) in the structural bridging
of tetrahedral dreierkette [132], [133]. The degree of the substitution is dependent on the C/S ratio,
which again follows the arrangement of the C-S-H and C-A-S-H. It is also influenced by the
amount of Al2O3, which is readily available for the processes of substitution [29], [117], [132].
2.4.3 Effects of Slag on the Properties of Concrete
The use of slag in concrete or paste and mortar has several effects. The effects of such
concretes will vary depending on the proportion and type of the slag used alongside other materials,
the procedure of mixing, placement, and held conditions.
2.4.3.1 Bleeding and Segregation
The concept of bleeding is a displacement of water to the surface of the concrete, which
has been recently placed [66]. Bleeding stops when either the flow of water is inhibited by
producing hydration contents or the solids that primarily come into contact with each other.
Bleeding results in the difference in the efficient water content of the whole concrete component,
resulting in corresponding changes in characteristics of the concrete [13], [14], [134]. Furthermore,
34
segregation is termed as the non-uniform blend that causes the separation of the elements (typically
coarse aggregate) from the fresh concrete [29], [66]. Segregation and bleeding in concrete are
considered to be more determined by the particle size, surface area, and reactivity or composition
of the slag. Less bleeding can occur with the incorporation of finer and reactive slags to the
concrete mixture; on the other hand, the bleeding can be increased by slow reactive coarser slags
[14], [29], [118], [134]–[136]. Therefore, segregation and bleeding are relatively lowered when
slag is used as the admixture of the mineral concrete, thereby improving the concrete pumpability.
Also, it is due to the ratio of solids to liquids which increases, thereby making the concrete to be
less susceptible to segregation and the pumpability increase of the concrete.
2.4.3.2 Workability
Typically, slags have less density than ordinary Portland cement. Hence, the volume of
paste is increased for the given amount of water, and thus increases cohesiveness and plasticity of
fresh concrete either as part of the cement replacement component or addition to the concrete.
Besides, several other features of the slags would have an impact on the workability of the concrete
[29]. The influence of slags on concrete's workability is generally related to changes in the demand
for water on the concrete mixture. The surface area increase correlates to an increase in the
interparticle surface force and contributes to a greater cohesiveness of the blended system, which
typically implies less mobility. Therefore, the water content generally needs to be increased for
specific workability when even using fine SCMs [13], [29], [137]. Even though particles of slag
have angular form, their surface is smooth and hard. Therefore, the water demand of slag-cement
blended concrete is lower than regular cement concrete. As a result, workability is affected by the
proportion of slag in the mixture and the slag and cement particle size distributions [19], [29].
35
Furthermore, some of the slag particles are known to be glassy in shape, as well as being spherical.
Such kinds of particles tend to allow for the greater workability for the ratio of w/c. This implies
that the ratio of w/c will be lowered or reduced during the operations. Based on the fact that slag
is finer compared to the cement paste, adding it to the concrete improves both particle parking and
surface smoothness, which, in turn, result in reduced water adsorption. Further, it also enhances
bleed capacity, particularly increasing the bleeding rate [2], [13], [14], [29], [118], [129], [137]–
[140].
2.4.3.3 Setting Time
The setting is known as a mortar or concrete state, which is beginning to lose its plasticity
[56], [66]. It differs from hardening representing a measurable strength condition. Hydration
occurs when the cement is in contact with water. Therefore, the hardening and setting are the
results of a continuous hydration process. The initial and final settings are the arbitrary aspects
where the mortar or concrete begins to stiffen and achieve considerable rigidity [14], [29], [139].
The importance of the first and last setting times indicates that the first gives the handling limit of
the fresh cement blended system (mixing, pouring, placement, and so on), and the second has
demonstrated to begin to a substantial strength gain [29], [118], [139]. Moreover, the parameters
that influence the setting time of cement-slag mixture are w/c ratio in the blended system, ambient
conditions, chemical and physicals properties of cement and slag (particle size, chemical
compositions, and so on), amount of slags in the mixture [2], [29], [134], [137], [139], [141]. In
general, the partial cement substitution with slag improves the setting time. It is possible to extend
the setting time to lower temperatures. Depending on the ambient conditions and the volume of
36
the replacement, the final setting time would be considerably delayed [14], [29], [105], [134],
[137], [138].
2.4.3.4 Compressive Strength
The initial slag-water reaction rate is less than ordinary Portland cement-water reaction,
meaning that the slow reaction of slag with water can lead to a lower strength gain at the early ages
of concrete, which contains varying slag levels [14], [29]. Besides, it lowers the strength gain and
this delay may increase as the quantity of slag is increased. However, the slag concrete strength
level is close to the control of concrete strength within 7 to 28 days and moderately stronger than
that of later ages (This may not be true all the time, especially when the w/cm is lower [14]).
Further, there is a general improvement in the ultimate strength and the durability of the material.
Several factors can affect the development of compressive strength. Mineralogical, chemical
compositions and physical properties of slag and cement, the slag reactivity (slag activity index),
slag amount in the blended system, mixture design, and curing ambient conditions have an impact
on the strength [13], [14], [29], [140].
In ASTM C 989 slag is categorized into three major groups, namely; Grade 120, Grade
100, and Grade 80, based on the slag activity index (SAI). The SAI represents a proportion of the
compressive strength of 50:50 blended slag-cement mortar cubes to the compressive strength of
cement mortar cubes, at a given age [14], [29], [125]. Grade 120 has the highest slag activity index,
which is known to contain high comprehensive strength, especially after 7 to 28 days.
Nevertheless, Grade 80 GGBS has the lowest compressive strength at any age. It is mostly used in
mass structures as it produces less heat than OPC. Generally, Grade 100 and 120 slags are
commonly used in the blended cement-slag systems [29], [118].
37
In the detailed research carried out by Oner and Akyuz [16] for the optimum use of slag in
the concrete mixture, the slag was mixed in varying quantities with various concentrations of
ordinary cement in four concrete groups. The volume of cement in the control samples increased
from 250 kg/m3 to 400 kg/m3. First, the mass of cement was lowered by 30%, and then the slag
was replaced in the concrete. The slag ratio was approximately between 15% and 60 % of overall
cementitious material mass. The slumps were stable at 120 ± 10 mm in all mixtures. The volume
of aggregates and water was also modified. Except for slag blended concretes with the lowest ratio
of slag, the rate of strength production of the concrete series was higher than those of the control
concrete at all ages from 7 days. The compressive strength increases proportionally by up to 50%
of the overall volume of slag in all cementitious contents. Nevertheless, over and above 50% of
slag blended concrete, the compressive strength declines, even though it is still stronger than the
control concrete samples. Moreover, the other significant aspect is that the development of strength
rate after 28 days in the slag mixed concretes are higher than in the control of concrete samples.
There was a more significant improvement in the strength gain of all slag blended concretes after
28 days, and this also improves with the rising slag volume [16], [29].
The researchers’ analyses on the substitution of partial cement with slag show that early
strength loss is balanced at and above seven days. The volume and fineness of the slag are mainly
affecting that compensation time. As the slag amount increase, the longer time required for
balancing the compressive strength, however increasing the fineness of the slag, will lead to shorter
times [13], [14], [16], [29], [111]–[119].
38
2.4.3.5 Durability
The durability is known as the capability to exist in the long-term without significant
corruption, by resisting the ambient conditions that lead to a disruption in the material properties.
Besides, the durability and service life are commonly regarded as related concepts. Considering
such description, concrete durability is distinguished as being able to resist environmental
conditions, corrosion, chemical attacks, or several other degeneration processes [2], [35], [56],
[118], [149]. Numerous scholars have researched the concrete durability containing slag-cement
under aggressive environmental conditions. Therefore, the durability of concrete is essential not
only from a technical and economic viewpoint but also from an ecological perspective; long-term
service life materials mean more natural resource conservation [13], [14], [29], [146], [150], [151].
Concrete must have sufficient strength and air induction to provide corrosion resistance
during the freezing and thawing cycle. For this reason, proper air content and adequate air spacing
are essential to provide sufficient protection against frost and thawing. Also, concrete must have
good compressive strength and be adequately cured [2], [14]. Many experiments on the impact of
slag on freezing and thawing resistance in water have comparable findings on OPC concretes, both
of which are sufficiently air entrained [13], [14], [29]. The level of GGBS also impacts the
concrete's frost-thawing resistance. The tolerance of freezing and thawing was the high agreement
in 50% slag-cement blended concrete with convenient air-containing agents’ comparison to
cement-based concrete [14], [122], [152]. Researchers conducted an air-entrained concrete
experiment with the GGBS and without GGBS for the resistance of freezing-thaw cycle durability.
The samples were stored in lime-saturated water at 23 ℃ for 14 days and tested according to
ASTM C666 [153] using Method A for 300 cycles. The concrete made from GGBS did not work
as well as OPC concrete, but all of them had a reasonable level of durability [122].
39
The creep is the concrete deformation is due to long-term steady stress applied to hardened
concrete [56]. Separately, internal water evaporation in the concrete causes drying shrinkage in the
concrete. Owing to the chemical and physical properties variations in slag and cement, testing
techniques used in analysis and curing conditions affect the creep and drying shrinkage [14].
Therefore, drying shrinkage and creep are essential time-related features of concrete [2], [13], [56].
Researchers tested three high-performance concrete mixtures for drying shrinkage and creep [154].
Sample A had 100 percent OPC, while sample B had 30% slag, and specimen C had 30% slag +
10% silica fume substitution by mass. All the other blending variables remained constant. The
shrinkage and creep experiments were carried out 180 days at 20±3 ℃. Furthermore, the creep
was applied, loading continuously to 40% of their 28-day compressive strength of concrete
specimens for 180 days. Researchers observed that the rate of creep, creep strain, and drying
shrinkage was significantly smaller in slag incorporated concretes than control concrete samples.
They concluded that high strength and workability could be reached by using slag with OPC. Slag
can reasonably increase ion and reactivity in cement, promote C-S-H gel, and improve AFt
hydration in the cement paste. These are provided a more robust structure and durability to applied
forces [13], [29], [154].
The addition of GGBS to concrete enhances its resistance to chloride diffusion, particularly
in later ages. It is quite useful to preserve structural steel against corrosion owing to the penetration
of chloride [155]. One cause for the declined chloride penetration is the decreased permeability of
slag-containing concrete. Besides, the chemical resistance of GGBS in concrete is another factor.
The chloride reaction with blended slag-cement hydration products improves the chloride binding
ability of the concrete. Its corrosion resistance efficiency is, therefore, very strong with high slag
content in concrete [14], [35], [129], [155], [156]. Researchers [157] determined the chloride
40
binding ability of cement-slag paste specimens containing 100% cement as a control sample and
mixtures of cement with varying amounts of slag (33.3%, 50%, and 66.7% ). The ratio of w/cm
was held at 0.55. They stated that chloride binding ability was increased with increasing slag
content for any levels of Cl-. At 66.7% slag-cement mixture, the chloride binding capacity was
approximately five times higher than the control sample in 5 mol/liter chloride exposure levels. In
addition, the binding capacity for any slag-cement samples was enhanced as the concentration of
chloride increased. Finally, they reported that the ratio of Cl-/OH- was directly proportional to the
capacity of chloride binding [13], [157].
Alkali aggregate reactivity (AAR) is typically known as the chemical reactions between
different aggregates components of mortar or concrete and alkalis (sodium and potassium) in the
cement and other sources (such as supplementary cementitious materials). If reactive amorphous
silica in aggregates is involved in the reaction, it is referred to as alkali-silica reactivity (ASR)
[56], [66], [158]–[160]. The reaction of ASR is like the reaction of pozzolanic, which is a
straightforward acid-base reaction between portlandite (Ca (OH)2) and silicic acid (Si (OH)4) [13].
The siliceous aggregate reacts with the alkaline solution (Na+, K+, or Ca2+) to turn the mixture
into a viscous alkali silicate gel. Water in the concrete pore includes all these alkalis solutions;
concentrations of alkalis depend on the chemical compositions of cement. Using cement-
containing high alkali increases the pH level of the system. Higher solubility and dissolution of
amorphous silica represent an increasing of pH level [2], [29], [35]. The product of the alkali-silica
gel reaction can cause a cracking and unusual expansion in the mortar and concrete. This gel is
capable of unrestricted moisture absorption. The expansion of volume induces internal stress,
which triggers the cement aggregate matrix cracks as it reaches the limit of tensile strength. The
local cracks occur. After these cracks interconnect with each other, finally, the concrete inevitably
41
fails [13], [29], [158], [161]. The researcher tested possible AAR for blended OPC and GGBS
using ASTM C227 [162]. A highly reactive aggregate component (Pyrex) was used in this mixture.
The study concluded that the partial substitution of high-alkali OPC with GGBS significantly
decreases the risk of AAR in concrete [35], [122]. In handling the ASR reaction, slag is extremely
useful because slag decreases concrete alkalinity and the alkali/silica ratio [14], [36], [147].
Besides, slag decreases concrete alkaline mobility and lowers the free lime (CaO) in the concrete,
which is a significant element in the ASR [2], [13], [105].
2.5 Sulfate Optimization
The term "sulfate optimization" is typically slightly relative and thus difficult to describe.
Volume stability and strength are commonly considered as two parameters for the determination
of the optimal sulfate level. The parameters that influence the optimum sulfate content of the
mixture are the source of the sulfate, the solubility of the sulfate, the mineral and chemical
compositions of the cement, the cement fineness, the mixture age, w/c ratio, curing temperature,
the type of SCMs, and the chemical admixtures [19], [20], [48], [163], [164]. Optimum SO3 level
for one mixture is most likely not to be usable for another blended system.
Researchers and scientists have been using ASTM C563 [164] to determine the optimal
level of SO3. There are, however, some discrepancies between this approach and the use of cement
in the field. The ASTM sulfate optimization experiments are carried out with cement paste or
mortar. No chemical or mineral admixtures are used during tests, although they are frequently used
in concrete. The ASTM experiments are conducted at 23 ℃, which does not represent several field
conditions.
42
Lerch [20] investigated the first most comprehensive research on sulfate optimization of
cement paste and mortar. Researcher optimized sulfate content of cement-based on compressive
strength, the heat of hydration (HOH), and length change of mortar bars stored in water. As
detected by Lerch [20], there are many certain factors affecting sulfate demand on cement and
cementitious system. Tricalcium aluminate (C3A), fineness, and alkalis are the most well-known
factors on the sulfate demand for cement. Lerch has been a pioneer and inspired many studies on
sulfate optimization of cement and cementitious systems. Therefore, Lerch's work [20] is the
cornerstone of these ASTM standards and is still the most important study of optimal amount of
sulfate. Although the factors influencing the optimum level of SO3 have been well established
since Lerch's study [20], the concrete industry has not adopted any numerical models to calculate
the optimum sulfate level. The complication, changeability, and lack of basic knowledge of the
cement hydration mechanisms lead to mechanistic approaches for optimum SO3 level
determination by measuring strength, total heat of hydration, and volume stability [19], [23], [24],
[164].
2.5.1 Sulfates
The task of SO42− (sulfate) in calcium sulfates is to regulate the hydration of aluminate. The
uncontrolled reaction of aluminate triggers the "flash set" [2], [19], [23], [39], [163]. The quantity
of C3A hydrated within the first minutes of hydration is sensitive to the amount of sulfate and the
form of sulfate used in cement [20], [163], [165]–[167]. Alternative sulfate forms have
demonstrated a substantial impact on cement performance. The several performances are linked to
the sulfate solubility. The solubility of sulfate depends upon both the type and the size of the
particle. The effect of sulfate form on the optimum sulfate level is influenced by the solubility
43
rates. Thus, higher solubility leads to reduce optimum sulfate concentrations in the system [165],
[166], [168]–[171].
The particle size distribution effect was illustrated by the intergrinding of gypsum versus
separate blending of gypsum experiments. The intergrinding of gypsum in the cement displays
better paste characteristics and decreased rate of C3A hydration for up to 24 hours and manages
the early C3A hydration efficiently. Better distribution of particles leads to a faster supply of
SO42− and Ca2+ to the C3A hydrating surfaces [23], [163], [165], [166], [168]–[171].
Several forms of sulfate can be found in the OPC in a variety of ways: Several forms of
sulfate can be found in the OPC in various ways: existing in the clinker, directly applied to the
finishing mill, and/or transferred from one form to another due to high temperatures or variations
in dew points. Calcium sulfate is typically supplied with additives into the finishing mill.
Impurities mostly associated with grade calcium sulfate's processing sources may also affect early
reactions, setting behaviors, and stiffening [20], [23], [48], [163], [165], [166], [168], [169].
Calcium sulfate dihydrate or gypsum (CaSO4•2H2O) naturally occurs in earth crust and
artificial source materials developed as by-products of other industrial sectors, including flue gas
desulfurization (FGD) [23], [163], [166], [169], [172]. Calcium sulfate hemihydrate
(CaSO4• 1
2 H2O), commonly known as bassanite or plaster of paris, it is naturally found or formed
by heating gypsum. The type of calcium sulfate hemihydrate depends on the preparation
techniques. The α form of hemihydrate can be generated with gypsum under autoclave conditions
at 200 °C in the presence of water. The β type begins to form at 40 °C in open boilers and ovens.
It is generally known as bassanite or plaster of paris. Besides, it commonly occurs in the OPC. The
surface area of the β is larger than the α type [23], [163], [166], [171]–[173].
44
Calcium sulfate III or soluble anhydrite (CaSO4 (III)) is formed under the same
circumstances as the β type of the hemihydrate, but it is formed at 200 °C. Calcium sulfate (III) is
structurally similar to the hemihydrate and can effectively convert water molecules into
subhydrates under atmospheric conditions. Relative humidity and material temperature must be
controlled; otherwise, it can convert into hemihydrate in a matter of hours [165], [174], [175].
Moreover, calcium sulfate II or insoluble anhydrite (CaSO4 (II)) is formed by heating either of the
phases mentioned earlier at 500 °C to 900 °C for approximately one hour. Calcium sulfate (II) has
a denser structure. Thus, there is no space for extra water molecules. This is in contrast with
calcium sulfate (III), which is very similar to the form of β hemihydrate. On the other side, CaSO4
(I) is not generally contained in cement because it demands more than 1200 °C [19], [23], [168],
[174], [175].
Changes in the sulfate form are typically found in cement mills. The gypsum's dehydration
to hemihydrate forms to anhydrite types occurs as a result of temperature rise [23], [168], [176],
[177]. CaSO4•2H2O → CaSO4• 1
2 H2O → CaSO4 (III) → CaSO4 (II) → CaSO4 (I)
CaSO4 (III) and hemihydrate are relatively more soluble than calcium sulfate dihydrate
[23], [163], [166], [175]. Nevertheless, using reactive soluble SO42− sources, such as CaSO4 (III)
and hemihydrate, are initially beneficial but do not have a long-term retardant effect if the physical
distribution is not proper. Due to its high solubility, the hemihydrate is suitable for high-aluminate
cements, although CaSO4 (III) could cause many issues, such as early stiffening and high demand
for water. On the other hand, low aluminate cements tend to react favorably to a wide variety of
sulfate sources, including CaSO4 (III), as set-control agents. Too much CaSO4 is inappropriate as
a primary set retarder since it is so slowly dissolved in water [19], [23], [163], [165], [166], [175],
[178].
45
2.5.2 Alkalis
Alkali accelerates both C3S and C3A hydration, while later silicate hydration slows down.
The hydration of C3A in high-alkali OPC was roughly double that in low-alkali OPC at 24 hours
[163], [166], [179]. Lerch [20] points out that since some of the alkalis in OPC are found in the
C3A phases, these phases have a quicker reaction than comparable non-alkali or low-alkali phases.
OPCs with alkalis require more gypsum than them for convenient retardation time. Tang [166]
states that high concentrations of soluble alkalis, that cement has more than one percent of
Na2Oeq, demand additional gypsum due to the alkalis' accelerating effect on the aluminate
hydration. On the other hand, the original cement sulfate content can be adequate and demand less
gypsum since alkalis in the clinker are mostly in alkali sulfate form. They are also more effective
than calcium sulfates in retarding the aluminate hydration [20], [23], [163], [165], [166], [179]–
[181].
Alkalis typically improve the early cement strength; however, they reduce later cement
strength. This could be linked to the accelerated aluminate hydration, leading to quicker SO3
consumption and an improved conversion process from ettringite to C3A solid solution sequence
[163], [165], [166]. Besides, alkalis can influence gypsum and calcium hydroxide solubility.
Higher alkali content result in more calcium hydroxide and the relatively lower gypsum
concentration both remain saturated in the pore solution. The higher the alkali content, the higher
the content of sulfate in the solution [20], [23], [182], [183].
Some of the high alkali-low sulfate Portland cements can have an HOH curve similar to a
properly retarded and hydrated cement curve, but further experiments with higher sulfate contents
demonstrate that they are not properly hydrated [20]. High alkali Portland cements need higher
volumes of gypsum for similar hydration rates. Higher alkali Portland cements are also able to
46
withstand higher sulfate levels without unnecessary expansion [163], [166]. While low-aluminate
cements are relatively less sensitive to the alkali effects, Lerch [20] pointed that a higher proportion
of gypsum is needed for cements with low-aluminate and high in sodium oxide for appropriate
retardation than cements high in potassium oxide.
Increasing the alkalis in the cement typically improves the early strength and decreases
late strength. They work as a catalyst for the hydration of calcium silicates and C3A. However, this
impact on strength is reduced above the ideal amount of sulfate. Na2O demonstrates shorter setting
times, while K2O is indifferent. It also adversely impacts 28-day strength more than K2O. The
adverse impact on 28-day strength and improved 1-day strength have been correlated with
increased C3A consumption [19], [20], [23], [163], [165], [166], [179]–[181].
2.5.3 Slag and Supplementary Cementitious Materials
The interesting fact is that very little research has been conducted to investigate the impact
of slag on sulfate optimization and sulfate interactions. The main explanation that slag has affected
the SO3 level is the aluminate, alkalis, and free-lime frequently present in high alumina slag-
cement system. Kocaba [10] investigated the effects of slag on the cement clinkers and blended
cement system. The researcher found that the use of slag significantly improved the hydration of
ferrite. On the other side, slag with a high alumina and alkali content is more reactive than the rest
of the slags and has increased the hydration rate [10], [184].
The addition of limestone and calcium sulfate to the high-alumina slag increases the
consistency and quality of mortar and concrete. Researchers [9], [126] aim to show different results
from applying sulfates and limestone on stabilizing ettringite in the high-alumina slag. While
ettringite remains stable in the high-alumina slag after sulfates addition, the addition of lime can
47
produce hemi-or mono-carboaluminate. Despite the fact that hemi-or mono-carboaluminate would
ultimately transform to ettringite, the external sulfate attack would have been considerably
delayed. The researchers found that the cement's and slag's sulfate content impacts sulfate
durability by a 50% substitution for high-alumina slags. The sulfate durability of the slag blending
systems was greatly affected by the sulfate to alumina ratio of the slag and OPC. [9], [126], [185]–
[187].
Even though there was no specific standard to determine sulfate optimization in blended
cement systems, the researcher also used compressive strength test and isothermal calorimetry to
determine sulfate demand on fly ash-OPC systems in his dissertation and paper [23], [24]. The
researcher demonstrated that fly ashes increase in optimum sulfate demand on the OPC-fly ash
system. The researcher examined in detail the effect of sulfate optimization on the OPC-fly ash
system, which not only depends on the physical properties of fly ash, like particle size but also
depends on the chemical properties of fly ash [23], [24].
Zhu et al. [188] demonstrated SO3 optimization for limestone cements and an OPC with
two unique sources of clay. The impact on many factors, including replacement level of clay and
source of clay, on the optimum SO3 content of the blended system, was addressed. Researchers
used isothermal calorimetry to measure the HOH at 23 ℃. Researchers determined that SO3
optimization in the blended system depended on the clay sources, the chemical composition of
clay used, and types of cement. Besides, the optimum SO3/Al2O3 ratio rises with time for all
blended systems. They recommended that particle size distribution and mineralogical composition
of clay may affect sulfate optimization [188].
Mei-zhu et al. [22] carried out experiments to check the impacts of SO3 content and types
of gypsum content on the strengths and fluidity of different types of cementitious materials or
48
systems. Such results have indicated that the gypsum influences cementitious materials differently
from one compound element to another. The findings have shown that the synergistic effects and
combined effects of fly ash and slag play a crucial role during hydration processes. In the
cementitious materials with only 45% of the clinkers, 5% limestone, 30% slag, and 20% fly ash,
the content of optimized SO3 in the calcined gypsum and gypsum itself are 3.51% and 3.13%,
respectively. This is translated to the optimum content of the gypsum as 6.5% [22].
2.5.4 Fineness and Age
The fineness or particle size of the OPC and SCMs is also an essential factor for
determining the optimum amount of SO3 in the mixture [20], [23]. The fineness of cement and
SCMs affects the rate of hydration. Increased surface area enhances the solubility and hydration
rate [2], [19], [189]–[191]. Besides, since surface area increases due to fineness increase, C3A
reactivity increases; as the reactive of C3A increases, an additional SO3 compound will be required
to control the hydration [19], [20], [23], [163].
Age is one of the parameters that determine the optimum SO3 level. The later the strength
measurements, the higher the SO3 content in the samples, the higher the strength [164], [192]. The
optimum sulfate level increases with age, likely since more C-S-H volume can accommodate,
requiring more ettringite to reduce porosity. The shifts in the optimum SO3 level duration between
one-day and three-days are significant, and it appears that sulfate level rises after seven-days but
by much less amount. These shifts increased with the levels of the aluminate and alkali content in
OPC [20], [23], [163], [164], [166], [192].
49
Chapter 3: Materials and Methodology
This section introduces detailed information on the materials used, the methods used, and
the experiments carried out during this research. All methods and experimental procedures
followed the ASTM standards referred to in this chapter. However, some standard mixing designs
have been modified for consistency during the testing. These modifications have been presented
in the following chapter.
3.1 Materials Characterization
3.1.1 Chemical oxide composition
Slags were chosen depending on the contents of alumina (Al2O3). Two blast furnace slags,
which contain low and high alumina contents, and four ordinary Portland cements (OPC), which
are varied from chemical, mineralogical composition, and fineness, were selected for this research.
Calcium sulfate hemihydrate (CaSO4• 1
2 H2O) was used for increasing sulfate content in the slag-
cement system. The proportion of calcium sulfate hemihydrate was replaced as cement by mass.
The chemical oxide compositions of the slags, hemihydrate, and cements were carried out
by using X-ray fluorescence spectroscopy (XRF), according to ASTM C114 and ASTM C989
[125], [193] (Table 3.1 and Table 3.2). In accordance with ASTM C1365 [194], the mineralogical
composition of the cements and slag were determined by using X-ray diffraction (XRD). Before
XRD measurements, cements and slags were ground in a McCrone Micronising Mill for 10
50
minutes with the addition of approximately 2 ml of ethanol (200 proof) for each gram of cement
or slag samples to a particle size less than 10 µm. The samples dried at 40 ± 1 ℃ in the oven [195].
It was known that at 40 ℃ ground cement oven drying has no phase shifting impact on gypsum
due to temperature because the transformation of gypsum to hemihydrate or anhydrate be formed
at higher temperatures [196], [197].
Table 3.1 Oxide chemical compositions for cements and calcium sulfate hemihydrate
Analyze Cement A
(wt %)
Cement B
(wt %)
Cement C
(wt %)
Cement D
(wt %)
Hemihydrate
(wt %)
SiO2 21.20 20.10 19.00 18.80 <0.01
Al2O3 5.15 5.60 5.90 5.70 <0.01
Fe2O3 3.61 2.00 2.80 2.50 0.01
CaO 63.91 64.40 60.80 61.00 38.70
MgO 0.70 0.90 2.50 2.70 <0.01
SO3 2.61 3.60 4.00 4.20 55.34
Na2O 0.14 0.08 0.32 0.30 0.02
K2O 0.31 0.47 1.10 1.02 <0.01
TiO2 0.29 0.18 0.26 0.25 <0.01
P2O5 0.15 0.33 0.26 0.25 <0.01
Mn2O3 0.03 0.03 0.11 0.09 <0.01
SrO 0.06 0.07 0.28 0.27 0.02
Cr2O3 0.02 0.01 0.01 <0.01 <0.01
ZnO 0.06 0.03 0.07 0.06 <0.01
L.O.I (950 ℃) 1.66 1.80 2.40 2.90 6.24
Total 99.89 99.56 99.82 100.00 100.34
Na2Oeq 0.35 0.39 1.05 0.97 -
51
Table 3.2 Oxide chemical composition of slags
Analyze Slag1 (wt %) Slag4 (wt %)
SiO2 38.59 32.86
Al2O3 8.09 16.29
Fe2O3 0.51 0.36
CaO 38.11 37.98
MgO 10.83 8.88
Sulfide S 0.89 0.95
Sulfate as SO3 0 0.23
Na2O 0.30 0.37
K2O 0.38 0.44
TiO2 0.37 1.21
P2O5 <0.01 <0.01
Mn2O3 0.59 0.25
SrO 0.05 0.10
Cr2O3 <0.01 <0.01
ZnO <0.01 <0.01
BaO 0.03 0.08
L.O.I (950 ℃) 1.15 0.40
Total 99.89 100.41
Na2Oeq 0.55 0.66
3.1.2 Mineralogical composition
XRD measurements were performed using the PANalytical diffractometer via X'Celerator
Scientific detector and Cu-Kα X-ray radiation. Voltage and current were fixed to 45 kV and 40
mA. The scans were collected in the range of 4–70° 2θ, with a step size of 0.0167° and a counting
time per step of 130.2 mm. In addition, anti-scatter slits and five millimeters divergent were used
52
in the auto mode. The back-loading method was used for specimens loading in the sample holder.
Phase identification of specimens was determined by using HighScore Plus 4.5 software. The
external standard method was used to identify unidentified amorphous content for each specimen
through Rietveld refinement [198]–[201]. The mineralogical composition of cements and slags is
presented in Table 3.3 and Table 3.4.
Table 3.3 Mineralogical composition of cements
Phases Cement A (wt %) Cement B (wt %) Cement C (wt %) Cement D (wt %)
Alite 48.08 54.03 48.11 49.52
Belite 23.13 17.28 15.60 13.40
Aluminate 5.54 8.36 8.29 8.09
Ferrite 9.89 5.56 7.56 6.76
Syngenite 0.72 0.43 0.86 0.70
Quartz 0.14 0.07 0.20 0.28
Calcite 1.17 0.35 1.62 1.84
Anhydrite 0.02 0.07 0.00 0.03
Hemihydrate 1.55 1.41 1.74 2.95
Gypsum 2.61 4.34 3.78 2.22
Portlandite 0.00 0.18 0.22 0.24
Aphthitalite 0.00 0.17 0.54 0.44
Periclase 0.00 0.00 1.33 1.33
Dolomite 0.00 0.00 1.28 1.82
Amorphous 7.16 7.75 9.06 10.49
53
Table 3.4 Mineralogical composition of the slags
Phases Slag1 (wt %) Slag4 (wt %)
Calcite 0.68 0.28
Melilite 0.46 0.73
Amorphous 98.86 98.99
3.1.3 Fineness
The Blaine fineness of the slags and cements were measured following ASTM C204-16
[202]. Particle size distribution (PSD) of as-received dry materials powders were measured via
Horiba LA-950 particle size analyzer [203]. The refractive index of 1.7-1.0i was selected for the
diffraction measurements of cement particles based on "Certification of SRM 114q: Part II
(Particle size distribution) and NIST Special Publication 260-166". The PSD measurements were
performed in the auto mode using tiny nozzle at 0.3 MPa air pressure. The same test repeats three
times were conducted all cements and slags. The results are represent in the Table 3.5.
Table 3.5 Particle size analysis for the slags and cements
Physical properties of
cements and slags Cement A Cement B Cement C Cement D Slag1 Slag4
Blaine Fineness,
(m²/kg) 485 474 436 571 652 466
Mean particle size (μm) 12.3 13.8 16.5 10.3 9.2 11.8
Median particle size
(μm) 10.0 11.5 12.6 9.1 7.8 10.0
D10 (µm) 2.0 2.7 2.7 2.1 1.0 1.6
D50 (µm) 10.0 11.5 12.6 9.1 7.8 10.0
D90 (µm) 25.0 27.6 35.3 19.6 18.8 23.7
54
3.2 Methodology
Optimization of SO3 content is typically conducted for paste or mortar systems using
compressive strength testing and isothermal calorimetry for quality control. The objective of these
procedures is to ensure adequate strength and set properties by controlling the time of occurrence
of the aluminate peak by adjusting the SO3 content by incremental addition of sulfate in the form
of hemihydrate at four or more levels, at the same time as attempting to maximize either strength
or heat of hydration. A single replacement level of 50% by mass of slag was used, and the
proportion of calcium sulfate hemihydrate was replaced as cement by mass. Their optimum SO3
was determined according to ASTM C563 [204] using both isothermal calorimetry and
compressive strength measurements at the age of 1 day and 3 days.
3.2.1 Compressive strength
Compressive strength of 2 in x 2 in mortar cubes was tested in accordance with the ASTM
C109 [28]. For each varying particular sulfate levels, six mortars cubes were prepared according
to ASTM C109, C305, and C511 [28], [205], [206], for compressive strength tests. To mixture
proportion and consistency of cement, slag, sand, and hemihydrate with water were used according
to ASTM 109C and ASTM C563. Furthermore, ASTM C305 and C511 were used for determining
the mortar mixture. The water to cementitious ratio (w/cm) of samples was 0.485 following ASTM
C109, and samples were prepared and molded at 23±2 ℃. The mixing procedure of compressive
strength is presented in Table 3.6.
55
Table 3.6 Mixing procedure for compressive test
Mixing Procedure Time
Add Water and Calcium Hemihydrate to the mixer -
Start a timer and operate the mixer on low speed (140 ± 5 rpm) for 20 sec 0:00-0:20
Stop the mixer and add cement-slag mixture to the mixer for 10 sec 0:20-0:30
Start the mixer and operate the mixer on low speed (140 ± 5 rpm) for 30 sec 0:30-1:00
Stop the mixer, change speed of the mixer to medium (285 ± 10 rpm) speed
and mix 30 sec 1:00-1:30
Stop the mixer, scrape down bowl first 15 sec and cover the bowl with the
lid 1:30-3:00
Start the mixer again on medium speed (285 ± 10 rpm) and mix for 60 sec 3:00-4:00
Following the completion of molding, samples were placed into moist closet for 24 hours.
After the end of 24 hours, the rest of the samples were kept in saturated lime water storage
containers until the experiment time. Mortar cubes were tested after one day and three days
according to the ASTM C109. The results of compression tests also evaluated according to ASTM
C109. After all these compression test results satisfied with ASTM C109 standard range, optimum
sulfate level estimated by using Gaussian distribution and second-order polynomial. The
researcher assumed that the sulfate optimization peak was symmetric; hence Gaussian distribution
was used for fitting to compression and isothermal calorimetry test results.
3.2.2 Isothermal calorimetry
The parallel test was conducted using isothermal calorimetry to determine the heat of
hydration at 23 ℃. For the measuring heat of hydration of samples, the researcher used TAM AIR
3-channel isothermal conduction calorimetry manufactured by TA Instruments. Ottawa sand was
used as reference material in the isothermal calorimetry. Mass of the sand matched with the heat
56
capacity of cement-slag paste [207]. Calibration of calorimetry was conducted with the instrument
specifications. Each channel of calorimetry was cautiously calibrated, and the baseline of
calorimetry stabilized at ±10 μW before running experiments.
Cement-slag samples were prepared and tested following ASTM C1702 Method B [208],
external mixing. Before starting the isothermal experiment, cement, slag, calcium sulfate
hemihydrate, and distilled water were preconditioned at 23 ℃. First, cement, slag, and
hemihydrate were dry mixed manually inside the glass vial. At the time of mixing, the distilled
water was added to the cement-slag-hemihydrate and manual mixing over 60 seconds with a hand
mixer inside the glass vial. Afterward, the vial was immediately sealed and placed into the
calorimetry cell. The data logging was quickly started putting the vial into the calorimeter.
Specimens for calorimetry were weighed 40g. All calorimetry experiment results were
normalized by the solid mass of the mixture [23], [24]. Because of the initial reaction of cement-
slag solid mixtures with water, to specify heats of hydration on the paste, the measured data were
considered after 90 min. This notion associate with the "traditional maturity method where the
temperature history before set is excluded from the evaluation" [27], [209].
57
Chapter 4: Results and Discussion
4.1 Determination of Optimum SO3 content
The determination of optimum SO3 content in this research has been made considering the
chemical and physical characteristics of these materials. Table 3.1 shows the oxide chemical
composition of the materials used in the current study. Cement C has the highest Al2O3 content
and Cement A has the lowest Al2O3 content than other cements. On the other hand, cement A has
the greatest Fe2O3 content, and cement B has the least Fe2O3 content than others. The sulfate
content, expressed as SO3, shows the difference among the cements studied here. Cement D has
the highest SO3 content, followed by cement C, cement B, and cement A in decreasing order.
Furthermore, cement C and D have higher alkali content than cement A and B, although cement
A and B have greater CaO and SiO2 content than cement C and D. Cement A and B are the same
Blaine fineness (see Table 3.5). However, cement B has higher alite (C3S), aluminate (C3A), and
gypsum content than cement A (see Table 3.3). Moreover, cement A has the lowest C3A content
than others, which are almost the same level. Cement B has the highest gypsum content pursued
by cement C, cement A, and cement D in a declined order. Cement C and cement D have almost
the same mineralogical and chemical composition. However, cement D has most fineness and
highest hemihydrate contents others than which are almost the same levels. Moreover, Slag4 has
higher Al2O3 than Slag1. On the other hand, Slag1 has higher fineness than Slag4.
58
The amounts of cement, slag, and calcium sulfate hemihydrate were calculated by
considering the sulfate level before every single experiment. The sulfate level is introduced as
sulfate (SO3) coming from the cement, slag, and added calcium sulfate hemihydrate divided by
the total mass of the cementitious materials in the blended system. The following equation is
expressed for the calculation of SO3 contents in the cement-slag systems:
SO3 Content (%) = (𝑀𝐶𝑒𝑚𝑥𝑆𝑂3𝐶𝑒𝑚)+(𝑀𝑆𝑙𝑎𝑔𝑥𝑆𝑂3𝑆𝑙𝑎𝑔)+(𝑀𝐻𝑒𝑚𝑖𝑥𝑆𝑂3𝐻𝑒𝑚𝑖)
𝑀𝑐+𝑀𝑆𝑙𝑎𝑔+𝑀𝐻𝑒𝑚𝑖 Eq. (4.1)
where
MCem is mass of the cement in cement-slag system
SO3Cem is sulfate content of cement from XRF analysis (%)
MSlag is mass of the slag in cement-slag system
SO3Slag is sulfate content of slag from XRF analysis (%)
MHemi is mass of the hemihydrate in cement-slag system
SO3Hemi is sulfate content of hemihydrate from XRF analysis (%)
Second-order polynomial and Gaussian distribution were used for estimating the optimum
sulfate levels in the cement-slag system. One assumed that the peaks were symmetric. The
researcher did not observe any asymmetric distribution in this research. A sample of Gaussian and
2nd-order polynomial fit for three-day Cement B-Slag1 compressive strength results versus sulfate
levels, as seen in Figure 4.1. The Gaussian and 2nd order polynomial fits overlap, and they have a
solid agreement. Consequently, the optimum compressive strength and heat of hydration test
results were estimated with a second-order polynomial and Gaussian distribution.
59
Figure 4.1 Gaussian and 2nd-order polynomial fit for three-day Cement B-Slag1 compressive
strength results versus various sulfate levels.
4.2 Optimization of SO3 Content in the Cement
According to calorimeter experiments, HOH was measured, and the optimum sulfate level
was determined for only cement components. The Figure 4.2 indicates examples of the heat flow
of 100% cement hydration for 72 hours. The heat flow curves provide a general idea about the
overall hydration of all used types of cement. Based on the Lerch's work, a properly retarded
cement should contain the minimum amount of gypsum necessary to generate a curve that
demonstrates two cycles of increasing and declining heat release rates and shows no significant
change with larger additions of gypsum within the first 30 hours of hydration [20], [163].
Following Lerch's statement, the figure indicates that 24-hours is enough to determine optimum
sulfate level for cements. In this research, cement-slag hydration worked took 3 days. That is why
2800
2900
3000
3100
3200
3300
3400
0 1 2 3 4 5 6 7
Co
mp
ress
ive
stre
ng
th (
psi
)
SO3 Content %
Cement B-Slag1
Gaussian Distribution
(Stress)
Avg stress (psi)
2nd-order Polynomial
Function
60
the researcher continued to measure the HOH of cements for the three days. The researcher did
not prepare a mortar sample for sulfate optimization because of previous research. Many papers
and studies already prove that compressive stress tests and HOH give the same result for
determining sulfate optimization for OPC [20], [22], [23], [163]. However, the optimum one-day
sulfate level was double-checked, with mortar samples prepared for all types of cement. Table 4.1
indicates the optimum sulfate content in Cement A, B, C, and D.
Table 4.1 Optimum SO3 level estimates for Cement A-D
Optimum
SO3 Level (%) Cem A Cem B Cem C Cem D
1-Day 3.3 3.7 4.0 4.4
3-Day 3.7 4.1 4.2 4.6
Cement D shows higher hydration heat and higher optimum sulfate content at all ages.
Considering aluminate (C3A) and alumina (Al2O3) are some of the most important factors to
consider when optimizing SO3 content in a Cement mixture. Furthermore, alkali solubility can
affect optimum sulfate level. These could help to explain why Cement D has high SO3 optimum
amounts. However, Cement C and cement D have almost the same mineralogical and chemical
composition, although cement D has most fineness than other cements (see Table 3.5). The
increase in fineness increases the reactivity of the aluminate due to increases in the surface area.
This increase in reactivity in aluminate would necessitate additional sulfate to control the reaction.
No optimum sulfate level was observed for Cement C at one day. When the amount of sulfate in
cement C is increased, HOH and compressive stress dropped immediately.
61
Figure 4.2 The heat of hydration flow for 100% Cement A, B, C, and D
4.3 Optimization of SO3 Content in the Slag-Cement Blended System
Several techniques were used for determining the optimum sulfate level on the one and
three days of compressive strength. Six mortar cubes were tested for each sulfate level. The
optimum strength level estimated on the individual strength test results at the same age and the
average of the particular strength results in the same period (shown in Figure 4.3 and Figure 4.4).
There was a minor gap between the compression test results of individual samples fitting and the
results of the average of the six mortar cube samples, as anticipated by the researcher. As seen in
Figure 4.5, there is an excellent substantial agreement between Gaussian and second-order
polynomial fitting.
0
1
2
3
4
5
6
0 8 16 24 32 40 48 56 64 72
Hea
t F
low
(m
W/g
of
cem
ent)
Time (hour)
Cement Heat Flow
Cement A
Cement B
Cement C
Cement D
62
Figure 4.3 Gaussian and 2nd-order polynomial fit for Cement D-Slag1 individual specimens
compressive strength results versus sulfate level at three days.
Figure 4.4 Gaussian and 2nd-order polynomial fit for Cement D-Slag1 average of specimens
compressive strength results versus sulfate level at three-day.
2400
2700
3000
3300
3600
3900
4200
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0
Com
pre
ssiv
e S
tren
gth
(psi
)
SO3 Content %
Cement D-Slag1
Gaussian
Distribution
(Stress)
Avg stress (psi)
2nd-order
Polynomial Function
2300
2600
2900
3200
3500
3800
4100
4400
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0
Com
pre
ssiv
e S
tren
gth
(psi
)
SO3 Content %
Cement D-Slag1
Avg stress (psi)
Gaussian Distribution
(psi)
2nd-order Polynomial
Function
63
Figure 4.5 The Gaussian versus 2nd order polynomial estimate the average strength of the
specimens' optimum sulfate level at 1-day and 3-days
The heat flow of samples and total heat of hydration (HOH) were measured by isothermal
calorimetry. The calorimetry was calibrated before each experiment to ensure the best
experimental results and verify the stability of the baseline. Figure 4.6 indicates an example of the
heat flow of Cement C-Slag 4 at various sulfate levels. The heat flow curves provide a general idea
about the overall hydration of the cement-slag system. They mainly deliver remarkable changes in
the reaction of the aluminate phases from increasing the sulfate level in the blended OPC-slag
system.
R² = 0.9988
2.5
3.0
3.5
4.0
4.5
5.0
2.5 3.0 3.5 4.0 4.5 5.0
2nd o
rder
poly
nom
ial
esti
mat
e by a
ver
age
of
spec
imen
s
Gaussian estimate by average of specimens
Compressive strength
Average
Linear (Average)
64
Figure 4.6 Heat flow of Cement C and Slag 4 in different sulfate level
The total heats of hydration increased with increase SO3 contents in the cement-slag
system; likewise, the strength increased. The heats of hydration declined with the higher SO3 levels
in the blended system; in the same way, the strength decreased. Notwithstanding, the types of
cement and slags come by own various chemical and mineralogical composition, this stable
relation sustained even the different sulfate levels. As seen in Figure 4.7 and Figure 4.8, this
relationship was observed even at different age of the samples.
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
0 8 16 24 32 40 48 56 64 72
Hea
t F
low
(m
W/g
of
cem
enti
tiou
s)
Time (hour)
Cement C-Slag4
Cem C-Slag4-SO3 2.1%
Cem C-Slag4-SO3 2.7%
Cem C-Slag4-SO3 3.2%
Cem C-Slag4-SO3 3.7%
Cem C-Slag4-SO3 4.3%
Cem C-Slag4-SO3 4.8%
65
Figure 4.7 Comparison of compressive strength and total heat of hydration of one-day Cement
A-Slag 1 on the different sulfate levels
Figure 4.8 Comparison of compressive strength and total heat of hydration of three-day Cement
A-Slag 1 on the different sulfate levels
105.00
110.00
115.00
120.00
125.00
1150
1250
1350
1450
1550
1650
1.00 2.00 3.00 4.00 5.00
Tota
l H
eat
(J/g
cem
enti
tious)
Com
pre
ssiv
e S
tren
gth
(psi
)
SO3 Content %
CementA-Slag1 1day
Strength
HOH
Poly. (Strength)
Poly. (HOH)
170.00
175.00
180.00
185.00
190.00
195.00
200.00
2200
2400
2600
2800
3000
3200
1.00 2.00 3.00 4.00 5.00 6.00
Tota
l H
eat
(J/g
cem
enti
tious)
Com
pre
ssiv
e S
tren
gth
(psi
)
SO3 Content %
CementA-Slag1 3day
Strength
HOH
Poly.
(Strength)
Poly.
(HOH)
66
Furthermore, the optimum sulfate level was estimated with the total heat of hydration in a
similar way to compressive strength. Gaussian distribution and second-order polynomial function
were used for determine optimum sulfate levels. The estimation of optimum sulfate levels for the
compression strength and total heat of hydration is seen in Table 4.2. This table indicates that there
is a substantial relationship between compression strength and heat of hydration to determine
optimum SO3 content in the slag-cement system.
67
Table 4.2 Optimum SO3 level estimates on compressive strength and calorimetry test
Age Sample
HOH Compressive Stress
Average Individual
Gaussian 2nd
order Gaussian
2nd
order Gaussian
2nd
order
1 D
AY
CemA-S1 2.1 2.1 2.2 2.2 2.2 2.2
CemB-S1 1.9 1.8 2.3 2.3 2.3 2.3
CemC-S1 2.0 2.0 2.0 2.0 2.0 2.0
CemD-S1 2.5 2.4 2.4 2.4 2.4 2.4
CemA-S4 3.0 3.0 3.0 3.0 3.1 3.1
CemB-S4 2.8 2.8 3.0 3.0 3.1 3.1
CemC-S4 2.6 2.6 2.6 2.6 2.6 2.6
CemD-S4 2.6 2.6 2.8 2.8 2.8 2.8
3 D
AY
CemA-S1 2.8 2.8 2.7 2.7 3.0 3.0
CemB-S1 2.7 2.7 2.7 2.7 2.8 2.8
CemC-S1 2.7 2.7 2.6 2.6 2.6 2.6
CemD-S1 3.2 3.2 3.1 3.1 3.1 3.1
CemA-S4 3.6 3.6 3.8 3.8 4.4 4.5
CemB-S4 3.3 3.3 3.7 3.7 4.3 4.4
CemC-S4 3.2 3.2 3.8 3.8 4.0 4.0
CemD-S4 4.2 4.2 4.2 4.2 4.1 4.1
68
Additionally, this strong correlation was verified using Matlab. The optimal sulfate data
points were fitted with a linear function, and then the coefficient of determination, R2, was
estimated. After that, the data's standard error and 95% prediction interval was calculated by the
Matlab code. The original data, linear fit, and 95% prediction interval are plotted in Figure 4.9.
A correlation between compressive strength and total heat of standard mortar was obtained
in this study. Besides, as demonstrated by previously conducted research [20], [23], [24], [27],
[210], the relation between compressive stress and HOH was established. Thus, this correlation
provides trustworthy information for determining optimum sulfate levels. Consequently,
isothermal calorimetry can be used to estimate the optimum sulfate level of cement-slag systems
instead of using traditional compressive strength testing at an early age. Calorimetry provides more
reliable, sensitive results than the compressive stress tests, and is less labor intensive. It can provide
a broader range of results.
69
Figure 4.9 Optimum sulfate level for the compression strength versus optimum sulfate level for
total heat of hydration (HOH). The linear fit and 95% prediction interval on the data
4.4 Factors Affecting the Sulfate Demand in the OPC-Slag System
4.4.1 The Effect of Alumina (Al2O3) in the Slag
It is obviously seen that (see Table 4.2), the same cement groups are mixed and reacted
with different slag, they require different amount of SO3. In this research, Slag 4 requires more
SO3 than Slag 1. The reason for this phenomenon that Slag4 has higher Al2O3 than Slag1 (see
Table 3.1). Increases in the amount of alumina in the slag increase slag reactivity in the blended
system so that it would allow for more monosulfoaluminate and ettringite to be produced [126],
[211]. Previous studies noted that even though slag addition reduces aluminate content, the total
cementitious alumina content may increase, depending on the slag's alumina content. Researchers
who supported this theory have argued that while slag's alumina is not readily available to shape
70
Afm and Aft. Thus, it is stabilized with the addition of sulfate (extra hemihydrate in this research)
in the high-aluminum slag [9], [126], [211], [212]. Since ettringite incorporates sulfate in its
structure it would require additional sulfate. Therefore, alumina is one of the most important
factors to consider when optimizing SO3 content in Cement-Slag mixture.
4.4.2 The Effect of Cement Particle Size and Fineness
The fineness of cement and slag affects the rate of hydration. Increased surface area
enhances the solubility and hydration rate. Besides, since surface area increases due to fineness
increase, C3A and Al2O3 reactivity increases; as the reactive of C3A and alumina increases, an
additional SO3 compound will be required to control the hydration.
Cement C and cement D have almost the same mineralogical and chemical composition
although cement D has most fineness than other cements (see Table 3.5). The increase in fineness
increases the reactivity of the aluminate due to increases in the surface area. This increase in
reactivity in aluminate would necessitate additional sulfate to control the reaction. It seems that
the fineness of the cement would be an important consideration in determining the quantity of SO3
required.
4.4.3 The Effect of the Alumina with Mean Particle Size Distribution and Slag Activity
In hydrated cement-slag system, SO3 is consumed by Al2O3 that is present as crystalline
aluminate and ferrite of cement and amorphous slag. Generally, two main factors have an effect
on SO3 optimization: total amount of Al2O3 and particle sizes of cement and slag. Higher Al2O3
content of cement-slag system demands higher SO3 content. If mean particle size of cement or slag
71
is lower, the hydration rate of these materials will be higher, and, respectively, will be higher Al2O3
content of material in reaction with SO3.
Many hydraulic chemical modules have been described as indices of slag activity (ISA).
Hydraulic moduli are typically associated with GGBS compressive strength. The major oxides
CaO (C), Al2O3 (A), SiO2 (S), and MgO (M) are part of the common modules used for the
hydraulic reactivity estimation for slags, or compressive strength in blended mortars [10], [14],
[117], [128], [129]. Some of the most important well-known basicity index or hydraulic index
formulae are 𝑀1 =𝐶
𝑆, 𝑀2 =
𝐶+𝑀
𝑆, and 𝑀3 =
𝐶+𝑀+𝐴
𝑆 [10], [128], [129]. However; all modules
are convenient instruments for quickly assessing slag quality, the index M3 is the most widely used
[14]. Considering these points; researcher plotted two graphs.
𝑃𝑐𝑒𝑚𝑒𝑛𝑡𝐴𝑙2𝑂3𝑐𝑒𝑚
𝑑𝑐𝑒𝑚+ 𝑃𝑠𝑙𝑎𝑔
𝐴𝑙2𝑂3𝑠𝑙𝑎𝑔
𝑑𝑠𝑙𝑎𝑔𝛼𝐼𝑆𝐴 Eq. (4.2)
where,
𝑃𝑐𝑒𝑚𝑒𝑛𝑡 is the fraction of cement in OPC-Slag system
𝐴𝑙2𝑂3𝑐𝑒𝑚 is the alumina content of cement from XRF analysis
𝑃𝑠𝑙𝑎𝑔 is the fraction of slag in the OPC-Slag system
𝐴𝑙2𝑂3𝑠𝑙𝑎𝑔 is the alumina content of slag from XRF analysis
𝛼𝐼𝑆𝐴 is index of slag activity ( 𝐶+𝑀+𝐴
𝑆 )
𝑑𝑐𝑒𝑚 and 𝑑𝑠𝑙𝑎𝑔 are the mean particle size/mean diameter (in microns) for cement and slag,
respectively
72
Figure 4.10 The ratio of the cement alumina content to mean particle size of cement plus slag
alumina content to mean particle of slag size versus optimum SO3 level.
Figure 4.10 represents how the ratio of alumina content of cement to mean particle size
plus the normalized ratio of slags alumina content to mean particle size of slags with optimum SO3
content in mixtures (50S1 and 50S4 with Cem A-B-C-D). The researcher also shows how the ratio
of the cement's total amount of aluminate and ferrite to mean particle size of cement and the ratio
of alumina content of slag to the mean particle size of slags with optimum SO3 content in mixtures.
Figure 4.11 displays this relationship by using the determined optimum sulfate level in the
calorimeter experimental results.
y = 0.838x - 0.9189
R² = 0.8044
y = 0.5825x - 0.7267
R² = 0.954
0.00
0.30
0.60
0.90
1.20
1.50
1.80
0.00 1.00 2.00 3.00 4.00 5.00
(Al 2
O3/M
PS
(cem
)+(A
l 2O
3/M
PS
(sl)
)*α
ISA
SO3 optimum (%)
1-Day 3-Day
Linear (1-Day) Linear (3-Day)
73
𝑃𝑐𝑒𝑚𝑒𝑛𝑡(𝐶3𝐴+𝐶4𝐴𝐹)𝑐𝑒𝑚
𝑑𝑐𝑒𝑚+ 𝑃𝑠𝑙𝑎𝑔
𝐴𝑙2𝑂3𝑠𝑙𝑎𝑔
𝑑𝑠𝑙𝑎𝑔𝛼𝐼𝑆𝐴 Eq. (4.3)
where,
𝑃𝑐𝑒𝑚𝑒𝑛𝑡 is the fraction of cement in OPC-Slag system
(𝐶3𝐴 + 𝐶4𝐴𝐹)𝑐𝑒𝑚 is the aluminate and ferrite content of cement from XRD analysis
𝑃𝑠𝑙𝑎𝑔 is the fraction of slag in the OPC-Slag system
𝐴𝑙2𝑂3𝑠𝑙𝑎𝑔 is the alumina content of slag from XRF analysis
𝛼𝐼𝑆𝐴 is index of slag activity ( 𝐶+𝑀+𝐴
𝑆 )
𝑑𝑐𝑒𝑚 and 𝑑𝑠𝑙𝑎𝑔 are the mean particle size/mean diameter (in microns) for cement and slag,
respectively
Figure 4.11 Total amount of aluminate and ferrite to mean particle size of cement and the ratio of
alumina content of slag to the mean particle size of slags versus optimum SO3 level
y = 0.8731x - 0.5607
R² = 0.8296
y = 0.61x - 0.4515
R² = 0.9772
0.00
0.50
1.00
1.50
2.00
2.50
0.00 1.00 2.00 3.00 4.00 5.00
(C3A
+C
4A
F/M
PS
(cem
)+(A
l 2O
3/M
PS
(sl)
)*α
ISA
SO3 optimum (%)
1-Day 3-Day
Linear (1-Day) Linear (3-Day)
74
There is an excellent substantial agreement between the ratio of alumina content of cement
to mean particle size plus the normalized ratio of slags alumina content to mean particle size of
slags with optimum SO3 content in mixtures. Researcher also found a similar strong agreement
between how the ratio of the cement's total amount of aluminate and ferrite to mean particle size
of cement and the ratio of alumina content of slag to the mean particle size of slags with optimum
SO3 content in mixtures. As seen in the figures; 3-day blended samples correlation is much better
than 1-day samples. The reason is slag hydration rate is slower than cement hydration rate. This
contribution demonstrates itself at higher ages. Moreover, equation 4.3 can prove that the
aluminate effect on sulfate optimization and ferrite may affect sulfate optimization in the slag-
OPC system. Mainly, Slag 4 significantly improved the hydration rate of ferrite because of its high
slag activity. Slag 4, which is high alumina and alkali content, is more reactive than Slag1 and has
increased the hydration rate.
75
Chapter 5: Conclusions
5.1 Summary and Conclusions
From a historical perspective, the determination of sulfate content on cement and blended
cement has become a crucial problem for researchers. This research investigates how the sulfate
level could be conceptually optimized for slag and Portland cement are blended. The determination
of optimum SO3 content in this research has been made considering these materials' chemical and
physical characteristics. The chemical oxide compositions of the slags, hemihydrate, and cements
were carried out by using XRF and the mineralogical composition of the cements and slag were
determined by using XRD. The particle size distribution of as-received dry materials powders was
measured via Horiba LA-950 particle size analyzer.
A matrix consisting of four OPC with variable fineness, C3A, C4AF content of cement, and
two slags with low and high alumina contents was investigated for optimum SO3 content. A single
replacement level of 50% by mass of slag was used only at 23 ℃. The researcher studied the
effects of fineness, aluminate, ferrite content of cement, and alumina content of slag, based on
compressive strength and isothermal calorimetry testing at one and three days.
The researcher obtained a correlation between compressive strength and total heat of
standard mortar corroborate in this study. The total heat of hydration increased with increased SO3
contents in the cement-slag system; likewise, the strength increased. The heats of hydration
declined with the higher SO3 levels in the blended system; in the same way, the strength decreased.
76
Thus, isothermal calorimetry can estimate the optimum sulfate level of cement-slag systems
instead of traditional compressive strength testing at an early age. Calorimetry provides more
reliable, sensitive results than the compressive stress tests, and calorimeter tests are less labor
intensive than the compressive test. It can provide a broader range of results.
The findings indicate a strong influence of aluminate, ferrite content of cement, and Al2O3
content of slag and mean particle size and fineness of both cement and slag on optimum SO3
content based on the heat of hydration and compressive strength testing. Higher alumina amount
in the slag increases slag reactivity in the blended system. Therefore, alumina is one of the most
important factors to consider when optimizing SO3 content in the OPC-slag mixture. Moreover,
increased surface area enhances the solubility and hydration rate. Besides, since surface area
increases due to fineness increase, aluminate and Al2O3 reactivity increases; as the reactive of C3A
and alumina increases, an additional SO3 compound will be required to control the hydration. In a
hydrated cement-slag system, SO3 is consumed by Al2O3 that present mainly in crystalline
aluminate and ferrite of cement and amorphous of slag. Generally, two main factors affect SO3
optimization: the total amount of Al2O3 and particle sizes of cement and slag. Higher Al2O3 content
of cement-slag system demands higher SO3 content. If the mean particle size of cement or slag is
lower, the hydration rate of these materials will be higher, and, higher Al2O3 content will react
with SO3.
Since alumina and aluminate consume more sulfate over time, the degree of hydration of
slag and cement increases; thereby, the optimal sulfate level increases with age. However, adapting
Lerch's work [20] to this study, a properly retarded cement-slag system needs to contain the
minimum amount of sulfate sources (e.g., gypsum and hemihydrate) necessary to generate a curve
that demonstrates two cycles of increasing and decreasing heat release rates. Also, as seen in
77
Figure 4.6, after some point heat flow rate stabilizes and gets close to zero. Thus, a convenient
optimum SO3 level and proper optimization age may be specified depending on the hydration heat
flow.
5.2 Future Work
This optimization research can be developed and extended in several ways. In order to
improve the accuracy and reliability of the findings, more cements, slags, different calcium sulfate
forms (especially gypsum and hemihydrate), and chemical admixture can be tested at various
temperatures. It can help recognize whether the slag and other phases affect the optimum sulfate
concentration. Additionally, it enables increasing the number of parameters for the optimum
cement-slag mixture model, if it will be needed. Furthermore, the increased temperature can
increase the hydration rate of slag and cement, but also, particularly over 40 ℃, some phases can
be converted into some other phase (e.g., the transformation from ettringite to monosulfoaluminate
[213] ). For all these factors, the optimal sulfate amount on the cement-slag system may shift.
A more detailed understanding of sulfate's effect on hydration can be obtained through
developing a thermodynamic model. It may be possible to better understand the hydration
processes by better characterizing and identifying the slag's amorphous phases. An alternative
mixture with a different slag/cement ratio can be prepared and examined to how increasing or
decreasing slag amount in the system affects the sulfate optimization. It can be used to design high-
volume slag-OPC systems to deal with the issues encountered when developing that application,
including poor early age strength.
78
References
[1] K. Hanson, “SCMs in Concrete,” National Precast Concrete Association, 2020. [Online].
Available: https://precast.org/2017/03/scms-in-concrete/. [Accessed: 02-Jul-2020].
[2] S. Mindess, J. F. Young, and D. Darwin, Concrete, 2nd ed. Upper Saddle River, NJ:
Prentice Hall, 2003.
[3] S. Dubey, A. Singh, and S. S. Kushwah, “Utilization of iron and steel slag in building
construction,” in International Conference on Sustainable Materials and Structures for
Civil Infrastructures (SMSCI2019), 2019, vol. 020032, no. 2158, pp. 1–6, doi:
https://doi.org/10.1063/1.5127156.
[4] N. G. Shanahan, “Interaction of Cementitious Systems with Chemical Admixtures,”
University of South Florida, 2016.
[5] U.S. Department of Transportation, “Use of Supplementary Cementitious Materials (
SCMs ) in Concrete Mixtures.” [Online]. Available: rosap.ntl.bts.gov › view › dot ›
dot_43714_DS1. [Accessed: 02-Oct-2020].
[6] K. Hanson, “SCMs in Concrete,” National Precast Concrete Association, 2020. .
[7] Global Cement, “World crude steel production grows by 4.9% to 925Mt in first half of
2019,” Global Slag News, pp. 1–2, Jul-2020.
[8] K. C. Curry, “Iron and steel slag,” 2020.
[9] F. Nosouhian, M. Fincan, N. Shanahan, Y. P. Stetsko, K. A. Riding, and A. Zayed,
“Effects of slag characteristics on sulfate durability of Portland cement-slag blended
systems,” Constr. Build. Mater., vol. 229, pp. 1–11, 2019, doi:
10.1016/j.conbuildmat.2019.116882.
[10] V. Kocaba, “Development and evaluation of methods to follow microstructural
development of cementitious systems including slags,” ÉCOLE POLYTECHNIQUE
FÉDÉRALE DE LAUSANNE, 2009.
[11] Y. Ueki, “History and Utilization of Portland Blast Furnace Slag Cement,” Fukuoka,
2015.
79
[12] K. Yang, Y. Jung, M. Cho, and S. Tae, “Effect of supplementary cementitious materials
on reduction of CO2 emissions from concrete,” J. Clean. Prod., vol. 103, pp. 774–783,
2015, doi: 10.1016/j.jclepro.2014.03.018.
[13] R. Siddique and M. I. Khan, Supplementary Cementing Materials. Berlin, Heidelberg:
Springer-Verlag, 2011.
[14] A. A. Ramezanianpour, Cement Replacement Materials. Berlin Heidelberg: Springer
Geochemistry/Mineralogy, 2014.
[15] H. Lee, X. Wang, L. Zhang, and K. Koh, “Analysis of the Optimum Usage of Slag for the
Compressive Strength of Concrete,” Materials (Basel)., vol. 8, no. 3, pp. 1213–1229,
2015, doi: 10.3390/ma8031213.
[16] A. Oner and S. Akyuz, “An experimental study on optimum usage of GGBS for the
compressive strength of concrete,” Cem. Concr. Compos., vol. 29, no. 6, pp. 505–514,
2007, doi: 10.1016/j.cemconcomp.2007.01.001.
[17] S. Kourounis, S. Tsivilis, P. E. Tsakiridis, G. D. Papadimitriou, and Z. Tsibouki,
“Properties and hydration of blended cements with steelmaking slag,” Cem. Concr. Res.,
vol. 37, pp. 815–822, 2007, doi: 10.1016/j.cemconres.2007.03.008.
[18] W. Qiang, Y. Peiyu, Y. Jianwei, and Z. Bo, “Influence of steel slag on mechanical
properties and durability of concrete,” Constr. Build. Mater., vol. 47, pp. 1414–1420,
2013, doi: 10.1016/j.conbuildmat.2013.06.044.
[19] H. F. W. Taylor, Cement chemistry, 2nd ed. New York: Thomas Telford, 1997.
[20] W. Lerch, “The Influence of Gypsum on the Hydration and Properties of Portland Cement
Pastes,” Proc. Am. Soc. Test. Mater., vol. 46, pp. 1–42, 1946.
[21] P. Sandberg and S. Bishnoi, “Calcined Clays for Sustainable Concrete,” in 2nd
International Conference on Calcined Clays for Sustainable Concrete, 2018, pp. 422–426,
doi: https://doi.org/10.1007/978-94-024-1207-9.
[22] C. Mei-zhu, L. Wen-quan, H. Zhen, and L. Bei-xing, “Optimization of SO3 content in
blended cementitious materials,” J. Wuhan Univ. Technol. Sci. Ed., vol. 18, no. 3, pp. 89–
93, 2003, doi: 10.1007/BF02838470.
[23] M. D. Niemuth, “EFFECT OF FLY ASH ON THE OPTIMUM SULFATE OF
PORTLAND CEMENT,” Purdue University, 2012.
80
[24] M. Niemuth, L. Barcelo, and J. Weiss, “Effect of fly ash on optimum sulfate levels
measured using heat and strength at early ages,” Adv. Civ. Eng. Mater., vol. 1, no. 1, pp.
1–18, 2012, doi: 10.1520/ACEM20120012.
[25] S. Mohammed and O. Safiullah, “Optimization of the SO3content of an Algerian Portland
cement: Study on the effect of various amounts of gypsum on cement properties,” Constr.
Build. Mater., vol. 164, pp. 362–370, 2018, doi: 10.1016/j.conbuildmat.2017.12.218.
[26] S. Kucharczyk, J. Deja, and M. Zajac, “Effect of slag reactivity influenced by alumina
content on hydration of composite cements,” J. Adv. Concr. Technol., vol. 14, pp. 535–
547, 2016, doi: 10.3151/jact.14.535.
[27] L. Frølich, L. Wadsö, and P. Sandberg, “Using isothermal calorimetry to predict one day
mortar strengths,” Cem. Concr. Res., vol. 88, pp. 108–113, 2016, doi:
10.1016/j.cemconres.2016.06.009.
[28] ASTM C109/C109M-16a, Standard Test Method for Compressive Strength of Hydraulic
Cement Mortars ( Using 2-in . or [ 50-mm ] Cube Specimens ). West Conshohocken, PA:
ASTM International, 2016.
[29] M. Tokyay, Cement and concrete mineral admixtures. Boca Raton, FL: CRC Press, 2016.
[30] D. Herfort and D. E. Macphee, “Components in Portland Cement Clinker and Their Phase
Relationships Duncan,” in Lea’s Chemistry of Cement and Concrete, 5th ed., Elsevier
Ltd., 2019, pp. 57–86.
[31] A. V. Soin, L. J. J. Catalan, and S. D. Kinrade, “A combined QXRD/TG method to
quantify the phase composition of hydrated Portland cements,” Cem. Concr. Res., vol. 48,
pp. 17–24, 2013, doi: 10.1016/j.cemconres.2013.02.007.
[32] D. D. Double, A. Hellawell, and S. J. Perry, “The Hydration of Portland Cement,” Proc.
R. Soc. Lond. A. Math. Phys. Sci., vol. 359, no. 1699, pp. 435–451, 1978, doi:
10.1098/rspa.1978.0050.
[33] I. Odler, “Hydration, Setting and Hardening of Portland Cement,” in Lea’s Chemistry of
Cement and Concrete, 4th ed., P. Hewlett, Ed. Elsevier Science, 2003, pp. 241–297.
[34] Portland Cement Association, “Portland cement, concrete, and heat of Hydration,” Concr.
Technol. Today, vol. 18, no. 2, pp. 1–4, 1997.
[35] P. K. Mehta and P. J. M. Monteiro, Concrete: microstructure, properties, and materials,
Third. the United States of America: McGraw-Hill, 2006.
81
[36] J. D. Bapat, Mineral admixtures in cement and concrete. Boca Raton, FL: CRC Press,
2012.
[37] I. G. Richardson, “Nature of C-S-H in hardened cements,” Cem. Concr. Res., vol. 29, no.
8, pp. 1131–1147, 1999, doi: 10.1016/S0008-8846(99)00168-4.
[38] I. Odler and R. Wonnemann, “Effect of alkalies on portland cement hydration II. Alkalies
present in form of sulphates,” Cement and Concrete Research, vol. 13, no. 6. pp. 771–777,
1983, doi: 10.1016/0008-8846(83)90078-9.
[39] J. Beaudoin and I. Odler, “Hydration, Setting and Hardening of Portland Cement,” in
Lea’s Chemistry of Cement and Concrete, 5th ed., P. Hewlett and M. Liska, Eds. Elsevier
Ltd., 2019, pp. 157–250.
[40] D. Marchon and R. J. Flatt, “Mechanisms of cement hydration,” Sci. Technol. Concr.
Admixtures, vol. 41, pp. 129–145, 2015, doi: 10.1016/B978-0-08-100693-1.00008-4.
[41] P. Juilland, E. Gallucci, R. Flatt, and K. Scrivener, “Dissolution theory applied to the
induction period in alite hydration,” Cem. Concr. Res., vol. 40, no. 6, pp. 831–844, 2010,
doi: 10.1016/j.cemconres.2010.01.012.
[42] K. L. Scrivener, P. Juilland, and P. J. M. Monteiro, “Advances in understanding hydration
of Portland cement,” Cem. Concr. Res., vol. 78, pp. 38–56, 2015, doi:
10.1016/j.cemconres.2015.05.025.
[43] K. L. Scrivener, P. Juilland, and P. J. M. Monteiro, “Advances in understanding hydration
of Portland cement,” Cem. Concr. Res., vol. 78, pp. 38–56, 2015, doi:
10.1016/j.cemconres.2015.05.025.
[44] A. Bentur, N. B. Milestone, J. F. Young, and S. Mindess, “Creep and drying shrinkage of
calcium silicate pastes IV. Effects of accelerated curing,” Cement and Concrete Research,
vol. 9, no. 2. pp. 161–169, 1979, doi: 10.1016/0008-8846(79)90022-X.
[45] F. Begarin, S. Garrault, a Nonat, and L. Nicoleau, “Hydration of alite containing
aluminium,” Adv. Appl. Ceram., vol. 110, no. 3, pp. 127–130, 2011, doi:
10.1179/1743676110Y.0000000007.
[46] J. W. Bullard et al., “Mechanisms of cement hydration,” Cem. Concr. Res., vol. 41, no.
12, pp. 1208–1223, 2011, doi: 10.1016/j.cemconres.2010.09.011.
[47] S. Pourchet, L. Regnaud, J. P. Perez, and A. Nonat, “Early C3A hydration in the presence
of different kinds of calcium sulfate,” Cem. Concr. Res., vol. 39, no. 11, pp. 989–996,
2009, doi: 10.1016/j.cemconres.2009.07.019.
82
[48] E. M. Gartner, J. F. Young, D. Damidot, and I. Jawed, “Hydration of Portland cement,” in
Structure and Performance of Cements, 2nd Editio., J. Bensted and P. Barnes, Eds. Spon
Press, 2002, pp. 57–114.
[49] A. Quennoz, “Hydration of C3A with Calcium Sulfate Alone and in the Presence of
Calcium Silicate,” Ecole Polytechnique Fédérale de Lausanne, 2011.
[50] H. Wang, D. De Leon, and H. Farzam, “C4AF reactivity-chemistry and hydration of
industrial cement,” ACI Mater. J., vol. 111, no. 2, pp. 201–210, 2014, doi:
10.14359/51686504.
[51] I. Odler, “Strength of cement (final report),” Mater. Struct., vol. 24, no. 2, pp. 143–157,
1991, doi: 10.1007/BF02472476.
[52] I. Odler, Special Inorganic Cements. London: E&FN Spon, 2000.
[53] P. Sandberg, “Optimization of Cement Sulfate Part I - Cement without admixture,” Ta,
vol. 46, pp. 1–3, 2004.
[54] T. Kishi and K. Maekawa, “Multi-component model for hydration heating of Portland
cement,” Concr. Libr. JSCE, vol. 28, no. JANUARY 1995, pp. 97–115, 1996, doi:
10.2208/jscej.1995.526.
[55] W. Lerch and R. H. Bogue, “Heat of hydration of portland cement pastes,” Bur. Stand. J.
Res., vol. 12, pp. 645–664, 1934, doi: 10.6028/jres.012.057.
[56] ACI Committee 116, “116R-00: Cement and Concrete Terminology,” Farmington Hills,
MI., 2000.
[57] P. K. Mehta, D. Pirtz, and M. Polivka, “Properties of alite cements,” Cem. Concr. Res.,
vol. 9, no. 4, pp. 439–450, 1979, doi: 10.1016/0008-8846(79)90041-3.
[58] K. L. Scrivener, P. Juilland, and P. J. M. Monteiro, “Advances in understanding hydration
of Portland cement,” Cem. Concr. Res., vol. 78, pp. 38–56, 2015, doi:
10.1016/j.cemconres.2015.05.025.
[59] E. S. Newman, “A study of the determination of the heat of hydration of portland
cement,” J. Res. Natl. Bur. Stand. (1934)., vol. 45, no. 5, p. 411, 1950, doi:
10.6028/jres.045.045.
[60] B. Guo et al., “Immobilization mechanism of Pb in fly ash-based geopolymer,” Constr.
Build. Mater., vol. 134, pp. 123–130, 2017, doi: 10.1016/j.conbuildmat.2016.12.139.
83
[61] A. Markandeya et al., “Chemical shrinkage and cracking resilience of metakaolin
concrete,” ACI Mater. J., vol. 116, no. 4, pp. 99–106, 2019, doi: 10.14359/51716714.
[62] V. Tydlitát, J. Zákoutský, P. Volfová, and R. Černý, “Hydration heat development in
blended cements containing fine-ground ceramics,” Thermochim. Acta, vol. 543, pp. 125–
129, 2012, doi: 10.1016/j.tca.2012.05.022.
[63] B. Lothenbach, K. Scrivener, and R. D. Hooton, “Supplementary cementitious materials,”
Cem. Concr. Res., vol. 41, pp. 1244–1256, 2011, doi: 10.1016/j.cemconres.2010.12.001.
[64] R. Snellings, G. Mertens, and J. Elsen, “Supplementary cementitious materials,” Rev.
Mineral. Geochemistry, vol. 74, pp. 211–278, 2012, doi: 10.2138/rmg.2012.74.6.
[65] E. M. J. BERODIER, “Impact of the Supplementary Cementitious Materials on the
kinetics and microstructural development of cement hydration PAR,” ÉCOLE
POLYTECHNIQUE FÉDÉRALE DE LAUSANNE, 2015.
[66] ASTM C125-16, “Standard Terminology Relating to Concrete and Concrete Aggregates,”
in ASTM International, West Conshohocken, PA: ASTM International, 2016, pp. 1–4.
[67] M. J. McCarthy and T. D. Dyer, “Pozzolanas and pozzolanic materials,” in Lea’s
Chemistry of Cement and Concrete, 5th ed., P. Hewlett and M. Liska, Eds. Elsevier Ltd.,
2019, pp. 363–467.
[68] R. J. Myers, “Thermodynamic Modelling of CaO-Al2O3-SiO2-H2O-Based Cements,” no.
May, p. 362, 2015.
[69] E. M. J. BERODIER, “Impact of the Supplementary Cementitious Materials on the
kinetics and microstructural development of cement hydration,” vol. 6417, p. 154, 2015.
[70] B. Lothenbach, K. Scrivener, and R. D. Hooton, “Supplementary cementitious materials,”
Cem. Concr. Res., vol. 41, no. 12, pp. 1244–1256, 2011, doi:
10.1016/j.cemconres.2010.12.001.
[71] ASTM C 1240-15, “Standard Specification for Silica Fume Used in Cementitious
Mixtures,” in Annual Book of ASTM Standards, West Conshohocken, PA: ASTM
International, 2015, pp. 1–7.
[72] M. Mazloom, A. A. Ramezanianpour, and J. J. Brooks, “Effect of silica fume on
mechanical properties of high-strength concrete,” Cem. Concr. Compos., vol. 26, pp. 347–
357, 2004, doi: 10.1016/S0958-9465(03)00017-9.
84
[73] Z. Q. Wu and J. F. Young, “The hydration of tricalcium silicate in the presence of
colloidal silica,” J. Mater. Sci., vol. 19, pp. 3477–3486, 1984, doi: 10.1007/BF00552262.
[74] X. Wang, P. Hou, J. Yu, X. Zhou, and X. Cheng, “The effects of silica fume on C3A
hydration,” Constr. Build. Mater., vol. 250, pp. 1–10, 2020, doi:
10.1016/j.conbuildmat.2020.118766.
[75] J. Zelić, D. Rušić, D. Veža, and R. Krstulović, “Role of silica fume in the kinetics and
mechanisms during the early stage of cement hydration,” Cem. Concr. Res., vol. 30, no.
10, pp. 1655–1662, 2000, doi: 10.1016/S0008-8846(00)00374-4.
[76] K. L. Scrivener et al., “TC 238-SCM: hydration and microstructure of concrete with
SCMs,” Mater. Struct., vol. 48, no. 4, pp. 835–862, 2015, doi: 10.1617/s11527-015-0527-
4.
[77] J. T. Ding and Z. Li, “Effects of metakaolin and silica fume on properties of concrete,”
ACI Mater. J., vol. 99, no. 4, pp. 393–398, 2002.
[78] M. G. Alexander and B. J. Magee, “Durability performance of concrete containing
condensed silica fume,” Cem. Concr. Res., vol. 29, no. 6, pp. 917–922, 1999, doi:
10.1016/S0008-8846(99)00064-2.
[79] R. D. Hooton, “Influence of silica fume replacement of cement on physical properties and
resistance to sulfate attack, freezing and thawing, and alkali-silica reactivity,” ACI Mater.
J., vol. 90, no. 2, pp. 143–151, 1993, doi: 10.14359/4009.
[80] A. Mardani-Aghabaglou, G. Inan Sezer, and K. Ramyar, “Comparison of fly ash, silica
fume and metakaolin from mechanical properties and durability performance of mortar
mixtures view point,” Constr. Build. Mater., vol. 70, pp. 17–25, 2014, doi:
10.1016/j.conbuildmat.2014.07.089.
[81] R. Duval and E. H. Kadri, “Influence of Silica Fume on the Workability and the
Compressive Strength of High-Performance Concretes,” Cem. Concr. Res., vol. 28, no. 4,
pp. 533–547, 1998.
[82] W. Wongkeo, P. Thongsanitgarn, and A. Chaipanich, “Compressive strength and drying
shrinkage of fly ash-bottom ash-silica fume multi-blended cement mortars,” Mater. Des.,
vol. 36, pp. 655–662, 2012, doi: 10.1016/j.matdes.2011.11.043.
[83] S. Wild, B. B. Sabir, and J. M. Khatib, “Factors influencing strength development of
concrete containing silica fume,” Cem. Concr. Res., vol. 25, no. 7, pp. 1567–1580, 1995,
doi: 10.1016/0008-8846(95)00150-B.
85
[84] G. A. Rao, “Development of strength with age of mortars containing silica fume,” Cem.
Concr. Res., vol. 31, no. 8, pp. 1141–1146, 2001, doi: 10.1016/S0008-8846(01)00540-3.
[85] H. A. Toutanji and T. El-Korchi, “The influence of silica fume on the compressive
strength of cement paste and mortar,” Cem. Concr. Res., vol. 25, no. 7, pp. 1591–1602,
1995, doi: 10.1016/0008-8846(95)00152-3.
[86] E.-H. Kadri, R. Duval, S. Aggoun, and S. Kenai, “Silica Fume Effect on Hydration Heat
and Compressive Strength of High Performance Concrete,” ACI Mater. J., no. 106, pp.
107–113, 2009.
[87] R. Siddique and J. Klaus, “Influence of metakaolin on the properties of mortar and
concrete: A review,” Appl. Clay Sci., vol. 43, pp. 392–400, 2009, doi:
10.1016/j.clay.2008.11.007.
[88] A. M. Rashad, “Metakaolin as cementitious material: History, scours, production and
composition-A comprehensive overview,” Constr. Build. Mater., vol. 41, pp. 303–318,
2013, doi: 10.1016/j.conbuildmat.2012.12.001.
[89] B. Sabir, S. Wild, and J. Bai, “Metakaolin and calcined clays as pozzolans for concrete: A
review,” Cem. Concr. Compos., vol. 23, no. 6, pp. 441–454, 2001, doi: 10.1016/S0958-
9465(00)00092-5.
[90] F. K. Moghaddam, R. S. Ravindrarajah, and V. Sirivivatnanon, “Properties of Metakaolin
Concrete – a Review,” in Int. Conference on Sustainable Structural Concrete, 2015, pp.
1–10.
[91] M. Murat, “Hydration reaction and hardening of calcined clays and related minerals. I.
Preliminary investigation on metakaolinite,” Cem. Concr. Res., vol. 13, no. 2, pp. 259–
266, 1983, doi: 10.1016/0008-8846(83)90109-6.
[92] S. Wild, J. M. Khatib, and L. J. Roose, “Chemical shrinkage and autogenous shrinkage of
Portland cement-metakaolin pastes,” Adv. Cem. Res., vol. 10, no. 3, pp. 109–119, 1998,
doi: 10.1680/adcr.1998.10.3.109.
[93] J. M. Khatib and S. Wild, “Pore size distribution of metakaolin paste,” Cem. Concr. Res.,
vol. 26, no. 10, pp. 1545–1553, 1996, doi: 10.1016/0008-8846(96)00147-0.
[94] P. De Silva and F. Glasser, “Pozzolanic activation of metakaolin,” Adv. Cem. Res., vol. 4,
no. 16, pp. 167–78, 1992, doi: 10.1680/adcr.1992.4.16.167.
86
[95] F. Moodi, A. A. Ramezanianpour, and A. S. Safavizadeh, “Evaluation of the optimal
process of thermal activation of kaolins,” Sci. Iran., vol. 18, no. 4 A, pp. 906–912, 2011,
doi: 10.1016/j.scient.2011.07.011.
[96] J. J. Brooks and M. A. Megat Johari, “Effect of metakaolin on creep and shrinkage of
concrete,” Cem. Concr. Compos., vol. 23, no. 6, pp. 495–502, 2001, doi: 10.1016/S0958-
9465(00)00095-0.
[97] C. S. Poon, L. Lam, S. C. Kou, Y. L. Wong, and R. Wong, “Rate of pozzolanic reaction of
metakaolin in high-performance cement pastes,” Cem. Concr. Res., vol. 31, no. 9, pp.
1301–1306, 2001, doi: 10.1016/S0008-8846(01)00581-6.
[98] C. S. Poon, S. C. Kou, and L. Lam, “Compressive strength, chloride diffusivity and pore
structure of high performance metakaolin and silica fume concrete,” Constr. Build.
Mater., vol. 20, no. 10, pp. 858–865, 2006, doi: 10.1016/j.conbuildmat.2005.07.001.
[99] K. A. Gruber, T. Ramlochan, A. Boddy, R. D. Hooton, and M. D. A. Thomas, “Increasing
concrete durability with high-reactivity metakaolin,” Cem. Concr. Compos., vol. 23, no. 6,
pp. 479–484, 2001, doi: 10.1016/S0958-9465(00)00097-4.
[100] H. E. D. H. Seleem, A. M. Rashad, and B. A. El-Sabbagh, “Durability and strength
evaluation of high-performance concrete in marine structures,” Constr. Build. Mater., vol.
24, pp. 878–884, 2010, doi: 10.1016/j.conbuildmat.2010.01.013.
[101] L. Mlinárik and K. Kopecskó, “Impact of metakaolin - a new supplementary material - on
the hydration mechanism of cements,” Acta Tech. Napocensis Civ. Eng. Archit., vol. 56,
no. 2, pp. 100–110, 2013.
[102] T. R. Jones, “Metakaolin as a pozzolanic addition to concrete,” in Structure and
Performance of Cements, 2nd ed., J. Bensted and P. Barnes, Eds. New York, NY: Spon
Press, 2002, pp. 372–398.
[103] Astm C618-17, “Standard Specification for Coal Fly Ash and Raw or Calcined Natural
Pozzolan for Use in Concrete,” West Conshohocken, PA, 2017.
[104] ACI committee 232, “Use of Fly Ash in Concrete,” ACI Mater. J., vol. 96, pp. 1–34, 2002.
[105] “Fly Ash , Slag , Silica Fume , and Natural Pozzolans.” [Online]. Available:
http://www.ce.memphis.edu/1101/notes/concrete/PCA_manual/Chap03.pdf. [Accessed:
03-May-2020].
[106] M. Ahmaruzzaman, “A review on the utilization of fly ash,” Prog. Energy Combust. Sci.,
vol. 36, no. 3, pp. 327–363, 2010, doi: 10.1016/j.pecs.2009.11.003.
87
[107] Y. Kocak and S. Nas, “The effect of using fly ash on the strength and hydration
characteristics of blended cements,” Constr. Build. Mater., vol. 73, pp. 25–32, 2014, doi:
10.1016/j.conbuildmat.2014.09.048.
[108] P. L. J. Domone, Construction materials: Their nature and behavior, Third. New York:
Spon Press, 2002.
[109] W. vo. Berg and H. Kukko, “FRESH MORTAR AND CONCRETE WITH FLY ASH,” in
Fly Ash in Concrete Properties and Performance (RILEM Report), K. Wesche, Ed. New
York: E&FN SPON, 1990, pp. 24–41.
[110] K. D. Stuart, D. A. Anderson, and P. D. Cady, “Compressive strength studies of Portland
cement mortars containing fly ash and superplasticizers,” Cem. Concr. Res., vol. 11, pp.
805–806, 1981.
[111] S. Alsadey, “Influence of Fly Ash Cement Replacement on Workability and Strength of
Concrete,” Aust. J. Manag. Policy Law SCIE Journals, pp. 7–12.
[112] E. Sakai, S. Miyahara, S. Ohsawa, S. H. Lee, and M. Daimon, “Hydration of fly ash
cement,” Cem. Concr. Res., vol. 35, pp. 1135–1140, 2005, doi:
10.1016/j.cemconres.2004.09.008.
[113] ACI Committee 211, “Standard Practice for Selecting Proportions for Normal,
Heavyweight, and Mass Concrete (ACI 211.1-91),” Am. Concr. Inst., pp. 1–38, 2002.
[114] C. S. Poon, L. Lam, and Y. L. Wong, “A study on high strength concrete prepared with
large volumes of low calcium fly ash,” Cem. Concr. Res., vol. 30, pp. 447–455, 2000, doi:
10.1016/S0008-8846(99)00271-9.
[115] G. Li and X. Zhao, “Properties of concrete incorporating fly ash and ground granulated
blast-furnace slag,” Cem. Concr. Compos., vol. 25, no. 3, pp. 293–299, 2003, doi:
10.1016/S0958-9465(02)00058-6.
[116] M. Moranville-Regourd and S. Kamali-Bernard, “Cements made from blastfurnace slag,”
in Lea’s Chemistry of Cement and Concrete, 5th ed., Elsevier Ltd., 2019, pp. 469–507.
[117] H. S. Pietersen, “Reactivity of Fly Ash and Slag In Cement,” Delft University of
Technology, 1993.
[118] ACI Committee 233, “Slag Cement in Concrete and Mortar: ACI 233R-03,” Farmington
Hills, MI, 2011.
88
[119] ACI Committee 233, “Ground Granulated Blast-Furnace Slag As a Cementitious
Constituent in Concrete. ACI 233R-95,” 2000.
[120] D. W. Lewls, “History of Slag Cements,” Natl. Slag Assoc., vol. MF 181-6, pp. 1–9, 1981.
[121] G. C. Wang, The utilization of slag in civil infrastructure construction. Woodhead
Publishing, 2016.
[122] F. Hogan and J. Meusel, “Evaluation for durability and strength development of a ground
granulated blast furnace slag,” Cem. Concr. Aggregates, vol. 87, no. 3, pp. 40–52, 1981,
doi: 10.1520/CCA10201J.
[123] W. Chen, “Hydration of slag cement. Theory, modeling and application,” University of
Twente, 2007.
[124] B. Kolani, L. Buffo-lacarrière, A. Sellier, G. Escadeillas, L. Boutillon, and L. Linger,
“Hydration of slag-blended cements,” Cem. Concr. Compos., vol. 34, no. 9, pp. 1009–
1018, 2012, doi: 10.1016/j.cemconcomp.2012.05.007.
[125] ASTM C989/C989M-17, “Standard Specification for Slag Cement for Use in Concrete
and Mortars,” in ASTM Standards, West Conshohocken, PA: ASTM International, 2017,
pp. 1–7.
[126] A. Zayed et al., “Effects of Blast Furnace Slag Characteristics on Durability of
Cementitious Systems for Florida Concrete Structures,” Tallahassee,,FL, 2019.
[127] M. Cheron and C. Lardinois, “The Role of Magnesia and Alumina in the Hydraulic
Properties of Granulated Blast-Furnace Slags,” 5th Int. Congr. Chem. Cem. Part 4, pp.
277–285, 1968.
[128] S. C. Pal, A. Mukherjee, and S. R. Pathak, “Investigation of hydraulic activity of ground
granulated blast furnace slag in concrete,” Cem. Concr. Res., vol. 33, pp. 1481–1486,
2003, doi: 10.1016/S0008-8846(03)00062-0.
[129] E. Lang, “Blastfurnace cements,” in Structure and Performance of Cements, 2nd ed., J.
Bensted and P. Barnes, Eds. New York, NY: Spon Press, 2002, pp. 310–326.
[130] EN197-1, “Cement Part 1: Composition, Specifications and Conformity Criteria for
Common Cements,” 2000.
[131] BS-EN197-1, “Cement Part 1: Composition, Specifications and Conformity Criteria for
Common Cements,” 2011.
89
[132] I. G. Richardson and G. W. Groves, “Microstructure and microanalysis of hardened
cement pastes involving ground granulated blast-furnace slag,” J. Mater. Sci., vol. 27, no.
22, pp. 6204–6212, 1992, doi: 10.1007/BF01133772.
[133] I. Richardson and G. Groves, “The structure of the calcium silicate hydrate phases present
in hardened pastes of white Portland cement/blast-furnace slag blends,” J. Mater. Sci., vol.
32, pp. 4793–4802, 1997, doi: 10.1023/A:1018639232570.
[134] P. J. Wainwright and H. Ait-Aider, “The influence of cement source and slag additions on
the bleeding of concrete,” Cem. Concr. Res., vol. 25, no. 7, pp. 1445–1456, 1995, doi:
10.1016/0008-8846(95)00139-4.
[135] ASTM C232/C232M − 14, “Standard Test Methods for Bleeding of Concrete,” in ASTM
Standards, West Conshohocken, PA: ASTM International, 2014, pp. 1–3.
[136] P. J. Wainwright and N. Rey, “Influence of ground granulated blastfurnace slag (GGBS)
additions and time delay on the bleeding of concrete,” Cem. Concr. Compos., vol. 22, no.
4, pp. 253–257, 2000, doi: 10.1016/S0958-9465(00)00024-X.
[137] ACI committee 226, “Ground Granulated Blast-Furnace Slag As a Cementitious
Constituent in Concrete.,” ACI Mater. J., vol. 84, no. 4, pp. 327–342, 1987.
[138] W. Matthes et al., Properties of Fresh and Hardened Concrete Containing Supplementary
Cementitious Materials: State-of-the-Art Report of the RILEM Technical Committee 238-
SCM, Working Group 4. Cham,Switzerland: Springer Nature, 2018.
[139] M. Thomas, Supplementary Cementing Materials in Concrete. Boca Raton, FL: CRC
Press, 2013.
[140] E. Özbay, M. Erdemir, and H. I. Durmuş, “Utilization and efficiency of ground granulated
blast furnace slag on concrete properties - A review,” Constr. Build. Mater., vol. 105, pp.
423–434, 2016, doi: 10.1016/j.conbuildmat.2015.12.153.
[141] D. P. Bentz, E. J. Garboczi, C. J. Haecker, and O. M. Jensen, “Effects of cement particle
size distribution on performance properties of Portland cement-based materials,” Cem.
Concr. Res., vol. 29, no. 10, pp. 1663–1671, 1999, doi: 10.1016/S0008-8846(99)00163-5.
[142] I. K. Labarca, R. D. Foley, and S. Cramer, “Effects of Ground Granulated Blast Furnace
Slag in Portland Cement Concrete ( PCC ) - Expanded Study,” Madison,WI, 2007.
[143] K. Ganesh Babu and V. Sree Rama Kumar, “Efficiency of GGBS in concrete,” Cem.
Concr. Res., vol. 30, no. 7, pp. 1031–1036, 2000, doi: 10.1016/S0008-8846(00)00271-4.
90
[144] H. Wan, Z. Shui, and Z. Lin, “Analysis of geometric characteristics of GGBS particles and
their influences on cement properties,” Cem. Concr. Res., vol. 34, pp. 133–137, 2004, doi:
10.1016/S0008-8846(03)00252-7.
[145] E. Güneyisi and M. Gesoğlu, “A study on durability properties of high-performance
concretes incorporating high replacement levels of slag,” Mater. Struct., vol. 41, no. 3, pp.
479–493, 2008, doi: 10.1617/s11527-007-9260-y.
[146] A. M. Rashad, “An overview on rheology, mechanical properties and durability of high-
volume slag used as a cement replacement in paste, mortar and concrete,” Constr. Build.
Mater., vol. 187, pp. 89–117, 2018, doi: 10.1016/j.conbuildmat.2018.07.150.
[147] G. J. Osborne, “Durability of Portland blast-furnace slag cement concrete,” Cem. Concr.
Compos., vol. 21, no. 1, pp. 11–21, 1999, doi: 10.1016/S0958-9465(98)00032-8.
[148] B. Abdelkader, K. El-Hadj, and E. Karim, “Efficiency of granulated blast furnace slag
replacement of cement according to the equivalent binder concept,” Cem. Concr.
Compos., vol. 32, pp. 226–231, 2010, doi: 10.1016/j.cemconcomp.2009.11.004.
[149] A. Cheng, R. Huang, J. K. Wu, and C. H. Chen, “Influence of GGBS on durability and
corrosion behavior of reinforced concrete,” Mater. Chem. Phys., vol. 93, pp. 404–411,
2005, doi: 10.1016/j.matchemphys.2005.03.043.
[150] E. Douglas, A. Bilodeau, and V. M. Malhotra, “Properties and durability of alkali-
activated slag concrete,” ACI Mater. J., vol. 89, no. 5, pp. 509–516, 1992, doi:
10.1617/s11527-014-0289-4.
[151] M. Gesoǧlu, E. Güneyisi, and E. Özbay, “Properties of self-compacting concretes made
with binary, ternary, and quaternary cementitious blends of fly ash, blast furnace slag, and
silica fume,” Constr. Build. Mater., vol. 23, no. 5, pp. 1847–1854, 2009, doi:
10.1016/j.conbuildmat.2008.09.015.
[152] G. J. Osborne, “Durability of Portland blast-furnace,” Cem. Concr. Compos., vol. 21, pp.
11–21, 1999.
[153] Astm C666/C666M-15, “Standard Test Method for Resistance of Concrete to Rapid
Freezing and Thawing,” in ASTM Standards, West Conshohocken, PA: ASTM
International, 2015, pp. 1–7.
[154] J. Li and Y. Yao, “A study on creep and drying shrinkage of high performance concrete,”
Cem. Concr. Res., vol. 31, pp. 1203–1206, 2001, doi: 10.1016/S0008-8846(01)00539-7.
91
[155] J. H. Potgieter, D. J. Delport, S. Verryn, and S. S. Potgieter-Vermaak, “Chloride-binding
Effect of Blast Furnace Slag in Cement Pastes Containing Added Chlorides,” South
African J. Chem. Tydskr. Vir Chemie, vol. 63, pp. 108–114, 2011.
[156] O. R. Ogirigbo, “Influence of Slag Composition and Temperature on the Hydration and
Performance of Slag Blends in Chloride Environments,” The University of Leeds School,
2016.
[157] R. K. Dhir, M. A. K. El-Mohr, and T. D. Dyer, “Chloride binding in GGBS concrete,”
Cem. Concr. Res., vol. 26, no. 12, pp. 1767–1773, 1996, doi: 10.1016/S0008-
8846(96)00180-9.
[158] M. D. A. Thomas, B. Fournier, and K. J. Folliard, “Alkali-Aggregate Reactivity (AAR)
Facts Book,” Austin,TX, 2013.
[159] ASTM C 1778-16, “Standard Guide for Reducing the Risk of Deleterious Alkali-
Aggregate Reaction,” in Annual Book of ASTM Standards, West Conshohocken, PA:
ASTM International, 2016, pp. 1–11.
[160] T. Bakharev, J. G. Sanjayan, and Y. B. Cheng, “Effect of admixtures on properties of
alkali-activated slag concrete,” Cem. Concr. Res., vol. 30, no. 9, pp. 1367–1374, 2000,
doi: 10.1016/S0008-8846(00)00349-5.
[161] F. Jackson, “The durability of concrete in service,” ACI J. Proc., vol. 43, no. 165, pp.
165–180, 1946.
[162] ASTM C227, “Standard Test Method for Potential Alkali Reactivity of Cement-Aggregate
Combinations ( Mortar-Bar Method ),” in Annual Book of ASTM Standards, West
Conshohocken, PA: ASTM International, 2010, pp. 1–5.
[163] K. A. Evans, “The Optimum Sulphate Content in Portland Cement,” University of
Toronto, 1997.
[164] ASTM C563-18, “Standard Guide for Approximation of Optimum SO3 in Hydraulic
Cement,” in ASTM International, West Conshohocken, PA, 2018, pp. 1–8.
[165] F. J. Tang and E. M. Gartner, “Influence of sulphate source on Portland cement
hydration,” Adv. Cem. Res., vol. 1, no. 2, pp. 67–74, 1988.
[166] F. J. Tang, “Optimization and Content of Sulfate Form,” Skokie,IL,USA, 1992.
92
[167] H. K. Choudhary et al., “Observation of phase transformations in cement during
hydration,” Constr. Build. Mater., vol. 101, pp. 122–129, 2015, doi:
10.1016/j.conbuildmat.2015.10.027.
[168] G. Tzouvalas, N. Dermatas, and S. Tsimas, “Alternative calcium sulfate-bearing materials
as cement retarders: Part I. Anhydrite,” Cem. Concr. Res., vol. 34, no. 11, pp. 2113–2118,
2004, doi: 10.1016/j.cemconres.2004.03.020.
[169] A. Papageorgiou, G. Tzouvalas, and S. Tsimas, “Use of inorganic setting retarders in
cement industry,” Cem. Concr. Compos., vol. 27, no. 2, pp. 183–189, 2005, doi:
10.1016/j.cemconcomp.2004.02.005.
[170] A. Ren, B. Guo, Y. Ma, Z. Zhu, and T. Zhu, “Effects of calcium sulfite on retarding of
cement,” in 3rd International Conference on Bioinformatics and Biomedical Engineering,
iCBBE 2009, 2009, pp. 1–5, doi: 10.1109/ICBBE.2009.5162731.
[171] B. Samet and S. L. Sarkar, “The influence of calcium sulfate form on the initial hydration
of clinkers containing different alkali combinations,” Cem. Concr. Res., vol. 27, no. 3, pp.
369–380, 1997, doi: 10.1016/S0008-8846(97)00030-6.
[172] W. Lou, B. Guan, and Z. Wu, “Dehydration behavior of FGD gypsum by simultaneous
TG and DSC analysis,” J. Therm. Anal. Calorim., vol. 104, pp. 661–669, 2011, doi:
10.1007/s10973-010-1100-6.
[173] R. Jadhav and N. C. Debnath, “Computation of X-ray powder diffractograms of cement
components and its application to phase analysis and hydration performance of OPC
cement,” Bull. Mater. Sci., vol. 34, no. 5, pp. 1137–1150, 2011, doi: 10.1007/s12034-011-
0134-0.
[174] S. Seufert, C. Hesse, F. Goetz-Neunhoeffer, and J. Neubauer, “Quantitative determination
of anhydrite III from dehydrated gypsum by XRD,” Cem. Concr. Res., vol. 39, pp. 936–
941, 2009, doi: 10.1016/j.cemconres.2009.06.018.
[175] A. N. Christensen, M. Olesen, Y. Cerenius, and T. R. Jensen, “Formation and
transformation of five different phases in the CaSO 4-H2O System: Crystal structure of
the subhydrate β-CaSO4·0.5H2O and soluble anhydrite CaSO 4,” Chem. Mater., vol. 20,
pp. 2124–2132, 2008, doi: 10.1021/cm7027542.
[176] G. Goswami, B. Mohapatra, and J. D. Panda, “Gypsum Dehydration During Comminution
and Its Effect on Cement Properties,” Am. Ceram. Soc., vol. 73, no. 3, pp. 721–723, 1990,
doi: 10.1201/9781315274508-20.
93
[177] P. K. Panigrahy, G. Goswami, J. D. Panda, and R. K. Panda, “Differential comminution of
gypsum in cements ground in different mills,” Cem. Concr. Res., vol. 33, pp. 945–947,
2003, doi: 10.1016/S0008-8846(02)00992-4.
[178] E. Gartner, I. Maruyama, and J. Chen, “A new model for the C-S-H phase formed during
the hydration of Portland cements,” Cem. Concr. Res., vol. 97, pp. 95–106, 2017, doi:
10.1016/j.cemconres.2017.03.001.
[179] I. Jawed and J. Skalny, “Alkalies in cement: A review. II. Effects of alkalies on hydration
and performance of Portland cement,” Cement and Concrete Research, vol. 8, no. 1. pp.
37–51, 1978, doi: 10.1016/0008-8846(78)90056-X.
[180] I. Jawed and J. Skalny, “Alkalies in cement: A review I. Forms of Alkalies and their effect
on clinker formation,” Cem. Concr. Res., vol. 7, no. 6, pp. 719–729, 1977, doi:
10.1016/0008-8846(77)90056-4.
[181] I. Odler and R. Wonnemann, “Effect of alkalies on portland cement hydration I. ALKALI
OXIDES INCORPORATED INTO THE CRYSTALLINE LATTICE OF CLINKER
MINERALS,” Cem. Concr. Res., vol. 13, no. 4, pp. 477–482, 1983, doi: 10.1016/0008-
8846(83)90005-4.
[182] W. Hansen and E. Pressler, “Solubility of Ca (OH) 2 and CaSO4. 2H2O in Dilute Alkali
Solutions,” Ind. Eng. Chem., vol. 39, no. 10, pp. 1280–1982, 1947.
[183] W. Ma, P. W. Brown, and D. Shi, “Solubility of Ca(OH)2 and CaSO4·2H2O in the liquid
phase from hardened cement paste,” Cem. Concr. Res., vol. 22, no. 4, pp. 531–540, 1992,
doi: 10.1016/0008-8846(92)90003-E.
[184] V. Kocaba, E. Gallucci, and K. L. Scrivener, “Methods for determination of degree of
reaction of slag in blended cement pastes,” Cem. Concr. Res., vol. 42, no. 3, pp. 511–525,
2012, doi: 10.1016/j.cemconres.2011.11.010.
[185] S. Ogawa, T. Nozaki, K. Yamada, H. Hirao, and R. D. Hooton, “Improvement on sulfate
resistance of blended cement with high alumina slag,” Cem. Concr. Res., vol. 42, no. 2,
pp. 244–251, 2012, doi: 10.1016/j.cemconres.2011.09.008.
[186] J. L. Poole, K. A. Riding, M. c. G. Juenger, K. J. Folliard, and A. K. Schindler, “Effects of
Supplementary Cementitious Materials on Apparent Activation Energy,” J. ASTM Int.,
vol. 7, no. 9, p. 102906, 2010, doi: 10.1520/JAI102906.
[187] ACI Committee 201, “ACI 201.2R-16: Guide To Durable Concrete.,” Farmington Hills,
MI, 2016.
94
[188] H. Zhu, D. G. Mapa, B. Lorentz, K. Riding, and A. Zayed, “Sulfate Optimization for
CCIL Blended Systems,” in Calcined Clays for Sustainable Concrete, S. Bishnoi, Ed.
Springer Singapore, 2020, pp. 323–330.
[189] Y. Zhou and Z. Zhang, “Effect of fineness on the pozzolanic reaction kinetics of slag in
composite binders: Experiment and modelling,” Constr. Build. Mater., pp. 1–17, 2020,
doi: 10.1016/j.conbuildmat.2020.121695.
[190] L. Ting, W. Qiang, and Z. Shiyu, “Effects of ultra-fine ground granulated blast-furnace
slag on initial setting time, fluidity and rheological properties of cement pastes,” Powder
Technol., vol. 345, pp. 54–63, 2019, doi: 10.1016/j.powtec.2018.12.094.
[191] M. Behim, M. Beddar, and P. Clastres, “Reactivity of Granulated Blast Furnace Slag,”
Slovak J. Civ. Eng., vol. XXI, no. 2, pp. 7–14, 2013, doi: 10.2478/sjce-2013-0007.
[192] I. Odler and S. Abdul-Maula, “Effect of Chemical Admixtures on Portland Cement
Hydration,” Cement, Concrete, and Aggregates, vol. 9, no. 1. pp. 38–43, 1987, doi:
10.1016/0008-8846(84)90039-5.
[193] ASTM C114-15, “Standard Test Methods for Chemical Analysis of Hydraulic Cement,”
West Conshohocken, PA, 2015.
[194] C1365-06, “Standard Test Method for Determination of the Proportion of Phases in
Portland Cement and Portland-Cement Clinker Using X-Ray Powder Diffraction
Analysis,” Annu. B. ASTM Stand. ASTM Int. West Conshohocken, PA, 2011, doi:
10.1520/C1365-06R11.2.
[195] N. Shanahan, A. Markandeya, A. Elnihum, Y. P. Stetsko, and A. Zayed, “Multi-technique
investigation of metakaolin and slag blended portland cement pastes,” Appl. Clay Sci., pp.
449–459, 2016, doi: 10.1016/j.clay.2016.07.015.
[196] A. Sedaghat, “Cement Heat of Hydration and Thermal Control,” 2016.
[197] D. L. Hudson-Lamb, C. A. Strydom, and J. H. Potgieter, “The thermal dehydration of
natural gypsum and pure calcium sulphate dihydrate (gypsum),” Thermochim. Acta, vol.
282/283, pp. 483–492, 1996, doi: 10.1016/0040-6031(95)02819-6.
[198] P. E. Stutzman, “Guide for X-Ray Powder Diffraction Analysis of Portland Cement and
Clinker.” pp. 1–38, 1996.
[199] P. E. Stutzman, P. Feng, and J. W. Bullard, “Phase Analysis of Portland Cement by
Combined Quantitative X-Ray Powder Diffraction and Scanning Electron Microscopy,” J.
Res. Natl. Inst. Stand. Technol., vol. 121, pp. 47–107, 2016, doi: 10.6028/jres.121.004.
95
[200] B. H. O’Connor and M. D. Raven, “Application of the Rietveld Refinement Procedure in
Assaying Powdered Mixtures,” Powder Diffr., vol. 3, no. 1, 1988, doi:
10.1017/S0885715600013026.
[201] D. Jansen, C. Stabler, F. Goetz-Neunhoeffer, S. Dittrich, and J. Neubauer, “Does Ordinary
Portland Cement contain amorphous phase? A quantitative study using an external
standard method,” Powder Diffr., vol. 26, no. 01, pp. 31–38, 2011, doi:
10.1154/1.3549186.
[202] ASTM C204-16, “Standard Test for Fineness of Hydraulic Cement by Air-Permeability
Apparatus,” West Conshohocken, PA: ASTM International, 2016.
[203] Horiba, “A Guidebook To Particle Size Analysis,” Horiba Instruments, Inc, p. , 2013.
[204] ASTM C563-18, “Standard Guide for Approximation of Optimum SO3 in Hydraulic
Cement Using,” West Conshohocken, PA, 2018.
[205] ASTM C305-14, Standard Practice for Mechanical Mixing of Hydraulic Cement Pastes
and Mortars of Plastic Consistency. West Conshohocken, PA,: ASTM International,
2014.
[206] ASTM C511-13, “Standard Specification for Mixing Rooms, Moist Cabinets, Moist
Rooms, and Water Storage Tanks Used in the Testing of Hydraulic Cements and
Concretes,” West Conshohocken, PA: ASTM International, 2013.
[207] P. F. Hansen, The Science of Construction Materials. Berlin Heidelberg: Springer, 2009.
[208] ASTM C1702-17, “Standard test method for measurement of heat of hydration of
hydraulic cementitious materials using isothermal conduction calorimetry,” Annu. B.
ASTM Stand., pp. 1–9, 2017, doi: 10.1520/C1702-17.2.
[209] N. Carino and H. Lew, “The maturity method: from theory to application,” Struct. Congr.
Expo. 2001, pp. 1–19, 2001, doi: doi:10.1061/40558(2001)17.
[210] A. Kumar, T. Oey, G. P. Falla, R. Henkensiefken, N. Neithalath, and G. Sant, “A
comparison of intergrinding and blending limestone on reaction and strength evolution in
cementitious materials,” Constr. Build. Mater., vol. 43, pp. 428–435, 2013, doi:
10.1016/j.conbuildmat.2013.02.032.
[211] M. J. Whittaker, “The Impact of Slag Composition on the Microstructure of Composite
Slag Cements Exposed to Sulfate Attack,” School of Civil Engineering, 2014.
96
[212] M. Whittaker, M. Zajac, M. Ben Haha, and L. Black, “The impact of alumina availability
on sulfate resistance of slag composite cements,” Constr. Build. Mater., vol. 119, pp. 356–
369, 2016, doi: 10.1016/j.conbuildmat.2016.05.015.
[213] K. Harada and H. Matsushita, “Effect of Gypsum Content in Cement on the Autogenous
Shrinkage of Low-Heat Portland Blast Furnace Slag Cement,” Memoris Fac. Eng. Univ.,
vol. 63, no. March 2003, pp. 51–66, 2003.
