Case Studies in Construction Materials 15 (2021) e00776

Contents lists available at ScienceDirect

Case Studies in Construction Materials

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

Case study

Influence of low-grade materials as clinker substitute on the

rheological behavior, hydration and mechanical performance of
ternary cements
C.S. Malacarne a, M.A. Longhi a, *, M.R.C. Silva a, J.P. Gonçalves b, E.D. Rodríguez c, A.
P. Kirchheim a, d, *
a
Postgraduate Program in Civil Engineering: Construction and Infrastructure, Building Innovation Research Unit (NORIE/UFRGS), Federal
University of Rio Grande do Sul (UFRGS), Av. Osvaldo Aranha, 99/706, CEP: 90035-190 Porto Alegre, RS, Brazil
b
Escola Politécnica, Centro Interdisciplinar de Energia e Ambiente (CIENAM), Universidade Federal da Bahia (UFBA), Rua Aristides Novis, 02,
Campus da Federação, Salvador CEP: 40210-630, Brazil
c
Departamento de Estruturas e Construção Civil (DECC), Universidade Federal de Santa Maria (UFSM), Santa Maria CEP: 97105-900, Brazil
d
Departamento de Engenharia Civil (DECIV), Núcleo Orientado para a Inovação da Construção, Universidade Federal do Rio Grande do Sul
(NORIE/UFRGS), Av. Osvaldo Aranha, 99, Porto Alegre CEP 90035-190, Brazil

A R T I C L E I N F O A B S T R A C T

Keywords: This study aims to evaluate the influence of low-grade local materials as clinker substitute in
Sustainability ternary cements, with calcined clay (30 wt%) and limestone (15 wt%), also known as LC3 ce­
Ternary cement ments. A coal mining waste (based on 35 wt% kaolinite content, AR) and a dolomitic limestone
Waste
(LLS) were compared to more pure materials such as natural clay (constituted by 83% kaolinite
Coal mining residue
content, NC) and high-grade limestone (HLS), respectively. LC3 cements were compared to
reference system (45 wt% quartz filler instead of limestone and calcined clays), as well as a fly ash
blended cement (40 wt% fly ash, 40FA). Hydration kinetics, water demand, plastic viscosity, and
yield stress were evaluated in a fresh state. The fineness of the calcined clays and limestone had a
greater impact on the water demand and thixotropy of the pastes than their mineralogical
compositions. However, the higher kaolinite content of NC led to higher pozzolanic activity,
greater formation of carboaluminate type phases, and compressive strength than LC3 cements
with AR. The HLS and LLS had similar impact on the proprieties assessed despite their different
mineralogical composition. Thus, both low-grade materials are feasible as raw materials for the
production of LC3 type cements with similar performance when compared to conventional
cements.

1. Introduction

The use of supplementary cementitious materials (SCMs) as clinker replacement is a well-established alternative for reducing
greenhouse gases (GHG) emissions related to the cement production [1–3]. Brazil, for instance, has long used high amount of SCMs in

* Corresponding author.
* Corresponding author at: Postgraduate Program in Civil Engineering: Construction and Infrastructure, Building Innovation Research Unit
(NORIE/UFRGS), Federal University of Rio Grande do Sul (UFRGS), Av. Osvaldo Aranha, 99/706, CEP: 90035-190 Porto Alegre, RS, Brazil.
E-mail addresses: marlonlonghi@gmail.com (M.A. Longhi), anapaula.k@ufrgs.br (A.P. Kirchheim).

https://doi.org/10.1016/j.cscm.2021.e00776
Received 21 September 2021; Received in revised form 6 November 2021; Accepted 10 November 2021
Available online 11 November 2021
2214-5095/© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
C.S. Malacarne et al. Case Studies in Construction Materials 15 (2021) e00776

cement production, but its clinker factor has been relatively fixed at approximately 67–69% in the past decade, with a slight upward
trend after 2014. This can be partly associated with the limited availability of some of the most traditionally used SCMs, as granulated
blast furnace slag (GGBFS) and fly ash (FA), which is seen in a global level as a consequence of the new recycling initiatives (steel and
iron) and the higher interest in renewable energy generation (without coal combustion process) [4–8].
The expectation of SCMs’ scarcity was the driving factor for the development of the so-called LC3 cement (Limestone Calcined Clay
Cement) in the last years, a cement with higher clinker substitution which combines calcined clay and limestone, taking advantage of
the wide availability of these materials [9–11]. It was demonstrated that up to 50% clinker substitution by calcined clay and limestone
(usually at 30% and 15% content, respectively) could provide good mechanical properties comparable to ordinary cements due to a
synergistic reaction that takes place between limestone and aluminum-containing SCMs, such as calcined clays [9,12–15].
Concerning the use of calcined clays in cement, metakaolin (MK) is a well know material with high pozzolan reactivity, after
thermal or mechanical treatments of rich-kaolinite clays [16–18]. On the other hand, there are some obstacles related to its major use
by the cement industry. MK is a high cost commercial product due to competitive use of purer kaolinite clays by other industries
(mainly by the paper, ceramic, and painting industries) [19]. Additionally, its layered structure and high surface area that increase the
higher water demand or the use of chemical admixtures to achieve an acceptable or required workability [20,21] or due to envi­
ronmental impacts and energy costs associated with thermal processes [22]. Related to the content of kaolinite, Favier et al. [23], who
suggests that when the higher is the content of calcined kaolinite of the clay, the greater is the amount of water captured from the
mixture, decreasing its effective interaction with the cement particles. Recent studies have demonstrated that clays with high kaolinite
content (> 60 wt%) contribute to early ages and there is no great mechanical strength benefit after 7 days of curing, which was
associated with a limiting pore refinement [24–26].
Additionally, studies on the production of LC3 cements demonstrate that clays with at least ~40% of kaolinite are feasible to
achieve up to 50% of Portland clinker replacement. Thus, use of lower purity clays, i.e., with a lower-kaolinite content, may be able to
overcome such limitations and even offer savings when adopting waste materials in cement production. Some efforts have been made
in this direction, such as the use of 30% Fired Rejected Clay Bricks (FRCB) to compose an LC3-50 (clinker content of 50%) [26] or even
a water treatment plant sludge ash (WTPSA) and limestone by measuring the compressive strength of systems with different SCMs
content.
Besides the necessity of further investigation on effects of impure clays or waste materials containing lower kaolinite content in LC3
cements, there is another gap in the literature, which is the influence of lower-grade limestone in such cements. Depending on the
geological formation, limestone can present a different oxide composition in each location, thus affecting its physical and chemical
properties. A few studies suggested that ternary cements with calcined clay and limestone are not significantly affected by the car­
bonate source used as SCM [28,29]. However, these studies were made using clays with high content of kaolinite. Therefore, the
question also arises about the feasibility of using lower quality limestones with lower amounts of calcium carbonate and higher im­
purities to produce ternary blended cements with low kaolinite content clays.
Thus, to allow extensive use of local materials, there is a great interest in finding and assessing feasible clays and limestone at low
cost, even with low kaolinite content, whose appropriated treatment can present a similar performance when compared to traditional
SCMs. The viability associated with the mechanical and hydration behavior of low-grade and locally available materials is evaluated to
produce LC3 cements. Thus, the reaction extent of LC3 cements with different kaolinite content and limestone with different contents of
MgO and CaO was assessed. This provides new insight into the reaction products and hydration behavior related to the use of these
materials. In the same way, a comparison with a pozzolanic cement currently used on a large scale was done enabling an open market
for LC3 cement.

2. Experimental program

2.1. Test conducted

The overall raw materials as well as hydrated samples were characterized through different techniques. Particle size distribution
(PSD) were measured by dry laser diffraction with a Malvern Mastersizer 3000. BET specific surface area was measured in a Micro­
meritics ASAP 200. X-ray diffractometry (XRD) measurements were performed in a Bruker D8 ADVANCE with Cu Kα radiation, copper
tube, a step size of 0.02◦ and 1 s/step, a voltage of 40 kV, 40 mA, obtained from 5◦ to 70◦ 2Ɵ. Qualitative phase analysis was performed
using X′ Pert High Score Plus software with the ICDD PDF 4 + 2016 database. Thermogravimetric analysis (TGA) was performed in a
Shimadzu DTG-60 H equipment, with approximately 10 mg crushed paste samples. A heating rate of 10 ◦ C/min was used, from 35 ◦ C to
1000 ◦ C, using alumina crucibles and nitrogen as purge gas (50 ml/min).

2.2. Materials and sample preparation

Two kaolinite sources from southern Brazil were used. A coal mining residue, an argillite type clay (AR) with 35% kaolinite was
collected at an open pile at Rio Grande do Sul (RS, Brazil). A natural clay (NC) from Pântano Grande/RS with a high amount of
kaolinite (83%) was used for comparison. The kaolinite content was estimated from thermogravimetric analysis (TGA), according to
the method proposed by (Avet and Scrivener 2018a). The XRD of both clays showed the presence of kaolinite (Al2Si2O5(OH)4; PDF 01-
078-2109), muscovite (KAl2(Si3Al)O10(OH)2; PDF 01-074-1392), and quartz (SiO2; PDF 01-083-0539) (Fig. 1). The clays AR and NC
were calcined at 750 ◦ C in a static laboratory oven for 1 h, considering the protocols used by Longhi et al. [31], ground using a ball mill
(Servitech CT-242), for 40 min and 10 min until Dv90 ≤ 35 µm was obtained, respectively.

2
C.S. Malacarne et al. Case Studies in Construction Materials 15 (2021) e00776

Two limestones grades were used. A high purity limestone (HLS), with the predominant presence of 99.7% calcite (CaCO3, PDF
00–005–0586), with traces of quartz (SiO2; PDF 01-083-0539) from Ijací /MG (southeastern Brazil), and a lower grade limestone
(LLS), with higher amount of impurities and composed by 78.3% calcite, dolomite (CaMg(CO3)2, PDF 00–036–0426) and muscovite
(KAl2(Si3Al)O10(OH)2), from Candiota/RS (southern Brazil) (Fig. 1). Both materials were homogenized and dried in an oven at 100 ºC
until constant mass and ground for 10 until a Dv90 ≤ 25 µm was obtained.
A gray Portland clinker, gypsum and fly ash were supplied ground by a Brazilian cement industry. The commercial quartz filler
was used as inert material in a reference cement LC3 . The Fly ash and quartz filler were used to compose reference cements. The
mineral composition of all materials are presented in Supplementary Material (See Fig. S.1–4).
Table 1 shows the chemical composition (obtained by XRF) and main physical properties of these materials. The particle size
distributions of the overall raw materials are also presented in Fig. 2. In order to reduce the effect of particle size, the mean particle size
and Dv [50] for the clays are slightly different, as well as the calcium carbonate materials. Pozzolanic activity of the kaolinite sources
and fly ash were assessed using the modified Chapelle method, which is described in the Brazilian Standard (ABNT NBR 15895:2010)
were possible to identify a significant difference in the reactivity degree between the natural and argillite. In this sense, the argillite
shows slightly similar pozzolanic reactivity degree, when compared to the fly ash.

2.3. Cements formulation and characterization

The cements were prepared according to the formulation shown in Table 2. Six blended systems were produced, with a fixed
percentage of clinker and gypsum of 50% and 5%, respectively, and a total percentage of SCMs of 45%, following the proportions from
Antoni et al. (Antoni et al., 2012). These are: four LC3 cements, varying the source of kaolinite and limestone; a reference cement (REF),
with Portland clinker, gypsum, and an inert filler (quartz), and a Portland pozzolanic cement (40FA) with the addition of fly ash. 40FA
cement also has 5% LLS, which was based on information provided by a cement industry on the composition of locally available
commercial cements. The nomenclature of LC3 cements was established as follows: the first two terms refer to the kaolinitic source,
“NC” for natural clay and “AR” for the residue, and the last two refer to the type of limestone, “HLS” for the purest limestone and “LLS”
for the limestone with higher impurities content.
The mini-slump test was used to evaluate the water demand of cements, according to Hunger and Brouwers [32] methodology.
The relative spread (Γp) is calculated, which represents a measure of the deformability of the mixture and plotted according to the ratio
between the volumes of water and cement in the pastes. A straight line is adjusted through linear regression, resulting in a function as
Eq. 1, which allows determining the minimum water demand (βp) of a cement paste, i.e., the maximum water content that can be
retained by the particles of a cement, and the deformation coefficient (Ep), which indicates the sensitivity of the material to water
addition.
Vw
= Ep Γ p + β p (1)
Vp

where:
Vw: volume of water.
Vp: cement volume.
Ep: deformation coefficient.
Γp: relative spread.

Fig. 1. Shows the XDR for the kaolinite sources and limestones.

3
C.S. Malacarne et al. Case Studies in Construction Materials 15 (2021) e00776

Table 1
Physical and chemical properties of the materials.
Natural Argillite High-grade Lower-grade Clinker Gypsum Fly ash Quartz
Clay (NC) Clay (AR) limestone limestone (LLS) (FA) filler
(HLS)

Oxide
Composition CaO 0.31 1.12 55.83 43.87 62.15 36.01 3.22 –
(%)
SiO2 45.70 57.39 0.47 11.91 19.29 1.08 56.75 96.11
Al2O3 38.20 24.29 0.28 3.65 3.75 0.47 30.10 2.20
Fe2O3 0.65 4.26 0.13 2.73 3.27 0.27 3.32 –
K2 O 0.39 2.14 0.02 1.16 1.70 0.09 2.79 0.10
TiO2 0.13 1.16 – 0.29 – – 1.81 0.06
SO3 – 0.61 0.06 0.17 1.56 40.92 0.56 0.34
MgO – 0.25 – 3.13 7.34 0.21 0.19 –
MnO – – – 0.18 0.14 – –
P2O5 – – 0.13 0.08 0.08 – 0.13 –
ZnO – – – – 0.05 – 0.04 0.10
SrO – – 0.16 0.02 0.04 0.09 0.02 –
Y2O3 – – – – – – 0.02 –
ZrO2 – 0.02 – – – – 0.03 –
Cr2O3 – – – – – – 0.05 –
L.O.I (loss on 13.74 7.92 42.92 32.81 0.53 20.86 0.97 0.90
ignition at
950 ◦ C)
Particle size Dv10 1.50 1.08 1.61 1.54 1.93 1.23 1.25 1.41
(µm)
Dv50 6.34 5.35 7.14 5.82 11.30 7.64 7.86 5.74
Dv90 22.07 33.30 20.40 21.57 41.10 48.63 24.97 13.32
Median particle 9.38 11.70 9.33 9.25 17.10 17.07 10.85 6.67
size
Specific mass (kg/m3) 2460 2590 2750 2730 3080 2340 2430 2650
BET Superficial area (m2/g) 11.82 24.75 1.02 6.21 1.43 13.73 1.63 2.55
Pozzolanic activity (mg Ca(OH)2/ 1168 527 – – – – 370 –
g)

Fig. 2. influence of SCMs on the (A) water demand (B) Ep and βp parameters.

βp: minimum water demand.

Rheological parameters were assessed by rotational rheometry using a Brookfield R/S Plus model 4SST 115LS rotational
rheometer with concentric cylinder geometry. The mixture of the pastes was performed at 0.60 w/b, as for lower w/b pastes were too
stiff, and the maximum torque of the equipment was reached. The shear cycles were initiated 5 min after the cement contact with
water in an environment with a constant temperature of 25 ± 1 ◦ C. The application profile adopted was based on the recommenda­
tions of C1749 - 17a [33]: the measurement was initiated at the lowest shear rate (0.129 s− 1) and increased in logarithmic scale until
50 s− 1, followed by a decrease in the same scale. This process was repeated once, the first cycle being only used as a pre-shear phase to

4
C.S. Malacarne et al. Case Studies in Construction Materials 15 (2021) e00776

Table 2
Binders formulations.
ID Portland clinker (%) Gypsum (%) Kaolinite sources Limestone Fly ash (%) Quartz filler (%) Blaine fineness (m2/g)
(%) (%)

NC AR HLS LLS

REF 50 5 – – – – – 45 6270
40FA 50 5 – – – 5 40 – 5460
NC-HLS 50 5 30 – 15 – – – 6930
NC-LLS 50 5 30 – – 15 – – 7750
AR-HLS 50 5 – 30 15 – – – 7460
AR-LLS 50 5 – 30 – 15 – – 8670

homogenize and break down the internal structure of the pastes, and the data corresponding to the second cycle was used to determine
the plastic viscosity and yield stress, and hysteresis areas [34,35]. A Bingham behavior was assumed, with a linear relationship be­
tween shear stress and shear rate.
The hydration was evaluated by isothermal calorimetry using a TAM Air microcalorimeter (TA Instruments). The temperature of
the calorimeter was set at 25 ◦ C and the analysis was performed for 72 h. Cements, water, and test accessories were kept in the
equipment room for 24 h before experiment. Cement pastes were made with external mixer of 10,000 rpm, mixing 100 g of cement
and deionized water at 0.45 w/b. Approximately 6 g of the mixed material was placed in a 20 ml glass ampoule, hermetically sealed
with a lid, and placed in the calorimeter channel.
Hydration products were followed by X-ray diffraction (XRD) and thermogravimetric analysis (TGA) in pastes with 1, 7, 28, and
91 days, in pastes (0.48 w/b) without hydration stoppage. Compressive strength was performed at 1, 7, 28 and 91 days in mortars
(0.48 w/b) according to the Brazilian standard NBR 7215 [36].

3. Results and discussion

3.1. Water demand

From the results of Mini-slump test at 0.6, 0.7, and 0.8 w/b the relative spreading (Γp) was calculated and plotted according to the
water and cement ratio volumes of the pastes in Fig. 2.
The minimum water demand, βp, corresponding to a theoretical relative spread equal to zero (or the maximum amount of water
retained by the powder) was obtained for REF and 40FA, respectively. The highest βp values was achieved by the LC3 cements with AR.
Regarding the type of limestone, the LC3 with LLS presented higher water demand than those with HLS. Despite the lower purity and
coarser PSD of the AR and LLS materials compared to their peers (Table 1), their higher inner surface area (BET measurements) led to
cements of higher Blaine fineness. Consequently, the amount of water required to cover the particles of the cements and promote a
dispersed system was greater than the others.
Regarding the strain coefficient, Ep, which indicates the workability sensitivity of the pastes to the addition of water, the mixtures
based on LC3 proportions showed the higher values than FA40. A linear correlation between Ep and Blaine specific surface area was
previously found in blended cements with limestone and fly ash only [32,37], but this property does not directly explain the results

Fig. 3. Flow curves of the cement pastes.

5
C.S. Malacarne et al. Case Studies in Construction Materials 15 (2021) e00776

observed here to the NC and AR cements. AR, for instance, although its higher water demand (βp) presented greater fluidity for a given
amount of water than cements with NC, independently of limestone used. As a result, from a relative spread of approximately 8
(corresponding to 106 mm), cements with AR demanded lesser amount of water to obtain same spread as cements with NC. This
suggests a higher impact of kaolinite content of the calcined clay than fineness, retaining water to provide further spread of the pastes,
as previously discussed by Favier and Scrivener [23,38].

3.2. Plastic viscosity and yield stress

Fig. 3 shows the linear fit (dashed lines) of the flow curves to calculate the yield stress and plastic viscosity of the pastes (Table 3).
The hysteresis areas were calculated from the second cycle in Fig. S.5 (See SM).
The results obtained are consistent with the characterization of the SCMs, such as PSD and surface area, which greatly influence the
plastic viscosity, as increasing surfaces lead to significant interparticle contact and surface interlocking [39,40]. This led to lower
plastic viscosity and yield stress and hysteresis areas of 40FA comparing to the LC3 cements. The use of limestone and calcined clays
increased such rheological parameters considering REF cement. In case of the calcined clays, a portion of the water from the mixture is
adsorbed into their surface, reducing the amount available for lubricating, thus increasing the internal friction during flow [41,42].
Finally, the angular shape of the calcined clays particles also contributes to increasing the plastic viscosity, yield stress and decreasing
the homogeneity of the pastes [41,43]. While limestone may have opposite effects.
Calcined clay`s surface area had more influence over the rheological properties of LC3 cements than the kaolinite content (i.e., the
AR resulted in higher increases in plastic viscosity and yield stress than NC, even though it contains a lower kaolinite content),
corroborating the results found by Muzenda et al. [44], who studied different calcined clay/limestone ratios and levels of addition.
These results contradict what some authors have previously argued about a greater impact of kaolinite content than the physical
properties of the calcined clays on the rheological behavior. In fact, the higher the kaolinite content of the clay, the lower its sensitivity
to spreading by water additions to the mixture (according to Ep results), but a higher BET of a calcined clay (e.g., RA) leads to more
cohesive, viscous and less homogeneous cement pastes.
Concerning limestones, pastes with LLS presented higher values of plastic viscosity and hysteresis areas than pastes with HLS. These
results are also in the same order as the spread results observed in the Mini-slump test for a similar w/b ratio of 0.60. However, the
yield stress of cements with HLS presented higher than those with LLS. Such results are not solely explained by the surface area and PSD
of the SCMs. As yield stress is also influenced by properties such as colloidal forces and non-contact interactions between material
surfaces [45], more detailed research would be required to justify these results.

3.3. Isothermal calorimetry

The composition of the ternary cements directly influenced the hydration kinetics as shown in Fig. 4.
Three events can be analyzed in the heat flow curve. The first peak is attributed to the dissolution of C3S and precipitation of C-S-H
and occurs between 5 and 10 h, depending on the cement type. The second peak is attributed to the dissolution of C3A and further
ettringite formation. Moreover, the third peak, less intense and larger, occurs after 35 h of hydration and can be associated with AFm
formation. More details about the time and heat released can be found in the SM (Table S.1-S.2).
The results show a similar trend for LC3 cements. The first hydration peak is accelerated in approximately 2 h compared to REF and
40FA cements, which can be associated with their higher specific surface area. The fine particle size additions act as a nucleation point
for the precipitation of hydration products [46] accelerating the reactions. Similarly, the aluminate reactions occur in advance and
with greater intensity in LC3 cements, indicating an accelerated sulfate depletion that occurs approximately 5.6 h earlier than for the
other cements, due to the higher fineness and alumina content present in LC3 cements. This result agrees with Antoni et al. . The heat
flow released by the pastes with higher kaolinite content (NC) is higher than the pastes with lower (AR), as observed in the second
peak. This result is in line with the observations made by Avet and Scrivener [48] who accounted for this effect due to the higher
kaolinite content which allows further alumina ions into solution. This is contrary to a more recent study from Zunino and Scrivener
[49] which has attributed this earlier sulfate depletion to filler effect solely (note that AR has a superficial area 2.1 times higher than
NC, as showed in Table 1).
After the 36 h of testing, NC induces an acceleration of the total heat of LC3 cement pastes compared to the pastes with AR (Fig. 4B).
This performance can be associated with its higher pozzolanic reactivity and the formation of calcium–aluminate–silicate–hydrate
(C–A–S–H) gel and higher formation of carboaluminate phases. Furthermore, after 72 h of testing, a higher total cumulative heat per

Table 3
Rheological parameters of the different cements.
Cement Plastic Viscosity (Pa⋅s) Yield stress (Pa) Hysteresis area (Pa/s)

REF 0.48 8.97 28.71

40FA 0.31 6.62 32.25
NC-HLS 0.84 9.09 156.97
NC-LLS 1.46 7.18 262.88
AR-HLS 1.11 13.49 156.75
AR-LLS 1.64 11.13 336.11

6
C.S. Malacarne et al. Case Studies in Construction Materials 15 (2021) e00776

Fig. 4. Heat flow measurements (A) Heat flow (B) Total Heat.

gram of cement is observed in the pastes with NC, as expected due to its higher kaolinite content, corroborating results previously
reported [30,50].
Regarding the type of limestone present in cements, HLS or LLS, a more pronounced filler effect is observed for LLS due to its higher
specific surface area. It can be observed that these pastes presented earlier hydration peaks and with more intense heat flow than
similar pastes with HLS. This behavior can be possibly associated with the higher initial formation of carboaluminate phases by the
pastes with LLS with one day of hydration (as shown in XRD/TGA results in item 3.3). Over time, however, as it is seen in the cu­
mulative heat at 72 h, the difference between the two limestones is reduced.
Comparing the results obtained by 40FA with LC3 cements, it is noticed that reactions occur more slowly, releasing less heat over

Fig. 5. Shows the XDR for the (A) REF and (B) 40FA cement.

7
C.S. Malacarne et al. Case Studies in Construction Materials 15 (2021) e00776

time. This may be related to a physical effect due to the smaller surface area and low reactivity of FA, as observed by the Chapelle
method (presented in Table 1), and to a faster dissolution rate of the Si and Al phases of the calcined clays [51]. Its behavior is similar to
REF cement, which contains 45% of an inert material, but presents a similar value of the specific surface area. It is also noted that the
heat flow at the main peak of hydration is higher in the REF, supporting the impact of fineness in the hydration kinetics.
It is possible to evaluate the sulfate balance in the systems related to the first and second peaks. The sulfate depletion occurred with
an average delay of 6 h concerning the main peak in LC3 cements and 10 h in 40FA and REF cements, indicating an excess of sulfates in
the systems. This also explains the low intensity of the aluminate peak in LC3 cements, as increasing sulfate content results in lower and
broader peaks . This observation could also explain the absence of an intense monosulfoaluminate phase formation by REF cement, a
point that was raised in the analysis of XRD results (item 2.4). The sulfate content of cements balances the formation of ettringite and
AFm phases (SO4-AFm or CO3-AFm); increasing sulfate content favors ettringite formation over AFm phases [53].

3.4. Hydration products

The XRD analysis of the cements are presented in Figs. 5 to 7.

The REF cement is shown in Fig. 5A, where is identified, over time, the formation of ettringite (C6A$3H32; PDF 01-072-0646),
portlandite (Ca(OH)2; PDF 01-072-0156), and monosulfoaluminate (C4A$3H12; PDF 01-083-1289) in lower intensity. The main
peak of portlandite increases over time, indicating the hydration process of clinker and the non-consumption of this hydrate due to the
absence of pozzolanic materials. The addition of fly ash in 40FA (Fig. 5B) induces a different behavior, with the consumption of
portlandite over time, visible from 28 days and more intense at 91 days. This consumption is attributed to the pozzolanic reaction
between fly ash and portlandite and results in the formation of calcium silicate hydrates. The crystalline peaks identified around 17 and
21 are attributed to mullite and quartz and related to these phases in fly ash.
Diffractograms of LC3 cements (Figs. 6 and 7), in addition to the previously observed phases, show the presence of the carboa­
luminate phases – monocarboaluminate (Mc) (C4ACH11; PDF 01-087-0493) and hemicarboaluminate (Hc) (C4AC.0.5H12; PDF 00-041-
0221) – due to the reaction between the calcium carbonate (CC) and the alumina provided by the calcined clays [13]. Ettringite
formation seems approximately stable at early ages (1 day) of hydration. According to previous studies, it can be expected that the
ettringite amount remains constant for all cements, as it depends solely on the sulfate content , which was kept constant. The presence
of Mc is noticeable from 24 h of hydration (except for system NC-HLS), which supports the assumption that all ettringite has been
formed at this age, as carboaluminate phases will only form after that . Also, portlandite peaks are more intense in both reference
systems (compared to ettringite), which indicates an early pozzolanic reaction of metakaolin.
The purity of limestone present in LC3 cements does not lead to a significant or visible change in the presence of hydration products,
which indicates that limestone with high dolomite (AR) content does not visibly affect the formation of hydration products, corrob­
orating the results for higher kaolinite content [28,29]. Concerning the sources of kaolinite two main differences are observed. The first
concerns the portlandite consumption by the pozzolanic reaction. In cements with NC, the consumption of portlandite was more
pronounced from the first to the seventh day, remaining approximately constant from that age, while for AR the consumption of
portlandite is noticeable only at 28 days. This difference results from the higher pozzolanic reactivity and faster reactions of the NC,
also identified in the Chapelle test (Table 1).
The second difference between LC3 cements with NC and AR is the intensity and type of carboaluminate formed. At later ages (28
and 91 days) there is a greater formation of Hc in NC-HLS and NC-LLS cements than AR-HLS and AR-LLS; which showed Mc and Hc

Fig. 6. Shows XDR for the LC3 : (A) NC-HLS and (B) NC-LLS.

8
C.S. Malacarne et al. Case Studies in Construction Materials 15 (2021) e00776

Fig. 7. Shows XDR for the LC3 : (A) AR-HLS and (B) AR-LLS.

with quite similar intensities. A similar result was observed by Avet and Scrivener [55], while evaluating clays with various kaolinite
contents in LC3 cements with equal clinker substitution percentage and calcined clay:limestone ratio to this study. The higher presence
of metakaolin, the higher formation of Hc was observed. This result is directly related to the content of reactive alumina available in
the calcined clay. Thus, it is visible the influence of the purity of the clay on the hydration products.
TGA analysis are in good agreement with XRD results. The thermal behavior of the cements produced after 1, 7, 28, and 90 days of
curing is shown in Fig. 8. Four main regions of mass loss are identified at temperatures of approximately 80 ◦ C, 125 ◦ C, 415 ◦ C, and
700 ◦ C. The main hydration products decomposed at these temperatures are C-S-H and ettringite; carboaluminates (both Mc and Hc);
portlandite, and; finally, above 600 ◦ C, calcium carbonate and dolomite [56–60].
40FA cement has a sensible shoulder around 150 ◦ C attributed to the formation of carboaluminates. This behavior is possible due to
the addition of 5% of limestone, which is also visible as calcium carbonate around 700 ◦ C, in less intensity when compared to
LC3 cements. In LC3 cements, the formation of carboaluminates is visible after 7 days and increases over time, which agrees with the
XRD results. It is noticeable that the peaks related to carboaluminates increase in intensity up to 91 days.
The mass loss related to the decomposition of CH at approximately 400 ◦ C increases overtime for the REF cement paste, as ex­
pected, while for 40FA and LC3 pastes, the highest amount of CH is observed at one day and later decreases as a result of the pozzolanic
reactions [61]. Due to the absence of limestone in the REF cements, carboaluminate phases are not observed, and any loss of mass
around 700 ◦ C can be attributed to the carbonation during the preparation process. For the 40FA and LC3 cements, the mass loss
observed above 600 ◦ C is related to calcium carbonates from the limestone [60]. Therefore, it is possible to observe a reduction in
intensity over time, associated with a its partial consumption, however, most part remains as filler. At long-term, clays with different
purity/reactivity do not exhibit a clear difference in thermal behavior. On the other way, a notable difference in the shape of the curve
is observed in the limestone variation, indicating a narrower range of decarbonation in the presence of dolomite. However, this change
does not alter the hydration products and possibly does not influence the mechanical behavior.

3.5. Compressive strength

The compressive strength results at 1, 7, 28 and 91 days are shown in Fig. 9.

The results of REF represent the mechanical performance provided by the interaction between clinker, gypsum stimulated by the
quartz filler. Due to the absence of pozzolanic addition, the reaction rate is accelerated until 7 days and then establishes a slower
increase until the clinker consumption. The additional strength presented by LC3 and 40FA cements is entirely provided by the
presence of SCMs, either through the pozzolanic reaction, which has a greater impact on the strength at later ages or through a physical
effect, which has a greater influence on the strength with 1 day, corroborating the previous study [62].
In general, cements produced with NC exhibited the highest strength, independently of the type of limestone used. The behavior
over time of NC-HLS and NC-LLS cements are characterized by their accelerated strength gain from 1 day to 7 days. At this age, these
cements already presented in average 87% and 73% of their strength at 28 days and 91 days, respectively, while for AR-HLS and AR-
LLS, these averages were 68% and 56% and for 40FA 64% and 42%. A similar result was observed by Antoni, Baquerizo, and Matschei
[63] , who attributed this behavior to the reaction between the extra alumina from metakaolin and carbonates, forming the car­
boaluminate phases. The formation of Mc and Hc is observed in XRD results (Figs. 5–7) and TGA (Fig. 8). Compared to LC3 cements
produced with AR, the average strength of NC-HLS and NC-LLS was approximately 46% higher at 28 days and 38% at 91 days. Better
mechanical performance of cements with NC was predicted since the kaolinite content has been demonstrated to be the most impacting
factor in the compressive strength of LC3 cements [64,65].

9
C.S. Malacarne et al. Case Studies in Construction Materials 15 (2021) e00776

Fig. 8. TGA results (A) REF (B) 40FA (C) NC-HLS (D) NC-LLS (E) AR-HLS and (F) AR-LLS.

Comparing the strength of LC3 cements with different limestones, it is noticeable that LLS provided a similar or slight increase of
strength to HLS. This result suggests that the type of carbonate in the system, its quantity, and other minor components did not affect
the strength development of LC3 cement since a very similar trend was observed in the presence of both HLS and LLS. This result is in
agreement with other previously reported [28].
Cement 40FA, on the other hand, presents a distinct behavior for strength development over time. A significant increase from 28 to
91 days was observed, compared to LC3 cements, reaching 95% of the average strength obtained by NC-HLS and NC-LLS cements. This
behavior agrees with the high consumption of portlandite between 28 and 91 days identified by TGA (Fig. 7). Similar performance was
observed by Tiwari and Chowdhury (Tiwari and Chowdhury 2016), who also compared LC3 cements to fly ash cements. The lower
initial strength is characteristic of commercial cements using fly ash as SCM [66,67] and directly associated with lower amount of
reactive Al2O3 compared to metakaolin, which delays reactions.
Even with the reduced performance of cements with AR compared to NC, AR-HLS and AR-LLS still reach compressive strength

10
C.S. Malacarne et al. Case Studies in Construction Materials 15 (2021) e00776

Fig. 9. Development of the compressive strength over time.

requirements of the Brazilian standard NBR 16697 [68] for pozzolanic cements as a type 32 strength class, alongside with 40FA, with
the minimum strength at 7 days (≥ 20.0 MPa), at 28 days (≥ 32.0 MPa) and 91 days (≥ 40 MPa). This classification is slightly similar
when compared to the CEM IV/B described in EN-197 in the UNE standards. Cements with NC would be categorized in a higher
strength class (type 40), with minimum strength of 23 MPa at 7 days, 40 MPa at 28 days and 48 MPa at 91 days.

4. Conclusions

The experimental results have shown that the tested low-grade materials are a promising source of SCMs, even with low purity. The
following main conclusions are:

• Regarding rheological behavior, the physical properties of the SCMs had a greater impact on the water demand and thixotropy of
the pastes than their mineralogical compositions. In LC3 cements, the finer the calcined clay (AR), the higher was the water de­
mand, plastic viscosity, yield stress and hysteresis areas than a cement with higher kaolinite content (NC). Which contradicts some
authors about the greater impact of clays kaolinite content on this. The same trend on the impact of fineness was observed in­
dependent of the type of limestone used.
• The hydration products formed in LC3 cements with NC and AR were similar, with different contents: while NC cements form higher
intensity peaks of Hc, cements with AR form Hc and Mc in equal proportions. For LC3 cements with NC, an intense consumption of
portlandite occurs at 7 days in LC, while for AR the pozzolanic reaction has less intensity. In contrast, in 40FA cement, the CH
consumption is more pronounced at 28 and 91 days, confirming the slow pozzolanic reaction of fly ash. Regarding the type of
limestone, no differences in phase formation were observed.
• The kaolinite content provides a strong influence on mechanical strength. LC3 cements with NC reached the highest compressive
strength; especially at 7 days. The type of carbonate in the system, its quantity, and other minor components did not affect the
strength development of LC3 cement since a very similar trend was observed in the presence of both HLS and LLS. All LC3 cements
achieved the strength requirements suggested by the Brazilian standard of cements [68], in different classes of strength, which
means they could be used in applications with different performance requirements.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Acknowledgments

The participation of Brazilian authors was sponsored by CNPq (Brazilian National Council for Scientific and Technological
Development) through the research project UNIVERSAL grant number 458597/2014-7 and the research fellowships PQ2017 303753/
2017-0 and 305530/2017- 8. The authors would like to acknowledge the financial support of the PROCAD Program, funded by CAPES
(Grant Nr. 88887.124147.2014-00), LNLS for the use of the XRD1 beamline (proposal 20180361) and Laboraty and Institute of
Chemistry CIENAM at Federal University of Bahia (UBFA). The authors are also thankful to Samile Matos and Jefferson Silva from
UFBA for experimental support.

Appendix A. Supporting information

Supplementary data associated with this article can be found in the online version at doi:10.1016/j.cscm.2021.e00776.

11
C.S. Malacarne et al. Case Studies in Construction Materials 15 (2021) e00776

References

[1] M. Schneider, M. Romer, M. Tschudin, H. Bolio, Sustainable cement production-present and future, Cem. Concr. Res. 41 (2011) 642–650, https://doi.org/
10.1016/j.cemconres.2011.03.019.
[2] J.S. Damtoft, J. Lukasik, D. Herfort, D. Sorrentino, E.M. Gartner, Sustainable development and climate change initiatives, Cem. Concr. Res. 38 (2008) 115–127,
https://doi.org/10.1016/j.cemconres.2007.09.008.
[3] CSI/ECRA, Development of state of the art techniques in cement manufacturing: Trying to look ahead; CSI/ECRA technology papers 2017, Dusseldorf, Geneva,
2017.
[4] IEA/CSI, Technology roadmap: Low-carbon transition in the cement industry, 2017, 66. 〈https://doi.org/10.1007/SpringerReference_7300〉.
[5] N. Müller, J. Harnisch, A blueprint for a climate friendly cement industry, Gland, Switzerland, 2008.
[6] K.L. Scrivener, V.M. John, E.M. Gartner, Eco-efficient cements: potential economically viable solutions for a low-CO2 cement-based materials industry, Cem.
Concr. Res. 114 (2018) 2–26, https://doi.org/10.1016/j.cemconres.2018.03.015.
[7] SNIC, ROADMAP tecnológico do cimento: Potencial de redução das emissões de carbono da indústria do cimento brasileira até 2050 (Roadmap cement
technology: Potential to reduce carbon emissions of the Brazilian cement industry by 2050, in english). coordenado por Gonzalo Visedo e Marcelo Pecchio, Rio
de Janeiro, 2019.
[8] G. Habert, S.A. Miller, V.M. John, J.L. Provis, A. Favier, A. Horvath, K.L. Scrivener, Environmental impacts and decarbonization strategies in the cement and
concrete industries, Nat. Rev. Earth Environ. 1 (2020) 559–573, https://doi.org/10.1038/s43017-020-0093-3.
[9] M. Antoni, J. Rossen, F. Martirena, K. Scrivener, Cement substitution by a combination of metakaolin and limestone, Cem. Concr. Res. 42 (2012) 1579–1589,
https://doi.org/10.1016/j.cemconres.2012.09.006.
[10] K. Scrivener, F. Martirena, S. Bishnoi, S. Maity, Calcined clay limestone cements (LC3), Cem. Concr. Res. 114 (2018) 49–56, https://doi.org/10.1016/j.
cemconres.2017.08.017.
[11] Y. Cancio Díaz, S. Sánchez Berriel, U. Heierli, A.R. Favier, I.R. Sánchez Machado, K.L. Scrivener, J.F. Martirena Hernández, G. Habert, S. Sánchez-Berriel,
U. Heierli, A.R. Favier, I.R. Sánchez Machado, K.L. Scrivener, F. Martirena, G. Habert, Limestone calcined clay cement as a low-carbon solution to meet
expanding cement demand in emerging economies, Dev. Eng. 2 (2017) 82–91, https://doi.org/10.1016/j.deveng.2017.06.001.
[12] M. Moesgaard, D. Herfort, M. Steenberg, L.F. Kirkegaard, Y. Yue, Physical performances of blended cements containing calcium aluminosilicate glass powder
and limestone, Cem. Concr. Res. 41 (2011) 359–364, https://doi.org/10.1016/j.cemconres.2010.12.005.
[13] G. Puerta-Falla, M. Balonis, G. Le Saout, G. Falzone, C. Zhang, N. Neithalath, G. Sant, Elucidating the role of the aluminous rource on limestone reactivity in
cementitious materials, J. Am. Ceram. Soc. 98 (2015) 4076–4089, https://doi.org/10.1111/jace.13806.
[14] K. Vance, A. Kumar, G. Sant, N. Neithalath, The rheological properties of ternary binders containing Portland cement, limestone, and metakaolin or fly ash, Cem.
Concr. Res. 52 (2013) 196–207, https://doi.org/10.1016/j.cemconres.2013.07.007.
[15] K. De Weerdt, M. Ben Haha, G. Le Saout, K.O. Kjellsen, H. Justnes, B. Lothenbach, Hydration mechanisms of ternary Portland cements containing limestone
powder and fly ash, Cem. Concr. Res. 41 (2011) 279–291, https://doi.org/10.1016/j.cemconres.2010.11.014.
[16] C. Vizcayno, R.M. de Gutiérrez, R. Castello, E. Rodriguez, C.E. Guerrero, Pozzolan obtained by mechanochemical and thermal treatments of kaolin, Appl. Clay
Sci. 49 (2010) 405–413, https://doi.org/10.1016/j.clay.2009.09.008.
[17] A. Tironi, M.A. Trezza, E.F. Irassar, A.N. Scian, Thermal treatment of kaolin: effect on the pozzolanic activity, Proc. Mater. Sci. 1 (2012) 343–350, https://doi.
org/10.1016/j.mspro.2012.06.046.
[18] A. Souri, H. Kazemi-Kamyab, R. Snellings, R. Naghizadeh, F. Golestani-Fard, K. Scrivener, Pozzolanic activity of mechanochemically and thermally activated
kaolins in cement, Cem. Concr. Res. 77 (2015) 47–59, https://doi.org/10.1016/j.cemconres.2015.04.017.
[19] M.S. Barata, R.S. Angélica, Pozzolanic activity of kaolin wastes from kaolin mining industry from the amazon region, Matéria 16 (2011) 795–810.
[20] E. Badogiannis, G. Kakali, G. Dimopoulou, E. Chaniotakis, S. Tsivilis, Metakaolin as a main cement constituent. Exploitation of poor Greek kaolins, Cem. Concr.
Compos. 27 (2005) 197–203, https://doi.org/10.1016/j.cemconcomp.2004.02.007.
[21] F. Cassagnabère, P. Diederich, M. Mouret, G. Escadeillas, M. Lachemi, Impact of metakaolin characteristics on the rheological properties of mortar in the fresh
state, Cem. Concr. Compos. 37 (2013) 95–107, https://doi.org/10.1016/j.cemconcomp.2012.12.001.
[22] F.N. Stafford, F. Raupp-pereira, J.A. Labrincha, D. Hotza, Life cycle assessment of the production of cement: a Brazilian case study, J. Clean. Prod. 137 (2016)
1293–1299, https://doi.org/10.1016/j.jclepro.2016.07.050.
[23] A. Favier, F. Zunino, I. Katrantzis, K. Scrivener, The effect of limestone on the performance of ternary blended cement LC3: limestone, calcined clays and cement
BT, in: F. Martirena, A. Favier, K. Scrivener (Eds.), Calcined Clays for Sustainable Concrete, Springer Netherlands, Dordrecht, 2018, pp. 170–175.
[24] S.N. Lim, T.H. Wee, Autogenous shrinkage of ground-granulated blast-furnace slag concrete, Mater. J. 97 (2000) 587–593.
[25] S. Krishnan, D. Gopala Rao, S. Bishnoi, Why low-grade calcined clays are the ideal for the production of limestone calcined clay cement (LC3), in: S. Bishnoi
(Ed.), Calcined Clays for Sustainable Concrete, RILEM Bookseries, 2020: pp. 125–130. 〈https://doi.org/10.1007/978–981-15–2806-4_14〉.
[26] J.M. Marangu, Physico-chemical properties of Kenyan made calcined Clay -Limestone cement (LC3), Case Stud. Constr. Mater. 12 (2020), e00333, https://doi.
org/10.1016/j.cscm.2020.e00333.
[28] S. Krishnan, S. Bishnoi, Understanding the hydration of dolomite in cementitious systems with reactive aluminosilicates such as calcined clay, Cem. Concr. Res.
108 (2018) 116–128, https://doi.org/10.1016/j.cemconres.2018.03.010.
[29] A. Machner, M. Zajac, M. Ben Haha, K.O. Kjellsen, M.R. Geiker, K. De Weerdt, Portland metakaolin cement containing dolomite or limestone – Similarities and
differences in phase assemblage and compressive strength, Constr. Build. Mater. 157 (2017) 214–225, https://doi.org/10.1016/j.conbuildmat.2017.09.056.
[30] F. Avet, K. Scrivener, Investigation of the calcined kaolinite content on the hydration of Limestone Calcined Clay Cement (LC3), Cem. Concr. Res. 107 (2018)
124–135, https://doi.org/10.1016/j.cemconres.2018.02.016.
[31] M.A. Longhi, E.D. Rodríguez, S.A. Bernal, J.L. Provis, A.P. Kirchheim, Valorisation of a kaolin mining waste for the production of geopolymers, J. Clean. Prod.
115 (2016) 265–272, https://doi.org/10.1016/j.jclepro.2015.12.011.
[32] M. Hunger, H.J.H. Brouwers, Flow analysis of water-powder mixtures: application to specific surface area and shape factor, Cem. Concr. Compos. 31 (2009)
39–59, https://doi.org/10.1016/j.cemconcomp.2008.09.010.
[33] ASTM International, C1749 − 17a Standard Guide for Measurement of the Rheological Properties of Hydraulic , (2017) 3–6. https://doi.org/https://doi.org/
10.1520/C1749-17A.
[34] J.S. Raucci, R.T. Cecel, R.C.O. Romano, R.G. Pileggi, V.M. John, Efeito do método de mistura sobre o espalhamento mini-slump de pastas de cimento Portland,
Rev. IBRACON Estrut. Mater. 11 (2018) 410–431, https://doi.org/10.1590/s1983-41952018000200010.
[35] K. Vance, A. Dakhane, G. Sant, Observations on the rheological response of alkali activated fly ash suspensions: the role of activator type and concentration,
Rheol. Acta 53 (2014) 843–855, https://doi.org/10.1007/s00397-014-0793-z.
[36] Associação Brasileira de Normas Técnicas. ABNT, NBR 7215 - Portland cement - Determination of compressive strength of cylindrical test specimens., 2019.
[37] H.J.H. Brouwers, H.J. Radix, Self-compacting concrete: theoretical and experimental study, Cem. Concr. Res. 35 (2005) 2116–2136, https://doi.org/10.1016/j.
cemconres.2005.06.002.
[38] K. Scrivener, F. Avet, H. Maraghechi, F. Zunino, J. Ston, W. Hanpongpun, A. Favier, Impacting factors and properties of limestone calcined clay cements (LC3),
Green Mater. 7 (2018) 3–14, https://doi.org/10.1680/jgrma.18.00029.
[39] P.F.G. Banfill, The rheology of fresh cement and concrete - A review, in: H. Justnes (Ed.), Proceedings of the 11th International Cement Chemistry Congress,
Durban, South Africa, 2003: pp. 50–62.
[40] K. Vance, A. Arora, G. Sant, N. Neithalath, Rheological evaluations of interground and blended cement-limestone suspensions, Constr. Build. Mater. 79 (2015)
65–72, https://doi.org/10.1016/j.conbuildmat.2014.12.054.
[41] J. da, S. Andrade Neto, T.A. Santos, S. de, A. Pinto, C.M.R. Dias, D.V. Ribeiro, Effect of the combined use of carbon nanotubes (CNT) and metakaolin on the
properties of cementitious matrices, Constr. Build. Mater. 271 (2021), 121903, https://doi.org/10.1016/j.conbuildmat.2020.121903.

12
C.S. Malacarne et al. Case Studies in Construction Materials 15 (2021) e00776

[42] N. Shanahan, V. Tran, A. Williams, A. Zayed, Effect of SCM combinations on paste rheology and its relationship to particle characteristics of the mixture, Constr.
Build. Mater. 123 (2016) 745–753, https://doi.org/10.1016/j.conbuildmat.2016.07.094.
[43] I.P. Sfikas, E.G. Badogiannis, K.G. Trezos, Rheology and mechanical characteristics of self-compacting concrete mixtures containing metakaolin, Constr. Build.
Mater. 64 (2014) 121–129, https://doi.org/10.1016/j.conbuildmat.2014.04.048.
[44] T.R. Muzenda, P. Hou, S. Kawashima, T. Sui, X. Cheng, The role of limestone and calcined clay on the rheological properties of LC3, Cem. Concr. Compos. 107
(2020), 103516, https://doi.org/10.1016/j.cemconcomp.2020.103516.
[45] N. Roussel, A. Lemaître, R.J. Flatt, P. Coussot, Steady state flow of cement suspensions: a micromechanical state of the art, Cem. Concr. Res. 40 (2010) 77–84,
https://doi.org/10.1016/j.cemconres.2009.08.026.
[46] B. Lothenbach, K. Scrivener, R.D. Hooton, Supplementary cementitious materials, Cem. Concr. Res. 41 (2011) 1244–1256, https://doi.org/10.1016/j.
cemconres.2010.12.001.
[48] F. Avet, X.R. Li, K. Scrivener, Determination of the amount of reacted metakaolin in calcined clay blends, Ceme. Concr. Res. 106 (2018) 40–48, https://doi.org/
10.1016/j.cemconres.2018.01.009.
[49] F. Zunino, K. Scrivener, The influence of the filler effect on the sulfate requirement of blended cements, Cem. Concr. Res. 126 (2019), 105918, https://doi.org/
10.1016/j.cemconres.2019.105918.
[50] F. Avet, R. Snellings, A. Alujas Diaz, M. Ben Haha, K. Scrivener, Development of a new rapid, relevant and reliable (R3) test method to evaluate the pozzolanic
reactivity of calcined kaolinitic clays, Cem. Concr. Res. 85 (2016) 1–11, https://doi.org/10.1016/j.cemconres.2016.02.015.
[51] Ch Panagiotopoulou, E. Kontori, Th Perraki, G. Kakali, Dissolution of aluminosilicate minerals and by-products in alkaline media, J. Mater. Sci. 42 (2007)
2967–2973, https://doi.org/10.1007/s10853-006-0531-8.
[53] M. Zajac, P. Durdzinski, C. Stabler, J. Skocek, D. Nied, M. Ben Haha, Influence of calcium and magnesium carbonates on hydration kinetics, hydrate assemblage
and microstructural development of metakaolin containing composite cements, Cem. Concr. Res. 106 (2018) 91–102, https://doi.org/10.1016/j.
cemconres.2018.01.008.
[55] F. Avet, K. Scrivener, Hydration study of Limestone Calcined Clay Cement (LC3) using various grades of calcined kaolinitic clays, in: F. Martirena, A. Favier,
K. Scrivener (Eds.), Calcined Clays for Sustainable Concrete, 16th ed., Springer, Dordrecht, 2018, pp. 35–40, https://doi.org/10.1007/978-94-024-1207-9_6.
RILEM Bookseries.
[56] K.L. Scrivener, A. Alujas Diaz, L.M. Vizcaíno Andrés, J.F. Martirena Hernández, M.G. Antoni, Effect of fineness in clinker-calcined clays-limestone cements, Adv.
Cem. Res. 27 (2015) 546–556, https://doi.org/10.1680/adcr.14.00095.
[57] G. Puerta-Falla, M. Balonis, G. Le Saout, N. Neithalath, G. Sant, The influence of metakaolin on limestone reactivity in cementitious materials, in: K. Scrivener,
A. Favier (Eds.), Calcined Clays for Sustainable Concrete, 10th ed., Springer, Dordrecht, 2015, pp. 11–19, https://doi.org/10.1007/978-94-017-9939-3_2.
RILEM Bookseries.
[58] Z. Shi, B. Lothenbach, M.R. Geiker, J. Kaufmann, A. Leemann, S. Ferreiro, J. Skibsted, Experimental studies and thermodynamic modeling of the carbonation of
Portland cement, metakaolin and limestone mortars, Cem. Concr. Res. 88 (2016) 60–72, https://doi.org/10.1016/j.cemconres.2016.06.006.
[59] S.A. Bernal, M.C.G. Juenger, X. Ke, W. Matthes, B. Lothenbach, N. De Belie, J.L. Provis, Characterization of supplementary cementitious materials by thermal
analysis, Mater. Struct./Mater. Constr. 50 (2017), https://doi.org/10.1617/s11527-016-0909-2.
[60] M. Antoni, J. Rossen, F. Martirena, K. Scrivener, Cement substitution by a combination of metakaolin and limestone, Cem. Concr. Res. 42 (2012) 1579–1589,
https://doi.org/10.1016/j.cemconres.2012.09.006.
[61] Y. Kocak, Effects of metakaolin on the hydration development of Portland–composite cement, J. Build. Eng. 31 (2020), 101419, https://doi.org/10.1016/j.
jobe.2020.101419.
[62] M. Antoni, L. Baquerizo, T. Matschei, Investigation of ternary mixes made of clinker limestone and slag or metakaolin: Importance of reactive alumina and silica
content, in: K. Scrivener, A. Favier (Eds.), Calcined Clays for Sustainable Concrete, RILEM Bookseries, 2015: pp. 545–553. 〈https://doi.org/10.1007/978–94-
017–9939-3_68〉.
[63] Antoni, M. Baquerizo, L. Matschei, T. Investigation of Ternary Mixes Made of Clinker Limestone and Slag or Metakaolin: Importance of Reactive Alumina and
Silica Content. In: Calcined Clays for Sustainable Concrete. 10 (2015).
[64] K. Scrivener, F. Avet, H. Maraghechi, F. Zunino, J. Ston, W. Hanpongpun, A. Favier, Impacting factors and properties of Limestone Calcined Clay Cements (LC3),
Green Mater. 7 (2018) 3–14, https://doi.org/10.1680/jgrma.18.00029.
[65] A.K. Tiwari, S. Chowdhury, Relative evaluation of performance of limestone calcined clay cement compared with Portland pozzolana cement, J. Asian Concr.
Fed. 2 (2016) 110–116, https://doi.org/10.18702/acf.2016.12.2.2.110.
[66] A. Favier, F. Avet, M. Antoni, K. Scrivener, Limestone calcined clay cement for a sustainable development, in: G. Habert, A. Schlueter (Eds.), Sustainable Built
Environment (SBE) Regional Conference Zurich 2016., vdf Hochschulverlag AG, Zurich, Switzerland, 2016: pp. 644–648. 〈https://doi.org/10.3218/3774–6_
104〉.
[67] J. Berodier, Impact of the supplementary cementitious materials on the kinetics and microstructural development of cement hydration, Ecole Polytechnique
Federal de Lausanne, Lausanne, Switzerland, 2015. 〈https://doi.org/10.5075/epfl-thesis-6417〉.
[68] Brazilian Association of Technical Standards, ABNT NBR 16697: Portland cement - Requirements, 2019.

13