oward low-carbon cements: maximizing supplementary cementitious materials incorporation in binary, ternary, and quaternary blends

Rumo a cimentos de baixo carbono: maximizando o uso de materiais cimentícios suplementares em cimentos binários, ternários e quartenários

Abstract

The aim of this study was to investigate low-clinker binary, ternary, and quaternary cement systems using fly ash (28–70%), metakaolin (20–46%), limestone (10–23%), and their interactions with calcium hydroxide (0–12%). For this purpose, hydration kinetics, hydration products, rheological behavior, and compressive strength of cement pastes were evaluated. Based on the obtained results, the following conclusions can be drawn: (1) the addition of calcium hydroxide increased the degree of hydration; (2) in binary systems, the incorporation of portlandite by 8% and 12% were not effective; (3) quaternary cements exhibited cumulative heat values similar to those of ternary cements, while presenting higher compressive strength; and (4) the addition of portlandite by 4% and 8% provided the best mechanical performance in quaternary cements, whereas these additions proved insufficient for ternary systems. Overall, the proposed cement systems demonstrated potential as a technological solution to increase the use of supplementary cementitious materials in the production of low-clinker cements.

Keywords
Supplementary cementitious materials; Fly ash; Metakaolin; Calcium hydroxide

Resumo

O objetivo deste estudo foi analisar sistemas binários, ternários e quaternários, com baixo teor de clínquer, utilizando uma cinza volante (28 – 70%), um metacaulim (20 – 46%), e um calcário (10 – 23%) e suas interações com hidróxido de cálcio (0 – 12%). Para tal, analisaram-se o calor de hidratação, os produtos de hidratação, reologia e a resistência à compressão em pastas de cimento. Diante dos resultados obtidos, as seguintes suposições podem ser feitas: (1) a adição de hidróxido de cálcio aumenta o grau de hidratação; (2) em sistemas binários, a adição de teores entre 8 e 12% de Ca(OH)₂ não foi efetiva; (3) o cimento quaternário apresentou calor acumulado semelhante aos cimentos ternários porém houve o aumento da resistência à compressão; (4) a adição de 4% e 8% de Ca(OH)₂ apresentou os melhores resultados mecânicos nos cimentos quaternários e se mostraram insuficientes para os cimentos ternários. Com base no exposto, os cimentos propostos apresentaram potencial como uma solução tecnológica para o aumento do uso de materiais cimentícios suplementares na produção de cimentos com reduzido fator clínquer.

Palavras-chave
Materiais cimentícios suplementares; Cinza volante; Metacaulim; Hidróxido de cálcio

1 Introduction

The cement industry, a pillar of global construction, faces one of the greatest challenges in its history: the urgent need to mitigate its environmental impact. Traditionally, cement production – the primary binder in concrete – has been one of the largest sources of carbon dioxide (CO₂) emissions worldwide, mainly due to the calcination of limestone, a fundamental step in clinker manufacture. It is estimated that the cement industry accounts for approximately 8% of global CO₂ emissions, totaling about 2.4 gigatons in 2019, which represents roughly 26% of total industrial emissions. From this amount, 50% to 60% originates from the thermal decomposition of calcium carbonate (CaCO₃), which releases CO₂ to form calcium oxide (CaO). Another 30 to 40% results from the combustion of fossil fuels and the energy required to operate clinker kilns and other stages of the production process (SNIC, 2019; Fankhauser et al., 2022). In order to change this reality, the pursuit of technologies capable of reducing emissions has become imperative in global climate mitigation efforts. The field of materials, civil and environmental engineering has therefore undergone a deep technological reorientation, directing efforts toward the development of solutions that reconcile technical performance, energy efficiency, and environmental responsibility. This transition represents a new industrial paradigm in which the sector moves from a linear model of extraction, use, and disposal to a circular economy framework. In this context, the valorization of waste from several sectors, along with industrial by-products use, plays a central role by converting materials previously destined for landfills into high-value raw materials. Such an approach significantly reduces environmental impacts, optimizes the use of natural resources, and enhances the resilience of the concrete production chain.

Conceptually, the principle of Net Zero Emissions – which refers to carbon neutrality – has gained prominence. It denotes a balance between the amount of CO₂ emitted and the amount removed from the atmosphere: the main goal is not to eliminate emissions entirely, but rather to offset each ton emitted with an equivalent removal so that the net carbon balance is zero. The concept of negative emissions goes beyond this equilibrium, aiming to remove more CO₂ than is emitted, thereby reducing atmospheric greenhouse gas concentrations. Though more ambitious, complex, and costly, negative emissions are considered essential for long-term mitigation scenarios. Among the main global strategies for CO₂ reduction, Guo et al. (2024) highlight several initiatives recommended by international cement associations: improving industrial energy efficiency, reducing cement consumption through structural optimization, lowering the clinker factor in blended cements, increasing the use of supplementary cementitious materials (SCMs), replacing fossil fuels with renewable alternatives, developing low-carbon cements, optimizing construction and modularity, enhancing carbonation of cementitious materials, and, increasingly, implementing carbon capture, utilization, and storage (CCUS) technologies.

According to the report “Eco-efficient cements: Potential economically viable solutions for a low-CO₂ cement-based materials industry” published by the United Nations Environment Program (U. N. Environment et al., 2018), two major pathways for achieving sustainable concretes and mortars are: “increasing the use of low-CO₂ materials as partial replacements for Portland cement clinker” and “more efficient use of Portland cement clinker in mortars and concretes”. Thus, if the challenge of supplementary cementitious material availability is overcome, determining the upper technical limit for average clinker substitution becomes the next question for academia (U. N. Environment et al., 2018). The adoption of blended cements incorporating higher amounts of SCMs to replace the clinker fraction has been increasingly investigated to meet the demand for reduced clinker use (Zunino; Scrivener, 2022). Furthermore, the use of SCMs as cement replacements is currently the most economical and effective method for reducing carbon dioxide emissions, as normalized CO₂ emissions decrease linearly with increasing substitution levels – potentially reducing global cement-related emissions by 30-40% (Han et al., 2022).

Although developing low-clinker cements presents significant challenges – particularly in ensuring fresh and hardened-state performance comparable to Portland cement – several opportunities arise using different SCMs. One such approach is the combination of multiple materials, which may enhance cement performance and further reduce the clinker factor due to the formation of distinct hydration products and the synergistic development of combined phases. Among the SCMs most commonly used in Brazil, either to supply specific functional components in cement formulations or to reduce clinker content, are fly ash (FA), ground granulated blast-furnace slag (GGBF), calcined clays (MK), limestone filler (LS), dregs, phosphogypsum, flue-gas desulfurization (FGD) gypsum, among several others.

Fly ash, derived from coal combustion in thermoelectric power plants, is produced particularly in southern Brazil. Its characteristics in the fresh state often improve workability due to its cenospheres or spherical particles, which contribute positively to reducing the clinker factor. However, its reactivity must be sufficient to ensure that mechanical strength is not compromised at early or later ages. On one hand, kaolinitic calcined clays can promote strength development in cementitious matrices when used at replacement levels of up to 50% kaolinitic content. On the other hand, in the fresh state, the matrix is negatively affected by the fineness and morphology of clay mineral particles, which typically appear as plate-like structures. This limits higher levels of substitution unless supplemented with superplasticizer admixtures to maintain a flowable concrete.

In LC³ cements, for example, Py et al. (2024) observed in their ternary formulations a complete depletion of portlandite due to the high reactivity of the calcined clays and the intensified pozzolanic reaction occurring from the earliest ages. Considering the use of lower clinker contents, the addition of portlandite in ternary or quaternary systems becomes important to maximize the pozzolanic reaction of SCMs within the cementitious matrix and to prevent the persistence of unreacted anhydrous phases. Thus, there is an inherent balance between reducing clinker content and incorporating pozzolanic materials. To enable pozzolanic reactions under low-clinker conditions, the addition of small amounts of calcium hydroxide may enhance the dissolution of reactive compounds, promote hydrate formation, and consequently increase portlandite consumption (Biernacki; Williams; Stutzman, 2001). These strategies can intensify hydration reactions and maximize SCM utilization in cementitious systems.

The aim of this study is to assess the feasibility of producing low-carbon cement, evaluating workability, hydration products, and mechanical performance in binary, ternary, and quaternary systems. This assessment was carried out using formulations containing only 30% clinker + gypsum, while varying the proportions of fly ash, metakaolin, limestone, and calcium hydroxide.

2 Materials and methodology

2.1 Materials characterization

The materials used in the study were one clinker, a natural gypsum, limestone, fly ash and metakaolin supplied by an industrial partner. The calcium hydroxide used in the formulations had an analytical purity greater than 95%. Based on these materials, twelve systems with different mix formulations were proposed to produce pozzolanic cements containing 30% clinker + gypsum (with 3.9% SO₃ relative to the clinker content). These combinations were classified as binary, ternary, and quaternary systems, in which the pozzolan contents were progressively reduced while the proportion of calcium hydroxide was increased (substituting the pozzolan at levels of 0%, 4%, 8%, and 12%).

A preliminary study was conducted to determine the required water content for each mixture to ensure adequate workability without the use of chemical admixtures. The compositions of binary (B), ternary (T) and quaternary (Q) formulations are presented in Figure 1.

Figure 1
Mix proportion (% wt. by mass)

The chemical composition of the raw materials is presented in Table 1. The clinker is predominantly composed of CaO (62.15%), confirming its high potential for hydration and formation of calcium silicate phases (C₃S and C₂S). The presence of SiO₂ (19.29%), Al₂O₃ (3.75%), and Fe₂O₃ (3.27%) is consistent with typical Portland clinker compositions, governing the formation of silicate and aluminate phases. Gypsum is characterized by a high SO₃ content (45.07%) and CaO (33.38%), confirming its role as a calcium sulfate source responsible for regulating the hydration of aluminates and controlling setting time. Its high loss on ignition (LOI ≈ 19.69%) is associated with its water molecule dehydration. Fly ash presents a typical pozzolanic composition, with high contents of SiO₂ (59.75%) and Al₂O₃ (30.1%), and relatively low CaO (3.22%), indicating a low-calcium (Class F-type) material. This composition suggests lower early reactivity but significant long-term pozzolanic potential. The presence of Fe₂O₃ (3.32%) and minor oxides further reflects its heterogeneous nature. Metakaolin exhibits high Al₂O₃ (42.47%) and SiO₂ (54.82%) contents, confirming its highly aluminosilicate nature. The very low CaO content (0.17%) reinforces that its reactivity is primarily pozzolanic, consuming calcium hydroxide to form additional C–S–H and C–A–S–H phases. The limestone filler is mainly composed of CaO (55.83%), associated with calcium carbonate (CaCO₃), as also evidenced by the high LOI (42.92%) from the decarbonization of carbon dioxide. Its low silica and alumina contents confirm its weakly reactive behavior, acting primarily as a filler and nucleation agent, while also participating in carboaluminate formation. Overall, the chemical characterization confirms a complementary system, where clinker provides hydraulic phases, metakaolin and fly ash act as pozzolanic materials with distinct reactivity levels, gypsum regulates hydration, and limestone contributes mainly through physical and chemical filler effects. This combination supports the development of synergistic interactions in blended cement systems, particularly in ternary and quaternary formulations.

Table 1
Chemical composition of the raw materials by XRF (% wt. mass)

Oxides	Clinker	Gypsum	Fly ash	Metakaolin	Filler
SiO2	19.29	0.81	59.75	54.82	0.47
Al2O3	3.75	0.2	30.1	42.47	0.28
Fe2O3	3.27	0.1	3.32	0.48	0.13
CaO	62.15	33.38	3.22	0.17	55.83
MgO	7.34	0.44	0.19	-	-
SO3	1.56	45.07	0.56	0.11	0.06
K2O	1.7	0.05	2.79	0.17	0.02
P2O5	-	0.01	0.13	0.17	0.13
ZnO	-	-	0.04	0.09	-
SrO	-	0.16	0.03	-	0.16
TiO2	-	0.02	1.81	1.23	-
ZrO2	-	0.01	0.03	-	-
LOI	0.53	19.69	0.97	0.11	42.92

Physical properties of the raw materials are illustrated in Figure 2. The particle size distribution (PSD) and physical properties of the raw materials reveal a well-structured multimodal system, which plays a fundamental role in the fresh and hardened behavior of the studied binders. Metakaolin presents the finest particle size distribution (D₅₀ ~ 3.5 µm) and the highest specific surface area (BET~ 13.5 m²/g), indicating its high reactivity and strong influence on water demand. In contrast, fly ash and limestone exhibit intermediate particle sizes (D₅₀ ~ 7–8 µm) and lower surface areas, which makes fly ash less reactive while also contributing to particle packing, whereas limestone additionally assists through filler effects. Calcium hydroxide and clinker show coarser distributions (D₅₀ ~ 10–11 µm), which reduce surface area-related water demand.

Figure 2
Particle size distribution (PSD) and physical properties – continued lines for cumulative and traced lines for the frequency distribution

The cumulative PSD curves confirm that metakaolin is shifted toward finer sizes, while fly ash, limestone, and Ca(OH)₂ are progressively coarser. The distinct curves further highlight different particle size domains, indicating a complementary granulometric distribution. This multimodal packing improves particle arrangement by filling voids across different scales, reducing interparticle friction and optimizing the packing density of the system.

Such granulometric characteristics directly influence rheological behavior. The finer and more angular particles of metakaolin increase yield stress and viscosity due to higher water adsorption and particle interactions. Conversely, the coarser and more spherical particles of fly ash improve workability by reducing friction and enhancing flowability. The partial replacement of fine particles by calcium hydroxide further contributes to lowering water demand and improving dispersion.

From a hydration perspective, the high surface area of metakaolin accelerates early reactions, while the lower reactivity of fly ash leads to a more gradual contribution over time. The combined granulometric distribution therefore supports both early-age reactivity and long-term performance. Overall, the results demonstrate that the optimized particle size distribution is a key factor governing rheology, hydration kinetics, and mechanical performance in low-clinker blended cement systems.

The pozzolanic reactivity of the supplementary cementitious materials was assessed using the modified Chapelle test, revealing a significant difference between fly ash and metakaolin. The fly ash exhibited relatively low reactivity, with a Chapelle value of 510 mg of Ca(OH)₂ per gram of pozzolan, indicating limited lime consumption and low early-age reactivity. In contrast, the metakaolin showed a markedly higher pozzolanic activity, reaching 1419 mg of Ca(OH)₂/g in the test. These results highlight the substantially greater reactivity of metakaolin, which is consistent with its highly disordered structure and elevated aluminum silicate content, promoting faster reaction with calcium hydroxide during early hydration.

2.2 Experimental procedures

The materials were weighed one day prior to mixing, stored at 22 °C and kept in vacuum-sealed plastic bags in order to prevent the hydration and carbonation of the clinker and calcium hydroxide. The cementitious pastes were prepared in two consecutive steps following the laboratory methodology. First, the solid materials were manually homogenized for 60 seconds. Then, mixing water was added and manually homogenized for 30 seconds. Subsequently, mechanical mixing was performed for 90 seconds using a high-shear mixer operating at 10,000 rpm to ensure uniform dispersion of the components. Cement pastes were cast in cube specimens (20mm), demolded after 24 hours and kept undersaturated with water until the day of analysis. Hydration stoppage was employed with isopropanol as solvent exchange method according to Snellings et al. (2018) and stored in vacuum plastic bags to prevent carbonation for all analysis.

Raw materials were characterized by X-Ray Fluorescence using XRF 1800 Shimadzu spectrometer, with data acquired from 400 to 4000 cm^-1. Particle size distribution (PSD) was obtained by Anton Paar PSA 1090L laser diffraction analyzer, with 50 mL of isopropanol as fluid during the test with 1 min ultrasound and Fraunhofer method. For supplementary cementitious materials PSD analysis, the procedure prescribed by Zunino and Scrivener (2020) to prevent agglomeration was adopted. The specific surface area by BET method was measured with materials preparation according to Mantellato, Palacios and Flatt (2016).

In parallel, the modified Chapelle test enables the determination of the lime consumption capacity of the materials, serving as an indicator of their pozzolanic activity, according to NBR 15895 (ABNT, 2010).

Thermogravimetric analysis (TGA) was performed in a Mettler Toledo TGA 2 instrument, using a heating rate at 20 °C/min in an air atmosphere with nitrogen purge protection of 20 mL/min, to quantify the combined water content and to determine the amounts of calcium hydroxide consumed and remaining in the mixtures at 1, 28, and 91 days, as proposed by Lothenbach, Durdzinksi and Weerdt (2016). Portlandite content (CH) and combined water fraction (CWF) were calculated using the tangent method according to these authors. As a limitation of the present study, qualitative X-ray diffraction (XRD) analyses were not performed.

Heat flow and cumulative heat curves were obtained through isothermal calorimetry as indicators of the hydration reaction kinetics using a TAM Air calorimeter and water as the reference samples as proposed by Wadsö (2010). The procedures were conducted in accordance with C1679-17 (ASTM, 2017) and C1753-15 (ASTM, 2015). Heat flow and total heat released curves were normalized by the mass of cement and the total heat evolved in the first 1 hour was subtracted from the values.

Rheology tests were evaluated using Thermo Fisher Rotational Rheometer HAAKE MARS equipped with parallel-plate geometry (35 mm diameter and 1 mm gap). An initial pre-shear stage was applied for 60 s at a shear rate of 100 s^-1 to ensure that all pastes were evaluated under a consistent reference state of stress and strain, thereby providing a uniform dispersion condition. Subsequently, after a resting period of 60 s, the ascending flow curve was determined by increasing the shear rate from 0.1 to 100 s⁻¹, applying a logarithmic variation between 0.1 and 10 s^-1 and a linear variation between 10 and 100 s^-1. The descending flow curve was obtained following the same procedure as the ascending curve. Rheological measurements were conducted at 5 minutes (300 s). As proposed by Hu and de Larrard (1996) and presented below, equations 1 and 2 were used to determine the dynamic yield stress (τ₀) and the equivalent plastic viscosity (μ_eq) of the samples, based on the Herschel–Bulkley (H–B) model. Data fitting was performed using the RheoWin Job Manager software provided with the equipment.

Eq. 1 $\tau ={\tau }_0+K·\dot{ɣ}·n$

Eq. 2 ${\mu }_{eq}={3K}_n+2{\left({\dot{ɣ}}_{\max }\right)}^{n-1}$

Where:

$\tau$ is the shear stress (Pa) – represents the stress required to maintain the material flow under shear;

${\tau }_0$is the yield stress (Pa) – minimum stress necessary to initiate flow. Below this value, the material behaves like a solid;

$K$is the consistency index – parameter related to the material viscosity or flow consistency. Higher values indicate greater resistance to flow; and

$\dot{\gamma }$is the shear rate ($s^{-1}$) – describes the rate of deformation applied to the material during shearing.

$n$is the flow behavior index (dimensionless) indicates the type of flow behavior:

$n<1$: shear-thinning (pseudoplastic) behavior;

$n=1$: Bingham/Newtonian tendency; and

$n>1$: shear-thickening (dilatant) behavior.

μ_eq is the equivalent viscosity (Pa·s) – represents an apparent or effective viscosity calculated for the material under specific shear conditions.; and

${\dot{\gamma }}_{\max }$is the maximum shear rate ($s^{-1}$) – represents the highest shear rate applied or considered during the rheological test.

The compressive strength test was carried out on the cement paste cubes (20mm), cured under water and lime solution at 22 °C after 1, 7, 28, and 120 days. To perform the test, an EMIC hydraulic testing machine was employed using a 50 kN load cell and a loading rate adjusted to 0.25 MPa according to the cube specimen cross-sectional area, according to NBR 7215 (ABNT, 2019). Mean values and standard deviation from three specimens were used. Statistical analysis using the multi-way ANOVA method was conducted with Tukey’s test using OriginPro 2025 software with a significance level of 5% (p < 0.05).

3 Results and discussion

3.1 Rotational rheometry

In order to avoid the use of chemical admixtures and to ensure that all samples exhibited adequate workability for handling the pastes, it was necessary to investigate the most suitable water-to-binder (w/b) ratio for each cement group, considering the chemical, physical, and morphological differences of the raw materials used in their respective proportions. Three w/b ratios (0.45, 0.50, and 0.55) were evaluated. Since the behavior among the systems was highly variable with respect to the rheological parameters – particularly for the ternary mixtures, in which a w/b ratio of 0.55 was the only condition that allowed proper paste preparation and testing – the study was limited to the rheological behavior of the cement pastes. However, given the large number of systems (12 samples × 3 w/b ratios), the tests were conducted only on mixtures without CH addition and on those with the highest CH content (12%), namely B0, B12, T0, T12, Q0, and Q12.

Figure 3 shows that higher water-to-cement ratios resulted in lower shear stress and reduced resistance of the pastes. This behavior was expected, as the increased availability of free water in the mixture reduces interactions among solid particles. In addition, it is generally observed that pastes containing fly ash (mixture B) exhibit lower yield stress and plastic viscosity compared to formulations with metakaolin and limestone (mixture T). This trend is consistent with findings reported in previous studies, which highlight the distinct effects of fly ash and metakaolin on the workability of cementitious systems (Silvestro et al., 2024; Vance et al., 2013). The reduction in yield stress and plastic viscosity in fly ash mixtures is associated with the spherical shape of its particles, which enhances paste fluidity. Furthermore, the particle size distribution of fly ash, slightly coarser than that of metakaolin, intensifies this positive effect on workability. This occurs because the increased spacing between neighboring particles reduces surface friction, facilitating their relative movement (Vance et al., 2013). In contrast, metakaolin, due to its finer and more angular particles, tends to increase both viscosity and yield stress, reinforcing the distinct rheological behavior between these two materials.

Figure 3
Dynamic yield stress and equivalent viscosity by rotational rheometry – (a) binary, (b) ternary and (c) quaternary cement blends

For the ternary mixtures, it was not possible to establish the same trend observations across different water-to-cement ratios, as reliable rheometer measurements could not be obtained due to the low plasticity of the pastes at w/b ratios of 0.45 and 0.50. However, at a w/b ratio of 0.55, a significant reduction in yield stress (74%) and paste viscosity (40%) was observed with the incorporation of CH and the corresponding reduction in metakaolin and limestone contents. This behavior is attributed to the replacement of finer and more reactive materials, such as metakaolin, by CH, which has a coarser particle size distribution. Materials with higher specific surface area tend to adsorb more water onto their particle surfaces, thereby reducing the amount of free water available to lubricate the system and promote flowability (Vance et al., 2013).

In contrast, mixtures containing 12% calcium hydroxide (B12 and Q12) exhibited higher yield stress and viscosity compared to B0 and Q0, regardless of the w/b ratio. This behavior is associated with the reduction in fly ash content to accommodate the incorporation of CH. These results further highlight the role of fly ash in enhancing the workability of cement pastes, as previously discussed.

When comparing the rheological performance of mixtures B0, T0, and Q0 at a w/b ratio of 0.55, the effectiveness of fly ash in improving the workability of cements produced with metakaolin and limestone becomes evident. The development of a quaternary cement system resulted, even with the addition of CH, in mixtures with lower yield stress and viscosity compared to the ternary cement proposed in this study.

Thus, it was possible to prepare and mold quaternary cement pastes using a lower w/b ratio (0.50) compared to that required for the ternary systems (0.55). These results highlight the influence of CH incorporation and the varying proportions of fly ash on the fresh-state behavior of the investigated binders. Based on the rheological performance and workability obtained within the investigated range, the w/b ratios adopted to produce the pastes were 0.45 for binary cements, 0.55 for ternary cements, and 0.50 for quaternary cements.

It is important to emphasize that the differences observed among the investigated systems are associated not only with the intrinsic reactivity of the supplementary cementitious materials, but also with variations in particle size distribution, particle morphology, and specific surface area. Consequently, physical packing effects, filler behavior, and surface-related interactions occur simultaneously with chemical effects and cannot be completely dissociated in the present study. Therefore, the rheological and mechanical performance discussed throughout this work should be interpreted considering the combined contribution of both physical and chemical factors.

3.2 Portlandite and combined water fraction content

The differential thermogravimetric curves, portlandite and combined water fraction (CWF) content after 1, 28, and 91 days of curing are illustrated in Figures 4 and 5. The behavior of binary, ternary, and quaternary cement pastes indicates differences in the formation of hydration products. All cement types exhibit three main mass loss regions. The relationship between CWF and calcium hydroxide consumption is linear, meaning that CH consumption increases with increasing combined water content (Biernacki; Williams; Stutzman, 2001).

Figure 4
DTG curves from thermogravimetric analysis – (a) binary, (b) ternary and (c) quaternary cement blends

Figure 5
Portlandite and combined water fraction content – (a) binary, (b) ternary and (c) quaternary cement blends

The first main mass loss regions were identified at temperatures of approximately 80–100 °C and around 125 °C, corresponding to the decomposition of primary hydration products such as C–S–H, ettringite, and carboaluminates (both monocarboaluminate, Mc, and hemicarboaluminate, Hc). The second region occurs at approximately 440–450 °C, associated with the decomposition of portlandite. Finally, above 600 °C, carbonates decompose, characterizing the third mass loss region (Bernal et al., 2017; Shi et al., 2016).

The medium to high intensity of the portlandite peak (440-450 °C) at all ages in binary cements indicates the limited pozzolanic reaction of B0 during the evaluated period, as well as the excess Ca(OH)₂ present in B4, B8, and B12. However, the higher alkalinity provided by Ca(OH)₂ addition may enhance pozzolanic reactions at later ages. Therefore, the addition of higher CH content was not necessary for these mixtures. Over time, the CH content decreases as a result of pozzolanic reactions. Due to the absence of limestone filler in binary cements, the mass loss observed around 700 °C can be attributed to carbonation during sample preparation (Malacarne et al., 2021). Consequently, the quantification of portlandite may have been affected, since the reaction between calcium hydroxide and atmospheric CO₂ results in the formation of calcium carbonates (CaCO₃). As already described in the methodology section, special care was taken to keep the materials vacuum-sealed until the moment of mixing. Nevertheless, during the weighing procedures, during paste preparation and the first day of hydration carbonation might have occurred due to exposure to the environment.

As shown in Figure 5, binary cements exhibited the highest residual CH contents that might be attributed to the lack of pozzolanic activity and the stronger dilution effects in the hydrated pastes.

A different behavior was observed for ternary binders. In respect to the high reactivity of metakaolin, a lower amount of Ca(OH)₂ in the system was observed only at 1 day of hydration for all addition levels tested. This indicates the potential for incorporating higher CH contents in ternary systems with high reactive SCM. Nevertheless, CH consumption led to the formation of additional C–S–H, as also reported in studies on low-clinker LC³ systems (<50%) with higher CH additions (15, 17.5, and 20%) (Sun; Zunino; Scrivener, 2024; Py et al., 2024).

For quaternary cements, a similar behavior was observed. However, CH additions above 8% were excessive due to the reduced metakaolin content compared to ternary systems. In both cases, the addition of Ca(OH)₂ acted as a promoter of pozzolanic reactions, as evidenced by the increase in cumulative heat and combined water fraction. Therefore, since fly ash is less reactive at early ages when compared to metakaolin, the behavior observed in the quaternary cement systems is consistent with the expected results. In addition, another important aspect to highlight concerns the use of metakaolin and other highly reactive pozzolans in cementitious systems, which may promote rapid densification of the cement matrix and consequently limit the continued growth of hydration products at later ages (Antoni et al., 2012). Thus, a more balanced hydration process promoted by the incorporation of fly ash may contribute to improvements in both long-term and even early-age regarding hydration and compressive strength development.

In ternary and quaternary cements, the mass loss observed above 600 °C is associated with the decomposition of calcium carbonates from limestone (Antoni et al., 2012, Lothenbach; Durdzinski; Weerdt, 2016). A reduction in the intensity of this region over time was observed, indicating partial consumption of limestone, particularly for the formation of carboaluminate phases detected around 150 °C. These phases form after 1 day of curing and tend to increase over time. However, despite this partial consumption, most of the limestone remains acting as a filler (Malacarne et al., 2021).

The fraction of combined water provides a general indication of the amount of hydration products formed as higher hydration corresponds to a higher degree of hydration (DoH) (John et al., 2019). This behavior is evident in Figure 5, where ternary and quaternary cements exhibit higher CWF values than binary cements, due to differences in system composition and hydration kinetics.

3.3 Hydration kinetics

From the isothermal calorimetry curves illustrated in Figure 6, it can be observed that the different cement compositions significantly influenced the hydration kinetics. Immediately after the induction period, the initial reactions begin and hydration products start to form (Talero et al., 2017). The first peak, associated with alite dissolution and C–S–H precipitation, occurred between 8 and 11 hours for binary cements and between 7 and 8 hours for quaternary systems. In ternary cements, however, this peak was not clearly defined, likely due to its overlap with the second peak, as the ternary system exhibited slight undersulfation because of metakaolin’s greater surface specific area.

Figure 6
Heat flow and total heat released curves from isothermal calorimetry normalized per gram of solids – (a) binary, (b) ternary and (c) quaternary cement blends

The second peak (related to C₃A dissolution and ettringite formation) was more pronounced in ternary cements, occurring between 7 and 8 hours of hydration. In quaternary cements, this peak showed lower intensity and was shifted to approximately 12 hours, while in binary systems it was less significant, appearing only after 20 hours of hydration.

The third peak, associated with the formation of AFm phases, was not identified in binary cements within the evaluated period. In ternary systems, however, the formation of hemi- and monocarboaluminates was observed after approximately 24 hours of hydration. In quaternary cements, this peak was also present, although with lower intensity, occurring at around 38 hours.

Ternary cements exhibited undersulfated behavior, as sulfate depletion occurred shortly after the first peak (Andrade Neto; de la Torre; Kirchheim, 2021). This phenomenon was also observed in quaternary cements, whereas to a lesser extent. It was related to the fact that, in quaternary blends, part of the metakaolin content was replaced by fly ash. The aluminate peak was most pronounced in ternary cements, followed by quaternary and then binary systems. The highest value was recorded for mixture T0 (2.97 mW/g) at 7.4 h, while B4 reached 1.31 mW/g at 14.6 h, and Q4 reached 2.00 mW/g at 12.1 h.

In ternary and quaternary systems, the incorporation of metakaolin provided additional surface area and a source of aluminum beyond that supplied by C₃A hydration (Silva et al., 2021; Zunino; Scrivener, 2019). The presence of highly reactive alumina accelerated the reactions but also contributed to sulfate depletion in ternary systems (Silva et al., 2021).

When comparing the cumulative heat at 120 h, mixture Q12 exhibited the highest value (200.58 J/g), slightly higher than T12 (199.32 J/g) and 6.45% greater than B12 (187.64 J/g). As expected, ternary and quaternary cements, due to the presence of metakaolin, showed higher cumulative heat (Talero et al., 2017). Furthermore, the increased availability of CH (12%) proved effective in enhancing the pozzolanic reaction over time, resulting in higher cumulative heat with increasing portlandite content in ternary and quaternary systems.

As observed for the binary system, differently from the other investigated formulations, the addition of portlandite slightly accelerated the hydration kinetics during the first 24 hours when compared to the mixture without calcium hydroxide. This behavior may be associated with the simpler composition of the binary mixtures and with the partial replacement of fly ash — a less reactive SCM at early ages — by calcium hydroxide. As a consequence, the systems containing CH exhibited slightly higher cumulative heat release values and a modest acceleration of the main hydration peak. However, despite this initial acceleration, the overall evolution of hydration remained relatively similar among the binary formulations, indicating that the low early-age reactivity of fly ash still governed the hydration process.

In ternary and quaternary systems, however, this effect became less visible and less pronounced due to the presence of a greater amount of SCMs simultaneously involved in dissolution reactions and hydrate nucleation processes. Furthermore, since the ternary system presented a higher specific surface area, likely associated with its higher metakaolin content, the differences in sulfate depletion time and in the occurrence of the maximum aluminate peak did not differ by more than approximately one hour. On the other hand, the overall shape of the calorimetric curve changed in comparison with the other systems, mainly due to the combined presence of limestone filler and metakaolin in the ternary formulation.

Another important aspect observed when comparing the calorimetric curves of the ternary systems is that, in these formulations, metakaolin was progressively replaced by calcium hydroxide. In other words, the ternary system without portlandite addition contained approximately 47% metakaolin, whereas the formulation with the highest CH addition contained only about 36% metakaolin. In contrast, for the quaternary systems, the metakaolin content remained constant in all formulations, while the incorporation of calcium hydroxide occurred through the replacement of fly ash.

Therefore, since fly ash is less reactive at early ages when compared to metakaolin, the behavior observed in the quaternary cement systems is consistent with the expected results. In addition, another important aspect to highlight concerns the use of metakaolin and other highly reactive pozzolans in cementitious systems, which may promote rapid densification of the cement matrix and consequently limit the continued growth of hydration products at later ages. Thus, a more balanced hydration process promoted by the incorporation of fly ash may contribute to improvements in both long-term and even early-age compressive strength development. Overall, the calorimetry results indicate that the combined interaction between SCM reactivity, particle packing, filler effects, and calcium hydroxide incorporation strongly influences the hydration kinetics of the investigated low-clinker cement systems.

3.4 Compressive strength

The compressive strength results of paste cube specimens (20 mm), evaluated at 1, 7, 28, and 120 days, are presented in Figure 7. In general, all cements exhibited an increase in compressive strength. The addition of 12% Ca(OH)₂ negatively affected the 1-day strength of all mixtures when compared to the others. On the other hand, all mixtures containing 8% calcium hydroxide showed an increase of up to 42% in compressive strength at 28 days. For mixtures with 4% CH, no significant differences in strength were observed across all cement types, indicating that the availability of portlandite at this dosage did not significantly influence strength development up to 120 days.

Figure 7
Cement pastes compressive strength at 1, 7, 28 and 120 days

At 1 day of hydration, pastes produced with ternary cements exhibited higher strength than the other systems, due to the intense early-age reactivity of metakaolin. The high fineness of its particles promotes rapid dissolution and formation of hydration products, surpassing the slower strength development typically associated with fly ash-based systems. This behavior persisted at 7 days, when synergistic reactions among limestone, metakaolin, and clinker led to the formation of additional hydration phases, such as hemicarboaluminate (Hc) and monocarboaluminate (Mc) (Cancio Díaz et al., 2017).

For quaternary cements, a significant increase in compressive strength was observed at 7 days, mainly attributed to the contribution of metakaolin hydration. The cumulative heat curves (Figure 6) support these findings: at 5 days, quaternary cements already exhibited cumulative heat values equal to or higher than those of ternary systems, reaching 129.41 J/g (Q12) and 128.59 J/g (T12), respectively.

At 28 days, a marked increase in strength was observed for binary cements, whose composition is predominantly based on fly ash. A similar trend was observed for quaternary cements, which also contain a significant proportion of this material. In both cases, the lower reactivity of fly ash contributes to a gradual development of strength over time. In contrast, ternary cements, due to the high reactivity of metakaolin, tend to reach their peak strength at early ages, followed by a tendency toward stabilization at later ages, as clinker hydration stabilizes and the availability of reactive pozzolans decreases.

The best mechanical performance within each group was observed for mixtures B8, T8, and Q8. Considering the 28-day compressive strength, quaternary cements demonstrated promising results compared to the other systems investigated in this study. These compressive strength results are consistent with the portlandite content and CWF values presented in Figure 5: binary cements exhibited lower strength at all ages, along with lower combined water fractions and higher residual portlandite contents in the systems.

Similarly, the compressive strength development observed among the investigated systems should not be interpreted exclusively as a consequence of differences in pozzolanic reactivity. Variations in particle packing, surface area, flowability and filler effects resulting from the distinct PSDs of the raw materials may also have significantly influenced mechanical performance.

Tables 2 and 3 present the results of the multi-way ANOVA and Tukey’s multiple comparison test performed for the compressive strength results considering the investigated factors: cement type (binary, ternary, and quaternary systems), calcium hydroxide content (0%, 4%, 8%, and 12%), and curing age (1, 7, 28, and 120 days). Statistical analysis was conducted using OriginPro 2025 software with a significance level of 5% (p < 0.05). The ANOVA evaluated the significance of the main factors and their interactions, while Tukey’s test was used to identify statistically significant differences among the investigated formulations. Formulations sharing at least one common letter belong to statistically equivalent groups and therefore do not present significant differences in compressive strength at the adopted confidence level.

Table 2
Three-way ANOVA

Factors and interactions	DoF	Sum of Squares	Mean Square	F Value	p-value
A- Cement (B, T or Q)	2	2774.95	1387.47	588.47	<0.0001
B- CH content	3	267.06	89.02	37.76	<0.0001
C- Age	3	16834.64	5611.55	2380.02	<0.0001
AxB	6	241.59	40.26	17.08	<0.0001
AxC	6	1280.78	213.46	90.54	<0.0001
BxC	9	538.13	59.79	25.36	<0.0001
AxBxC	18	215.28	11.96	5.07	<0.0001
Model	47	22152.42	471.33	199.90	<0.0001
Error	96	226.35	2.36	-	-

Table 3
Tukey test

Cement	CH content	Age	Mean	Groups
Q	8	28	41.48	A
Q	12	120	40.88	A	B
T	8	120	38.15	A	B	C
Q	8	120	37.20		B	C	D
T	4	120	34.11			C	D	E
Q	4	28	33.68				D	E	F
T	8	28	32.60					E	F	G
B	0	120	32.15					E	F	G	H
Q	12	7	31.45					E	F	G	H
B	8	120	31.26					E	F	G	H
Q	0	120	30.99					E	F	G	H
Q	4	120	30.76					E	F	G	H
T	0	120	30.54					E	F	G	H
Q	12	28	30.26					E	F	G	H
T	12	120	30.05					E	F	G	H	I
B	12	120	29.88						F	G	H	I
T	4	28	29.82						F	G	H	I
Q	8	7	29.14							G	H	I
Q	0	28	28.96							G	H	I	J
T	0	28	28.18								H	I	J	K
B	4	120	25.98									I	J	K	L
T	12	7	24.78										J	K	L	M
T	12	28	24.14											K	L	M	N
Q	4	7	23.89												L	M	N	O
T	8	7	23.79												L	M	N	O
Q	0	7	22.27												L	M	N	O
B	8	28	21.83												L	M	N	O
T	4	7	21.11													M	N	O
T	0	7	20.57														N	O
B	4	28	20.52														N	O
B	0	28	19.69															O
B	12	28	12.14																P
B	12	7	9.37																P	Q
B	4	7	7.49																	Q	R
T	0	1	7.15																	Q	R
B	0	7	6.84																	Q	R	S
B	8	7	6.68																	Q	R	S	T
T	4	1	6.62																	Q	R	S	T
T	8	1	5.52																	Q	R	S	T	U
Q	8	1	4.87																		R	S	T	U
Q	4	1	4.48																		R	S	T	U
Q	0	1	4.28																		R	S	T	U
B	8	1	2.75																			S	T	U
Q	12	1	2.74																			S	T	U
B	4	1	2.62																				T	U
B	0	1	2.27																					U
T	12	1	1.92																					U
B	12	1	1.48																					U

The three-way ANOVA results demonstrated that all investigated factors: cement type (binary, ternary, and quaternary systems), CH content, and hydration age, significantly influenced compressive strength development (p < 0.0001). In addition, all interactions between the investigated factors were also statistically significant, indicating that the mechanical behavior of the systems resulted from the combined interaction between binder composition, calcium hydroxide content, and hydration evolution over time.

Complementarily, Tukey’s multiple comparison test allowed the identification of statistically significant differences among the investigated formulations. The results indicated that the highest compressive strength values were obtained for ternary and quaternary systems at advanced curing ages, particularly for Q8 at 28 days (41.48 MPa), Q12 at 120 days (40.88 MPa), and T8 at 120 days (38.15 MPa), which belonged to the highest statistical groups (A–C). These findings confirm the strong mechanical potential of systems combining metakaolin and fly ash.

The quaternary systems exhibited particularly relevant behavior, since formulations containing 4%, 8%, and 12% CH remained among the statistically superior groups at later ages. This behavior suggests that the incorporation of fly ash promoted a more balanced hydration process, reducing the rapid early-age densification caused by metakaolin and favoring the continuation of pozzolanic reactions over time. This interpretation is consistent with the calorimetry and combined water fraction results, which indicated a more gradual hydration evolution in the quaternary systems.

For the ternary systems, high compressive strength values were observed at early and intermediate ages, especially for T8, reflecting the high reactivity of metakaolin. However, some formulations showed a tendency toward strength stabilization at later ages, which may be associated with the rapid densification of the cementitious matrix and the consequent limitation of available space for continued hydrate growth over time, as discussed in the literature for systems containing highly reactive pozzolans.

Binary systems exhibited the lowest overall mechanical performance, particularly at early ages, as observed for B0, B4, and B12 at 1 and 7 days, which were positioned within the lowest statistical groups (P–U). This behavior is associated with the lower early-age reactivity of fly ash, reflecting its slower hydration kinetics. Nevertheless, at later ages, some binary formulations exhibited significant strength development, indicating the delayed pozzolanic contribution of fly ash.

Regarding CH incorporation, the statistical analysis indicated that intermediate calcium hydroxide contents, particularly 4% and 8%, positively influenced the mechanical performance of ternary and quaternary systems. On the other hand, higher CH additions (12%) did not provide proportional improvements for all investigated systems, especially for binary mixtures, suggesting that excessive CH incorporation does not necessarily result in additional compressive strength gains.

Overall, the ANOVA and Tukey test results confirm that the mechanical performance of the investigated systems cannot be attributed exclusively to the intrinsic pozzolanic reactivity of the supplementary cementitious materials. Instead, the observed behavior results from the combined interaction between chemical composition, particle size distribution, filler effects, particle packing, and hydration kinetics throughout the curing period.

4 Conclusions

The present study evaluated the effect of binder composition in binary, ternary, and quaternary cement systems with 70 wt.% clinker replacement by supplementary cementitious materials – fly ash (28 – 70%), metakaolin (20 – 46%), limestone (10 – 23%) and Ca(OH)₂ (0–12%) to enhance pozzolanic reactions. Based on the experimental results, the main conclusions can be summarized as follows:

1. the incorporation of fly ash led to a reduction in the cumulative heat of hydration, reflecting its lower intrinsic reactivity and slower reaction kinetics compared to metakaolin-based systems;

2. quaternary cements were able to maintain relatively high cumulative heat release while simultaneously improving compressive strength, indicating a more balanced synergy between filler, pozzolanic, and nucleation effects;

3. the addition of portlandite proved to be beneficial in ternary and quaternary systems, as it enhanced the availability of calcium ions, promoting additional pozzolanic reactions and increasing the formation of hydration products, even in cases where CH was later depleted;

4. a clear relationship was observed between the fraction of combined water and compressive strength, confirming that higher degrees of hydration directly contribute to improved mechanical performance;

5. high levels of clinker replacement by fly ash, although beneficial for sustainability, resulted in reduced compressive strength, particularly at an early age, due to its delayed reactivity;

6. quaternary cements exhibited the best overall mechanical performance over time, combining the early reactivity of metakaolin with the long-term contribution of fly ash;

7. among the evaluated compositions, cements containing overall 8% Ca(OH)₂ showed the most favorable balance between hydration kinetics and mechanical strength at all tested ages;

8. the combined use of metakaolin and fly ash in quaternary systems proved to be an effective strategy to optimize both early- and late-age performance, mitigating the limitations associated with each material when used individually; and

9. overall, the results highlight the importance of tailoring binder composition to achieve an optimal balance between reactivity, workability, and mechanical performance in low-clinker cement systems.

• FROENER, M. S.; PY, L. G.; LONGHI, M. A.; KIRCHHEIM, A. P. Toward low-carbon cements: maximizing supplementary cementitious materials incorporation in binary, ternary, and quaternary blends. Ambiente Construído, Porto Alegre, v. 26, e154614, jan./dez. 2026. ISSN 1678-8621 Associação Nacional de Tecnologia do Ambiente Construído. http://dx.doi.org/10.1590/s1678-86212026000101007
• Declaration of Generative AI and AI-Assisted Technologies in the Writing Process
  The authors declare that AI was used solely for manuscript formatting and language revision. AI was not used for data analysis, interpretation of results, or the generation of scientific content. The authors reviewed and approved the final manuscript and take full responsibility for its content.
• Financial Support
  The authors gratefully acknowledge the financial support provided by the Brazilian funding agencies the National Council for Scientific and Technological Development (CNPq), the Coordination for the Improvement of Higher Education Personnel (CAPES), and the Research Support Foundation of the State of Rio Grande do Sul (FAPERGS), as well as by InterCement Brazil.

Data Availability Statement

Data will be made available on reasonable request.

References

• AMERICAN SOCIETY FOR TESTING AND MATERIALS. C1679-17: standard practice for measuring hydration kinetics of hydraulic cementitious mixtures using isothermal calorimetry. West Conshohocken, 2017.
• AMERICAN SOCIETY FOR TESTING AND MATERIALS. C1753-15: standard practice for evaluating early hydration of hydraulic cementitious mixtures using thermal measurements. West Conshohocken, 2015.
• ANDRADE NETO, J. da S.; DE LA TORRE, A. G.; KIRCHHEIM, A. P. Effects of sulfates on the hydration of Portland cement: a review. Construction and Building Materials, v. 279, abr. 2021.
• ANTONI, M. et al Cement substitution by a combination of metakaolin and limestone. Cement and Concrete Research, v. 42, n. 12, p. 1579–1589, dez. 2012.
• ASSOCIAÇÃO BRASILEIRA DE NORMAS TÉCNICAS. NBR 15895: materiais pozolânicos: determinação do teor de hidróxido de cálcio fixado: método Chapelle modificado. Rio de Janeiro, 2010.
• ASSOCIAÇÃO BRASILEIRA DE NORMAS TÉCNICAS. NBR 7215: cimento Portland: determinação da resistência à compressão de corpos de prova cilíndricos. Rio de Janeiro, 2019.
• BERNAL, S. A. et al. Characterization of supplementary cementitious materials by thermal analysis. Materials and Structures/Materiaux et Constructions, v. 50, n. 1, fev. 2017.
• BIERNACKI, J. J.; WILLIAMS, P. J.; STUTZMAN, P. E. Kinetics of reaction of calcium hydroxide and fly ash. ACI Materials Journal, v. 98, n. 4, 2001.
• CANCIO DÍAZ, Y. et al. Limestone calcined clay cement as a low-carbon solution to meet expanding cement demand in emerging economies. Development Engineering, v. 2, p. 82–91, 2017.
• U. N. ENVIRONMENT et al. Eco-efficient cements: potential economically viable solutions for a low-CO2 cement-based materials industry. Cement and concrete Research, v. 114, p. 2-26, 2018.
• FANKHAUSER, S. et al The meaning of net zero and how to get it right. Nature Climate Change, v. 12, p. 15–21, dec. 2022.
• GUO, Y. et al. A review of low-carbon technologies and projects for the global cement industry. Journal of Environmental Sciences. v. 136, p. 682–697, 2024.
• HAN, Y. et al. Hydration–strength–workability–durability of binary, ternary, and quaternary composite pastes. Materials, v. 15, n. 1, jan. 2022.
• HU, C.; de LARRARD, F. The rheology of fresh high-performance concrete. Cement and Concrete Research, v. 26, n. 2, p. 283–294, 1996.
• JOHN, V. M. et al. Rethinking cement standards: Opportunities for a better future. Cement and Concrete Research, v. 124, out. 2019.
• LOTHENBACH, B.; DURDZINSKI, P.; WEERDT, K. Thermogravimetric analysis. In: SCRIVINER, K.; SNELLINGS, R.; LOTHENBACH, B. (org.). A pratical guide to microstructural analysis of cementitious materials New York: Taylor & Francis Group, 2016.
• MALACARNE, C. S. et al. Influence of low-grade materials as clinker substitute on the rheological behavior, hydration and mechanical performance of ternary cements. Case Studies in Construction Materials, v. 15, dez. 2021.
• MANTELLATO, S.; PALACIOS, M.; FLATT, R. J. Impact of sample preparation on the specific surface area of synthetic ettringite, Cement and Concrete Research, v. 86, p. 20–28, 2016.
• PY, L. G. et al. Evaluation of ultrafine calcined clays on LC3 cements on the sulfate requirement, water demand and strength development. Materials and Structures/Materiaux et Constructions, v. 57, n. 1, fev. 2024.
• SHI, Z. et al. Experimental studies and thermodynamic modeling of the carbonation of Portland cement, metakaolin and limestone mortars. Cement and Concrete Research, v. 88, p. 60–72, out. 2016.
• SILVA, M. R. C. da et al. Valorization of kaolin mining waste from the Amazon region (Brazil) for the low-carbon cement production. Case Studies in Construction Materials, v. 15, dez. 2021.
• SILVESTRO, L. et al. Rotational rheometry test of Portland cement-based materials: a systematic literature review. Construction and Building Materials, v. 432, jun. 2024.
• SINDICATO NACIONAL DA INDÚSTRIA DO CIMENTO. Roadmap Tecnológico tecnológico do cimento: potencial de redução das emissões de carbono da indústria do cimento brasileira até 2050. 2019. Available: https://www.dropbox.com/s/9cbtj1c9oquigf8/Roadmap Tecnológico do Cimento_Brasil.pdf?dl=0
  » https://www.dropbox.com/s/9cbtj1c9oquigf8/Roadmap Tecnológico do Cimento_Brasil.pdf?dl=0
• SNELLINGS, R. et al. Report of TC 238-SCM: hydration stoppage methods for phase assemblage studies of blended cements—results of a round robin test. Material Structure Construction v. 51, 2018.
• SUN, J.; ZUNINO, F.; SCRIVENER, K. Hydration and phase assemblage of limestone calcined clay cements (LC³) with clinker content below 50 %. Cement and Concrete Research, v. 177, mar. 2024.
• TALERO, R. et al. Role of the filler on Portland cement hydration at very early ages: Rheological behaviour of their fresh cement pastes. Construction and Building Materials, v. 151, p. 939–949, out. 2017.
• VANCE, K. et al. The rheological properties of ternary binders containing Portland cement, limestone, and metakaolin or fly ash. Cement and Concrete Research, v. 52, p. 196–207, 2013.
• WADSÖ, L. Operational issues in isothermal calorimetry. Cement and Concrete Research. v. 40, n. 7, p. 1129–1137, 2010.
• ZUNINO, F.; SCRIVENER, K. Increasing the kaolinite content of raw clays using particle classification techniques for use as supplementary cementitious materials. Construction and Building Material, v. 244, 118335, may. 2020.
• ZUNINO, F.; SCRIVENER, K. L. The influence of the filler effect on the sulfate requirement of blended cements. Cement and Concrete Research, v. 126, 105918, dez. 2019.
• ZUNINO, F.; SCRIVENER, K. Microstructural developments of limestone calcined clay cement (LC³) pastes after long-term (3 years) hydration. Cement and Concrete Research, v. 153, 106693, mar. 2022.

Edited by

• Editor-in-chief:
  Enedir Ghisi

• Guest editor:
  Marlova Piva Kulakowski

Publication Dates

• Publication in this collection
  03 Aug 2026
• Date of issue
  Jan-Dec 2026

History

• Received
  30 Mar 2026
• Reviewed
  21 May 2026
• Accepted
  26 June 2026