Initial hydration of tricalcium silicate as studied by secondary neutrals mass spectrometry: II. Results and discussion

The hydration of tricalcium silicate (C₃S) was studied by secondary neutrals mass spectrometry (SNMS), a method that enables determination of the Ca/Si ratio of the formed calcium silicate hydrate (C-S-H) phase with an extremely low information depth. It was found that the magnitude of this parameter within the hydrate layer formed at the surface of the nonhydrated C₃S is not constant and increases with increasing distance from the liquid-solid interface. It was also found that, at a constant distance from the surface, the Ca/Si ratio declines with hydration time. The kinetics of the hydration process is characterized by a very fast initial reaction, followed by a dormant period and a subsequent period of renewed hydration. The rate of hydration becomes distinctly accelerated by elevated temperature and retarded by the presence of sucrose, while NaCl affects the initial hydration kinetics only to a small degree.     

A Reaction Zone Hypothesis for the Effects of Particle Size and Water‐to‐Cement Ratio on the Early Hydration Kinetics of C3S

The hydration kinetics of tricalcium silicate (C₃S) has been the subject of much study, yet the experimentally observed effects of the water‐to‐cement (w/c) ratio and particle size distribution have been difficult to explain with models. Here, we propose a simple hypothesis that provides an explanation of the lack of any significant effect of w/c on the kinetics and for the strong effect of the particle size distribution on the amount of early hydration associated with the main hydration peak. The hypothesis is that during the early hydration period the calcium–silicate–hydrate product forms only in a reaction zone close to the surface of the C₃S particles. To test the hypothesis, a new microstructure‐based kinetics (MBK) model has been developed. The MBK model treats the C₃S particle size distribution in a statistical way to save computation time and treats the early hydration as essentially a boundary nucleation and growth process. The MBK model is used to fit kinetic data from two published studies for C₃S with different size distributions, one for alite (impure C₃S) pastes and one for stirred C₃S suspensions. The model is able to fit all the data sets with parameters that show no significant trend with particle size, providing support for the reaction zone hypothesis.     

Time dependent driving forces and the kinetics of tricalcium silicate hydration

Simulations of tricalcium silicate (C₃S) hydration using a kinetic cellular automaton program, HydratiCA, indicate that the net rate depends both on C₃S dissolution and on hydration product growth. Neither process can be considered the sole rate-controlling step because the solution remains significantly undersaturated with respect to C₃S yet significantly supersaturated with respect to calcium silicate hydrate (C–S–H). The reaction rate peak is attributed to increasing coverage of C₃S by C–S–H, which reduces both the dissolution rate and the supersaturation of C–S–H. This supersaturation dependence is included in a generalized boundary nucleation and growth model to describe the kinetics without requiring significant impingement of products on separate cement grains. The latter point explains the observation that paste hydration rates are insensitive to water/cement ratio. The simulations indicate that the product layer on C₃S remains permeable; no transition to diffusion control is indicated, even long after the rate peak.     

Early hydration of tricalcium silicate I. Kinetics of the hydration process and the stoichiometry of the hydration products

The initial hydration of C₃S in paste form at room temperature was studied. The process is initiated by a short lasting rapid hydration in which about 1 – 2% of C₃S is hydrated and a hydrate with low C/S and high H/S ratio is formed. After a subsequent induction period of 4 hours a renewed rapid hydration is observed in which a hydrate of constant stoichiometric composition, independent on the time of hydration is formed. This hydrate has a higher C/S and lower H/S ratio than the one formed initially. The liquid phase stays supersaturated with respect to calcium hydroxide for several hours after the induction period is terminated.     

Analyses of the aqueous phase during early C3S hydration

The concentrations of calcium and silica in solution during the first 4 hours of C₃S hydration were measured. The results of these analyses indicate that a solid calcium silicate hydrate forms within 30 seconds of the start of hydration and that an equilibrium between the solution and the solid hydrate is rapidly established. A strong dependence of the rate of early hydration on the w:C₃S ratio was observed, while the dependence on the surface area of the C₃S was minimal.     

The Effect of Magnesium and Zinc Ions on the Hydration Kinetics of C3S

Impure tricalcium silicate (C₃S) in portland cement may contain various foreign ions. These ions can stabilize different polymorphs of C₃S at room temperature and may affect its reactivity. In this paper, the effects of magnesium and zinc on the polymorph type, hydration kinetics, and the hydrate morphology of C₃S were investigated. The pure C₃S has the T1 structure while magnesium and zinc stabilize polymorphs M3 and T2/T3, respectively. The two elements have distinct effects on the hydration kinetics. Zinc increases the maximum heat released. Magnesium increases the hydration peak width. The C–S–H morphology is modified, leading to longer needles in the presence of zinc and thicker needles in the presence of magnesium. Zinc is incorporated into C–S–H, while magnesium is only incorporated slightly, if at all, but rather seems to inhibit nucleation. Implementing experimentally measured parameters for C–S–H nucleation and growth in the μic hydration model captured well the observed changes in hydration kinetics. This supports C–S–H nucleation and growth to be rate controlling in the hydration of C₃S.     

Implications for C3S kinetics from combined C3S/CA hydration

This study was performed to understand the influence of a critical amount of CA on the kinetics of C₃S hydration. For this purpose, monoclinic C₃S was blended with 15 wt.% CA and investigated by heat flow calorimetry and in situ XRD at 23°C at a water to cement ratio of 0.5. The binary mixture shows 3 distinct heat flow maxima where the underlying C₃S dissolution is proceeding stepwise. The C₃S dissolution rates during the 3 steps are varying strongly, depending on the hydrate phase precipitated during the respective reaction step. Comparison of these dissolution rates with a pure C₃S reference sample allows the conclusion that the dissolution rate of pure C₃S after the heat flow maximum might be governed by either the remaining available reactive surface of C₃S or a diffusion‐controlled process, which would both be influenced by the respective hydrate phase precipitating on the C₃S surface.     

Hydration of calcium aluminates at 60°C – Development paths of C2AHx in dependence on the content of free water

The influence of water loss during the hydration of calcium aluminates on the phase development is investigated at 60°C. This is relevant for applications in which calcium aluminate cement (CAC) based formulations are exposed to quick drying during hydration. The presented results provide new insights into the well-known conversion processes occurring in CAC pastes. Using in situ XRD two different routes of the development of initially formed C₂AH₈ are determined: (a) transformation to C₃AH₆ + AH₃ in the presence of sufficient free water and (b) dehydration to C₂AH₅ at a lack of free water. Moreover, the influence of precuring of the pastes at 23°C before heating to 60°C is investigated. The increasing loss of free water with increasing precuring time resulting from both, precipitation of hydrate phases and evaporation, causes incomplete hydration of CA or CA₂ as well as dehydration of C₂AH₈ instead of conversion into C₃AH₆. Comparative investigations of sealed samples always revealed complete hydration of CA and CA₂ as well as complete conversion of C₂AH₈.     

The Influence of Water Activity on the Hydration Rate of Tricalcium Silicate

Tricalcium silicate does not undergo hydration at relative humidities (RH's) below 80%. But, the rate at which its hydration rate decreases as a function of the RH has not yet been elucidated. By invoking correspondence between RH and water activity (a_H, unitless), both of which are related to the chemical potential of water, the reaction evolutions of triclinic tricalcium silicate (i.e., T1‐Ca₃SiO₅ or C₃S) are tracked in water + isopropanol (IPA) mixtures, prepared across a wide range of water activities. Emphasis is placed on quantifying the: (a) rate of hydration as a function of a_H, and (b) the critical (initial, a_H0c or the achieved) water activity at which hydration effectively ceases, i.e., does not progress; here identified to be ≈ 0.70. The hydration of tricalcium silicate is arrested even when the system remains near saturated with a liquid phase, such that small, if any, capillary stresses develop. This suggests that changes in chemical potential induced via a vapor‐phase or liquid‐phase route both induce similar suppressions of C₃S hydration. A phase boundary nucleation and growth (pBNG) model is fit to measured hydration rates from the onset of the acceleration period until well beyond the rate maximum when the water activity is altered. The simulations suggest that for a fixed hydrate nucleation density, any water activity reductions consistently suppress the growth of hydration products. Thermodynamic considerations of how water activity changes may influence reactions/hydrate evolutions are discussed. The outcomes improve our understanding of the chemical factors that influence the rate of Ca₃SiO₅ hydration.     

Action of triethanolamine on the hydration of tricalcium aluminate

Hydration characteristics of tricalcium aluminate and tricalcium aluminate + gypsum were studied following addition of 0.5, 1.0, 5.0 or 10.0% triethanolamine (TEA) at a solution/C₃A ratio of 1.0 after hydration periods of 1 to 60 min. TEA accelerated the hydration of C₃A to the hexagonal aluminate hydrate and its conversion to the cubic aluminate hydrate. The rate of hydration increased with increased amounts of TEA, which also accelerated the formation of ettringite in the C₃A-gypsum-H₂O system.     

The influence of an ion-exchange resin on the kinetics of hydration of tricalcium silicate

The addition of a finely-ground ion-exchange resin makes it possible to modify the hydration kinetics of C₃S pastes. Analyses of the liquid phase in pastes and more dilute suspensions show that the resin exchanges calcium ions for sodium ions very rapidly during the early stage of hydration and therefore the concentration of silica in solution increases. The resin impacts the hydration of C₃S by other mechanisms which depends on the resin quantity added. For a high resin quantity, the induction period is very short, but the longer-term hydration is enhanced compared to a reference sample without resin. We hypothesize that the surface of the resin can provide sites for the nucleation and growth of C-S-H hydrates and/or portlandite far away from the surface of the C₃S grains. This consequently increases the quantity of hydrates that can precipitate before a continuous hydrate layer forms over the surfaces of C₃S particles.     

The hydration characteristics of MSWI fly ash slag present in C3S

This paper reports on an investigation of the hydration characteristics of C₃S and the mixing of C₃S with municipal solid waste incinerator (MSWI) fly ash slag. The results can be summarized as follows: TGA observations show lower amounts of CSH and Ca(OH)₂ in samples that incorporated MSWI slag into C₃S, possibly due to the partial replacement of the C₃S by slag with less activity. In general, the incorporation of slag into C₃S decreases the initial hydration reactions, but it increases the pozzolanic reactions at a later stage by consuming Ca(OH)₂. As the hydration precedes, the C₃S peaks decrease, up to 90 days, due to the activation of the slag by liberated Ca(OH)₂. As well, the amount of hydration products (Ca(OH)₂) from the pure C₃S pastes are more than for the C₃S-slag pastes. Moreover, the degree of hydration of the C₃S pastes increases with the curing time, the C₃S-slag paste also shows that lower hydration degree values at all ages.     

Influence of pozzolanic additives on hydration mechanisms of tricalcium silicate

Pozzolanic mineral additives, such as silica fume (SF) and metakaolin (MK), are used to partially replace cement in concrete. This study employs extensive experimentation and simulations to elucidate and contrast the influence of SF and MK on the early age hydration rates of tricalcium silicate (triclinic C₃S), the major phase in cement. Results show that at low replacement levels (i.e., ≤10%), both SF and MK accelerate C₃S hydration rates via the filler effect, that is, enhanced heterogeneous nucleation of the main hydration product (C–S–H: calcium‐silicate‐hydrate) on the extra surfaces provided by the additive. The filler effect of SF is inferior to that of MK because of agglomeration of the fine particles of SF, which causes significant reduction (i.e., up to 97%) in its surface area. At higher replacement levels (i.e., ≥20%), while SF continues to serve as a filler, the propensity of MK to allow nucleation of C–S–H on its surface is substantially suppressed. This reversal in the filler effect of MK is attributed to the abundance of aluminate [Al(OH)₄^?] ions in the solution—released from the dissolution of MK—which inhibit topographical sites for C–S–H nucleation and impede its subsequent growth. Results also show that in the first 24 hours of hydration, MK is a superior pozzolan compared to SF. However, the pozzolanic activities of both SF and MK are limited and, thus, do not produce significant alterations in the early age hydration kinetics of C₃S. Overall, the outcomes of this study provide novel insights into the mechanistic origins of the filler and pozzolanic effects of SF and MK, and their impact on cementitious reaction rates.     

Hydration of C3S in presence of CA: Mineral-pore solution interaction

C₃S and CA are the main phases of OPC and Fe-rich CAC, respectively. The objective of this research was to investigate the influence of CA on C₃S hydration, representing an under sulfated OPC-rich binder, and to shed light on the underlying hydration mechanisms. To this end, C₃S was blended with 1-30 wt-% CA and the pastes (w/c 0.5) were investigated by heat flow calorimetry, in situ X-ray diffraction and analysis of the pore solution chemistry. CA additions ≥5 wt-% reveal a separation into three distinct heat flow maxima, whereas additions ≤3 wt-% just retard the start of the main reaction. The silicate reaction (dissolution of C₃S and precipitation of C–S–H with or without CH) can be retarded for 4 to ≥22 hours in comparison to pure C₃S depending on the admixed CA content. The start of the silicate reaction seems to be related to a decrease in Al- and increase in Ca-concentration in the pore solution. However, it can be shown in this study that C₃S is able to dissolve even at high Al concentrations in the pore solution.     

Etude de l'effet retardateur du zinc sur l'hydratation de la pate de ciment Portland

The retarding effect of zinc on the hydration of C₃S and C₃A, the two principal Portland cement components, has been investigated by X - ray diffraction. The results show that the C₃S retardation is more important than that of C₃A. This retardation is due to the precipitation of an amorphous layer of zinc hydroxide around the anhydrous grains. The effect of this coating depends on its permeability. The hydration reaction starts again through the transformation of the zinc hydroxide into the crystalline calcium zinc hydroxide Ca Zn₂ (OH)₆, 2H₂ O.     

Hydration of cubic tricalcium aluminate in the presence of calcined clays

Calcined clay blended cements play a major role in cement industry's strategy to reduce CO₂ emissions. During their hydration, an accelerated aluminate reaction is often observed to affect the sulfate balance. The objective of this study was to provide insights into the influence of different calcined clays on the hydration of cubic tricalcium aluminate (C₃A). A cementitious model system consisting of cubic C₃A, quartz powder, calcium sulfate and a model pore solution was investigated. The influence of three different calcined clays and one nanolimestone was examined by a successive replacement of the quartz powder and variation of the calcium sulfate. Heat flow and hydrate phase development were followed by isothermal calorimetry and quantitative in-situ X-ray diffraction. Accelerated ettringite formation and sulfate depletion were observed for all calcined clays, while the nanolimestone exhibited the opposite effect. It was found that adsorption of SO₄ ions and/or Ca-SO₄-complexes at the surface of calcined clay particles is the main factor inhibiting retardation of the C₃A hydration in absence of a silicate reaction. In the Al-rich systems a retardation through sulfate adjustment seems to be impeded by additional Al ions, which react with Ca adsorbed onto and leached from the C₃A surface.     

Effect of the densification of C–S–H on hydration kinetics of tricalcium silicate

Effect of water to cement (w/c) ratio and temperature profiles on the densification of C–S–H (calcium silicate hydrate gel) and hydration kinetics of triclinic tricalcium silicate (C₃S) is studied beyond the first day of hydration. Calorimetry and quantitative X‐ray diffraction/Rietveld analysis show that degree of hydration is unaffected by w/c up to 7 days and marginally thereafter. Coupling the degree of hydration with the portlandite content measured from thermal analysis indicate that C/S ratio of C–S–H decreases with increasing w/c. There is a clear increase in the portlandite content with increasing w/c, even though the degree of hydration is unchanged, due to the variations in C/S ratio of C–S–H. On the other hand, when C₃S is initially cured at a lower temperature (20°C) and then at a higher temperature (40°C), there is a significant increase in the reactivity even until 28 days and vice versa. These experimental results were explained using the densified volumetric growth hypothesis, which assumes that hydration kinetics are dependent on the internal surface area of C–S–H.     

Effect of Temperature on C3S and C3S + Nanosilica Hydration and C–S–H Structure

This study aimed to monitor the effect of temperature and the addition of nanosilica on the nanostructure of the C–S–H gel forming during tricalcium silicate (C₃S) hydration. Two types of paste were prepared from a synthesized T₁ C₃S. The first consisted of a blend of deionized water and C₃S at a water/solid ratio of 0.425. In the second, a 90 wt% C₃S + 10 wt% of nanosilica blend was mixed with water at a water/solid ratio of 0.7. The pastes were stored in closed containers at 100% RH and 25°C, 40°C, or 65°C. The hydration reaction was detained after 1, 14, 28, or 62 d with acetone, and then pastes were studied by ²⁹Si magic angle spinning nuclear magnetic resonance (²⁹Si MAS NMR).The main conclusion was that adding nSA expedites C₃S hydration at any age or temperature and modifies the structure of the C–S–H gel formed, two types of C–S–H gel appear. At 25°C and 40°C, more orderly, longer chain gels are initially (1 d) obtained as a result of the pozzolanic reaction between nSA and portlandite (CH) (C–S–H_II gel formation). Subsequently, ongoing C₃S hydration and the concomitant flow of dimers shorten the mean chain length in the gel.     

Low pressure steam hydration of tricalcium silicate

Pastes of C₃S (w/c ratio = 0.5) were steam cured at 25, 40, 60 and 90°C for 1 hour to 30 days. The results obtained have shown that, as the curing temperature rises, the induction period is shortened and the initial rate of hydration of C₃S is increased; at longer curings, on the other hand, such hydration rate is considerably lowered. In order to explain the influence of temperature on the hydration reaction a new hypothesis has been proposed, which takes into account the C/S molar ratio as well as the surface properties of the hydrated silicate.     

Effect of particle size distribution of metakaolin on hydration kinetics of tricalcium silicate

The focus of this study is to elucidate the role of particle size distribution (PSD) of metakaolin (MK) on hydration kinetics of tricalcium silicate (C₃S–T₁) pastes. Investigations were carried out utilizing both physical experiments and phase boundary nucleation and growth (pBNG) simulations. [C₃S + MK] pastes, prepared using 8%_mass or 30%_mass MK, were investigated. Three different PSDs of MK were used: fine MK, with particulate sizes <20 µm; intermediate MK, with particulate sizes between 20 and 32 µm; and coarse MK, with particulate sizes >32 µm. Results show that the correlation between specific surface area (SSA) of MK's particulates and the consequent alteration in hydration behavior of C₃S in first 72 hours is nonlinear and nonmonotonic. At low replacement of C₃S (ie, at 8% _mass), fine MK, and, to some extent, coarse MK act as fillers, and facilitate additional nucleation and growth of calcium silicate hydrate (C–S–H). When C₃S replacement increases to 30% _mass, the filler effects of both fine and coarse MK are reversed, leading to suppression of C–S–H nucleation and growth. Such reversal of filler effect is also observed in the case of intermediate MK; but unlike the other PSDs, the intermediate MK shows reversal at both low and high replacement levels. This is due to the ability of intermediate MK to dissolve rapidly—with faster kinetics compared to both coarse and fine MK—which results in faster release of aluminate [Al(OH)₄⁻] ions in the solution. The aluminate ions adsorb onto C₃S and MK particulates and suppress C₃S hydration by blocking C₃S dissolution sites and C–S–H nucleation sites on the substrates’ surfaces and suppressing the post-nucleation growth of C–S–H. Overall, the results suggest that grinding-based enhancement in SSA of MK particulates does not necessarily enhance early-age hydration of C₃S.