Hydration of cementitious materials, present and future

This paper is a keynote presentation from the 13th International Congress on the Chemistry of Cement. It discusses the underlying principles of hydration and recent evidence for the mechanisms governing this process in both Portland cements and other cementitious materials. Given the overriding imperative to improve the sustainability of cementitious materials, routes to reducing CO₂ emissions are discussed and the impact of supplementary materials on hydration considered.     

Recent advances in modeling for cementitious materials

Since the 12th International Congress on the Chemistry of Cement (Montreal, 2007) there has been an intensive research activity devoted to describe the cementitious materials by computational means. This review aims to summarize the collective effort performed on this topic during the last 4 years, and highlights the most relevant results.     

Resistivity of cementitious materials measured in diaphragm migration cells: The effect of the experimental set-up

Experimental measurements and numerical analysis were carried out to study the effect of the cell geometry in resistivity determinations. The resistance of the diaphragm in cement paste and mortar samples was determined using impedance spectroscopy. Numerical simulations were performed using finite element method (FEM). Several surface ratios (geometrical diaphragm surface to electrolyte-diaphragm surface, S/S′) were investigated. The thickness of the diaphragm, L, was also considered.The experimental results show a significant decrease of the apparent resistivity when the ratio S/S′ increases. Similar trend was observed for increasing values of the L/S′ ratio. The numerical simulations can explain the experimental findings and also allow to formulate a general rule for the design of migration and diffusion experiments in porous materials.     

Desiccation shrinkage of cementitious materials as an aging, poroviscoelastic response

This paper describes and evaluates a new model that utilizes aging poroviscoelasticity for predicting the shrinkage of cementitious materials induced by loss of moisture from the pore structure (i.e. desiccation). The new model incorporates well-accepted mechanisms for desiccation shrinkage and accounts for the effect of changing concentrations of dissolved species in the pore fluid. Additionally, the model is used to interpret viscoelastic behavior during the drying process via comparisons of model predictions with measured shrinkage of hardened portland cement paste. It was found that while a poroelastic model under predicts the measured shrinkage, the poroviscoelastic model significantly over predicts the shrinkage unless intrinsic aging of the C–S–H gel is included in the model.     

Assessment of the long-term stability of cementitious barriers of radioactive waste repositories by using digital-image-based microstructure generation and reactive transport modelling

Cement-based grout plays a significant role in the design and performance of nuclear waste repositories: used correctly, it can enhance their safety. However, the high water-to-binder ratios, which are required to meet the desired workability and injection ability at early age, lead to high porosity that may affect the durability of this material and undermine its long-term geochemical performance.In this paper, a new methodology is presented in order to help the process of mix design which best meets the compromise between these two conflicting requirements. It involves the combined use of the computer programs CEMHYD3D for the generation of digital-image-based microstructures and CrunchFlow, for the reactive transport calculations affecting the materials so simulated. This approach is exemplified with two grout types, namely, the so-called Standard mix 5/5, used in the upper parts of the structure, and the “low-pH” P308B, to be injected at higher depths.The results of the digital reconstruction of the mineralogical composition of the hardened paste are entirely logical, as the microstructures display high degrees of hydration, large porosities and low or nil contents of aluminium compounds.Diffusion of solutes in the pore solution was considered to be the dominant transport process. A single scenario was studied for both mix designs and their performances were compared. The reactive transport model adequately reproduces the process of decalcification of the C-S-H and the precipitation of calcite, which is corroborated by empirical observations. It was found that the evolution of the deterioration process is sensitive to the chemical composition of groundwater, its effects being more severe when grout is set under continuous exposure to poorly mineralized groundwater. Results obtained appear to indicate that a correct conceptualization of the problem was accomplished and support the assumption that, in absence of more reliable empirical data, it might constitute a useful tool to estimate the durability of cement-based structures.     

Development of low-pH cementitious materials for HLRW repositories: Resistance against ground waters aggression

One of the most accepted engineering construction concepts of underground repositories for high radioactive waste considers the use of low-pH cementitious materials. This paper deals with the design of those based on Ordinary Portland Cements with high contents of silica fume and/or fly ashes that modify most of the concrete “standard” properties, the pore fluid composition and the microstructure of the hydrated products. Their resistance to long-term groundwater aggression is also evaluated. The results show that the use of OPC cement binders with high silica content produces low-pH pore waters and the microstructure of these cement pastes is different from the conventional OPC ones, generating C-S-H gels with lower CaO/SiO₂ ratios that possibly bind alkali ions. Leaching tests show a good resistance of low-pH concretes against groundwater aggression although an altered front can be observed.     

Supplementary cementitious materials for mitigating degradation of kraft pulp fiber-cement composites

Kraft pulp fiber reinforced cement-based materials are being increasingly used where performance after exposure to environmental conditions must be ensured. However, significant losses in mechanical performance due to wet/dry cycling have been observed in these composites, when portland cement is the only cementitious material used in the matrix. In this research program, the effects of partial portland cement replacement with various supplementary cementitious materials were investigated. Binary, ternary, and quaternary blends of silica fume, slag, Class C fly ash, Class F fly ash, metakaolin, and diatomaceous earth/volcanic ash blends were examined for their effect on the degradation of kraft pulp fiber-cement composite mechanical properties (i.e., strength and toughness) during wet/dry cycling. After 25 wet/dry cycles, it was shown that binary composites containing 90% slag, 30% metakaolin, or greater than 30% silica fume did not exhibit any signs of degradation, as measured through mechanical testing and microscopy. Ternary blends containing 70% slag/10% metakaolin or 70% slag/10% silica fume were also effective in preventing degradation. A reduction in calcium hydroxide content and the stability of the alkali content due to supplementary cementitious material addition were shown to be primary mechanisms for improved durability.     

Investigation of the carbonation front shape on cementitious materials: Effects of the chemical kinetics

Carbonation depth-profiles have been determined by thermogravimetric analysis and by gammadensitometry after accelerated carbonation tests on ordinary Portland cement (OPC) pastes and concretes. These methods support the idea that carbonation does not exhibit a sharp reaction front. From analytical modelling, this feature is explained by the fact that the kinetics of the chemical reactions become the rate-controlling processes, rather than the diffusion of CO₂. Furthermore, conclusions are drawn as to the mechanism by which carbonation of Ca(OH)₂ and C-S-H takes place. Carbonation gives rise to almost complete disappearance of C-S-H gel, while Ca(OH)₂ remains in appreciable amount. This may be associated with the CaCO₃ precipitation, forming a dense coating around partially reacted Ca(OH)₂ crystals. The way in which CO₂ is fixed in carbonated samples is studied. The results indicate that CO₂ is chemically bound as CaCO₃, which precipitates in various forms, namely: stable, metastable, and amorphous. It seems that the thermal stability of the produced CaCO₃ is lower when the carbonation level is high. It is also shown that the poorly crystallized and thermally unstable forms of CaCO₃ are preferentially associated with C-S-H carbonation.     

Modelling and simulations of the chemo-mechanical behaviour of leached cement-based materials: Interactions between damage and leaching

The assessment of the durability of cement-based materials, which could be employed in underground structures for nuclear waste disposal, requires accounting for deterioration factors, such as chemical attacks and damage, and for the interactions between these phenomena. The objective of the present paper consists in investigating the long-term behaviour of cementitious materials by simulating their response to chemical and mechanical solicitations. In a companion paper (Stora et al., submitted to Cem. Concr. Res. 2008), the implementation of a multi-scale homogenization model into an integration platform has allowed for evaluating the evolution of the mineral composition, diffusive and elastic properties inside a concrete material subjected to leaching. To complete this previous work, an orthotropic micromechanical damage model is presently developed and incorporated in this numerical platform to estimate the mechanical and diffusive properties of damaged cement-based materials. Simulations of the chemo-mechanical behaviour of leached cementitious materials are performed with the tool thus obtained and compared with available experiments. The numerical results are insightful about the interactions between damage and chemical deteriorations.     

A method based on isothermal calorimetry to quantify the influence of moisture on the hydration rate of young cement pastes

Cement hydration needs water to proceed and if water is lost by drying, the hydration rate will decrease. This can be of importance in cases when concrete surfaces are exposed to drying so that their strength development will be retarded. We describe a method based on isothermal calorimetry to assess how the rate of cement hydration is influenced by removal of water (drying) at different times up to three days after mixing. Thin samples of cement pastes are hydrated in a calorimeter and at different times exposed to one hour drying periods. The resulting decrease in thermal power following the removal of water is quantified as a measure of the reduction in hydration rate. The mass loss is found by weighing the samples before and after a measurement, and the change in water activity of a sample during drying can be found from the slope of the thermal power during the drying period.     

Evolution of penetration resistance in fresh concrete

The objective of this research was to examine the setting of concrete through its penetration resistance; an experimental device, especially developed for this purpose was used, which consists of a system that lets a sphere fall on concrete from a certain height and then measures the depth of the crater. Forty-five samples were made with four different types of cement varying its quantity, water-cement ratio, at ambient temperature and humidity. A semiempirical model which explains the penetration resistance evolution in fresh concrete was proposed and experimentally demonstrated. The results are compared with the consolidation curves for soils represented by the logarithmic method. The proposed setting time was defined as the elapsed time between the placement of the concrete and the time when the depth of the crater is 18% of the initial one.     

Photocatalytic degradation of air pollutants — From modeling to large scale application

Indoor as well as outdoor air quality and their limiting values remain a major problem to our present-day society.This paper addresses the modeling of the decomposition process of nitrogen monoxide (NO) on reactive concrete surfaces under the controlled exposition of a UV source. Within this model the external mass transfer of the pollutant and the internal molecule diffusion-reaction were considered. A first-order kinetics equation is derived with respect to the NO concentration and a site-competitive adsorption between NO/NO₂ and water molecules, resulting in a dependence of the reaction kinetics on the relative humidity. Using the proposed model, a reaction rate constant k and an adsorption equilibrium constant K_d can be derived which describe an active paving stone accurately. Experimental results from a self-developed photoreactor with continuous flow mode were used to validate the proposed kinetic expression. Furthermore, the effect of variations in process conditions such as irradiance and relative humidity on the two constants k and K_d is investigated. All modeling work provides a sound foundation for the translation of this process to real outside conditions. In this regard an upcoming project in a Dutch city is described in brief.     

Modelling of ageing effects on crack-bridging behaviour of AR-glass multifilament yarns embedded in cement-based matrix

This article focus on modelling of ageing effects on crack-bridging behaviour of AR-glass multifilament yarns embedded in cement-based matrix. In the first step, age-dependent changes in the crack-bridging behaviour of AR-glass multifilament yarns were investigated at the meso and micro levels. Two cementitious matrices were considered where the binder contained Portland cement clinker and ground granulated blast furnace slag cement, respectively. Mechanical characteristics of the bond between matrix and multifilament yarns after accelerated ageing were measured by means of double-sided yarn pullout tests. In these tests the multifilament yarns bridged a single crack in the matrix arising in a notched area of the specimen. Losses in performance with increasing age differed widely depending on matrix material composition. The essential cause of such losses was discovered to be the microscopic densification of the fibre-to-matrix interface. This led to increased bond intensity and restricted slip-ability of the filaments. Subsequently, these micro-structural phenomena were related to the mesoscopic material behaviour by means of a phenomenological bond model. This cross-linkage model describes the crack-bridging effect of the entire multifilament yarn at the single filament level. According to the model, each filament possesses a specific deformation length depending on its position in the cross-section of the yarn. This deformation length depends on bond characteristics between single filament and cementitious matrix, which vary with age. Characteristic values of the model were computed from load-crack width curves obtained from the yarn pullout tests. The changes in the microstructure were represented by the characteristic values of the model.     

International Summit on Cement Hydration Kinetics and Modeling Special Edition Québec City, July 27, 28 and 29, 2009

The performance of portland cement concrete relies upon a series of complex events that begin with raw minerals and end many years after the concrete is placed. Between these points, the life of this dynamic material is dominated by chemical reactions called hydration. While much is known about hydration, unfortunately, there is no unifying theory that describes the kinetics (rates) of these complex transformations from anhydrous cement to hydrous cement paste. Other industries including metalugy, petrochemicals, pharmaceuticals and semiconductors have asserted process control by developing a fundamental, mechanistic understanding of the kinetics of the chemical reactions and phase transformations that define their products. Might the concrete industry be moving along a similar trajectory?     

Influence of the cementitious paste composition on the E-modulus and heat of hydration evolutions

E-modulus and heat of hydration are features of cement-based materials that follow a rapid rate of change at early ages. This paper analyses the influence of the composition of cementitious pastes on these features by using two methods: (i) a novel technique for continuously monitoring the E-modulus of cement-based materials, based on evaluating the first resonant frequency of a composite beam containing the material under testing, and (ii) an isothermal calorimeter to determine the released heat of hydration. Seventeen mixes are tested, encompassing pastes with five w/c ratios, as well as different contents of limestone filler, fly ash, silica fume and metakaolin. The results permit the comparison of the E-modulus and heat of hydration sensitivities to mix composition changes, and to check possible relations between these features. This work also helps to establish the technique (i) as a non-destructive method for monitoring the E-modulus evolution in cement-based materials since casting.     

Modeling and simulation of cement hydration kinetics and microstructure development

Efforts to model and simulate the highly complex cement hydration process over the past 40 years are reviewed, covering different modeling approaches such as single particle models, mathematical nucleation and growth models, and vector and lattice-based approaches to simulating microstructure development. Particular attention is given to promising developments that have taken place in the past few years. Recent applications of molecular-scale simulation methods to understanding the structure and formation of calcium–silicate–hydrate phases, and to understanding the process of dissolution of cement minerals in water are also discussed, as these topics are highly relevant to the future development of more complete and fundamental hydration models.