publications: articles

Link

Set in Stone: Transforming Concrete into a Sustainable Infrastructure Material

Link

From electrons to infrastructure: Engineering concrete from the bottom up

An approach rooted in fundamental, mechanistic models of concrete materials offers the only viable path for handling the enormous number of variables that are being introduced as new materials are added to the design space, and as new properties are mandated for a sustainable infrastructure. These models must begin at the smallest length scales relevant for concrete properties; in some cases this is the scale of electron interactions among atoms and ions. But concrete has complex chemical and structural properties that are manifested at greater length and time scales, so atomic scale models must ultimately be integrated with new models that capture behavior at mesoscopic and macroscopic scales. We refer to this methodology as the "bottom-up" approach because it proceeds from the smallest length scales. We describe this kind of modeling approach, include some recent results, and suggest some principles for collaboratively integrating multi-scale models.

Link

Understanding and Controlling the Reactivity of the Calcium Silicate phases from First Principles

First principles calculations are employed to provide a fundamental understanding of the relationship between the reactivity of synthetic calcium silicate phases and their electronic structure. Our aim is to shed light on the wide range of hydration kinetics observed in different phases of calcium silicate. For example, while the dicalcium silicate (Ca2SiO4) phase slowly reacts with water, the tricalcium silicate (Ca3SiO5) shows much faster hydration kinetics. We show that the high reactivity of Ca3SiO5 is mainly related to the reactive sites around its more ionic oxygen atoms. Ca2SiO4 does not contain these types of oxygen atoms, although experiments suggest that impurities may play a role in changing the reactivity of these materials. We analyze the electronic structure of a wide range of possible substitutions in both Ca3SiO5 and Ca2SiO4 and show that while the influence of different types of impurities on structural properties is similar, their effect on reactivity is very different. Our calculations suggest that the variation of electronic structure is mainly related to the formation of new hybridized orbitals and the charge exchange between the impurity atoms and the bulk material. The charge localization upon introducing impurities is quantified to predict candidate substitutions that could increase the reactivity of Ca2SiO4, which would broaden the applicability of this lower temperature and thus less costly and energetically less demanding phase.

Link

Hydration of Calcium Oxide Surface Predicted by Reactive Force Field Molecular Dynamics

In this work, we present the parametrization of Ca−O/H interactions within the reactive force field ReaxFF, and its application to study the hydration of calcium oxide surface. The force field has been fitted using density functional theory calculations on gas phase calcium−water clusters, calcium oxide bulk and surface properties, calcium hydroxide, bcc and fcc Ca, and proton transfer reactions in the presence of calcium. Then, the reactive force field has been used to study the hydration of the calcium oxide {001} surface with different water contents. Calcium oxide is used as a catalyzer in many applications such as CO2 sequestration and biodiesel production, and the degree of surface hydroxylation is a key factor in its catalytic performance. The results show that the water dissociates very fast on CaO {001} bare surfaces without any defect or vacancy. The surface structure is maintained up to a certain amount of water, after which the surface undergoes a structural rearrangement, becoming a disordered calcium hydroxyl layer. This transformation is the most probable reason for the CaO catalytic activity decrease.

Link

Impact of Chemical Impurities on the Crystalline Cement Clinker Phases Determined by Atomistic Simulations

The presence of chemical substitutions is believed to play a crucial role in the hydration reactions, structure, and elastic properties of cement clinker phases. Hence, substitutions are of great technological interest, as more efficient production of cement clinkers would result in a reduction of CO2 emissions, as well as possible economic benefits. Here we use a combination of classical and quantum mechanical simulation methods to study the detailed physicochemical changes of the clinker phases alite (Ca3SiO5) and belite (Ca2SiO4) when Mg2+, Al3+ and Fe3+ guest ions are incorporated into their structure. Using classical force field methods, we considered random substitutions among possible sites and different compositions in order to identify the preferential substitution sites on the crystalline structures. Then, the resulting structural changes that take place to accommodate the guest ions are investigated and discussed in detail. Using quantum mechanical density functional theory calculations the electronic structure of representative configurations has been computed to determine the potential impact of impurities on the reactivity.

Link

Mechanisms of cement hydration

The current state of knowledge of cement hydration mechanisms is reviewed, including the origin of the period of slow reaction in alite and cement, the nature of the acceleration period, the role of calcium sulfate in modifying the reaction rate of tricalcium aluminate, the interactions of silicates and aluminates, and the kinetics of the deceleration period. In addition, several remaining controversies or gaps in understanding are identified, such as the nature and influence on kinetics of an early surface hydrate, the mechanistic origin of the beginning of the acceleration period, the manner in which microscopic growth processes lead to the characteristic morphologies of hydration products at larger length scales, and the role played by diffusion in the deceleration period. The review concludes with some perspectives on research needs for the future.

Link

Influence of nucleation seeding on the hydration kinetics and compressive strength of alkali activated slag paste

Addition of pure calcium silicate hydrate (C–S–H) to alkali-activated slag (AAS) paste resulted in an earlier and larger hydration rate peak measured with isothermal calorimetry and a much higher compressive strength after 1 d of curing. This is attributed to a nucleation seeding effect, as was previously established for Portland cement and tricalcium silicate pastes. The acceleration of AAS hydration by seeding indicates that the early hydration rate is controlled by nucleation and growth. For the experiments reported here, the effect of C–S–H seed on the strength development of AAS paste between 1 d and 14 d of curing depended strongly on the curing method. With sealed curing the strength continued to increase, but with underwater curing the strength decreased due to cracking. This cracking is attributed to differential stresses arising from chemical and autogenous shrinkage. Similar experiments were also performed on Portland cement paste.

Link

Hydration Kinetics and Microstructure Development of Normal and CaCl2-Accelerated Tricalcium Silicate Pastes

Microstructure development and the kinetics of hydration of pure tricalcium silicate (C3S) and CaCl2-accelerated C3S pastes were investigated by performing isothermal calorimetry and in situ small-angle neutron scattering (SANS) measurements on parallel specimens during the first few days of hydration, as well as on 28-d old specimens hydrated under the same curing conditions (water:cement ratio = 0.5, 20 C). Calorimetry experiments were also performed over a range of hydration temperatures from 10 C to 40 C. The calorimetry data were analyzed by applying a previously described boundary nucleation and growth model. CaCl2 significantly increases the rate of nucleation of hydration product on the surface of the C3S particles, but has relatively little effect on the product growth rate. The SANS measurements indicate important differences in the nanostructural development associated with CaCl2 acceleration, although the composition and density of the calcium silicate hydrate (C S H) nanoparticles are unchanged. In the CaCl2-accelerated paste the surface fractal scattering associated with the deposition of hydration product onto the initially smooth surfaces of the C3S particles rapidly declines in intensity and essentially disappears by the age of 1 d, while in the pure C3S paste the surface fractal scattering remains prominent throughout the hydration process. The key observations from both the calorimetry and SANS analysis can be explained if in a pure C3S paste the C S H hydration product forms initially with a low packing density and then densifies with time.

Link

Nanostructure and Nanomechanics of Cement: Polydisperse Colloidal Packing

Cement setting and cohesion are governed by the precipitation and growth of calcium-silicate-hydrate, through a complex evolution of microstructure. A colloidal model to describe nucleation, packing, and rigidity of calcium-silicate-hydrate aggregates is proposed. Polydispersity and particle size dependent cohesion strength combine to produce a spectrum of packing fractions and of corresponding elastic properties that can be tested against nanoindentation experiments. Implications regarding plastic deformations and reconciling current structural characterizations are discussed.

Link

Nanoscale Elastic Properties of Montmorillonite upon Water Adsorption

Smectites are an important group of clay minerals that experience swelling upon water adsorption. This paper uses molecular dynamics with the CLAYFF force field to simulate isothermal isobaric water adsorption of interlayer Wyoming Na-montmorillonite, a member of the smectite group. Nanoscale elastic properties of the clay–interlayer water system are calculated from the potential energy of the model system. The transverse isotropic symmetry of the elastic constant matrix was assessed by calculating Euclidean and Riemannian distance metrics. Simulated elastic constants of the clay mineral are compared with available results from acoustic and nanoindentation measurements.

Link

Development of Frangible Concrete to Reduce Blast-Related Casualties

A new high-performance material known as frangible concrete was developed with the goal of eliminating the potential for lethal-sized fragments of broken concrete when a structure is impacted by a dynamic load, such as the blast force from a terrorist vehicle bomb. Based on a strategy of developing concrete with moderate, quasi-static, load-bearing properties and high frangibility under dynamic loading conditions, the investigation made use of the shrinkage and fracture properties of cement and slag binders, along with the proper choice of aggregate and aggregate gradations. The development process included designed experiments to optimize mixture proportions against compressive strength and laboratory-scale impact tests followed by the manufacture of concrete masonry units (CMUs) for testing a wall under full-scale field blast conditions.

Link

Adsorption Induced Deformation of Microporous Materials

Carbon dioxide injection in coal seams is known to improve the methane production of the coal seam, while ensuring a safe and long-term carbon sequestration. This improvement is due to the preferential adsorption of CO2 in coal with respect to CH4: an injection of CO2 thus results in a desorption of CH4. However, this preferential adsorption is also known to cause a differential swelling of coal, which results in a significant decrease in the reservoir permeability during the injection process. Recent studies have shown that adsorption in coal micropores (few angströms in size) is the main cause of the swelling. In this work, we focus on the competitive adsorption behavior of CO2 and CH4 in micropores. We perform molecular simulations of adsorption with a realistic atomistic model for coal. The competitive adsorption is studied at various temperatures and pressures representative of those in geological reservoirs. With the help of a poromechanical model, we then quantify the subsequent differential swelling induced by the computed adsorption behaviors. The differential swelling is almost insensitive to the geological temperatures and pressures considered here and is proportional to the CO2 mole fraction in the coal.

Link

Confined Water Dissociation in Microporous Defective Silicates: Mechanism, Dipole Distribution, and Impact on Substrate Properties

Interest in microporous materials has risen in recent years, as they offer a confined environment that is optimal to enhance chemical reactions. Calcium silicate hydrate (C-S-H) gel, the main component of cement, presents a layered structure with sub-nanometer-size disordered pores filled with water and cations. The size of the pores and the hydrophilicity of the environment make C-S-H gel an excellent system to study the possibility of confined water reactions. To investigate it, we have performed molecular dynamics simulations using the ReaxFF force field. The results show that water does dissociate to form hydroxyl groups. We have analyzed the water dissociation mechanism, as well as the changes in the structure and water affinity of the C-S-H matrix and water polarization, comparing the results with the behavior of water in a defective zeolite. Finally, we establish a relationship between water dissociation in C-S-H gel and the increase of hardness due to a transformation from a two- to a three-dimensional structure.

Link

Evidence on the Dual Nature of Aluminum in the Calcium-Silicate-Hydrates Based on Atomistic Simulations

Hydration of tri-calcium silicate (C3S) and di-calcium silicate (C2S) precipitates calcium-silicate-hydrate (CSH) which is the bonding phase responsible for the strength of cementitious materials. Substitution of part of C3S and C2S with aluminum-containing additives alters the chemical composition of hydration products by precipitating calcium-aluminate-silicate-hydrate (CASH). Incorporation of aluminum in the molecular building blocks of CSH entails structural and chemo-mechanical consequences. These alterations can be measured through solid state nuclear magnetic resonance (NMR) experiments. By conducting a wide spectrum of atomistic simulation methods on thousands of aluminum-containing molecular CASH structures, an overall molecular approach for determination of CASH nanostructure is presented. Through detailed analysis of different order parameters, it is found that aluminum can exhibit a tetra-/penta-/octahedral behavior which is fully consistent with the recent NMR observations. This corresponds to the formation of a class of complex three-dimensional alumino-silicate skeletons with partial healing effect in the CASH nanostructure potentially increasing durability and strength of hydration products. We explored the variation of mechanical observables by increasing aluminum content in CASH structures of varying calcium to silicon ratio. Finally, deformation of CSHs and CASHs of different chemical formula in a multi-scale fashion unravels the effect of chemical composition on the strength and kinematics of deformation in this particular type of composites.

Link

Density and water content of nanoscale solid C–S–H formed in alkali-activated slag (AAS) paste and implications for chemical shrinkage

Alkali-activated slag (AAS) paste was analyzed using small-angle neutron scattering (SANS). The scattering response indicates that the microstructure consists of a uniform matrix of hydration product with a high surface area studded with unhydrated cores of slag particles. In contrast with portland cement paste, no surface fractal scattering regime was detected, and elevated temperature curing (at 60 °C) had no detectable effect on the microstructure at any length scale studied. The specific surface area of the AAS pastes is about 25% higher than that of a portland cement paste cured under the same conditions. The composition and mass density of the nanoscale solid C–S–H phase formed in the AAS paste was determined using a previously developed neutron scattering method, in conjunction with a hydration model. The result ((CaO)0.99–SiO2–(Al2O3)0.06–(H2O)0.97, d = (2.73 ± 0.02) g/cm3) is significantly lower in calcium and in water as compared to portland cement or pure tricalcium silicate paste. These values were used to calculate the chemical shrinkage that would result from complete hydration of the AAS paste. The result, (12.2 ± 1.5) cm3 of volumetric shrinkage per 100 g of unhydrated cement, is about twice the amount of chemical shrinkage exhibited by normal cement pastes.

Link

Nano-Engineering of Concrete

This paper summarizes recent developments in the field of nanoindentation analysis of highly heterogeneous composites. The fundamental idea of the proposed approach is that it is possible to assess nanostructure from the implementation of micromechanics-based scaling relations for a large array of nanoindentation tests on heterogeneous materials. We illustrate this approach through the application to calcium-silicate-hydrate (C-S-H), the binding phase of all cement-based materials. For this important class of materials, we show that C-S-H exists in at least three structurally distinct but compositionally similar forms: low density, high density and ultra-high density. These three forms differ merely in the packing density of 5-nm sized particles. The proposed approach also gives access to the solid particle properties of C-S-H, which can now be compared with results from atomistic simulations. By way of conclusion, we show how this approach provides a new way of analyzing complex hydrated nanocomposites, in addition to classical microscopy techniques and chemical analysis. This approach will turn out invaluable in our quest of adding the necessary “green” value to a commodity, concrete, by nano-engineering higher strength and toughness from first principles.

Link

Fracture scaling relations for scratch tests of axisymmetric shape

Scratch testing and scratch test analysis continues to gain momentum in Applied Mechanics, due to the possibility offered by this method to assess fracture properties at very fine scales. In this paper, we derive general scratch force scaling relations for axisymmetric scratch probes defined by single variable monomial functions. These relations are used to define fracture criteria with and without consideration of the development of shear stresses at the probe–material interface. The approach is illustrated for common scratch probe geometries: conical probe, flat punch, and hemi-spherical probe. Application of the proposed method to micro-scratch tests on two materials (an aluminum alloy and a thermoplastic polymer) using a Rockwell probe (a conical probe ending in a hemi-spherical shape) illustrates the versatility of the approach: First, the scratch force-depth scaling relations provide a means to determine the degree of the homogeneous function characterizing the scratch probe. Second, the fracture criteria enable an experimental assessment of the fracture toughness. The good agreement between the fracture toughness determined by scratching and values reported in the open literature show the potential of the proposed method for determining fracture properties of materials at even smaller scales.

Link

Scratch hardness–strength solutions for cohesive-frictional materials

In this paper we develop analytical solutions for scratch hardness–strength relations for cohesive-frictional materials of the Mohr–Coulomb and Drucker–Prager type. Based on the lower bound yield design approach, closed-form solutions are derived for frictionless scratch devices, and validated against computational upper bound and elastoplastic finite element solutions. The influence of friction at the blade–material interface is also investigated, for which a simple computational optimization is proposed. Illustrated for scratch tests on cement paste, we show that the proposed solutions provide a convenient way to determine estimates of cohesion and friction parameters from scratch data, and may serve as a benchmark to identify the relevance of strength models for scratch test analysis.

Link

Quantifying plasticity-independent creep compliance and relaxation of viscoelastoplastic materials under contact loading

Here we quantify the time-dependent mechanical properties of a linear viscoelastoplastic material under contact loading. For contact load relaxation, we showed that the relaxation modulus can be measured independently of concurrent plasticity exhibited during the loading phase. For indentation creep, we showed that the rate of change of the contact creep compliance L.(t) can be measured independently of any plastic deformation exhibited during loading through L.(t)=2a(t)h.(t)/Pmax, where a(t) is the contact radius, h(t) is the displacement of the contact probe, and Pmax is the constant applied load during the creep phase. These analytical relations were compared with numerical simulations of conical indentation creep for a viscoelastoplastic material and validated against sharp indentation creep experiments conducted on polystyrene. The derived relations enable extraction of viscoelastic material characteristics, even if sharp probes confer concurrent plasticity, applicable for a general axisymmetric contact probe geometry and a general time-independent plasticity.

Link

Experimental determination of the fracture toughness via microscratch tests: Application to polymers, ceramics, and metals

This article presents a novel microscratch technique for the determination of the fracture toughness of materials from scratch data. While acoustic emission and optical imaging devices provide quantitative evidence of fracture processes during scratch tests, the technique proposed here provides a quantitative means to assess the fracture toughness from the recorded forces and depth of penetration. We apply the proposed method to a large range of materials, from soft (polymers) to hard (metal), spanning fracture toughness values over more than two orders of magnitude. The fracture toughness values so obtained are in excellent agreement with toughness values obtained for the same materials by conventional fracture tests. The fact that the proposed microscratch technique is highly reproducible, almost nondestructive, and requires only small material volumes makes this technique a powerful tool for the assessment of fracture properties for microscale materials science and engineering applications.

Link

Poromechanics of microporous media

Microporous media, i.e., porous media made of pores with a nanometer size, are important for a variety of applications, for instance for sequestration of carbon dioxide in coal, or for storage of hydrogen in metal-organic frameworks. In a pore of nanometer size, fluid molecules are not in their bulk state anymore since they interact with the atoms of the solid: they are said to be in an adsorbed state. For such microporous media, conventional poromechanics breaks down.

In this work we derive poroelastic constitutive equations which are valid for a generic porous medium, i.e., even for a porous medium with pores of nanometer size. The complete determination of the poromechanical behavior of a microporous medium requires knowing how the amount of fluid adsorbed depends on both the fluid bulk pressure and the strain of the medium. The derived constitutive equations are validated with the help of molecular simulations on one-dimensional microporous media. Even when a microporous medium behaves linearly in the absence of any fluid (i.e., its bulk modulus does not depend on strain), we show that fluid adsorption can induce non-linear behavior (i.e., its drained bulk modulus can then depend significantly on strain). We also show that adsorption can lead to an apparent Biot coefficient of the microporous medium greater than unity or smaller than zero.

The poromechanical response of a microporous medium to adsorption significantly depends on the pore size distribution. Indeed, the commensurability (i.e., the ratio of the size of the pores to that of the fluid molecules) proves to play a major role. For a one-dimensional model of micropores with a variety of pore sizes, molecular simulations show that the amount of adsorbed fluid depends linearly on the strain of the medium. We derive linearized constitutive equations which are valid when such a linear dependence of the adsorbed amount of fluid on the strain is observed.

As an application, the case of methane and coal is considered. Molecular simulations of an adsorption of methane on a microporous realistic model for coal are performed with a flexible solid skeleton. The applicability of the set of linearized constitutive equations to this case is discussed and the results are shown to be consistent with swelling data measured during a classical adsorption experiment.

Link

A nanoscale numerical model of calcium silicate hydrate

This manuscript presents a numerical model of the low-density and high-density calcium–silicate–hydrate (C–S–H) gel phases in cement paste. Generated using an autocatalytic growth algorithm, C–S–H is introduced as an assemblage of discrete granular particles at nanoscale with realistic particle-level properties, such as elastic modulus, friction, and cohesion. Using the discrete element method, nanoindentation simulations are performed on each phase, demonstrating that its mechanical contact properties compare well to the results from nanoindentation experiments in the literature. By creating an additional loosely packed phase of C–S–H and maintaining constant particle-level material properties, the results further show that the indentation modulus, as a function of the volumetric packing fraction of the C–S–H gel phase, compares well to a linear self-consistent scaling relation while the hardness most closely fits a nonlinear self-consistent scaling relation.

Link

The scratch test for strength and fracture toughness determination of oil well cements cured at high temperature and pressure

Recent advances in scratch test analysis provide new ways to relate measured scratch test properties not only to strength properties but fracture properties of materials as well. Herein, we present an application of such tools to oil well cements cured at high temperatures and pressures. We find a concurrent increase of strength and toughness of different oil well cement baseline formulations which we relate to the water-to-binder ratio for a series of cementitious materials prepared with cement and silica flour. The scratch test thus emerges as a self-consistent technique for both cohesive–frictional strength and fracture properties that is highly reproducible, almost non-destructive, and not more sophisticated than classical compression tests, which makes this ‘old’ test highly attractive for performance-based field applications.

Link

Scratching as a Fracture Process: From Butter to Steel

We present results of a hybrid experimental and theoretical investigation of the fracture scaling in scratch tests and show that scratching is a fracture dominated process. Validated for paraffin wax, cement paste, Jurassic limestone and steel, we derive a model that provides a quantitative means to relate quantities measured in scratch tests to fracture properties of materials at multiple scales. The scalability of scratching for different probes and depths opens new venues towards miniaturization of our technique, to extract fracture properties of materials at even smaller length scales.

Link

Scratch test model for the determination of fracture toughness

We revisit the scratch test within the framework of linear elastic fracture mechanics. In the analysis, we employ an Airy stress function approach to determine stresses and displacement in the vicinity of the scratch-blade–material interface, which serve as input for the evaluation of the energy release rate by means of the J-Integral. In contrast to previous models, the energy release rate thus found scales with the sum of the applied forces squared. This entails a linear relation between the applied forces and w.sqrt(d), where w is the scratch width and d the scratch depth. This analytical scaling is validated using experimental scratch data on cement paste and sandstone, which shows that the proposed approach provides a convenient way to determine the fracture toughness from scratch tests carried out with different scratch widths and depths.

Link

Comparison of Computational Water Models for Simulation of Calcium-Silicate-Hydrate

Calcium silicate hydrate, or C-S-H, is the chief hydration product of Portland cement. The structure of the C-S-H phase within cement has been proposed and developed via molecular simulations. In such simulations, empirical interatomic potentials for water molecules within C-S-H are adopted to govern the position and relative motion of this key constituent. Initial simulations and force fields of C-S-H have assumed the simplest molecular model of H2O termed "single point charge" or SPC, but this choice has not been validated by comparison with other computational models of water that confer additional bond flexibility or charge distribution. To enable efficiently computational modeling of C-S-H and to explore the role that H2O plays in maintaining C-S-H structure and properties, the choice of an efficient and accurate water model is critical. Here, we consider five distinct, classical atomistic water models (SPC, TIP3P, TIP4P, TIP4P05, and TIP5P) to determine the effects of these computational simplifications on C-S-H properties. Quantitative comparison of all five water models shows that the appropriate water model depends on the C-S-H characteristics of interest. Among these models, both SPC and TIP5P models successfully predict key properties of the structure and elastic constants of C-S-H, as well as the dynamics within C-S-H.

Link

A novel and general form of effective stress in a partially saturated porous material: The influence of microstructure

A recently published constitutive model for drying of a partially saturated porous material is extended to take into account finite air (gas) pressure as well as finite external load, variables that are absent during simple drying under atmospheric conditions. We further use the result to derive a general form of effective stress in a partially saturated material. The proposed framework overcomes a primary shortcoming of the classic Bishop effective stress expression and offers a novel way to incorporate important morphological features. In addition, we show how existing micromechanical homogenization techniques aided by basic descriptions of material morphology may be used to inform the study of elastic deformations in the partially saturated media. We further show that for two very specific schemes, and thus for two particular morphologies of porous materials, the proposed effective stress framework remarkably reduces to a volumetric average pressure form which is commonly encountered in literature. In this work, we also provide an extensive discussion and critique of the classic Bishop effective stress approach.

Link

Glassy Nature of Water in an Ultraconfining Disordered Material: The Case of Calcium−Silicate−Hydrate

We present the structural and dynamic nature of water ultraconfined in the quasi-two-dimensional nanopores of the highly disordered calcium−silicate−hydrate (C-S-H), the major binding phase in cement. Our approach is based on classical molecular simulations. We demonstrate that the C-S-H nanopore space is hydrophilic, particularly because of the nonbridging oxygen atoms on the disordered silicate chains which serve as hydrogen-bond acceptor sites, directionally orienting the hydrogen atoms of the interfacial water molecules toward the calcium−silicate layers. The water in this interlayer space adopts a unique multirange structure: a distorted tetrahedral coordination at short range up to 2.7 Å, a disordered structure similar to that of dense fluids and supercooled phases at intermediate range up to 4.2 Å, and persisting spatial correlations through dipole−dipole interactions up to 10 Å. A three-stage dynamics governs the mean square displacement (MSD) of water molecules, with a clear cage stage characteristic of the dynamics in supercooled liquids and glasses, consistent with its intermediate-range structure identified here. At the intermediate time scales corresponding to the β-relaxation of glassy materials, coincident with the cage stage in MSD, the non-Gaussian parameter indicates a significant heterogeneity in the translational dynamics. This dynamic heterogeneity is induced primarily because of the heterogeneity in the distribution of hydrogen bond strengths. The strongly attractive interactions of water molecules with the calcium silicate walls serve to constrain their motion. Our findings have important implications on describing the cohesion and mechanical behavior of cement from its setting to its aging.

Link

Empirical force fields for complex hydrated calcio-silicate layered materials

The use of empirical force fields is now a standard approach in predicting the properties of hydrated oxides which are omnipresent in both natural and engineering applications. Transferability of force fields to analogous hydrated oxides without rigorous investigations may result in misleading property predictions. Herein, we focus on two common empirical force fields, the simple point charge ClayFF potential and the core–shell potential to study tobermorite minerals, the most prominent family of Calcium–Silicate–Hydrates that are complex hydrated oxides. We benchmark the predictive capabilities of these force fields against first principles results. While the structural information seem to be in close agreement with DFT results, we find that for higher order properties such as elastic constants, the core–shell potential quantitatively improves upon the simple point charge model, and shows a larger degree of transferability to complex materials. In return, to remedy the deficiencies of the simple point charge potential for hydrated calcio-silicates, we suggest using both structural data and elasticity data for potential calibration, a new force field potential, CSH-FF. This re-parameterized version of ClayFF is then applied to simulating an atomistic model of cement (Pellenq et al., PNAS, 2009). We demonstrate that this force field improves the predictive capabilities of ClayFF, being considerably less computational intensive than the core–shell model.

Link

C-Crete: From atoms to concrete structures

We review recent developments of a multiscale bottom-up approach for concrete. The approach starts at the electron and atomic scale to nanoengineer the fundamental building block of concrete; to assess the properties by nanoindentation; and upscale strength, fracture and stiffness properties from nanoscales to macroscales of day-to-day concrete engineering applications. The key to all this is mechanics at the interface of physics and engineering.

Link

Fracture Toughness BottomUp: Experiments and Simulations

Link

Simulation of hydration and Elastic Properties of Montmoriollnite using Molecular Dynamics

This article describes molecular-level simulations of the adsorption isotherm and elastic properties for a Wyoming Na-montmorillonite using the General Utility Lattice Program. The selected clay mineral includes isomorphous substitutions in both the silica tetrahedral and octahedral gibbsite sheets. The resulting negative charges are balanced by Na+ cations. The authors have investigated the importance of different interatomic potentials in understanding the adsorption of water molecules (at ambient temperature, 300K). A unique combination of NPT and Grand Canonical ensemble simulations using Core–Shell potential shows intra-layer adsorption, while the same approach based on empirical CLAYFF potential produces only interlayer water adsorption. There is a correspondingly large difference in the theoretical elastic modulus normal to the clay sheets using the two different sets of force potentials. This article discusses the underlying reasons for these differences and reviews the experimental evidence supporting intra-layer water absorption in montmorillonite.

Link

Adsorption and Strain

Link

Roughness of fracture surfaces and compressive strength of hydrated cement pastes

A new type of roughness number Rno is formulated as a possible analytical tool for surface studies using confocal microscopy. The formulation accommodates fractal dimension and both of the boundary length scales limiting the fractal region of the fracture surface under investigation. Besides the number Rno other roughness characteristics are discussed and their effectiveness in the field of cementitious materials is tested. 3D-surface profile SP and surface roughness SR parameters designed for surface 3D-analysis were calculated for fracture surfaces of hydrated Portland cement pastes with different values of water-to-cement ratio. The surface profile SP parameters monotonically increased with water-to-cement ratio and monotonically decreased with compressive strength. A short discussion of possible reasons for such a behavior is presented.

Link

The effect of drying on early-age morphology of C–S–H as observed in environmental SEM

The morphology of early-age C–S–H, often referred to as outer product or low-density C–S–H, is generally accepted to be fibrillar and forms mainly during the early stages of hydration. This paper reports the effect of drying on the microstructure in young tricalcium silicate paste, which provides insight into the mechanism that leads to the fibrillar morphology. During the first few days after C3S is mixed with water, the morphology of C–S–H is more affected by drying rate than by relative humidity. This sensitivity is most apparent at partial pressures greater than 85%. During this time, the fibrillar C–S–H product can be suppressed by drying C3S paste samples very slowly prior to imaging. This approach is interpreted as evidence that this fibrillar morphology, which naturally form over time, grow as tiny colloidal particles that rearrange during the early stages of hydration, leading to well-established larger scale morphologies.

Link

A coupled nanoindentation/SEM-EDS study on low water/cement ratio Portland cement paste: Evidence for C-S-H/Ca(OH)2 nanocomposites

A low water/cement ratio (w/c=0.20) hydrated Portland cement paste was analyzed by grid-indentation coupled with ex situ scanning electron microscope-energy-dispersive X-ray spectra (SEM-EDS) analysis at each indentation point. Because finite element and Monte-Carlo simulations showed that the microvolumes probed by each method are of comparable size (approximately 2 μm), the mechanical information provided by nanoindentation was directly comparable to the chemical information provided by SEM-EDS. This coupled approach provided the opportunity to determine whether the local indentation response was a result of a single- or a multiphase response--the latter being shown predominant in the highly concentrated w/c=0.20 hydrated cement paste. Results indicate that, in the selected microvolumes where C-S-H and nanoscale Ca(OH)2 (CH) are present, increasing fractions of CH increase the local indentation modulus (and hardness), yielding values above those reported for high-density (HD) C-S-H. Micromechanical analyses show that C-S-H and CH are associated, not merely as a simple biphase mixture, but as an intimate nanocomposite where nanoscale CH reinforces C-S-H by partially filling the latter's gel pores. The paper discusses the mechanism of forming the C-S-H/CH nanocomposite, as well as the impact of nanocomposites on various macroscopic properties of concrete (e.g., shrinkage, expansion). On a general level, this study illustrates how a coupled nanoindentation/X-ray microanalysis/micromechanics approach can provide otherwise inaccessible information on the nanomechanical properties of highly heterogeneous composites with intermixing at length scales smaller than the stress field in a nanoindentation experiment.

Link

Nanogranular packing of C–S–H at substochiometric conditions

Herein, we present a comprehensive nanoindentation investigation of cement pastes prepared at substoichiometric water-to-cement (w/c) mass ratios between 0.15 and 0.4 with and without heat treatment. Based on a statistical indentation technique, we provide strong evidence of the existence of a statistically significant third hydrated mechanical phase in addition to the already known Low-Density (LD) and High-Density (HD) C–S–H phases. The nanomechanical properties of this third phase are found to follow similar packing density scaling relations as LD C–S–H and HD C–S–H, while being significantly greater. This third phase, whose nano-packing density is measured at 0.83 ± 0.01, is therefore termed Ultra-High-Density (UHD) phase. All three phases are present in concrete materials in different volume proportions: LD dominates cement-based materials prepared at high w/c mass ratios; HD and UHD control the microstructure of low w/c ratio materials. In addition, heat treatment favors the formation of HD and UHD. The insight thus gained into the link between composition, processing and microstructure makes it possible to monitor packing density distributions of the hydration products at the nanoscale.

Link

Does microstructure matter for statistical nanoindentation techniques?

In their paper, Trtik et al. (2009) identify spurious peaks in the application of statistical nanoindentation technique as a critical obstacle for mechanical phase identification. In this discussion, we show that Trtik et al.’s finding is a consequence of an unrealistic virtual 3-D checkerboard microstructure considered by the authors. These peaks are not a general feature of indentation on multiphase materials, nor can the presence of such peaks be attributed to an intrinsic shortcoming of the grid-indentation technique. We also show that the authors’ assertion of the absence of homogeneous material regions extending beyond 3 μm in cementitious materials is groundless.

Link

First-Principles Study of Elastic Constants and Interlayer Interactions of Complex Hydrated Oxides: Case Study of Tobermorite and Jennite

It is a common perception that layered materials are soft in the interlayer direction. Herein, we present results of first-principles calculations of the structure and elastic constants of a class for hydrated oxides, tobermorite, and jennite, which illustrate that this is not the case, if (1) the interlayer distance is such that coulombic interlayer interactions become comparable to the iono-covalent intralayer interactions and (2) the existence of interlayer ions and water molecules do not shield the coulombic interlayer interactions. In this case, the mechanically softest directions are two inclined regions that form a hinge mechanism. The investigated class of materials and results are relevant to chemically complex hydrated oxides such as layered calcium–silicate–hydrates (C–S–H), the binding phase of all concrete materials, and the principle source of their strength and stiffness. In addition, the first-principles results may serve as a benchmark for validating empirical force fields required for the analysis of complex calcio–silicate oxides.

Link

A realistic molecular model of cement hydrates

Despite decades of studies of calcium-silicate-hydrate (C-S-H), the structurally complex binder phase of concrete, the interplay between chemical composition and density remains essentially unexplored. Together these characteristics of C-S-H define and modulate the physical and mechanical properties of this “liquid stone” gel phase. With the recent determination of the calcium/silicon (C/S = 1.7) ratio and the density of the C-S-H particle (2.6 g/cm3) by neutron scattering measurements, there is new urgency to the challenge of explaining these essential properties. Here we propose a molecular model of C-S-H based on a bottom-up atomistic simulation approach that considers only the chemical specificity of the system as the overriding constraint. By allowing for short silica chains distributed as monomers, dimers, and pentamers, this C-S-H archetype of a molecular description of interacting CaO, SiO2, and H2O units provides not only realistic values of the C/S ratio and the density computed by grand canonical Monte Carlo simulation of water adsorption at 300 K. The model, with a chemical composition of (CaO)1.65(SiO2)(H2O)1.75, also predicts other essential structural features and fundamental physical properties amenable to experimental validation, which suggest that the C-S-H gel structure includes both glass-like short-range order and crystalline features of the mineral tobermorite. Additionally, we probe the mechanical stiffness, strength, and hydrolytic shear response of our molecular model, as compared to experimentally measured properties of C-S-H. The latter results illustrate the prospect of treating cement on equal footing with metals and ceramics in the current application of mechanism-based models and multiscale simulations to study inelastic deformation and cracking.

Link

A model for the C-A-S-H gel formed in alkali-activated slag cements

For first time, an experimental and computational study has been conducted to define a structural model for the C-A-S-H gel forming in alkali-activated slag (AAS) pastes that would account for the mechanical properties of these materials. The study involved a comparison with the C-S-H gel forming in a Portland cement paste.

The structure of the C-A-S-H gels in AAS pastes depends on the nature of the alkali activator. When the activator is a NaOH, the structure of the C-S-H gel falls in between tobermorite 1.4 nm with a mean chain length of five, and tobermorite 1.1 nm with a mean length of 14. When waterglass is the activator the structure of the C-A-S-H gel is indicative of the co-existence of tobermorite 1.4 nm with a chain length of 11 and tobermorite 1.1 nm with a chain length of 14. This very densely packed structure gives rise to excellent mechanical properties.

Link

Time scale bridging in atomistic simulation of slow dynamics: viscous relaxation and defect activation

Atomistic simulation methods are known for timescale limitations in resolving slow dynamical processes. Two well-known scenarios of slow dynamics are viscous relaxation in supercooled liquids and creep deformation in stressed solids. In both phenomena the challenge to theory and simulation is to sample the transition state pathways efficiently and follow the dynamical processes on long timescales. We present a perspective based on the biased molecular simulation methods such as metadynamics, autonomous basin climbing (ABC), strain-boost and adaptive boost simulations. Such algorithms can enable an atomic-level explanation of the temperature variation of the shear viscosity of glassy liquids, and the relaxation behavior in solids undergoing creep deformation. By discussing the dynamics of slow relaxation in two quite different areas of condensed matter science, we hope to draw attention to other complex problems where anthropological or geological-scale time behavior can be simulated at atomic resolution and understood in terms of micro-scale processes of molecular rearrangements and collective interactions. As examples of a class of phenomena that can be broadly classified as materials ageing, we point to stress corrosion cracking and cement setting as opportunities for atomistic modeling and simulations.