CONCREEP 10: Mechanics and Physics of Creep, Shrinkage, and Durability of Concrete and Concrete Structures

It has generally been accepted that the volume of cement hydration products is slightly smaller than the original volume of cement and water. However, this does not mean that the hydration reaction causes the hardened cement paste and concrete to contract. In fact, C-S-H shells that grow around anhydrous cement grains push the neighbors apart by crystallization pressure and thus cause the solid framework of cement paste to expand. Proposed here is a new idea—this expansion always dominates over the contraction, i.e., the hydration is, in the bulk, always expansive, while the source of all of the observed shrinkage, whether autogenous or due to external drying, is a compressive elastic strain in the solid caused by a decrease of chemical potential of pore water, with the corresponding changes in pore humidity, surface tension and disjoining pressure. From recent observations of autogenous shrinkage growing logarithmically in time over many years it follows that the growing C-S-H shells surrounding cement grains must act as diffusion barriers for water and ions, which slow down the hydration process and can extend it over many years and even decades. The new idea implies that all of the autogenous shrinkage must be caused by elastic compression (probably with no, or almost no, creep) of crystalline nano-sheets in the solid framework subjected to stresses that arise as a reaction to pore water stresses. Swelling under water immersion is explained by insufficient elastic compression when water is permanently supplied to the pores. The lecture first presents the aforementioned theory and then summarizes some recent advances in related phenomena, particularly a model for oriented damage due to alkali-silica reaction and a method for shrinkage extrapolation.

It has long been argued that shrinkage in cementitious materials can occur even in the absence of drying phenomena. Herein we argue that such shrinkage is attributable to colloidal forces that develop between C-S-H particles during the hydration in an out-of-equilibrium process. We base our argument on meso-scale simulation results, and show via microporomechanics theory that such colloidal forces can lead to the build-up of solid eigenstresses causing macroscopic shrinkage deformation.

Aggregates in cementitious composites restrain the volume changes of the matrix, which reduces the overall shrinkage of the composite and may lead to internal microcracking. In this paper, autogenous shrinkage of cement pastes and mortars with different amounts of aggregates was measured. While in the case of thermal deformations it is possible to estimate the volume changes of the composites with reasonable accuracy using effective medium approaches, predictions of early-age shrinkage tend to be less accurate, possibly because it is difficult to fully take into account the viscoelastic behavior of the cement paste, which is particularly pronounced at early ages.

This paper deals with important issue of the aging and deterioration of concrete structures: a problem at the forefront of civil engrineering research that has profound impact on our society and our way of living. Despite the recognized importance of such issue, entrenched practice and remediation approaches are, for the most part, empirical and fundamentally related only to the lifetime experience of few capable professionals. We all need to thank the work of these people but this paper argues that such “modus operandi” is only slightly better than predicting the future of infrastructure by looking into a crystal ball—it is more magic than science. On the contrary, this paper presents a different vision in which “predictions” are based on a rigourous structural reliability framework equipped with physically-sound computational models that are rigorously validated and continually updated on the basis of reliable experimental data obtained from both destructuve and non-destructuve techniques.

By the end of cement hydration calcium-silicate-hydrate (C-S-H) gels extends over tens and hundreds of nanometers. Their complex texture affects directly, and to a large extent, the macroscopic hygrothermal and mechanical behavior of cement. Here we review a statistical physics approach recently developed, which allows us to investigate the gel formation under the out-of-equilibrium conditions typical of cement hydration and the role of the nano-scale structure in C—S—H mechanics upon hardening. Our investigations have unveiled the role, in the C-S-H gels, of nano-scale structural and mechanical heterogeneities that develop due to the the far-from-equilibrium physico-chemical environment in which the material forms. A subtle interplay between the out-of-equilibrium evolution and the effective interactions emerging between the nano-scale units of the gels at different stages of the hydration process ultimately determines the mesoscale texture of cement hydrates and their material properties.

Despite their impact on longevity, serviceability, and environmental footprint of our built infrastructure, the chemo-physical origins of nanoscale properties of cementitious materials, and their link to macroscale properties still remain rather obscure. Here, we discuss a multi-scale approach that describes different aspects of physical properties of C-S-H at the nano- and meso-scales. These include dynamics of water, thermal properties and mechanical behavior of C-S-H and its effect on properties of cement paste at different scales.

The purpose of this work is to achieve a better understanding of the relationship between mechanical damage and the transport properties of cementitious materials. In the literature, analyses are usually restricted to intrinsic permeability of the material and the evolution of the apparent permeability with respect to the pressure gradient and to the nature of the fluid considered are left aside. A new model capable to provide the apparent permeability of a porous material to gas, directly from the pore size distribution and from the properties of the gas is discussed. Comparisons with experimental data on mortar specimens show that the model can reproduce the intrinsic permeability and its evolution when the material is subjected to mechanical damage, provided the pore size distributions are available. For a given pore size distribution, the evolution of the apparent permeability is also provided and test data with several types of gases compare quite well with the model.

This paper presents a summary of recent experimental researches about basic creep of concretes. From the analysis of these researches, a proposal is made concerning the dominant physical mechanism at the origin of this basic creep. Based on this physical mechanism, a new way to model, by using finite elements method, basic creep of concrete structures is proposed.

Recent advances in the application of the three fluids model for concrete will be discussed where creep matters such as in repair problems. As known, this model considers concrete as composed of a deformable solid matrix with pores filled by dry air, capillary water and water vapor. The interaction of the constituents as well as phase change is taken into account. It will also be shown that the same material free balance equations can successfully be adopted for the simulation of tumor growth and drug delivery. In this case the solid phase is the extracellular matrix and the three fluids are formed by the healthy cells, the tumor cells and the interstitial fluid which transports the chemical species (therapeutic agents). The extracellular matrix is an elasto-visco-plastic porous solid of the Perzyna type. Applications for both forms of the general model will be shown.

Creep is not observed for cementitious materials only: most materials creep, in particular materials very relevant for civil engineering applications, namely clay or wood. Although each of those materials possesses its own specific features, some analogies between their respective creep behaviors can be drawn. From these analogies, we can try to obtain meaningful pieces of information on the physical processes at stake during creep. After having introduced the topic, I present a few of those analogies and discuss briefly what can be learned from them.

In this contribution selected fundamental properties of concrete are presented and discussed. Basic mechanisms of shrinkage are first described in some detail. With increasing age of concrete different mechanisms of shrinkage become activated. During the first hours after mixing dissolution of cement is the main shrinkage mechanism. Depending on the humidity of the surrounding air capillary pressure in the pore water may cause serious damage in the first hours. After this initial loss of water, the remaining aqueous solution is retreated in the free spaces between fine aggregates and first hydration products. The role of disjoining pressure is pointed out in detail. In this context it becomes clear that the influence of capillary pressure on shrinkage of hardened concrete can be neglected. In equilibrium with RH lower than 50 % the dominating mechanism is the increasing surface energy of the drying gel particles. Shrinkage of high strength concrete has a totally different time evolution as the time-dependent moisture distribution in aging high strength concrete is completely different. If high strength concrete is exposed to an environment with RH higher than 80% there will be no shrinkage at all but swelling. Creep of concrete can be described in a realistic way by means of rate theory. Elements of rate theory are briefly described in the last section. Prediction of creep can be significantly improved if these useful fundamentals are taken into consideration.

Although amorphous materials are usually brittle, and break without deforming irreversibly at observable sclaes, their fracture involves some dissipation. By examining the post-mortem fracture surfaces, by following in situ the progession of a stress corrosion crack in silicate glasses and by measuring the depth of penetration of water in the material during slow fracture, we show that glass behaves in fact in a quasibrittle manner, with a damage zone of ≈ 10 nm developping ahead of the crack tip. In order to make these damage scales observable optically, we have started to work on colloidal systems, where “atoms” have the size of a few micrometers. Direct observation of crack propagation through these materials show that damage is not of plastic origin as in ductile metals, for example, but of a quasi-brittle nature.

A concrete member is subjected to loads for a long period of time, during which creep and shrinkage of concrete develop gradually. The prediction of this time-dependent behaviour is important as it may cause serious serviceability problems in concrete structures. A time-dependent analysis is commonly based on empirical equations according to design codes where the function describing the time dependent increment of shrinkage and creep is commonly, among others, defined based on the notional size of the element. In case of imbedded steel or insulated boundaries the moisture transport can be partially affected or prevented. Also, the geometry and size of the cross-section have an important effect on the shrinkage behaviour of a concrete member. Hence, the performance of commonly used empirical formulas may be improved by applying a correction factor on the notional size. In order to investigate the impact of these various factors on the net macroscopic shrinkage used in analysis and design, a discretized 2D physical model was developed. The model was used to simulate drying of a concrete cross-section by determining the moisture distribution in the cross-section as function of time.

Chloride-induced corrosion of steel reinforcement in concrete is one of the major causes for deterioration of reinforced concrete (RC) structures. RC structures exposed to aggressive environmental conditions, such as structures close to the sea or highway bridges and garages exposed to de-icing salts, very often exhibit damage due to corrosion. Therefore, to predict durability of RC structure it is important to have a numerical tool, which is able to predict corrosion processes and their consequences for the structural safety. In order to speed up corrosion of reinforcement in concrete the experimental tests are almost always accelerated by imposing external electric potential and by adding chlorides to concrete mix. Therefore, the corrosion rate becomes much faster than the fastest corrosion rate in the nature, which can have significant consequences on the non-mechanical and mechanical processes related to the corrosion of reinforcement, i.e. the experimental results can lead to unrealistic conclusions. In the present paper recently performed experiments under accelerated conditions are simulated and compared with corrosion process that approximately corresponds to natural conditions. For the investigated geometry and environmental conditions it is shown that accelerated corrosion rate, which is approximately 20 times faster than corrosion rate under natural conditions, leads to corrosion induced damage of concrete, which is similar to the damage predicted assuming natural conditions.

The deteriorating effect of alkali-silica reaction in concrete has a relevant impact both on the material and on the structural performance. This effect is mainly considered in terms of concrete expansion and microscopic damage within the aggregates. Limited attention is reserved to the degradation of mechanical properties, which are often used in the structural assessment. In this paper the attention is focussed on the relationship between ASR-induced concrete expansion and mechanical degradation (e.g. elastic modulus, tensile strength, compressive strength). The results of a statistical analysis are presented. The elastic modulus results the best indicator to identify the reaction, by showing the earliest degradation and the lowest residual values. The non-linear stress-strain behaviour of ASR-affected concrete appears substantially different from the one of unaffected concrete. Experimental observations are simulated via a multiscale micro-poro-fracture-mechanical model. The linear elastic pressure-based single porosity model overestimates the stiffness-expansion relationship. It reasonably approximates the degradation ratio between strengths and stiffness.

In this study, an experimental investigation on a high-performance concrete used for the construction of a pre-stressed bridge is presented, together with the numerical assessments of the material properties, such as strength, modulus of elasticity, creep, and shrinkage. In this experimental investigation cylindrical specimens were used to measure shrinkage strains under both sealed and drying conditions, with environmental relative humidity of 50% and temperature of 20°C. In addition, basic and drying creep tests were also conducted under the same aforementioned environmental conditions starting at the age of 2 and 28 days. The rate-type creep model, entitled solidification-microprestress-microplane (SMM) model (Di Luzio and Cusatis, 2013), which amalgamates a microplane model and the solidification-microprestress theory is adopted to simulate the age-dependent response of sealed and unsealed specimens observed in the experimental investigation previously mentioned. Comparison with experimental data shows that the SMM model can explain well the interplay of shrinkage (autogenous and drying), creep, and cracking phenomena.

In 1979, 16 concrete prisms with dimensions 140x150x4000 mm were casted in the Magnel Laboratory for Concrete Research. These prisms differ by the amount of passive reinforcement and by the applied loading level. Four reinforcement ratios were considered, i.e. 0%, 1.5%, 3% and 6%. For each reinforcement ratio a prism was subjected to an axial load corresponding to a concrete stress of 0, 5, 10 or 15 MPa. The combination of both parameters results in a total of 16 specimens. The compressive stress was applied to the prisms by means of post-tensioned unbonded strands at an age of 28 days. The stress level was kept constant during the first 12 years of the experiment by re-adjusting the force in the strands when the deviation exceeded 2% of the initial value. Afterwards, no re-adjusting of the strands took place. Significant redistribution of the stresses between the concrete and the steel can be expected due to the creep and shrinkage of concrete. This redistribution is larger for the prisms with a higher reinforcement ratio, resulting in lower creep and shrinkage strains. The time-dependent response of these prisms was modelled taking into account the prestress losses and stress redistribution between the concrete and the steel. For reasons of numerical efficiency, the compliance function was approximated by a Dirichtlet series using continuous retardation spectra. A comparison between the predicted results and the measurements is given for two prisms.

Selected results of extensive fracture testing of steel fibre reinforced concrete (developed in industry for the production of prefabricated concrete elements) after different period of hardening are presented in the paper. Experiments were performed on notched specimens in a three-point bending test configuration. The fracture parameters were evaluated using three models/approaches: double-K fracture model, effective crack length method, and work-of-fracture method. The obtained parameters can be used to quantify the structural resistance against crack initiation and propagation — or the brittleness and toughness/ductility of the structural members — as well as for the comparison of the studied or developed cement based composites or the structural members of buildings. They can be also employed for the definition of the material models for deterministic or stochastic simulation of the quasi-brittle/ductile response of cement based composites/members using Stochastic Finite Element Method model with nonlinear fracture mechanics principles implemented — this approach can help to the optimal design of developed prefabricated concrete elements.

Concrete is a time-dependent and multi-scale composite. Alkalis in cement react with reactive silica in aggregate in the presence of water and forms a gel in what is known as alkali-silica reaction (ASR). When ASR gel absorbs water, it swells. Swelling of ASR gels can damage concrete and cause cracking which reflects on long-term structural performance. In this paper, the mechanical consequences of ASR on concrete are simulated using a discrete homogenization model. Representative volume element (RVE) of concrete is established as a cluster of cement paste and aggregate particles connected using an interfacial transition zone (ITZ), which includes contact and cracking mechanisms to consider particle interaction. The mechanical evolution of concrete subjected to ASR is achieved from microstructural behaviors: volume expansion of reactive elements induces internal stresses and cause debonding and cracking. The mechanical evolution of concrete characteristics due to continuous cement hydration is also considered. The constitutive model of concrete is attained via the homogenization approach. The proposed method allows considering the effect of ASR on performance of concrete structures. A case study, in the literature, is used to demonstrate the efficiency of the proposed meso-scale homogenization approach. The proposed model shows good agreement with experimental observations of concrete behavior with ASR.

Ultra-high performance concretes (UHPCs) are cementitious composite materials with high level of performance including but not limited to high compressive strength, tensile strength and durability, reached by a low water-to-binder ratio, an optimized aggregate size distribution, thermal activation, and fiber reinforcement. In the past couple of decades, more and more UHPCs have been developed and found their way into practice. Thus, also the demand for computational models capable of describing and predicting relevant aging phenomenon to assist design and planning is increasing. This paper presents the early age experimental characterization as well as the results of subsequent simulations of a typical UHPC. Performed and simulated tests include unconfined compression, splitting, and 3-point-bending tests. The computational framework is formulated by coupling a hygro-thermal-chemical (HTC) theory with a comprehensive mesoscale discrete model: the HTC component allows taking into account various types of curing conditions with varying temperature and relative humidity and predicting the level of concrete aging, while the mechanical component, based on a recently formulated discrete model, the Lattice Discrete Particle Model (LDPM), permits the simulation of the failure behavior of concrete at the coarse aggregate level. The obtained results provide both insight in UHPC early age mechanisms and a computational model for the analysis of aging UHPC structures.

The design and long-term performance of long-span bridges is to a large part determined by serviceability related considerations. For prestressed concrete bridges the relevant phenomena are concrete creep and shrinkage, coupled with steel relaxation. It has been shown that stochastic processes and in particular continuous Gamma processes are suitable to describe the deterioration of infrastructure systems within the frameworks of performance assessment and life-time prediction. This paper investigates the applicability of gamma processes for the prediction of creep related deformations in prestressed concrete bridges and attempts to provide guidance regarding the parameter selection. Deflection and shortening histories obtained by a probabilistic nonlinear rate-type analysis of the Colle Isarco Viaduct utilizing different creep models serves as data basis for this case study.

The macroscopic continuum creep simulation always neglect internal creep/relaxation at lower scales due to the internal self-equilibrated stresses. So far, a comprehensive model for concrete creep at the meso-scale level has been lacking. In this paper, such a shortcoming is over come by the explicit implementation of the solidification-microprestress (SM) theory within the Lattice Discrete Particle Model (LDPM). Aging effect is obtained using a global reaction degree of concrete obtained by a multi physics model evolving temperature, humidity and cement degree of reaction in full coupling over time and space leading to an elegant and simple implementation within the LDPM framework through an imposed eigenstrain. This leaves the features of the LDPM constitutive equation simulating material strength and toughness completely unaltered. To show the superiority of the proposed model, extensive calibration and validation of the model is pursued by numerical simulations of experimental data from literature.

This paper presents a computational multi-level model for the description of alkali and moisture transport in concrete structures coupled to a macroscopic ASR induced phase-field damage model. Concrete is modeled as a heterogeneous material consisting of a partially saturated pore space with diffusively distributed microcracks and the solid skeleton (cement paste and potentially reactive and inert aggregates). The influence of the topology of the pore space and the presence of oriented microcracks on ion diffusion and moisture transport is taken into account through a novel continuum micromechanics homogenization model. The transport model is connected to a phenomenological reaction kinetics model to account for the ASR induced volume expansion of the affected aggregates. At the macroscopic scale, crack propagation and effects of induced topological changes on the fluid and ion transport are taken into account using a phase-field model.

In this study, authors measured strain and temperature of existing reinforced concrete structure using embedded strain gauge for the purpose to predict the shrinkage crack risk of the concrete structure. Authors also measured shrinkage and restrained of stresses relationship of concrete at laboratory. As a result, the shrinkage of the concrete under actual environmental condition was approximately 300*10^-6 at 1.5 years after casting of the concrete. The shrinkage stress of wall was approximately 1.0N/mm².In addition, shrinkage strain of underground beam was 70 ×10^-6 and it’s temperature changes ranged from -5 °C to +5 °C throughout a year.

Concrete pavement joints have shown the potential to deteriorate, thereby compromising the service life of an otherwise healthy pavement. It is proposed that this damage is due to fluid collecting at the joint. The collection of fluid makes concrete 1) susceptible to freeze and thaw damage when it reaches a critical degree of saturation and 2) susceptible to damage due to a chemical reaction between the salt and matrix. Currently, models that describe the rate of water absorption in concrete consider the concrete as uncracked. In service however, concrete can develop cracks that increase the rate of transport. The first part of this paper will begin by discussing water absorption in undamaged concrete. The second part of this paper discusses the influence of distributed cracking. The third part of this paper discusses a potential chemical reaction between cement and deicing salt that forms calcium oxychloride.

The influence of microbiologically induced corrosion on the bearing capacity of concrete sewer pipes is modeled as a time dependant and stochastic problem. The potential of the model and software is shown in this presentation of a parametric study of a corroding pipe, the comparison of crushing strength to test results and an approach for the complex analysis of a buried pipe.

One of the challenges in bridge engineering is to predict the long-term deflection and damage accumulation in large-span prestressed concrete girders during their lifespan. For prestressed concrete bridges, a salient by-product of time-dependent deformation resulting primarily from concrete creep is unexpected stress redistribution, which triggers damage and cracking in the critical structural components. To deepen the understanding of the coupled effects of deflection and damage on bridge performance, a case study is carried out in this investigation, in which a large-span prestressed concrete bridge is modeled in ABAQUS based on its blueprints available. To capture the concrete behavior during service, a constitutive law formulated based on a damage model is employed to approximate the nonlinear softening after concrete cracking. This formulation is further integrated with a rheological model resting on Kelvin units for concrete creep. Using a rate-type algorithm, the long-term behavior of the bridge is probed.

This article is a continuation of the series of publications of authors to produce instant nonlinear properties of the concrete. The theory, which is a substantial development of the methods for calculating the long resistance of reinforced concrete structures, allows almost exactly the instant nonlinear properties of the concrete.

Most of large civil engineering concrete structures have been instrumented for decades with embedded sensors. To prevent the eventual loss of data, complementary instrumentation of external surface has recently been deployed. This new instrumentation can take different forms but in all cases, to avoid damaging the structure, it will be only superficially anchored. Near the outer surfaces, thermo-hydro-mechanical concrete behaviour is more sensitive to varying environmental conditions than in the centre of the structures. Therefore, the strain measured near the outer surfaces is not identical to the strain measured by embedded sensors. Consequently the methods of classical physical-statistical analysis must be reviewed. Using a thermo-hydro-mechanical finite element modeling calibrated on a representative concrete and applied on a current part of a thick structure, this work confirms a dependence of strain on the depth. First results show that the depth impact affects both kinetic and amplitude strain.

Nanoindentation testing suggests that creep of hydration products is the microscopic reason for macroscopic creep of cementitious materials. This is supported by a multiscale creep model which explains aging creep of young concretes as the consequence of universal creep of hydration products (Scheiner and Hellmich, 2009), whereby the latter is described with a rheological model consisting of linear springs and dashpots. We here extend the investigation of the origin of creep of cementitious materials further down to the nanoscale of hydration products, where we envision solid matter sliding (upon loading) along interfaces which are filled with lubricating thin layers of adsorbed water, i.e. water in a “glassy”, “liquid crystal” state. As for the viscous behavior of the interfaces, we follow (Shahidi et al., 2014) and consider that the shear traction acting on an adsorbed water layer is proportional to the shear dislocation rate of the interface, with an interface viscosity as the proportionality constant. Our analysis starts from corresponding anisotropic creep and relaxation tensors of matrix-interface composites containing parallel interfaces (Shahidi et al., 2014). Considering that hydration products contain interfaces oriented isotropically in all space directions, we here compute complete spatial averages of parallel interface-related anisotropic creep and relaxation tensors, in order to derive isotropic creep and relaxation tensor bounds. Comparing them with creep and relaxation functions of the aforementioned rheological model for universal creep of hydration products allows for identification (i) of the interface density and (ii) of the product of interface size and viscosity. Based on the Reusstype creep tensor bound, we obtain, interesting quantitative insight into microstructural features of hydration products.

Paper deals with creep effects being combined with alkali-silica reaction (ASR) in concrete. At first description of reaction kinetics is discussed, then the creep model is presented based on Model Code 2010 (MC2010) approach. An approximation of MC2010 analytical formulation in the form of Kelvin chain is proposed. Presented numerical examples show that it renders analytical formulation really well. Next a numerical algorithm developed previously by authors in order to solve chemomechanical problems is recalled. Performed exemplary numerical computations for a gravity dam show clearly that the influence of creep on the overall behavior of the structure is relatively small in comparison with effects caused by ASR itself.

To fundamentally understanding the drying behavior of cementitious materials, this paper presents a microstructure-based approach to model the drying process in pores and the shrinkage of cement specimens. At high relative humidity (>84% RH), on the basis of Kelvin-Laplace equation and a microstructure model HYMOSTRUC3D, the drying process in capillary pores is simulated by several algorithms. Considering the drying deformation of C-S-H and assuming the capillary pressure in capillary pores as the main responsible stress for the drying deformation, the shrinkage of cement paste at different RH is calculated by three-dimensional lattice analyses.

Creep and shrinkage are complicated time-dependent processes taking place in cementitious materials. Creep refers to a stress-induced increasing deformation at constant stress while shrinkage is stress-independent. In typical concretes, the most significant part of shrinkage is represented by drying shrinkage; it is caused by a decrease in relative humidity of pores which leads to an increase in capillary tension of pore water and in the solid surface tension at pore walls. Experimental data indicate that the ultimate value of drying shrinkagemeasured on concrete and mortar specimens is a nonlinear function of the ambient relative humidity. The nonlinear behavior is partially caused by microcracking, and partially by creep of the cement matrix. On the other hand, drying shrinkage of hardened cement paste, measured on very thin specimens at gradually decreasing relative humidity, is found to be a linear function of relative humidity. The aim of this paper is to assess the nature of the macroscopic shrinkage computed employing finite element simulations at the meso-scale level in which the drying cement paste is described by a viscoelastic model based on the Microprestress- Solidification Theory (MPS) with tensile cracking and with a constant shrinkage coefficient.

Delayed deformations in concrete under sustained loading may depend on complex chemo-hydro-mechanical phenomena. Generally, the creep deformation of a cement paste is distinguished in the short-term and long-term. Simply speaking, the former can be associated to the diffusion of gel water in pores and cracks and it is mainly volumetric (i.e., volumetric deformation), reversible, and bounded. The latter can be associated to the sliding of the sheets of Calcium-Silicate-Hydrate (C-S-H), which is mainly deviatoric (i.e., shape deformation), irreversible and with an unbounded asymptotic deformation. From an experimental point of view, recent studies have shown that microindentation is a powerful technique to quickly characterize basic creep thanks to the reduced characteristic time of a sample size of few micrometers. The objective of this project is twofold: (1) investigating the effect of relative humidity on the creep behavior by microindentation; (2) characterize the viscous behavior of the interface transition zones (ITZ) around a rigid inclusion. In particular, the project investigated two different relative humidity, 33% and 85%, by considering both creep and relaxation microindentation tests. The presented results show the effect of the moisture content on the creep behavior of cement paste both at short and long terms. In addition, the results showed that the ITZ zone around an aggregate showed higher creep than the bulk cement paste. Those results provide new insights to understand the role of water and interface porosity on the creep mechanisms of concrete.

Shrinkage can be critical for the strength and durability of drying cement pastes. Shrinkage becomes particularly severe at very low relative humidity, < 20%, which can be met in some activities involving extreme temperatures. Experiments and simulations suggest that small pores in the cement paste, with approximate thickness ≤ 1 nm, stay saturated unless the humidity drops below 20%. Here we suggest that this pore size can define two different categories of pores in the paste: pores thicker than 1 nm, where the Kelvin’s equation and the corresponding capillary (Laplace) pressure apply, and pores thinner than 1 nm, which can be considered as part of the solid skeleton if the humidity stays above 20%. We show that a continuum model, incorporating a pore-blocking mechanism for desorption and equilibrium thermodynamics for adsorption, explains well the sorption hysteresis for a paste that remains above ~ 20%. At lower humidities, we assume that (1) during adsorpion water re-enters the smallest pores throughout the entire RH range (supported by experiments and simulations) and (2) there exists a simple linear relationship between water and strain in the smallest pores. These minimal assumptions are sufficient to explain the low-humidity hysteresis of water content and strain, but the underlying mechanistic explanation is still an open question. Combining the low-humidity and high-humidity models allows capturing the entire drying and rewetting hysteresis, and provides parameters to predict the corresponding dimensional changes.

Creep of cementitious materials is known to originate mostly from the creep of calcium silicate hydrates (i.e., C-S-H). Here, we study numerically a model for the C-SH structure, and explicitly calculate the energy barriers that impact the rate at which two C-S-H layers slide over each other. In particular, we calculate how the viscosity scales with the microprestress. The obtained scaling is consistent with the microprestress theory which can explain why long-term creep of cementitious materials is a logarithmic function of time. Calculations are performed at various basal spacings (i.e., at various water contents), and model predictions are compared with experimental data obtained by micro-indentation of how the long-term logarithmic creep rate of C-S-H depends on the relative humidity. Considering properties of the interlayer water identical to those of bulk water, we obtain a good agreement with the experimental observations.

It has long been understood that creep and desiccation-driven shrinkage of concrete are coupled phenomena. Many mechanisms have been proposed over the past six decades by which creep and shrinkage are intertwined. Here, we propose that dissolution of solid, microscale constituents while under a state of stress leads to time-dependent deformation. As the state of stress may be incurred via external tractions on the microstructure, such time-dependent deformation may yield creep. As the state of stress may also be incurred through internal pore pressures, such time-dependent deformation may likewise generate shrinkage. By integrating the THAMES microstructural evolution model with a time-stepping finite element routine, simulations were performed to assess the magnitude of shrinkage deformations that may result from time-dependent dissolution. The simulation results indicate that the dissolution mechanism is significant, and may help explain the source of the irreversible component of desiccation shrinkage.

This paper is part of the mini-symposium on the interplay between water and shrinkage. The paper provides a short discussion of observations from a recent neutron radiography experiment where radiography was performed on plain concrete and concrete containing a shrinkage reducing admixture (SRA) when each sample was exposed to drying. While the results of the radiography test provided information on the moisture content and degree of saturation, the goal of this paper is to relate the moisture (degree of saturation) gradients, the relative humidity gradients, the shrinkage gradient and a corresponding stress gradient. The work provides evidence for the reduced curling that is often seen in concrete containing SRA. The reduced curling with SRA appears to be a combination of the reduced surface tension that limits shrinkage and the increased viscosity that reduces fluid flux.

Geomaterials present a mechanical behaviour with a more or less pronounced time dependency. This sensitivity to time effects encompasses creep and effects of strain rate and its changes. This time dependent behaviour of geomaterials is affected by humidity, and also by humidity changes. In this paper, these effects are first reviewed, with an emphasis put on a series of experiments on chalk saturated by different fluids or pairs of fluids. The main observation is that less humidity implies less creep and a smaller sensitivity of the mechanical behaviour to strain rates effects. A constitutive model accounting for saturation level and time effects is then developed in the framework of elastoplasticity, at the macroscopic scale. The constitutive laws for the description of the time-dependent mechanical behaviour generalise the isotach concept to partially saturated states. It is shown that the model can predict creep or relaxation of geomaterials, when the applied load is close to the elastic limit. Finally, numerical predictions by the proposed model are compared with available experimental results on saturated and partially saturated chalk. It is shown that the model can capture satisfactorily creep and strain rate effects on the mechanical response of this soft rock at various water saturation states.

The influence of applied stresses on adsorbed water in cement paste has been proposed long ago (independently by Powers and by Ruetz) as a possible mechanism able to explain delayed deformations (creep) of concrete. Following this approach, the application of compressive stresses should lead to an increase in internal relative humidity (RH). This hypothesis has never been confirmed experimentally. Even though water redistribution due to loading was applied in some modeling approaches for description of short-term creep, the RH dependence upon the stress state has generally been neglected over the past decades. This study shows that instantaneous RH changes do accompany applied load and can be accurately measured. Moreover, the measured RH changes may in fact explain a large part of the short-term creep deformation and also the commonly-observed difference between static and dynamic elastic moduli. In this study, mortars made with water-to-cement ratio of 0.3 were loaded in uniaxial compression or tension, while internal RH was measured by means of embedded water activity sensors. At the same time, elastic moduli were measured in compression using both the standard static method and ultrasound resonance spectroscopy.

Cementitious materials are particularly creep active at early ages. We here perform an ageing creep test on a cement paste sample exhibiting an initial water-tocement mass ratio amounting to w/c = 0.50. The sample is conditioned to 20°C. The creep test is started 24 hours after production and runs for 48 hours. The loading amounts to 15% of the uniaxial compressive strength of the material at the time instant of loading. Inspired by a recently developed new testing protocol consisting of hourly-repeated three-minutes long creep tests [6], we exploit force and deformation readings recorded during the first three minutes of our 2-days creep test, i.e. we identify Young’s modulus and power-law creep properties. Given that the chemical hydration process does not progress significantly during the analyzed three minutes, the identified creep properties refer to non-ageing creep. The identified properties are combined with measurements of autogenous shrinkage, in order to predict the strain evolution over the entire 2-days runtime of our ageing creep test. When fitting of the creep function targets, through an appropriately chosen weight functions, the end of three minutes creep interval, in terms of both absolute creep strain and creep strain rate, then the resulting creep function remarkably well predicts creep up to a time period of 2 days.

The mutual effect between electric corrosion of reinforcing bars and concrete cracks development caused by corrosion is investigated experimentally and analytically. In the experiment, a salt-water pool was set on the concrete cover of each specimen containing single round rebar for the implementation of electric corrosion. Then the radius losses around the corroded rebars were measured for the corrosion patterns. It was shown that the corrosion rates at various positions around the rebar were related to the crack width nearby. In the analysis, a numerical model based on the rigid body spring method combined with a truss network is proposed to simulate the coupled process by assuming the relation of current efficiency to the crack width. Satisfied agreement with the test results of non-uniform corrosion process and associated crack propagation is obtained using this model, offering a good tool to assist in the employment of the electric corrosion method.

A network model for the time dependence of transport properties of porous materials is proposed. A statistical volume element (SVE) of a capillary pore network of porous materials is idealised by a discrete lattice model consisting of spheres and pipes representing pores and throats, respectively, which can be filled with either liquid water or gas. The time dependence of the transport properties is modelled by incrementally decreasing the radius of the water filled pipes and spheres accounting for the evolution of the microstructure. The microstructural model is applied to investigate the change of macroscopic transport properties due to microstructural changes.

A three-dimensional lattice approach for modelling the coupling of fracture and transport was applied to micro-cracking induced by aggregate restrained shrinkage in concrete, which was idealised as a three phase material consisting of matrix, aggregates and interfacial transition zones. Shrinkage of the matrix and interfacial transition zones was modelled by applying incrementally a uniform eigenstrain. The influence of aggregate diameter and specimen thickness on cracking and transport properties was investigated. The results show that increasing aggregate diameter at constant volume fraction increases the crack widths and, therefore, permeability, which confirms previously obtained 2D modelling results.

This paper investigates the effect of microcracks on transport properties of concrete using a three-dimensional model. Concrete is idealised as consisting of aggregate particles (1-10 mm, 60% vol.), cement paste matrix and microcracks with widths of 1-50 µm. Using aligned meshing, aggregate particles are discretised with tetrahedral elements and the microcracks are incorporated as interface elements. The microcracks are either bond cracks at the aggregate-paste interface or matrix cracks that span nearest neighbouring aggregate particles. A finite-element model is applied to simulate diffusion and permeation. The model is used to perform a sensitivity analysis to examine the effect of microcrack width, density and percolation. It is found that the effect of microcracks is more pronounced for concretes with denser matrix. Furthermore, the effect of microcracks in non-percolated networks increases up to a finite limit, the value of which is independent of crack width but depends on other crack characteristics and transport property of the matrix. In all cases, microcracking has a greater effect on permeation than diffusion and this is more pronounced for percolated crack networks.

Cracking of concrete, which may induced by environmental action or mechanical loading, has been identified as one of the major factors controlling the long-term performance of concrete structures because it can provide preferential pathways for the ingress agents (e.g. water or chloride ions). In the current work, the transport coefficients of water and chloride through a single crack are developed by treating it as a parallel-plate configuration. The physical migration models coupled with capillary absorption and convection-diffusion are utilized to predict water penetration and chloride ingress into cracked concrete. By means of the mesoscale computational method that concrete is modeled as a composite of impermeable coarse aggregate embedded in the porous matrix separated by vulnerable interfacial transition zone, the lattice network model is presented to investigate water and chloride transport properties in cracked concrete. The influence of cracking and the actual saturation level on the transport process of water and chlorides are investigated numerically. The numerical results indicated that the computational method is able to well represent the mass movement within cracked concrete. The effect of saturation level and cracks on water and chloride transport process is of great importance.

An overview of a recently developed thermo-hygro-mechanical-hydration model (THMH) model for concrete is presented. The coupled flow aspect of the model solves a mass balance equation for moisture and an enthalpy balance equation for heat transport. A key assumption used to simplify the moisture flow component of the model is that the gas pressure is assumed to remain constant at atmospheric pressure. The model simulates early age concrete behaviour and therefore a hydration component is included in the formulation. This uses the approach of Schindler & Folliard, which was based on a comprehensive range of data and gives good predictions for a range of cement types. The mechanical component of the model simulates the dependence of the strength and stiffness properties of the model on the degree of hydration. Furthermore, the plastic-damage-contact model of Jefferson and Mihai; which simulates cracking, crushing and crack closure; has been extended to include hydration dependent behaviour. The hydration sub-model is also linked to a new creep model. The paper includes an example of the analysis of beam RL1 from the Concrack benchmark programme. The focus of the work from this benchmark is the accurate prediction of crack widths in reinforced concrete elements and reference is made to a recent study that explored the accuracy of the present model in this context. A critique of the model is presented which concludes that the moisture flow component of the model would be greatly improved by including coupling between the flow parameters and mechanical damage.

Strain hardening cementitious composites are a class of cement based materials which show strain hardening behaviour in tension. This is achieved by multiple microcracking, which results in a tightly spaced crack pattern with relatively small crack widths (50-80μm, in general) and high strain capacity (up to 4-5%). Due to their ductile behaviour and tight crack widths, SHCCs are commonly used for concrete repair applications. However, due to the tight crack width and crack spacing, moisture uptake by capillary suction can take place very fast. This could result in rapid access of deleterious substances, such as chloride ions, resulting in corrosion initiation. In this study, X-ray tomography is used for monitoring and quantification of water uptake in SHCC. Specimens were first loaded to different strain levels in uniaxial tension. They are subsequently subjected to a capillary suction test. The performed test was subsequently modelled using a lattice model.

In this work, we analyzed x-ray microtomographic images to quantify the porosity of small mortar specimens, with a particular focus on the porosity of the interfacial transition zone (ITZ). Specimens were nominally 5 mm diameter, 4 mm long cylinders with 0.5 mm diameter glass bead aggregates. Specimens were scanned via synchrotron-based x-ray microtomography while they were positioned in an in situ loading frame in a split cylinder configuration. Scans of undamaged specimens were evaluated for porosity both in the bulk paste and in the ITZ. Specifically, voids in the paste and porosity in the ITZ were superimposed onto a map of the tensile stress in the specimen in an attempt to identify critical flaws, and to measure their role in split cylinder strength. Preliminary results indicate that the ITZ porosity has a lower effect on split cylinder strength than the large flaws that occur either in the ITZ or the bulk cement past.

Coupled elasticity and potential flow problems, such as drying shrinkage, can be solved using lattice models. Typically, both the displacement and relative humidity fields are represented by the same set of nodes. When considering crack development and its influences on transport, however, it may be advantageous to use a different lattice for each field. In this research, the structural problem is to be solved using a lattice defined by the edges of the Delaunay tessellation of a set of nodal points. The edges of the dual Voronoi diagram are used to solve the flow problem. In this way, flow elements are better positioned to simulate transport along cracks, the effects of crack opening on such flow, and transport from the crack faces into the bulk material. A complicating factor is the inevitable occurrence of arbitrarily small and/or poorly-shaped geometric objects associated with the Voronoi diagram. These features can both degrade the physical fidelity of the simulation results, and negatively impact the reliability of the overall simulation process. This research introduces the concept of automated tolerance-driven domain discretization, by which undesirable features of the Voronoi diagram are avoided at the mesh-generation stage. Corresponding improvements in the ability to reliably model flow in homogeneous and heterogeneous solids are demonstrated. Potential improvements in the modeling of fracture and flow through fractured media are discussed.

Differential shrinkage between repair material and concrete substrate is considered to be the main cause of premature failure of repair systems. Magnitude of induced stresses depends on many factors, for example the degree of restraint, moisture gradients caused by curing and drying conditions, type of repair material, etc. Numerical simulations combined with experimental observations can be of great use when determining the influence of these parameters on the performance of a repair system. In this work, a lattice type model is used first to simulate moisture transport inside a repair system and then to model time dependent damage. 3D simulations are performed, and damage patterns and crack widths are qualitatively verified with some experimental observations.

Cover cracking and spalling in concrete due to corrosion of reinforcement bars is one of the major concerns for durability of reinforced concrete structures and has been widely researched during recent years. Most approaches to the problem are based on a uniform corrosion and expansion pressure around the rebar. However, corrosion rust tends to accumulate around the steel circumference that faces the concrete cover. From this outer part of the rebar, the corrosion front gradually advances to the inner, which entails a non-uniform expansive pressure around the rebar. The purpose of the study is to simulate the effect of non-uniform rust distribution around the corroded rebar, taking advantage of an embedded cohesive crack finite element. The objective is to develop more realistic models for the estimation of the service life of reinforced concrete structures.

Corrosion has known as a major problem on deterioration of RC structures because it may induce cracking of concrete cover as well as degradation of structural performance. Most of studies on the corrosion cracking have used a single bar in a concrete prism or a cylinder and absence of stirrup bars, which may not represent the real boundary condition of reinforced concrete members. The main objective of this paper is to investigate the influence of stirrup confinement, longitudinal bar diameter, concrete strength and corrosion rate on the cover crack propagation. Six beams were designed to experience a specified corrosion by performing an accelerated corrosion. The current flowed on each bar and the strains on the stirrups were recorded using data logger. The corrosion-crack behaviors were frequently observed using digital microscope. Furthermore, finite element analysis was performed to study the internal cracking behavior of RC beams and the structural behavior of corroded RC columns.

In the context of a project to characterize and model the Alkali-Silica Reaction in concrete, the results of a series of tests on specimens with a single interface between aggregate and cement paste, and the comparison to the numerical results of a simplified diffusion-reaction Finite Element model are presented. The tests are conceived as the simplest test to characterize the basic constituent reactions of the process. The model implemented is a first operational version of a more general three-species diffusion-reaction model already formulated and in the process of implementation. In spite of its relative simplicity, the model described seems capable of reproducing the experimental observations with specimens tested under different diffusion boundary conditions, and provides a satisfactory basis for further development.

Delayed Ettringite formation (DEF) is an autogenous expansive reaction that can affect concrete. A long enough exposure to high enough temperature are the necessary conditions to develop DEF. The results of experimental laboratory investigations that aim to quantify the effect of thermal history on DEF characteristics (namely magnitude and kinetics of expansion) are presented. A threshold of temperature for the concrete at early age, a pessimum effect with the heating duration and a relation between thermal energy and swelling parameters (kinetics and magnitude) are highlighted.

Concrete degradation is analyzed due to alkali silica reaction (ASR) in combination with creep and shrinkage. Environmental factors such as temperature and relative humidity are included in this study since they accelerate hydration and aging of cement-based materials due to the underlying transport of multiple chemical reaction processes. This study examines the performance of concrete at the meso-scale level of observation where aggregate particles and the embedding hydrated cement paste form interacting continua. Macro-scale numerical analyses are compared with corresponding meso-scale ones, in order to capture the chemo-physical damage and understand the role of the different concrete constituents. For this purpose a 3D fully coupled thermo-hygro-mechanical finite element code developed at Padua University, which treats concrete as a multi-phase material, has been adopted and updated to take into account the effects of ASR expansion in the interface transition zone, the aggregate particles and the hydrated cement based mortar.

Internal cracking tend to arise from the edge of knots present in deformed bars embedded in tension zone of reinforced concrete structure when loaded up to 100 N/mm². This could be further exacerbated by drying shrinkage depending on types of aggregate used in concrete mixtures under dry environment. The objective of this study was to examine the durability performance of internally cracked RC beam specimens cast using aggregate including crushed limestone and sandstone with different absorption capacity obtained in regional resources. Based on accelerated chloride-induced corrosion tests carried out via a cycle of 3-day wet and 11-day dry until the age of 1 year, although corrosion tended to occur severely in the cracked parts with higher chloride content depending on the drying shrinkage of concrete specimens, the results showed that internal cracking could trigger localized corrosion initiated especially at the knots of deformed steel bar.

The motivation of this contribution is to apply computational diffusion homogenization to the microscale and mesoscale of concrete sequentially, while taking into account both physical and chemical bindings. The effective diffusivity of cement paste obtained by homogenization at the microscale is considered as the input for diffusion homogenization at the mesoscale, thereby establishing the link between three length scales of concrete. Comparisons between numerical results of each step of the diffusion homogenization and experimental data from the literature are reported and discussed.

Durability of the concrete depends on the mass transfer properties of the concrete. This paper investigates the chloride diffusion in polymer-concrete composites subjected to pure diffusion process for 720 days. Chloride profiles for all types of concrete were obtained and then chloride diffusion coefficients were calculated. Extensive experimental program showed a significant enhancement of concrete durability by means of polymer-concrete composites. Reducing numbers and size of cracks and blocking the interconnected pores in concrete can enhance the durability of concrete. The results of this experimental study revealed that utilising both synthetic fibres and polymers in concrete improved durability of the concrete significantly.

The present paper gives an overview of creep data of UHPC and an appropriate creep function based on this data which was used to compute age-dependent relaxation spectra, i.e. Maxwell chain parameters. The relaxation spectra computed was validated with the help of cracking frame tests. The numerically obtained relaxation spectra enables mechanical studies of members made from UHPC taking into account the agedependent viscoelastic behavior.

Understanding the physical origin of creep in calcium–silicate–hydrate (C–S–H) is of primary importance, both for fundamental and practical interest. Here, we present a new method, based on molecular dynamics simulation, allowing us to simulate the long-term visco-elastic deformations of C–S–H. Under a given shear stress, C–S–H features a gradually increasing shear strain, which follows a logarithmic law. The computed creep modulus is found to be independent of the shear stress applied and is in excellent agreement with nanoindentation measurements, as extrapolated to zero porosity.

Calcium Silicate Hydrate gel (C-S-H) is the basic product of the cement hydration process responsible for strength and durability of concrete. Synthetic C-S-H was synthesized by mixing Calcium oxide (CaO denoted as C) and silicate (SiO₂ denoted as S) with a C/S mixture ratio of 1.5. C-S-H powder was filtered, and dried to a relative humidity of 11% and then compacted at 400 MPa. Nanoindentation tests incorporating time-dependent effect (creep) were performed on polished C-S-H specimens using Berkovich indenter tip. The reduced elastic modulus and creep compliance of C-S-H specimens were extracted from nanoindentation measurements. An innovative approach is suggested to extract cracking energy. A viscoelastic-plastic finite element model was developed to simulate the nanoindentation of C-S-H incorporating creep and cracking. The finite element model was used to isolate the viscoelastic from the cracking energy components. The proposed approach enabled estimating fracture toughness of C-S-H.

A wide variety of fluid systems including organic liquids, mineral melts, colloidal suspensions and granular materials, undergo a dynamical arrest transition when the temperature is lowered, or the density is increased, or the applied stress is decreased. This is known as the glass transition in the first case and as the jamming transition in two latter cases. The purpose of this paper is to show that the coagulation and the setting of cement pastes share with these systems several key attributes which bring them under the same umbrella.

This work focuses on the creep of clay-based materials, which exhibit significant analogies with cement-based materials. Here, we studied the creep of clay at two scales and with two techniques: numerically (with molecular simulations) at the scale of a clay layer (nm), and experimentally at the scale of thin self-standing clay films (few dozen μm). At the scale of the clay layer, numerical simulations showed that the shear rate was constant over time and an affine function of the shear stress. Creep experiments showed that, after a transient period, the creep function of our thin self-standing clay films was a logarithmic function of time. A comparison of the results obtained at the two scales shows that the origin of the logarithmic feature of clay creep must at least partly originate from a scale greater than that of an individual clay layer. By analogy, such result is likely to hold for cementitious materials, which are also known to creep logarithmically with respect to time in the long term: the origin of this logarithmic feature is likely to stem at least partly from a scale greater than the scale of an individual C-S-H layer.

Long-term performance of aging concrete in nuclear power plants (NPPs) requires a careful examination of the physical phenomena taking place in the material. Concrete under high neutron irradiation is subjected to large irreversible deformations as well as mechanical damage, caused by a swelling of the aggregates. Mesoscale numerical simulations are performed to separate the underlying mechanisms and their interactions. The cement paste creep-damage model accounts for the effect of the loading rate on the apparent damage properties of the material and uses an event-based approach to capture the competition between creep and damage. The model is applied to the simulation of irradiation experiments from the literature and shows a good agreement with the experimental data.

Current advances in micro and nano X-ray tomography allow to image multiscale porous materials with spatial resolutions up to about 10 nm. This is particularly the case of natural or industrial geo-materials like sedimentary rocks, shale or cement pastes. These experiments reveal complex multiscale morphologies and raise the question to couple these various length scales with some functional properties such as the transport of gas or liquid in strong confinement. In this conference, we focus on the diffusive transport of molecule in porous media where adsorption and geometrical confinement at different time and length scales are observed. We propose some possible ways to quantitatively decouple the fluid confinement within “elementary pores”, involving adsorption, intermittent dynamics, and the long distance exploration, sensitive to the porous network connection.

Novel fabrication techniques for non-standard concrete structures rely on the interplay of robotic fabrication with the evolution of the material performance, mainly during early hydration. When the material transforms from a yield stress fluid to a cohesive frictional material, detailed knowledge on the evolution of the strength envelope becomes essential. This is even truer, as the dead weight acting on the previously fabricated zones increases with the size of the fabricated structures. A set of non-standard mechanical tests for compression, tension and shear, particularly suited for measuring failure from early hydration states on, exhibits two distinct scaling regimes. The first one is characteristic for large local plastic deformations, followed by the transition to the second regime with material behavior dominated by crack growth. Finally we condense all test results to a limit surface in principal stress space and discuss its evolution with the advancing state of hydration.

The logarithmic deviatoric creep of cement paste is a technical and scientific challenge. Transition State Theory (TST) indicates that some nanoscale mechanisms of shear deformation, associated with a specific kind of strain hardening, can explain the type of deviatoric creep observed experimentally in mature cement pastes. To test this possible explanation, we simulate the shear deformations of a colloidal model of cement hydrates at the nanoscale. Results from quasi-static simulations indicate a strain hardening analogous to that postulated by the TST approach. Additional results from oscillatory shear (fatigue) simulations show an increase of deformation with number of loading cycles that is consistent with the observed creep. These findings indicate that nanoscale simulations can improve our current understanding of the mechanisms underlying creep, with potential to go beyond the logarithmic creep and explore the onset of failure during tertiary creep.

Calcium-silicate hydrate (C-S-H) is the main binder in cement and concrete. It starts forming from the early stages of cement hydration and it progressively densifies as cement sets. C-S-H nanoscale building blocks form a cohesive gel, whose structure and mechanics are still poorly understood, in spite of its practical importance. Here we review a statistical physics approach recently developed, which allows us to investigate the C-S-H gel formation under the out-of-equilibrium conditions typical of cement hydration. Our approach is based on colloidal particles, precipitating in the pore solution and interacting with effective forces associated to the ionic environment. We present the evolution of the space filling of C-S-H with different particle interactions and compare them with experimental data at different lime concentrations. Moreover, we discuss the structural features of C-S-H in the mesoscale in terms of the scattering intensity. The comparison of our early stage C-S-H structures with small angle neutron scattering (SANS) experiments shows that long range spatial correlations and structural heterogeneties that develop in that early stages of hydration persist also in the hardened paste.

C-S-H gel densifies progressively during hydration. Consideering that the available precipitation space may be the key parameter driving gel densification, we here use NMR relaxometry data in combination with basic stoichiometric relations of hydration chemistry, in order to develop a model for the evolution of C-S-H gel density as well as the phase volume fractions of hydrating cement pastes. Then, we combine the hydration model with a multiscale microstructural representation of cement paste. Considering full anisotropy of solid C-S-H crystals, we are able to predict the experimentally determined stiffness increase during proceeding hydration.

Cyclic water freezing/thawing in concrete is a major cause of civil engineering infrastructure deterioration in cold climates. Ice exerts some additional pressure on the skeleton, causing its damage, which is dependent on the material microstructure. The total open porosity, pore size distribution, contribution of ink-bottle pores, capillary suction coefficient and strength properties were experimentally investigated for the virgin cement mortar and after 25 and 50 freezing cycles. The results indicate that the narrowest throats between larger pores are broken by frost damage, causing increase of the material permeability. A mathematical model of coupled heat and water transport in deformable, fully saturated porous materials, accounting for the kinetics of water/ice phase change, is developed and validated experimentally. The kinetics of water freezing - ice thawing is described with non-equilibrium Thermodynamics. The average crystallization pressure is determined for the measured parameters of the frost-deteriorated microstructure of cement mortar. The model equations are solved numerically with FE and FD methods. Considering the evolution of material microstructure allows for modeling the progress of frost induced damage in cementitious materials at low temperatures.

Delayed Ettringite Formation is a type of chemical degradation of moist concrete exposed at early ages to heating up to temperature of about 70-80°C. The material deterioration is caused by expanding phases produced by the ettringite reaction. A phenomenological, rate-type model of the DEF strains, considering effect of the temperature and duration of heat-treatment of concrete at early ages, and the evolution of ambient temperature and relative humidity, on the development of strains is introduced into the mathematical model of heat and moisture transport in deformable porous media. The model equations are solved numerically by means of finite element, finite difference and Newton-Raphson methods. The computer code was developed and used for the model validation by comparing its’ results with the published experimental data.

A quantitative influence of the water migration in pores of a cement-based material on the creep phenomenon was estimated in a viewpoint of micromechanics and then the effect of the pore water pressure as a disjoining pressure, the friction resistance between water and pore walls due to the water migration and the viscoelastic behavior of the solid phase on the creep phenomenon were mutually discussed. At first, the coupled model of the liquid and solid phase of a cement-based material as a porous permeable material developed by authors, which simultaneously satisfies the force equilibrium equation and the mass conservation law of pore water, was expanded so as to be taken the diffusion phenomenon and seepage phenomenon for pore water and the viscoelastic behavior of the solid phase into account. Secondary, an experimental apparatus, which can reproduce the creep phenomenon due to the water migration in micro pores, was developed.

In the light of the developing use of cementitious materials in safety critical, high temperature applications associated with energy industries (nuclear, oil and gas) this work looks to the nano-scale to consider the origins of macro-scale phenomena. It enables a first comparison of nano-scale results with macro-scale observations, toward the inclusion of physico-chemico-mechancial processes in models at larger scales and a move away from phenomenological models. Assuming the development of nano-scale porosity to be the principal effect of temperature at the nano-scale, comparisons are here developed between nano-scale measurements, simulations and macro-scale experimental results. The results suggest that the effect of temperature on the nano/micro mechanical properties of cementitious materials might be responsible for a significant part of the experimentally observed trends at the macro-scale, although more work is required to understand scaling of fracturing. It is concluded that modifying the nano-scale material response to temperature gradients could eventually impact the engineering performance of structures at elevated temperatures.

This work aims at putting in evidence the coupled effects of both delayed deformations of concrete and relaxation of prestressing steel on the long term behavior of long span bridges. After recalling the possible origins of delayed deformations of prestressed concrete structures, the example of the Savines bridge is presented. This 65-year-old bridge shows excessive deflections at mid-spans which can not be predicted by the current Eurocode 2 creep laws, unless artificial coefficients are adjusted. Then a simplified prestressed cantilever beam is considered allowing to quasi-analytically address the problem of interaction between concrete creep and steel relaxation. The resolution of this problem with ageing material laws which are adapted from Eurocode 2 highlights the importance of the evolution of steel tendons after decades and allows to question the validity of laws used in models and the difficulty to identify the contributions of concrete and steel in the evolution of the deflection.

The typical French reactor building consists of two concentric containments. The outer containment is made with reinforced concrete. The inner containment is biaxially prestressed, from 80 to 120 cm thick, and designed to withstand an internal pressure of 0.5 MPa (absolute) in case of an accident. In order to avoid tensile stresses in concrete, a prestress is applied, corresponding to compressive stresses in concrete of about 8.5 MPa and 12 MPa along vertical and tangential directions, respectively. In order to maintain enough prestressing in the structure, the delayed behavior of concrete on the very long term should be correctly assessed. The measurements of the very long term deformations of larges bridges and of nuclear containments seem to indicate that, on the very long term, creep strains are function of the logarithm of time. In particular, the measurements of delayed strains in the containment wall show two distinct kinetics: (1) the difference between tangential and vertical strains, which exhibits a fast kinetics and becomes almost constant after 500 days, (2) the average of tangential and vertical strains undergoes a slow kinetics. If vertical and tangential strains rates are equal (despite different values of the prestressing), this means that the long term biaxial creep deformations are only sensible to the mean stress and independent of deviatoric ones. In the recent Model Code 2010 creep is expressed as a combination of an asymptotic and a logarithmic function. The logarithmic function corresponds to basic creep while drying creep is asymptotic. Using this model for delayed strains, and adding the possibility to adjust the parameters of the relations proposed by MC2010 on experimental results, the evolution of the strains is calculated and compared to in situ measurements. We finally discuss the set of adjusted possible parameters.

The balanced lift method for bridge construction proposes to build the bridge girders in a vertical position and to rotate them into the final horizontal position with the aid of compression struts. During the construction process, when the bridge girders are rotated from the initial vertical position into the final horizontal position, it is of utmost importance for an economic application of the balanced lift method that the structural members are as light as possible. In a test beam (length = 30.02 m, height = 1.44 m, width = 0.7 m) the actual filling process with concrete was carried out. Subsequently, the longitudinal strains on the outside of the beam were monitored over a period of four years in order to investigate the creep behaviour. Due to creep, the initially high stresses in the prefabricated outer shell are gradually reduced and the in-fill concrete is subjected to axial compressive stresses.

Evaluation of the internal stress will be a key for understanding the cracking potential of concrete structures. One of the reasons for generation of internal stress in concrete is the volume change of cementitious material such as shrinkage and creep in the vicinity of aggregates and reinforcements in concrete. To overcome the complexity of concrete as a heterogeneous material, stress generation needs to be evaluated by the effect of difference in stiffness between materials. The objective of this research is to evaluate the internal stress of concrete in an early age through the measurement of strain. Image analysis will be helpful in clarifying the development and the quantity of strain. In order to have the various potential of internal stress, specimens with different restrained conditions were prepared in the experiment. Furthermore, the difference of cracking load was explained by the development of internal stress with age of concrete.

The efficient and permanently safe design of fastening systems requires a thorough understanding of all involved load carrying mechanisms and their potential change in course of time due to deterioration and material aging. The latter are determined by the involved materials—concrete, steel, and, depending on the system, the mortars. In this contribution we focus on concrete cone failure, the critical failure mechanism for bonded as well as mechanical anchors under pre-dominant tension loading. Both mechanical as well as chemical anchors undergo an approval process during which their performance is certified. These tests are performed according to strict guidelines and for concretes of a specified composition and age range. Nonetheless, differences between laboratories, the age of concrete at testing, the storage conditions, the coarse aggregate type exist. In this contribution we attempt to quantify the effect of testing age in combination with storage conditions on obtained pull-out loads. We will present the numerical framework that is able to model the coupled problem of heatand moisture transport, hydration, aging of material properties, shrinkage and creep in a rate–type, point–wise form. The framework will be calibrated based on an experimental study of creep and shrinkage on a typical C25/30. The mechanical model is calibrated and validated on standard tests for uniaxial compression, three point bending as well as pull-out tests. Afterwards, the calibrated framework is applied in a purely predictive way to fastenings under sustained loads.

The paper deals with drying and shrinkage of concrete specimens. Drying of concrete was measured first on small concrete cubes (edge length 200 mm), which were sealed on all with exception of one side (top side). The pore relative humidity was measured in different depth from the surface exposed to the environment (the top side, which were not sealed). The pore humidity is recorded at the moment for a period of about 14 months. After a good experience with the measurements, the experimental program was extended. The segments of walls of the thickness 200 mm, 400 mm and 800 mm were cast. Four sides were sealed in order to simulate the drying of the infinite walls of different thicknesses. The relative pore humidity is measured in different distances from the surface of individual walls. The elements are equipped with vibrating wire strain gauges which measure strains inside the elements. The strains in vertical as well as in horizontal directions are recorded. Additional tests are executed. The concrete compression strength is measured on cubes and uniaxial shrinkage is measured on standard cylinders using embedded vibrating wire strain gauges. The cylinders are stored in both, in laboratory conditions and in water.

The purpose of this study is to clarify shrinkage of full-scale wall made with Blast Furnace Slag Fine Aggregate (BFS). Authors investigated the influence of BFS on strength properties, drying shrinkage, uni-axial restraint stress, tensile creep and drying shrinkage cracking of full-scale wall. Following findings were obtained. The concretes with BFS had high shrinkage cracking resistance. The shrinkage stress could be calculated with incremental effective young’s modulus method. This analytical approach also indicated that the shrinkage cracking resistance of concrete with BFS was good enough.

In this study, we investigated the drying shrinkage and fracture energy of concrete with lime stone and other aggregates and their effect on the shear strength of RC beams. The results of material experiments showed that the use of the lime stone aggregate having large aggregate crushing value decreased the fracture energy and drying shrinkage compared to the other aggregates. The results of structural experiments showed that the measured shear strength of RC beam with the lime stone aggregate was smaller than that with rhyolite aggregate, while it reduced internal stress of RC beams caused by the shrinkage of concrete. It was suggested that the decrease in the shear strength was caused by the small fracture energy and characteristic length of the concrete with the lime stone aggregate.

The paper is directed to the cross section warping effect and to the warping induced long term prestress loss due to creep of concrete, which is analyzed deterministically and stochastically. The statistical variability of material creep parameters and environmental factors has a significant effect on the statistical scatter of the calculated prestress loss. In order to obtain the real development of concrete creep it is important to use a model with strong theoretical basis. Sophisticated model B4 provides very accurate concrete creep and shrinkage prediction. The statistical randomness of boundary and environmental conditions is solved by application of Monte Carlo — Latin hypercube sampling method. It is shown that the severity of cross section warping can be of fundamental importance, particularly in the case of unevenly distributed and isolated tendons in cross-section and in the case of short spans. Both the 3D effects and the statistical variability of creep must be taken into account if a reliable design of box girder bridges from the viewpoint of long-term serviceability should be achieved.

A robust advanced model for creep and shrinkage of fibre reinforced concrete is presented; the influence of various types of fibres, their shapes and bonding to the matrix is respected in the models. The intention is to present a method for better understanding rheological performance of fibre reinforced concrete, to improve conventional calculation methods and to offer suitable design tool for practice. The prediction models will create instruments enabling efficiency of materials and offer objective and effective tools increasing in the same time the level of reliability, durability and efficiency of fibre reinforced concrete structural members.

The restrained ring test has been used to assess the potential for early-age cracking of concrete and other cement-based materials. Recently, a novel elliptical ring test method is proposed to replace the circular ring test method for the purpose of shortening test duration and more conveniently examining crack evolution in restrained concrete rings. In order to explore the mechanism of this test method, a numerical model is developed to analyze crack propagation process in restrained concrete rings, in which the effect of concrete shrinkage is simulated by a fictitious temperature drop applied on concrete causing the same strain as that induced by shrinkage. Finite element analysis is carried out to simulate the evolution of SIF in restrained concrete rings subject to circumferential drying. It is found that the failure mechanism of concrete rings changed with the increase of concrete ring wall thickness. Elliptical rings with appropriate geometry can enable crack initiation much earlier than circular rings in the case that crack is caused by the restraint of the central steel ring.

The Temperature-Stress Testing Machine (TSTM) System at NTNU is well established, but the software for logging and management is new. The degree of restraint can now be varied in the range between 0 and 100%, and thus the real temperature- and stress history of critical positions in large concrete structures can be experimentally investigated. Introductory tests, calculations and evaluations are on-going, and the test equipment gives reliable results and good reproducibility. The verification of the TSTM System includes comparisons of test results with three different calculations approaches: Excel, CrackTeSt COIN and DIANA. Tests conducted in the TSTM show that the restrained stress development is reduced when increasing the concrete fly ash content. However, this has to be considered in combination with the simultaneously reduced tensile strength. The performed restrained stress tests and appurtenant calculations show that good knowledge regarding the creep/relaxation behavior of the concrete in question is of major importance. In addition, the obtained results indicate a correlation between temperature history and the magnitude of the E-modulus for fly ash concretes. A mechanical test series of concretes with varying fly ash content has been carried out, and the determined material properties: E-modulus, tensile strength and compressive strength, were all found to be decreasing with an increasing amount of fly ash content. The described activities have been a part of the COIN-project «Crack free concrete structures».

The design of structures is more and more directed towards the entire lifetime design with multiple concurrent objectives. Apart from durability, the most important factor in the whole life design of reinforced and, in particular, prestressed concrete bridges, is the Service Limit State. From this point of view, prestressed concrete bridges are very sensitive to long-term increase of deflections. A survey of many bridges monitored in various countries showed that all of them have experienced similar deflection histories. The intention of the paper is to describe methods and mathematical models which should be used for reliable prediction of long-term behaviour of prestressed concrete bridges. The comparisons between results of numerical simulations and measurement data will be also mentioned.

The present study includes the experimental investigation on the structural mechanism of reinforced concrete (RC) beams using high strength reactive powder concrete (RPC) and containing high-strength steel short fibers. The steel fiber reinforced high-strength RPC beams was design lightly reinforced with shear reinforcement but to still fail in shear in the loading tests. The parameter was set as shear reinforcement ratio (0, 0.3%, and 0.6%), volume fraction of fiber (0, 1%, and 2%) and volume change due to mainly shrinkage which was controlled by the water dissipation during the curing with drying or sealed sequences until the loading test for the beams. The strains of the concrete and reinforcing bars in the RC beams and the strain in the reference specimen were measured during the curing. In the loading test, an image analysis using a digital image correlation method was carried out in order to detect the micro- and macro-cracks in the RC beams. We have confirmed from the loading experiments that the increase in the shear capacity of the RC beams by containing the shear reinforcement and steel fibers even when the amount of shear reinforcement was set close to the minimum requirement. The interval of the diagonal cracks in the RC beams at the ultimate stage was changed according to the amount of the shear reinforcement and the volume fraction of fiber. In addition, the results of loading tests indicate that the initiation and propagation of a diagonal crack in the RC beams was changed and finally the shear capacity of the beams was influenced by the parameters, especially in the case of high shrinkage condition.

During the period 1967-1985 the Magnel Laboratory for Concrete Research participated in an extensive Belgian research campaign with respect to the influence of creep and shrinkage on the long-term behaviour of reinforced and prestressed concrete beams. This research campaign, jointly conducted at several Belgian research institutes, comprised the investigation of concrete and reinforced concrete beams (phase 1), prestressed concrete beams (phase 2) and partially prestressed concrete beams (phase 3). The main aim of the research campaign was the determination of the long-term behaviour subjected to permanent loads, considering the influence of the magnitude of the loads, different reinforcement ratios and/or prestressing degrees and/or different cross-sectional shapes. These results were obtained by a joint collaboration of 4 Belgian research institutes, each focussing on a different reinforcement ratio and reinforcement arrangement. With respect to the reinforced concrete beams (phase 1), at each institute 12 beams were tested in a 4-point bending configuration, namely 2 static tests at 28 days and 10 long-term tests with a duration of 2 to 4 years, considering different loading levels. In this contribution some results of the reinforced concrete beams (phase 1) will be documented and analysed, comprising the results obtained on 48 reinforced beam specimens with a length of 3.4 m (span of 2.8 m) and cross-section of 0.28 m x 0.15 m. A cross-sectional calculation tool developed at our department — incorporating the current creep and shrinkage models in standards and guidelines — will be employed in order to investigate the accuracy of the available models with respect to their ability to predict the structural behaviour of the documented reinforced concrete beams.

To properly identify anomalous data in a structural monitoring system, the expected behavior of the structure under changing environmental conditions must be accurately determined. The St. Anthony Falls Bridge, a post-tensioned concrete box girder bridge, was examined as a case study for testing a data normalization scheme for temperature- and time-dependent deformations of an in situ structure. A methodology based on linear regression was developed to extract time-dependent behavior from the measured data. An Arrhenius-adjusted time was proposed to correct for the slowing and accelerating creep and shrinkage rates during the winter and summer, respectively. The extracted time-dependent deformations were compared to estimates using the finite element method. Bayesian regression was used to update the finite element model predictions using the measured data, substantially narrowing the expected bounds and facilitating anomaly detection.

The efficient and permanently safe design of fastening systems requires a thorough understanding of all involved load-carrying mechanisms and their potential change in the course of time, due to deterioration and material ageing. The latter effects are mainly ruled by the involved materials: concrete, steel, and, depending on the system, the mortars. Considering the required extrapolation in time that is necessary for e.g. a 50 year design life-time and the required low failure probabilities, especially for the ultimate limit case, a suitable framework for prediction and assessment as well as accurate modelling concepts for all involved elements and processes with a quantification of the associated modelling and prediction uncertainties are paramount. Within this contribution we will present the framework for predicting the life-time of a (fastening) system subjected to a sustained load as well as extreme events following a stochastic process. For this formulation it is clear that any damage resulting from an early occurrence of the extreme event will influence the behaviour under sustained load. Conversely, the behaviour during an extreme event will be influenced by the sustained load effect up to this point in time. The stochastic ageing nature of concrete, which is the source of changes in mechanical properties, further complicates the analysis.

As structural concrete is a continuously-developing material, evolution in the estimation of its material properties can be seen with the development of concrete technology. The paper reviews methods for predicting concrete creep in bridge structures. The principal variables that influence creep are outlined. Code-related procedures have been drawn up in recently-used models for concrete creep with a higher range of concrete grades. This enables a more realistic estimation of delayed deformations and prestress losses in bridge structures.

Among the approximately 5,000 bridges which are maintained by the infrastructure operator ASFiNAG a total number of 1,500 objects are overpasses. Particularly their substructures are subject to increased chloride contamination due to the application of de-icing salt on the road. Sooner or later this results in damage to the infrastructure and, hence, in a reduced service life. This paper provides a critical discussion of data collected from an on-site monitoring and from weather stations. It reveals important interrelations and dependencies to the corrosion rate. The monitored time variant corrosion rate indicates a significant mathematical detectable periodicity. In particular, the paper focuses on the interactions between environmental loads and corrosion rates of the reinforcement of concrete columns. Finally a temperature compensated empirically model for prediction of the corrosion rate by using environmental and monitoring data will be proposed.

Chloride transport mechanisms into road structures are multifarious. Because the current level of knowledge is not sufficient to model all relevant transport mechanism, the approach chosen to model chloride ingress should capture the decisive mechanisms involved. Based on this idea, Fick’s 2^nd law of diffusion has been selected as the basic physical approach. With this approach the apparent diffusion coefficient is one of the key variables. To improve the accuracy of the approach this variable has been adapted empirically. As for concrete exposed to intermittent splash water the profile may have a maximum content inside, Fick’s 2^nd law of diffusion cannot be applied without restrictions. The two published approaches regarding this issue will be discussed and assessed. For the selected approach, the empirical adaptation of the apparent chloride diffusion coefficient has been carried out by evaluation of chloride profiles. The regression variables derived have been quantified stochastically to enable a full-probabilistic calculation of the predicted chloride profile. Finally, an example calculation has been carried out to demonstrate the application of this model within the framework of service life design.

Corrosion of reinforcing steel in concrete is on one hand a major security problem and on the other a cost driver, which is very often not appropriately considered. Corrosion induced by chloride leads to localized corrosion phenomena, which reduces the cross section rapidly. Already corroded steels have to be replaced. When the initiation of the corrosion can be predicted accurately, the reinforcing steels can be prevented from corrosion and a comparatively small rehabilitation measure is sufficient. The cost for rehabilitation, which is necessary approximately every 20 years, can be reduced to half the costs. For an approximate structure lifetime of 100 years this leads to enormous cost savings. Accurate measurements are the basis for understanding and defining the process of Cl-ingress into concrete structures. In this paper, the deficiencies of the standard methods are demonstrated and a new evaluation method is applied. The calibration process as well as real life applications on retaining walls are presented and discussed.

Cracking and transverse bars strain behaviors of reinforced concrete (RC) columns induced by expansive pressure of corrosion products around corroding reinforcement were investigated through an accelerated corrosion test. Ten full-scale specimens which were element of RC columns having a dimension of 600 x 600 x 300mm were prepared. To get closer to actual phenomenon of chloride attack, a different amount of current was given into the longitudinal and transverse bars in response to the diffusion progresses of chloride ions depending on their cover thickness. Variable factors were concrete strength, bar arrangement and cover thickness. Test results indicated that concrete strength affects reinforcing bar strain and surface crack width growth considerably. A corrosion product generated in high strength concrete with less pores does not diffuse and concentrates into a crack. A specimen of smaller cover thickness (20mm) increased surface crack width and transverse bars strain was small compared with basic cover thickness specimen (40mm). On the other hand, bar arrangement didn’t show a remarkable effect on cracking behaviors and strain of transverse bar.

Laser-induced breakdown spectroscopy (LIBS) is a combination of plasma generation on the sample surface by a high power laser pulse and optical emission spectroscopy (OES). It is a direct measurement method without an extensive sample preparation. Concrete is a multi phase system. With LIBS it is possible to distinguish the different phases and based on that to correlate the chloride content to the cement mass. This is done by scanning the surface with a resolution of up to 0.1 mm. The high measurement frequency of 100 Hz or even higher allows the in-situ visualization of the chloride content over the measured area. With a calibration a quantification of chloride concentration is possible. As an example the LIBS measurement on a 50 mm by 70 mm area with a resolution of 0.5 mm and the visualization of the chloride distribution takes only 10 minutes. Additionally it is possible to evaluate the carbonation depth from the same measurement. Examples of practical application to determine the chloride-to-cement-content on concrete samples are presented. The possibilities and the limitations of LIBS are discussed.

The paper presents the results of an experimental program related with the damage behavior of Recycled Aggregate Concrete (RAC) under uniaxial loading-unloading cycles. To this purpose, the effect of the percentage of recycled aggregates (RA) on the compressive mechanical properties of concrete was investigated. Eight mixes with C25/30 and C35/45 target class of compressive strength were prepared using normal or recycled coarse aggregates. For both series, a target S4 class of workability has been chosen. Cylindrical specimens were fabricated with different RCA replacement percentages of 0%, 30%, 100% fine and coarse RA and 100% of coarse RA. Uniaxial loading-unloading compression tests were performed using controlled displacement rate of 0.5 mm/min. From the obtained stress-strain curves, the scalar damage is quantified as the variation of the elastic modulus. Based on the experimental results, it was found that the damage depends on the replacement percentage of recycled aggregates and increases when the replacement ratio is high.

Nowadays, there is a significant increase in researches of concretes based on recycled aggregate. In this study, the recycled aggregates are provided from demolition waste. This work aims to characterize recycled gravels (RG) and to estimate their deterioration after their exposure to freezing and thawing cycles. The main objective is to determine the quality and suitability of RG by comparison to natural ones (NG) in manufacturing concretes dedicated to buildings, in particular when they are exposed to freeze/thaw. For each granular type, sieving, water absorption, porosity (n), Los Angeles (LA) and Micro-Deval (MDE) tests were conducted before and after their exposure to freezing/thawing in water according to European standards. The coefficient of water absorption, WA, was evaluated by three different methods. Results show that the law of mixtures can be used to predict all the studied properties of gravels mixes (natural and recycled gravels) except for MDE. Relationships are established between WA-density, LA-WA and LA-density on the basis on the results of this work and those of literature. Good correlation between the Micro-Deval (MDE) and the WA, as well as between MDE and LA was found. The analysis of the durability performance allowed concluding that, RGs are less resistant to freeze/thaw than NGs but their degradation, estimated through WA, n and MDE, is not significant.

This work aims to study the possibility of using recycled sand to prepare an epoxy resin mortar mechanically resistant and resistant against chloride ions diffusion. Comparative studies were performed on standardized hydraulic mortar and epoxy / standardized sand. The effect of the incorporation of a thinner, methyl octanoate, on the workability, mechanical properties and durability of the epoxy resin mortar is analyzed. It appears from this study that investigated mortars have a workability which is improved with the increase in the percentage of resin and with the addition of thinner. The total substitution of standardized sand by recycled sand alters this property. The resin mortar have a lower porosity than the hydraulic mortar, better mechanical properties and better resistance to chloride ions diffusion regardless of sand nature. The total substitution of standardized sand by recycled sand does not significantly affect the mechanical properties of resin mortars having the same binder percentages (20%). However, it causes a slight decrease in the resistance to chloride ions diffusion attributed to a higher total porosity accessible to water for recycled sand mortar. Similarly, the migration of these ions into mortar decreases with increasing the resin percentage, decrease attributed to the lowering of total porosity. Thinner addition affects the apparent diffusion coefficient.

In recent years new concrete types have been investigated, which include in the mix design a certain percentage of recycled aggregates, in view of a reduction of the production costs. Notably the re-use of material looks like an attractive alternative to landfill disposal and goes towards a more sustainable approach in the construction process. The present work addresses the realistic modelling of such a concrete mix design, focusing on both the geometric and constitutive aspect of the composite material, which is investigated at a mesoscale level, i.e. distinguishing concrete into its single components: matrix (cement paste) and coarse aggregates. A specific procedure for geometrically modelling such a kind of aggregates in a realistic way within a trial mix design is shown. The comparison between numerical and experimental results is in progress, to prove the goodness of the suggested approach, but here the challenging features represented by recycled aggregates in concrete from the point of view of numerical modelling, namely the geometric issue and the one related to the mechanics of the constituent materials are disclosed.

This paper presents the results of experimental research on recycled concrete at early age and long term. It consists in studying the effect of the proportion of recycled gravel (RG) on shrinkage and cracking risk. Three mixtures were investigated to study the effect of substitution of natural gravel (NG) by the recycled gravel (0% RG, 30% RG and 100% RG). A stress/strength approach was used to quantify the risk of cracking at early age. It is based on experimentally assessed parameters. At long term ring tests were performed to assess the cracking sensitivity induced by drying shrinkage. At early age the results indicate RG affected the properties of the recycled concrete and increase the potential risk of cracking. At long term, the effect of substitution of the recycled gravel on drying shrinkage was not significant. However, the three concrete showed different cracking sensitivity, thus shrinkage-induced cracking was influenced by the evolution of stiffness and relaxation. Finally, compared to the conventional concrete recycled concrete showed higher cracking sensitivity at early age and long term, but the effect was not proportional to the proportion of recycled gravel. 30% RG actually showed higher cracking sensitivity than 100% RG.

Concrete is the most used materials in the world. It is also one of the most versatile while complex materials which human have used for construction. However, concrete is weak in tension. Therefore, over the past thirty years many studies were focused to improve the tensile properties of concrete (cement-based composites) using a variety of physical and chemical methods. One of the most successful attempts is to use polymer fibers in the structure of concrete to obtain a composite with high tensile strength and ductility. However, a thorough understanding of the mechanical behavior of fiber reinforced concrete requires the knowledge of fiber/matrix interfaces at the small scale. In this study, experimental techniques have been used to study the nanostructure of fiber/matrix interfaces. Based on Scanning Electron Microscopy (SEM) and Energy-dispersive X-ray spectroscopy (EDX) analysis, the calcium-silicate-hydrate (C-S-H)/fiber interfaces of two different polymer fibers, polyvinyl alcohol and polypropylene, was numerically studied at the atomistic level.

The addition of external nanoparticles, mainly nano silica during the hydration of cement is a field of investigation in high performance cements. The added particles act as seeds and initiate early nucleation and subsequent growth of C-S-H gel. Nucleation is triggered very early, before enough clinker grains are dissolved or in other words, before the super saturation condition is attained. Hence, depending on the amount of added seed, the morphology and the mechanical properties of the product differ. Experimental studies in this area are less favored due to economic reasons and simulation studies are rare in the literature. An earlier work by some of the authors introduced a Monte Carlo model which dealt with the kinetics of the hydration process at early ages. The colloidal model incorporated random nucleation in the bulk, followed by an Avramian style layer by layer growth of 5nm sized C-S-H particles. The model was based on a Random Sequential Addition scheme and enabled a satisfactory rationalization of the early growth of C-S-H gel. In the present study, we extend this model for the addition of extra seeds and for different water to cement ratios.

In this work we have studied the dynamics of water confined in C-S-H gel and C-S-H gel synthesized with the addition of nano-silica (SiO2) and aminopropyl (-(CH2)3-NH2,) functionalized silica nanoparticles. Structural characterizations showed that the addition of nanoparticles increases the average length of the silicate chains in C-S-H gel. The dynamics of water confined in the microstructure of C-S-H with and without the addition of nanoparticles is altered and it can be related to a different porous structure in these materials.

Energy plays a key role in the technological, industrial, economical and societal development. The energy-intensive building construction sector accounts for approximately 42% of the total EU final energy consumption and is responsible for approximately 36% of EU CO2 emissions. The building construction sector is considered to be the one with the biggest cost effective energy-saving potential impact. Consequently, there is a strong need for implementing: i) Quick low-cost energy efficient retrofitting solutions to adapt the non-efficient residential European building stock to present energy regulation and ii) Low-cost & low-energy new buildings in line with the “nearly zero energy buildings” standards. The building envelope is the key element to address, in a building, in order to achieve the best improvements in terms of energy efficiency while maintaining the indoor environment quality. Therefore, the development of new lightweight materials and components for the building envelope with outstanding properties is terms of high thermal performance is a must that is being addressed in this work. These new materials and components will not only reduce considerably the heating and cooling demands but will maintain the indoor comfort of the buildings. The general objective of this work is to design and synthesize high thermal insulating cement-based products by incorporation of superinsulating aerogel-like materials. The develop materials will be lightweight, with a low value for the thermal conductivity while maintaining their mechanical properties.

The properties of the cement paste are defined to a high extend by those of the C-S-H gel, and irreversible deformations such as creep and shrinkage are not an exception. It is believed that molecular scale processes play an important role on the stress accumulation and relaxation, and molecular dynamics simulation a great tool to identify and quantify those mechanisms. In this work, we use NEMD simulation to investigate first the effect of the temperature and shear rate on the shear strength, Then, we study the stress development on a C-S-H interface with an increasing distance between adjacent particles. We discuss the results in terms of water, and we define a limiting distance, 1 nm, at which the shear strength is lost.

Nanoindentation is increasingly being used for the determination of nanomechanical properties of cementitious materials. The method is based on elastic punch theory to extract mechanical properties of interest from experimental load-displacement data. This theory requires initial unloading stiffness and contact area to be known. In this study, we propose a simplified method for the evaluation of stiffness and contact area based on more accurate and realistic interpretation of the load-displacement data. Such interpretation allows us to evaluate both the initial unloading stiffness and the contact area directly as a function of normalized elastic work. The area function derived herein may be used to compute contact area whenever the ratio of indenter tip to maximum penetration depth ratio is in the range of 0.1-1.67. Indentation moduli obtained using these expressions are in good agreement with those obtained by the Oliver and Pharr method for all the phases of hardened cement paste.

In recent years, the additions of nanomaterials as reinforcements for cement have attracted great interests. However, the understanding of the interfacial properties between these nanomaterials and the cement matrix is still limited. In this study, molecular dynamics (MD) simulations are conducted to investigate the graphene pull-out from calcium silicate hydrate (C-S-H). The jennite is simulated to model the structure of C-S-H. The graphene sheet is pulled out from the jennite matrix by 5 Å each time. The pull-out energy is found to increase linearly with the pull-out displacement. The pull-out force keeps constant (0.538N/m) throughout the pull-out process. This reinforcing mechanism is different with that of the microfibers (or macrofibers) and further research is needed to understand the role of nanomaterials in cement composites.

Ultra-high strength concretes (UHSC) with self-healing system based on the combined action of two additions have been developed in order to seal developing cracks, thus improving their long-term durability. The present work evaluates the efficiency of the self-healing system in aggressive environmental conditions with chloride attack. UHSC with and without the addition of the self-healing system were fabricated and cracked at 28 days of curing considering two crack width. Once cracked, the specimens were subjected to a salt spray test during 90 days and, after that, they were characterized by means of capillarity tests and colorimetric methods (chloride penetration evaluation). Unlike in the concretes without the self-healing system, the initial water absorption rate of the cracked specimens of self-healing concretes is similar to that of the corresponding un-cracked samples and almost insensitive to the crack width, thus confirming a robust self-healing mechanism that even limits chloride penetration.

This contribution aims at describing the creep behaviour of early age concrete with particular focus on the hardening process and the stress history. The influence of the hardening process on basic creep behaviour is investigated through tensile creep tests at different loading ages under constant stress. Influences of preloading are analysed through creep tests with several different stress histories. Creep tests with a stepped stress change are performed to check the applicability of superposition principle. Tests with slow and continuous stress application are targeted at describing the influence of loading velocity. The modelling of the viscoelastic behaviour is done using a rheological approach in terms of a generalized Maxwell model with aging spring stiffnesses and dashpot viscosities. The results show a good correlation for all investigated stress histories. This implies that the viscoelastic behaviour of early age concrete in tension can be described by theory of linear viscoelasticity.

Shrinkage, restrained by reinforcement or structural supports, induces tensile stresses in the concrete that might significantly affect the cracking resistance and short-term deformations of reinforced concrete (RC) elements. Incorrect assessment of this effect might result in noticeable errors of constitutive modeling. Recently the authors proposed a methodology for eliminating the shrinkage effect from short-term moment-curvature and tension stiffening relationships. However, due to rather complex issues of convergence and numerical integration, this approach was difficult to apply in practice. To simplify the methodology, the authors introduce a new concept of tension stiffening, the hybrid approach. Similar to the reinforcement-attributed approach, it assumes that tension stiffening stresses act in the area equivalent to the area of reinforcement. As in the concrete-related concept, tension stiffening is attributed to the material having the ability to shrink. The constitutive modeling based on the hybrid concept has become straightforward and easily performed by hand that is illustrated using experimental data of RC beam tested by the authors.

Within the CEOS.FR national research project, several experiments on massive concrete structures were conducted in order to improve the knowledge on the cracking phenomenon and, at last, to provide the designers with reliable crack design codes, able to give a reasonable appreciation of crack width and spacing for these structures as well as for classical ones. In this paper, experiments where deformations at early age are restrained are presented. Three restrained shrinkage testing bodies that are I-shaped, with a central part which has a cross-section of 0.80 mx0.50 m were tested. Two largely dimensioned steel struts (diameter equal to 32.3 cm and thickness of 4 cm) are placed laterally between the two transverse heads to prevent almost any shrinkage. They are equipped with strain gauges to know the level of restraint and to obtain the forces generated in the struts. Examples of the evolution of temperature, strains and forces are given. A first analysis of the cracking process shows that the cracks could appear for stresses below the tensile strength. With mainly strain information, global strain and local strain in concrete and on rebars, the force and then the stresses in concrete and rebars could be deduced. From numerical simulation and more simple elastic approach, state in each beam section could be obtained, and thus concordance between all the measures established. This huge amount of data allows verifying the phenomenology of the concrete in such boundary conditions. Various hypotheses were analyzed to explain strain measured and the corresponding forces in each component (concrete, rebars and struts) during specific period of early age. A particular attention was paid on the early age before the occurrence of the first crack. A realistic scenario is then proposed to explain the measurements. These hypotheses are done using a complementary numerical modelling and measurement analysis.

The current study reports the test results on deformations of cracked reinforced concrete (RC) ties. Four ties were tested under short- and long-term loading for 315 days. To perform the long-term tests, four testing rigs were constructed. The experimental results were compared with the Model Code 2010 predictions. It was revealed that the code overestimates the cracking load by 65-90%, and it is most likely due to the neglect of the shrinkage effect. It has been shown that 30-45% of the strain increment due to the long-term loading occurred during the first two-three days. The deformations of RC ties have practically stabilized after almost 300 days. The code quite accurately predicted the ultimate long-term strains. However, the prediction model was found to be too simplified to assess the increment of the long-term deformations.

In order to evaluate concretes in terms of crack sensitivity in the hardening phase it is essential to quantify thermal dilation (TD) and autogenous deformation (AD) in laboratory tests. Many laboratories performed such tests using different techniques for linear and volumetric measurements. The paper presents the results from a RR-program performed within the RILEM committee TC 195-DTD where 10 laboratories were represented and totally 30 Dilation Rig tests (linear test set-ups) were performed. For simplicity the nominal test condition for the RR was 20°C isothermal temperature; i.e. the scope was to measure only the effect of AD. The results showed that fresh concrete properties varied significantly in the various laboratories and this, amongst others, appears to have contributed to variations in the measured AD. The coefficient of variation (COV) of AD is significantly larger than for mechanical properties.

Work Group 1 (WG1) of COST Action TU1404 entitled “Testing of cement based materials” deals with determination of different properties of early age and hardened cement based materials with the following main objectives: (1) to recommend new advanced standard techniques which have been developed during the last years to become standard in the (near) future, (2) to test the ability of using different waste, recycled, and by-products as raw materials to design sustainable concrete mixture, and (3) to prepare a database of concrete mixtures which could serve designers and engineers to better predict lifespan, durability, and serviceability of concrete mixtures and structures. The paper describes organization and strategic view of WG1, presents its main objectives and gives important details about Round Robin Test, as a main tool that will be employed within WG1.

The study presented in this paper deals with the influence of the measurement technique on autogenous shrinkage. A state of the art shows that many setups based on different techniques have been designed to measure autogenous deformations. In this experimental study, four different setups and setting devices were gathered in the same laboratory. All the methods were found to fulfill the criteria for reliable assessment of autogenous deformation. The different methods showed good repeatability and consistent evolutions of autogenous shrinkage. The autogenous shrinkage was initialized at the time of minimum shrinkage rate. Good correlations could be found between the times of minimum and maximum shrinkage rate, the evolutions of ultrasonic velocities, and the times of maximum temperature.

The work presented here was performed in the context of the French national project CEOS.fr. The study is part of the second axis of the project dealing with the concrete behavior under thermo-hydro-chemo-mechanical (THCM) loads. One of the objectives of this component was to construct some massive structures with strain restraint in order to test and validate THCM model for predicting the early-age cracking of concrete structures. Once validated, this THCM model is used in the last part of the project to perform some numerical parametric studies and a more precise analysis of the effect of the reinforcement on the crack pattern of massive structures at early age.

Restrained volume changes in concrete structures at early age may be concomitant with mechanical loadings and lead to cracking. If this is not precisely addressed in service-life design, long-term consequences on the serviceability, the durability and the structural behavior of concrete structures may occur. Qualification of advanced models and development of appropriate standards and specification are two main target points of a recently launched COST Action. The main features of the program and objectives of the related working groups are detailed in this paper.

The quantitative verification of an inkbottle pore and a continuative pore determined by mercury intrusion porosimetry (MIP) was conducted by comparing the volume of air obtained through the linear-traverse method (image analysis) with the mass change due to water absorption and drying. The c pore had a high correlation with the volume of air voids and the continuative pore with the amount of rapid water absorption and drying up to 3 hours from the start of the test.

Due to chemical hydration processes, cementitious materials exhibit a strong evolution of their physical (mechanical, transport, etc…) properties as a function of time. More precisely, as hydration proceeds, phase transformations occur, such as dissolution of anhydrous phases and precipitation of hydrates. The microstructure strongly evolves. Micromechanics can be used to translate this time-dependent microstructure into mechanical properties evolution. This is straightforward when the investigated behaviour does not involve time by itself, such as elasticity. The case of creep is much more complex, as another time scale steps in, through the intrinsic delayed behaviour of hydrates. Taking advantage of recent works on ageing viscoelastic homogenization of random media, and on a micromechanical extension of Bažant’s solidification theory, a simplified model of cement paste ageing basic creep is proposed. Assuming an evolving morphological model with three phases (anhydrous, hydrates and capillary porosity), the effective ageing creep of the paste is estimated.

Nano-indentation is used to study the mechanical properties of cement paste. A method is developed to prepare small cement paste cubes (100 µm). These samples are loaded to failure with a Berkovich indenter to obtain the global mechanical properties of the material. Local mechanical properties are obtained by standard nan-indentation measurements on the same material. These local properties are used as input parameters for a lattice-type model to simulate the failure of heterogeneous microstructures. With the model the failure of the small cement paste cubes is simulated and the outcome of test and simulation is compared. The proposed technique forms the basis for upscaling of modelling techniques and validating the results on every scale.

Most carbonation models only account for diffusion as the main transport mechanism rather than advection. Nevertheless, in the case of concrete used for underground waste disposal facilities, concrete may be subjected to a high hydrostatic pressure and the surrounding environment may contain a high dissolved CO₂ concentration. Therefore, a combination of diffusion and advection should be taken into account. This is also the case in accelerated carbonation where a high CO₂ pressure gradient is applied in which advection in the gas phase has a significant contribution to the carbonation process. This study aims at developing a model to predict the evolution of the microstructure and transport properties of cement pastes due to carbonation under accelerated conditions in which a pressure gradient of pure CO₂ is applied. The proposed model is based on a macroscopic mass balance for carbon dioxide in gaseous and aqueous phases. Besides the prediction of the changes in transport properties (diffusivity, permeability), the model also enables to predict the changes in microstructure. Data from accelerated tests were used to validate the model. Preliminary verification with experimental results shows a good agreement.

This paper deals with the process of non-expansive alkali-carbonate reaction (ACR) in cement mortar prepared with specific dolomite aggregate. ACR was studied for one year in accelerated conditions. Results obtained at microstructural level show that ACR is progressive with time and reflects in the process of aggregate grains’ dedolomitisation, the formation of secondary calcite in binder and the formation of new reaction products rich in Mg, Al and Si atoms along the dedolomitised parts of the grains. Observed mineralogical and microstructural changes reflect in relatively high increase in compressive strength of the mortar over time. On the other hand, its modulus of elasticity started to decrease already after five months of the exposure and was gradually decreasing until one year of the exposure.

The durability of the concrete structure is determined by transport properties of concrete. The transport of chloride ions greatly changes by the amount of adsorption by hydrates. However, there are few studies on change of the amount of adsorption with the calcium leaching. In this study, the purpose is to investigate a change of the amount of adsorption of the chloride ion with the calcium leaching of hardened cement pastes. The amount of chloride ion adsorption increased with the increase of the replacement ratio of the blast furnace slag. We established an equation to predict the amount of chloride ion adsorption on C-S-H and Afm. It was clarified that the amount of chloride ion adsorption on C-S-H of leached hardened cement paste depends on Ca/Si and Al/Si.

We propose an investigation combining numerical and analytical tools to estimate the ageing viscoelastic properties of cementitious materials within a multiscale framework. With analytical homogenization the properties at the cement paste and mortar scale are estimated by a combination of Generalized Self-Consistent (GSC) and Mori-Tanaka (MT) schemes. With numerical homogenization the effective properties at the concrete scale are estimated. Numerical homogenization has the advantage of allowing to assess local information as well as to study more complex geometries. This combined strategy constitutes a promising tool to investigate how different mechanisms leading to ageing at the hydrate products scale, as well as other features of cement-based materials such as the Interfacial Transition Zone (ITZ), affect the viscoelastic behavior at superior scales. In this context, we study the solidification of non-ageing constituents as the mechanisms leading to the ageing behavior. Relaxation and creep results at different scales are presented.

For evaluating damages of structural concrete by ASR, an analytical platform to rationally deal with the complex interaction of multi-scale chemo-physics events is being developed. For experimental verification of the predictive model proposed, ASR expansion tests under several magnitudes of confinement are conducted and the results are compared with the multi-scale simulation. It is experimentally found that the highly deviatoric compression may bring about isotropically confined ASR expansion. The poro-mechanics based multi-phase modeling can simulate this nonlinearity by considering the quasi-hydro static pressure of created ASR gels in concrete composites and its injection into the micro-pores.

The magnitude and rate of development of early-age deformation of concrete is often not known reliably and restraint to shrinkage is often poorly modelled in structural design. In order to accurately quantify the early-age deformational characteristics of Australian concrete, including shrinkage and tensile creep, a comprehensive experimental program is underway at the Centre for Infrastructure Engineering and Safety at UNSW Australia, where a series of experiments have been performed to quantify the early-age properties of concrete. A restrained cracking test was devised and reinforced concrete prisms were tested in tension under sustained loading to measure tensile creep and restrained shrinkage. Free shrinkage was measured on companion unreinforced prism specimens. Dog-bone shaped specimens were also tested in direct tension in portable creep rigs. All the specimens were tested at a stress-strength level of 50 % of the tensile strength of concrete at the age of 2 days. It was found that RC prisms cracked with time under sustained tension before reaching the instantaneous tensile strength of concrete. A strength reduction factor of 0.7 and 0.87 was observed based on split tensile strength and direct tensile strength, respectively. Tensile creep coefficients calculated from tension prisms agree well with the tensile creep coefficients of dog-bone shaped specimens.

Water vapour sorption isotherms (WVSIs) provide a relationship between the mass of adsorbed water within pores and the prevailing relative humidity; they are a reflection of the material microstructure. One aspect of cementitious material WVSIs that has not been firmly established is the effect of elevated temperatures. In this paper, two newly developed multi-chamber WVSI testing apparatus are described. These apparatus have been used to study the adsorption and desorption isotherms and kinetics of HCP and concrete at various temperatures. Results show a two-stage equilibrium process in desorption studies, where the initial stage is associated with diffusive moisture transport, and the second appears associated with a time-dependent loss of hysteresis moisture.

In recent years there has been a rapid increase in the use of supplementary cementitious materials, such as fly ash and ground granulated blastfurnace slag, for high performance and durable concrete, as well as to reduce clinker requirement. Such materials being fine and not as reactive as cement, increase the setting time and water demand, and reduce bleeding in concrete. If the rate of evaporation is high enough, plastic shrinkage cracking occurs in slabs and other elements that have high surface area to volume ratios. Such premature surface cracks may accelerate the ingress of harmful agents, and impair durability and serviceability of the structures. The present study has two objectives: firstly, to investigate the influence of supplementary cementitious materials on plastic shrinkage cracking, in hot and dry environmental conditions; and secondly, to control plastic shrinkage cracks using fibers and shrinkage reducing admixtures. Four synthetic fibers — polyacrylonitrile, polyester, glass and polypropylene fibers at commonly-used dosages are evaluated. The results indicate that the addition of supplementary cementitious materials increases the cracking potential as the dosage increases. The incorporation of fibers and shrinkage reducing admixtures reduce crack widths, suggesting that they are viable solutions for combating plastic shrinkage cracking in concretes with supplementary cementitious materials.

In this paper we investigate analytically and numerically the creep behavior of concrete at mesoscale. The simulations are carried out with the finite element (FE) code (Cast3M) on 3D concrete specimens consisting in about 4600 polyhedral aggregates of various size and shapes randomly distributed in a box. Both matrix and Interfacial Transition Zone (ITZ) are considered as linear viscoelastic materials while the aggregates are elastic. Specific interface finite elements are introduced between the aggregates and the matrix to model the ITZ. The overall and intra-phase response of the numerical specimens when subjected to classical creep loadings is investigated, and then compared to the results of analytical estimations obtained with classical mean-field approximation schemes. These schemes are applied in the Laplace-Carson (LC) space, and the effects of the ITZ are accounted for via appropriate interface models. The results obtained for different ITZ thicknesses are analyzed so as to evaluate their respective influence on mortar and concrete materials.

The aim of the present study is to experimentally investigate the water vapor desorption and drying shrinkage of cement-based materials during a step-by-step desorption process, with the objective to link drying shrinkage to water desorption. Water vapor desorption isotherms of two cement pastes and two mortars (with water-to-cement ratio of 0.5 and 0.8) were studied using thin slices with three thicknesses 1 mm, 2 mm and 3 mm. Drying shrinkage was measured with 2 mm slices for each material. The slice thickness has an impact on the relative mass variation, but the difference between 2 mm slices and 3 mm slices is very small or even negligible. The results obtained in the present study are compared with a former study on larger prismatic beams. The curing condition influences the water desorption isotherm: the water content in the present study is always higher at each relative humidity after a longer curing period with thin slices. The size effect is pronounced for drying shrinkage: the drying shrinkage obtained with slices is always bigger than larger prismatic beams. The correlation between drying shrinkage and water mass loss shows two almost linear parts with different slopes for both cement pastes and exhibits three characteristic phases for both mortars.

The aim of this paper was to quantify crack initiation and length change of ordinary Portland cement-based mortar and three types of alkali activated slag mortars: reference one that only slag was used as binder, the second one with same amount of slag and with shrinkage reducing admixture in amount 1 % weight of slag and the third one with 5 % weight of slag replaced by fluidized bed combustion fly ash. Czech standard sand consisting of three different fractions (fulfilling requirements of ČSN EN 196-1) was used in all mixtures. The sand to binder ratio was 2:1 and water to binder ratio was 0.40 in all mixtures. Curing and length change testing of the specimens were based on ASTM C596-07 and ASTM C490-7. After demolding and three days of water curing mortar prisms were placed in air storage for 24 days. During this period length change (shrinkage) was measured. After that the three point bending test was performed on these specimens at the age of 28 days to obtain the fracture parameters, which was determined via Double-K fracture model. The initiation of cracks during the fracture tests was also monitored by the acoustic emission method. For comparison, specimens cured in water until the age of 28 days were tested in the same manner.

The quality of curing is a significant initial condition, which changes regularly during construction. A longer sealed curing results in higher strength and stiffness of concrete. It also provides a better durability and a lower carbonated depth. However, few studies deal with the influence of the duration of sealed curing on the long-term drying shrinkage and cracking sensitivity. In this paper, one self-consolidating concrete and one vibrated concrete, referred to as SCC and VC respectively are tested to study the effect of sealed curing duration on the long-term behaviour of concrete. The corresponding specimens were demolded after three different curing durations: 16, 24, and 48 hours. Total and autogenous shrinkages and mass-loss evolutions were monitored. Mercury intrusion porosimetry was performed in the three cases. The carbonated depth was measured after 6 months and one year. A long curing period actually gives more time to the cement to hydrate without water loss, and the pores become finer under these conditions. The application of the Kelvin-Laplace equation results in a higher stress for finer capillary pores. The experimental results confirmed this theoretical assumption, for a higher drying shrinkage for concrete associated with a longer sealed period. The experimental cracking sensitivity also increases with this duration owing to higher stiffness and shrinkage.

Oil palm shell (OPS) is the hard endocarp which surrounds the palm kernel and received as crushed pieces during the extraction of palm oil. It is an agricultural waste and can be found in abundance in Malaysia. Economic and environmental benefits as well as decreasing the use of natural resources can be achieved if OPS can be utilised as a building material. Promising results showed that OPS has good potential to be used as coarse aggregate replacement in making structural lightweight concrete. It has also been previously established that polyvinyl alcohol (PVA) can be used to improve the properties of OPS concrete. Chloride induced reinforcement corrosion is an important area of study which needs to be investigated, especially for concrete structures exposed to marine environments. Therefore, this paper presents an experimental study on the chloride penetrability of PVA treated OPS concrete in comparison to that of raw OPS concrete using the 90-day salt ponding test. The effects of different curing conditions (full water curing, air-dry curing and site curing) on the chloride penetration resistance of PVA treated OPS concrete and raw OPS concrete were also studied. Both PVA treated OPS and raw OPS concrete surface were ponded with 3% sodium chloride solution up to 90 days, and the chloride profiles were measured by analysing concrete dust samples extracted from different depth intervals from the chloride exposed concrete surface. Subsequently, chloride diffusion coefficients were determined by fitting the experimental data to Fick’s second law of diffusion. From the test results, it was found that PVA treated OPS concrete had high surface chloride concentration as compared to raw OPS concrete. However, PVA treated OPS concrete showed higher resistance to chloride penetration as compared to raw OPS concrete as indicated from the chloride diffusion coefficient. The results indicated that curing conditions also have a significant influence on the chloride resistance of PVA treated OPS concrete and raw OPS concrete.

Coefficient of thermal expansion of concrete is an essential part of cracking risk analysis. A formula for coefficient of thermal expansion based on the theory of poromechanics was derived. This formula utilizes ultrasonic pulse velocity measurements. To validate the formula, ultrasonic pulse velocity and free shrinkage of cement paste with water to cement ratio of 0.33 were measured starting from the casting. A good agreement was found between the calculated values and the data found in the literature. In addition, the calculated coefficient of thermal expansion was applied for decoupling of thermal and autogenous shrinkage of cement paste. Decoupled linear autogenous shrinkage was compared to the measured volumetric autogenous shrinkage.

The extensive research activity carried out over the last decades on fibre-reinforced concrete (FRC) has shown that such material has enhanced mechanical and durability properties compared to plain concretes. The presence of short fibres in the concrete mass allows to control cracking and have moderate time-dependent effects. Compared to plain concrete, FRC flexural members show a higher number of cracks with reduced mean width. The experimental study presented herein discusses the mechanical behavior of FRC flexural beams subjected to sustained service load and environmental exposure for 72 months. The effects of different short fibers (polyester and steel), sustained loading and aging were investigated. A comparison with the results of a previous research is shown, with reference to the same kind of beams exposed for 17 months under the same conditions. The results show the beneficial effects of the fibers in terms of reduced crack width and increased flexural stiffness. The mechanical tests also highlighted how the presence of short structural fibers could play an effective role in mitigating creep effects in the concrete elements.

Because of their high volume of paste, self-compacting concretes (SCCs) cost more expansive than ordinary concrete and exhibit more shrinkage values which increases their susceptibility to cracking and may alter their durability performances. For economical and environmental considerations, many studies were done these last years on the use of mineral admixtures as filler in SCCs. The aim of this paper is to contribute to the characterization and the understanding of shrinkage of SCCs where Portland cement is partially substituted by slag, fly ash or metakaolin. The shrinkage from casting to 360 days of aging of four SCC mixtures based on CEM I 52.5N, CEM III/A 52.5L (slag cement), CEM V 42.5N (blended cement containing fly ash and slag) and CEM I + 15% of metakaolin is investigated. Experimental tests include plastic, autogenous, free drying and restrained shrinkage. Results show that shrinkage depends on the mineral admixture type and more particularly on the developed microstructure. Replacing 15% of Portland cement by metakaolin increases the shrinkage at the plastic state and decreases it significantly at the hardened state in free drying conditions. The use of CEM III instead of CEM I involves an inverse behavior. In restrained conditions, mineral admixtures seem to reduce the age of crack initiation. However, the crack opening is reduced by at least a factor of 2. The results are in a good agreement with water porosity, mass loss, compression strength and thermal measurements.

The drying of concrete induces surface microcracking due to hydric gradients, and intern microcracking due to drying shrinkage incompatibilities between cement paste and aggregates. The difficulty of separating each size effect on concrete cracking makes it hard to quantify aggregates effects with classical experiments. An ongoing parametrical experimental study aims at quantifying the impact of drying and of inclusions by using morphologically controlled materials. In these materials, the aggregates size and volume fraction is set. The samples are then submitted to drying and regularly controlled for mechanical and transport properties. Also, microcracks development is followed using X-Ray tomography. The first results indicate the low impact of aggregates size, whereas the one of volume fraction is significant. The study is to be enhanced of several results: different formulations and tests to assess the evolution of mechanical and transport properties at several ages are currently undertaken.

This paper discusses key processes and phenomena that occur when concrete is exposed to elevated temperatures up to 800 degrees Celsius which is of significant importance to high fire risk infrastructures such as Nuclear power plants. The majority of research in this field to date has concentrated on the mechanical properties of concrete such as compressive strength at elevated temperatures. In order to fully develop accurate computational analyses of the performance of concrete in fire, the long-term deformation of concrete known as creep at elevated temperature must be included. This is particular important since high fire risk infrastructures are designed for operation for a very long period of time and therefore their long-term deformation under elevated temperatures is of significant concern to design engineers. This paper presents a brief state of the art literature review regarding the mechanical properties of concrete and creep of concrete after exposure to elevated temperatures.

This paper investigates the uniaxial compression behaviour of concrete under different strain rates and its relation to sustained loading. The aim is to observe the strain rate effect on the pre- and post-peak behaviour of concrete. In particular, the development of nonlinear creep strains for different strain rates has been investigated and compared to the sustained load behaviour. The development of creep strains is in accordance with the progress of cracking as observed by acoustic emissions. An analogy is found between the concrete compression behaviour under various strain rates and the concrete compression behaviour under sustained loads at different stress levels.

The volumetric change is a very important problem e.g. the cracking of concrete. Therefore, the evaluation of the detailed water migration mechanism and the accurate predicted model of the volumetric change are indispensable. In this study, the diffusion model depending on the pore diameter was constructed by introducing the dominant substance transfer forms in each diameter of pore focusing on the water migration in a microscopic viewpoint. Moreover the proposed model was expanded by considering the ratio of CaO to SiO₂ (hereafter C/S ratio) in C-S-H that is a main cement hydrate. Furthermore the volumetric changes, which occur in the state during the dehydration from inside to outside of the specimen and in the state after the finish of dehydration, were modeled. Finally, the comparisons between the experimental and analytical results for the volumetric change of the cement paste due to the drying were performed and then the applicability of the proposed model was confirmed.

One of the most important issues in civil engineering about structures is their durability. A structure during his lifetime faces different environmental situations, like different relative humidities, and different cycles of drying and saturation will be imposed on it, so it is very critical to characterize materials behavior against these changes. Some informations are needed to predict it, one of these informations is desorption and sorption isotherm. In cement based materials just as cement paste, there remain generally extremely fine pores, consequently obtaining isotherms experimentally is considered as a substantially time-consuming experience. Therefore, it is of interest to model the isotherm from a numerical point of view. This physical phenomenon has been a field of study of intense researches, however, there are not many models relying on true physical effects which take place during drying and saturation. In the present work, in order to model a porous media, a numerical morphology of the geometry of the porous media has been presented by means of a random field excursion through the selection of a random field and a threshold. Random field parameters are defined based on experimental values. Once the porous media has been reproduced, the morphological analysis must be done in order to model water transfers.

Compressive strength is one of the most important and tested mechanical properties of cement paste. This paper presents a new four-level micromechanical model for compressive strength applicable on both pure and blended cement pastes. The model assumes that intrinsic tensile strength of C-S-H globules governs the compressive strength of cement paste. Crack propagation on all hierarchical levels starts once tensile stresses on a randomly inclined ellipsoidal inclusions within C-S-H exceed cohesive stress. The inclusion of unhydrated clinker, supplementary cementitious materials, other hydration products, or entrapped (or entrained) air further decreases the compressive strength of cement paste. The multiscale model uses volume fractions of principal chemical phases as input parameters as well as introduces a spatial gradient of C-S-H between individual grains which has a pronounced impact on predicted compressive strength. Calibration of the model on 95 experimental compressive strength values shows that the intrinsic tensile strength of C-S-H globule amounts to 320 MPa. Principal factors influencing compressive strength evolution are “C-S-H/space” ratio, volume of entrapped (entrained) air, and the spatial gradient of C-S-H.

The viscoelastic Poisson’s ratio is an important parameter for assessing the multiaxial creep behavior of concrete. However, its definition in viscoelasticity can generate some ambiguity: at least five different ways of defining a viscoelastic Poisson’s ratio are presented in the literature. As to their difference either in theory or in practice, little is known. In this work, we focus on the most intuitive two ways of defining a viscoelastic Poisson’s ratio, which we call “relaxation Poisson’s ratio” and “creep Poisson’s ratio”. First, we derive the analytical expressions of the two Poisson’s ratios and a relationship between them. We show that their initial values are identical, and that their asymptotic values when time tends towards infinity as well. We also show that such is the case for their derivatives with respect to time. Then, considering concrete as a non-aging linear viscoelastic material, the results of multiaxial basic creep tests on concrete available in the literature are analyzed to compare the relaxation and the creep Poisson’s ratios. The results show that the difference between the two Poisson’s ratios is rather small but does exist in some cases. In such cases, whether this difference is significant should be considered with respect to the application considered.

In order to understand the interaction between creep and damage, some experimental methods were developed to follow the microcracking evolution during creep tests as the acoustic emission technique. However, these techniques are unable to analyze the real coupling between creep and damage. In fact, the creep deformations lead to the damage which in turn modifies the creep deformations. Thanks to the modeling, we can take in consideration the microscopic mechanisms to better evaluate the creep deformations of young concrete. For this, a multi-scale creep model was developed to calculate the viscoelastic parameters at the concrete scale from those of hydrates. Then, the creep model was coupled to an isotropic damage model to understand the interaction between creep and damage during creep tests and to evaluate its effect on the mechanical behavior after creep. Finally, the coupled model was used to simulate the interaction creep-damage of pre-damaged materials in order to assess the effect of an initial damage on the creep deformations and on the residual capacity.

The experimental results of isotropic compression tests performed at 20°C and 90°C on a class G hardened cement paste (Ghabezloo et al., 2008, Cem. Conc. Res. 38, 1424-1437) have been revisited considering time-dependent response. Within the frame of a viscoplastic model, the non-linear responses of the volumetric strains as observed in drained and undrained tests and of the pore pressure in undrained tests are analysed. The calibration of model parameters based on experimental data allows to study the effect of the test temperature on the time-dependent response of hardened cement paste.

A new bond model for reinforced concrete members considering material deterioration induced by freezing-thawing action has been developed. The developed model with the extensive applicability from non-damaged to severely damaged members can reasonably describe deterioration mechanism between steel bar and concrete.

How volume change of the cement hydrates affects mechanical characteristics of mortar after calcium leaching are discussed and prediction equations for elastic modulus, tensile strength, fracture energy and shear strength are proposed. Furthermore an analysis of mass transfer with chemical reaction based on a finite volume method is developed and is coupled with a mechanical analysis based on the RBSM. The validity of the coupled analytical method is discussed.

In this paper we are presenting the development of a new concept, the embedded unit cell (EUC) approach, used to calculate local responses in elastic media. In addition, the suggested formulation provides homogenization and multi scale analysis of composite materials, structures and domains; where the classical theory of homogenization does not valid. The EUC approach is based on a multi-scale formulation of the asymptotic homogenization theory to evaluate structure response in several special cases, such as response of non-periodic domains or local/micro response at zones that are expected to develop stress concentrations. The suggested approach is based on the zone-adapted unit cell, restricted by alternative boundary conditions and surrounded by micro scale domain that represents non periodic features of the macroscopic structure. By using the alternative boundary conditions, the periodic assumption of unit cell response that is essential in the classical theory, is no longer required yet preserving an accurate micro-scale response evaluation. This approach offers a reduced computational cost model of the macroscopic/global problem however the precision of the microscale problem solution is retained. The EUC concept broadens the applicability of multiscale analysis techniques, used to evaluate mechanical response of variety of composite materials, in particular highly heterogeneous materials (i.e. concrete, etc.), which are widely used in modern construction industry.

This contribution examines how the properties of a viscoelastic model are affected by the choice of the creep compliance function and by the properties of the corresponding continuous retardation spectrum. Attention is paid to the phenomenon called divergence of creep curves, which arises for certain types of models but is questionable or even unacceptable from the physical point of view. Divergence occurs when the distance between creep curves (plotted for different ages at loading as a function of current age) increases, and its consequence is that, if the principle of superposition is accepted, the recovery of strain after unloading becomes non-monotonic. Analysis of this undesirable phenomenon is performed for creep models recommended by major design codes (ACI, CEB, fib, JSCE), and for simple as well as advanced models developed by researchers (DPL, LDPL, B3/B4, GL2000). The ranges of ages at loading and load durations for which divergence occurs are detected and illustrated by graphs. It is suggested that the origin of divergence could be explained by examining the corresponding continuous retardation spectra.

In this study, for the purpose of improving the prediction technique of cracking with volume change of concrete structures, it is to develop the system which can simulate shrinkage stress with volume change of concrete in the laboratory. For the behavior of shrinkage of concrete at early age, it was carried out to compare the data measured in the existing structures with experimental results using the equipment (TSTM) in which it is possible to simulate the behavior of thermal stress, autogenous stress and creep stress of existing structures in the laboratory. As the results of experiments, it was confirmed that it was possible to comparatively reproduce the stress in concrete at the good accuracy by using TSTM. By comparing the direct tensile strength with experimental results, it was possible to acquire the useful data in which it was possible to evaluate cracking using direct tensile strength.

Thermal stress in a gravity concrete dam was simulated by three dimensional FEM analyses taking autogenous shrinkage of concrete with full-sized aggregate into account. It is proved from experimental results that external concrete with blast-furnace slag Portland cement exhibits autogenous shrinkage of 130×10-6 at one year after casting, which is much larger than moderate-heat Portland cements with and without fly ash. Calculated thermal stress in upstream-downstream direction at the center of the concrete blocks near foundation rock is increased by autogenous shrinkage if concrete with large autogenous shrinkage is used. Thermal stresses in dam-axis direction at upstream and downstream surfaces are increased by the difference in autogenous shrinkage strains between internal and external concretes, if blast-furnace slag Portland cement is used. It can be said that autogenous shrinkage which simultaneously occurs with thermal strain should be taken into account in order to control thermal cracking during construction period which can adversely affect the stability, water tightness and durability of concrete gravity dams.

Many existing mass concrete structures are unreinforced because of economic aspects. Based upon the fact that no national or international standards for the design of such structures are available, each of them has to be treated as individual case. This often leads to safety concerns or concerns of reliability. An efficient design of unreinforced mass concrete structures needs a realistic evaluation of any crack formation to be expected. This calls for a realistic quantification of the decisive stress-distribution. This contribution shows first results of a preliminary study to develop a new design concept for unreinforced mass concrete structures that is in accordance with the design philosophy of Eurocode 2 (Ö NORM B 1992-1-1, 2011). This contribution also deals with the development of a new, efficient time-step FEM-based approach for the consideration of viscoelasticity in order to allow a comprehensie parametric study on the restraint behaviour of mass concrete structures.

Remaining performance of corroded reinforced concrete beams is numerically evaluated by means of poro-mechanics approach. Corrosion substance is treated as the combined phases of liquid and solid gels. The static loading experiment of corroded reinforced concrete beams is fairly simulated by the multi-directional fixed crack model combined with corrosion gel migration. Further, fatigue life of corroded reinforced concrete beams is studied numerically. Even if static shear strength is not reduced so much, remaining fatigue life could be shortened significantly due to the cyclic degradation of cracked cover concrete around expanded steel bars.

A numerical simulation system for long term stress and deflection of concrete structures is developed. Environmental actions on structures, which are temperature, humidity, rain fall, sunlight and their time-dependent changes, are precisely considered. Time-dependent behaviour of concrete member is calculated by the moisture transport and drying shrinkage model based on pore structure of concrete. Coupled transport model of heat, vapour and liquid water in concrete pore structure is employed. In addition, capillary suction from the surface, condensation of vapour on the surface and acceleration of evaporation by direct sunlight are taken into account. Automated Meteorological Data Acquisition System provided through Internet is utilized to evaluate climatic action. It is demonstrated that time-dependent change of moisture content, free shrinkage of concrete specimen and deflection in prestressed concrete beam under natural environmental action and sustaining load can be adequately simulated.

Experiments have shown that shrinkage of concrete is larger in those specimens which are loaded in tension compared to non-loaded specimens. The hypothesis is that this is a consequence of microcracking and increase of water diffusivity. Water diffusivity is indeed increased by tensile loading as several researchers have demonstrated. That shrinkage is increased by tensile loading is in contradiction to other theories which state that surface cracking is caused by eigenstresses during drying and thus the shrinkage strain is reduced. Several experimental series have been carried out on concretes with various strength and composition which have confirmed the phenomenon. Another phenomenon has been observed on concrete which was mixed with superabsorbent polymers (SAP). Super-absorbent polymers are a new class of admixture. They absorb a manifold of water compared to the own weight. Mixed in concrete, they store the water and release it during the hydration of cement. They can be used for internal curing of high-performance concrete, further, they can heal cracks and can make concrete more impervious. Tensile creep tests have been performed on dogbone shaped specimens. It appeared that tensile creep of SAP modified concrete was considerably reduced while shrinkage was not. The reason for this behavior is attributed to self-healing of microcracks. Results will be shown and discussed.

The creep behavior of concrete members is of major importance for their serviceability. Besides its negative influence on deflections under load impact, creep strains are supposed to positively reduce the hardening induced restraint stresses after temperature equalization - which is an essential assumption for today's design practice of neglecting the superposition of hardening induced restraint stresses with later restraint stresses during life time. A number of different experimental studies, however, have indicated that the assumed positive effect of creep is highly overestimated. This can mainly be explained by disproportionately diminishing tensile creep responses after relaxation under the maximum of all ever experienced tensile stresses. Based on detailed information on the experimental studies with passive restraint frames as well as comparative tensile creep tests, this contribution presents new insights on tensile creep behavior under realistic conditions in concrete structures.

A new concept for creep coefficient defined as effective creep coefficient was proposed. This concept was derived based on strain in prestressing strand directly measured in a prestressed concrete slab-girder as well as shrinkage strain of concrete specimen, which were made of ultra high strength concretes containing a molten slag fine aggregate and both the molten slag fine aggregate and roof tile waste coarse aggregate. The effective creep coefficient was verified by comparing the prestress loss obtained using the effective creep coefficient with that obtained using experiment as well as that by the creep coefficient obtained from the conventional creep test. The results showed that the stress losses obtained using effective creep coefficients agreed with experimentally obtained stress better than those obtained using creep coefficients from the conventional creep test.

A mini-beam test has been designed to assess the flexural creep and shrinkage behaviors of paste and mortar samples. It is a four-point bending test carried out on a plate sample of 350 mm long. In total seven sets of samples, pastes and mortars made of ordinary Portland cement (OPC) and geopolymer were tested. Test results of OPC and geopolymer pastes and mortars from this test showed agreement with the general long-term performance of OPC and geopolymer concretes from the literature. The mini-beam test can be performed under various environmental conditions, such as high temperature and sealed condition, to assess the environmental effects on the samples. Furthermore, prediction of long-term behavior of OPC mortar and concrete can be done by using the developed models from the literature, based on the long-term behavior of OPC paste obtained from the mini-beam test.

The assessment of concrete mechanical properties during construction of concrete structures is of paramount importance for many intrinsic operations. However many of the available non-destructive methods for mechanical properties have limitations for use in construction sites. One of such methodologies is EMM-ARM, which is a variant of classic resonant frequency methods. This paper aims to demonstrate the efforts towards in-situ applicability of EMMARM, as to provide real-time information about concrete mechanical properties such as E-modulus and compressive strength. To achieve the aforementioned objective, a set of adaptations to the method have been successfully implemented and tested: (i) the reduction of the beam span; (ii) the use of a different mould material and (iii) a new support system for the beams. Based on these adaptations, a reusable mould was designed to enable easier systematic use of EMMARM. A pilot test was successfully performed under in-situ conditions during a bridge construction.

The experimental evaluation of viscoelastic properties of concrete is traditionally made upon creep tests that consist in the application of sustained loads either in compression or in tension. This kind of testing demands for specially devised rigs and requires careful monitoring of the evolution of strains, whereas assuring proper load constancy. The characterization of creep behaviour at early ages offers additional challenges due to the strong variations in viscoelastic behaviour of concrete during such stages, demanding for several testing ages to be assessed. The present research work aims to assist in reducing efforts for continuous assessment of viscoelastic properties of concrete at early ages, by application of a dynamic testing technique inspired in methodologies used in polymer science: Dynamic Mechanical Analyses. This paper briefly explains the principles of the proposed methodology and exhibits the first results obtained in a pilot application. The results are promising enough to encourage further developments.

The post-tensioned inner concrete containment building of nearly half of the French nuclear power plants has the role of barrier against potential radiological release, since these containments buildings have no metallic liner. Hence, the integrity of concrete and the stability of post-tensioning are of paramount importance. Since the variation of post-tensioning is partly due to concrete creep, EDF has been interested in the subject for more than two decades. This article presents the results of experiments performed on a concrete similar to that of a concrete containment building, with an emphasis on the bidimensional nature of creep, on the basis of 1D and 2D tests that were performed at EDF. The evolution of the creep Poisson ratio is discussed as well as its use by different authors. Creep is then analyzed in terms of deviatoric and volumetric components. We show that assuming a constant Poisson ratio is a good assumption.

The creep behavior of concrete structures at long term depends on the initial state and therefore on an eventual damage of the main material. Most studies were interested in creep phenomenon of mature concrete while a few works were carried out on young concrete. However, the concrete structures may be solicited earlier as the tensioning of the pre-stress cables induces an early creep and an eventual cracking, thus reducing the structure capacity at the long term. An original device for creep tests has been developed to follow the evolution of the displacement, the force and the crack opening during creep since early ages for concrete beams. Creep tests were performed on concrete beams aged of 24h during 28 days. Results show that creep influences the rigidity and the strength when the load-to-strength ratio is maintained constant and if the beam is damaged before the creep test.

Basic creep of concrete is the part of the delayed deformation that occurs when concrete is loaded without drying. In this work results from the literature are analyzed: the derivative of the compliance function J of basic creep tests with different loading ages t0 are compared. It has been observed that considering dJ/dt as a function of time t the long term behavior is a logarithmic one with the same slope. Here it is shown that if we consider dJ/dt as a function of the age since loading (t-t₀) a unique behavior is observed. This indicates that basic creep could be expressed as a logarithmic function in which the age of loading affects only the term (t-t0). In the recent Model Code 2010 basic creep is exactly expressed like that. Using the experimental data the parameters of the model are calculated for each test and compared with the values proposed by MC2010. It is shown that MC2010 gives a rather good set of values for the two parameters but it could be improved by adjusting some parameters on experimental results.

The adiabatic temperature rise model of concrete is very important in simulated computations of temperature field of mass concrete. A combination model from different kinds of function models of adiabatic temperature rise of concrete was proposed to fit the adiabatic temperature rise test data of Ultra-High-Volume Fly Ash Conventional Dam Concrete (UHVFACDC) to obtain a reasonable adiabatic temperature rise model. This model can improve a 3-D FEM simulation program for calculating the temperature field of a dam whose material is UHVFACDC (content of fly ash reaches up to 80%). The 1.2m×1.2m×1.2m large cubic specimen of UHVFACDC was tested to obtain the temperature history curve in the middle of the specimen. The temperature history curve obtained by the FEM simulation for the large specimen was compared with the tested one. The two results coincided to a substantial degree which revealed that the obtained model is valid. Thus, the temperature history curve of UHVFACDC can be achieved. This model can improve the precision of temperature field simulation for the dam. It provides the temperature history curve as input for the experimental study on temperature-stress testing machine test of UHVFACDC which is crucial for the evaluation of crack resistance performance at early age.

This research presents a novel uniaxial compression test system for concrete specimens. The motivation for this development comes from the need to calibrate and validate multi-scale numerical methods for the analysis of concrete structures, a research field which requires new experimental systems for concrete. The aim of this study is to offer an experimental methodology and to develop an experimental uniaxial compression system which provides the results needed for the calibration and validation of multi-scale models. The uniqueness of the suggested experimental system is related to its ability to measure the lateral displacements of a specimen with a rectangular cross section. The main challenges to achieving reliable results using electro-mechanical means are concerned with attachment of the lateral gauges so that it can follow the targeted spot with progressing longitudinal deformation. The unique experimental system developed for this study was found to be accurate and was validated by comparing macroscopic experimental results such as: displacements, modulus of elasticity and Poisson’s ratio with the literature on concrete and aluminum specimens.

One of the main issues of the ongoing Andra’s research program is the development of construction methods and optimization of radioactive waste structures of the disposal facility. A new method for tunnel construction was tested in the underground research laboratory (URL) using a TBM with segment emplacement technique. The lining is composed of reinforced concrete segments, backfilled with two types of grouting mortar: a classical and a compressible grout. The major objectives of this experiment are to check the technological feasibility, and to analyze the hydromechanical behavior of the host rock and of the lining. Different investigation methods were used; the host rock and the concrete lining were instrumented, and the different grouts and concrete used are being characterized. The global data analyses, underline an anisotropic loading, which is more pronounced on the classical mortar zone. The major radial loading is identified in the horizontal direction which is consistent with the existing drifts along this orientation.

Even there are many influential parameters on both the shrinkage and the creep of concrete, one of the most important indices is the moisture movement that is affected by the concrete size as well as the relative humidity of environment surrounding the concrete. The objective of this study is to evaluate the difference of the creep coefficient of concrete by the location by monitoring used for the prestressed concrete (PC) girders of railway structures. Based on the monitored data, the value of creep coefficient was different according to the location: e.g. web, flange, and slab in PC girders.

A finite element model updating technique based on measured strain responses is used to monitor the health of prestressed concrete structures. In this work health corresponds to the existing prestressing force present and the associated stiffness properties of the structure. Measurements are taken on post-tensioned beams, both bonded and unbonded. The measured response is the total strain of the beams at predefined locations and the natural frequencies of the structures. In this work uncertainties in the measured data and those in the stiffness parameters are considered. The uncertainties are modeled using both Bayesian technique and interval analysis. The uncertain parameters in the finite element model are then estimated using Markov chain Monte Carlo technique in Bayesian framework and set inversion via interval analysis in interval analysis framework. The existing prestressing force estimated by both the mentioned techniques are then compared with the measured actual prestressing force present in the structure. To check the robustness of the proposed algorithm tests are conducted at different time spans corresponding to different prestress loss and different material stiffness.

While models to study prestress losses based on cylinder data are valuable for their flexibility and ability to investigate micro-phenomena, calibration of models based on girder data is also valuable for some applications. More specifically, data obtained from the monitoring of girders captures the effects of physical phenomena taking place within a girder section, effects that are not accurately captured with cylinder testing (e.g. irregular sectional geometry and variable stress history). While general methods are aimed at a broad range of cases, it is intuitive that less complexity is necessary in a method aimed to a singular type of structure. The development of such a method, for use on pretensioned, simply-supported girders is briefly summarized herein. This model provides simple numerical estimates for strain-related prestress losses, and retains the transparency of basic creep and shrinkage models. The model was calibrated using full-scale structural test data.

Metakaolin is one of the most attracting gradients of high and ultra high performance concretes which are used in different types of buildings such as nuclear waste storages, impact-resistant military buildings, fire-resistant high-rise buildings and reinforced concrete bridges. The metakaolin induced concretes are very dense and more homogeneous than normal strength concrete. However, these properties cause significant increase in brittleness of hardened concrete and change of volume during the hardening of these materials. In this work, cement pastes were produced at three different water/binder (w/b) ratios of 0.42–0.35 and 0.28 for which cement has been replaced by metakaolin with at different weight fractions of 8%, 16% and 24%. 48 hours volumetric autogenous shrinkage measurements showed that in pastes with 0.42 w/b ratios autogenous shrinkage reduced with increasing metakaolin content. In mixtures with 0.35 and 0.28 w/b ratios, on the other hand, it has been seen that the autogenous shrinkage increased with increasing the amount of metakaolin. Using thermo-gravimetric analysis and mercury porosimetry measurements, it has been found that with increasing metakaolin content the amount of calcium hydroxide decreased and critical pore size reduced. It has been determined that as a result of the pozzolanic reaction the amount of fine gel pores increased and those of the coarser pores significantly decreased at 28 days. One year linear autogenous shrinkage measurements showed that in pastes with 0.42 w/b ratios the autogenous shrinkage increased with increasing metakaolin content, while it decreased in pastes with lower w/b ratios.

Shrinkage in concrete structures may lead to cracks which reduce the durability of concrete against aggressive mediums. In prestressed concrete elements shrinkage results in a significant loss of prestressing forces. Limestone-rich cements with limestone contents beyond the values of DIN EN 197-1 were developed in order to reduce the influence of cement production on environment. In this paper the shrinkage behavior of mortar and concrete samples made of limestone-rich cements up to 70 wt.% with different water/cement-ratios was analyzed. The results indicate that shrinkage of such cements is strongly dependent on the amount, chemical and physical properties and the mineral components of limestone. The mortar samples made of such cements had shown both higher or lower shrinkage than reference samples made of pure Portland cement (CEM I 52,5 R) depending on the used limestone type. Specific surface area measured by BET-method seems to play a key role on the shrinkage behavior by controlling the disjoining pressure between hydration products. Meanwhile, it was found that in contrary to reference samples the prolonged curing in water does not influence the shrinkage of specimens made of limestone-rich cements significantly.

Steel Fibre Reinforced Concrete (SFRC) is currently being used as a structural material even though little is known about the creep behaviour of SFRC, specially the cracked tensile creep. In recent studies it has been revealed that even though the tensile creep of SFRC is similar to that of ordinary concrete, the creep increased significantly after the SFRC had cracked. This phenomenon can be associated with time-dependent crack widening that is caused by fibre pull-out over time. Single fibre pull-out rate and sustained load tests were performed to investigate the time-dependent pull-out behaviour of hooked-end steel fibres. Four pull-out rates ranging from 2.5 mm/s to 0.00025 mm/s were considered. The sustained load varied from 30 % to 85 % of the maximum (short term) pull-out resistance. Finally, an analytical study was conducted to predict the time-dependent pull-out behaviour.

In fiber reinforced concrete, fibers are able to transmit forces across a crack. Macrofibers can be used in structural applications to arrest crack growth and increase the long-term durability of the structural element. However, time-dependent phenomena such as creep can gradually widen cracks. In fiber reinforced concrete, this time-dependent crack widening can be mainly attributed to two mechanisms, fiber creep and gradual fiber pull-out out of the concrete matrix. This paper presents the results of an experimental campaign into these two crack-widening mechanisms for polypropylene fiber reinforced concrete. Fiber creep is evaluated at different load levels. The tests show that creep fracture may occur after mere days to several months for high and low load levels, respectively. Furthermore, the total strain at failure can easily exceed 50 % for low load levels. Pull-out experiments have been done on the same polypropylene fiber at different embedded lengths. The results show that the maximum measured force is not linear dependent on the embedded length of the fiber. Furthermore, post-peak pull-out behavior can be related to the surface profile of the fiber.

The early-age shrinkage of concrete has a pronounced influence on the performance of the material during the whole service life. The casting of concrete in hot climatic conditions, which are harsh for early-age concrete, often lead to cracking which subsequently has an negative impact on various properties of the structural element, especially on its durability. The article at hand provides some additional insights on plastic and early-age shrinkage of two very different types of concrete: an ordinary concrete and a relatively new type of fiber-reinforced concrete — a strain-hardening cement-based composite (SHCC). These materials were produced and tested both without and with addition of a superabsorbent polymer (SAP), which was used as an internal curing agent for mitigating early-age shrinkage. Three climatic parameters were varied in a certain range: temperature of the surrounding air, its relative humidity and the wind velocity. The addition of SAP clearly reduces the capillary pressure within fresh concrete, which indicates the efficiency of its action as internal curing agent. The SHCC is able to successfully withstand the early-age shrinkage in the hot climatic conditions under investigation. In contrary, in the most cases a major crack developed in the ordinary through the entire height of the specimen and separated it in two parts.

Early age (EA) drying shrinkage of cement based materials is a complex process because it comprises reaction evolution, microstructure development, stiffening and water evaporation due to environmental conditions. The occurrence of EA shrinkage can increase early age cracking risk which can compromise durability. Due to the larger amount of paste, moderate strength powder-type self-compacting concrete (SCC) is more prone to EA drying shrinkage than conventional concrete. To evaluate the effect of limestone filler, microsilica (MS), nanosilica (NS) and metakaolin (MK) on EA free drying shrinkage of cement pastes and SCC, an experimental program was carried out. Free shrinkage and mass loss were monitored on paste and SCC samples during 24 hours, subjected to surface desiccation to evaporate all the bled water during 6 hours, while temperature and Relative Humidity remained constant. It was observed that evaporation and free shrinkage were not related directly, although a physical relationship could be identified. Evaporation during EA shrinkage of pastes and SCC compositions showed similar values, while drying shrinkage depended on the presence of the aggregates. The effect of the paste components’ amount and type, the volumetric fraction of aggregates and the evaporation were assessed and some equations were proposed to estimate EA drying shrinkage of moderate strength powder-type self-compacting pastes and concretes.

A Predicting delayed strains (creep and shrinkage) in concrete proves to be critical to a large number of pre-stressed concrete structures, such as containment vessels of nuclear power plants, post-tensioned bridge girders, etc. These delayed strains include two components of shrinkage, namely, Autogeneous shrinkage and Drying shrinkage and two components of creep strains, namely, Basic creep and Drying creep. The present study focuses to predict the creep and shrinkage of Self Compacted Concrete (SCC) and High volume fly ash normal concrete experimentally. And a Hygro thermo chemo mechanical model (D.Gawin et al, 2006) is implemented to estimate gaseous pressure, capillary pressure and temperature and displacements which in turn varies with the hydration process. The Solidification theory proposed by Bazent (Bazant and Prasannan, 1989), is used to estimate the creep and shrinkage.

Use of blended cements is growing rapidly in the construction industry due to economical, ecological and technical benefits. Syria is relatively rich in volcanic scoria. In the study, three types of cement were prepared; one plain Portland cement CEM I (control) and two blended cements: CEM II/A-P and CEM II/B-P (EN 197-1), each of them with three replacement levels of scoria: (10%, 15%, 20%) and (25%, 30%, 35%), respectively. Evaluation of sulfate resistance of mortars containing scoria-based cements was performed in accordance with ASTM C1012. Length measurements were performed up to 52 weeks after immersing the specimens into 5% Na₂SO₄ solution. Drying shrinkage has been evaluated using mortar bar method in accordance with ASTM C596. Lengths of mortar bars subjected to drying shrinkage were measured periodically up to 360 days. SRPC cement was employed for comparison in both tests. Test results revealed that the increase of scoria significantly improves the sulfate resistance of mortars. The expansion of scoria-based mortar bars decreased with the increase of scoria replacement level. Further, according to the expansion test results, CEM II/B-P can be used instead of SRPC in sulfate-bearing environments. The results of drying shrinkage showed that the CEM II/B-P-based mortar bars exhibited greater contraction when compared to the control mortar, especially at early ages. However, the long-term drying shrinkage of mortars was not influenced much by the addition of volcanic scoria.

The long term behaviour of Fibre Reinforced Concrete on cracked state is an important research topic, since it is essential to understand this property when using FRC as structural material. The RILEM CCF Technical Committee was created in order to focus the efforts of all researchers on the better understanding of creep behaviour. One of the objectives of the TC was to compile a global database of results of creep test in order to analyse this phenomenon in an overall review. In order to achieve a complete enough database, several minimal variables and parameters data have been proposed on this first attempt. This first database of creep results is based on flexural creep test with different environmental conditions. In future, it can be extended to other testing methodologies. Thanks to this database, first conclusions start reveal those variables which have a higher significance level in creep deferred behaviour.

The mix-design of self-compacting concrete (SCC) with both fine and coarse concrete waste from construction and demolition (C&DW) is investigated. The effects of concrete recycled aggregates on fresh state behavior and short- and long-term mechanical and physical properties of new structural SCC are investigated to ascertain the feasibility of SCC with medium-high mechanical strength. A reference SCC mix, prepared with natural aggregates and the same water/powder ratio as those prepared with recycled aggregates, is reported for comparison. An integrated approach involving porosimetric analysis and mechanical characterizations has been carried out. Porosity measurements and mechanical investigations combined with time-dependent properties, such as shrinkage and creep are reported as fundamental features to assess structural SCC behavior.

Three 200x320x2300mm RC beams were tested under three different loading conditions to investigate the effects of loading rate on the shear performance of RC beam. Tested at loading rate 10mm/1hr, RC Beam 1 represents specimen under normal loading while RC Beam 2 represents specimen under slow loading at loading rate 10mm/100hr from 60kN. Tested at loading rate 10mm/100hr from 100kN, RC Beam 3 represents specimen under both normal and slow loading. The results showed that RC Beam 1, 2 and 3 had diagonal cracking load of 134.1kN, 133.6kN and 133.9kN and had ultimate failure load of 137.5kN, 146.2kN and 161.6kN respectively. Although these results indicated similar diagonal cracking strength, slow loading had influences on the ultimate failure load and crack pattern. The locations of diagonal bending crack for RC Beam 1, 2 and 3 were approximately 350mm, 530mm and 390mm from the center of the beam on the failure side. The location of bending cracks was affected by slow loading rate. As a consequence, slow loading had significant effects on the location of diagonal cracks so that it increased the ultimate failure load.

Concrete that has been damaged by alkali-silica reaction (ASR) is used in many cases still for a longer period of time. To calculate the residual strength and the deformation of ASR-damaged structures knowledge of the strength, of the elastic modulus and creep are required. The focus of most studies [ICAAR] is put on the damage mechanisms of expansion occurring, the crack formation and decreasing short-term strength. In addition, the concretes investigated were not made with slow/late aggregates in most cases. The research project has the goal to investigate the creep of concrete with slow/late aggregates. These react only after long-term (10 to 20 years). The slow/late aggregates stem from quartz porphyries (rhyolites), greywacke, quartzite, granite, andesite and high-grade crushed gravel from the Upper Rhine. The results presented in this study include concrete with different aggregates, namely tree alkali-sensitive ones (greywacke from the Lower Harz, quartz porphyry from the Halle Porphyry Complex and crushed gravel from the Upper Rhine valley) and one alkali-insensitive aggregate (quartz porphyry from the Black Forest).

Long-term excessive deformation of underground RC box culverts in service was monitored over 20 years and its mechanism is analytically discussed in this study. The long-term excessive deformation possibly attributes to synergy effects accompanying delayed shear failure of RC slabs subjected to vertical soil pressures and the time-dependent creep-shrinkage of structural concrete. Special attention is directed to the delayed shear cracking which was actually found in real underground box culverts over the service life. The delayed shear crack is experimentally reproduced in the laboratory as well as the multi-scale computational simulation.

Some concrete structures are serviced in lower high temperature (80~240 degree C), such as nuclear containment vessels or underground structure on geothermal area. However, the creep behaviour of concrete for these special structures remains as an issue that is neglected or not implemented consideration during the design process, which may lead to inappropriate estimation for the service behavior and durability to concrete structure. Although there are already many creep models for concrete, lots of them are based on the normal temperature (20 or 23 degree C). Even though the influence of high temperature is considered in some studies, most of them are focused on the short-term creep strain under extremely high temperature. On the other hand, the widely application of lightweight aggregate concrete (LAC) brings considerable beneficial results of economy and society due to its lighter dead-weight and better thermal performance. What’s more, the heat resistance of LAC is much better than the normal concrete. However, the high creep behavior of LAC is also a deterrent. Is LAC a suitable material for some special concrete structure under long term lower higher temperature? In this paper, the creep behaviour of LAC is studied at various temperatures ranging from 80 degree C to adopted maximal 240 degree C, together with the parallel test at the 20 degree C. The specimen is cylindrical, which is 100mm in diameter and 300mm in height. The compressive strength of the specimen at 28 day is 30 MPa, with the water binder ratio is 0.5. Four kinds of strain, such as the instantaneous stress-related strain, free thermal strain, transient strain and creep strain are obtained by unstressed tests, transient strain test and steady state creep test. The revised creep model will be developed to describe the sensitivity to the temperature rise under stress fields as a result of enhancement in the existing creep predication model. Finally, the microscopic observation for the specimen will be done by scanning electron microscope to obtain the microstructure evolution induced by thermal-chemo-mechanical effects. The creep behaviour of LAC in some high temperature range is obtained in this paper. Although the research is relatively conventional, it provides an additional accuracy of creep in terms of lower high temperature (80~240 degree C) in the long term for special concrete structures, which will be demonstrated in the cracking control and durability of some special infrastructure. What’s more, it also provides the possibility of thermal-chemo-mechanical analysis on the creep behaviour, and helps in analyzing the multi-scales model of the creep, which can be improved via an Arrhenius law.

In this paper long-term creep and shrinkage behavior of vertical dry cask structures are investigated and discussed. Environmental aging is considered in the form of deleterious mismatch effects of steel and concrete creep at the interface of two materials. Additionally, the performance in the case of a tip-over impact is studied. The aging simulations include creep and shrinkage predictions of the representative cask models for 40 years of service using the B4 creep formulation of concrete [RILEM, 2014] and the Bailey-Norton creep formulation for steel [Betten, J., 2008]. The results illustrate cask behavior for hygro-thermal-mechanical aging before it is subjected to tip-over impact. The time history results of creep and shrinkage illustrate the mismatch of deformations and residual stresses in the steel liner and contiguous concrete. Both, full bond and no bond cases are studied in order to assess progressive damage at the interface of the two materials. It is observed that the concomitant shear transfer induces significant damage in the interfacial region of the two cask components. Results of tip-over simulations show that the strains and stresses in the concrete are up to 20% different for young and aged concrete.

The long term analysis of structures and structural elements depends on accounting both for the creep stress relaxation and the development of structural cracks. The development of such cracks may take place over long periods and therefore be difficult to distinguish from the viscous effects. Modelling structures under such slow-loading conditions requires the development of appropriate numerical models for material behaviour, and novel strategies for computing the link between creep and damage. In this paper, we present a new empirical model combining both aspects. The creep component is represented here using a Maxwell chain, and the material laws are based on the modified compression field theory (MCFT) and the Mazars model. The effects of speed-of-loading are investigated, and the effect of creep on the post-peak behaviour is assessed. A particular attention is given to the implementation which integrates models developed with widely differing aims.