International Concrete Abstracts Portal

The action of high temperature in concrete is a field of much interest and attention due to its strong influence in strength, durability and serviceability conditions. Long-term exposures to high temperature fields strongly affect the most relevant mechanical properties of concrete materials such as cohesion, friction, stiffness and strength. In this work, two alternatives approaches for the analysis of failure behavior of concrete subjected to high temperatures are discussed and their predictions analyzed. Specifically, a thermodynamic gradient poro-plastic model based on the continuous or smeared-crack approach and an interface model based on the discrete crack approach are developed. After describing the main aspects of both models, this work focuses on the analysis of their results in terms of the degradation of concrete durability and strength capacities when subjected to severe thermal fields. The results demonstrate the comparative advantages of the discrete approach to analyze at both the macroscopic and mesoscopic scale the complex degradation processes of concrete constituents at high temperature, thanks to the robustness, stability and overall simplicity of the discrete model approach. Furthermore, the results show the capabilities of the continuous model to analyze the durability degradation of concrete at material level.

The durability of concrete structures is particularly susceptible to aggressive environments, in particular to the penetration and diffusion of chloride ions. Hence, a reliable prediction of the chloride diffusivity is mandatory to schedule efficient maintenance as well as to estimate the service and ultimate life of concrete structures. This is a non-trivial task because the chloride diffusion process is clearly a multiscale problem since it is influenced by different factors acting at different length and time scales, including the ability of some phases of the hardened cement paste (HCP) to interact with chloride ions. In the present work the chloride diffusivity of HCP is numerically simulated using a modified version of Fick’s law accounting for the ability of some HCP phases to bind chloride ions. The 3D HCP microstructures for the analyses are generated artificially, using the software CEMHYD3D, as well as segmented starting from real X-ray images, and in all cases are discretized using a voxel-based mesh. The effective (homogenized) coefficient of diffusivity, to be used for mesoscale analyses, is obtained through upscaling and is validated using data from the literature. Finally, comparisons between real and artificially generated HCP microstructures are performed and discussed.

In many countries, structures in reinforced concrete “of historical interest” are covered by preservation legislation. In striving to restore them, scholars make use of knowledge accumulated over time. Less well known is the technological research that was part of the production and use of cements and concrete mixtures for reinforced concretes, whose durability has always been a prime concern. Historic works bear witness to their ability to last over time and to the ways in which structures and materials age and deteriorate, thus providing evidence as to the validity of the expectations of durability which existed when the work was designed. The systematic collation of data relating to such artefacts and the repairs they have undergone would be of great use (e.g. with regard to the components used in the original work and in the repairs). Furthermore, collaboration with manufacturing companies and research laboratories should allow us to make use of recently-developed prepacked mortars and concrete in new repair work, assessing their compatibility with old materials and monitoring their performance over time. The resultant database and experimental results would provide clues useful in moving beyond current rudimentary practices, laying the basis for a shift from “concrete repair to concrete conservation”.

Service life calculations are now treated more in detail in standards and in particular in fib Model Code 2010. The new performance is introduced in addition to probabilistic calculations. In spite of the advances made it is necessary to apply the new approaches with care because any of the existing models have been calibrated in the same concrete more than a couple of decades and then, the predictions may present substantial errors. In addition, the calculation of the propagation period and the limit state of corrosion are aspects not well defined as the MC2010 does not considers any model for the propagation stage. In present paper are analysed some aspects of service life models proposing improvements. A universal statistical distribution of chloride threshold is presented and the consideration of an Initiation Limit State as defined by ISO 13283 is proposed for the depassivation onset. Finally, are illustrated trough examples that the probability of failure for deterioration processes should not be a fixed value but it would depend on the rate of deterioration.

In the present work, a coupled environmental-mechanical damage model for structural analysis of RC elements subjected to aggressive action - originally developed by some of the authors - is presented and further enhanced, introducing some innovative formulations. In particular the effect on structural performance of rebars corrosion, induced both by chloride attack and carbonation, and of freeze-thaw cycles is analyzed. To this aim, the environmental damage parameter is re-formulated and splitted in two contributions in order to better represent the degradation of concrete caused by cracking due to different processes (e.g. development of expansive products during corrosion, pressure induced by internal ice formation). The proposed models are implemented in the finite-element framework OpenSees, developed at Berkeley, University of California and validated by comparison with a number of experimental tests. In the first part of the paper the proposed constitutive models are introduced discussing the most relevant features and characteristics. Then, in the second part of the paper, the validation tests are presented and the obtained results are compared with experimental ones, proving that the model is suitably accurate in reproducing the main aspects observed during experiments: i.e. failure load, ultimate displacement and failure mode.