E.A.B. Koenders

NUMERICAL MODELLING OF AUTOGENOUS SHRINKAGE OF HARDENING CEMENT PASTE

Hardened cement paste is a porous material. As hydration proceeds, pores become emptied and the relative humidity reduces. This reduction of the relative humidity goes along with a reduction of the pressure in the emptied pore space. Thermodynamic equilibrium requires an increase of the surface tension in the boundary layer that develops at the inner pore wall area. For validation of the model, experiments have been carried out. Good agreement is reached between the numerical simulations and the results obtained from experiments.

Modelling of Transport Phenomena at Cement Matrix—Aggregate Interfaces

The performance of a heterogeneous material like concrete is largely determined by the many interfaces in this material. This contribution focuses on the potential of numerical simulation models to investigate the character of the matrix-aggregate interfacial zone and to simulate hydration-induced moisture transport from the water-rich interfacial zone to the drying bulk paste. Typical features of the simulation model are presented, as well as results of the numerical analysis of the effect of moisture transport within the hardening paste.

Numerical simulation of hydration-driven moisture transport in bulk and interface paste in hardening concrete

Abstract In real concrete two types of cement paste can be distinguished, i.e., bulk paste and interface paste. Initially the paste in the interface zone will generally contain more water than the bulk paste and will therefore hydrate differently. Differences in relative humidity and associated differences in pore water pressure will result as well. If the interface paste and the bulk paste could hydrate individually, a situation will result where a relatively porous water-rich interfacial zone coexists with a relatively dry bulk paste. However, due to gradients in porosity, permeability, relative humidity and pore water pressure, a flow of moisture will start from the water-rich interfacial zone to the bulk paste. It will be shown how the moisture transport can be simulated numerically and how this transport phenomenon influences the overall rate of hydration of cement in concrete. Numerical results are compared with experimental data presented in literature. The relevance of modelling of this kind of transport phenomena is briefly dealt with.

Towards Simulation of Fresh Granular-Cement Paste Material Behavior

The mechanical performance and durability of hardened granular- cement paste material such as mortar and concrete are affected by its earlier stage behavior which is defined in terms of consistency and/or workability. Therefore understanding the behavior of the mixture during this early period is a crucial requirement for the technical and economical success of professional production sites. This paper presents a numerical approach based on Discrete Element Method (DEM) to simulate the macroscopic behavior of a fresh granular-cement paste mixture in which the pasty behavior is dominant. The research approach is based on a conceptual idea where the particle-paste interaction system is explicitly modelled by means of a two phase particle interaction system. The effect of the characteristics of paste on the macroscopic fresh behavior of a mixture has been studied and compared by the mini slump cone test regarding the excess paste theory. Modelling and experimental laboratory test results show good agreement.

Rheological investigation of specific interactions in Na Alginate and Na MMT suspension

Here we report on a study of a rheological behavior of sodium alginate and montmorillonite suspension. We find that viscoelastic behavior of this suspension is dramatically affected with increasing volume fraction of montmorillonite platelets. Addition of montmorillonite generally leads to gel formation, which is attributed to interactions of montmorillonite and alginate via H-bonding and attraction between the positive edges of the platelets and the anionic backbone of the biopolymer. A critical concentration for the measured system was observed at 20wt.% montmorillonite, where a crossover to a gel-like structure was detected. The observed gel has a rubber plateau, which develops further with higher montmorillonite concentration. In this physical gel the relaxation maximum was detected, which is associated with the breaking and reformation of the bonds between the platelets and the biopolymer. For this transient behavior, we find that a Maxwell type viscoelasticity quite well describes the relaxation time and the observed G-G crossover. We believe that this gel-like behavior plays an important role in formation of highly ordered nanostructures that develop during the drying of these bio-nanocomposite suspensions.

Modelling Rheological Behavior of Fresh Cement-Sand Mixtures

In this paper, the effect of the mix composition on rheological behavior of fresh cement-sand mixture is studied by considering a mixture as a two-phase model that is decomposed into a granular and a paste phase. The paste itself is subdivided into void paste and excess paste. Void paste fills the void space between the grains when they are in a fully compacted state while excess paste will use the remaining paste to form a paste layer around each individual grain particle, with equal thickness. By considering each grain particle covered with the excess paste layer as a single element, the rheological behavior of cement-sand mixtures can be related to their inter-element interactions for all sets of particle combinations.Copyright

Representative Volumes for Numerical Modeling of Mass Transport in Hydrating Cement Paste

Representative elementary volume (REV) has major importance in numerical estimation of effective transport properties of porous materials. The increasing focus on the durability and reliability aspects of cementitious materials, calls for a better understanding of the mass transport phenomenon through their evolving porous microstructure. A multi-scale nature of the cementitious materials imposes a great challenge to modeling efforts. This paper investigates the REV size for numerical modeling of transport in hydrating cement paste. Numerous series of virtual 3D microstructures with different porosities and capillary pore morphologies were generated using Hymostruc platform, a numerical model for cement hydration and microstructure development. Effective diffusion was obtained by using a finite difference scheme. The effect of numerical resolution was also investigated. Based on a statistical analysis, it was concluded that the REV size depends on the complexity of the pore morphology, which further primarily depends on the porosity and employed numerical resolution.

Mitigating Resolution Problems in Numerical Modeling of Effective Transport Properties of Porous Materials

Generating a realistic microstructure for a computational grid, which accurately represents heterogeneities while maintaining a computational efficiency, is still an unsolved problem. This paper presents a novel multi-scale approach to mitigate the resolution problems in numerical methods for calculating effective transport properties of porous materials using 3D digital images. The method up-scales sub-voxel information of the fractional occupancy of interface voxels (i.e. voxels containing phase-boundary) in order to increase the accuracy of the pore schematization and hence the accuracy of the numerical transport calculation as well. The method is validated on a simple periodic arrangement of mono-sized particles. The numerical algorithm is implemented within the Hymostruc platform (numerical model for cement hydration and microstructure development) and backed up with a 3D graphics visualization. The new approach significantly reduces computational efforts, is easy to implement, and improves the accuracy of the transport property estimation.

Numerical Algorithm for Pore Size Distribution Characterization of Materials from 3D Images

This paper presents an image based numerical method proposed to obtain information regarding pore structure and organization of pores within materials based on 3D digital image input. The output of the numerical algorithm is a pore size distribution of materials. The algorithm is based on the combination of the two digital image processing algorithms: 1) a medial axis thinning algorithm to obtain 3D skeleton of the pore structure, and 2) the distance transform of an image. The method is tested on simple 2D and 3D microstructures of packed spheres, demonstrating the performance of the proposed method.

Hydration Process of Portland Cement Blended with Silica Fume

The enormous carbon footprint associated with the global cement production (5-7%) asks for a radical change in the use of sustainable replacement materials in concrete. Replacement of cement by pozzolanic waste materials, being a by-product from industrial processes, has been widely recognized as the most promising route towards sustainable construction materials. This paper presents experimental study on hydration of commercial Portland cement blended with silica fume in replacement ratio of 15 mass %. Isothermal calorimetry was employed to monitor the hydration kinetics. Thermogravimetric analysis coupled by differential scanning calorimeter (TG/DSC) was used to investigate the formed hydration products at 1, 3, 7, and 28 days of hydration. Two different approaches for a dispersion of silica fume in cement paste were compared: ultrasound bath and addition of superplasticizer (polycarboxylic ether based).