Guang Ye

Study on the development of the microstructure in cement-based materials by means of numerical simulation and ultrasonic pulse velocity measurement

The formation of microstructure in cementitious materials was simulated with a numerical model. Simulation results have been verified by measuring the evolution of the ultrasonic pulse velocity (UPV). In this contribution, the applied computer-based cement hydration model is presented. The UPV measurements are also presented and evaluated. Experiments were performed on concrete mixtures with water/cement ratio 0.40, 0.45 and 0.55. The concrete was cured isothermally at 10, 20, 30 and 40 °C. Correlations between the development of the microstructure and the evolution of UPV were found. Two critical processes were individuated. The first is the percolation threshold of the solid phase. The second is the full connectivity of the solid phase. Both in the experiments and in the numerical simulations it was possible to distinguish these critical stages. These stages are discussed and conclusions are drawn regarding the potential of numerical simulation models in the study of early age cementitious materials for quantitative analysis of hydration processes.

THREE-DIMENSIONAL MICROSTRUCTURE ANALYSIS OF NUMERICALLY SIMULATED CEMENTITIOUS MATERIALS

In order to predict the transport properties of porous media, such as permeability and electrical conductivity of cementitious materials, a better understanding of the microstructural characteristics, including the geometrical and topological properties, is required. In this contribution, the microstructure of cementitious materials is simulated by using the cement hydration model HYMOSTRUC. In this computer-based numerical model, the hydrating cement grains are modeled as gradually growing spheres, which become in contact while growing. The simulated porous medium can be described as a series of sections taken from three orthogonal directions, in which each unit (pixel) is filled either with a solid or a fluid phase (pores). Various algorithms based on a random walk process are utilized to determine the local geometrical information, such as gravity centers coordinate, perimeter and area of each individual pore. The percolating path of the fluid in three dimensions is traced by using an overlap algorithm. Both three-dimensional (3D) geometrical information and topological space characterization including branch node network and genus of the pores are derived. Calculation results of these algorithms are compared with results obtained by other microstructural models at various degree of hydration.

EXPERIMENTAL STUDY AND NUMERICAL SIMULATION ON THE FORMATION OF MICROSTRUCTURE IN CEMENTITIOUS MATERIALS AT EARLY AGE

Abstract The formation of microstructure in early age cement paste and concrete was examined with an ultrasonic experimental set-up. Research parameters included the influence of curing temperature (isothermal curing at 20, 30 and 40 °C), water/cement ratio (0.40, 0.45 and 0.55) and amount of aggregate. In parallel with the experiments, the cement hydration model HYMOSTRUC was utilized to simulate the formation of the microstructure. In this study, the cement paste was considered as a four-phase system consisting of water, unhydrated cement, hydration products and that part of the hydration product that causes the contact between the hydrating cement grains (so called “bridge volume”). A correlation has been found between the growth of bridge volume calculated with the model and the changes in the pulse velocity. It is believed that ultrasonic pulse velocity (UPV) measurements can represent a valuable tool to investigate the development of the microstructure at early age.

Influence of Boundary Conditions on Pore Percolation in Model Cement Paste

Fresh model cement mixtures, with the same w/c ratio and particle size distribution, were simulated by the SPACE system that is based on a dynamic mixing algorithm. Thereupon, they were hydrated by the HYMOSTRUC 3D system. Boundary conditions were varied, rendering possible assessment of their influence on percolation of capillary porosity by serial sectioning and using the overlap of slices. Simulation results revealed increases in total porosity and in connected fraction of capillary pores due to the existence of aggregate. The de-percolation threshold of capillary porosity was found not only related to total porosity and image resolution, but also governed by the spatial distribution of capillary pores.

The effect of activating solution on the mechanical strength, reaction rate, mineralogy, and microstructure of alkali-activated fly ash

Alkali-activated fly ash (AAF) is a promising material that exhibits comparable material properties as cement-based materials but with much less CO2 emission. In the present work, the effect of activating solution (SiO2 and Na2O content) on the performance of AAF was studied by means of isothermal calorimetry and X-ray diffraction analysis. Meanwhile, the pore structure of AAF was examined by mercury intrusion porosimetry combined with environmental scanning electron microscope. The results indicate that increasing the sodium oxide content leads to a higher extent of reaction, denser matrix and higher possibility of crystallization, corresponding to a higher compressive strength of AAF. The addition of silica in the alkaline solution retards the reaction rate and zeolite formation, while improves the microstructure of the matrix. Therefore, there is an optimal value for SiO2 with respect to the Na2O content for the AAF in this study.

Effect of Moisture Exchange on Interface Formation in the Repair System Studied by X-ray Absorption

In concrete repair systems, material properties of the repair material and the interface are greatly influenced by the moisture exchange between the repair material and the substrate. If the substrate is dry, it can absorb water from the repair material and reduce its effective water-to-cement ratio (w/c). This further affects the hydration rate of cement based material. In addition to the change in hydration rate, void content at the interface between the two materials is also affected. In this research, the influence of moisture exchange on the void content in the repair system as a function of initial saturation level of the substrate is investigated. Repair systems with varying level of substrate saturation are made. Moisture exchange in these repair systems as a function of time is monitored by the X-ray absorption technique. After a specified curing age (3 d), the internal microstructure of the repair systems was captured by micro-computed X-ray tomography (CT-scanning). From reconstructed images, different phases in the repair system (repair material, substrate, voids) can be distinguished. In order to quantify the void content, voids were thresholded and their percentage was calculated. It was found that significantly more voids form when the substrate is dry prior to application of the repair material. Air, initially filling voids and pores of the dry substrate, is being released due to the moisture exchange. As a result, air voids remain entrapped in the repair material close to the interface. These voids are found to form as a continuation of pre-existing surface voids in the substrate. Knowledge about moisture exchange and its effects provides engineers with the basis for recommendations about substrate preconditioning in practice.

3D Lattice Fracture Model: Application to Cement Paste at Microscale

The fracture processes in cement paste at microscale are simulated by the 3D lattice fracture model based on the microstructure of hydrating cement paste. The uniaxial tensile test simulation is carried out to obtain the load-displacement diagram and microcracks propagation for a Portland cement paste specimen in the size of 100×100×100 µm3 at the degree of hydration 69%. The Youngs modulus, tensile strength, strain at peak load and fracture energy are computed on the basis of the load-displacement diagram.

3D Lattice Fracture Model: Theory and Computer Implementation

The lattice fracture model is presented in this paper, which is intended to simulate the fracture processes in multiphase materials to obtain the mechanical behavior in terms of load-displacement diagram and the cracks propagation. The basic procedures of lattice fracture analysis is that imposing a prescribed displacement on a lattice structure, finding the critical lattice element with the highest stress/strength ratio, removing it from the system and repeating until the system fails globally. One of the challenges in computer implementation of 3D lattice fracture model is the huge demand for computer memory. Matrix free technique is adopted to solve this problem.

Modeling Framework for Fracture in Multiscale Cement-Based Material Structures

Multiscale modeling for cement-based materials, such as concrete, is a relatively young subject, but there are already a number of different approaches to study different aspects of these classical materials. In this paper, the parameter-passing multiscale modeling scheme is established and applied to address the multiscale modeling problem for the integrated system of cement paste, mortar, and concrete. The block-by-block technique is employed to solve the length scale overlap challenge between the mortar level (0.1–10 mm) and the concrete level (1–40 mm). The microstructures of cement paste are simulated by the HYMOSTRUC3D model, and the material structures of mortar and concrete are simulated by the Anm material model. Afterwards the 3D lattice fracture model is used to evaluate their mechanical performance by simulating a uniaxial tensile test. The simulated output properties at a lower scale are passed to the next higher scale to serve as input local properties. A three-level multiscale lattice fracture analysis is demonstrated, including cement paste at the micrometer scale, mortar at the millimeter scale, and concrete at centimeter scale.

A 3D Lattice Modelling Study of Drying Shrinkage Damage in Concrete Repair Systems

Differential shrinkage between repair material and concrete substrate is considered to be the main cause of premature failure of repair systems. The 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 repair systems. In this work, a lattice type model was used to simulate first the moisture transport inside a repair system and then the resulting damage as a function of time. 3D simulations were performed, and damage patterns were qualitatively verified with experimental results and cracking tendencies in different brittle and ductile materials. The influence of substrate surface preparation, bond strength between the two materials, and thickness of the repair material were investigated. Benefits of using a specially tailored fibre reinforced material, namely strain hardening cementitious composite (SHCC), for controlling the damage development due to drying shrinkage in concrete repairs was also examined.