International Concrete Abstracts Portal

Stresses develop in portland cement concrete pavement at early ages when volume changes associated with hydration reactions, moisture loss, and temperature variations are restrained. Saw-cuts are placed in concrete pavements to provide a weakened plane that enables cracks to form as intended, thereby relieving developed residual stresses. Although the idea of creating a weakened plane by saw-cutting is relatively straight forward, practically determining the timing and depth of saw-cut can be complicated in field construction. This study uses a finite element model (FEMMASSE) to evaluate influence of saw-cut timing on cracking behavior of concrete pavements. The model considers the influence of ambient temperature, cooling effect of wind, and time of casting. It is shown that the saw-cutting time window was reduced as ambient temperature was increased. Higher wind speeds influence the saw-cutting time window to a lesser degree at high ambient temperatures than they do at lower ambient temperatures. It was also shown that the time of casting influences the saw-cutting time window and it needs to be considered in estimating the saw-cutting time window especially at high ambient temperatures.

A simulation algorithm was developed for modeling the dense packing of large assemblies of particulate materials (in the order of millions). These assemblies represent the real aggregate systems of portland cement concrete. Two variations of the algorithm are proposed: Sequential Packing Model and Particles Suspension Model. A developed multi-cell packing procedure as well as fine adjustment of the algorithm’s parameters were useful to optimize the computational resources (i.e., to realize the trade-off between the memory and packing time). Some options to speed up the algorithm and to pack very large volumes of spherical entities (up to 10 millions) are discussed. The described procedure resulted in a quick method for packing of large assemblies of particulate materials. The influence of model variables on the degree of packing and the corresponding distribution of particles was analyzed. Based on the simulation results, different particle size distributions of particulate materials are correlated to their packing degree. The developed algorithm generates and visualizes dense packings corresponding to concrete aggregates. These packings show a good agreement with the standard requirements and available research data. The results of the research can be applied to the optimal proportioning of concrete mixtures.

Alkali silica reaction (ASR) causes premature and unrecoverable deteriorations of numerous civil engineering structures. ASR-expansions and induced cracking can affect the functional capacity of bridges and dams. Several hydraulic dams of Electricité de France (EDF) are concerned by ASR. Therefore, a behaviour model implemented in a finite element code has been developed in order to assess the safety level and the maintenance choices of these degraded structures. This approach has the particularity of modelling the ASR structural effects from the construction of the structure until today. It uses several ASR advancement variables, one for each aggregate size range of the affected concrete. These advancement variables depend on both the saturation degree and the temperature in the dam. The difficulty of using a classical residual expansion test on core samples to fit the model is pointed out, particularly when the swelling rate is slow due to low alkali content in the concrete. Thus, the authors propose an original approach combining additional tests and physical modelling to assess the chemical advancement of the ASR for each aggregate size of the affected concrete. Only the chemical advancement, which is a normalized variable linked to the residual reactive silica content, is measured in laboratory. The concrete residual potential expansion is not measured on laboratory tests but fitted through an inverse analysis based on a finite element structural calculation.

The primary purpose of this paper is to introduce and demonstrate the applicability of a statistical concept, the average, for the modeling of the deformations of two-phase composites under load. Concrete is modeled as a well-compacted two-phase composite, the hardened paste as the matrix, and the aggregate as the dispersed phase. Only the paste has creep. The demonstration is done by the development of novel viscoelastic models and their mathematical equivalents for the instantaneous as well as time-dependent deformations of concrete, as a two-phase composite, under load. The underlying principle of the work is based on an extension of earlier publications by the writer in which averages of the averages of the related the phases, the composite averages are offered for the estimation of the modulus of elasticity of composites. Since experimental results supported the composite average method, CAM, quite well for this, it seemed worthwhile to investigate whether the method can be extended for the calculation of time-dependent deformations. The extension consists of the addition of dashboard elements to the existing composite average spring models for the modulus of elasticity of concrete, for the estimation of creep. This is the combinations of two existing spring-dash models for the calculation of the creep: the Poyinting-Thomson model with the Maxwell model the results of which are two CAMs that are determined by the type of combination between these two: one for normal-weight concretes when the two models are connected in a series, and the other, when they are in parallel for lightweight-aggregate concretes. Experimental data on creep with uniaxial loading taken from the literature support these composite models well. Among others, the data and the models show that during the period when the creep development is gradually decelerating: 1. creep values as a function of loading time, give straight lines, let us call them creep lines (compliance functions) in semi-log system as well as in log-log system of coordinates. Consequently, they can be approximated both by logarithmic as well as power functions. Such formulas are suitable for the estimation of creep at a later time from an earlier measurement; and 2. various creep lines of comparable concretes may be parallel, regardless at what age t' the loading started. It is shown that the new models are: well supported by experimental results within reasonable time limits; they are conceptually simple and logical; they are novel; they can consider the composition of the concrete; they represent both E and creep; and they are valid both normal-weight and lightweight-aggregate concretes.

The long-term durability of concrete is related to its ability to impede or reduce fluid transport. The long-term durability performance of concrete pavement can be dramatically influenced by the ingress of water or other fluids at saw-cut joints. Research is needed to better understand the role of complex geometries, like saw-cuts, on fluid transport. This paper uses x-ray attenuation to study the unsaturated fluid transport in systems containing a saw-cut (notch). The rate of water transport is greater in the direction perpendicular (i.e., horizontal) to the wall of the saw-cut when compared to the penetration below the tip of the saw-cut. This can be explained by the geometry of the source. To study the influence of fluid properties on transport, two fluids were tested with dramatically different viscosities and surface tensions. The results indicate that for the solution with higher viscosity and lower surface tension the absorption rate is reduced significantly. A finite element based code (Hydrus) is used to simulate the unsaturated flow based on solution of Richard’s equation. Results of simulations show good agreement with experimental results and confirm the effects of the geometry of the saw-cut on fluid transport.