Effective diffusion coefficient of chloride in porous concrete

A simplified diffusion model is given in the present paper to represent the chloride diffusion in porous concrete. The model with appropriate effective diffusion coefficient is shown to represent equally well as the more complex diffusion/reaction model proposed by Pereira and Hegedus.⁴ However the present model has the main advantages of being amenable to an analytic solution and ease of model parameter estimation.     

Post-peak cyclic behaviour of concrete in uniaxial tensile and alternating tensile and compressive loading

Deformation controlled uniaxial tests have been carried out on cylindrical concrete specimens in order to establish the complete stress-deformation curve in tensile loading. Furthermore, cyclic tests have been performed under three types of loading: cycling from a tensile low stress, cycling from a compressive lower stress, and cycling between a lower and an upper stress to the envelope curve, respectively. The results seem to show that the tensile envelope curve is unique and the residual strength is the more affected the lower the stress at unloading is.     

Determination of chloride diffusion coefficient of concrete using open-circuit potential measurements

The rate of chloride ion ingress into concrete is of great importance for the performance of reinforced concrete structures exposed to chloride-contaminated environments. The service life of reinforced concrete structures subjected to such exposure conditions is closely related to the rate of chloride ion diffusion through the concrete. This paper presents the determination of the apparent chloride diffusion coefficient of concrete using open-circuit potential measurements. The chloride diffusion coefficients obtained are in the range of 6.4×10⁻⁸ to 12.4×10⁻⁸ cm²/s for a simulated seawater tidal condition, which is quite consistent with those reported in the literature. This indicates that open-circuit potential measurements can be considered as an approximate but simple method of assessing the diffusivity of chloride through concrete. Limited with the testing conditions and the characteristics of concrete used, results indicated that the time necessary for corrosion initiation of concrete with a cover depth of 7 cm ranges from 3 to 6 years for the seawater exposure, whereas it is only 1.5 years for a 3% sodium chloride exposure.     

On the mathematics of time-dependent apparent chloride diffusion coefficient in concrete

This paper provides an improved mathematical analysis of chloride penetration into concrete employing a time-dependent diffusion coefficient for the solution of Fick's second law of diffusion. In the paper the possible errors caused by the application of oversimplified mathematical expressions used in some models for the evaluation of service life of reinforced concrete structures are discussed. The results from this mathematical analysis demonstrate that some models based on the oversimplified error function complement (ERFC) solutions may easily overestimate the service life by orders of magnitude, especially when the age factor is high. Some chloride profiles after up to 10 years' field exposure were used to compare the oversimplified with the improved models. The results show that both the oversimplified and the improved models fairly well predict the 10 years' chloride ingress in Portland cement concrete, but the oversimplified ERFC model significantly underestimates the chloride ingress in concrete with fly ash.     

Properties of confined fibre-reinforced concrete under uniaxial compressive impact

The effect of confinement on plain and fibre-reinforced concrete (FRC) prisms subjected to uniaxial compressive impact was studied. It was found that the response of the material changed with the degree of confinement. Confined concrete exhibited more ductile behaviour, with both strength (f_c′) and ultimate strain (ε_ult) increasing with the degree of confinement. However, the elastic modulus (E) of the confined specimens was found to be about the same as or slightly lower than those of unconfined prisms. In addition, the relationship between stress and stress rate (n value) was also determined. It was found that, with confinement, the material became more rate sensitive.     

Influence of aggregates on chloride diffusion coefficient into mortar

In order to determine the influence of aggregates on chloride ion ingress, mortar specimens were cast with different aggregate size distribution and the same aggregate volume content. One side of the specimen was in contact with alkaline solution containing sodium chloride. The process of chloride ingress is due to diffusion. The effect of the interfacial transition zone (ITZ) around aggregate and of the tortuosity due to aggregates have been taken into account for interpreting the experimental data obtained. These two competing effects have been quantified by using the theory of composite materials. It appears that the diffusion coefficient varies as a function of the ITZ volume content and of the tortuosity.     

Mechanical properties of recycled aggregate concrete under uniaxial loading

In this paper, the compressive strength and the stress-strain curve (SSC) of recycled aggregate concrete (RAC) with different replacement percentages of recycled coarse aggregate (RCA) are investigated experimentally. Concrete specimens were fabricated and tested with different RCA replacement percentages of 0%, 30%, 50%, 70% and 100%, respectively. Uniaxial compression loading is applied in the experiments. Special attention of the analysis is devoted to the failure behaviour and the influences of the RCA contents on the compressive strength, the elastic modulus, the peak and the ultimate strains of RAC. Analytical expressions for the peak strain and the stress-strain relationship of RAC are given, which can be directly used in theoretical and numerical analysis as well as practical engineering design of RAC structures.     

Correlation between water permeability of latex-modified concrete (LMC) and water diffusion coefficient of latex film

Cement-based materials are generally known as weak materials in flexure and tension in comparison with compression. Polymers are used in cement-based materials to improve their flexural and tensile behaviors. The composite is called as polymer-modified concrete/mortar. Furthermore, polymers decrease permeability of water into cementitious matrices. Polymers are usually used as admixtures in concretes in form of latexes. Latexes are water-based polymers, which are consistent with water-based concrete matrices. On this basis, these kinds of products are called latex-modified concretes (LMCs). However, it has been found that chemical composition, particle size distribution, molecular weight, physical/mechanical properties of latexes affect performance of modified concretes. In this investigation, six latexes in three categories (acrylic, SBR and polyvinyl acetate) were used as concrete admixtures. They were characterized for chemical composition (by FTIR analysis), minimum film formation temperature, pH, glass transition temperature (T _g), particle size and particle size distribution to evaluate the effect of each property on LMC performance. Due to the formation of latex film in the microcracks and pores of concrete microstructure, it was suggested that diffusion of water into films controls permeability of whole concrete structures. On this basis, the diffusion coefficient of the latex films subjected to water was measured using a new method (continuous FTIR analysis). Capillary water absorption test was performed on concrete specimens to verify validity of the suggestion. It was found that there is a correlation between capillary water absorption of LMCs and water diffusion coefficient of latex films.     

Aggregate influence on chloride ion diffusion into concrete

An attempt is made to predict the probable effect of the aggregate on chloride ion diffusion into saturated concrete. It is shown that if the chloride ion diffusion coefficient of an aggregate ranges from 0.2 to 10 times that of the cement paste matrix, then this could result in variations in the concrete chloride ion diffusion coefficient of up to 10:1. Such a variation is equivalent to a change in free water-cement ratio from 0.77 to 0.45.     

Influence of traversing crack on chloride diffusion into concrete

This study examined the effects of traversing cracks of concrete on chloride diffusion. Three different concretes were tested: one ordinary concrete (OC) and two high performance concretes with two different mix designs (HPC and HPCSF, with silica fume) to show the influence of the water/cement ratio and silica fume addition. Cracks with average widths ranging from 30 to 250 μm, were induced using a splitting tensile test. Chloride diffusion coefficients of concrete were evaluated using a steady-state migration test. The results showed that the diffusion coefficient of uncracked HPCSF was less than HPC and OC, but the cracking changed the material behavior in terms of chloride diffusion. The diffusion coefficient increased with the increasing crack width, and this trend was present for all three concretes. The diffusion coefficient through the crack D_cr was not dependent of material parameters and becomes constant when the crack width is higher than  80 μm, where the value obtained was the diffusion coefficient in free solution.     

Effective diffusion coefficient for the gas contained in closed cell polyethylene-based foams subjected to compressive creep tests

In this paper, a new method to obtain the effective diffusion coefficient of the gas contained in closed cell foams under static loading is presented. Compressive creep experiments were performed on six low-density polyethylene (LDPE) and two polyethylene-co-vinyl acetate (EVA) foams of different densities using a home-designed compressive creep apparatus. The modelling of the evolution of pressure inside the cells was performed using an isothermal compression model. The effective diffusion coefficients were obtained from pressure decrease with time using an analytical solution of the diffusion equation. The values obtained agreed with those in literature and were dependent on both foam density and chemical composition.     

Effect of crack width on chloride diffusion coefficients of concrete by steady-state migration tests

The purpose of the present study is to explore the diffusion characteristics of cracked concrete according to the width of cracks. Major test variables include crack width, concrete strength, fly ash addition, and maximum aggregate size. The diffusion characteristics have been measured by steady-state migration test. The present study indicates that the diffusion coefficients do not increase with increasing crack widths up to the so-called “threshold crack width.” The threshold crack width for diffusion is found to be around 55–80 μm. Above this threshold value, the diffusion coefficients start to increase with crack width. A composite model with the introduction of “crack geometry factor” was derived to identify the diffusion coefficient in cracked concrete. It was shown that the crack geometry factor ranges from 0.067 to 0.206. Finally, the effects of concrete strength, fly ash addition and maximum aggregate size on diffusion coefficients are also discussed.     

Development of the map cracking in concrete under compressive loading

Preliminary to the modeling of the fracture of concrete, it is necessary to make observations at low scale for a better understanding of the phenomena. In this paper, the author shows how to follow the development of the microcracks in a sample during a compressive loading experimentation by the mean of non-destructive takings of replicas observed under S.E.M. The quantification by the stereological method of total projection. Enables to establish a relation between the observed damages and the global behaviour of concrete.     

Study of chloride binding and diffusion in GGBS concrete

Ordinary Portland cement (OPC) and OPC/ground granulated blastfurnace slag (GGBS) 70%, with or without 5% sulfates, were widely investigated in respect to their pore structures, chloride diffusion coefficients, internal and external chloride-binding capabilities by expression method and leaching method and the microstructure analysis on Friedel's salt such as differential thermal analysis (DTA), X-ray diffraction (XRD) and scanning electron microscopy (SEM). It can be concluded that GGBS can improve the pore structure of OPC and decrease the chloride diffusion coefficient greatly, and that sulfates do not do good for the pore structure and chloride diffusion for GGBS. GGBS increases the chloride-binding capability greatly without reference to the internal or external chloride and sulfates decrease the chloride-binding capability of GGBS greatly. It can be also found that the maximum intensity peak corresponding to Friedel's salt appears at about 8.0 Å, its endothermic effect appears at about 360 °C, its morphology is hexagonal slice whose size is about 2-3 μm; that the output of Friedel's salt formed by GGBS is much more than OPC; and that sulfates influence the output of Friedel's salt greatly. The corresponding mechanism was also analyzed.     

Water permeability and chloride penetrability of high-strength lightweight aggregate concrete

This paper presents an experimental study on the water permeability and chloride penetrability of high-strength lightweight concrete (LWC) in comparison to that of high-strength normal-weight concrete (NWC) with or without silica fume. The results were also compared with those of the concrete at a normal-strength level of about 30-40 MPa. In order to compare the water permeability and chloride-ion penetrability, LWC and NWC had the same proportion by volume. The only difference between them was the coarse aggregate used. The results indicated that at the strength level of about 30-40 MPa, the water permeability of the LWC was lower than that of the corresponding NWC. However, the water permeability of the high-strength LWC and NWC was of the same order. The resistance of the LWC to the chloride penetration was similar to that of the corresponding NWC both in the normal-strength and high-strength levels. As the compressive strength of the LWC was lower than that of the corresponding NWC, the results indicated that for a given 28-day strength, the LWC would probably have high resistance to water and chloride-ion penetration than the NWC. The results indicated that the resistance to the chloride penetration does not seem to be correlated to the water permeability of the concrete. There appears to be, however, a direct relationship between the rapid chloride penetrability determined by ASTM C1202 and the observed chloride penetration depth determined by the immersion and salt ponding tests.     

Capillary suction and diffusion model for chloride ingress into concrete

A numerical approach, named TransChlor, is proposed to simulate transport phenomena of various substances in concrete. This approach is a theoretical model based on finite elements and finite differences methods. The model consists of coupled nonlinear partial differential equations based on Fick's diffusion law and on kinematics equations. Simulation results from a parametrical study highlight the influence of microclimatic conditions, exposure to deicing salts and concrete cover permeability and thickness on chloride ingress in concrete. The results show that the chloride ion concentration increases quickly in concrete cover when a structure is exposed to deicing salts at a mountainous location; whereas permeability of concrete cover is an insignificant parameter when the concrete is in direct or splash water contact.     

The use of bulk diffusion tests to establish time-dependent concrete chloride diffusion coefficients

The proper determination of chloride diffusion values, including how they change with time, is important for service life modelling. Currently, there are two major approaches for using chloride diffusion coefficients to predict the service life of structures. The average diffusion coefficient or the instantaneous diffusion coefficient can be used. Using instantaneous diffusion coefficients is a more flexible technique, but requires a more advanced evaluation of bulk diffusion test results to establish material parameters. This paper describes an analytical procedure for determining the instantaneous chloride diffusion value as a function of time for a concrete using data from bulk diffusion tests. The importance of interpreting the data correctly is illustrated with simulated bulk diffusion test data, generated using a finite-difference model for diffusion. In addition, the application of this procedure for evaluating diffusion values is illustrated with experimental data.     

Effect of loading rate on damage of concrete

In this study, damage mechanics theory was used directly to determine the damage of concrete subjected to both static and impact compressive loadings. Two approaches based on the variations of (1) elastic modulus (E) and (2) strain rate (ε), were used. Results from both approaches indicated that the damage to concrete at peak load depended mainly on the rate of loading, as it increased with increasing loading rate. The specimens subjected to impact loading were found to suffer higher damage than those subjected to static loading, as seen by the larger value of D at peak load (0.8-0.9 for concrete subjected to impact loading and 0.65 for concrete subjected to static loading). Beyond the peak, the strain energy was sufficient to cause the damage to increase to one (D=1) without any further applied load.     

New method to measure the rapid chloride migration coefficient of chloride-contaminated concrete

The apparent chloride diffusion coefficient, D_app, which is obtained by fitting chloride profiles as the result of time-consuming immersion tests can be substituted in a model on chloride ingress by the rapid chloride migration (RCM) coefficient of concrete, D_RCM, which is determined under electrically accelerated conditions. Until now, it was not possible to measure D_RCM of chloride-contaminated concrete, as already inherent chlorides interfere with the common colorimetric indicator used for penetration depth measurements. Furthermore, carbonation may also interfere with the penetration depth reading. To overcome these problems, the regular test has been modified by using iodide as penetrating ion and iodate-starch acetic acid as indicator. A strong linear relationship between the regular RCM test and the newly developed Rapid Iodide Migration (RIM) Test was found. Carbonated specimens can be tested using the RIM test without additional interference. Thus, the new method enables the quantification of the actual concrete quality sampled from existing structures during a condition assessment.