Journal of Materials in Civil Engineering

A single fiber pullout test was used in this study to measure the bond properties of polyvinyl alcohol fibers that are available at various diameters in a mortar matrix. Despite short fiber embedment lengths, the small diameter fibers ruptured during the pullout tests. However, it is shown that even if full fiber pullout is not achieved, it is still possible to determine a chemical debonding energy, G_d, and an initial interfacial frictional bond strength, τ₀. Despite high G_d values, the fibers did not rupture during the fiber chemical debonding process, but during fiber pull-out, a strong slip-hardening effect, characterized by the high values of the slip-hardening coefficient, β, induced severe abrasion damage visible under scanning electron microscope on the fiber surface. As a consequence, when the fiber apparent tensile strength was exceeded, fibers ruptured by delamination. Finally, an attempt was made to lower the values of the bond properties to minimize fiber rupture during pullout. This goal was partially achieved by applying an oil coating on the hydrophilic polyvinyl alcohol fiber surface to reduce any strong fiber/mortar interaction.

A singular hybrid finite element with seven nodes and triangular shape is used to determine the stress intensity factors for the splitting tension test. For different types of marble the finite-element simulations were used to compute the fracture toughness values. The small-grained marble types exhibit a high fracture toughness and stable crack growth. The smallest toughness was obtained for the coarse grained Georgia marble.

The recovery of container glass (cullet) for recycling can often lead to the generation of surplus quantities of the material. The pozzolanicity of finely ground glass cullet (GGC) can be exploited by using it as a cement component in concrete. This paper examines two of the chemical reactions most significant to this mode of use, namely, the pozzolanic reaction and the potentially deleterious alkali-silica reaction. Although the presence of GGC has little effect on the reactions of portland cement, larger quantities of calcium silicate hydrate gel are formed, which are thought to contain quantities of sodium. Blends containing the largest quantities of calcium silicate hydrate gel also displayed the highest strengths. Expansion due to alkali-silica reaction of mortars containing GGC is reduced relative to controls. This can be understood when it is considered that the pozzolanic and alkali-silica reactions are chemically identical. The fine cullet particles have a high surface area and, hence, favor the rapid pozzolanic rate of reaction over the slower alkali-silica reaction rate.

The results of punching shear tests on 31 square ferrocement slabs are reported in this paper. The slabs were simply supported on all four sides and tested under a central concentrated load. The parameters investigated include the width of square loaded area w, mortar strength $\mathit{f}{\prime }_{\mathit{c}}$, volume fraction of reinforcement V_f, depth of slab h, and the effective span ℓ. All slabs failed first in punching without total separation, and then exhibited a second peak in the load-deflection history. Both cracking load and punching shear load increased with an increase in w, V_f, $\mathit{f}{\prime }_{\mathit{c}}$, or h, but decreased as ℓ is increased. The critical perimeter for punching shear failure was found to be located at a distance of 1.5h from the edge of the loading plate. Based on the present test data as well as those available in the literature, an equation has been proposed for predicting the punching shear strength of ferrocement.

This paper presents the results of a statistical analysis aimed to optimize a concrete mix design for hot climates. A full factorial experiment with 3 × 4 × 4 × 3 treatment combinations (432) samples) of 48 mixes at three levels of temperature was used. The influences of the water/cement ratio (0.4, 0.5, and 0.6), coarse aggregate/total aggregate ratio (0.55, 0.6, 0.65, and 0.7), total aggregate/cement ratio (3, 4, 5, and 6), and temperature (24, 38, and 52°C) on compressive strength were characterized and analyzed using polynomial regression. Mathematical polynomials were developed for concrete strength as a function of temperature and mix proportion. Based on the statistical analysis, recommendations are provided on the optimum concrete mix for different temperatures as well as the mix that is least sensitive to temperature variations.

The changes in the stiffness of two asphalt concrete mixtures due to temperature, fatigue damage growth, and healing during rest periods are evaluated using the impact resonance method. The impact resonance method is a means of determining the dynamic modulus of elasticity of a specimen nondestructively. The dynamic modulus of elasticity decreases as temperature increases and as microcrack damage growth occurs in the specimen due to fatigue. The impact resonance method also detects increases in dynamic modulus of elasticity after the application of rest periods. A gain in flexural stiffness was also observed from measurements and is attributed to closure of microcracks or healing during the rest period. The amount of healing or stiffness gain appeared to increase when specimens were subject to a higher temperature during the rest period. A qualitative study of the two asphalt mixtures showed that there is a difference between the two with respect to healing performance.

This paper investigates effects of several factors such as lime content and the types of lime on water requirement, initial porosity, strength development, pozzolanic reaction rate, and pozzolanic reaction chemistry of lime-pozzolan cement. The water requirement of the lime-pozzolan cement did not change when lime content increased from 0 to 20% but it increased with higher lime content. A calculation indicated that lime content had a similar effect on the initial porosity of fresh lime-pozzolan cement mortars as on the water requirement. Strength measurement indicated that the optimum lime content for lime-pozzolan cement was around 20%. For a given CaO content, use of quicklime showed a higher lime consumption rate than use of hydrated lime before 28 days but no difference could be observed thereafter. Lime-pozzolan cements made with quicklime gave higher strengths than the cement made with hydrated lime.

This paper presents a semianalytical method to predict fatigue behavior in flexure of concrete in the case of maximum flexural moment larger than that of the first crack moment based on force equilibrium in the critical cracked section. The model relies on the cyclic bridging law, the so-called stress–crack-width relation under cyclic tensile load as the fundamental constitutive relationship in tension. The structural size effect on fatigue in bending of concrete beams is studied by the present model. Eight series of beams, with heights from 50 to 800 mm, are analyzed. The model results show that the fatigue performance in bending, normally expressed in terms of the maximum flexural stress versus fatigue life diagram (S-N) is strongly dependent on the structural size, even when the same material parameters are used in the model. Under the same cyclic flexural stress levels, the smaller the beam height, the longer the fatigue life is. Besides this, the deformation characteristics, such as fatigue crack growth history as well as the final fatigue crack length, are also size dependent. Finally, the model predictions are compared with experimental results. Good agreement between model predictions and experiments has been obtained.

Galvanized coating is used to prevent corrosion of reinforcing bars. Contrary to steel, this coating reacts with the surrounding aqueous phase of the cement paste and induces a delay in the hydration of the interfacial zone. This paper describes an experimental study of this effect on bond strength of the reinforcement. This effect is evaluated by testing specimens with smooth bars, ribbed bars placed in a horizontal or vertical position, or machined bars with different rib heights. The results indicate that the bond performance of galvanized bars depends on the failure zone of the concrete surrounding the reinforcement and on the time of testing. The type of bar and the rib height of the reinforcing bar are designated as the parameters affecting the bond failure of galvanized bars.

The behavior of marble was investigated to evaluate the damage and failure mechanisms produced from cyclic loading. Approximately 100 cycles of increasing load steps, representing low cycle fatigue, were imposed on beams in three and four-point bending. Damage was recorded through an array of strain gauges and an interferometric technique that provided a full-field assessment of the surface deformation. It was shown that large strains in the marble beams created permanent deformations that, with cyclic loading, collected in an irreversible way with a progressive loosening of the calcite grains. The failure surfaces of the specimens under monotonic and cyclic loading were visually different in that macroscopic fracture of cyclically loaded specimens were covered with powder, revealing the decohesion process. Permanent elongation was measured at the bottom of the beams, while shortening appeared at the top. This permanent deformation is characteristic of a bowing type of behavior that has been observed in thin marble cladding damaged from weathering.

Although considerable research has been conducted on the shear strength of reinforced concrete members without web reinforcement, the debate as to the effects of the major variables on the ultimate shear strength (v_u) and cracking shear strengths (v_c) has yet to be unequivocally established. Previous research on the subject has at times been unclear or often yielded contradictory results. The intent of this paper is to shed some light and clarify some of the misconceptions regarding the effects of the compressive strength (f′c), tensile reinforcement ratio (ρ), and shear span-to-depth ratio (a/d) on v_u and v_c of steel-reinforced concrete beams without web reinforcement. The analysis uses extensive shear strength data retrieved from existing literature for both normal strength and high strength concrete members. Results show that a/d significantly affects v_u, whereas its effect on v_c is much less significant. The effect of f′c on v_c is not negligible and should be accounted for in strength analysis. The impact of ρ on v_c is insignificant in the case of short beams (a/d < 2.5).