International Concrete Abstracts Portal

Behavior of polyester polymer concrete (PC) with and without notch and graphite fiber was investigated using nondestructive and destructive testing techniques. The flexural strength of polyester PC was 14 MPa (2,000 psi). The effect of up to 6% chopped graphite fibers on the elastic modulus, shear modulus. Poisson’s ratio, flexural strength and fracture parameters were investigated. Nondestructive methods such as impact resonance and pulse velocity were used to determine the effect of notch depth on the mechanical and damping properties of PC. Fracture parameters, critical stress intensity factor Ktc and critical Jtc-integral were determined using single edge notched beam loaded in four-point bending by varying the initial notch-to-depth ratio from 0.2 to 0.7. By measuring the crack mouth openin g displacement (CMOD) during loading, the crack extension in the test specimen was determined. The critical stress intensity factor and critical J-integral for the polyester PC were 1.3 MN/m’ and 0.27 kN/m respectively. The addition of 6 mm long 6% chopped graphite fibers to the polyester polymer concrete improved the tlexural strength by 20% and Ktc and Jtc by over 25% and 125% respectively. Impact resonance test results were sensitive to the notch-to-depth ratio in the test specimen.

Experiments of three kinds of FRP-strengthened concrete beams under three-point bending are reviewed first. Two different cracking behaviors, with and without distributed crack in concrete, were observed. To analyze the different cracking behaviors affected by the FRP strengthening through interfacial bond, the FRP strengthened concrete beam is subdivided into a plain concrete beam, under three-point bending, and a FRP sheets bonded concrete prism, subjected to shearing load, to address the fracture mechanism. Nonlinear fracture mechanics is used to model the cohesive crack along the FRP-concrete bond interface and concrete cracking. Finite element simulation is also performed to demonstrate the applicability of the fracture mechanism. Based on both experimental observations and finite element results, it can be concluded that the FRP strengthening effect occurs after the first flexural crack. The bond strength and interfacial fracture energy of bond interface determine the ability of stress transfer. The occurrence of the new flexural cracks after the first one is governed by the relation between the concrete tensile strength and the maximum concrete stress obtained by combining effects of shear stress transfer and bending moment including the stress release due to flexural cracks. Further strengthening effect is archived by the formation of new cracks.

This work describes a modification to the two-parameter fracture method’s experimental procedure aimed at removing this operator/equipment dependence. With this method, three compliances are used to determine the focal point at which these compliances intersect. This focal point is then used to determine the slope of the unloading compliance that corresponds to the peak of the load vs. CMOD curve. The unloading compliance that corresponds to unloading at the peak load and initial compliance are then used to determine Ktc and CTODc as normally done with the Two Parameter Fracture Model. Use of this method makes it possible to remove operator and machine dependence, especially if the materials are extremely brittle, such as in pastes or high strength concrete, thereby permitting the loading and unloading to be programmed using testing software removing the need for manual operator loading changes. Tests on 15 mortar beams with 4 different notch lengths and initial unloading points ranging from 97% to 75% of maximum load are used to validate this approach. The experimental results are typically more consistent and better correlate to results from the peak load test method. These results indicate that utilizing the focal point correction typically reduces Ktc and CTODc by 12% and 38% respectively for the mortar tested thereby causing the TPFM and peak load method results to coincide even more closely.

Early-age cracking can occur in concrete if free shrinkage is prevented by the surrounding structure. This paper highlights recent findings to illustrate that shrinkage cracking is influenced by the geometry of the structure. A series of experimental results are presented from three different ring specimen geometries to illustrate that although these specimens had the same residual strain level (and similar residual stress), the age of cracking varied with specimen geometry. A second series of experiments was performed to illustrate that a geometry dependence also exists in specimens with moisture gradients. This paper describes how fracture mechanics concepts can explain this geometry dependent behavior under a uniform moisture distribution. Residual stress levels are computed, non-linear fracture mechanics failure criterion is applied to develop the time and geometry dependent tensile stress resistance curves, and the age of cracking is predicted. The theoretical simulations were found to compare reasonably with the experimental observations. A discussion is provided to illustrate how these considerations may be extended to the specimens with moisture gradients.

Fracture surface provides valuable information on internal structure and mechanical behavior of composite materials. Loading rate affects the roughness of the fracture surface of composites. A higher loading rate, in general, results in a smoother fracture surface. Similarly, aggregate size influences the roughness of the fracture surface. Larger aggregates cause a rougher fracture surface under the same loading rate. The roughness of the fracture surface of concrete is experimentally studied using concrete specimens made of the different aggregate sizes under different loading rates. Fractal dimension is used to evaluate the surface roughness of concrete specimens. A new fractal fracture model is developed which correlates the fractal dimension with concrete mix design parameters, such as volume fraction and size of aggregate, as well as loading rate. The model prediction agrees with test data very well.