Papers by Author: Veerle Boel

Abstract: This study focuses on the fracture mechanics aspect of self-compacting concrete, compared to vibrated concrete. The most commonly used experiments to investigate the toughness and cracking behaviour of concrete are the three-point bending test (3PBT) on small, notched beams, and the wedge-splitting test (WST) on cubic samples with guiding groove and starter notch. From the resulting P-CMOD curves (applied load versus crack mouth opening displacement), different fracture parameters, such as fracture energy and fracture toughness, can be extracted. Moreover, using inverse analysis, the σ-w relationship (tensile stress versus crack width) can be derived. This paper lists the results of a series of tests on samples, made of VC, SCC of equal strength, and SCC with identical w/c factor. Subsequently, a comparison of the mechanical characteristics is made, revealing important differences regarding several fracture parameters.

Abstract: The stress intensity factor and the T-stress describing the near-crack-tip fields for selected specimen shapes of a test geometry based on wedge splitting and three point bending tests with several variants of boundary conditions are computed using finite element software ANSYS. The test configuration in question is expected to be a convenient alternative to classical fracture tests (especially the tensile ones) for investigation of the quasi-brittle fracture of building materials, when low constraint is requested. These specimens are investigated within the framework of two-parameter fracture mechanics; near-crack-tip stress field parameters are determined and compared with those of the wedge splitting test due to their shape similarity. The sensitivity of the values of these parameters to the boundary conditions is also shown. Suitable choice of the shape of the specimens is discussed.

Abstract: Vibrated concrete (VC) and self-compacting concrete (SCC) have a substantially different composition, resulting in dissimilar mechanical properties regarding cracking behaviour. The critical value of the mode I stress-intensity factor K_IC is an appropriate fracture parameter for evaluating fracture toughness and can be obtained from three-point bending tests (3PBT) on small, notched specimens. Subsequent determination of the energy release rate thus allows to examine the crack propagation and fracture process of both concrete types. This paper describes the results of such 3PBTs on samples, made from VC and SCC. Evaluation of the cracking behaviour, derived from these results, reveals remarkable differences.

Abstract: Connection between steel and concrete parts is frequently required in constructions where the steel-concrete joints are often realized by welded shear studs. In order to avoid stress concentrations, corrosion proneness, and other negative consequences of the welding process, steel-concrete connection without welded mechanical shear connectors is sought nowadays. Connection can be realized via an epoxy adhesive layer and gritted with granules. In the paper, the assessment of the push-out test configuration was performed from the generalized fracture mechanics point of view. The numerical-analytical modelling of a steel-concrete connection is performed without and with the epoxy interlayer, while 2D and 3D modelling is used. Thus conditions of crack initiation can be predicted from knowledge of the standard mechanical and fracture-mechanics properties of particular materials. The model of a bi-material notch with various geometry, and material properties is used to simulate various singular stress concentrators that can be responsible for failure initiation. Various manners of preparation of the epoxy interlayer are tested experimentally. Results of the fracture-mechanics studies are compared with each other and with experimental results. On the basis of the comparison, the 2D simulation of the steel-concrete connection without the epoxy interlayer is shown to be suitable for the estimation of failure conditions.

Abstract: The wedge-splitting test (WST) is a frequently used test configuration for performing stable crack fracture experiments on concrete specimens, thus allowing to determine the fracture process and crack propagation in the heterogeneous material. However, there are no standard rules regarding the wedge-splitting specimens geometry, groove dimensions or notch length. This paper concentrates on the influence of the initial notch length in geometrically identical, cubical specimens, cast from vibrated concrete. The experimental results of nine WSTs under monotonic loading, including F_sp-CMOD curves - splitting force versus crack mouth opening displacement - and fracture energy G_f, are presented. An important effect of the starting notch length on the fracture properties is observed.

Abstract: Steel-concrete joints are often provided with welded shear studs. However, stress concentrations are induced in the structure due to the welding. Moreover, a reduction in toughness and ductility of the steel and a decreased fatigue endurance of the construction is observed. In this paper the shear bond strength between steel and ultra-high performance concrete (UHPC) without mechanical shear connectors is evaluated through push-out tests. The test samples consist of two sandblasted steel plates with a thickness of 10 mm and a concrete core. The connection between steel and concrete is obtained by a 2-component epoxy resin. Test samples with a smooth adhesive layer are compared with those with an epoxy layer, which is applied with a toothed paddle and/or gritted with small aggregates. In this research, specimens prepared with river gravel, crushed stone, and steel grit are compared and also two different epoxy resins are used. During the tests, the ultimate shear force is recorded as well as the slip between steel and concrete. All test specimens exhibited a concrete-adhesive or concrete failure. Furthermore, test results show that the use of a more fluid epoxy resin improves the anchorage of the gritted aggregates in the adhesive layer, resulting in higher shear bond stresses. No significant difference is found between specimens, gritted with river gravel or crushed stone. Applying the adhesive layer with the toothed paddle in horizontal direction slightly improves the bond behaviour. Finally, the experimental results of the test members with a smooth epoxy layer without gritted aggregates, provide test data for a fracture mechanics approach, which uses a 2D numerical model of the test specimen, composed of steel, epoxy resin, and concrete.

Abstract: Steel-concrete joints can suffer from premature fail due to inadequate shear bond between the two surfaces. In this paper the shear bond strength between steel and self-compacting concrete (SCC) without mechanical shear connectors is evaluated through push-out tests. The test samples consist of two sandblasted steel plates (10 and 6 mm) and a concrete core, with connection between steel and concrete obtained by a 2-component epoxy resin, gritted with granulates. During the tests, the ultimate shear force is recorded as well as the slip between steel and concrete. All test members exhibited a concrete - adhesive failure, and indicate nominal shear bond stresses between 2.20 and 4.22 MPa. In addition, a substantial difference in measured shear bond stresses is found between the 6 and 10 mm steel plates, indicating unwanted secondary effects with the 6 mm plates. During testing, maximum slip values between 0.02 and 0.05 mm are recorded. In addition to the experimental tests, shear stress distribution in the epoxy – concrete interface is examined by finite element analysis (FEA). In this way, a non-uniform stress distribution between steel and concrete is found with the maximum shear value about 2.5 times higher than the nominal shear stress value. The FEA combined with the experimental results provide a reasonable understanding of the shear induced failure conditions at a steel-concrete joint, and create test data for a fracture mechanics approach.

Abstract: In order to evaluate the shear bond strength of a steel-concrete joint using an epoxy adhesive interlayer, push-out tests were carried out. The test samples consisted of two sandblasted steel plates and a self-compacting concrete sample, with the epoxy layer applied on the steel plates and gritted with granulates. During testing, an external force was applied to the concrete core and continuously recorded. To investigate the failure mechanism in detail, a fracture mechanics approach is required. In this paper theoretical-numerical assessment of the push-out test is performed. Regarding the finite element calculations, the locations suitable for failure initiation match bi-material (steel-concrete) notches. The most dangerous locations are evaluated from a generalized linear elastic fracture mechanics point of view. The critical load corresponding to the conditions of failure initiation is estimated and compared with the experimental results.

Abstract: This paper presents the results of experimental tests carried out on steel fibre reinforced self-compacting concrete (SFR-SCC) beams without stirrups. Sixteen beams are cast using four mixtures of SCC with different steel fibre content, while the longitudinal reinforcement is kept constant in all test members. The beams are subjected to four point bending tests at a shear span-to-depth ratio of 2. The ultimate shear stress is recorded, as well as the crack pattern and the mid-span beam deflection. Test results show that as fibre content increases, higher ultimate shear stresses are achieved. When fibres are included, test members exhibit an increase in ductility and a more extensive crack pattern is observed. The experimental values of the ultimate shear stresses are also compared with theoretical values as given by empirical expressions in literature.

Abstract: Many civil structures are subjected to dynamic loading with load cycles exceeding 100.000 to 100.000.000 or more depending on the type of structure. At such high amounts of load cycles concrete can fail due to fatigue damage. The phenomenon is well understood and documented in the literature for traditional concrete. It remains unclear whether these conclusions can be transferred to self-compacting concrete (SCC), due to changes in concrete composition, decrease in the size of the coarse aggregates in combination with the addition of filler and superplasticizer. For this paper several reinforced concrete beams with varying reinforcement ratios are subjected to dynamic loads in a four point bending rig until failure. The imposed stress levels for the dynamic loading range between 0,80fcc and 0,85fcc. The main conclusions of the experimental tests are (1) clearly different failure mechanisms determine the static and dynamic loading for a single type of beam, (2) SCC appears to have a somewhat lower shear resistance during fatigue, (3) the differences in flexural degradation between SCC and TC beams are rather small.