الفهرس 
Overview of UHPFRC
stress. The idealized modelling approach divides the tensile behavior into three categories as shown in Fig. 2-20:Only 14 pages are availabe for public view
 |  | from | 278 |  |  |
| | | from | 278 | | |
Abstract
Due to changes in load, design faults, or to comply with new standards and guidelines, reinforced concrete (RC) structural elements such as beams may require strengthening or repair. There are numerous materials that are efficient for strengthening and repairing RC beams, such as fiber-reinforced polymers (FRP), steel, and ultra-highperformance fiber-reinforced concrete (UHPFRC). The load-carrying capacity of RC beams flexural-strengthened with an FRP plate bonded to the tension face is often limited by one of the observed debonding failure modes known as intermediate or end debonding. In the intermediate crack (IC) debonding failures, the stress state of the interface is almost similar to that of a shear test specimen in which a plate is bonded to a concrete prism and is subject to tension. The experimental and theoretical results from FRP-concrete bonded joints showed that the ultimate load (i.e., the maximum transferable load) of the joint and the effective bond length (the length beyond which an extension of the bond length cannot increase the ultimate load) depend on the properties of the concrete substrate, the tensile properties of the FRP, and the geometry of the FRP. The intermediate crack (IC) debonding failures were observed in the experimental studies on RC beams flexural-strengthened with an UHPFRC plate bonded to the tension face, so there was a need to study these failures in the case of using UHPFRC with high tensile strength as a strengthening material as this failure will occur before the rupture occurs in the UHPFRC plate. Also, the tensile behavior of UHPFRC is different from that of FRP, as the relation between stress and strain is linear (one modulus of elasticity until rupture), while this relation is bi-linear in the case of UHPFRC (two slopes of the tensile behavior, one in the elastic phase and the other after the matrix is cracked (hardening phase) till the maximum stress). Hence, an experimental study was conducted to estimate the effective bond length and the debonding loads through pull-push shear tests, which can simulate the intermediate crackinduced debonding in RC beams strengthened with prefabricated UHPFRC plates. Concrete strength (20, 25, 30, 35, 40, 45, and 50 MPa), bond length (150, 300, and 500 mm), and bond area (160X200 mm2 vs. 140X250 mm2) were the variables. Then, a finite element model (FEM) using ABAQUS software has been developed for simulating the experimental work and studying the effect of various parameters, such as the bi-linear behavior and thickness of UHPFRC, on the behavior. Finally, an analytical model in which the bi-linear behavior of UHPFRC in tension is employed is proposed to predict the effective bond length and the debonding load. First of all, the single pull-push shear test, given its simplicity and reliability, is therefore a good candidate as a standard bond test in the case where the UHPFRC plate is subjected to tension. The results of the experimental and FEM studies showed that the effective bond length and debonding load are significantly affected by the bi-linear behavior of UHPFRC in tension and the mechanical properties of the concrete substrate. The strain profiles showed that the effective bond length decreased with increasing concrete substrate strength. e.g., the decreasing ratio in the effective bond length for concrete grade (35 MPa) was equal to 19 percent and 13 percent compared with specimens of concrete grade (20 and 25 MPa, respectively). Among the specimens where only the concrete grade was the variable parameter and the bonded length was greater than the effective bond length, the specimen of concrete grade 50 MPa had the largest maximum debonding load where the increasing ratios were equal to 49 and 13% compared with concrete grades 20 and 35 MPa, respectively. The proposed finite element approach was in good agreement with the experimental work, as the ratios between experimental and FEM results of debonding load and effective bond length ranged from 89 percent to 108 percent and 101 percent to 107 percent, respectively. Also, the proposed analytical model is suitable for use in the case of UHPFRC or bi-linear materials. Contrary to previous models for FRP/steel material, the ratios between the proposed analytical model and experimental/FEM results of effective bond length and debonding stress ranged from 85 percent to 111 percent and 92 percent to 114 percent, respectively.