FRP-to-concrete debonding - global and local bond behaviour

FRP (fiber reinforced polymer) strips are used to flexurally strengthen reinforced concrete (RC) beams. The global bond strength namely the maximum transferable bond stress due to the increase of FRP tensile stress in a cracked concrete element (subsequently named as global bond shear stress) is a relevant aspect of flexural strengthening design. The differences between the bond strength in an uncracked concrete (end anchorage) and in a cracked concrete element have been highlighted in the existing literature; however, despite a large number of research works on the simple lap-shear test (local bond shear stress), less effort has been dedicated to investigate and model debonding on a global view. The current version of Swiss code on externally bonded reinforcement (SIA 166-2004) limits the local and global bond strength by means of the shear stresses, τf,max=τl0 and τl,lim, which are both assumed to be only functions of the concrete tensile strength. Nevertheless, it has been demonstrated in the literature that the maximum interfacial shear stress between two adjacent flexural (or flexural-shear) cracks depends also on the stress level in the FRP (σo). In the current paper, the difference between the local and global bond behavior and its debonding processes is explained and discussed with analytical and numerical models. In analogy to the existing approach developed to study the bond behavior of prestressed carbon fiber reinforced polymer (CFRP) strips during force release, a fracture energy-based model is proposed in the current study to determine the maximum global bond developed in a so-called intermediate crack element (ICE). The proposed model is a function of the FRP tensile stress (σo) and assumes a constant shear stress law. It is here demonstrated that the proposed model provides similar results to the more complicated models available in the literature. The main findings are here discussed with the aim to evaluate the feasibility of this new model for a future enhancement of the Swiss code.

An experimental investigation into bond behavior of prestressed CFRP to steel substrate

Elyas Ghafoori, Ardalan Hosseini, Masoud Motavalli

Nowadays, strengthening of existing steel structures using advanced carbon fiber reinforced polymer (CFRP) composites is becoming a common technique due to the unique advantages of the material such as light weight, high strength and fatigue endurance, as well as their relatively easy installation. Recently, the tendency to strengthen steel structures with prestressed CFRP plates is growing, especially in case of fatigue strengthening, where a reduction in acting stresses on the structures would be possible using prestressed reinforcements. However, the efficiency of using prestressed bonded CFRP reinforcements is severely affected by the undesirable premature debonding of CFRP from the steel substrate, before reaching the ultimate tensile capacity of the strengthening material. Consequently, it is of great importance to investigate the anchorage resistance and bond behavior of prestressed CFRP plates bonded to steel substrate. In the current study, a set of single lap-shear and prestress release tests was performed on CFRP-to-steel joints. All the tests were monitored using a 3D digital image correlation (DIC) system to understand the bond behavior of CFRP plates during lap-shear and prestress release tests. Moreover, the effect of accelerated curing on bond anchorage of the prestressed CFRP plates was investigated. Prestress release test results of the current study showed that the governing phenomenon in CFRP-to-steel joints is completely different than that in CFRP-to-concrete joints. The reason is that, despite in concrete, no tensile failure can occur in steel; consequently, relatively high prestressing forces can be transferred to the steel substrate before reaching debonding.

2017

Carbon Fiber-Reinforced Polymer Punching Reinforcement and Strengthening of Concrete Flat Slabs

Thomas Keller, Robert Koppitz

Three different carbon fiber-reinforced polymer (CFRP) systems were used to reinforce new or strengthen existing concrete slabs against punching. Untensioned CFRP plates or only very low prestressed CFRP straps could not prevent brittle punching failure. Strap prestressing of at least 15% of the tensile strength led to a ductile two-peak slab response although the CFRP material systems were brittle. The ultimate load of the unreinforced slabs could be increased by 77 to 118% via a redistribution of forces from the concrete to the strap system. An empirical model was established that is able to estimate the first peak load of all the applied CFRP systems. It takes into account the parameters CFRP plate or strap strength, CFRP geometry, CFRP prestressing level, and punching load of the unreinforced slab. A cantilever strut-and-tie model was established to estimate the second peak load. Both models provide acceptable accuracy.

American Concrete Institute2013

Punching shear strengthening of flat slabs using prestressed carbon fiber-reinforced polymer straps

Thomas Keller, Robert Koppitz

An experimental study of full-scale reinforced concrete flat slabs crosswise strengthened with prestressed carbon fiber-reinforced polymer (CFRP) straps against punching shear was performed. The effects of two strap anchoring systems and of slab thickness on punching behavior were compared. In one system the anchors were adhesively bonded to the concrete surface, while an external steel frame balanced the horizontal strap force components in the second system. Strap activation and thus strap force increments were higher in cases with either lower prestressing or higher strap stiffness. The deformability of the steel frame allowed a balancing of the strap forces. In the cases with steel frame, the friction bond between the steel anchors and the concrete surface reduced the concrete compressive zone. The system stiffness was thus increased and the rotation capacity reduced compared to the case with bonded anchors. The thickest slab exhibited a reduced rotation capacity and thus strap activation. The slab therefore failed at the lowest load increase, which however was still 67% above the theoretical resistance of the non-strengthened slab.

Elsevier2014