FRP strengthening of masonry arches towards an enhanced behaviour

This paper deals with the experimental behaviour of brick masonry arches strengthened with glass composite materials (GFRP). Eight 1:2 scale models of 1.5 m span arches were tested under a monotonic vertical load applied at the quarter span. The FRP strengthening was applied either at the extrados or at the intrados of the specimens. The experimental results presented in this paper show that the adopted GFRP strengthening provides an enhanced arch behaviour, with respect to the unstrengthened specimens. The ultimate strength was considerably increased and also a noticeable improvement in ductility was possible. The collapse mechanism of the strengthened arches was no longer related to the formation of a classical four-hinge mechanism, but it was characterized by the occurrence of new failure modes at some critical sections instead. 2 BEHAVIOUR AND FAILURE MECHANISMS OF MASONRY ARCHES Assuming that masonry has zero tensile strength, which can be justified by its relatively low or even zero tensile strength, an arched masonry structure is kept in compression as long as the thrust line (or pressure line), which represents the eccentricity of the compressive force at every cross-section, is kept inside the central core. When the thrust line moves outside the central core, at a given cross-section, the formation and consequent opening of a crack takes place. In this way, safety is maintained as long as the thrust line is kept inside the thickness of the arch. Naturally, the crack development leads to the formation of a plastic hinge at the compressed edge of the arch. However, in most cases masonry crushing is not likely to occur. Then, the formation of successive hinges leads to the formation of a mechanism that causes the arch failure. This means that unstrengthened masonry arches fail essentially by the occurrence of plastic hinges enough to form a mechanism (Heyman, 1982). Figure 1 represents the classical fourhinge mechanism of a masonry arch submitted to an asymmetrical loading. Figure 1. Four-hinge failure mechanism of a semi-circular masonry arch submitted to an asymmetrical loading. As expected, the presence of a bonded FRP strengthening changes completely the structural behaviour of a masonry arch. The fibers, which possess a high tensile strength, prevent the aforementioned hinge formation and may change significantly the failure mechanism. Since the use of FRP strips provides bending moment resistance, the thrust line may now safely move outside the thickness of the arch. For the arch illustrated in Figure 1 and considering the reinforcement located either at the extrados or at the intrados of the arch, the formation of a fourth hinge mechanism is prevented. Therefore, only three hinges are able to rise, transforming the arch into an isostatic structure. This means that new failure mechanisms different from the one afore-mentioned have to be considered. Due to the FRP high tensile strength, the compressive stress in masonry may now assume higher values so failure of the arch caused by masonry crushing has to be taken into account. The presence of the reinforcement also allows the development of higher shear stresses in masonry and, therefore, shear failure due to sliding along a mortar joint may occur. Moreover, in addition to the usual stresses parallel to the fibers, the curved shape of arches originates stresses with a component normal to the fibers, which may lead to the detachment of the reinforcement from masonry, namely in arches strengthened at the intrados. Consequently, the following failure mechanisms are usually added to the afore-mentioned one: Failure due to masonry crushing; Failure due to detachment of the fibers; Failure due to sliding along a masonry joint. Sliding between the fibers and its support is usually neglected since shear stresses at the FRPmasonry interface are of minor magnitude (Valluzzi et al., 2001). Also FRP tensile failure is not likely to occur due to its high tensile strength. It is known that, for a given arch shape, the type of failure to be obtained depends both on the mechanical properties of the materials (brick, mortar and FRP) and on the quantity and location of the reinforcement. In order to evaluate the behaviour of brick masonry arches strengthened with FRP, a combined experimental-numerical research project was started at Universidade do Minho, see also Lourenco & Martins (2001). This paper presents the first experimental results concerning the behaviour of brick masonry arches strengthened with glass composite materials (GFRP) and tested under a monotonic loading scheme. 3 EXPERIMENTAL STUDY The experimental program carried out consisted partially in the testing of twelve scaled semicircular brick masonry arches, plain and strengthened with GFRP strips. This experimental program was designed to attain the following main objectives: Characterization of the structural behaviour of both unstrengthened and strengthened masonry arches loaded monotonically until failure; Assessment of the influence of the reinforcement on the mechanical behaviour and failure mechanism; Creation of a reliable database on the experimental behaviour of masonry arches, able to be used in the calibration of both analytical and numerical tools. All arch specimens were constructed at scale 1:2 in order to optimize expenses related to raw materials and workmanship as well as to achieve a quicker construction process and a feasible testing setup. In order to replicate old masonry constructions, handmade bricks and a suitable mortar were selected. For that purpose, 100×50×25 mm clay bricks were especially made, reaching an average compressive strength of about 6.3 N/mm, whereas a pre-mixed hydraulic lime based mortar was adopted for the joints. Each semi-circular single-ring arch was composed of 59 brick courses and had a 750 mm radius, 500 mm width and 50 mm thickness (thickness/span ≈ 1/30), see Figure 2. The mortar joints were kept with a constant intrados thickness of approximately 10 mm.