CRM REINFORCEMENT OF BRICK MASONRY WALLS: EXPERIMENTAL AND NUMERICAL INVESTIGATIONS

The preservation of the building heritage is often accompanied by the necessity of ensuring an appropriate structural safety of existing buildings. Existing masonry structures are usually characterized by a high seismic vulnerability due to their low tensile strength; moreover, despite the significant compressive strength, often masonry panels are not able to guarantee the safety level required by the current Building Codes under gravitational loads, due to their slenderness and possible out-of-plane actions. In recent years, several systems have been developed for the reinforcement of masonry buildings based on the use of composite materials externally bonded to the structural elements. Among them, the CRM can be used for static and seismic reinforcement of masonry piers as an alternative to the traditional reinforced plaster. The CRM system consists of the application on the panel surfaces of a mortar layer about 3cm thick reinforced with a glass fibre grid and connected to the wall through transversal connectors. Despite its increasing use, to date limited experimental results on strengthened panels are available; in particular, few studies have been carried out focusing on the flexural behaviour of reinforced panels and on the effect of the reinforcement on the ultimate displacement, as well as few studies are available regarding the contribution provided by CRM layers on the buckling failure of panels induced by vertical loads. Due to the lack of experimental and theoretical studies, a standardized design procedure for the assessment of the response of retrofitted panels has not yet been defined. The present Thesis focuses on the study of CRM-reinforced panels loaded by vertical and horizontal actions, with the aim of analysing the parameters that most affect the effectiveness of the reinforcement and defining an analytical procedure for the evaluation of the load-bearing capacity of strengthened panels. An experimental campaign has been carried out at the Laboratory of Structures and Materials of the University of Florence: a series of tests for the mechanical characterization of materials composing the masonry and the reinforcement has been at first realized; then, shear-compression and uniaxial compression tests on solid brick panels reinforced with CRM have been performed. Tests allowed the assessment of the mechanical properties of the CRM system and its adhesion to the substrate and highlighted the advantages and limitations produced in the structural behaviour of reinforced walls. Based on the experimental results, a non-linear finite element model has been calibrated, capable of reproducing the behaviour of reinforced panels, both in terms of capacity curves and in terms of failure mechanisms. The model, calibrated and validated using experimental evidence, was used to carry out a parametric analysis by varying the most influential mechanical and geometric parameters, such as masonry typology, wall thickness, application of plaster on one or two sides of the walls, and, in the case of pure compression, the eccentricity of the vertical loads. The results of this analysis were processed to develop, through numerical regressions, analytical formulations to compute the lateral and compressive response of CRM-reinforced walls. In particular, the contribution provided by CRM to the in-plane lateral stiffness, to the shear strength by diagonal cracking and bending, and to the displacement capacity was expressed by analytical equations, as well as the compressive strength of brick panels subject to out-of-plane instability. These formulations are valid for brick walls, tested experimentally, and have been numerically extended to the masonry types provided by Italian standards under the assumption of monolithic behaviour through the thickness of panels. The proposed procedures are simple to be implemented in the common analysis methods required for unreinforced structures and can be used as a useful reference for the design of interventions with CRM.