Out-of-plane seismic safety assessment of URM infills accounting for the in-plane/out-of-plane interaction in a nonlinear static framework

Past and recent earthquakes showed the occurrence of Out-Of-Plane (OOP) failures of Unreinforced Masonry (URM) infills in Reinforced Concrete (RC) frames. Such a type of failure, which is promoted by In-Plane (IP) damage (IP/OOP interaction), is dangerous for human life safety and its effects can be associated with the attainment of the Life Safety limit state. In this work, the seismic capacity of Reinforced Concrete buildings at the first OOP collapse is evaluated with a linear analysis based on code provisions and with a procedure based on nonlinear static analysis. More specifically, 16 buildings designed to Eurocodes, different for number of storeys and design Peak Ground Acceleration (PGA), are infilled by two infill layouts different for thickness and masonry properties. For these buildings, the PGA at which the first OOP infill collapse occurs is evaluated by applying two different approaches. The first is based on linear analysis and consists in the simple application of the demand and capacity models currently provided by the Eurocodes. So, it accounts neither for the effect of the IP/OOP interaction on the OOP capacity nor for the effect of structural non-linearity on the OOP force demand. The second approach is based on nonlinear static analysis and on the application of refined literature formulations accounting for the IP/OOP interaction and for structural non-linearity for the definition of the OOP capacity and force demand. The PGAs at the OOP collapse of infills obtained by applying both approaches are compared to show their significant overestimation if the IP/OOP interaction and structural non-linearity are not considered, i.e., if current code provisions are applied. Considerations concerning the influence of the number of storeys and of the design PGA of buildings on the PGA at the OOP collapse of infills are reported. Frequency distributions of OOP collapses at different storeys and fragility curves relating the probability of OOP collapse to both the PGA acting in the OOP direction and the maximum IDR attained in the IP direction are shown. For all case-study buildings, a range of 42 infill layouts, different for thickness and masonry compressive strength, is considered and, with the application of the more refined, not code-based approach, a “limit state” curve defining the infill height-to-thickness ratio/masonry compressive strength couples for which the OOP safety check of infills can be neglected is reported