Numerical analyses for the assessment of the isolation effects by a gravel-rubber mixture layer at the base of a building

Seismic hazard is recognized as one of the main causes of structures and infrastructure failure worldwide. Moreover, in recent years, a need to safeguard the ecosystem has encouraged researchers to find new solutions that combine seismic risk mitigation and ecosystem protection. In this framework, Geotechnical Seismic Isolation (GSI) systems have been proposed as a new mitigation technique based on improving soil behaviour using natural or modified geomaterials. Among the geomaterials proposed as GSI systems, soil-rubber mixtures (SoRMs) have emerged as a valuable and eco-sustainable technique for protecting structures in earthquake-prone areas. The main idea is to improve the soil immediately underneath the foundations using SoRMs so that seismic energy will be partially dissipated within SoRMs before being transmitted to the structures. SoRMs are generally obtained by blending sand or gravel as the soil portion and granulated tyre rubber as the synthetic portion. Rubber grains for the mixtures are manufactured from End-Of-Life Tyres (ELTs), the disposal of which has become a severe environmental problem worldwide. Laboratory tests have been recently carried out on gravel-rubber (GRMs) and sand-rubber mixtures (SRMs), allowing us to highlight their good static and dynamic properties. Numerical investigations and small-scale experiments on the GRM effectiveness as GSI system have been performed in the last decades. Only one full-scale test was recently performed on the EuroProteas prototype structure located in the Euroseistest experimental facility at Thessaloniki (Greece), after replacing the foundation soil with GRMs characterized by different rubber contents. These tests investigated the rubber content effect of the GRM layer on the dynamic response and the overall performance of the soil-GRM-structure system, demonstrating that a GRM characterized by a rubber content per weight equal to 30% can effectively dissipate the seismic energy within it before being transmitted to the structure. The present study deals with a numerical investigation on the effectiveness of the seismic isolation offered by a GRM layer placed under a real building. The hypothesized GRM was equal to the mixture used for the large tests carried out in Greece, that is the GRM70/30, i.e., having the 30% rubber content per weight. The chosen structure was a typical reinforced-concrete Italian building, damaged by the 2018 Catania earthquake. Both the soil-structure system without the GRM and with the GRM were modelled, using a 2D FEM approach and adopting an equivalent visco-elastic constitutive model for the soil and the GRM and a visco-inelastic constitutive model for the structure. The GRM layer underneath the foundations leads to a decrease in the amplitude of the Fourier spectrum for the low periods. The peaks of the FAS move towards lower frequencies, comparing the system in absence of mixture with the system in presence of mixture. At the foundation level, a significant reduction of the spectral accelerations was achieved for periods lower than 0.28 s, as well as for periods in the range 0.52-0.94 s. At the roof of the structure, the system response was even better: a considerable decrease of the spectral accelerations was achieved for the period range 0-0.93 s. So, the investigated mixture could be effectively applied as a GSI technique underneath the foundations of new structures similar to that investigated.