Undrained behaviour of the sandy-silty levels of the Tiber River alluvial deposits in Rome (Italy)

Abstract

This paper reports the ndings from a laboratory study aimed at the physical and mechanical characterisation under undrained conditions of sands belonging to recent Tiber river alluvia (level D Auct.) located in historical centre of Rome. The study was conducted in preparation for subsequent analysis of possible seismically-induced permanent deformation and/or liquefaction under the worst expected seismic shaking scenarios. Tests were carried out on a laboratory-reconstituted granular mix, which was considered to be representative of the above deposits. This choice was justi ed by: i) the difficulty of collecting fairly undisturbed samples with standard boring techniques, and ii) the need for gathering about 50 kg of material for all the planned tests. The tests consisted of: undrained static mechanical tests and dynamic resonant-column tests. They were conducted at the Laboratorio di Geologia Applicata - Dipartimento di Scienze della Terra-CERI, Sapienza University of Rome - and at the Laboratoire de Méchanique des Sols - Départment Géotechnique, Eau et Risque (IFSTTAR-Paris) respectively. The representativeness of the granular mix was assessed on the basis of the: i) mineralogical composition and the weight per unit of volume of the solid (γs), ii) comparison of its grain size with a target curve of the in-situ deposits, and iii) Unified Soil Classification System (USCS). The Proctor optimum of the granular mix was attained at a saturation of 90%, with a water content (w) of 17.6%, corresponding to a density (γd) of 16.70 kN/ m3. Undrained strength - determined via unconsolidated undrained triaxal tests (TxUUI) - was in the range of 20 to 60 kPa for void ratio (e) values of 0.7 to 0.5, whereas the critical void ratio (ecr) values were comprised between 0.55 and 0.65, corresponding only in part to the ones obtainable for the in-situ deposits (0.6<e0<0.8). Resonant-column tests yielded a linear shear deformation threshold (γl) of 0.00005 and a volume shear deformation threshold (γv) of 0.0003. The resulting decay curves were fairly reliable in comparison with the ones available in the literature for materials of similar density and grain size.