Strain localization and shear band formation during tensile tests of disordered floating granular monolayers

Understanding and predicting strain localization and shear band formation during deformation of amorphous solids is an ongoing challenge. In this study, quasi-static tensile experiments were performed using a model disordered solid consisting of a monolayer of polydisperse granular spheres with capillary attractions floating on an air-oil interface. Under tensile deformation, the strain in the monolayer gradually localizes into an inclined shear band, upon which failure occurs. The ductility of the monolayer can be tuned by using different particle sizes compared to the capillary length, which sets the interaction range. Using machine learning methods, we developed a scalar field, softness, that relates the local structure around particles and particle-level dynamics. We found that softness tends to increase around particles that have rearranged. Furthermore, planes that develop into shear bands tend to have higher than average softness prior to deformation and become even softer as the shear band forms. These results reveal the relation between structure and dynamics that leads to strain localization, which could potentially apply to a broad range of amorphous solids.