A first-order physical model for the prediction of shear-induced particle migration and lubricating layer formation during concrete pumping

Abstract

For the first time, a first-order physical correlation between pumping parameters (pressure and flow rate), granular skeleton properties (particle diameter and packing properties) and LL properties (thickness and viscosity) is proposed. This correlation is derived from the analysis of the equations driving shear-induced particle migration in dense suspensions. The model predictions in terms of lubricating layer apparent viscosity are compared to experimental measurements in the case of model concretes prepared with glass beads of various diameters, conventional-vibrated concretes and self-compacting concretes. This comparison is carried out for two flow typologies, namely the Sliper, known for its ability to mimic concrete pumping, and a co-axial tribometer. In all cases, the proposed model is shown to be able to capture qualitatively the observed main features and their evolutions despite the absence of any fitting parameters.