Revealing the transverse diffusion of particles with different shapes in bed-load transport

A fluid flow imposed over a granular bed can cause the grains to move by rolling, sliding, and jumping within a thin layer known as bed-load layer. In this layer, it may be observed a transverse diffusion and segregation of grains by size, which has been extensively studied, and shape, which has only recently been recognized as an important control. Here, we perform numerical simulations using an eulerian-lagrangian approach (CFD-DEM) to predict the intriguing behavior of grains present in the bed-load layer. We set a periodic channel filled with spherical and non-spherical particles, such as cubes and cylinders of equal mean diameter, sheared by a viscous Couette flow which imposes enough shear stress to move the particles by bed-load transport. We investigate the statistical properties of the transverse diffusion that non-spherical grains experience in the bed-load layer by tracking hundreds of individual trajectories throughout the entire bed, and the mechanisms involved that are mainly driven by the particles collisions and the fluid-grains interactions. These results illuminate the role of grain shape in controlling sediment diffusion, with implications for natural rivers, hillslopes, marine environments, and aeolian systems.