Skip to content

Analysing the rheology of diverse volcanic granular flows with a new open-source coarse-graining tool.

Claudia Elijas-Parra1, Eric C. P. Breard1, Eliza S. Calder1, Samantha L. Engwell2, Mattia de' Michieli Vitturi3

Abstract

Granular flow processes with varying types of interstitial fluid and coupling regimes are ubiquitous in volcanology: from concentrated underflows in pyroclastic density currents, to debris flows, and crystal mushes in magma. However, the behaviour of these highly complex granular flows remains poorly understood. Modelling of granular flows through software packages that couple the Discrete Element Method (DEM) with Computational Fluid Dynamics (CFD) can provide detailed information at high temporal and spatial resolution of particle-particle and particle-fluid interactions. As such DEM-CFD approaches provide an invaluable tool to study the small-scale behaviour of granular flows, and are instrumental in the development of constitutive models used to understand rheology of granular materials. DEM-CFD models yield fundamental physical properties, for instance: particle force and velocity, (for solid phase) and fluid pressure (for fluid phase). To obtain continuum fields of relevant variables from discrete data (e.g., stress and strain tensors, pressure etc.), DEM-CFD outputs must be processed through a method called Coarse-Graining (CG). In this work we present a new python-based CG package designed to: maximise computational efficiency through -amongst other strategies- parallel computing; be user-friendly; and open source.  Our tool postprocesses output files of DEM-CFD software packages, such as MFiX and LAMMPS, and yields outputs ready to render on the ParaView visualisation software. We present exemplar applications for a fluidised bed and a granular flow impacting a bend and make the case that our tool could be usefully applied to other areas such as crystals and magma in a conduit, or impact cratering.