Insight into contact forces in crushable sand using experiments and predictive particle-scale modelling

John M. Harmon, Dawa Seo, Giuseppe Buscarnera, José E. Andrade*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

5 Scopus citations

Abstract

In this paper, an attempt is made to predict the evolving statistics of inter-particle contact forces during comminution using grain-scale computational modelling. A validation is first carried out by creating a one-to-one virtual avatar of an Ottawa sand specimen from three-dimensional X-ray tomography with level sets and comparing the data from an oedometric test to the model's prediction. The predictive capabilities are confirmed by comparing the constitutive response, grain size distribution and changes in particle shapes in both the experiment and model. Once validated, the predicted contact forces and particle stresses are investigated. It is found that the largest particles experience the largest forces. Despite larger particles being weaker on average, many survive because they are on the stronger side of the particle strength distribution and also have a higher coordination number producing a more isotropic stress state in the particle. These highest forces are largely aligned with the specimen axis, demonstrating that larger particles provide the strength in the loading direction. Meanwhile forces in the radial direction are more broadly distributed, indicating that small particles play a significant part in providing radial stability.

Original languageEnglish (US)
Pages (from-to)238-249
Number of pages12
JournalGeotechnique
Volume74
Issue number3
DOIs
StatePublished - May 6 2022

Keywords

  • discrete-element modelling
  • fabric/structure of soils
  • particle crushing/crushability

ASJC Scopus subject areas

  • Geotechnical Engineering and Engineering Geology
  • Earth and Planetary Sciences (miscellaneous)

Fingerprint

Dive into the research topics of 'Insight into contact forces in crushable sand using experiments and predictive particle-scale modelling'. Together they form a unique fingerprint.

Cite this