Breakage behavior of biomass pellets: An experimental and numerical study

Hamid Gilvari*, Wiebren de Jong, Dingena L. Schott

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

16 Citations (Scopus)
103 Downloads (Pure)


The presence and generation of fines and dust in the bulk of biomass pellets have inflicted several problems in the supply chain during transportation and storage, and the breakage behavior of pellets has been scarcely studied so far. Fines and dust are the consequences of impact and abrasive forces through the whole supply chain; however, the breakage happens at the particle level. Therefore, to study the fines generation, first, the breakage behavior of individual pellets should be understood, and then, the behavior of the bulk materials in operational conditions can be investigated. This paper aims to investigate the breakage behavior of individual pellets under experimental compression tests and to introduce a calibrated numerical model using discrete element method (DEM) in order to pave the way for further studies on pellet breakage. For that purpose, seven different types of biomass pellets were studied experimentally, and then, a calibrated model was introduced via the Timoshenko–Ehrenfest beam theory using DEM. Results show that the model could reasonably predict the breakage behavior of pellets under uniaxial and diametrical compressions. The findings could help to develop a new design of the equipment for transportation and handling of biomass pellets with the aim to reduce the amount of generating fines and dust.
Original languageEnglish
Pages (from-to)1047-1060
Number of pages14
JournalComputational Particle Mechanics
Issue number5
Publication statusPublished - 2020


  • Biomass pellet
  • Breakage behavior
  • Compression tests
  • Discrete element method
  • Timoshenko–Ehrenfest beam theory


Dive into the research topics of 'Breakage behavior of biomass pellets: An experimental and numerical study'. Together they form a unique fingerprint.

Cite this