Characterization of microcrystalline cellulose spheres and prediction of hopper flow based on a μ(I)-rheology model

Johan Remmelgas, Abdoulaye Fall, Srdjan Sasic, Henrik Ström, Pirjo Tajarobi, Håkan Wikström, Mariagrazia Marucci, Catherine Boissier

Research output: Contribution to journalArticlepeer-review

Abstract

The objective of this study was to characterize the rheology of a pharmaceutical material in the context of the µ(I)-rheology model and to use this model to predict powder flow in a manufacturing operation that is relevant to pharmaceutical manufacturing. The rheology of microcrystalline cellulose spheres was therefore characterized in terms of the μ(I)-rheology model using a modified Malvern Kinexus rheometer. As an example of an important problem in pharmaceutical manufacturing, the flow of these particles from a hopper was studied experimentally and numerically using a continuum Navier-Stokes solver based on the Volume-Of-Fluid (VOF) interface-capturing numerical method. The work shows that the rheology of this typical pharmaceutical material can be measured using a modified annular shear rheometer and that the results can be interpreted in terms of the μ(I)-rheology model. It is demonstrated that both the simulation results and the experimental data show a constant hopper discharge rate. It is noted that the model can suffer from ill-posedness and it is shown how an increasingly fine grid resolution can result in predictions that are not entirely physically realistic. This shortcoming of the numerical framework implies that caution is required when making a one-to-one comparison with experimental data.

Original languageEnglish
Article number105085
JournalEuropean Journal of Pharmaceutical Sciences
Volume142
DOIs
Publication statusPublished - 15 Jan 2020
Externally publishedYes

Keywords

  • Hopper
  • Inertial number
  • Powder
  • Rheology
  • Volume-Of-Fluid

Fingerprint

Dive into the research topics of 'Characterization of microcrystalline cellulose spheres and prediction of hopper flow based on a μ(I)-rheology model'. Together they form a unique fingerprint.

Cite this