Modeling and simulation of compact strength due to particle bonding using a hybrid discrete-continuum approach

A. Koynov, I. Akseli, A. M. Cuitiño

Research output: Contribution to journalArticlepeer-review

15 Scopus citations


The compaction of powder beds into solid bodies occurs by virtue of the formation of inter-particle bonds. The mechanical strength of the compact depends on the type of bonding interaction, as well as, the inter-particle contact area created in the compression process. A hybrid quasi-continuum computational approach has been implemented to study the bonding occurring in compressed granular assemblies. The approach resolves the powder bed on the particle level, allowing for the tracking of contact area generated by particle deformation and the computation of history-dependent inter-particle bonding forces. The magnitude of the bonding force is calculated using a synthetically constructed potential aiming to mimic the shape of typical molecular-type interactions. The uniaxial compaction and subsequent relaxation of powder beds, representative of pharmaceutical excipients have been simulated. Due to the bonding occurring between the individual particles the compressed beds acquire tensile strength. Post-relaxation tensile loading of the compacts is used to quantify the magnitude of this tensile strength. To validate the computational results, tablets are prepared using a compaction simulator under conditions closely resembling the simulated scenarios, where subsequently the tablets are subjected to tensile loads until failure. The predicted values for the tablet strength utilizing the present methodology capture the general trends exhibited by the experimental record.

Original languageEnglish (US)
Pages (from-to)273-285
Number of pages13
JournalInternational Journal of Pharmaceutics
Issue number2
StatePublished - Oct 14 2011

All Science Journal Classification (ASJC) codes

  • Pharmaceutical Science


Dive into the research topics of 'Modeling and simulation of compact strength due to particle bonding using a hybrid discrete-continuum approach'. Together they form a unique fingerprint.

Cite this