Repository logo

Predictive modelling of powder compaction for binary mixtures using the finite element method

Published version

Change log


van der Haven, Dingeman LH 
ørtoft, Frederik H 
Naelapää, Kaisa 
Fragkopoulos, Ioannis S 
Elliott, James A 


Despite the widespread use of solid-form drug delivery within the pharmaceutical industry, tablets remain challenging to formulate because their properties depend strongly on the powder composition and details of the compaction process. Powder compaction simulations, using the finite element method (FEM) in combination with the density-dependent Drucker-Prager Cap model, can be used to aid the design process of pharmaceutical tablets. Parametrisation is typically carried out manually and requires experimental data for each powder considered. This becomes cumbersome when considering different ratios of component powders. An automated parameterisation workflow was developed and validated using experimental powder mixtures of microcrystalline cellulose and dibasic calcium phosphate dihydrate. FEM simulations reproduced experimental compaction curves with a mean error of 2.5% of the maximum compaction pressure. Moreover, a mixing methodology was developed to estimate parameters of mixtures using only pure-component parameters as input. The experimental compaction curves of mixtures were predicted with a mean error of 4.8%.



Powder compaction, Drucker-Prager Cap model, Predictive modelling, Mixing model, Mechanical properties, Formulation

Journal Title

Powder Technology

Conference Name

Journal ISSN


Volume Title


Elsevier BV