Loh, Jonathan; Ketterhagen, William; Elliott, James
Multiscale Modelling of Pharmaceutical Powders: Macroscopic Behaviour Prediction
POWDERS AND GRAINS 2013, 1542:161-164, 2013

The pharmaceutical industry uses computer models at many stages during drug development. Quantum and molecular models are used to predict the crystal structures of potential active pharmaceutical ingredients (APIs), whereas discrete element models are used to optimise the mechanical properties of mixtures of APIs and excipient powders. The present work combines the strengths of modelling from all of the mentioned length scales to predict the behaviour of macroscopic powder granules from first principles using the molecular and crystal structures of acetazolamide as an example API. Starting with a single molecule of acetazolamide, ab initio self-consistent field calculations were used to calculate the equilibrium gas phase structure, vibrational spectra, interaction energy with water molecules and perform potential energy scans. By using these results and following the CHARMM General Force Field parameterisation process, all of the parameters required to perform a molecular dynamics simulation were iteratively determined using the CHARMM program. Next, by using crystallographic data from literature, the monoclinic and triclinic forms of the acetazolamide crystal were simulated. Material properties like the Young's modulus and Poisson ratio, and surface energies have been calculated. These material properties are then used as input parameters in a discrete element model containing Thornton's plastic model and the JKR cohesive force to predict the behaviour of macroscopic acetazolamide powder in angle of repose tests and tabletting simulations. Similar methodologies can be employed in the future to evaluate at an early stage the performance of novel APIs and excipients for tabletting applications.

DOI:10.1063/1.4811892

Find full text with Google Scholar.