Evaluation of droplet size distributions using univariate and multivariate approaches

Research output: Contribution to journalJournal articleResearchpeer-review

M.H. Gauno, C.C. Larsen, T. Vilhelmsen, Jørn Sonnergaard, J. Wittendorff, J. Rantanen

Pharmaceutically relevant material characteristics are often analyzed based on univariate descriptors instead of utilizing the whole information available in the full distribution. One example is droplet size distribution, which is often described by the median droplet size and the width of the distribution. The current study was aiming to compare univariate and multivariate approach in evaluating droplet size distributions. As a model system, the atomization of a coating solution from a two-fluid nozzle was investigated. The effect of three process parameters (concentration of ethyl cellulose in ethanol, atomizing air pressure, and flow rate of coating solution) on the droplet size and droplet size distribution using a full mixed factorial design was used. The droplet size produced by a two-fluid nozzle was measured by laser diffraction and reported as volume based size distribution. Investigation of loading and score plots from principal component analysis (PCA) revealed additional information on the droplet size distributions and it was possible to identify univariate statistics (volume median droplet size), which were similar, however, originating from varying droplet size distributions. The multivariate data analysis was proven to be an efficient tool for evaluating the full information contained in a distribution.
Original languageEnglish
JournalPharmaceutical Development and Technology
Volume18
Issue number4
Pages (from-to)926-934
Number of pages9
ISSN1083-7450
DOIs
Publication statusPublished - 1 Jul 2013

ID: 62955691