Grouping of residual solvents present in pharmaceuticals using experimental planning and chemometric methods.

The main effects of six experimental factors on the efficiency of HS (headspace) extraction in headspace gas chromatography--flame ionization detector (HS-GC-FID) determination of twenty organic solvents routinely used in production of pharmaceuticals were obtained on the basis of the results of experiments carried out according to the Plackett-Burman factorial design. The effects were used as a basis for grouping the solvents into five groups, the solvents belonging to a group responded similarly to changes of HS conditions. To this end, visualization approaches were used as well as chemometric methods: cluster analysis (CA) and principal component analysis (PCA). Moreover, the most important HS experimental factors were selected for further optimization of the HS-GC determination procedure.