Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
9918969 | International Journal of Pharmaceutics | 2005 | 11 Pages |
Abstract
There are a number of situations where there is a need to determine the concentrations of components in solid-state mixtures without dissolving the sample. Diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) coupled with partial-least-squares (PLS) data analysis has been used to determine the minor component in a mixture of structurally similar solid-state compounds, in this case mixtures of ephedrine and pseudoephedrine. Factors that limit the precision and accuracy of the determinations are discussed. It is shown that when care is taken to produce homogeneous calibration samples very good results can be obtained, in this case cross-validated standard error of predictions of 0.74Â wt% when the minor component spanned the concentration range of 0-50Â wt%, and 0.11Â wt% when the minor component spanned the concentration range of 0-5Â wt%. Results are presented that indicate that the amount of data available to the PLS calibration routine relative to the range over which the calibration is performed can limit the precision and accuracy of the determinations.
Keywords
Related Topics
Health Sciences
Pharmacology, Toxicology and Pharmaceutical Science
Pharmaceutical Science
Authors
Yanga K. Dijiba, Anding Zhang, Thomas M. Niemczyk,