Multicomponent analysis: A case report
Three cases are described. Case 1: kinetic studies often need high time resolution measurements in order to follow the pattern of reactions taking place during the experiment. This is often laborious to achieve with the collection of fractions for chromatographic separation. Some tool for separation...
Gespeichert in:
Veröffentlicht in: | Journal of pharmaceutical and biomedical analysis 1991, Vol.9 (8), p.643-654 |
---|---|
Hauptverfasser: | , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Three cases are described. Case 1: kinetic studies often need high time resolution measurements in order to follow the pattern of reactions taking place during the experiment. This is often laborious to achieve with the collection of fractions for chromatographic separation. Some tool for separation is, however, necessary in order to decompose the concentrations of reactants, products and intermediate species. The spectra of the intermediates may not be known at the time when the kinetic studies are needed. With unknown spectra there are still possibilities to use spectroscopy and multivariate techniques to obtain qualitative information. Case 2: it is possible to use Partial Least Squares (PLS) in order to describe the chromatographic profiles for the species even if the separation is insufficient for traditional peak measurement methods. This requires that mixtures are available with known concentrations of the species to be determined. Case 3: with modern diode array liquid chromatography detectors there is the possibility to capture the chromatogram and the spectra at the same time. The ability to reproduce the chromatographic profile between samples makes it possible to use the Generalized Rank Annihilation Method (GRAM) possible. Whereas PLS only treats one spectrum at a time, this method treats the full two-dimensional chromatogram as an entity. The GRAM calibration is claimed to be insensitive to interfering species which are not present in the calibration. Limitations are that GRAM requires a linear detector response and very good repeatability of the retention time. The use of GRAM for calibration with real samples is demonstrated. |
---|---|
ISSN: | 0731-7085 1873-264X |
DOI: | 10.1016/0731-7085(91)80190-K |