P3.11 Recommendations for optimal sampling and filter rates in liquid chromatography
- Event
- 20. GMA/ITG-Fachtagung Sensoren und Messsysteme 2019
2019-06-25 - 2019-06-26
Nürnberg, Germany - Chapter
- P3: Anwendungen
- Author(s)
- D. Marehn, D. Wilhelm - AnaTox GmbH & Co. KG, Fürstenwalde (Deutschland), R. Pizzoferrato - Università degli Studi Di Roma ”Tor Vergata“, Rom (Italien), H. Pospisil - Technische Universtität Wildau (Deutschland)
- Pages
- 800 - 805
- DOI
- 10.5162/sensoren2019/P3.11
- ISBN
- 978-3-9819376-0-2
- Price
- free
Abstract
In the strong regulated field of pharmacy, all substance analysis are performed by very well defined experiments using specific techniques and parameters for each substance, which are written down in the pharmacopoeias, mainly the United State Pharmacopoeia (USP), Japanese Pharmacopoeia (JP) or European Pharmacopoeia (EP). Some of these experiments use techniques which belong to the field of chromatography especially the high performance liquid chromatography (HPLC) and gas chromatography (GC). Nearly all possible parameters are listed in the so called monographs of the active substances. Nevertheless two editable parameters of the detectors are not mentioned and not restricted here: the sampling rate and the signal filtration. Both parameters have a clear impact on the peak width, peak height and peak distortion. Therefore, it would be desirable to define recommendable setups for chromatography tests.
In this article we present how to find the optimal values for the sampling rate and signal filtration for a given substance of a monograph. Chromatograms with the highest possible sampling rate and lowest filtration have been altered by performing a sampling reduction and various signal filtration algorithms considering the detector firmware of the major HPLC manufacturers. For the evaluation, several chromatographic properties describing a chromatogram have been taken for an optimization process based on the multi objective (Pareto) optimization. The optimization has been tested on simulated and real HPLC data. The suitability has been proved by comparing the results against the usual approach of 20 points per peak.
The criterion allows to make further investigations on the substance monographs in order to develop the sampling rate and signal filtration recommendations depending on the HPLC system firmware.