The effect of filtering and automated integration of chromatographic data on the calibration curve and detection limit was assessed. In a first approach simulated chromatograms were used to quantify the effects of data processing. Three types of filters were used: Savitzky-Golay, Fourier and Wavelet filter. The filter parameters chosen have been optimized in a previous study. The simulated data have been integrated by a commercial software package. The use of applying the DIN 32645 concept for the determination of the detection limit of chromatographic data is discussed and opposed to the concept of the method detection limit. Under the conditions investigated, filtering can improve the limit of detection up to a factor of three. This can be explained by the fact that filtering reduces the variance of the peak area and height and the limit of detection is mainly determined by their variance. However, the integration algorithm practically limits the possible improvements by filtering the data.
[1]
Ingrid Daubechies,et al.
Ten Lectures on Wavelets
,
1992
.
[2]
John C. Berridge,et al.
Quantitative column liquid chromatography — a survey of chemometric methods (journal of chromatography library, vol. 29.) : Elsevier, Amsterdam, 1984 (ISBN 0-444-42393-1). xiv + 300 pp. Price Dfl 165.00 ($63.50 in the U.S.A. and Canada).
,
1984
.
[3]
D. Massart.
Chemometrics: A Textbook
,
1988
.
[4]
Alan V. Oppenheim,et al.
Discrete-Time Signal Pro-cessing
,
1989
.