Baseline Noise in High-Performance Liquid Chromatography with Electrochemical Detection

The chromatographic baseline was converted to power spectrum by Fourier transform to show the frequency dependence of noise. The spectral analysis of chromatographic baseline was made under various experimental conditions for high-performance liquid chromatography with electrochemical detection (HPLC-ECD). From the spectral analysis, it was found easily that a system using a small stroke pump and a wall jet type cell with a plastic formed carbon working electrode had a low level of the baseline noise. The optimization of HPLC-ECD system was performed by evaluation of the detection limit based on a chemometric tool, called the Function of Mutual Information (FUMI) theory, which treats the chromatographic baseline fluctuation as a 1/f fluctuation model. We found that the observed relative standard deviation (RSD) of peak area by repetitive measurements (n = 5) of chromatograms was parallel to the predicted RSD curve based on the FUMI theory. The result indicates that the FUMI theory can provide a measure for estimating the precise and reliable detection limit from a single measurement of noise and signal in HPLC-ECD. The present spectral analysis of baseline noise and the prediction of RSD based on the FUMI theory are an efficient optimization strategy for determining catechins using HPLC-ECD.