Qualitative and Quantitative Analysis of Unresolved Responses In Liquid Chromatoraphy With Fourier Transform Infrared Spectroscopic Detection By Using the Kalman Filter

Abstract

The optimization of chromatographic separations is hindered by difficulties in establishing the identities of components, and by the need to have well-resolved chromatographic peaks for commonly-used methods of quantitation. Full spectra methods, such as Fourier transform infrared spectroscopy, may assist in establishing identities, but are hampered by poor sensitivity. A method is proposed which efficiently reduces the large amounts of spectral data obtained in such cases, and enhances signal-to-noise ratios to permit identification and quantitation of components present in well-and poorly-resolved chromatographic peaks obtained under nonoptimal conditions. the method is based on combined Gram-Schmidt orthogo-nalization for removal of the dominant response from the mobile phase, followed by Kalman filtering for qualitative and quantitative analysis of averaged spectra. the method is demonstrated for infrared detection of a reverse-phase separation of a five-component mixture. Although Kalman filtering is effective at enhancing weak

Key Words:

• Fourier transform infrared detection
• data analysis
• Kalman filtering