Procedure for automated background correction in flow systems with infrared spectroscopic detection and changing...

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Chromatography

Background:

• On-line Fourier transform infrared (FT-IR) detection in gradient high-performance liquid chromatography (HPLC) is challenged by spectral contributions from varying solvent mixtures.
• Accurate analysis requires effective correction for solvent spectral interference, especially in complex gradient separations.

Purpose of the Study:

• To develop and validate an automated partial least squares (PLS) based procedure for correcting solvent spectral contributions in on-line FT-IR detection during gradient LC.
• To enable robust and accurate solute spectral analysis by effectively removing solvent background variations.

Main Methods:

• Utilized a partial least squares (PLS) regression approach for automated spectral correction.
• Required a reference FT-IR dataset covering expected solvent composition variations.
• Focused on a specific spectral region (2300-2400 cm⁻¹) where solvents absorb but analytes do not, exemplified by Acetonitrile:Water (ACN:H₂O) mixtures.
• Developed a numerical method to select and subtract appropriate background spectra from analyte spectra.

Main Results:

• The PLS-based procedure successfully corrected for solvent spectral contributions in on-line FT-IR detection.
• The method demonstrated effectiveness across various isocratic and gradient reversed-phase LC-FT-IR datasets.
• Validated the ability to perform accurate eluent correction in the on-line coupling of FT-IR spectrometry with gradient LC.

Conclusions:

• The developed automated PLS procedure provides a reliable solution for eluent correction in on-line LC-FT-IR.
• This method significantly improves the quality and accuracy of spectral data obtained from gradient LC separations.