Second-order analyte quantitation under identical profiles in one data dimension. A dependency-adapted partial...

Area of Science:

• Analytical Chemistry
• Chemometrics

Background:

• Multivariate calibration methods like partial least-squares with residual bilinearization are used for analyte quantitation from second-order data.
• Existing methods fail when interfering agents and calibrated components share identical profiles in one data dimension.

Purpose of the Study:

• To introduce a novel residual bilinearization procedure to address limitations in analyte quantitation.
• To evaluate the new method's performance against established techniques for complex analytical problems.

Main Methods:

• Development of a new residual bilinearization procedure designed to handle linear dependencies.
• Validation using simulated data to assess performance against multivariate curve resolution-alternating least-squares and parallel factor analysis.
• Application to experimental data for determining ciprofloxacin in urine and serum samples.

Main Results:

• The new residual bilinearization model successfully handles analytical problems involving linear dependencies.
• Performance was comparable to multivariate curve resolution-alternating least-squares and adapted parallel factor analysis.
• Good analytical performance was observed in experimental determinations of ciprofloxacin.

Conclusions:

• The novel residual bilinearization procedure effectively overcomes limitations of classical methods for analyte quantitation.
• The approach demonstrates robust performance in both simulated and real-world analytical scenarios.
• This method offers a viable alternative for accurate analyte determination in complex matrices.