Recent applications of first- and second-order multivariate calibration to analytical chemistry

Area of Science:

• Analytical Chemistry
• Chemometrics
• Data Science

Background:

• Multivariate calibration techniques are crucial for analyzing complex samples in industrial, pharmaceutical, food, and environmental sectors.
• First-order data (e.g., spectra, chromatograms) and second-order data require specific analytical approaches.
• Process control benefits significantly from advanced data analysis methods.

Purpose of the Study:

• To review recent literature on multivariate calibration techniques for analytical determination.
• To highlight the advantages of combining specific multivariate data types with algorithms.
• To focus on the application of partial least-squares regression for first-order data and discuss second-order data advancements.

Main Methods:

• Review of scientific literature focusing on multivariate calibration.
• Description of common data processing tools and algorithms.
• Emphasis on partial least-squares regression for first-order data analysis.
• Discussion of emerging techniques for second-order data.

Main Results:

• Partial least-squares regression is the standard for first-order analytical data.
• Specific combinations of multivariate data and algorithms offer significant advantages.
• Second-order data analysis is a rapidly growing field with increasing potential.

Conclusions:

• Multivariate calibration, especially with partial least-squares regression, is a powerful tool for analytical determination.
• The application of these techniques spans diverse fields, including process control.
• Further research into second-order data methods is warranted due to their rapid development.