Data model for the elimination of matrix effects in enzyme-based flow-injection systems

Area of Science:

• Biotechnology
• Analytical Chemistry
• Food Science

Background:

• Process and quality control in food processing and biotechnology often requires complex analytical methods.
• Existing systems frequently necessitate extensive sample preparation and dilution, increasing time and cost.
• Interferences from heterogeneous sample matrices pose a significant challenge in direct analysis.

Purpose of the Study:

• To introduce a new conceptual enzyme-based flow injection analysis (FIA) system.
• To enable the determination of various analytes in diverse process media using a unified experimental setup.
• To eliminate the need for sample preparation and dilution steps common in comparable systems.

Main Methods:

• Optimization of intrinsic system parameters for measurement range adaptation.
• Development of a specific injection mode to manage heterogeneous sample matrices.
• Utilization of dehydrogenases as indicator enzymes.
• Implementation of a specially developed data model employing cognitive methods to compensate for interferences.

Main Results:

• Successful elimination of cross-sensitivities and compensation for disturbed enzyme reaction rates.
• Demonstrated applicability through the analysis of ethanol in non-alcoholic beer.
• Validated through the online determination of D-/L-lactate during lactic acid fermentation.

Conclusions:

• The developed enzyme-based FIA system offers a robust and efficient approach for process and quality control.
• The system effectively overcomes matrix interferences without sample preparation, simplifying analysis.
• The methodology proves advantageous for real-time monitoring in food and biotechnological applications.