Enhancing reproducibility in mixing time determination of stirred tank reactors via automated analysis and standardized inter-laboratory trials.

Determining the mixing time in a mixing apparatus enables the evaluation of mixing quality and, therefore, represents a valuable tool for characterizing unit operations in process engineering. Mixing is highly relevant in both upstream processes (e.g., bioreactors) and downstream processes (e.g., blending tanks in diafiltration) in biotechnology and pharmaceutics. The aim of combining a colorimetric method with video capture and automated image analysis is to provide a robust, standardized methodology for determining mixing times in transparent bioreactors, such as laboratory-scale glass bioreactors and single-use bioreactors. A Python-based tool for video analysis was developed for this purpose. The Python script provides different mixing indices that enable the evaluation of both global and local mixing times under varying stirring speeds and aeration rates. This approach offers deep insights into the mixing process and enables the identification of heterogeneities. A round-robin study involving members of the DECHEMA Working Group "Single-Use Technologies for Bio-Based Applications" is currently being conducted to enhance the analysis and ensure reproducibility across different laboratories, thereby providing a robust basis for validating computational-fluid dynamics simulations. Initial reproducibility issues were identified and addressed in the standard operating procedure, demonstrating that a framework for automated analysis and improved standardization has been established. KEY POINTS: • Mixing-time experiments conducted at different aeration rates and stirring speeds • Automated video analysis for determining mixing times on both global and local scales • Identification of heterogeneities through visualization of local mixing times.