(13f) Biomass Measurements in Filamentous Fermentations: Comparison of Advanced On-Line Sensors | AIChE

(13f) Biomass Measurements in Filamentous Fermentations: Comparison of Advanced On-Line Sensors

Authors 

Gernaey, K. V. - Presenter, Technical University of Denmark
Petersen, N. - Presenter, Technical University of Denmark (DTU)
Stocks, S. M. - Presenter, Novozymes A/S
Eliasson Lantz, A. - Presenter, Technical University of Denmark (DTU)


Reliable biomass measurements are critical for efficient monitoring and control of fermentation processes. Despite the development of a number of promising technologies for in situ monitoring, biomass measurements remain a challenge particularly for highly aerated fermentations with organisms of complex morphology such as filamentous bacteria and fungi. The objective of this study was to compare a range of different technologies for biomass monitoring in a challenging and industrially relevant system. Eight Streptomyces coelicolor fed batch fermentations were run as part of a process development in which pH, feeding strategy and the medium composition were varied. All fermentations were monitored in situ using near infrared spectroscopy, multi wavelength fluorescence spectroscopy, optical density and capacitance measurements. In addition, all of the traditional fermentation data such as the concentration of CO2 and O2 in the exhaust gas and the concentration of dissolved oxygen in the fermenter were logged. To our knowledge, a direct comparison between the different technologies in one set of fermentations has never previously been made. Prediction models for the biomass concentration were estimated based on the individual data sets as well as on combinations of the data sets. The models were validated on a new batch, which was independent of model development. The batches used for model calibration and validation covered a wide range of biomass concentrations, medium compositions and different pH values. All of the sensors gave satisfactory predictions of the biomass concentration in different subsets of the fermentation batches. The range of biomass concentrations and process conditions covered by the different sensors highlighted the strengths and the limitations of the technologies. By using combinations of the sensors it was possible to provide online validation of the sensors and thus provide more robust estimations of the biomass concentration. This study is an example of how process analytical technology can be integrated into the process development providing a monitoring strategy in addition to an optimized process recipe.