Assessment of a bio-molecular sensor in the operation of adsorption processes-A model based approach

Adoption of digitalisation and Industry 4.0 concepts is gaining pace in food manufacturing and bio-based processes [1]. An essential aspect is the introduction of real time sensors in order to improve process monitoring, decision making and process control. To this end, we assess the use of a novel technology for the real-time sensing of biomolecular substances [2, 3] for the monitoring and control of adsorption processes in the food industry.
In recent years, the mechanistic modeling of adsorption processes has got attention since they can be used in optimization, scale-up etc. However, the development of bottom-up models takes a long time and requires estimation of several process parameters from experimental data. Instead, we make use of simple mathematical models (double exponential or logistic function, [4]) to describe adsorption breakthrough curves, which express the concentration of an adsorbate in the fluid phase at the outlet of an adsorption column as a function of time [5]. Our work shows that such a simple model can be used to study how a biosensor placed at the inlet of a column can be used to steer decisions when to stop the adsorptions process. Furthermore, the model can also be used to design a feedback -feedforward control strategy to achieve a constant concentration at the outlet of a series of adsorption columns which was studied for an extraction process of an anti-nutritional factor. The results of our study shows that the integration of a real-time sensing technology could help to track, analyze and optimize production processes in the food industry.
This work was partly funded by The Netherlands Topsectors Agri&Food, HTSM, and Chemistry under contract number LWV20.117.