Addressing raw material variation: Maintaining a steady-state during cultivation by blending of lignocellulosic feed streams

Abstract

The transition to a sustainable bioeconomy necessitates the utilization of second-generation renewable resources. However, the inherent variability in the composition of these resources poses a challenge for establishing stable continuous bioprocesses. Industrial side-streams, such as spent sulfite liquor (SSL) from the pulp and paper industry, offer a promising starting point. Unlike first-generation processes, which rely on constant feed compositions, addressing the uncertainties of renewable resources requires robust monitoring and control strategies. To tackle the nonlinear dynamics of microbial conversion, we propose a model-based controller employing feedback linearization of a nonlinear model. Specifically, this study is focused on the uptake and bioconversion of sugars within SSL by Corynebacterium glutamicum. The controller dynamically adjusts mass flows into and out of the bioreactor to maintain steady-state conditions despite fluctuations in feed composition. Additionally, wood hydrolysate, rich in glucose and also from the paper and pulp industry, supplements the feed, enhancing control capabilities. Simulation studies showed that by actively controlling the process and adjusting the feed, harvest and bleed rates based on the sugar concentrations in the feed, process variability can be significantly reduced while sugar utilization increases. Experimentally, the introduced process was verified by continuous cultivation, whereby the process was effectively controlled and kept stable for more than 120 h with residual sugar concentrations always below 3.2 g/L. The decoupling of the multiple-input multiple-output system enables simultaneous and independent control of various process variables, establishing continuous cultivation processes resilient to raw material uncertainties and fluctuations.