Implementation of structured metal packing for the design of biofilm reactor : analysis by high energy X-ray tomography and application to the production of lipopeptides by Bacillus subtilis

1. Whereas multi-species biofilm reactors are commonly used for treatments of water and gas effluents, new strategies are arising for the development of mono-species biofilm reactors in order to produce high added value molecules. Thus, it is required to design new bioreactors able to promote the growth of the biomass on the form of a biofilm and to identify the key physico-chemical parameters involved in order to optimize the bioprocess.
2. Aim of this study was to investigate a pilot-scale biofilm reactor comprising a metal structured packing promoting growth of Bacillus subtilis as a biofilm for the production of lipopeptides, high added value compounds with high surface active properties.
3. In this work, the mechanical stirring system of a 20L stirred tank bioreactor has been removed and replaced by a metal structured packing positioned in the headspace of the vessel above a liquid phase. The culture medium is continuously recirculated on the packing thanks to a peristaltic pump and air supply is performed just above the liquid phase under the packing. High energy X-ray tomography was used to estimate non-invasively the biofilm distribution inside the packing and permitted to define parameters that affect scale-up. Performances of the biofilm reactor were compared with a submerged culture in a stirred tank reactor in terms of lipopeptides production.
4. After 72 hours of fermentation, 94 % of the total biomass adheres onto the metal packing on the form of a biofilm. The colonization of this latter has been visualized non-invasively by X-ray tomography directly inside the packing and shows a conical repartition of the biofilm mass (about 25% of the total volume of the packing) as well as the presence of clogging. However, unlike the submerged culture, no foam formation appeared during fermentation and surfactin yield reaches 345,4 ± 32,8 mg / L for the biofilm reactor against 277,3 ± 34,4 mg / L in the stirred tank reactor.
5. In conclusion, this experimental setting leads to a major technological progress avoiding foam formation and increasing surfactin production. Nevertheless, significant improvements are required at the level of the biofilm distribution in thin layers inside the packing in order to increase mass transfer and lipopeptides recoveries. Further investigations will be devoted to the optimization of the physico-chemical parameters involved in biofilm distribution.