Experimental and numerical investigations of mixing in raceway ponds for algae cultivation

The current high interest in the algae sector is leading to the development of several demo/commercial scale projects, either for the food market or bioenergy production. Raceway Ponds (RWPs) are a widely used technology for algae mass cultivation. RWPs were developed long time ago, and thus capital and operating costs are well assessed. Nevertheless, room still exists to further reduce operational costs. A possible route towards energy optimization and therefore operational cost reduction can be identified through a better understanding of the mixing phenomena. The focus of the present work is that vertical mixing, defined as the cyclical movement of the algal cells between surface and bottom layers of the culture, cannot be completely determined by considering only turbulence, and therefore it is not represented by the Re number. A 3D Computational Fluid Dynamic (CFD) analysis of a conventional RWP was carried out based on a multi-phase “Volume of Fluid” model, in order to investigate the flow field of the culture in the pond. The CFD results were compared with experimental measures on a 20 m2 pilot RWP. Once agreement among CFD and experimental results was shown, a statistical evaluation of the trajectories calculated for algae particles in the flow was carried out. The aim of this statistical evaluation was to define the level of vertical mixing in different sections of the pond. The model proposed was then used to scale-up the results to a demo/pre-commercial size RWP (500 m2). The standard deviation of the actual trajectory was calculated with respect to the undisturbed trajectory for each section modeled. The results of the simulation showed that a limited mixing is to be expected in RWPs. In the long straight parts of the pond vertical mixing is poor and algae tend to settle to the bottom. Only in the bends the vortexes produced by flow separation move part of the culture from the bottom to the top and vice-versa. This result does not fit with the practice, typically observed in large scale ponds, of reducing vortexes around the bends by placing baffles. The method described can be applied to different pond designs operated at different culture velocities.