A flat plate and a tubular packed-bed photobioreactor with an algal-bacterial biofilm attached onto Poraver (R) beads carriers, a flat plate and a tubular photobioreactor with the biofilm attached onto the reactor walls, and an algal-turf reactor were compared in terms of BOD removal efficiencies, elimination capacities, and stability. A control column photobioreactor with suspended algal-bacterial biomass was also tested to compare the performance of biofilm photobioreactors with conventional algal-based processes. When the algal-bacteria I biomass was immobilized onto Poraver (R) the process never reached a steady state due to a poor homogenization in the bioreactor. When the biofilm was formed onto the reactor wall (or reactor base) the process was stable. A Maximum degradation rate of 295 mg BODI-1 h(-1) was achieved in the algal-turf reactor although control experiments performed in the dark showed atmospheric O-2 diffusion represented 55% of the oxygenation capacity in this system. BOD removal rates of 108, and 92 mg BODI-1 h(-1) were achieved in the tubular and flat plate biofilm reactors, respectively, compared to 77 mg BODI-1 h(-1) in the control Suspended bioreactor. In addition, all biofilm photobioreactors produced an easily settleable biomass. Evidence was found that biomass attachment to the reactor's wall improved stability. (C) 2008 Elsevier B.V. All rights reserved.