Cross-flow microfiltration using ceramic membranes applied to the cuttlefish effluents treatment: effect of operating parameters and the addition of pre or post-treatment

Cross-flow microfiltration alone or associated with ultrafiltration or coagulation was conducted to investigate the efficiency of this process in the reducing of chemical oxygen demand (COD) and proteins (Npr) in the waste waters produced from cuttlefish conditioning and freezing process. The study of the effect of operating parameters on the fouling of a ceramic microfiltration membrane (velocities and pressures) using real effluent showed that the maximum permeate flux was obtained at 5.6 m/s and 2 bar; it was about 120 l/h.m2. For the same initial COD, the last value increased by 28% and 40% respectively when the conductivity decreased from 50 to 0.6 mS/cm and the pH decreased from 7 to 4. However, the turbidity of the permeate and the retention rates of COD and nitrogen proteins were not affected by the variation of these parameters (0.8 NTU, 55% and 60% respectively). The flux enhancement was also studied by using a combination of microfiltration-ultrafiltration and aluminium salt and MgO coagulation-microfiltation. The same performances were obtained in the case of the use of combined SA-MgO-MF and MF-UF. The total resistance of fouling was less important in the latter cases. However, the irreversible resistance remained the main component in all cases. Maximum steady state flux was then 140 l/h.m2 obtained at the previously determined optimum conditions. The corresponding rejection of COD was better for SA-MgO-MF treatment (75% against 65%). However, the retention rate of nitrogen proteins was better for MF-UF treatment (78% against 57%).

The membrane processes offer a very efficient method of treating cuttlefish effluents. Based on permeate flux performances, the best ceramic MF operating conditions applied to raw effluents were: 2 bars transmembrane pressures and 5.6 m/s cross-flow velocity. The variation in feed effluent quality was also studied through the use of various reconstituted cuttlefish effluent samples differing by COD, pH and conductivity. It is found that permeate flux was also improved compared to the real effluent conditions by 25% and 26% when the model effluent conductivity and pH decreased from 50 to 0.6 mS/cm and from 7 to 4 respectively. The variation of these different parameters is observed to be relatively ineffective in COD and nitrogen proteins retention rate, which are about 55% and 60% respectively. To yield higher permeate fluxes and to improve the treated effluent quality, MF was used in conjunction with UF or chemical coagulation. The use of SA coagulation prior to microfiltration did not affect the stabilised permeate flux. However, water flux after washing improved significantly when the treatment was accomplished by using the combined (MF/UF). Thus, it appears that the fluxes decrease in the MF/UF and SA-MgO/MF are affected by the filtration process and the pretreatment nature. This result may be attributed to the change in molecular size on the effluent after each treatment, which affects the fouling properties. Hence, we can say that membrane processes with a suitable selection of membrane cut-off and pre-treatment step could be used to preserve the environment with water recycling into the process and protein recovery. Acknowledgment The financial support for this work received from the Calembo Company, Sfax, Tunisia and Higher Education and Technology and Scientific Research Ministry within the VRR project, is greatly appreciated.