Étude préliminaire du traitement d’effluents contenant de l’encre de seiche par centrifugation et procédés à membranes Preliminary study of the treatment of effluents containing cuttlefish ink by centrifugation and membrane processes

Résumé

Il s’agit d’une étude préliminaire sur le traitement d’effluents de conditionnement de la seiche avant congélation en vue de réduire la charge polluante des rejets et de valoriser l’encre qu’ils contiennent. Deux types de procédés ont été mis en oeuvre : d’une part, la centrifugation, qui permet de fractionner la suspension d’encre de seiche entre un culot noir à DCO (Demande Chimique en Oxygène) élevée et un surnageant limpide et, d’autre part, l’ultrafiltration (UF) et la microfiltration (MF). Les flux de perméat obtenus par les deux procédés à membranes sont du même ordre de grandeur (25 à 30 L·h^‑1·m^‑2 sous 1,5 bar). La rétention moyenne en DCO est de 65 % et la rétention en COT (Carbone Organique Total) et azote protéique (NTK) de plus de 95 %. Cependant le colmatage irréversible de la membrane de MF conduit à préférer l’UF, plus facilement régénérable.

Mots clés:

• centrifugation,
• microfiltration,
• ultrafiltration,
• dépollution,
• encre de seiche

Summary

Industries that condition fish products have to cope with the problem of processing their usually protein-rich wastewaters. An example of such an industry that discards a large amount of wastewater is the CALEMBO Company (Sfax-Tunisia), which uses 50 m³ per metric ton a day to condition cuttlefish for freezing. In order to conserve water, high-salinity bore water is sometimes used. This high salinity water is responsible for the difficulties encountered during the biological treatment of wastewaters and the recovery of valuable by-products. In this respect, membrane processes, used in the treatment and exploitation of effluents from industries that process sea products, are very attractive. The first membrane filtration trials on sea-product effluents date back to the 1980’s, but they did not result in major developments. Legislative pressures and the increasing costs of water and effluent-processing, as well as the improvement of membrane efficiencies, have made membrane treatment processes much more interesting for wastewater treatment processes. The GEPEA Laboratory at Nantes University has carried out research on membrane technologies to clean up polluted process waters, enhance substances such as soluble fish proteins, and to recover substances responsible for the flavour of bivalves and shellfish.

This paper presents preliminary research on the treatment and exploitation of water used in cuttlefish conditioning. Treatment processes used include centrifugation, microfiltration and ultrafiltration. Centrifugation is used to determine the distribution of the effluent between the black residue and the clear supernatant, whereas membrane processing is used to reduce wastewater pollution and concentrate pigments.

The effluent studied was reconstituted from pure cuttlefish-ink samples taken directly from the animal and salt waters of the same salinity as the bore water used by the CALEMBO Company (Table 1). The samples were reconstituted in ratios of 1 to 50 for centrifugation and 1 to 100 for membrane filtration. Centrifugation trials were carried out using a KR 22i type JOUAN centrifuge, whereas ultrafiltration and microfiltration trials were carried out using the laboratory apparatus represented in figure 1. The main characteristics of membranes used are indicated in table 2. Operating conditions were determined according to the capacities of the feed pump: transmembrane pressure P_tm = 1.5 bar, circulation velocity U = 1.5 m·s^‑1 and temperature T = 25°C. The parameters measured on initial feed solutions and the fractions obtained were COD (Chemical Oxygen Demand), TOC (Total Organic Carbon) and nitrogen content (NTK). Filtration trials were carried out according to two different procedures, either with constant feed composition to determine the best operating conditions, or with increasing effluent concentration together with monitoring of the Volumetric Reduction Factor (VRF).

Centrifugation of the cuttlefish-ink suspension produced two phases: a very dense black residue and relatively clear supernatant. The volumetric distribution and the COD and TOC contents of the different fractions are presented in table 3. The supernatant represented about 75% of crude effluent volume. Organic matter was concentrated in the residue and consisted primarily of suspended particles.

At a constant concentration, the ultrafiltration (UF) and microfiltration (MF) processes behaved differently. A rapid drop in flux in the first minutes followed by stabilization at 30 L·h^‑1·m^‑2 after 30 min was observed for the MF process, whereas a rapid stabilization at approximately 25 L·h^‑1·m^‑2 was observed for the UF process. The drop in flux at the beginning of MF process may be due to the partial fouling of the membrane pores by melanin particles ranging in sizes from 55 to 160 nm, which are of the same order of magnitude as the membrane pores of 100 nm. On the other hand, the small decrease in flux in the case of ultrafiltration resulted essentially from the formation of a polarization layer and possible interactions between the membrane material and the solution.

Batch-concentration trials were carried out for 5 and 4 h using UF and MF respectively, the operating time being dictated by the dead volume of the equipment (0.75 L). The permeate flux variation as a function of the volumetric reduction factor (VRF) is illustrated in figure 3. The MF flux was slightly higher despite the higher initial concentration of organic substances. For a VRF of 2.64 (final concentration), J = 2.8 L·h^‑1·m^‑2 for MF and 15.2 L·h^‑1·m^‑2 for UF. Despite the significantly different permeabilities of the MR and UF membranes to pure water (2690 against 34 L·h^‑1·m^‑2·bar^‑1), their very similar J values are a consequence of the internal pore fouling of the MF membranes.

Analyses performed on the initial feed samples, and on the different fractions of ink suspensions obtained by MF and UF following concentration, are presented in table 5. Retention ratios for UF were very slightly higher than those found for MF, about 65% for COD, 98% for TOC and 95% for NTK. From the point of view of pollution remediation, and considering permeate COD values, the efficiency of the membrane technique does not seem sufficient.

Following ultrafiltration, membrane regeneration was possible by simply rinsing the membrane with water. On the other hand, the same procedure proved inefficient for the microfiltration (PVDF) membrane. The black pigment remained stuck to the membrane surface and most likely inside the pores as well. Furthermore, chemical regeneration (NaOH 0.1 M, 20 min, 25°C) was not enough to recover the membrane’s initial permeability.

To conclude, the ultrafiltration process is better adapted to the treatment of cuttlefish washing wastewater. However, considering the level of residual COD in the ultrafiltration permeate, more efficient post-treatment techniques must be developed.

Key Words:

• ultracentrifugation,
• microfiltration,
• ultrafiltration,
• pollution remediation,
• cuttlefish ink