Sulphate removal by membrane filtration minimizes the risk of hydrogen sulphide formation in fixed bed biofilters

• Published in: Aquacultural Engineering
• Main Authors: Fernandes, Paulo M., Aalto, Sanni L., Åsnes, Helga Ø., Rojas-Tirado, Paula, Åtland, Åse, Letelier-Gordo, Carlos O.
• Format: Article in Journal/Newspaper
• Language: English
• Published: 2024
• Subjects: Atlantic salmon; Brackish water; Dissolved oxygen; H2S; Oxidation-reduction potential; RAS; Norway
• Online Access: https://orbit.dtu.dk/en/publications/e4b95711-fc0d-499b-91ae-6af2b7df528a; https://doi.org/10.1016/j.aquaeng.2024.102426; https://backend.orbit.dtu.dk/ws/files/360800044/1-s2.0-S0144860924000372-main.pdf

Hydrogen sulphide (H 2 S) is one of the suspected reasons behind sudden mass fish mortalities in recirculating aquaculture systems (RAS) in recent years. H 2 S production in aquaculture systems depends on sulphate and organic matter availability, presence of specific microbial groups, and local anoxic conditions. Specific potential H 2 S production hotspots in RAS have been identified within biofilters and in accumulated sludge. Current H 2 S control methods have been identified, such as improved hydrodynamics, increasing degassing efficiency, chemical addition of hydrogen peroxide or ozone, but have not been efficient or widespread applied. In this study, a nanomembrane filtration system was installed at a brackish water (mixture of seawater and freshwater to 15 ppt) smolt production site in Norway to remove sulphate ions from the seawater intake line (15 times reduction). The hydrogen sulphide production potential of the nanofiltered seawater mixed with freshwater (n = 3) was compared to an unfiltered seawater and freshwater mixture (15 ppt, n = 3) for 42 days in experimental scale biofilters using industrial fixed bed media. In both treatments, the linear production of H 2 S started around the time that bulk water measurements of oxidation-reduction potential (ORP) and dissolved oxygen (DO) dropped below 0 mV and 1 mg/L, respectively. As expected, the highest H 2 S concentration was observed in unfiltered water reactors, which also reached the highest concentration faster than filtered-water reactors. A 15 times reduction in initial sulphate levels by the nanofiltration membrane led to overall three times lower H 2 S production and delayed the onset of production by two days. Hence, membrane-filtering intake water decreased the risk of H 2 S production. A limitation in this study, however, was that sulphate was not completely removed from the intake water, and the next steps should evaluate how increasing the effort of membrane operation to completely remove sulphate affect the dynamics of H 2 S production in ...