Registration Dossier

Ecotoxicological information

Short-term toxicity to fish

Currently viewing:

Administrative data

Link to relevant study record(s)

Description of key information

Two standardised test on Al2(SO4)3 and two non standardised tests on (NH4)2SO4 and AlNH4(SO4)2 were performed on two different freshwater fish species and one marine water fish species. The greatest sensitivity observed was related to aluminium toxicty at pH ranging between 4.8-5. The lowest reliable LC50(96h) correspond to 158 mg/L of AlNH4(SO4)2.12H2O.

Key value for chemical safety assessment

LC50 for freshwater fish:
158 mg/L
LC50 for marine water fish:
158 mg/L

Additional information

Two test performed according to OECD 203 guideline and GLP were performed with Al2(SO4)3on Danio rerio under static and semi static conditions. During the static test no pH adjustment was made and the pH decrease up 4.2. However, the study included further testing to demonstrate that toxicity was related to aluminium and not to low pH values.In this test an LC50 (96h) was found of 1 mg/L Dissolved Al or 158 mg/L AlNH4(SO4)2.12H2O at a pH of approximately 4.8-5.During the second test under semi-static, the pH was adjusted between 7.4 and 8.0. No effects were found in this test, therefore, an LC50 (96h) greater than 0.247 mg/L Dissolved Al or 1443 mg/LAlNH4(SO4)2.12H2O was determined. 

An 96h acute toxicity test of (NH4)2SO4to juvenile Salmo gairdneri and not performed according to international guidelines was reported in a peer review publication. The methodology used is scientifically acceptable and even if important information was not reported on tested concentrations the results are considered reliable. Ten fish per tank were exposed in each concentration and one control fish was realized. Tests were performed with a mean pH of 7.9; and a mean temperature of 12.4°C. Total ammonia nitrogen was measured and un-ionized ammonia concentrations were calculated. The LC50 (96h) was 0.764 mg NH3/L corresponding to 924 mg/L AlNH4(SO4)2.12H2O.

Another peer review publication reported the toxicity of the salt AlNH4(SO4)2.12H2O to the marine water fish Fundulus heteroclitus after 48h of exposure at two different salinities. The test was not performed according to international guidelines and the report is missing important information so the study was assessed as not assignable. The pH was not adjusted and is expected to have ranged between 4 to 7.6 and to have been the immediate cause of death in some test. However the methodology is scientifically acceptable and the results are sufficiently reliable to be use in a weight of evidence approach. EC50 (48h) of 3.6 mg/L and 27.5 mg/L Total Al corresponding to 60.5 and 461.4 mg/L AlNH4(SO4)2.12H2O at a salinity of 6.6 and 17 ppt, respectively have been determined. The result on the toxicity of the salt confirms the result on the dissociation products reinforcing the read across approach reliability.


The results show that the greatest sensitivity was observed on the marine water fish Fundulus heteroclitus during the 48h exposure test with AlNH4(SO4)2.12H2O. During this test, the pH ranged between 4 and 7.6 and therefore the toxicity observed seems related to aluminium. However, this result cannot be validated because the low pH might have been the immediate cause of death. On the other hand, this assumption is confirmed by the static test with Al2(SO4)3on Danio rerio where a comparable level of toxicity was found (taking into account the interspecies differences) at a pH of 4.8-5. No toxicity of aluminium is observed when the pH is maintained between 7.4 and 8.0 as shown in the semi static test on Al2(SO4)3. At the same range of pH (7.9-8), ammoniac show toxicity on Salmo gairdneri juveniles exposed to (NH4)2SO4 during 96h. However the observed acute toxicity of aluminium on fish is around 2 and 15 times higher than the acute toxicity of ammoniac related to the salt concentration independently of the pH value.