Registration Dossier

Ecotoxicological information

Long-term toxicity to fish

Currently viewing:

Administrative data

Link to relevant study record(s)

Description of key information

Aluminium nitride is susceptible to hydrolysis in aqueous compartments yielding aluminium and ammonium as degradation products with aluminium being the compound relevant for risk assessment. The lowest available NOEC value describing the long-term toxicity of aluminium towards fish is 13 µg/L. 

Key value for chemical safety assessment

EC10, LC10 or NOEC for freshwater fish:
13 µg/L

Additional information

Freeman et al. (1971) studied the toxic effects of aqueous aluminium complexes in neutral and basic media to rainbow trout fingerlings under constant flow conditions. The trout were hatched and reared to test size in well water and exposed to aluminium after acclimatization. The study was performed at various pH values and aluminium concentrations. Behaviour, growth and mortality were recorded. The lowest no-effect concentration (NOEC) was found to be 0.05 ppm (ca. 75 µg aluminium/L).

In addition, the 60-day chronic toxicity of aluminium (test material: aluminium sulphate) to early life stage of brook trout (Salvelinus fontinalis) was studied under flow-through conditions (Cleveland et al., 1989). Fertilised eggs (eye stage, 200 per treatment level) of brook trout were exposed to control, pH corrected control, and various test chemical concentrations. The test system was maintained at 12 ºC and a pH of 6.5 to 6.6. The 60-day NOEC values, based on sublethal effects (hatching success, length, DNA-RNA ratio), were 13 µg/L, respectively (dissolved Al, measured concentration).

In a 120-h acute toxicity study according to OECD 212 (Macova et al., 2010), juvenile Danio rerio were exposed to a preparation containing aluminium at nominal concentrations of 700, 750, 800, 850, 900mg/L under semi-static conditions.  The 96-h LC50was 67.5 +/- 2.8 mg Al/L.

As results obtained from tests according to OECD 212 (Fish, Short-term Toxicity Test on Embryo and Sac Fry Stages) are considered less sensitive compared to results obtained under the conditions of the FELS test (OECD 210), the results obtained by Macova et al. (2010) are not further considered in the risk assessment.

In the study by Rice et al. eggs and slevins of pink salmon (Oncorhynchus gorbuscha) were exposed to ammonia in a series of static and flow-through experiments to determine which levels of ammonia would decrease survival. Long-term tests (up to 61 days) with lower ammonia concentrations were conducted to determine effects on survival and size of fry at emergence. The value stated in the study which had an effect on the weight and length of migrant pink salmon was given to ca. 2.4 ppb. This has been recalculated to be ca. 2.24 µg/L of un-ionised ammonia. The amount of un-ionised ammonia under the given pH and temperature conditions of the study is stated as <0.1 %. This would lead to a very rough estimated amount of total (ionised) ammonia of > 2 mg/L. The possibility of ammonia stimulating emergence of immature fry was tested at various stages of development. Only for very high ammonia concentrations some immature fry was observed immediately after short-term exposure. These high concentrations are not likely to be encountered in natural or hatchery environments.

Solbe et al. (1989) examined the toxicity of ammonia (NH3) to early life stages of rainbow trout (Salmo gairdneri) in hard fresh water in a flow-through experiment with a total duration of 73 days. Two sets of experiments were performed, one where the exposure to ammonia (added in form of ammonium chloride) started within 24 h of fertilization (Test A) and one set where the exposure started 24 days after examination (Test B). Mortality > 70% was recorded at a concentration of 0.027 mg/L NH3 when exposure began within 24h of fertilization. In contrast, only 40 % mortality was noted at 0.27 mg/L NH3 when exposure did not start until the eyed-egg stage (c. 24 days). EC10 values were 0.14 mg total ammonia (NH3 +NH4) per litre and 0.19 mg total ammonia/L in Test A and Test B, respectively.

The NH3 effect values are not used as basis for the risk assessment because this would overestimate the toxicity of ammonium in aqueous solution drastically. If ammonia comes in contact with water it will form an equilibrium between the ionised and the unionised form which is highly on the side of the ionised form, slightly depending on temperature and pH. Nevertheless, the amount of ionised ammonium will be much higher compared to the NH3-form. For this reason the environmental risk characterisation will be performed on the basis of the lowest NOEC value identified for aluminium. The contribution of ammonia to the toxicity against fish, which is significantly lower, can be ignored.