Sodium hypochlorite

Short-term toxicity to aquatic invertebrates

Description of key information

Freshwater:
Initially available data, as reviewed in the European RAR and other dossiers (such as for biocide active substance) were of limited quality. Consequently, in the frame  preparation of this registration dossier a new state of the art study with Daphnia magna revealed a 48 h EC50 value of 141 µg/l. More recently, an additional study with Ceriodaphnia dubia, carried out in equivalent conditions gave an EC50 = 35 µg/l. Both are considered as relevant for Classification and Labelling and for Risk Assessment.
Marine water:
Different studies are  available. A Crassostrea virginica larvae acute test was determined as key, study revealing a EC50 (48 h) of 0.026 mg TRO/L.

Key value for chemical safety assessment

Fresh water invertebrates

Fresh water invertebrates

Effect concentration:

0.035 mg/L

Marine water invertebrates

Marine water invertebrates

Effect concentration:

0.026 mg/L

Additional information

Freshwater:

Several invertebrates belonging to different phyla (arthropods, mollusks, annelids) have been tested for sensitivity to sodium hypochlorite.

48–hr-acute toxicity of sodium hypochlorite to Daphnia magna was studied under flow through conditions in a new state of the art study ( Gallagher at al., 2009). Daphnids were exposed to control and test chemical at nominal concentration of 12.5, 25, 50, 100, 200 and 400 µg active chlorine/L for 48 hr. Mortality/immobilization and sublethal effects were observed daily. The 48 – hour EC50 was 141 µg active chlorine/L. The 48 – hr NOEC based on mortality/immobilization was 50 µg active chlorine /L. Based on the results of this study, sodium hypochlorite has to be classified as acute toxicity class 1 to daphnia in accordance with the classification system CLP (GHS). This study is classified as acceptable and satisfies the guideline requirements for an acute toxicity study with freshwater invertebrates.

Taylor (1993) tested the acute toxicity of various forms of free and combined chlorine to Ceriodaphnia dubia in standard 24h toxicity tests, carried out under static and flow through conditions. Sodium hypochlorite was tested at pH 7 for HOCl (70% HOCl and 30% OCl-) and pH 8 for OCl- (80% OCl and 20% HOCl). In static tests the decay of free chlorine was very rapid (1 minute and 7 hours in tests with or without food, respectively) and the results were not considered valid. Flow-through tests (without food) were carried out to maintain a constant concentration over the exposure time. The toxicity of free chlorine in these tests was much higher: EC50 -24h = 5 and 6 μg/l for HOCl and OCl -, respectively. These data were judged valid with restriction (rated 2) by RAR Rapporteur Member State because the test concentrations were calculated from measured chlorine concentration of the stock solution and dilution ratios, the number of concentrations/replicates are not specified, the performance of the controls not mentioned, and the 24h LC50s determined by graphical interpolation only.

Considering the uncertainty about the quality of this study, and concerns of some members states about differences of sensitivity between the 2 daphnia species, a litterature review showed that the available information tends to conclude that if differences can be observed between the 2 species, depending on the toxicant, there is no general rule about relative sensitivity of the two species. It has therefore been decided by the consortium of registrants to commission a new study on Ceriodaphnia dubia, in conditions strictly similar (OECD202, flow-through conditions, GLP) to those applied in the Gallagher (2009) study on Daphnia magna. The results (Gallagher, 2011) show an EC50 -48h = 35 µg/l and NOEC = 25 µg/l. This shows on the one hand that the Taylor study is indeed an outlier, and on the other hand confirms that the range of acute toxicities to be used for classification and labelling purpose is 10 -100 µg/l.

Marine water:

The acute toxicity of sodium hypochlorite to saltwater invertebrates has been widely studied with many organisms, including rotifers, mollusks, crustaceans and worms, but data adequate for the assessment are available only for a few classes.

Thatcher (1978) tested seven species, including shrimps, mysids, amphipods andcrabs, in flow through apparatus where animals were simultaneously exposed to chlorinated unfiltered seawater and 5°C thermal stress to mimic a common scenario in the real environment. The most sensitive species was the shrimp Pandalus goniurus for which a 96h LC50= 90 μg/l was calculated pooling data from different tests. These data have been rated 2 because of data pooling and, above all, because the additional thermal shock might have influenced the sensitivity to sodium hypochlorite.

Very short-term exposure to pulse chlorination, with and without thermal stress, has been evaluated by Capuzzo and coworkers (1976, 1979a,b), who conducted flowthrough bioassays with Brachionus plicatilis, Acartia tonsa, and larvae of Crassostrea virginica with exposure periods of 30-60 minutes at various temperatures. In bioassays performed at acclimation temperatures, lethality (measured after 48 h after exposure) was LC50= 820, 180 and 120 μg/l, respectively. For all species tested,except the copepods, the thermal stress showed a synergistic effect with residual chlorine concentration, lowering LC50 values down to 10 μg/l (ΔT=7.5°C) for B. plicatilis and 80μg/l (deltaT=5°C) for C. virginica.

Four invertebrate species were tested by Roberts et al. (1975) using estuarine river water, chlorinated by addition of Ca(OCl)2 in flow through or static systems. When static systems were used the TRC concentrations were kept as stable as possible by constant addition of the chemical and average values were used to calculate the endpoints. The lowest lethal concentrations to be used as supportive information are extrapolated values for Mercenaria mercenaria larvae (48h TL50=1 μg/l TRC, in the static test) and Crassostrea virginica juveniles (96hEC50(shell deposition)=23 μg/l TRC, in the flow through test). The authors report that when oyster larvae were exposed intermittently (manual additions at 6-8h intervals) the 48h LT50 was 0.11 mg/l, i.e. two orders of magnitude higher.

In a study from Raikow (2007), the 3 d-acute toxicity of Sodium Hypochlorite to Artemia sp. (brine shrimp) was studied under static conditions. Eggs of Artemia sp. were exposed to control and test chemical at nominal concentration between 1 and 300 mg/L for 3 days. Mortality/immobilization and sublethal effects of hatches were observed daily. The 3 – day LC50 was 34.6 mg/L. The 3 – day LC90 was 86.5 mg/L. This study is classified as acceptable as supporting study.

In later experiments Roberts and Gleeson (1978) observed a similar toxicity in oyster and copepods continuously exposed for 48 hours to Ca (OCl)2 in flowing estuarine river water. For Crassostrea virginica larvae the 48EC50 was 26 μg/l and for adults of Acartia tonsa the lowest 48hLC50 was 29 μg/l as residual chlorine measured amperometrically. In the absence of other, more reliable studies with the relevant chemical species (sodium hypochlorite) it was decided to use these data, because they provide valuable information for the toxicity assessment of chlorinated natural water, and assigned these data the rate ‘valid with restriction’ because of the different parent compound.

The range of acute toxicity of chlorinated seawater to invertebrates appears therefore to span almost two orders of magnitude. The relevant EC50 for Classification and Labelling and Risk Assessment is 26 µg/L (Roberts and Gleeson, 1978).