Chlorine

Short-term toxicity to fish

Description of key information

Read-across from sodium hypochlorite (justification see IUCLID5 section 6.1 or CSR section 7.1.1): 
Freshwater: Acute fish test performed according to a standard test design are not found in the literature. Most available studies from the public domain were designed to mimic intermittent exposure. The rainbow trout (S. gairdneri) and channel catfish (I. punctatus) were shown to be the species which are most sensitive to hypochlorite in freshwater.
Marine Water: In short term tests,  fish shows a similar sensitivity: LC50 = 32 µg/L (most sensitive species in salt water: Coho salmon).

Key value for chemical safety assessment

Fresh water fish

Fresh water fish

Effect concentration:

0.06 mg/L

Marine water fish

Marine water fish

Effect concentration:

0.032 mg/L

Additional information

Read-across from sodium hypochlorite (justification see IUCLID5 section 6.1 or CSR section 7.1.1):

Freshwater:

Concerning the short-term toxicity to fish, the collected and evaluated literature refers only to exposures of very short duration, often on an intermittent regime. Hence a 96h LC50 adequate for the assessment cannot be retrieved but the LC50 values provided are useful as supportive information.

In the late 70's a number of studies in which fish were exposed to a variety of exposure regimes (one or more short pulse doses for 1 or more days), were carried out to simulate intermittent exposure occurring in natural waters receiving power plant cooling effluents. In two studies (Bass et al., 1977 and Heath, 1978) six fish species (Salmo gairdneri, Oncorhynchus kisutch, Ictalurus punctatus, Notemigonus crysoleucas, Lepomis macrochirus, and Cyprinus carpio) were exposed for different time intervals (ranging from 24 to168 hours) to 40' calcium hypochlorite pulses (3 per day) in flow through tap water at different temperatures. S. gairdneri and I. punctatus were the most sensitive species in both studies, showing, respectively, an 120h LC50=50 μgTRC/l (at 12 °C) and an 120h. LC50= 33 μgTRC/l (at 24°C). At 96h, the LC50 for the trout was 60 μgTRC/l (at 5°C) and 64 μgTRC /l for the channel catfish (at 24°C).

In the paper of Brooks and Seegert (1977) the effects of single (30 min) and multiple (5 min exposure at 3 hours interval) exposure to sodium hypochlorite in tap water with and without the addition of a thermal stress on two species of fish - Perca flavescens and Salmo gairdneri (juveniles) - have been described. A single 30' dose was more toxic than the triple exposure. Generally, S. gairdneri was more sensitive than Perca flavescens; after 24h recovery, the LC50 was 990 μg/l. With the addition of a thermal stress, the median lethal concentration lowered to 430 μg/l.

Seegert and Brooks (1978) exposed for 30 min. four fish species - Oncorhyncus kisutch, Alosa pseudoharengus, Notropis hudsonius and Osmerus mordax - to sodium hypochlorite to various temperatures. Coho salmon (Oncorhyncus kisutch) showed a sensitivity comparable with that of rainbow smelt (Osmerus mordax) and both were more sensitive than the other two species. As observed in other studies, the LC 50 values showed an inverse relationship with test temperature and a typical species dependent variability; at 10°C the LC50s, measured after 48 hrs, were in the range of 1.26 - 2.41 mg/l.

In a study report of Buccafusco (1978) the 96 -hour LC50 for Lepomis macrochirus was dertemined to be 0.58 mg/L. For Oncorynchus mykiss Stiefel (1978) reported an LC50 of 0.2 mg/L after 96 -hour of exposure. Both studies were performed according to the U.S. EPA Guideline (1975) "Methods for acute toxicity tests with fish, macroinvertehrates, and amphibians" and meets scientific requirements. However, analytical dose verification were not performed.

The toxicity of hypochlorite following intermittent exposure was studied by Wilde et al. (1983a,b) in two studies conducted at two different temperatures (27.7 and 21.1 °C). Acute toxicity tests (96h) have been carried out on juvenile and adult fathead minnows (Pimephales promelas) and young-of-the-year bluegills (Lepomis macrochirus) in a mobile laboratory. For each test, sodium hypochlorite was dosed intermittently (for 1 hour at the beginning of each 24h interval) into a flow-through system using water from a reactor cooling reservoir as dilution water. The six test concentrations, as TRC and FAC, were measured at 10-minute intervals for the 2 hr per day that toxicant residuals were measurable. In the two studies, in the chamber receiving 100% biocide solution, FRC (the FAC measured after the reaction) contributed 50-70 and 83.2% to TRC during the 30 minute periods of maximum exposure. The lowest 96h LC50s, determined as intermittent exposure mean, was calculated as 0.08 mg/l for juvenile fathead minnow exposed to 27.7°C.

Tsai et al. (1990) exposed three species of fish (Cyprinus carpio, Dorosoma petenense and Gambusia affinis) of different age classes to sodium hypochlorite in a flow - through exposure system for one hour and evaluated the toxic effects 48h after exposure (fish) and 3 days after fertilization (eggs). C. carpio fertilized eggs were far less sensitive than later development stages (one-d-old eggs LC50 = 140 mg/l; 1 week-old prolarvae LC50 = 0.33 mg/l) but hardened fertilized eggs were much less sensitive than 1-h soft eggs (10-h-old eggs LC50 = 158 mg/l; 1-h-old (soft) eggs LC50 = 54). Sensitivity of D. petenense was slightly lower than that of C. carpio: LC50=0.260 mg/l for prolarvae. For G. affinis the sensitivity varied over the first year of life: the LC50 was 0.61 mg/l and 1.28 mg/l for 1 week-old postlarvae and 1 year-old young, respectively.

A toxicity testing system was developed by Mattice et al. (1981a) to test the effects of sodium hypochlorite, dosed for 30 and 60 minutes, to the fish Gambusia affinis. The LC50, determined after 48 hours, was 1.59 and 0.84 mg/l, respectively.

In a 24-h acute toxicity study (de pavia magalhaes, 2007), Zebrafish (Danio rerio) were exposed to Sodium Hypochlorite at nominal concentrations of 0 22.8, 28.5,34.2, 39.0, 45.6, 51.3, 57.0 mg/L under static conditions. The 24-h LC50 was 48 mg/L. Sublethal effects of (reduced swimming activity) were observed in the groups exposed to of 14.4 and 19.2 mg/L of sodium hypochlorite. This toxicity study is classified as acceptability and satisfies the guideline requirement (except duration of exposure) for an acute fish toxicity study.

The reliability of the above mentioned studies is good because they are all well described, but they are judged not adequate to retrieve an LC50 useful for the assessment of short-term toxicity because they would underestimate the toxicity from a continuous exposure (e.g. 96h LC50). The effect of the exposure regime on toxicity has also been observed in invertebrates: oyster larvae survival under intermittent chlorination was much higher than continuous chlorination (Roberts et al 1975). Nevertheless, these data are retrieved as useful supportive information because they mimic the short-term exposures expected in some scenarios and give an idea of the potential of sodium hypochlorite to produce acute effects in fish.

Marine water:

The short-term toxicity of chlorinated salt water has been tested on many estuarine and marine fish species in bioassays conducted at different exposure regimes, from a single pulse dose of few minutes to a continuous 96 hours standard exposure. Data valid for the assessment were retrieved from the papers by Bellanca and Bailey

(1977), Thatcher (1978) and Roberts et al. (1975).

Bellanca and Bailey (1977) evaluated the short-term toxicity of chlorine to the estuarine fish Leiostomus xanthurus (ocean spot) in a flow through laboratory experiment, using a continuous flow serial diluter fed with river water. The authors calculated a 96h-TLm (equivalent to an LC50)= 0.090 mg/l of TRC, which consisted

principally of free chlorine. This data is rated 1.

Thatcher (1978) conducted many laboratory flow-through bioassays on 8 species of estuarine and marine fish, belonging to different families including salmonidae,clupeidae and percidae. Since the main purpose of the study was to investigate the impact of chlorinated effluents from power plants, fish were simultaneously exposed to sodium hypochlorite and to a 5°C thermal stress. The 96h LC50 ranged from 0.032 mg/l (as TRO), for the most sensitive species (Oncorhynchus kisutch, coho salmon), to 0.167 mg/l (Gasterosteous aculeatus). These data were considered relevant for the assessment because heat is usually associated to chlorine in power plants effluents, but they were rated 2 because the authors report that in a previous study the addition of thermal stress resulted in a toxicity higher than chlorine alone and, moreover, LC50 was calculated pooling data from different tests.

Roberts et al. (1975) tested the toxicity of continuous exposure of chlorinated river water (salinity 20%o) to three estuarine fish species. The lowest LC50 was calculated for Menidia menidia: 96h LC50= 37 μg/l (as TRC). This data was judged valid with restriction because Ca(OCl2) was applied as chlorine source and some information on the experimental procedure is lacking.

Middaugh et al. (1977) tested the toxicity of chlorinated brackish pond water to earlylife stages of Morone saxatilis in a flow-throug test. Only data relative to eggs

hatchability could be retrieved. A rough estimate of the 48h LC50 = 8 μg/l TRC was calculated by us using the authors' raw data relative to percentages of hatched eggs per test concentration. This data can be used as indicative information of eggs sensitivity.