Lithium sulphate

Endpoint summary

Additional information

Short-term toxicity to fish

key study

A short-term toxicity test in fish was not available for lithium sulfate or its monohydrate. Therefore read-across was conducted with a structural similar compound, lithium carbonate.

A static freshwater toxicity test was conducted to determine the acute toxicity of lithium carbonate to rainbow trout, Oncorhynchus mykiss according to OECD Guideline 203. Mean measured concentration of lithium carbonate ranged from 4.99 to 77.7 mg /L and from 96 to 100 % of nominal. All test solutions appeared clear and colourless and concentrations remained stable throughout the test. The pH of the test solutions was affected by the presence of lithium carbonate (i.e. the pH increased as the test substance concentrations increased). The pH values of all test solutions ranged from 8.7 to 10.4 at test initiation and from 6.7 to 9.8 for the remainder of the test. Mortality of the rainbow trout exposed for 96 hours to lithium carbonate ranged from 0 % at test concentrations <= 19.1 mg /L to 100 % at 77.7 mg /L. No mortality occurred in the dilution water control. The 96-hour LC50 was 30.3 mg lithium carbonate/L with 95 % confidence limits of 19.1 and 38.9 mg/L. The NOEC of 19.1 mg lithium carbonate/L was based on a lack of significant mortality and the absence of sublethal effects at this and all lower test concentrations. Based on these data, the calculated LC50 for lithium ion was 5.69 mg/L and the NOEC was 3.59 mg/L. (Toxikon, 1997)

Based on read-across approach, the calculated 96-hour LC50 for lithium sulfate and lithium sulfate monohydrate is 45 mg/L and 52 mg/L, respectively.

supporting data

A static freshwater toxicity test was conducted to determine the acute toxicity of sodium sulfate in fathead minnow, Pimephales promelas. The fishes were exposed for 96 hours. A LC50 value of 7960 mg/L was determined. (Mount et al., 1997)

Based on read-across approach, the LC50 values for lithium sulfate and its monohydrate are 6161 mg/L and 7170 mg/L, respectively.

Long-term toxicity to fish

A long-term toxicity test in fish with lithium sulfate or its monohydrate was not available. Consequently, read-across was applied using study results obtained from lithium hydroxide monohydrate.

The purpose of this study was to evaluate the chronic toxicity of the test item lithium hydroxide monohydrate to early life stages (embryo, larvae and juveniles) of fish.

For this purpose, eggs were exposed in a semi-static test to aqueous test media containing the test item for 34 days at a range of concentrations under defined conditions. Lethal and sub-lethal effects were assessed and compared with control values to determine the lowest observed effect concentration (LOEC) and hence the no observed effect concentration (NOEC) for each response assessed.

The test was considered to be valid as assay acceptance criteria were fulfilled.

Based on the preliminary chronic fish toxicity test result the following nominal test concentrations were selected for the early life stage toxicity test: 1.8 mg/L, 3.0 mg/L, 7.2 mg/L, 15.0 mg/L and 21.1 mg/L.

Test item concentrations were analyzed in the test solutions by flame photometry. Lithium hydroxide monohydrate concentration in the test solutions was determined at 5 renewal periods by flame photometry. The measured concentrations of the lithium hydroxide monohydrate varied between 102 % and 151 % of the nominal concentration during the test. Mean measured concentrations were calculated for each parallel test chambers and for each treatment concentrations. The measured concentrations of each parallel treatment were within the range of ± 20 % of the mean measured values meeting the validity criteria. Since concentrations of the test substance were satisfactorily maintained within ± 20 % of the mean measured values during the test in all treatment groups, all biological results are reported on the basis of the mean measured concentrations: 2.43 mg/L, 3.82 mg/L, 8.60 mg/L, 17.35 mg/L and 24.35 mg/L.

79 to 82 eggs were tested per test concentrations and the control in two parallels in the test groups and in the control group (approximately 40-40 eggs per each parallel treatment). Based on microscopic observation embryos were at approximately 64 to 128 cell stages at start of treatment.

Environmental parameters (water temperature, pH, LDO) were monitored during the test. There were no deviations from the defined ranges. Animals were fed from elimination of the yolk sac (free-feeding stage) to the end of the test with appropriate type and amount of food ad libitum, and thus starvation did not have any effect on the results obtained.

Cumulative mortality (observed from Day 0 to the end of the test on Day 34), mortality observed at embryonic/eleutheroembryonic (i.e. early larval) stage (from Day 0 to the end of hatching on Day 5) and mortality observed at larval/juvenile stage (from Day 6 - free feeding stage - to the end of the test on Day 34) were statistically evaluated. Statistically significant (p < 0.01) cumulative mortality compared to the control was observed at a treatment concentration of 24.35 mg/L (21.1 mg/L nominal). No significant lethal effect was observed until embryonic/eleutheroembryonic stage. Since mortality observed from Day 6 to Day 34 was statistically significant (p < 0.01) it is considered that the test item influenced the survival of the test organism at larval/juvenile stage rather than at embryonic stage. No statistically significant lethal effect was observed in other test concentrations (neither during embryonic nor larval/juvenile stage).

No significant effect on hatching of the larvae was observed at any concentrations tested. No significant differences were observed either in starting or duration of hatching. Numbers of larvae hatched daily were statistically evaluated on Day 4 and Day 5 and no significant differences were observed.

At the end of the test, group body weights were measured. No significant differences were observed.

Body length was measured for each animal individually and the data were statistically evaluated. A statistically significant (p < 0.05), but biologically not relevant difference was observed at a test concentration of 8.61 mg/L (7.2 mg/L nominal).

Other sub-lethal effects were monitored with limited extent: numbers of embryos/larvae/juveniles with deformities or abnormal behavior were recorded. No obvious test item related changes in the behavior of fish could be observed.

Under the conditions of this early life stage toxicity study, the test item lithium hydroxide monohydrate had a significant lethal effect on early life stages of Zebrafish (Danio rerio). The observed effect was associated with larval/juvenile stages, but no significant effect was observed during the embryonic stage.

The following endpoints (34 days LOEC and NOEC) were calculated for lithium hydroxide monohydrate:

The 34 d LOEC 24.35 mg test item/L

The 34 d NOEC 17.35 mg test item/L (Toxi-Coop, 2012)

Based on read-across approach, the calculated LOEC and NOEC values for lithium sulfate anhydrous are 31.2 and 22.73 mg/L. The LOEC and NOEC values for its monohydrate are 37.12 and 26.45 mg/L, respectively.

Short-term toxicity to aquatic invertebrates

A short-term toxicity test in Daphnia magna with lithium sulfate or its monohydrate is not available. Consequently, read-across was applied using study results obtained from a supporting substance, lithium carbonate.

A static freshwater toxicity test was conducted to determine the acute toxicity of lithium carbonate to the water flea, Daphnia magna according to OECD Guideline 202. Mean measured concentrations of lithium carbonate ranged from 4.76 to 82.8 mg/L and from 95 to 109 % of nominal. All test solutions appeared clear and colourless and concentrations remained stable throughout the test. Mortality of the water flea exposed for 48 hours to lithium carbonate ranged from 0 % at test concentrations lower than 20.0 mg/L to 100 % at 82.8 mg/L. Control mortality was 0 %. The 48-hour EC50 was 33.2 mg lithium carbonate/L with 95 % confidence limits of 20.0 and 43.7 mg/L. The NOEC was 20.0 mg lithium carbonate/L, based on a lack of significant mortality and sublethal effects observed at this and all lower test concentrations. Based on these data, the calculated EC50 for lithium ion was 6.24 mg/L and the NOEC was 3.76 mg/L. (Toxikon, 1997)

Based on read-across approach, the calculated 48-hour EC50 of lithium sulfate and lithium sulfate monohydrate are 49 mg/L and 57 mg/L, respectively.

Long term toxicity to aquatic invertebrates

Long-term toxicity tests in daphnia with lithium sulfate or its monohydrate is not available. Consequently, read-across was applied using study results obtained from a supporting substance, lithium.

The purpose of the study was to evaluate the influence of the test item lithium on the reproductive output of Daphnia magna in a semi-static test system according to OECD Guideline 211. Young female Daphnia (the parent animals) aged less than 24 hours at the start of the test were exposed to aqueous test media containing the test item for 21 days at a range of concentrations. The nominal test item concentrations were 0.50, 0.75, 1.13, 1.70, 2.53, 3.80 and 5.70 mg lithium/L. The performed parallel running analytical determinations confirmed that the test item concentrations examined (lowest and highest test concentrations) remained within the range of ± 20 % of the nominal and of the initial concentrations (varied between 98 and 117 per cent of the nominal concentration); thus, all results were based on the nominal test item concentrations. In the three highest tested concentrations (2.53, 3.80 and 5.70 mg/L) all parent animals died by the 13th day of the test without producing any offspring. Therefore the results of these concentrations were excluded from the data analysis related to the reproductive output. In the control group two parent animals (20 %) died during the test which was within the acceptable validity criteria. In the concentration range of 0.50 – 1.70 mg/L mortality of parent animals was not observed during the experiment. The reproduction was not reduced significantly in the concentration range of 0.50 – 1.70 mg/L compared to the untreated control group. During the evaluation of the body length of parent animals at the end of the test, a statistically significant difference was not observed in the remaining living parent daphnids (in the concentration range of 0.50 – 1.70 mg/L) compared to the control group. Aborted broods, presence of male neonates or ephippia were not noticed during the test. Accordingly, the 21-day NOEC value related to reproduction was determined to be 1.70 mg/L and the LOEC value as 2.53 mg/L. The obtained results were not sufficient for an exact EC50 value estimation. The 21-day EC50 was determined to be higher than 1.70 mg/L. (Toxi-Coop, 2012)

Based on a read-across approach, the calculated NOEC and LOEC values for lithium sulfate are 13.5 and 20.0 mg/L. For lithium sulfate monohydrate, the NOEC and LOEC values are 15.7 mg/L and 23.3 mg/L respectively.

Toxicity to aquatic algae and cyanobacteria

A toxicity study in aquatic algae with lithium sulfate and its monohydrate was not available. Consequently, read-across was applied using study results obtained from a supporting substance, lithium carbonate.

The effect of lithium carbonate on the growth of an algal species Desmodesmus subspicatus over a 72-hour static exposure period was assessed according to OECD guideline 201. The test solutions were prepared in a dilution series from a stock solution of 100 mg/L of the test item in culture medium for the range-finding test and from a stock solution of 400 mg/L of the test item in culture medium for the definitive test, respectively. The stock solutions and the corresponding dilutions were prepared using a mixing device (Ultra-Turrax, Janke & Kunkel; 8000 rpm, 2 minutes). The test solutions were clear at all test concentrations. The controls were kept in culture medium. Vessels including culture medium (controls) and test solutions were kept under the same conditions as the test vessels for concentration analysis. Concentration analysis was performed at 0 h and 72 h. The Alga-test was carried without adjustment of the pH.AES-analysis confirmed that the test solutions were correctly dosed, i.e., the recoveries were within 92.4 to 103 % of the nominal concentrations at study start. At the end of the exposure, i.e., after 72 hours, the recoveries ranged from 95.9 to 103 % of the nominal concentrations demonstrating that the lithium concentrations were stable throughout the exposure period. Consequently, the results of the definitive test were based on nominal concentrations.In this 72-h algal growth inhibition test with Desmodesmus subspicatus the 72-h EC50 based on growth rate was determined as greater than 400 mg/L for lithium carbonate.

The overall NOEC was determined to be 50 mg/L for lithium carbonate (the calculated NOEC for lithium ion: 9.39 mg/L) . The results are based on the nominal concentrations. (Steinbeis, 2010)

Based on read-across approach, the 72-hour EC10, EC50 and overall NOEC of lithium sulfate anhydrous are 134, greater than 596 and 75 mg/L, respectively. Also based on read-across, the 72-hour EC10, EC50 and overall NOEC values for lithium sulfate monohydrate are 156, 692 and 87 mg/L, respectively.

Toxicity to microorganisms

A toxicity study with microorganisms for lithium sulfate and its monohydrate were not available. Therefore read-across was applied using study results obtained from a supporting substance, lithium hydroxide.

The influence of the test item lithium hydroxide on the activity of activated sludge by measuring the respiration rate was evaluated according to OECD Guideline 209 and EU method C.11. The respiration rate (oxygen consumption) of an aerobic activated sludge fed with a standard amount of synthetic sewage was measured in the presence of various concentrations of the test item after an incubation period of 3 hours. The inhibitory effect of the test item at the particular concentrations was expressed as percentage of the mean respiration rate of two controls. Following test concentrations were used: 10, 32, 100, 320 and 1000 mg lithium hydroxide/L; 3.2, 10 and 32 mg 3,5-Dichlorophenol/L and two inoculum controls. In comparison to the inoculum controls the respiration rate of the activated sludge was inhibited between –1.8% and 98.2 % up to the highest nominal test concentration of 1000 mg/L. Concentrations exceeding 1000 mg/L nominal were not tested. The 3-hour EC 50 for the positive control 3,5-Dichlorophenol, which was tested in the same way as the test item, was found to be 7.5 mg/L and is within the range of 5 – 30 mg/L recommended by the test guidelines; thus, confirming suitability of the activated sludge.

The 3-hour EC20, EC50, and EC80 values for the test substance lithium hydroxide in the Activated Sludge Respiration Inhibition Test were 114.3, 180.8, and 286.1 mg/L (based on measured inhibition rates), respectively. The EC10 value was calculated by linear regression to be 79.2 mg/L for lithium hydroxide anhydrous and 22.95 mg/L for lithium ion. (LAB, 2004)

Based on read-across approach, the 3-hour EC10, EC20, EC50, and EC80 values for the test substance lithium sulfate and lithium sulfate monohydrate in the Activated Sludge Respiration Inhibition Test are 182, 262, 415, 657 mg/L and 212, 305, 483, 764 mg/L, respectively.

Supporting data regarding sodium sulfate studies and publications on sulfate concentration of the aquatic environment

The supporting data provided in this section refers to read-across from sodium sulfate and sulfate ions as such. Literature citations from short-term toxicity tests with sodium sulfate showed that the lithium ion and not sulfate is the ecotoxicological relevant moiety. The table below summarises the LC50 values for lithium sulfate and its monohydrate obtained from these publications by calculation (read-across).

 

Aquatic Assay	LC50	Species
short-term toxicity fish	6161 mg/L (96-h LC50, Li2SO4)	Pimephales promelas
short-term toxicity fish	7170 mg/L (96-h LC50, Li2SO4*H2O)	Pimephales promelas
Short-term toxicity daphnia	1775 mg/L (48-h LC50, Li2SO4)	Daphnia magna
Short-term toxicity daphnia	2066 mg/L (48-h LC50, Li2SO4*H2O)	Daphnia magna

 

The read-across data received from sodium sulfate are much higher than those received from lithium carbonate (factor of 10 -100).

Sulfate content in drinking water is regulated (limit value 250 mg/L; geogenically caused values up to 500 mg/L are not considered). High sulfate values in drinking water may cause gastrointestinal effects in humans or may affect / destroy water lines (by spiky crystals of calcium sulfate).

With respect to chronic toxicity, the limit concentration of 250 mg/L for sulfate ions as laid down in  the drinking water Directive 98/83/EC was used for read-across, resulting in 286 mg Li2SO4/L and 333 mg Li2SO4*H2O/L. As can be seen after read-across from long-term toxicity studies conducted with lithium hydroxide, lithium carbonate and lithium as such, the NOEC values for lithium sulfate are all below those derived from the limit concentration:

NOEC (fish) = 22.73 mg Li2SO4/L

NOEC (fish) = 26.45 mg Li2SO4*H2O/L

NOEC (daphnia) = 13.5 mg Li2SO4/L

NOEC (daphnia) = 23.3 mg Li2SO4*H2O/L

NOEC (algae) = 75 mg Li2SO4/L

NOEC (algae) = 87 mg Li2SO4*H2O/L

Thus, also based on this information it can be assumed that lithium is the more toxicological relevant moiety (than sulfate) with respect to long-term exposure.