Reaction mass of aluminium chloride, iron dichloride, magnesium chloride and manganese dichloride

Ecotoxicological Summary

Hazard for aquatic organisms

Freshwater

Hazard assessment conclusion:

PNEC aqua (freshwater)

PNEC value:

114 µg/L

Assessment factor:

5

Extrapolation method:

sensitivity distribution

PNEC freshwater (intermittent releases):

400 µg/L

Marine water

Hazard assessment conclusion:

PNEC aqua (marine water)

PNEC value:

57 µg/L

Assessment factor:

10

Extrapolation method:

sensitivity distribution

STP

Hazard assessment conclusion:

PNEC STP

PNEC value:

737 mg/L

Assessment factor:

1

Extrapolation method:

assessment factor

Sediment (freshwater)

Hazard assessment conclusion:

PNEC sediment (freshwater)

PNEC value:

18.07 mg/kg sediment dw

Extrapolation method:

equilibrium partitioning method

Sediment (marine water)

Hazard assessment conclusion:

PNEC sediment (marine water)

PNEC value:

9.03 mg/kg sediment dw

Extrapolation method:

equilibrium partitioning method

Hazard for air

Air

Hazard assessment conclusion:

no hazard identified

Hazard for terrestrial organisms

Soil

Hazard assessment conclusion:

no exposure of soil expected

Hazard for predators

Secondary poisoning

Hazard assessment conclusion:

no potential for bioaccumulation

Additional information

General conceptual considerations

The standard risk assessment approaches are almost not appropriate as

1. the submission substance is contains solely naturally occurring chemical entities and
2. most of the relevant ones are essential for biota. The anions and the bioessential (ICMM 2007) magnesium are not discussed as it is obvious that the handling a use of the submission substance will not alter the natural concentration in a relevant magnitude.

The submission is regarded as a multi-constituent substance. The main metal, i.e. iron contributes less than 80% of the total metal molarity. The other main components, i.e. manganese, magnesium, and aluminium contribute 10% or more. About 1 to 2% falls upon vanadium, which is beneficial but not essential in animals (ICMM 2007), titanium, the bioessential calcium, chromium III, and zirconium. Les than 1% of the total metal is represented by nickel, whose role is sensitisation, is discussed in the respective chapter, and niobium. Only traces far below 1% of lead, and cadmium are analytically detectable. As a result of literature screening and due to the lack of particular ecotoxicity only iron and the metals above 9% were regarded in this section. In consequence the ecotoxicological endpoints were assessed on the basis of iron and the three main metals are discussed with regard to particular effects if relevant.

The assessment bases on OCEE / NRA considerations (Optimal Concentration Range for Essential Elements / No Risk Area, ICMM 2007) for the essential metals iron, as the main constituent, and manganese. No standard PNEC can be derived for them as their absence would be chronically lethal.

In consequence the additional load will be assessed following the added risk approach or total risk approach depending on the significance of the additional load compared to the natural background concentrations according to ECHA (2008, Guidance on information requirements and chemical safety assessment, Appendix R.7.13-2). A PNECadd (additional Predicted No Effect Concentration), i.e. the difference between the upper boundary of the NRA and the ambient natural background concentrations or a PNECtotal (total Predicted No Effect Concentration) will be the measure for the environmental safety assessment in a tiered approach, which considers where required bioavailability and acclimation/adaption.

As aluminium is not an essential element, no OCEE / NRA discussion is applicable. Its impact on the environment is to be assessed solely with regard to the potential to alter the environmentally fraction, i.e. where required by derivation of PNEC in the above mentioned tiered approach and evaluation on the basis of potential environmental risks (Risk Characterisation Ratio, RCR = PEC/PNEC).

• ICMM International Concil of Mining and Metals (2007). MERAG: Metals Environmental Risk Assessment Guidance. Self-published London, UK. ISBN: 978-0-9553591-2-5. 80 p

PNEC normalisation

As the threshold levels is based on the total dry mass of the submission item. The derived PNEC of the submission item is thus derived by assuming the threshold level of the most toxic constituent to be equal to the threshold level of the submission item dry matter. This consideration constitutes a significant additional safety margin as equalisation includes non-toxic compounds as chloride and magnesium. The dry matter content for the Chemical Risk Assessment is set to 35% mass.

Equilibrium Partitioning Method EPM

According to ECHA (2008, Guidance on information requirements and chemical safety assessment Chapter R.7c: Endpoint specific guidance, 235 p) Table R.7.11-2, p 131, the soil hazard category of the submission item was determined. High adsorption is to be considered as the constituents are ionisable. The acute toxicity of iron and aluminium is considered not relevant as discussed below in this section. The only acute toxicity value for manganese < 1 mg/L is reported from Reimer (1999) to be a 48 h LC50 for daphnids of 0.8 mg Mn/L at 25 mg CaCO3/L. Depending on the calcium carbonate concentration the value increased to 76.3 mg Mn/L. As the medium according to OECD TGD 202 (2004) foresees 293 mg CaCl2·2H2O/L, i.e. a concentration of 2 mmol Ca/L, which corresponds to 201 mg CaCO3/L, it is concluded that under standard test conditions no value < 1 mg/L exist for manganese as the most toxic component to the aquatic life. Accordingly the submission item, whereof manganese contributes to only 9.9 mol% of the metal content, is considered clearly not very toxic to aquatic organisms.

In conclusion the submission item is to be placed in Soil Hazard Category 3 and EPM according to ECB (2003) can be applied.

The related confirmatory long-term testing is considered obsolete as the natural background concentration of manganese in soils is 550 mg/kg soil d.w. or 1260 mg/kg soil d.w. (see discussion on environmental fate and pathways).

• ECB European Chemicals Bureau (2003) Technical Guidance Document in support of Commission Directive 93/67/EEC on Risk Assessment for new notified substances, Commission Regulation (EC) No 1488/94 on Risk Assessment for existing substances and Directive 98/8/EC of the European Parliament and of the Council concerning the placing of biocidal products on the market, Part II, 328 p

Iron

Results from toxicity tests may be reported relative to concentrations of iron or the iron compound that was tested. To show toxicity values across this iron category in an equivalent manner the results need to be expressed in terms of Fe. For hazard classification the results need to be expressed in terms of the concentration of the individual salts. Where necessary these can be calculated from the Fe concentration using the adjustment factors given in the Table below.

Table: Adjustment factors to convert numerical endpoints expressed as Fe concentrations to numerical endpoints for iron compounds

Chemical Species	Molecular Weight	Ratio
FeSO4.7H2O	278.0	4.98
Fe(SO4)1.5	199.8	3.58
FeSO4H2O	169.9	3.04
FeSO4	151.9	2.72
FeCl3	162.3	2.91
FeCl3.6H2O	270.3	4.84
FeCl2	126.8	2.27
Fe	55.8	1.0

For clarity and in order not to misrepresent the accuracy of the test methods, the results are presented to two significant figures in the following text and tables.

The data used in environmental hazard assessment have been reviewed and discussed in the peer-reviewed and published SIARs (refer to Section 4.1 of the SIAR for Iron Salts and Section 4.1 of the SIAR for Iron Dichloride). The data were obtained from reviewing the following sources:

• confidential test reports
• published studies, including open literature and IUCLID 2000
• an in-depth CEFIC review of ferrous sulphate carried out as part of the recent European Chemicals Bureau classification process (CEFIC 2002, ECB 2004)
• data for FeSO4.7H2O, provided by
• data for FeCl2 compiled as part of a SIDS submission by
• the US EPA Ecotox database (http://cfpub.epa.gov/ecotox/).

Acute and chronic toxicity data were screened to comply with standard acceptance criteria that have been used for the EU metal risk assessments (Euras 2005). However the data for “Key studies” for each REACH endpoint that have been assessed to be reliable without restrictions (reliability 1) or reliable with restrictions (reliability 2) in accordance with Klimisch codes they were assessed not conclusive on the basis of the following considerations. Accordingly the Klimisch ratings in the selected studies which were included in the present dossier may be at variance to the ones in the ones in the iron salt category dossier.

The effects of iron sulphate on aquatic organisms in short-term tests are observed at nominal exposure concentrations in the range 1 – 1000 mg/L salt with the majority being in the range 10 – 100 mg/L. Effects arising from long-term exposures are observed at nominal concentrations > 1 mg/L.

Given the half-time for oxidation and precipitation detailed in the chapter on hydrolysis it is anticipated that a significant proportion of any ferrous salts added to oxygenated aqueous test media will have converted to ferric within the timescale of the standard OECD test protocols. The known LC50, EC50 and NOEC measured for the substance, expressed in terms of iron concentration, significantly exceed the equilibrium concentrations of dissolved ferric iron given below.

The ferrous ion, Fe(II), is unstable when its solutions are exposed to oxygen, and it oxidizes to the ferric ion, Fe(III). Thus testing of the iron the submission item under aerobic conditions is not feasible as the test item is unstable.

Nonetheless it may be considered to conduct guideline testing with equilibrated aerobic solutions. Calculated maximum solubility of ferric iron in solution for a pH range of 4.0 to 8.0 and a temperature of 20 ºC in mg Fe/L (mmol Fe/L) are 6.16E-02 (1.1E-03), 6.16E-05 (1.1E-06), 6.16E-08 (1.1E-09), 6.16E-11 (1.1E-12), and 6.16E-14 (1.1E-15) at pH 4, 5, 6, 7, and 8 respectively

Were the effects observed in the tests to have arisen from exposure to the presence of the dissolved ferric iron this would have placed the substance on a par with some of the most toxic substances that are known. This is unrealistic given that iron is an essential trace element for plants and animals that is widely distributed and abundant in nature. Even more as the background concentrations range in the order of the solubility

In view of the low equilibrium concentrations of dissolved ferric iron under aerobic conditions it can therefore be expected that most of the iron to which test organisms are exposed will be present as undissolved and precipitated ferric hydroxide. The Kd have been set to a very large value (1E+05) indicating that under environmental condition the precipitates would strongly bind to sediments and soil and thus be removed from the benthic water bodies. Effects arising from exposure to undissolved iron, such smothering or clogging of the gills or respiratory membranes in fish and invertebrates and restriction of plant growth by impairment of photosynthesis or nutrient chelation (in particular phosphorous), are not toxic effects and should not be used as the basis for deriving a PNEC.A fortiori the from of the undissolved iron and the exposure to this from constitutes an unnatural situation and is assumed to represent a testing artefact. On the other hand testing at the level of environmental concentrations seems not insightful and meaningless. Accordingly it is concluded that acute and chronic aquatic toxicity testing under aerobic conditions is technically not feasible and scientifically unjustified.

Iron salts may present a toxic hazard to environmental species but only under very specific geochemical conditions, which are dominated by environmental release from natural reservoirs rather than from additional anthropogenic load.

As the natural background concentration is assumed to have no effects on biota and is due to the equilibrium with large undissolved reservoirs comparable to the level of environmental solubility, it is concluded that the background is significant compared to the PNEC. In conclusion the Added Risk Approach according to ECHA (2008, Guidance on information requirements and chemical safety assessment, Appendix R.7.13-2: Environmental risk assessment for metals and metal compounds, p 27) is to be applied.

Due to the fact that any additional iron would rapidly precipitate and integrate to the large natural sediment and/or soil reservoirs no meaningful PNECadd can be derived on the basis of iron effects - except for the Sewage Water Treatment Plant influent, where no water-sediment equilibrium exist and the precipitate would be in direct contact to the activated sludge micro-organisms.

It is thus to consider whether aquatic PNECs can be derived from the remaining major constituents, i.e. manganese and aluminium.

• CEFIC Conseil Européen de l'Industrie Chimique (2002). Review of ecological data relating to ferrous sulphate. Unpublished report.
• ECB European Chemicals Bureau (2004). Document ECBI/14/01 Add.12 Critical review on acute and chronic aquatic ecotoxicity data to be used for classification purposes of iron sulfate. Final Report 25th August 2004. Marnix Vangheluwe & Bram Versonnen, EURAS.
• Euras (2005). Fact Sheet 1: Data compilation, selection and derivation of ecotoxicity reference values. MERAG Program- Building Block: Classification for effects on the aquatic environment (Version 12 April 2005) http://www.euras.be/assets/files/Fact%20Sheets%201-4.pdf.)

Manganese

Hazardous concentration to protect 95% of species with 50% confidence, i.e. HC5(50%), of 200 µg Mn/L freshwater was derived in the section on Aquatic toxicity. Nonetheless manganese precipitates to insoluble species under environmental equilibrium conditions, at pH < 8.5 the kinetics of these processes are slow (Zaw & Chiswell 1999). In conclusion toxic manganese concentrations may be reached in the environment and the PNEC derivation can be based on manganes effects.

As the natural background concentration of 34 µg/L (see discussion on environmental fate and pathways) is significant compared to the resulting aquatic PNECs. In conclusion the Added Risk Approach according to ECHA (2008, Guidance on information requirements and chemical safety assessment, Appendix R.7.13-2: Environmental risk assessment for metals and metal compounds, p 27) is to be applied.

Accordingly the aquatic PNECadd can be based on manganse effects.

• Zaw M, Chiswell B (1999). Iron and manganese dynamics in lake water. Water Research 33(8):1900–10.

Aluminium

Studies reported in the literature have extensively used test solutions (soluble salts) with aluminium concentrations above that of its solubility limit. Due to physical effects of precipitated material most of these studies are meaningless for the investigation of intrinsic toxicity. Aluminium ions released to surface waters quickly form insoluble aluminium hydroxides in mixing zones. Formation of the complex hydroxide causes the aluminium to drop out of solution very rapidly in neutral and alkaline waters. The dissolved natural background concentrations of aluminium, in most cases, are at equilibrium therefore an addition of aluminium would lead to the precipitation of aluminium compounds from solution and not result in effects to aquatic life.

It is thus concluded that the aquatic PNEC derivation should not be based on aluminium effects.

Conclusion on classification

Degradation and Persistence

As defined in the Glossary of OECD (2001), the term “degradation” refers to the decomposition of organic molecules.

For inorganic compounds and metals, clearly the concept of degradability, as it has been considered and used for organic substances, has limited or no meaning. Rather, the substance may be transformed by normal environmental processes to either increase or decrease the bioavailability of the toxic species. Equally, the log Kow cannot be considered as a measure of the potential to accumulate. Nevertheless, the concepts that a substance or a toxic metabolite/reaction product may not be rapidly lost from the environment and/or may bioaccumulate are as applicable to metals and metal compounds as they are to organic substances (OECD 2001).

Speciation of the soluble form can be affected by pH, water hardness and other variables, and may yield particular forms of the metal ion which are more or less toxic. In addition, metal ions could be made non-available from the water column by a number of processes (e.g. mineralisation and partitioning). Sometimes these processes can be sufficiently rapid to be analogous to degradation in assessing chronic classification. However, partitioning of the metal ion from the water column to other environmental media does not necessarily mean that it is no longer bioavailable, nor does it mean that the metal has been made permanently unavailable (OECD 2001).

• OECD Organisation for Economic Co-operation and Development Environment Directorate (2001). Guidance Document on the Use of the Harmonised System for the Classification of Chemicals which are hazardous for the Aquatic Environment. OECD Environment, Health and Safety Publications Series on Testing and Assessment Number 27. 115 p

Iron

Responses to effects arising from other secondary factors such as lowered pH and nutrient complexation will be dependent upon the susceptibility of resident organisms to perturbations in these parameters and the characteristics of the receiving environment (e.g. buffering capacity and background nutrient concentrations). The results of the laboratory tests and the published values suggest that it would be necessary for dissolved iron concentrations to exceed 1 mg/L before significant effects could be anticipated. Such concentrations are only likely to occur and persist under conditions of low dissolved oxygen concentration and low pH. Therefore related effects are considered not arising from true toxicity.

Even if the test results reported in this assessment are taken at face value as indicative of toxic effects, it was considered in a recent EU classification of ferrous sulphate heptahydrate and monohydrate that the majority of acute toxicity end points had values > 10 mg/L iron salt and that the reduction of soluble iron concentrations below measured chronic NOECs could be considered as equivalent to ‘rapid degradability’ leading to a ‘no environmental hazard’ classification decision.

Manganese

Manganese sulphate (CAS 7785-87-7, EC# 232-089-9) is classified R51/53 according to 67/548/EEC Annex 1, index number 025 -003-00-4 but manganese chloride is not. As classification is based on weight but effects on molarity it seems indicated to compare the molecular weights of manganese sulphate and manganese chloride: MW (MnSO4) = 150.996 g/mol, MW (MnCl2) = 125.844. As there is only a minor difference similar Annex 1 classification should apply to manganese chloride.

Manganese is a constituent contributing to about 9.8 mol% of total metals in the submission item. The dry matter of the submission item contains 9.852 % weight manganese. On this basis it is clearly visible that more than the 10 fold amount of the submission item would release comparable Mn2+ quantities as the corresponding manganese sulphate amount. This places the submission item in the 10 mg/ < EC50 = 100 mg/L category. The biodegradation criterion is considered inappropriate for naturally occurring bioessential metals in accordance with OECD (2001) Guidenace (see above). The manganese bioavailability depends on geochemical equilibration. Additional environmentally released Mn2+ would be precipitated (in the order of days) and not increase the bioavailability according to the mechanisms discussed in the section on stability. Thus it is considered to be treated like biodegradable materials in the classification context. In consequence R52/R53 does not apply to the submission item on the basis of its manganese content.

It is assessed that the submission item does not present a danger to the structure and/or functioning of aquatic ecosystems. Thus no R52 classification applies.

No safety net classification according to GHS/CLP is considered appropriate.

Aluminium

Aluminium is a constituent contributing to about 9.3 mol% of total metals in the submission item. Available data indicate that aluminium salts are relatively non toxic in most waters with circumneutral pH and this was sufficient for the EU Classification and Labelling Committee (1999) to determine that there was no need for classification of aluminium chloride. Other aluminium compounds act similarly in water as aluminium chloride and are in many cases less soluble and non-hazardous.