Silver nitrate

Endpoint summary

Additional information

The hazard assessment has been conducted primarily to be consistent with ECHA guidance for Chemical Safety Assessment. However, this assessment is also built on aspects of guidance developed for the derivation of EQS under the Water Framework Directive (European Commission 2018) and the MERAG (Metals Environmental Risk Assessment Guidance) programme (ICMM 2016).

In line with the risk assessment/classification approach adopted for other metals and inorganic metal compounds (ECHA 2012), aquatic toxicity information requirements are “read across” from the properties of the dissolved silver ion (released from soluble inorganic silver species in aqueous solutions). Almost exclusively these are studies that used silver nitrate (AgNO₃) as the test substance. Silver nitrate is considered to be the form of silver with the greatest toxicity as it dissociates rapidly and completely in aqueous solution in Ag+ and NO3- ions (with the effect of the nitrate counterion being irrelevant compared to that of Ag+). Where data for silver nitrate was not available, data derived from other inorganic salts (e.g. silver chloride) have been used, but only after the exposure conditions were determined to be acceptable (e.g. testing was conducted within the limits of solubility and Ag+ ion likely to be fully dissociated). Where applicable, further justification of the validity of this read-across approach is made in endpoint summaries.

After review of the available information in the scientific and grey literature (last search data end 2021), long-term reliable freshwater toxicity data on silver are available for twelve taxonomic groups: cyprinid fish, salmonid fish, crustaceans (Cladocera and Amphipoda), insects (Ephemeroptera and Diptera), rotifers, molluscs (Bivalvia and Gastropoda), cyanobacteria, algae and higher plants.

The effects database for marine species is considerably smaller than that for freshwater organisms. Long-term toxicity data are available for five different taxonomic groups: algae, crustaceans, echinoderms, fish and molluscs. The saltwater toxicity data suggest that silver toxicity is modified by salinity, with increased toxicity occurring at lower salinities (Ward et al. 2006b, Pedroso et al. 2007). Because the speciation of silver across different media differs, in line with guidance, the freshwater and saltwater datasets have not been combined when deriving the PNEC for the marine environment.

The PNEC for the freshwater compartment has been derived from the available data using a statistical extrapolation methodology (species sensitivity distributions, SSDs). The quality criteria used for selection of chronic freshwater toxicity values were in line with available guidances from REACH (ECHA 2011) and the European Water Framework Directive 2000/60/EC (European Commission 2018) and with the criteria for reporting and evaluating ecotoxicity data (Moermond et al. 2016). Evaluation criteria for the reliability of the toxicity studies were as follows:

1)    only chronic toxicity values based on measured dissolved Ag concentrations (mostly filtered over a 0.45-µm filter) and using an ionic Ag test substance (usually Ag nitrate) were selected;

2)    considering the strong influence of water physicochemistry on metal toxicity, the test conditions that could influence the bioavailability and toxicity (pH, hardness, DOC, chloride) of Ag should be adequately described and should be within the tolerance limits of the test organisms as indicated in the corresponding test guidelines; and

3)    concentration–response modeling (e.g., regression methods) were preferred over hypothesis-testing methods (NOEC values), with the use of EC10 (i.e., the modeled concentration causing a 10% decrease in response) as the preferred endpoint for deriving safe thresholds.

Evaluation criteria for the relevance of chronic toxicity were also applied. Relevance covers the extent to which a test is appropriate for a particular risk assessment and the evaluation of the choice of test species, the test duration, and the test substance used. Regarding test species, all freshwater species for which Ag toxicity data are available were considered including species not usually tested in standardized test procedures. Regarding test duration, a relevant chronic test duration is a function of the life cycle of the test organism. Recommendations from standard ecotoxicity protocols were followed. Chronic toxicity tests were generally defined as >4 d for all invertebrates and fish. It is noted, though, that whether or not an effect concentration is considered chronic is not determined exclusively by the exposure duration limit of 4 d. For unicellular algae but also for specific invertebrates (e.g., rotifers), an exposure time of <4 d usually already covers one or more generations. Thus, for these organisms, chronic effect concentrations may be derived from experiments of <4 d. For algae, the minimum required exposure time is 48 h (OECD 2011). The relevance of specific exposure durations for the estimation of chronic effects for organisms with relatively long life cycles (e.g., fish) was evaluated on a case-by-case basis (e.g., by considering the use of sensitive life stages in the test).

A single “Key” NOEC/EC10 for each species (or species geometric mean NOEC/EC10, see below) was selected from the available reliable and relevant NOECs/EC10s based on the following criteria:

·        Where more than one endpoint from a test was generated, the most sensitive endpoint was selected for inclusion in the SSD;

·        In the absence of a chronic BLM for Ag, in case toxicity values at different hardness conditions were reported for the same species, only the EC10 from the test at the lowest hardness (potentially representing the highest bioavailability and thus highest toxicity) was retained for Ag threshold derivation purposes.This is so that the derived PNEC is protective across the range of physico-chemical conditions expected across the EU. It should be noted that the relationship between water physico-chemistry and chronic silver toxicity has yet to be fully elucidated. As such, this additional data selection criterion should be considered precautionary.

·        Where appropriate, the results for key studies using the same species tested under sufficiently similar conditions were combined as species geometric means.

An SSD was then fitted to these key data to derive an HC5 concentration. Additional reliable and relevant data that were not used for PNEC derivation (as they were not consistent with the criteria described above) were considered as supporting information only.

The PNECs for the marine compartment, sediments and aquatic microorganisms were calculated based on an assessment factor approach, using the most sensitive reliable and relevant data available in combination with standard assessment/safety factors.

Both the SSD and assessment factor approaches are compliant with Chapter R.10 of ECHA guidance (ECHA 2008). For full details of the calculated PNECs and the identification and selection of the data see the PNEC summary document attached in IUCLID Section 13.The freshwater PNEC has been re-derived in 2021 by Arijs et al. (2021); this revision is also attached in IUCLID Section 13. Since then, a literature search covering the period until end 2021 revealed 2 additional relevant and reliable studies (Wang et al. 2019 and Kusi and Maier 2022), which have been included in the PNEC derivation by statistical extrapolation.

Potential Mitigating Effect of Sulfide on Chronic Silver Ecotoxicity

Some freshwater physico-chemical properties are thought to have an influence on chronic silver toxicity. As well as the usual properties affecting metal bioavailability in freshwaters, such as dissolved organic carbon, sulfide is also thought to have an important role in influencing the ecotoxicity of silver. Historically, one of the practical difficulties of assessing the influence of sulfide on silver in freshwaters has been the complexity of the required chemical analysis and commensurate absence of any high quality routine monitoring data. These problems have been addressed by the National Laboratory Service of England and Wales and a study was undertaken by the Environment Agency of England and Wales and EPMF where silver and accompanying sulfide measurements were collected from a range of silver exposures in freshwaters.

The results of this study indicate that free sulfide (operationally defined as chromium reducible sulfide, CRS) could have a potentially significant mitigating effect upon the free silver ion concentration in solution. Further details are provided below.

The acute toxicity of silver has been shown to be mitigated, to some degree, by elevated levels of sodium, chloride, DOC and, to a lesser degree, hardness. The limited effectiveness of hardness cations at mitigating short-term silver toxicity is probably related to the high strength of binding of Ag+ at the site of action, (i.e. to the biotic ligand), a characteristic that makes it difficult for Ag+ to be displaced by competing cations. Chromium reducible sulfide (CRS) is another water quality constituent that has the potential to be very important, particularly at relatively low silver concentrations, as a result of the high affinity of CRS for Ag+.

Monitoring data from the UK from 2010 (wca environment 2010) indicates that the environmental levels of silver tend to be relatively low, in most cases less than the PNEC of 40 ng/L (0.04 µg/L, or ~0.37 nM). Kramer et al. (2007) estimate CRS concentrations in freshwater as 14.6 nmol/mg DOC. As the DOC concentrations typical range from 2 – 10 mg/L (or higher) in Europe, it is expected CRS levels will commonly be in the concentration range of about 30 – 150 nM, an estimate which is consistent with the monitoring results from the UK programme (Peters et al. 2009). The molar ratio of CRS to Ag+ should therefore be about 75 – 425 (using the silver PNEC as a conservatively high estimate of the ambient Ag+ concentration). Although other cationic metals will also bind to CRS, they are not effective at competing with Ag+ for CRS binding sites. Hence, even for the relatively high metal concentrations that were assumed above, Ag+ should effectively compete with the other cationic metals for complexation by CRS and there should normally be enough CRS to complex the Ag+ that is present. This should in turn markedly reduce the concentration of Ag+, an important bioavailable form of silver (Bianchini and Bowles, 2002). The question that remains to be addressed is whether or not Ag-CRS complexes are sufficiently low in bioavailability and that the residual Ag+ concentration is sufficiently low such that an adequate level of protection against silver toxicity, chronic toxicity in particular, will be realized. Some of the experimental evidence that is relevant to consider in regard to this topic is summarized below.

Bianchini and coworkers investigated the protective effect of CRS on toxicity due to Ag+. They considered the effect of CRS on the acute toxicity of Ag+ toD. magna, a relatively sensitive invertebrate (Bianchini et al. 2002). When CRS was present at 25 nM it increased the Ag LC50 by 5.6-fold relative to the LC50 in water without CRS (from 0.16 and 0.26 µg/L to 1.47 µg/L). This increase is comparable to, but somewhat less, than the Ag+ complexation capacity of the CRS (2.7 µg/L). The difference may in part reflect uncertainty in the measured concentrations, given that changes in silver concentration have been shown to occur over the course of toxicity tests (Bowles et al. 2002). It may also reflect the allowance of an insufficient time (3 hours) for pre-equilibration of the Ag-CRS complex prior to daily water renewal. This latter explanation is consistent with the results of Glover et al. (2005) who showed an increase in LC50 of approximately 55% to 100% (as much as a factor of two increase, depending on DOC level) when the equilibration time between silver and NOM (presumably associated with CRS) was increased from 3 hours to 24 hours.

Importantly, when the CRS concentration in the preceding study exceeded the total Ag by more than 10-fold (250 nM vs 18.5 nM), it provided complete protection toD. magnaafter short-term exposure over the full range of the silver concentrations used to define the dose-response curve in the absence of CRS (i.e. over a range of 0 - 2 µg/L Ag). This provides a clear demonstration of the protective effect of CRS, at least with regard to the short-term toxicity of silver. Interestingly, silver was accumulated by theD. magnato a greater degree in the presence of CRS, a result attributed to sulfide-bound silver in the digestive tract of the daphnids rather than silver bound to the exoskeleton (Bianchini et al. 2004). The observation that silver accumulation occurred to a greater extent in the presence of elevated CRS (when toxicity was reduced) is an indication that the accumulation occurred via a physical process such as ingestion of colloidal silver and that the accumulated silver was not toxicologically available to the organism. Additional measurements demonstrated that the gastrointestinal tract silver burden was in fact markedly elevated while depuration experiments showed that it was readily eliminated from the organism over a time course of several hours (Bianchini et al. 2004).

With regard to long-term toxicity, a more recent study evaluated the protective effects of either hardness or CRS (at an environmentally relevant concentration of 23 nM, as might be associated with DOC ~ 1.6 mg/L) on the acute (48 hour) and chronic (21 day) toxicity of silver toD. magna(Bianchini and Wood 2008). Effect levels were computed on the basis of both total and dissolved Ag. On the basis of a 1:1 stoichiometry for AgHS (Bowles et al. 2002a) the amount of CRS added to the test water was equivalent to 2.48 µg/L Ag+ complexation capacity. TheD. magnaeffect levels for dissolved silver in the presence of CRS were essentially the same as for silver in the absence of CRS due to the demonstrated sorption of Ag CRS to membrane filters and other surfaces (Bowles et al. 2002a,b). As a result, the dissolved Ag data are not of direct use in an assessment of the level of protection provided by CRS. On a total Ag basis, however, the effect level for the 48 hour EC50 for survival was increased from 6.9 to 8.3 µg/L in the presence of CRS. This increase of 1.4 µg/L is somewhat less than the estimated complexation capacity. Again, the difference may have resulted from the relatively short Ag-CRS equilibration time that was provided (3 hours) when the exposure water was renewed each day, as this would limit the degree of Ag complexation that could occur. The 21 day chronic tests exhibited a result that was more consistent with stoichiometric expectations, with the 21 day survival EC50 increased by 2.23 µg/L in the presence of CRS (slightly less than the estimated complexation capacity of CRS).

The 21 day chronic toxicity test results forD. magnaalso showed that that CRS provided a protective benefit for several of the end points that were reported. The tests were performed with and without the addition of 23 nM of CRS (complexation capacity ~ 2.48 µg/L). End points included total number of neonates produced (TN), time to first brood (TB), number of broods (NB), number of young per brood (YB), number of reproduction days (RD) and number of young per adult per reproduction day (YAD). The differences in total Ag LC50s (with and without CRS) in 48 hour and 21 day tests were 1.4 µg/L and 2.23 µg/L, respectively. The increase for sub-lethal effects on reproduction (YAD) was 1.65 µg/L. The collective average of 1.8 µg/L is similar to the complexation capacity of 23 nM CRS (2.48 µg/L). Again, the somewhat lower value that was observed likely reflects the short equilibration time that was provided when the test water was renewed each day. Other end points exhibit a limited benefit of the added CRS. However, the silver effect levels were relatively high in comparison to the CRS levels, so this is not surprising.

Chronic toxicity study results have also been reported forC. dubia(7 day static renewal) exposed to silver in the presence and absence of sulfide (added as CuS; Naddy et al. 2007). The CRS increase from <2.2 nM (Horsetooth Reservoir control water) to 75.4 nM (8.14 µg/L complexation capacity) led to a 43% increase in the IC20. The NOEC was 17.5 µg/L. The CRS complexation capacity of 8.14 µg/L was slightly higher than the increase in the survival and reproduction NOECs of 6.1 and 5.8 µg/L, respectively. Once again, only a three hour equilibration time was used, so the full benefit of the added CRS may not have been realized. Even so, the results further demonstrate the potential protective benefit of CRS.

With regard to the UK monitoring data, ambient silver levels are in most cases less than the PNEC of 40 ng/L (0.04 µg/L). Further, CRS is typically present at much higher concentrations than silver, such that CRS is available in considerable excess. As was seen above for the short-term toxicity results, when CRS was present at a 10-fold higher concentration than the effect level, silver toxicity was essentially eliminated. Based on these results, toxicity due to silver is likely to be effectively limited by CRS in most settings, even in instances where the freshwater PNEC (which was derived in the absence of CRS) may be exceeded.

Whilst the theoretical basis for the influence of sulfide on the speciation/bioavailability of silver in the freshwater environment is compelling, and the empirical evidence significant, there are several aspects of the relationship between abiotic water chemistry and silver hazard in the freshwater environment that require additional understanding before they are applied in a generic chemical safety assessment for silver and nanosilver. Specifically, additional research on the precise speciation behaviour of silver ions in freshwaters and the chronic toxicity of ligand bound silver are required. Equally, the role of speciation/bioavailability with respect to hazard of nanosilver is not well developed. As such, PNEC values and the associated risk characterisation of silver in the CSR do not incorporate the influence of abiotic factors on toxicity. Research is currently ongoing to refine this understanding.

References cited:

Arijs, K., Nys, C., Van Sprang, P., De Schamphelaere, K. and Mertens, J., 2021.“Setting a Protective Threshold Value for Silver Toward Freshwater Organisms”. Environmental Toxicology and Chemistry, 40, 1678-1693.

Bianchini, A., and K.C. Bowles, 2002. “Metal Sulfides in Oxygenated Aquatic Systems: Implication for the Biotic Ligand Model,”Comparative Biochemistry and Physiology, Part C 133: 51-64.

Bianchini, A., K.C. Bowles, C.J. Brauner, J.W. Gorsuch, J.R. Kramer and C.M. Wood, 2002. “Evaluation of the Effect of Reactive Sulfide on the Acute Toxicity of Silver (I) toDaphnia magna. Part 2: Toxicity Results,” Environmental Toxicology and Chemistry, 21(6): 1294-1300.

Bianchini, A., C. Rouleau and C.M. Wood, 2004. “Silver Accumulation in Daphnia magna in the Presence of Reactive Sulfide,” Aquatic Toxicology, 72(4): 339-349.

Bianchini, A., and C. M. Wood. 2008. “Does sulfide or water hardness protect against chronic silver toxicity in Daphnia magna? A critical assessment of the acute-to-chronic toxicity ratio for silver,” Ecotoxicology and Environmental Safety 71:32-40.  

Bowles K.C., A. Bianchini, C.J. Brauner, J.R. Kramer, C. M. Wood. 2002a. “Evaluation of the effect of reactive sulfide on the acute toxicity of silver (I) to Daphnia magna. Part 1: description of the chemical system,” Environmental Toxicology and Chemistry21:1286-1293. 

Bowles, K.C., R.A. Bell, M.J. Ernste, J.R. Kramer, H. Manolopoulos and N. Ogden, 2002b. “Synthesis and Characterization of Metal Sulfide Clusters for Toxicological Studies,”Environmental Toxicology and Chemistry, 21(4): 693-699.

European Chemical Agency [ECHA]. 2008. Guidance on information requirements and chemical safety assessment. Chapter R.10: Characterisation of dose [concentration]-response for environment. ECHA, Helsinki.

European Chemical Agency [ECHA]. 2011. Guidance on information requirements and chemical safety assessment. Chapter R.4: Evaluation of available information. European Chemicals Agency.

European Commission. 2018. Guidance document No. 27 - Updated version 2018. Technical Guidance for Deriving Environmental Quality Standards.

Glover, C., R.C. Playle and C.M. Wood, 2005. “Heterogeneity of Natural Organic Matter Amelioration of Silver Toxicity toDaphnia magna: Effect of Source and Equilibration Time,” Environmental Toxicology and Chemistry, 24(11): 2934-2940.

International Council on Mining and Metals [ICMM]. 2016. MERAG: Metals Environmental Risk Assessment Guidance. ISBN: 978-1-909434-20-2. ICMM, London, UK

Kramer J.R., R.A. Bell, D.S. Smith. 2007. “Determination of sulfide ligands and association with natural organic matter,” Applied Geochemistry 22:1606-1611.

Moermond CT, Kase R, Korkaric M, Ågerstrand M. 2016. CRED: Criteria for reporting and evaluating ecotoxicity data. Environ Toxicol Chem35:1297–1309.

Naddy, R.B., J.W. Gorsuch, A.B. Rehner, G.R. McNerney, R.A. Bell and J.R. Kramer, 2007. “Chronic Toxicity of Silver Nitrate to Ceriodaphnia dubia and Daphnia magna, and Potential Mitigating Factors,” Aquatic Toxicology, 84: 1-10.

OECD. 2011. Test No. 201: Freshwater alga and cyanobacteria, growth inhibition test. OECD Guidelines for the Testing of Chemicals. Paris, France.

wca environment. 2010. Silver emissions to freshwaters in England and Wales. A report to the Environment Agency of England and Wales and the Precious Metals Consortia from WCA Environment Ltd. Report Number P0181_09-10.