Silver

Long-term toxicity to aquatic invertebrates

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

key study

Justification for type of information:

1. HYPOTHESIS FOR THE READ-ACROSS APPROACH (ENDPOINT LEVEL)
The REACH registration of silver (powder and massive forms of zero-valent, elemental, silver) is underpinned, in common with other metals, by a read-across (or analogue) approach from the properties of the free ion. This principle of read-across from the free ion has been extended to also include nanosilver.
The scientific validity of read-across from the hazard properties of ionic silver (source) to nanosilver (target) under REACH is underpinned by both theoretical and empirical considerations.
The theoretical basis for the use of ionic silver data as the foundation of the risk assessment of nanosilver is based on the premise that the free metal ion (Me+) is the most toxic metal form/species (Starodub et al. 1987). This consideration was implicit in the development of the Free Ion Activity Model [FIAM] (Morel 1983, Paquin et al. 2002, Campbell 1985, Brown and Markich 2000) and, more recently, the Biotic Ligand Model [BLM] (Paquin et al 2002, Niyogi and Wood 2004) that has underpinned the risk assessment of several metals (e.g. Cu, Ni, Zn) under the Existing Substance Regulations and REACH; and most recently the development of the Environmental Quality Standard (EQS) for nickel and nickel substances under the Water Framework Directive (WFD). When considered on an equal mass basis ionic silver would therefore be expected to have greater toxicity than nanosilver simply on the basis that silver ions are released over time from the surface of particles (via oxidative dissolution). As the properties of nanosilver are read-across directly from ionic silver (not just to the fraction of silver ions released from nanosilver), this read-across is also expected to introduce considerable precaution into the hazard component of the risk assessment of nanosilver as all nanosilver, irrespective of coating, morphology or particle size distribution is assumed to behave similarly to the free ion.
This theoretical consideration has been tested by conducting a comprehensive review of the available scientific literature for nanosilver, with particular emphasis on the comparative effects on REACH relevant biotic systems (REACH information requirement) of ionic silver and nanosilver. This review is described for each endpoint in subsequent sections of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IULCID Section 13) and is summarised below. Furthermore, this theoretical consideration was confirmed by a specific ecotoxicity testing programme undertaken by the EPMF following the silver substance evaluation and designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms.

2. READ-ACROSS APPROACH JUSTIFICATION (ENDPOINT LEVEL)
In terms of ecotoxicity, nanosilver on an equal mass basis has been found to be significantly less toxic than ionic silver for the majority of REACH endpoints and of equivalent toxicity for some others. None of the empirical information available suggests that nanosilver is consistently more hazardous than ionic silver on an equivalent mass basis, or that “nano-specific” effects that would prejudice the validity of read across from ionic silver to nanosilver occur. In addition, with very limited exceptions which are described further in the report ‘Nanosilver: read-across justification for environmental information requirements’, none of the available data suggested consistent relationships between particle morphology, size, particle size distribution (raw or agglomerated) or coating (surface treatment) and effects.
Notter et al. (2014) presents a meta-analysis of published EC50 values for ionic silver and nanosilver. The authors demonstrate that almost 94% of acute toxicity values assessed for freshwater, seawater and terrestrial systems using algae, annelid, arthropoda, bacteria, crustacea, fish, nematoda, plant, protozoa and rotatoria show that the nanoform of silver is less toxic than the dissolved metal (when normalised for total metal concentration).
In addition, a specific ecotoxicity testing programme designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms was undertaken following the silver substance evaluation. This demonstrated that the nanoform of silver is less toxic than ionic silver (based on EC10 and EC50 values for total, ‘conventional’ dissolved (<0.45μm) and ‘truly’ dissolved (<3kDa) silver). Therefore, taking the full body of evidence into account, the read-across use of toxicity values from ionic to nanosilver as a ‘worst case’ approach is justified and scientifically defensible for environmental endpoints. The Substance Evaluation Conclusion document for silver agreed with this conclusion for the nanosilver forms covered in the REACH dossier (RIVM 2018).

Long-term toxicity to aquatic invertebrates
Published data from two long-term toxicity to Daphnia magna studies using various sizes of nanoparticles and coating types are included in the REACH dossier as Endpoint Study Records. The nanosilver NOEC values from these studies are in the same order of magnitude or higher than the EC10 of 2.14 μg/L for Daphnia magna for ionic silver (Bianchini and Wood 2008). A summary of these supporting studies is available under Section 4.1.4 of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IUCLID Section 13). There is insufficient chronic data for aquatic invertebrates to investigate the influence of particle size of coating on the observed toxicity.
In addition, following the silver substance evaluation, Fraunhofer (Schlich et al. 2017) undertook comparative studies for the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and ionic silver (silver nitrate) with Daphnia magna, in accordance to OECD guideline 211. The most sensitive endpoint in both tests was reproduction after 21 days. The EC10 value for reproduction was 33.4 µg/L measured dissolved silver (<0.45 µm) for nanosilver and 3.49 µg/L measured dissolved silver (<0.45 µm) for silver nitrate. Thus, the directly comparable EC10 values reported by Schlich et al (2017) were an order of magnitude more toxic for ionic silver than for nanosilver. These studies are included in the REACH dossier as Endpoint Study Records and a summary is available in the report ‘Summary of Comparative Ecotoxicity Testing Programme for Nanosilver and Silver Nitrate (2017)’ (attached in IUCLID Section 13).
Together with the theoretical basis for read-across based on the free-ion, above studies support the use of ionic silver as the ‘worst case’ basis to read across properties to nanosilver.



For further information and data matrix see 'CSR Annex 9 - Read Across Justification Nanosilver ENV_SUMMARY_200706' attached in IUCLID section 13.

Reason / purpose for cross-reference:

read-across source

Post exposure observation period:

None

Duration:

21 d

Dose descriptor:

EC10

Effect conc.:

2.14 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

not specified

Basis for effect:

growth

Details on results:

The EC10 for Daphnia magna is 2.14 µg/L for growth.

Results with reference substance (positive control):

Not applicable

Reported statistics and error estimates:

No data reported

Validity criteria fulfilled:

yes

Conclusions:

The EC10 for Daphnia magna is 2.14 µg/L for growth.

Executive summary:

This is a non-GLP, chronic toxicity study on Daphnia magna following ASTM guidelines for conducting D. magna life cycle tests. The study is considered reliable and suitable for use for this endpoint.

Reference 2

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

key study

Justification for type of information:

1. HYPOTHESIS FOR THE READ-ACROSS APPROACH (ENDPOINT LEVEL)
The REACH registration of silver (powder and massive forms of zero-valent, elemental, silver) is underpinned, in common with other metals, by a read-across (or analogue) approach from the properties of the free ion. This principle of read-across from the free ion has been extended to also include nanosilver.
The scientific validity of read-across from the hazard properties of ionic silver (source) to nanosilver (target) under REACH is underpinned by both theoretical and empirical considerations.
The theoretical basis for the use of ionic silver data as the foundation of the risk assessment of nanosilver is based on the premise that the free metal ion (Me+) is the most toxic metal form/species (Starodub et al. 1987). This consideration was implicit in the development of the Free Ion Activity Model [FIAM] (Morel 1983, Paquin et al. 2002, Campbell 1985, Brown and Markich 2000) and, more recently, the Biotic Ligand Model [BLM] (Paquin et al 2002, Niyogi and Wood 2004) that has underpinned the risk assessment of several metals (e.g. Cu, Ni, Zn) under the Existing Substance Regulations and REACH; and most recently the development of the Environmental Quality Standard (EQS) for nickel and nickel substances under the Water Framework Directive (WFD). When considered on an equal mass basis ionic silver would therefore be expected to have greater toxicity than nanosilver simply on the basis that silver ions are released over time from the surface of particles (via oxidative dissolution). As the properties of nanosilver are read-across directly from ionic silver (not just to the fraction of silver ions released from nanosilver), this read-across is also expected to introduce considerable precaution into the hazard component of the risk assessment of nanosilver as all nanosilver, irrespective of coating, morphology or particle size distribution is assumed to behave similarly to the free ion.
This theoretical consideration has been tested by conducting a comprehensive review of the available scientific literature for nanosilver, with particular emphasis on the comparative effects on REACH relevant biotic systems (REACH information requirement) of ionic silver and nanosilver. This review is described for each endpoint in subsequent sections of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IULCID Section 13) and is summarised below. Furthermore, this theoretical consideration was confirmed by a specific ecotoxicity testing programme undertaken by the EPMF following the silver substance evaluation and designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms.

2. READ-ACROSS APPROACH JUSTIFICATION (ENDPOINT LEVEL)
In terms of ecotoxicity, nanosilver on an equal mass basis has been found to be significantly less toxic than ionic silver for the majority of REACH endpoints and of equivalent toxicity for some others. None of the empirical information available suggests that nanosilver is consistently more hazardous than ionic silver on an equivalent mass basis, or that “nano-specific” effects that would prejudice the validity of read across from ionic silver to nanosilver occur. In addition, with very limited exceptions which are described further in the report ‘Nanosilver: read-across justification for environmental information requirements’, none of the available data suggested consistent relationships between particle morphology, size, particle size distribution (raw or agglomerated) or coating (surface treatment) and effects.
Notter et al. (2014) presents a meta-analysis of published EC50 values for ionic silver and nanosilver. The authors demonstrate that almost 94% of acute toxicity values assessed for freshwater, seawater and terrestrial systems using algae, annelid, arthropoda, bacteria, crustacea, fish, nematoda, plant, protozoa and rotatoria show that the nanoform of silver is less toxic than the dissolved metal (when normalised for total metal concentration).
In addition, a specific ecotoxicity testing programme designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms was undertaken following the silver substance evaluation. This demonstrated that the nanoform of silver is less toxic than ionic silver (based on EC10 and EC50 values for total, ‘conventional’ dissolved (<0.45μm) and ‘truly’ dissolved (<3kDa) silver). Therefore, taking the full body of evidence into account, the read-across use of toxicity values from ionic to nanosilver as a ‘worst case’ approach is justified and scientifically defensible for environmental endpoints. The Substance Evaluation Conclusion document for silver agreed with this conclusion for the nanosilver forms covered in the REACH dossier (RIVM 2018).

Long-term toxicity to aquatic invertebrates
Published data from two long-term toxicity to Daphnia magna studies using various sizes of nanoparticles and coating types are included in the REACH dossier as Endpoint Study Records. The nanosilver NOEC values from these studies are in the same order of magnitude or higher than the EC10 of 2.14 μg/L for Daphnia magna for ionic silver (Bianchini and Wood 2008). A summary of these supporting studies is available under Section 4.1.4 of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IUCLID Section 13). There is insufficient chronic data for aquatic invertebrates to investigate the influence of particle size of coating on the observed toxicity.
In addition, following the silver substance evaluation, Fraunhofer (Schlich et al. 2017) undertook comparative studies for the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and ionic silver (silver nitrate) with Daphnia magna, in accordance to OECD guideline 211. The most sensitive endpoint in both tests was reproduction after 21 days. The EC10 value for reproduction was 33.4 µg/L measured dissolved silver (<0.45 µm) for nanosilver and 3.49 µg/L measured dissolved silver (<0.45 µm) for silver nitrate. Thus, the directly comparable EC10 values reported by Schlich et al (2017) were an order of magnitude more toxic for ionic silver than for nanosilver. These studies are included in the REACH dossier as Endpoint Study Records and a summary is available in the report ‘Summary of Comparative Ecotoxicity Testing Programme for Nanosilver and Silver Nitrate (2017)’ (attached in IUCLID Section 13).
Together with the theoretical basis for read-across based on the free-ion, above studies support the use of ionic silver as the ‘worst case’ basis to read across properties to nanosilver.


For further information and data matrix see 'CSR Annex 9 - Read Across Justification Nanosilver ENV_SUMMARY_200706' attached in IUCLID section 13.

Reason / purpose for cross-reference:

read-across source

Duration:

10 d

Dose descriptor:

LC50

Effect conc.:

0.057 other: mg dissolved Ag/L

Nominal / measured:

meas. (geom. mean)

Conc. based on:

dissolved

Remarks:

silver

Basis for effect:

growth

Remarks on result:

other: (0.030 - 0.108)

Duration:

10 d

Dose descriptor:

EC10

Effect conc.:

0.014 other: mg dissolved Ag/L

Nominal / measured:

meas. (geom. mean)

Conc. based on:

dissolved

Remarks:

silver

Basis for effect:

growth

Duration:

10 d

Dose descriptor:

NOEC

Effect conc.:

0.013 other: mg dissolved Ag/L

Nominal / measured:

meas. (geom. mean)

Conc. based on:

dissolved

Remarks:

silver

Basis for effect:

growth

Duration:

10 d

Dose descriptor:

LOEC

Effect conc.:

0.066 other: mg dissolved Ag/L

Nominal / measured:

meas. (geom. mean)

Conc. based on:

dissolved

Remarks:

silver

Basis for effect:

growth

Details on results:

No data reported

Results with reference substance (positive control):

Not applicable

Reported statistics and error estimates:

No data reported

Validity criteria fulfilled:

not applicable

Remarks:

No applicable criteria but water quality appears acceptable.

Conclusions:

The 10 day EC10 (growth) of AgNO3 to Chironomus tentans was determined to be 0.0144 mg dissolved Ag/L.

Executive summary:

In a non-GLP, non-guideline, test the 10 day EC10 (growth) of AgNO3 to Chironomus tentans was determined to be 0.0144 mg dissolved Ag/L.

Reference 3

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

key study

Study period:

Not reported

Justification for type of information:

1. HYPOTHESIS FOR THE READ-ACROSS APPROACH (ENDPOINT LEVEL)
The REACH registration of silver (powder and massive forms of zero-valent, elemental, silver) is underpinned, in common with other metals, by a read-across (or analogue) approach from the properties of the free ion. This principle of read-across from the free ion has been extended to also include nanosilver.
The scientific validity of read-across from the hazard properties of ionic silver (source) to nanosilver (target) under REACH is underpinned by both theoretical and empirical considerations.
The theoretical basis for the use of ionic silver data as the foundation of the risk assessment of nanosilver is based on the premise that the free metal ion (Me+) is the most toxic metal form/species (Starodub et al. 1987). This consideration was implicit in the development of the Free Ion Activity Model [FIAM] (Morel 1983, Paquin et al. 2002, Campbell 1985, Brown and Markich 2000) and, more recently, the Biotic Ligand Model [BLM] (Paquin et al 2002, Niyogi and Wood 2004) that has underpinned the risk assessment of several metals (e.g. Cu, Ni, Zn) under the Existing Substance Regulations and REACH; and most recently the development of the Environmental Quality Standard (EQS) for nickel and nickel substances under the Water Framework Directive (WFD). When considered on an equal mass basis ionic silver would therefore be expected to have greater toxicity than nanosilver simply on the basis that silver ions are released over time from the surface of particles (via oxidative dissolution). As the properties of nanosilver are read-across directly from ionic silver (not just to the fraction of silver ions released from nanosilver), this read-across is also expected to introduce considerable precaution into the hazard component of the risk assessment of nanosilver as all nanosilver, irrespective of coating, morphology or particle size distribution is assumed to behave similarly to the free ion.
This theoretical consideration has been tested by conducting a comprehensive review of the available scientific literature for nanosilver, with particular emphasis on the comparative effects on REACH relevant biotic systems (REACH information requirement) of ionic silver and nanosilver. This review is described for each endpoint in subsequent sections of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IULCID Section 13) and is summarised below. Furthermore, this theoretical consideration was confirmed by a specific ecotoxicity testing programme undertaken by the EPMF following the silver substance evaluation and designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms.

2. READ-ACROSS APPROACH JUSTIFICATION (ENDPOINT LEVEL)
In terms of ecotoxicity, nanosilver on an equal mass basis has been found to be significantly less toxic than ionic silver for the majority of REACH endpoints and of equivalent toxicity for some others. None of the empirical information available suggests that nanosilver is consistently more hazardous than ionic silver on an equivalent mass basis, or that “nano-specific” effects that would prejudice the validity of read across from ionic silver to nanosilver occur. In addition, with very limited exceptions which are described further in the report ‘Nanosilver: read-across justification for environmental information requirements’, none of the available data suggested consistent relationships between particle morphology, size, particle size distribution (raw or agglomerated) or coating (surface treatment) and effects.
Notter et al. (2014) presents a meta-analysis of published EC50 values for ionic silver and nanosilver. The authors demonstrate that almost 94% of acute toxicity values assessed for freshwater, seawater and terrestrial systems using algae, annelid, arthropoda, bacteria, crustacea, fish, nematoda, plant, protozoa and rotatoria show that the nanoform of silver is less toxic than the dissolved metal (when normalised for total metal concentration).
In addition, a specific ecotoxicity testing programme designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms was undertaken following the silver substance evaluation. This demonstrated that the nanoform of silver is less toxic than ionic silver (based on EC10 and EC50 values for total, ‘conventional’ dissolved (<0.45μm) and ‘truly’ dissolved (<3kDa) silver). Therefore, taking the full body of evidence into account, the read-across use of toxicity values from ionic to nanosilver as a ‘worst case’ approach is justified and scientifically defensible for environmental endpoints. The Substance Evaluation Conclusion document for silver agreed with this conclusion for the nanosilver forms covered in the REACH dossier (RIVM 2018).

Long-term toxicity to aquatic invertebrates
Published data from two long-term toxicity to Daphnia magna studies using various sizes of nanoparticles and coating types are included in the REACH dossier as Endpoint Study Records. The nanosilver NOEC values from these studies are in the same order of magnitude or higher than the EC10 of 2.14 μg/L for Daphnia magna for ionic silver (Bianchini and Wood 2008). A summary of these supporting studies is available under Section 4.1.4 of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IUCLID Section 13). There is insufficient chronic data for aquatic invertebrates to investigate the influence of particle size of coating on the observed toxicity.
In addition, following the silver substance evaluation, Fraunhofer (Schlich et al. 2017) undertook comparative studies for the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and ionic silver (silver nitrate) with Daphnia magna, in accordance to OECD guideline 211. The most sensitive endpoint in both tests was reproduction after 21 days. The EC10 value for reproduction was 33.4 µg/L measured dissolved silver (<0.45 µm) for nanosilver and 3.49 µg/L measured dissolved silver (<0.45 µm) for silver nitrate. Thus, the directly comparable EC10 values reported by Schlich et al (2017) were an order of magnitude more toxic for ionic silver than for nanosilver. These studies are included in the REACH dossier as Endpoint Study Records and a summary is available in the report ‘Summary of Comparative Ecotoxicity Testing Programme for Nanosilver and Silver Nitrate (2017)’ (attached in IUCLID Section 13).
Together with the theoretical basis for read-across based on the free-ion, above studies support the use of ionic silver as the ‘worst case’ basis to read across properties to nanosilver.


For further information and data matrix see 'CSR Annex 9 - Read Across Justification Nanosilver ENV_SUMMARY_200706' attached in IUCLID section 13.

Reason / purpose for cross-reference:

read-across source

Duration:

14 d

Dose descriptor:

NOEC

Effect conc.:

1 µg/L

Nominal / measured:

not specified

Conc. based on:

dissolved

Basis for effect:

other: moult production

Details on results:

Survival NOEC 3.7 µg /L. LC85 = 5.8 µg/L. In the controls, the survival rate was 80% after 14 days. Within 14 days significant effects on Stenonema were found at ≥ 2.2 dissolved, 3.4 total Ag/L.

Results with reference substance (positive control):

Not applicable

Reported statistics and error estimates:

Larval survival was analysed at 7 and 14 days to yield NOECs using ANOVA and Dunnett’s Test. Survival data were transformed (arc-sine square root) before analysing by using a significance level of 0.05. Growth was analysed by examining 3 endpoints: A) molt or exuviae production at 7 and 14 days, B) body length after 14 days of exposure and C) head capsule width after 14 days exposure. Differences in the number of molts between treatments and controls were analysed by ANOVA and Dunnett’s test at a p value of 0.05. Treatments were compared to each other with either ANOVA and Tukey’s HSD test (if data satisfied parametric test assumptions) or Kruskal-Wallis nonparametric ANOVA and multiple means test. Differences in final body length and head capsule width between controls and treatments at 14 days were analysed with ANOVA and Bonferroni’s t test (p at 0.05) because these data met parametric test assumptions.

Validity criteria fulfilled:

not applicable

Conclusions:

14 day NOEC for molt production is 1.0 μg dissolved silver/L.

Executive summary:

The study presents the results of a subacute toxicity test with the mayfly (Stenonema modestum). A series of effluent toxicity tests were conducted, each consisting of a 14 day exposure using a static-renewal test mode. The study is considered to be reliable and suitable for use for this endpoint.

Reference 4

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

key study

Justification for type of information:

1. HYPOTHESIS FOR THE READ-ACROSS APPROACH (ENDPOINT LEVEL)
The REACH registration of silver (powder and massive forms of zero-valent, elemental, silver) is underpinned, in common with other metals, by a read-across (or analogue) approach from the properties of the free ion. This principle of read-across from the free ion has been extended to also include nanosilver.
The scientific validity of read-across from the hazard properties of ionic silver (source) to nanosilver (target) under REACH is underpinned by both theoretical and empirical considerations.
The theoretical basis for the use of ionic silver data as the foundation of the risk assessment of nanosilver is based on the premise that the free metal ion (Me+) is the most toxic metal form/species (Starodub et al. 1987). This consideration was implicit in the development of the Free Ion Activity Model [FIAM] (Morel 1983, Paquin et al. 2002, Campbell 1985, Brown and Markich 2000) and, more recently, the Biotic Ligand Model [BLM] (Paquin et al 2002, Niyogi and Wood 2004) that has underpinned the risk assessment of several metals (e.g. Cu, Ni, Zn) under the Existing Substance Regulations and REACH; and most recently the development of the Environmental Quality Standard (EQS) for nickel and nickel substances under the Water Framework Directive (WFD). When considered on an equal mass basis ionic silver would therefore be expected to have greater toxicity than nanosilver simply on the basis that silver ions are released over time from the surface of particles (via oxidative dissolution). As the properties of nanosilver are read-across directly from ionic silver (not just to the fraction of silver ions released from nanosilver), this read-across is also expected to introduce considerable precaution into the hazard component of the risk assessment of nanosilver as all nanosilver, irrespective of coating, morphology or particle size distribution is assumed to behave similarly to the free ion.
This theoretical consideration has been tested by conducting a comprehensive review of the available scientific literature for nanosilver, with particular emphasis on the comparative effects on REACH relevant biotic systems (REACH information requirement) of ionic silver and nanosilver. This review is described for each endpoint in subsequent sections of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IULCID Section 13) and is summarised below. Furthermore, this theoretical consideration was confirmed by a specific ecotoxicity testing programme undertaken by the EPMF following the silver substance evaluation and designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms.

2. READ-ACROSS APPROACH JUSTIFICATION (ENDPOINT LEVEL)
In terms of ecotoxicity, nanosilver on an equal mass basis has been found to be significantly less toxic than ionic silver for the majority of REACH endpoints and of equivalent toxicity for some others. None of the empirical information available suggests that nanosilver is consistently more hazardous than ionic silver on an equivalent mass basis, or that “nano-specific” effects that would prejudice the validity of read across from ionic silver to nanosilver occur. In addition, with very limited exceptions which are described further in the report ‘Nanosilver: read-across justification for environmental information requirements’, none of the available data suggested consistent relationships between particle morphology, size, particle size distribution (raw or agglomerated) or coating (surface treatment) and effects.
Notter et al. (2014) presents a meta-analysis of published EC50 values for ionic silver and nanosilver. The authors demonstrate that almost 94% of acute toxicity values assessed for freshwater, seawater and terrestrial systems using algae, annelid, arthropoda, bacteria, crustacea, fish, nematoda, plant, protozoa and rotatoria show that the nanoform of silver is less toxic than the dissolved metal (when normalised for total metal concentration).
In addition, a specific ecotoxicity testing programme designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms was undertaken following the silver substance evaluation. This demonstrated that the nanoform of silver is less toxic than ionic silver (based on EC10 and EC50 values for total, ‘conventional’ dissolved (<0.45μm) and ‘truly’ dissolved (<3kDa) silver). Therefore, taking the full body of evidence into account, the read-across use of toxicity values from ionic to nanosilver as a ‘worst case’ approach is justified and scientifically defensible for environmental endpoints. The Substance Evaluation Conclusion document for silver agreed with this conclusion for the nanosilver forms covered in the REACH dossier (RIVM 2018).

Long-term toxicity to aquatic invertebrates
Published data from two long-term toxicity to Daphnia magna studies using various sizes of nanoparticles and coating types are included in the REACH dossier as Endpoint Study Records. The nanosilver NOEC values from these studies are in the same order of magnitude or higher than the EC10 of 2.14 μg/L for Daphnia magna for ionic silver (Bianchini and Wood 2008). A summary of these supporting studies is available under Section 4.1.4 of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IUCLID Section 13). There is insufficient chronic data for aquatic invertebrates to investigate the influence of particle size of coating on the observed toxicity.
In addition, following the silver substance evaluation, Fraunhofer (Schlich et al. 2017) undertook comparative studies for the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and ionic silver (silver nitrate) with Daphnia magna, in accordance to OECD guideline 211. The most sensitive endpoint in both tests was reproduction after 21 days. The EC10 value for reproduction was 33.4 µg/L measured dissolved silver (<0.45 µm) for nanosilver and 3.49 µg/L measured dissolved silver (<0.45 µm) for silver nitrate. Thus, the directly comparable EC10 values reported by Schlich et al (2017) were an order of magnitude more toxic for ionic silver than for nanosilver. These studies are included in the REACH dossier as Endpoint Study Records and a summary is available in the report ‘Summary of Comparative Ecotoxicity Testing Programme for Nanosilver and Silver Nitrate (2017)’ (attached in IUCLID Section 13).
Together with the theoretical basis for read-across based on the free-ion, above studies support the use of ionic silver as the ‘worst case’ basis to read across properties to nanosilver.



For further information and data matrix see 'CSR Annex 9 - Read Across Justification Nanosilver ENV_SUMMARY_200706' attached in IUCLID section 13.

Reason / purpose for cross-reference:

read-across source

Duration:

20 d

Dose descriptor:

NOEC

Effect conc.:

0.31 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

other: Total silver

Basis for effect:

other: Moult production

Duration:

20 d

Dose descriptor:

EC10

Effect conc.:

1.48 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

other: Total silver

Basis for effect:

other: Moult production

Details on results:

The 20 day EC10 for Isonychia bicolor is 1.48 µg/L.

Results with reference substance (positive control):

Not applicable

Reported statistics and error estimates:

Results given as total silver. Recovery of silver from test solutions was 34 – 80% of theoretical values. Results based on measured total silver concentrations.

Results given as total silver. Stock solution was 85% of theoretical value such that concentrations are estimated to be 10 to 20% higher than expressed in paper, due to calculation methods. In addition sampling from the surface water layer would reduce analysis of precipitated or bound silver leading to further underestimation of silver levels within the test systems. The feeding regime is likely to have given rise to binding of silver and the consequences of dietary silver are unknown.

Validity criteria fulfilled:

not specified

Conclusions:

The 20 day EC10 for Isonychia bicolor is 1.48 µg/L.

Executive summary:

Although not a traditional endpoint it is indicative of effects on growth and valid for use as part of the overall dataset. Neither the weight nor length of individuals were sensitive indicators of sublethal toxicity and no effect was seen on survival. The percentage recovery of silver based on nominal concentrations ranged between 34 and 78% for unfiltered samples. The authors suggest that the method of sampling may also have led to low recovery of the test substance leading to an under estimate of exposure level. The losses of silver from the exposure solutions, although large, were well documented and the mean measured concentrations used to express the results are considered reasonable estimates of exposure.

Reference 5

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

key study

Justification for type of information:

1. HYPOTHESIS FOR THE READ-ACROSS APPROACH (ENDPOINT LEVEL)
The REACH registration of silver (powder and massive forms of zero-valent, elemental, silver) is underpinned, in common with other metals, by a read-across (or analogue) approach from the properties of the free ion. This principle of read-across from the free ion has been extended to also include nanosilver.
The scientific validity of read-across from the hazard properties of ionic silver (source) to nanosilver (target) under REACH is underpinned by both theoretical and empirical considerations.
The theoretical basis for the use of ionic silver data as the foundation of the risk assessment of nanosilver is based on the premise that the free metal ion (Me+) is the most toxic metal form/species (Starodub et al. 1987). This consideration was implicit in the development of the Free Ion Activity Model [FIAM] (Morel 1983, Paquin et al. 2002, Campbell 1985, Brown and Markich 2000) and, more recently, the Biotic Ligand Model [BLM] (Paquin et al 2002, Niyogi and Wood 2004) that has underpinned the risk assessment of several metals (e.g. Cu, Ni, Zn) under the Existing Substance Regulations and REACH; and most recently the development of the Environmental Quality Standard (EQS) for nickel and nickel substances under the Water Framework Directive (WFD). When considered on an equal mass basis ionic silver would therefore be expected to have greater toxicity than nanosilver simply on the basis that silver ions are released over time from the surface of particles (via oxidative dissolution). As the properties of nanosilver are read-across directly from ionic silver (not just to the fraction of silver ions released from nanosilver), this read-across is also expected to introduce considerable precaution into the hazard component of the risk assessment of nanosilver as all nanosilver, irrespective of coating, morphology or particle size distribution is assumed to behave similarly to the free ion.
This theoretical consideration has been tested by conducting a comprehensive review of the available scientific literature for nanosilver, with particular emphasis on the comparative effects on REACH relevant biotic systems (REACH information requirement) of ionic silver and nanosilver. This review is described for each endpoint in subsequent sections of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IULCID Section 13) and is summarised below. Furthermore, this theoretical consideration was confirmed by a specific ecotoxicity testing programme undertaken by the EPMF following the silver substance evaluation and designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms.

2. READ-ACROSS APPROACH JUSTIFICATION (ENDPOINT LEVEL)
In terms of ecotoxicity, nanosilver on an equal mass basis has been found to be significantly less toxic than ionic silver for the majority of REACH endpoints and of equivalent toxicity for some others. None of the empirical information available suggests that nanosilver is consistently more hazardous than ionic silver on an equivalent mass basis, or that “nano-specific” effects that would prejudice the validity of read across from ionic silver to nanosilver occur. In addition, with very limited exceptions which are described further in the report ‘Nanosilver: read-across justification for environmental information requirements’, none of the available data suggested consistent relationships between particle morphology, size, particle size distribution (raw or agglomerated) or coating (surface treatment) and effects.
Notter et al. (2014) presents a meta-analysis of published EC50 values for ionic silver and nanosilver. The authors demonstrate that almost 94% of acute toxicity values assessed for freshwater, seawater and terrestrial systems using algae, annelid, arthropoda, bacteria, crustacea, fish, nematoda, plant, protozoa and rotatoria show that the nanoform of silver is less toxic than the dissolved metal (when normalised for total metal concentration).
In addition, a specific ecotoxicity testing programme designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms was undertaken following the silver substance evaluation. This demonstrated that the nanoform of silver is less toxic than ionic silver (based on EC10 and EC50 values for total, ‘conventional’ dissolved (<0.45μm) and ‘truly’ dissolved (<3kDa) silver). Therefore, taking the full body of evidence into account, the read-across use of toxicity values from ionic to nanosilver as a ‘worst case’ approach is justified and scientifically defensible for environmental endpoints. The Substance Evaluation Conclusion document for silver agreed with this conclusion for the nanosilver forms covered in the REACH dossier (RIVM 2018).

Long-term toxicity to aquatic invertebrates
Published data from two long-term toxicity to Daphnia magna studies using various sizes of nanoparticles and coating types are included in the REACH dossier as Endpoint Study Records. The nanosilver NOEC values from these studies are in the same order of magnitude or higher than the EC10 of 2.14 μg/L for Daphnia magna for ionic silver (Bianchini and Wood 2008). A summary of these supporting studies is available under Section 4.1.4 of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IUCLID Section 13). There is insufficient chronic data for aquatic invertebrates to investigate the influence of particle size of coating on the observed toxicity.
In addition, following the silver substance evaluation, Fraunhofer (Schlich et al. 2017) undertook comparative studies for the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and ionic silver (silver nitrate) with Daphnia magna, in accordance to OECD guideline 211. The most sensitive endpoint in both tests was reproduction after 21 days. The EC10 value for reproduction was 33.4 µg/L measured dissolved silver (<0.45 µm) for nanosilver and 3.49 µg/L measured dissolved silver (<0.45 µm) for silver nitrate. Thus, the directly comparable EC10 values reported by Schlich et al (2017) were an order of magnitude more toxic for ionic silver than for nanosilver. These studies are included in the REACH dossier as Endpoint Study Records and a summary is available in the report ‘Summary of Comparative Ecotoxicity Testing Programme for Nanosilver and Silver Nitrate (2017)’ (attached in IUCLID Section 13).
Together with the theoretical basis for read-across based on the free-ion, above studies support the use of ionic silver as the ‘worst case’ basis to read across properties to nanosilver.



For further information and data matrix see 'CSR Annex 9 - Read Across Justification Nanosilver ENV_SUMMARY_200706' attached in IUCLID section 13.

Reason / purpose for cross-reference:

read-across source

Duration:

21 d

Dose descriptor:

NOEC

Effect conc.:

2.6 µg/L

Nominal / measured:

not specified

Conc. based on:

other: total silver

Basis for effect:

growth

Duration:

21 d

Dose descriptor:

EC10

Effect conc.:

1.68 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

other: Total silver

Basis for effect:

growth

Remarks:

(weight)

Details on results:

21 day NOEC growth (weight and length) is 2.6 μg total silver/L and the 21 day EC10 for growth (weight) is 1.68 µg/L.

Results with reference substance (positive control):

Not applicable

Reported statistics and error estimates:

Results given as total silver. Recovery of silver from test solutions was 34 – 80% of theoretical values. Results based on measured total silver concentrations.

Results given as total silver. Stock solution was 85% of theoretical value such that concentrations are estimated to be 10 to 20% higher than expressed in paper, due to calculation methods. In addition sampling from the surface water layer would reduce analysis of precipitated or bound silver leading to further underestimation of silver levels within the test systems. The feeding regime is likely to have given rise to binding of silver and the consequences of dietary silver are unknown.

Validity criteria fulfilled:

yes

Conclusions:

21 day NOEC growth (weight and length) is 2.6 μg total silver/L and the 21 day EC10 for growth (weight) is 1.68 µg/L.

Executive summary:

This study is a non-GLP, non-guideline study in which juvenile Corbicula were exposed in a 21 day static-renewal growth test. Minimum methodology is given in the paper, but references are given. The study is considered to be reliable and suitable for use for this endpoint.

Reference 6

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

key study

Study period:

Not reported

Justification for type of information:

1. HYPOTHESIS FOR THE READ-ACROSS APPROACH (ENDPOINT LEVEL)
The REACH registration of silver (powder and massive forms of zero-valent, elemental, silver) is underpinned, in common with other metals, by a read-across (or analogue) approach from the properties of the free ion. This principle of read-across from the free ion has been extended to also include nanosilver.
The scientific validity of read-across from the hazard properties of ionic silver (source) to nanosilver (target) under REACH is underpinned by both theoretical and empirical considerations.
The theoretical basis for the use of ionic silver data as the foundation of the risk assessment of nanosilver is based on the premise that the free metal ion (Me+) is the most toxic metal form/species (Starodub et al. 1987). This consideration was implicit in the development of the Free Ion Activity Model [FIAM] (Morel 1983, Paquin et al. 2002, Campbell 1985, Brown and Markich 2000) and, more recently, the Biotic Ligand Model [BLM] (Paquin et al 2002, Niyogi and Wood 2004) that has underpinned the risk assessment of several metals (e.g. Cu, Ni, Zn) under the Existing Substance Regulations and REACH; and most recently the development of the Environmental Quality Standard (EQS) for nickel and nickel substances under the Water Framework Directive (WFD). When considered on an equal mass basis ionic silver would therefore be expected to have greater toxicity than nanosilver simply on the basis that silver ions are released over time from the surface of particles (via oxidative dissolution). As the properties of nanosilver are read-across directly from ionic silver (not just to the fraction of silver ions released from nanosilver), this read-across is also expected to introduce considerable precaution into the hazard component of the risk assessment of nanosilver as all nanosilver, irrespective of coating, morphology or particle size distribution is assumed to behave similarly to the free ion.
This theoretical consideration has been tested by conducting a comprehensive review of the available scientific literature for nanosilver, with particular emphasis on the comparative effects on REACH relevant biotic systems (REACH information requirement) of ionic silver and nanosilver. This review is described for each endpoint in subsequent sections of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IULCID Section 13) and is summarised below. Furthermore, this theoretical consideration was confirmed by a specific ecotoxicity testing programme undertaken by the EPMF following the silver substance evaluation and designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms.

2. READ-ACROSS APPROACH JUSTIFICATION (ENDPOINT LEVEL)
In terms of ecotoxicity, nanosilver on an equal mass basis has been found to be significantly less toxic than ionic silver for the majority of REACH endpoints and of equivalent toxicity for some others. None of the empirical information available suggests that nanosilver is consistently more hazardous than ionic silver on an equivalent mass basis, or that “nano-specific” effects that would prejudice the validity of read across from ionic silver to nanosilver occur. In addition, with very limited exceptions which are described further in the report ‘Nanosilver: read-across justification for environmental information requirements’, none of the available data suggested consistent relationships between particle morphology, size, particle size distribution (raw or agglomerated) or coating (surface treatment) and effects.
Notter et al. (2014) presents a meta-analysis of published EC50 values for ionic silver and nanosilver. The authors demonstrate that almost 94% of acute toxicity values assessed for freshwater, seawater and terrestrial systems using algae, annelid, arthropoda, bacteria, crustacea, fish, nematoda, plant, protozoa and rotatoria show that the nanoform of silver is less toxic than the dissolved metal (when normalised for total metal concentration).
In addition, a specific ecotoxicity testing programme designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms was undertaken following the silver substance evaluation. This demonstrated that the nanoform of silver is less toxic than ionic silver (based on EC10 and EC50 values for total, ‘conventional’ dissolved (<0.45μm) and ‘truly’ dissolved (<3kDa) silver). Therefore, taking the full body of evidence into account, the read-across use of toxicity values from ionic to nanosilver as a ‘worst case’ approach is justified and scientifically defensible for environmental endpoints. The Substance Evaluation Conclusion document for silver agreed with this conclusion for the nanosilver forms covered in the REACH dossier (RIVM 2018).

Long-term toxicity to aquatic invertebrates
Published data from two long-term toxicity to Daphnia magna studies using various sizes of nanoparticles and coating types are included in the REACH dossier as Endpoint Study Records. The nanosilver NOEC values from these studies are in the same order of magnitude or higher than the EC10 of 2.14 μg/L for Daphnia magna for ionic silver (Bianchini and Wood 2008). A summary of these supporting studies is available under Section 4.1.4 of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IUCLID Section 13). There is insufficient chronic data for aquatic invertebrates to investigate the influence of particle size of coating on the observed toxicity.
In addition, following the silver substance evaluation, Fraunhofer (Schlich et al. 2017) undertook comparative studies for the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and ionic silver (silver nitrate) with Daphnia magna, in accordance to OECD guideline 211. The most sensitive endpoint in both tests was reproduction after 21 days. The EC10 value for reproduction was 33.4 µg/L measured dissolved silver (<0.45 µm) for nanosilver and 3.49 µg/L measured dissolved silver (<0.45 µm) for silver nitrate. Thus, the directly comparable EC10 values reported by Schlich et al (2017) were an order of magnitude more toxic for ionic silver than for nanosilver. These studies are included in the REACH dossier as Endpoint Study Records and a summary is available in the report ‘Summary of Comparative Ecotoxicity Testing Programme for Nanosilver and Silver Nitrate (2017)’ (attached in IUCLID Section 13).
Together with the theoretical basis for read-across based on the free-ion, above studies support the use of ionic silver as the ‘worst case’ basis to read across properties to nanosilver.


For further information and data matrix see 'CSR Annex 9 - Read Across Justification Nanosilver ENV_SUMMARY_200706' attached in IUCLID section 13.

Reason / purpose for cross-reference:

read-across source

Duration:

7 d

Dose descriptor:

EC50

Effect conc.:

0.8 µg/L

Nominal / measured:

not specified

Conc. based on:

not specified

Basis for effect:

reproduction

Remarks on result:

other: 95% 0.13-1.9

Duration:

7 d

Dose descriptor:

EC50

Effect conc.:

6.4 µg/L

Nominal / measured:

not specified

Conc. based on:

not specified

Basis for effect:

mortality

Duration:

7 d

Dose descriptor:

NOEC

Effect conc.:

1 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

not specified

Basis for effect:

reproduction

Details on results:

Authors state dilution water control populations in chronic tests met quality requirements as follows; first brook released within 7 to 9 d; three broods released by ceriodaphnia by day 7; a minimum of 20 young per adult released wihtin 14 d; none of the original organisms died; no ephippia were observed.

Results with reference substance (positive control):

Not applicable

Reported statistics and error estimates:

Chronic (7 and 14 day) EC50 values and 95% confidence limits were calculated using either Probit analysis or the moving average angle method. Reproductive and survival data were analysed by analysis of variance. Statements of statistical significance refer to p = 0.05 or less.

Conclusions:

The 7d NOEC is 1 µg/L for reproduction.

Executive summary:

This is a non-GLP, renewal chronic test conducted with Daphnia magna, Daphnia pulex and Ceriodaphnia reticulate to determine their relative sensitivities to silver. The organisms were exposed to silver for 7 days and the reproduction and mortality were recorded.

Reference 7

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

key study

Justification for type of information:

1. HYPOTHESIS FOR THE READ-ACROSS APPROACH (ENDPOINT LEVEL)
The REACH registration of silver (powder and massive forms of zero-valent, elemental, silver) is underpinned, in common with other metals, by a read-across (or analogue) approach from the properties of the free ion. This principle of read-across from the free ion has been extended to also include nanosilver.
The scientific validity of read-across from the hazard properties of ionic silver (source) to nanosilver (target) under REACH is underpinned by both theoretical and empirical considerations.
The theoretical basis for the use of ionic silver data as the foundation of the risk assessment of nanosilver is based on the premise that the free metal ion (Me+) is the most toxic metal form/species (Starodub et al. 1987). This consideration was implicit in the development of the Free Ion Activity Model [FIAM] (Morel 1983, Paquin et al. 2002, Campbell 1985, Brown and Markich 2000) and, more recently, the Biotic Ligand Model [BLM] (Paquin et al 2002, Niyogi and Wood 2004) that has underpinned the risk assessment of several metals (e.g. Cu, Ni, Zn) under the Existing Substance Regulations and REACH; and most recently the development of the Environmental Quality Standard (EQS) for nickel and nickel substances under the Water Framework Directive (WFD). When considered on an equal mass basis ionic silver would therefore be expected to have greater toxicity than nanosilver simply on the basis that silver ions are released over time from the surface of particles (via oxidative dissolution). As the properties of nanosilver are read-across directly from ionic silver (not just to the fraction of silver ions released from nanosilver), this read-across is also expected to introduce considerable precaution into the hazard component of the risk assessment of nanosilver as all nanosilver, irrespective of coating, morphology or particle size distribution is assumed to behave similarly to the free ion.
This theoretical consideration has been tested by conducting a comprehensive review of the available scientific literature for nanosilver, with particular emphasis on the comparative effects on REACH relevant biotic systems (REACH information requirement) of ionic silver and nanosilver. This review is described for each endpoint in subsequent sections of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IULCID Section 13) and is summarised below. Furthermore, this theoretical consideration was confirmed by a specific ecotoxicity testing programme undertaken by the EPMF following the silver substance evaluation and designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms.

2. READ-ACROSS APPROACH JUSTIFICATION (ENDPOINT LEVEL)
In terms of ecotoxicity, nanosilver on an equal mass basis has been found to be significantly less toxic than ionic silver for the majority of REACH endpoints and of equivalent toxicity for some others. None of the empirical information available suggests that nanosilver is consistently more hazardous than ionic silver on an equivalent mass basis, or that “nano-specific” effects that would prejudice the validity of read across from ionic silver to nanosilver occur. In addition, with very limited exceptions which are described further in the report ‘Nanosilver: read-across justification for environmental information requirements’, none of the available data suggested consistent relationships between particle morphology, size, particle size distribution (raw or agglomerated) or coating (surface treatment) and effects.
Notter et al. (2014) presents a meta-analysis of published EC50 values for ionic silver and nanosilver. The authors demonstrate that almost 94% of acute toxicity values assessed for freshwater, seawater and terrestrial systems using algae, annelid, arthropoda, bacteria, crustacea, fish, nematoda, plant, protozoa and rotatoria show that the nanoform of silver is less toxic than the dissolved metal (when normalised for total metal concentration).
In addition, a specific ecotoxicity testing programme designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms was undertaken following the silver substance evaluation. This demonstrated that the nanoform of silver is less toxic than ionic silver (based on EC10 and EC50 values for total, ‘conventional’ dissolved (<0.45μm) and ‘truly’ dissolved (<3kDa) silver). Therefore, taking the full body of evidence into account, the read-across use of toxicity values from ionic to nanosilver as a ‘worst case’ approach is justified and scientifically defensible for environmental endpoints. The Substance Evaluation Conclusion document for silver agreed with this conclusion for the nanosilver forms covered in the REACH dossier (RIVM 2018).

Long-term toxicity to aquatic invertebrates
Published data from two long-term toxicity to Daphnia magna studies using various sizes of nanoparticles and coating types are included in the REACH dossier as Endpoint Study Records. The nanosilver NOEC values from these studies are in the same order of magnitude or higher than the EC10 of 2.14 μg/L for Daphnia magna for ionic silver (Bianchini and Wood 2008). A summary of these supporting studies is available under Section 4.1.4 of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IUCLID Section 13). There is insufficient chronic data for aquatic invertebrates to investigate the influence of particle size of coating on the observed toxicity.
In addition, following the silver substance evaluation, Fraunhofer (Schlich et al. 2017) undertook comparative studies for the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and ionic silver (silver nitrate) with Daphnia magna, in accordance to OECD guideline 211. The most sensitive endpoint in both tests was reproduction after 21 days. The EC10 value for reproduction was 33.4 µg/L measured dissolved silver (<0.45 µm) for nanosilver and 3.49 µg/L measured dissolved silver (<0.45 µm) for silver nitrate. Thus, the directly comparable EC10 values reported by Schlich et al (2017) were an order of magnitude more toxic for ionic silver than for nanosilver. These studies are included in the REACH dossier as Endpoint Study Records and a summary is available in the report ‘Summary of Comparative Ecotoxicity Testing Programme for Nanosilver and Silver Nitrate (2017)’ (attached in IUCLID Section 13).
Together with the theoretical basis for read-across based on the free-ion, above studies support the use of ionic silver as the ‘worst case’ basis to read across properties to nanosilver.


For further information and data matrix see 'CSR Annex 9 - Read Across Justification Nanosilver ENV_SUMMARY_200706' attached in IUCLID section 13.

Reason / purpose for cross-reference:

read-across source

Duration:

21 d

Dose descriptor:

NOEC

Effect conc.:

0.32 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

element

Basis for effect:

mortality

Details on results:

Control survival of D. magna for the Ag+experiment was 100 %.
The measured dissolved Ag+ 21 day EC50 value was 0.81(SD±0.11) µgAg+/L. The LOEC and NOEC values of the Ag+ study were 0.63 µg/L and 0.32µgAg+/L, respectively.
Analysis of 21-day mean clutch indicated no significant differences from the control for both the NaCl and Ag+ experiments (p>0.05). A general dose response relationship as determined by mean clutch size across broods was apparent for treatments of Ag+, no broods were found to be significantly different from each other or controls (p>0.05).

Results with reference substance (positive control):

Not applicable

Reported statistics and error estimates:

A one-way analysis of variance (ANOVA) was used in Sigma Plot version 11.0 (Systat Software, Inc., San Jose, CA, USA) to identify differences between treatment levels and control (a=0.05). ANOVA assumptions of residual normality and homogenous variances were assessed and confirmed using the Shapiro–Wilk test. Thereafter, Dunnett’s test (a=0.05) was used to compare means of each treatment level to respective control means. Each treatment level of the 21 day chronic toxicity experiments of NaCl and Ag+ were compared to the control (n=10) for the endpoints of lethality, reproduction, and mean clutch size.

Validity criteria fulfilled:

not specified

Conclusions:

The 21 day NOEC for Daphnia magna exposed to dissolved silver was 0.32 µg Ag+/L based on mortality.

Executive summary:

The toxicity of dissolved silver ions to Daphnia magna neonates was determined in a non-GLP, static-renewal chronic toxicity study following USEPA guideline 850.1300 (1996). D. magna 21 day studies were conducted with model stressors of sodium chloride and dissolved silver and the 21 day NOEC for Daphnia magna exposed to dissolved silver was 0.32 µg Ag+/L based on mortality.

Reference 8

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

key study

Justification for type of information:

1. HYPOTHESIS FOR THE READ-ACROSS APPROACH (ENDPOINT LEVEL)
The REACH registration of silver (powder and massive forms of zero-valent, elemental, silver) is underpinned, in common with other metals, by a read-across (or analogue) approach from the properties of the free ion. This principle of read-across from the free ion has been extended to also include nanosilver.
The scientific validity of read-across from the hazard properties of ionic silver (source) to nanosilver (target) under REACH is underpinned by both theoretical and empirical considerations.
The theoretical basis for the use of ionic silver data as the foundation of the risk assessment of nanosilver is based on the premise that the free metal ion (Me+) is the most toxic metal form/species (Starodub et al. 1987). This consideration was implicit in the development of the Free Ion Activity Model [FIAM] (Morel 1983, Paquin et al. 2002, Campbell 1985, Brown and Markich 2000) and, more recently, the Biotic Ligand Model [BLM] (Paquin et al 2002, Niyogi and Wood 2004) that has underpinned the risk assessment of several metals (e.g. Cu, Ni, Zn) under the Existing Substance Regulations and REACH; and most recently the development of the Environmental Quality Standard (EQS) for nickel and nickel substances under the Water Framework Directive (WFD). When considered on an equal mass basis ionic silver would therefore be expected to have greater toxicity than nanosilver simply on the basis that silver ions are released over time from the surface of particles (via oxidative dissolution). As the properties of nanosilver are read-across directly from ionic silver (not just to the fraction of silver ions released from nanosilver), this read-across is also expected to introduce considerable precaution into the hazard component of the risk assessment of nanosilver as all nanosilver, irrespective of coating, morphology or particle size distribution is assumed to behave similarly to the free ion.
This theoretical consideration has been tested by conducting a comprehensive review of the available scientific literature for nanosilver, with particular emphasis on the comparative effects on REACH relevant biotic systems (REACH information requirement) of ionic silver and nanosilver. This review is described for each endpoint in subsequent sections of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IULCID Section 13) and is summarised below. Furthermore, this theoretical consideration was confirmed by a specific ecotoxicity testing programme undertaken by the EPMF following the silver substance evaluation and designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms.

2. READ-ACROSS APPROACH JUSTIFICATION (ENDPOINT LEVEL)
In terms of ecotoxicity, nanosilver on an equal mass basis has been found to be significantly less toxic than ionic silver for the majority of REACH endpoints and of equivalent toxicity for some others. None of the empirical information available suggests that nanosilver is consistently more hazardous than ionic silver on an equivalent mass basis, or that “nano-specific” effects that would prejudice the validity of read across from ionic silver to nanosilver occur. In addition, with very limited exceptions which are described further in the report ‘Nanosilver: read-across justification for environmental information requirements’, none of the available data suggested consistent relationships between particle morphology, size, particle size distribution (raw or agglomerated) or coating (surface treatment) and effects.
Notter et al. (2014) presents a meta-analysis of published EC50 values for ionic silver and nanosilver. The authors demonstrate that almost 94% of acute toxicity values assessed for freshwater, seawater and terrestrial systems using algae, annelid, arthropoda, bacteria, crustacea, fish, nematoda, plant, protozoa and rotatoria show that the nanoform of silver is less toxic than the dissolved metal (when normalised for total metal concentration).
In addition, a specific ecotoxicity testing programme designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms was undertaken following the silver substance evaluation. This demonstrated that the nanoform of silver is less toxic than ionic silver (based on EC10 and EC50 values for total, ‘conventional’ dissolved (<0.45μm) and ‘truly’ dissolved (<3kDa) silver). Therefore, taking the full body of evidence into account, the read-across use of toxicity values from ionic to nanosilver as a ‘worst case’ approach is justified and scientifically defensible for environmental endpoints. The Substance Evaluation Conclusion document for silver agreed with this conclusion for the nanosilver forms covered in the REACH dossier (RIVM 2018).

Long-term toxicity to aquatic invertebrates
Published data from two long-term toxicity to Daphnia magna studies using various sizes of nanoparticles and coating types are included in the REACH dossier as Endpoint Study Records. The nanosilver NOEC values from these studies are in the same order of magnitude or higher than the EC10 of 2.14 μg/L for Daphnia magna for ionic silver (Bianchini and Wood 2008). A summary of these supporting studies is available under Section 4.1.4 of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IUCLID Section 13). There is insufficient chronic data for aquatic invertebrates to investigate the influence of particle size of coating on the observed toxicity.
In addition, following the silver substance evaluation, Fraunhofer (Schlich et al. 2017) undertook comparative studies for the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and ionic silver (silver nitrate) with Daphnia magna, in accordance to OECD guideline 211. The most sensitive endpoint in both tests was reproduction after 21 days. The EC10 value for reproduction was 33.4 µg/L measured dissolved silver (<0.45 µm) for nanosilver and 3.49 µg/L measured dissolved silver (<0.45 µm) for silver nitrate. Thus, the directly comparable EC10 values reported by Schlich et al (2017) were an order of magnitude more toxic for ionic silver than for nanosilver. These studies are included in the REACH dossier as Endpoint Study Records and a summary is available in the report ‘Summary of Comparative Ecotoxicity Testing Programme for Nanosilver and Silver Nitrate (2017)’ (attached in IUCLID Section 13).
Together with the theoretical basis for read-across based on the free-ion, above studies support the use of ionic silver as the ‘worst case’ basis to read across properties to nanosilver.


For further information and data matrix see 'CSR Annex 9 - Read Across Justification Nanosilver ENV_SUMMARY_200706' attached in IUCLID section 13.

Reason / purpose for cross-reference:

read-across source

Duration:

7 d

Dose descriptor:

EC10

Effect conc.:

2.48 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

not specified

Basis for effect:

reproduction

Details on results:

The 7 day reproduction EC10 is 2.48 µg/L

Results with reference substance (positive control):

Not applicable

Reported statistics and error estimates:

No data reported

Validity criteria fulfilled:

yes

Conclusions:

The 7 day reproduction EC10 is 2.48 µg/L

Executive summary:

This is a non-GLP, long term toxicity study on Ceriodaphnia dubia following US-EPA (2002) guideline "Short-term methods for estimating the chronic toxicity of effluents and receiving waters to freshwater organisms". The study is considered reliable and suitable for use for this endpoint.

Reference 9

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

key study

Justification for type of information:

1. HYPOTHESIS FOR THE READ-ACROSS APPROACH (ENDPOINT LEVEL)
The REACH registration of silver (powder and massive forms of zero-valent, elemental, silver) is underpinned, in common with other metals, by a read-across (or analogue) approach from the properties of the free ion. This principle of read-across from the free ion has been extended to also include nanosilver.
The scientific validity of read-across from the hazard properties of ionic silver (source) to nanosilver (target) under REACH is underpinned by both theoretical and empirical considerations.
The theoretical basis for the use of ionic silver data as the foundation of the risk assessment of nanosilver is based on the premise that the free metal ion (Me+) is the most toxic metal form/species (Starodub et al. 1987). This consideration was implicit in the development of the Free Ion Activity Model [FIAM] (Morel 1983, Paquin et al. 2002, Campbell 1985, Brown and Markich 2000) and, more recently, the Biotic Ligand Model [BLM] (Paquin et al 2002, Niyogi and Wood 2004) that has underpinned the risk assessment of several metals (e.g. Cu, Ni, Zn) under the Existing Substance Regulations and REACH; and most recently the development of the Environmental Quality Standard (EQS) for nickel and nickel substances under the Water Framework Directive (WFD). When considered on an equal mass basis ionic silver would therefore be expected to have greater toxicity than nanosilver simply on the basis that silver ions are released over time from the surface of particles (via oxidative dissolution). As the properties of nanosilver are read-across directly from ionic silver (not just to the fraction of silver ions released from nanosilver), this read-across is also expected to introduce considerable precaution into the hazard component of the risk assessment of nanosilver as all nanosilver, irrespective of coating, morphology or particle size distribution is assumed to behave similarly to the free ion.
This theoretical consideration has been tested by conducting a comprehensive review of the available scientific literature for nanosilver, with particular emphasis on the comparative effects on REACH relevant biotic systems (REACH information requirement) of ionic silver and nanosilver. This review is described for each endpoint in subsequent sections of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IULCID Section 13) and is summarised below. Furthermore, this theoretical consideration was confirmed by a specific ecotoxicity testing programme undertaken by the EPMF following the silver substance evaluation and designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms.

2. READ-ACROSS APPROACH JUSTIFICATION (ENDPOINT LEVEL)
In terms of ecotoxicity, nanosilver on an equal mass basis has been found to be significantly less toxic than ionic silver for the majority of REACH endpoints and of equivalent toxicity for some others. None of the empirical information available suggests that nanosilver is consistently more hazardous than ionic silver on an equivalent mass basis, or that “nano-specific” effects that would prejudice the validity of read across from ionic silver to nanosilver occur. In addition, with very limited exceptions which are described further in the report ‘Nanosilver: read-across justification for environmental information requirements’, none of the available data suggested consistent relationships between particle morphology, size, particle size distribution (raw or agglomerated) or coating (surface treatment) and effects.
Notter et al. (2014) presents a meta-analysis of published EC50 values for ionic silver and nanosilver. The authors demonstrate that almost 94% of acute toxicity values assessed for freshwater, seawater and terrestrial systems using algae, annelid, arthropoda, bacteria, crustacea, fish, nematoda, plant, protozoa and rotatoria show that the nanoform of silver is less toxic than the dissolved metal (when normalised for total metal concentration).
In addition, a specific ecotoxicity testing programme designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms was undertaken following the silver substance evaluation. This demonstrated that the nanoform of silver is less toxic than ionic silver (based on EC10 and EC50 values for total, ‘conventional’ dissolved (<0.45μm) and ‘truly’ dissolved (<3kDa) silver). Therefore, taking the full body of evidence into account, the read-across use of toxicity values from ionic to nanosilver as a ‘worst case’ approach is justified and scientifically defensible for environmental endpoints. The Substance Evaluation Conclusion document for silver agreed with this conclusion for the nanosilver forms covered in the REACH dossier (RIVM 2018).

Long-term toxicity to aquatic invertebrates
Published data from two long-term toxicity to Daphnia magna studies using various sizes of nanoparticles and coating types are included in the REACH dossier as Endpoint Study Records. The nanosilver NOEC values from these studies are in the same order of magnitude or higher than the EC10 of 2.14 μg/L for Daphnia magna for ionic silver (Bianchini and Wood 2008). A summary of these supporting studies is available under Section 4.1.4 of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IUCLID Section 13). There is insufficient chronic data for aquatic invertebrates to investigate the influence of particle size of coating on the observed toxicity.
In addition, following the silver substance evaluation, Fraunhofer (Schlich et al. 2017) undertook comparative studies for the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and ionic silver (silver nitrate) with Daphnia magna, in accordance to OECD guideline 211. The most sensitive endpoint in both tests was reproduction after 21 days. The EC10 value for reproduction was 33.4 µg/L measured dissolved silver (<0.45 µm) for nanosilver and 3.49 µg/L measured dissolved silver (<0.45 µm) for silver nitrate. Thus, the directly comparable EC10 values reported by Schlich et al (2017) were an order of magnitude more toxic for ionic silver than for nanosilver. These studies are included in the REACH dossier as Endpoint Study Records and a summary is available in the report ‘Summary of Comparative Ecotoxicity Testing Programme for Nanosilver and Silver Nitrate (2017)’ (attached in IUCLID Section 13).
Together with the theoretical basis for read-across based on the free-ion, above studies support the use of ionic silver as the ‘worst case’ basis to read across properties to nanosilver.


For further information and data matrix see 'CSR Annex 9 - Read Across Justification Nanosilver ENV_SUMMARY_200706' attached in IUCLID section 13.

Reason / purpose for cross-reference:

read-across source

Duration:

7 d

Dose descriptor:

EC10

Effect conc.:

10.1 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

dissolved

Basis for effect:

reproduction

Duration:

7 d

Dose descriptor:

EC10

Effect conc.:

6.48 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

dissolved

Basis for effect:

reproduction

Duration:

7 d

Dose descriptor:

EC10

Effect conc.:

8.69 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

dissolved

Basis for effect:

reproduction

Remarks on result:

other: Additional dissolved organic carbon replicate

Details on results:

The 7 days reproduction EC10 for Ceriodaphnia dubia is 6.48 to 10.1 µg/L for dissolved organic carbon of 4.74 mg/L and 8.69 µg/L for dissolved organic carbon of 5.11 mg/L.

Results with reference substance (positive control):

Not applicable

Reported statistics and error estimates:

Survival for C. dubia studies was analysed using Fisher’s Exact Test to compare treatment group survival to control survival (p≤0.05) or LC50 methods. Normality and homogeneity assumptions for reproductive data were evaluated using Shapiro-Wilk’s or Chi-Square Test and Bartlett’s Test respectively (p≤0.01).

Validity criteria fulfilled:

yes

Conclusions:

The 7 day reproduction EC10 for Ceriodaphnia dubia is 6.48 to 10.1 µg/L for dissolved organic carbon of 4.74 mg/L and 8.69 µg/L for dissolved organic carbon of 5.11 mg/L.

Reference 10

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

key study

Justification for type of information:

1. HYPOTHESIS FOR THE READ-ACROSS APPROACH (ENDPOINT LEVEL)
The REACH registration of silver (powder and massive forms of zero-valent, elemental, silver) is underpinned, in common with other metals, by a read-across (or analogue) approach from the properties of the free ion. This principle of read-across from the free ion has been extended to also include nanosilver.
The scientific validity of read-across from the hazard properties of ionic silver (source) to nanosilver (target) under REACH is underpinned by both theoretical and empirical considerations.
The theoretical basis for the use of ionic silver data as the foundation of the risk assessment of nanosilver is based on the premise that the free metal ion (Me+) is the most toxic metal form/species (Starodub et al. 1987). This consideration was implicit in the development of the Free Ion Activity Model [FIAM] (Morel 1983, Paquin et al. 2002, Campbell 1985, Brown and Markich 2000) and, more recently, the Biotic Ligand Model [BLM] (Paquin et al 2002, Niyogi and Wood 2004) that has underpinned the risk assessment of several metals (e.g. Cu, Ni, Zn) under the Existing Substance Regulations and REACH; and most recently the development of the Environmental Quality Standard (EQS) for nickel and nickel substances under the Water Framework Directive (WFD). When considered on an equal mass basis ionic silver would therefore be expected to have greater toxicity than nanosilver simply on the basis that silver ions are released over time from the surface of particles (via oxidative dissolution). As the properties of nanosilver are read-across directly from ionic silver (not just to the fraction of silver ions released from nanosilver), this read-across is also expected to introduce considerable precaution into the hazard component of the risk assessment of nanosilver as all nanosilver, irrespective of coating, morphology or particle size distribution is assumed to behave similarly to the free ion.
This theoretical consideration has been tested by conducting a comprehensive review of the available scientific literature for nanosilver, with particular emphasis on the comparative effects on REACH relevant biotic systems (REACH information requirement) of ionic silver and nanosilver. This review is described for each endpoint in subsequent sections of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IULCID Section 13) and is summarised below. Furthermore, this theoretical consideration was confirmed by a specific ecotoxicity testing programme undertaken by the EPMF following the silver substance evaluation and designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms.

2. READ-ACROSS APPROACH JUSTIFICATION (ENDPOINT LEVEL)
In terms of ecotoxicity, nanosilver on an equal mass basis has been found to be significantly less toxic than ionic silver for the majority of REACH endpoints and of equivalent toxicity for some others. None of the empirical information available suggests that nanosilver is consistently more hazardous than ionic silver on an equivalent mass basis, or that “nano-specific” effects that would prejudice the validity of read across from ionic silver to nanosilver occur. In addition, with very limited exceptions which are described further in the report ‘Nanosilver: read-across justification for environmental information requirements’, none of the available data suggested consistent relationships between particle morphology, size, particle size distribution (raw or agglomerated) or coating (surface treatment) and effects.
Notter et al. (2014) presents a meta-analysis of published EC50 values for ionic silver and nanosilver. The authors demonstrate that almost 94% of acute toxicity values assessed for freshwater, seawater and terrestrial systems using algae, annelid, arthropoda, bacteria, crustacea, fish, nematoda, plant, protozoa and rotatoria show that the nanoform of silver is less toxic than the dissolved metal (when normalised for total metal concentration).
In addition, a specific ecotoxicity testing programme designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms was undertaken following the silver substance evaluation. This demonstrated that the nanoform of silver is less toxic than ionic silver (based on EC10 and EC50 values for total, ‘conventional’ dissolved (<0.45μm) and ‘truly’ dissolved (<3kDa) silver). Therefore, taking the full body of evidence into account, the read-across use of toxicity values from ionic to nanosilver as a ‘worst case’ approach is justified and scientifically defensible for environmental endpoints. The Substance Evaluation Conclusion document for silver agreed with this conclusion for the nanosilver forms covered in the REACH dossier (RIVM 2018).

Long-term toxicity to aquatic invertebrates
Published data from two long-term toxicity to Daphnia magna studies using various sizes of nanoparticles and coating types are included in the REACH dossier as Endpoint Study Records. The nanosilver NOEC values from these studies are in the same order of magnitude or higher than the EC10 of 2.14 μg/L for Daphnia magna for ionic silver (Bianchini and Wood 2008). A summary of these supporting studies is available under Section 4.1.4 of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IUCLID Section 13). There is insufficient chronic data for aquatic invertebrates to investigate the influence of particle size of coating on the observed toxicity.
In addition, following the silver substance evaluation, Fraunhofer (Schlich et al. 2017) undertook comparative studies for the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and ionic silver (silver nitrate) with Daphnia magna, in accordance to OECD guideline 211. The most sensitive endpoint in both tests was reproduction after 21 days. The EC10 value for reproduction was 33.4 µg/L measured dissolved silver (<0.45 µm) for nanosilver and 3.49 µg/L measured dissolved silver (<0.45 µm) for silver nitrate. Thus, the directly comparable EC10 values reported by Schlich et al (2017) were an order of magnitude more toxic for ionic silver than for nanosilver. These studies are included in the REACH dossier as Endpoint Study Records and a summary is available in the report ‘Summary of Comparative Ecotoxicity Testing Programme for Nanosilver and Silver Nitrate (2017)’ (attached in IUCLID Section 13).
Together with the theoretical basis for read-across based on the free-ion, above studies support the use of ionic silver as the ‘worst case’ basis to read across properties to nanosilver.


For further information and data matrix see 'CSR Annex 9 - Read Across Justification Nanosilver ENV_SUMMARY_200706' attached in IUCLID section 13.

Reason / purpose for cross-reference:

read-across source

Duration:

24 mo

Dose descriptor:

NOEC

Effect conc.:

5 µg/L

Nominal / measured:

not specified

Conc. based on:

element

Basis for effect:

reproduction

Details on results:

Larval releases in the parental stock of C. fornicata were significantly reduced in the 0.010 mg/L silver exposure group. The generation times, from egg to adulthood and first spawning of females at 0.001 and 0.005 mg/L were significantly different from controls. Females at 0.005 mg/L were significantly smaller than controls upon fist larval release. No differences were found in larval size at release for F1 C. fornicata.

Results with reference substance (positive control):

Not applicable

Reported statistics and error estimates:

No data reported

Validity criteria fulfilled:

not specified

Conclusions:

The 2 year NOEC is 5 µg/L silver.

Executive summary:

C. fornicata was exposed for 24 month to silver as nitrate. 10 mated pairs of C. fornicata per concentration were exposed to 1, 5 and 10 µg Ag/L (1.34, 4.97 and 10.93 µg/L actual silver concentrations). After copulation and fertilization the females produced egg sacs, which allowed observation of developing larvae. The number of larval releases for each of 10 females from each test concentration of silver and controls was recorded for the entire test period. Fifteen larvae were then measured to determine their size at release. F1 larvae from each release of a single mating pair of the parental stock at each test concentration and control were held at the same silver exposure concentrations as the parents (15-25 °C). Each culture was changed daily. The larvae were reared through metamorphosis and the percentage that set was determined. F1 juveniles were maintained at ambient seawater conditions (4.5 - 25 °C) and at the same silver concentrations in which they were hatched. F1 juveniles were then reared to adulthood. A maximum of 10 mated pairs from each release of the single mated pair of parents was retained to observe fecundity. Results: Larval releases in the parental stock of C. fornicata were significantly reduced in the 0.010 mg/L silver exposure group. The generation times, from egg to adulthood and first spawning of females at 0.001 and 0.005 mg/L were significantly different from controls. Females at 0.005 mg/L were significantly smaller than controls upon fist larval release. No differences were found in larval size at release for F1 C. fornicata.

Reference 11

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

key study

Justification for type of information:

1. HYPOTHESIS FOR THE READ-ACROSS APPROACH (ENDPOINT LEVEL)
The REACH registration of silver (powder and massive forms of zero-valent, elemental, silver) is underpinned, in common with other metals, by a read-across (or analogue) approach from the properties of the free ion. This principle of read-across from the free ion has been extended to also include nanosilver.
The scientific validity of read-across from the hazard properties of ionic silver (source) to nanosilver (target) under REACH is underpinned by both theoretical and empirical considerations.
The theoretical basis for the use of ionic silver data as the foundation of the risk assessment of nanosilver is based on the premise that the free metal ion (Me+) is the most toxic metal form/species (Starodub et al. 1987). This consideration was implicit in the development of the Free Ion Activity Model [FIAM] (Morel 1983, Paquin et al. 2002, Campbell 1985, Brown and Markich 2000) and, more recently, the Biotic Ligand Model [BLM] (Paquin et al 2002, Niyogi and Wood 2004) that has underpinned the risk assessment of several metals (e.g. Cu, Ni, Zn) under the Existing Substance Regulations and REACH; and most recently the development of the Environmental Quality Standard (EQS) for nickel and nickel substances under the Water Framework Directive (WFD). When considered on an equal mass basis ionic silver would therefore be expected to have greater toxicity than nanosilver simply on the basis that silver ions are released over time from the surface of particles (via oxidative dissolution). As the properties of nanosilver are read-across directly from ionic silver (not just to the fraction of silver ions released from nanosilver), this read-across is also expected to introduce considerable precaution into the hazard component of the risk assessment of nanosilver as all nanosilver, irrespective of coating, morphology or particle size distribution is assumed to behave similarly to the free ion.
This theoretical consideration has been tested by conducting a comprehensive review of the available scientific literature for nanosilver, with particular emphasis on the comparative effects on REACH relevant biotic systems (REACH information requirement) of ionic silver and nanosilver. This review is described for each endpoint in subsequent sections of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IULCID Section 13) and is summarised below. Furthermore, this theoretical consideration was confirmed by a specific ecotoxicity testing programme undertaken by the EPMF following the silver substance evaluation and designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms.

2. READ-ACROSS APPROACH JUSTIFICATION (ENDPOINT LEVEL)
In terms of ecotoxicity, nanosilver on an equal mass basis has been found to be significantly less toxic than ionic silver for the majority of REACH endpoints and of equivalent toxicity for some others. None of the empirical information available suggests that nanosilver is consistently more hazardous than ionic silver on an equivalent mass basis, or that “nano-specific” effects that would prejudice the validity of read across from ionic silver to nanosilver occur. In addition, with very limited exceptions which are described further in the report ‘Nanosilver: read-across justification for environmental information requirements’, none of the available data suggested consistent relationships between particle morphology, size, particle size distribution (raw or agglomerated) or coating (surface treatment) and effects.
Notter et al. (2014) presents a meta-analysis of published EC50 values for ionic silver and nanosilver. The authors demonstrate that almost 94% of acute toxicity values assessed for freshwater, seawater and terrestrial systems using algae, annelid, arthropoda, bacteria, crustacea, fish, nematoda, plant, protozoa and rotatoria show that the nanoform of silver is less toxic than the dissolved metal (when normalised for total metal concentration).
In addition, a specific ecotoxicity testing programme designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms was undertaken following the silver substance evaluation. This demonstrated that the nanoform of silver is less toxic than ionic silver (based on EC10 and EC50 values for total, ‘conventional’ dissolved (<0.45μm) and ‘truly’ dissolved (<3kDa) silver). Therefore, taking the full body of evidence into account, the read-across use of toxicity values from ionic to nanosilver as a ‘worst case’ approach is justified and scientifically defensible for environmental endpoints. The Substance Evaluation Conclusion document for silver agreed with this conclusion for the nanosilver forms covered in the REACH dossier (RIVM 2018).

Long-term toxicity to aquatic invertebrates
Published data from two long-term toxicity to Daphnia magna studies using various sizes of nanoparticles and coating types are included in the REACH dossier as Endpoint Study Records. The nanosilver NOEC values from these studies are in the same order of magnitude or higher than the EC10 of 2.14 μg/L for Daphnia magna for ionic silver (Bianchini and Wood 2008). A summary of these supporting studies is available under Section 4.1.4 of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IUCLID Section 13). There is insufficient chronic data for aquatic invertebrates to investigate the influence of particle size of coating on the observed toxicity.
In addition, following the silver substance evaluation, Fraunhofer (Schlich et al. 2017) undertook comparative studies for the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and ionic silver (silver nitrate) with Daphnia magna, in accordance to OECD guideline 211. The most sensitive endpoint in both tests was reproduction after 21 days. The EC10 value for reproduction was 33.4 µg/L measured dissolved silver (<0.45 µm) for nanosilver and 3.49 µg/L measured dissolved silver (<0.45 µm) for silver nitrate. Thus, the directly comparable EC10 values reported by Schlich et al (2017) were an order of magnitude more toxic for ionic silver than for nanosilver. These studies are included in the REACH dossier as Endpoint Study Records and a summary is available in the report ‘Summary of Comparative Ecotoxicity Testing Programme for Nanosilver and Silver Nitrate (2017)’ (attached in IUCLID Section 13).
Together with the theoretical basis for read-across based on the free-ion, above studies support the use of ionic silver as the ‘worst case’ basis to read across properties to nanosilver.



For further information and data matrix see 'CSR Annex 9 - Read Across Justification Nanosilver ENV_SUMMARY_200706' attached in IUCLID section 13.

Reason / purpose for cross-reference:

read-across source

Duration:

28 d

Dose descriptor:

NOEC

Effect conc.:

19 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

dissolved

Basis for effect:

mortality

Remarks on result:

other: salinity 30‰

Details on results:

The effect values from the toxicity test demonstrate that increasing salinity has a mitigating effect on silver toxicity if toxicity is determined using measured concentrations of dissolved silver.

Results with reference substance (positive control):

Not applicable

Reported statistics and error estimates:

Values for NOEC and LOEC were calculated using ANOVA and a parmetric Dunnettt's test or nonparametric William's test. The LC20 calculated by weighted least-squares nonlinear regression, and the EC50 calculated by binomial/nonlinear interpolation. (p<0.05).

Salinity	NOEC	LOEC	EC20	ACR
‰	µg/L dissolved Ag
Mysids
10	6.0	13	3.9	>3.2
20	34	60	60	5.8
30	19	37	35	9.3

Validity criteria fulfilled:

yes

Conclusions:

28 day NOEC mortality is 19 μg dissolved Ag/L (salinity 30‰)

Executive summary:

This is a GLP, guideline study and is considered reliable and fully acceptable for use for this endpoint

Reference 12

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

key study

Justification for type of information:

1. HYPOTHESIS FOR THE READ-ACROSS APPROACH (ENDPOINT LEVEL)
The REACH registration of silver (powder and massive forms of zero-valent, elemental, silver) is underpinned, in common with other metals, by a read-across (or analogue) approach from the properties of the free ion. This principle of read-across from the free ion has been extended to also include nanosilver.
The scientific validity of read-across from the hazard properties of ionic silver (source) to nanosilver (target) under REACH is underpinned by both theoretical and empirical considerations.
The theoretical basis for the use of ionic silver data as the foundation of the risk assessment of nanosilver is based on the premise that the free metal ion (Me+) is the most toxic metal form/species (Starodub et al. 1987). This consideration was implicit in the development of the Free Ion Activity Model [FIAM] (Morel 1983, Paquin et al. 2002, Campbell 1985, Brown and Markich 2000) and, more recently, the Biotic Ligand Model [BLM] (Paquin et al 2002, Niyogi and Wood 2004) that has underpinned the risk assessment of several metals (e.g. Cu, Ni, Zn) under the Existing Substance Regulations and REACH; and most recently the development of the Environmental Quality Standard (EQS) for nickel and nickel substances under the Water Framework Directive (WFD). When considered on an equal mass basis ionic silver would therefore be expected to have greater toxicity than nanosilver simply on the basis that silver ions are released over time from the surface of particles (via oxidative dissolution). As the properties of nanosilver are read-across directly from ionic silver (not just to the fraction of silver ions released from nanosilver), this read-across is also expected to introduce considerable precaution into the hazard component of the risk assessment of nanosilver as all nanosilver, irrespective of coating, morphology or particle size distribution is assumed to behave similarly to the free ion.
This theoretical consideration has been tested by conducting a comprehensive review of the available scientific literature for nanosilver, with particular emphasis on the comparative effects on REACH relevant biotic systems (REACH information requirement) of ionic silver and nanosilver. This review is described for each endpoint in subsequent sections of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IULCID Section 13) and is summarised below. Furthermore, this theoretical consideration was confirmed by a specific ecotoxicity testing programme undertaken by the EPMF following the silver substance evaluation and designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms.

2. READ-ACROSS APPROACH JUSTIFICATION (ENDPOINT LEVEL)
In terms of ecotoxicity, nanosilver on an equal mass basis has been found to be significantly less toxic than ionic silver for the majority of REACH endpoints and of equivalent toxicity for some others. None of the empirical information available suggests that nanosilver is consistently more hazardous than ionic silver on an equivalent mass basis, or that “nano-specific” effects that would prejudice the validity of read across from ionic silver to nanosilver occur. In addition, with very limited exceptions which are described further in the report ‘Nanosilver: read-across justification for environmental information requirements’, none of the available data suggested consistent relationships between particle morphology, size, particle size distribution (raw or agglomerated) or coating (surface treatment) and effects.
Notter et al. (2014) presents a meta-analysis of published EC50 values for ionic silver and nanosilver. The authors demonstrate that almost 94% of acute toxicity values assessed for freshwater, seawater and terrestrial systems using algae, annelid, arthropoda, bacteria, crustacea, fish, nematoda, plant, protozoa and rotatoria show that the nanoform of silver is less toxic than the dissolved metal (when normalised for total metal concentration).
In addition, a specific ecotoxicity testing programme designed to compare the effects of the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and silver nitrate to algae, Daphnia (long-term) and soil microorganisms was undertaken following the silver substance evaluation. This demonstrated that the nanoform of silver is less toxic than ionic silver (based on EC10 and EC50 values for total, ‘conventional’ dissolved (<0.45μm) and ‘truly’ dissolved (<3kDa) silver). Therefore, taking the full body of evidence into account, the read-across use of toxicity values from ionic to nanosilver as a ‘worst case’ approach is justified and scientifically defensible for environmental endpoints. The Substance Evaluation Conclusion document for silver agreed with this conclusion for the nanosilver forms covered in the REACH dossier (RIVM 2018).

Long-term toxicity to aquatic invertebrates
Published data from two long-term toxicity to Daphnia magna studies using various sizes of nanoparticles and coating types are included in the REACH dossier as Endpoint Study Records. The nanosilver NOEC values from these studies are in the same order of magnitude or higher than the EC10 of 2.14 μg/L for Daphnia magna for ionic silver (Bianchini and Wood 2008). A summary of these supporting studies is available under Section 4.1.4 of the report ‘Nanosilver: read-across justification for environmental information requirements’ (attached in IUCLID Section 13). There is insufficient chronic data for aquatic invertebrates to investigate the influence of particle size of coating on the observed toxicity.
In addition, following the silver substance evaluation, Fraunhofer (Schlich et al. 2017) undertook comparative studies for the smallest nanosilver form registered under REACH (‘Nano 8.1’ or ‘Silberpulver Typ 300-30’) and ionic silver (silver nitrate) with Daphnia magna, in accordance to OECD guideline 211. The most sensitive endpoint in both tests was reproduction after 21 days. The EC10 value for reproduction was 33.4 µg/L measured dissolved silver (<0.45 µm) for nanosilver and 3.49 µg/L measured dissolved silver (<0.45 µm) for silver nitrate. Thus, the directly comparable EC10 values reported by Schlich et al (2017) were an order of magnitude more toxic for ionic silver than for nanosilver. These studies are included in the REACH dossier as Endpoint Study Records and a summary is available in the report ‘Summary of Comparative Ecotoxicity Testing Programme for Nanosilver and Silver Nitrate (2017)’ (attached in IUCLID Section 13).
Together with the theoretical basis for read-across based on the free-ion, above studies support the use of ionic silver as the ‘worst case’ basis to read across properties to nanosilver.


For further information and data matrix see 'CSR Annex 9 - Read Across Justification Nanosilver ENV_SUMMARY_200706' attached in IUCLID section 13.

Reason / purpose for cross-reference:

read-across source

Duration:

30 d

Dose descriptor:

NOEC

Effect conc.:

8.6 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

dissolved

Basis for effect:

other: development NOEC effects on spines

Details on results:

Adult NOEC 8.6 µg/L; adult LOEC 19 µg/L. 96 h LC50 40 µg/L. Fertilization EC50 14 µg/L (95% cf 12-16 µg/L)

Results with reference substance (positive control):

Not applicable

Validity criteria fulfilled:

yes

Conclusions:

30 day development NOEC effect on spines is 8.6 μg dissolved silver/L

Reference 13

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

OECD Guideline 211 (Daphnia magna Reproduction Test)

Deviations:

yes

Remarks:

The test item dispersion was prepared in modified Elendt M7 medium

Qualifier:

according to guideline

Guideline:

EU Method C.20 (Daphnia magna Reproduction Test)

Deviations:

yes

Remarks:

The test item dispersion was prepared in modified Elendt M7 medium

GLP compliance:

yes (incl. QA statement)

Specific details on test material used for the study:

SOURCE OF TEST MATERIAL
- Expiration date of the lot/batch: April 19, 2017
- Purity test date: Not reported


STABILITY AND STORAGE CONDITIONS OF TEST MATERIAL
- Storage condition of test material: Keep tightly closed in a dry, cool and well-ventilated place at 5 – 15 °C.
- Stability under test conditions: Stable under normal storage conditions.

Analytical monitoring:

yes

Details on sampling:

-Sampling: To assess the test item concentrations and the release of silver ions from the test item ‘truly’ and ‘conventional’ dissolved silver was measured by chemical analysis at test initiation, after each medium renewal and test termination. Total silver was measured at test initiation, after each medium renewal and test termination, in test media samples of each treatment taken directly from test vessels (total silver), after centrifugal filtration (for ‘truly’ dissolved silver) and after filtration through 0.45 µm membrane filters (for ‘conventional’ dissolved silver).

Additionally, the particle size and the zeta potential was measured from samples of an extra analytical vessel without daphnia and algal food to characterize the test item in test media once per week. Total and dissolved silver (using both centrifugal filtration and filtration through 0.45 µm membrane filters) were also measured in this additional vessel (in addition to the measurements from the test vessels) once per week and at test termination.

Differentiation between nanoparticles and ions
For differentiation of particle and ions (’truly dissolved silver) the samples were filtered using special centrifugal vials (Vivaspin 20, Sartorius, 3kDa). The samples were centrifugated at 3000 x g for 1 hour at 20 °C stepwise. The centrfugal filters were conditioned twice before use (such that the third centrifuged sample was analysed). In addition a separate sample was filtered through 0.45 µm polyethersulfone membrane filter for determination of the ‘conventional’ dissolved silver after precondition of the filter with an adequate amount of the sample. The determination of ‘conventional’ and ‘truly’ dissolved silver for the control and each treatment was performed at test initiation, after each medium renewal and at test termination in test vessels and once per week also in additional vessels (no Daphnia or algal food).

- Sample storage conditions before analysis: All samples for the test item concentration measurements were acidified with HNO3 (1.25 mL concentrated HNO3) and stored in a refrigerator (about 4 °C) until further analysis. For the determination of size and zeta potential single 10 mL samples of the test dispersion were taken and measurements were performed immediately at the day of sampling.

Vehicle:

no

Details on test solutions:

PREPARATION AND APPLICATION OF TEST SOLUTION
- Method: The test item dispersion was prepared in modified Elendt M7 medium. Initially, a stock dispersion was prepared by pipetting an adequate amount of the test item, and transferring it to purified water. Directly after addition of the test item the stock dispersion was sonicated for 3 minutes in a sonication bath and then the dispersion was shaken carefully.
The test concentrations were prepared by adding 55.6, 176, 556, 1760 and 5560 µL of stock dispersion into modified Elendt M7 medium and the dispersion was sonicated for 3 minutes in a sonication bath and then shaken carefully. The freshly prepared test dispersions were then left to stand for 2 hours to reach equilibrium. Following equilibration, the test solutions were filled into the test vessels after more careful shaking.

Test organisms (species):

Daphnia magna

Details on test organisms:

TEST ORGANISM
- Common name: Daphnia magna
- Strain/clone: Not reported
- Source: bred in the testing facility, originally from German Federal Environment Agency, Institut für Wasser-, Boden- und Lufthygiene.
- Age of parental stock (mean and range, SD): at least 3 weeks old
- Feeding during test: yes
- Food type: The daphnids were fed during the test with a mixture of suspensions of unicellular alga Desmodesmus subspicatus and JBL ArtemioFluid in a ratio of 9 : 1 (vol/vol).
- Amount: The content of alga food in the test media, measured at 585 nm, was kept constant at 0.2 mg C/(Daphnia x day).

ACCLIMATION
- Acclimation period: at least three weeks
- Acclimation conditions: No, daphnids were gradually acclimated to the modified Elendt M7 medium. First batch of daphnids was prepared by moving from purified drinking water to 33 %, 66 % and 100 % of the modified Elendt M7 mediumover a three week period. Second batch was prepared by moving the daphnids from purified drinking water to 50 % and 100 % of the modified Elendt M7 medium.
- Health during acclimation (any mortality observed): The first batch animals appeared to be less vital and a higher mortality of adult and juvenile animals was observed in the culture. Therefore, a new culture was established from animals held in Cu-free dilution water (second batch). No mortalities was observed in the second batch culture.

Test type:

semi-static

Water media type:

freshwater

Limit test:

no

Total exposure duration:

21 d

Hardness:

Dilution water: 1.0 mmol/L (total hardness), 0.8 mmol/L (Ca-hardenss)
Modified Elendt M7 medium: 2.2 mmol/L (Ca-hardenss)
Overlaying water: 2.6 - 2.8 mmol/L (fresh), 2.3 - 2.8 mmol/L (old)

Test temperature:

20.6 - 20.9 oC

pH:

8.25 - 9.40

Dissolved oxygen:

7.80 - 10.10 mg/L

Salinity:

Not applicable

Conductivity:

252.0 µS/cm

Nominal and measured concentrations:

Nominal: 3.16, 10.0, 31.6, 100 and 316 µg Ag/L
Arithmetic mean measured concentrations of total Ag: 2.29, 7.34, 23.0, 70.1 and 193 µg Ag/L
Arithmetic mean measured conventional dissolved Ag concentrations: 1.09, 3.93, 12.8, 43.6 and 104 µg Ag/L
Arithmetic mean measured concentrations of truly dissolved Ag: 0.001, 0.010, 0.061, 0.400 and 1.130 µg Ag/L

Details on test conditions:

TEST SYSTEM
- Test vessel: polypropylene beakers (50 mL nominal volume) cleaned as per standard cleaning methods to remove any deposits derived from the manufacturing process or previous use, and then were cleaned with a HNO3 solution (11.5 mL of concentrated HNO3 per L of pure water) and rinsed thoroughly with ultrapure water.
- Material, size, headspace, fill volume: The containers were filled up with about 50 mL test solution and covered with glass panes to prevent evaporation as much as possible, but also permit gaseous exchange
- Aeration: no
- Renewal rate of test solution (frequency/flow rate): Test dispersion was exchanged three times a week.
- No. of organisms per vessel: 1 young female Daphnia per vial
- No. of vessels per concentration (replicates): 12
- No. of vessels per control (replicates): 12, (2 additional replicates per treatment and control were prepared similar to the test vessels (with daphnia and algal food) for chemical analysis purposes only; and 2 additional replicate without daphnia and algal food for the measurement of particle size and zeta potential (and additional chemical analysis)).

TEST MEDIUM / WATER PARAMETERS
- Source/preparation of dilution water: synthetic, modified Elendt M7 medium. To avoid reactions of the silver with chloride (Cl-), the trace elements and macro nutrients containing Cl- were replaced by suitable elements containing nitrate (NO3-) instead of Cl¬-. EDTA was completely excluded from the test media.
- Culture medium different from test medium: yes

OTHER TEST CONDITIONS
- Photoperiod: light/dark cycle of 16/8 hours
- Light intensity: 700 – 831 lux (corresponding to 9.3 – 11.1 µE/(m² *s))

EFFECT PARAMETERS MEASURED (with observation intervals if applicable) : The numbers of immobile daphnids were visually determined and any abnormalities in appearance and behaviour were recorded daily. The newborn daphnids per beaker were counted and removed daily until all daphnids start to reproduce (at least 10 days and at a maximum of 14 days). After that newborn daphnids were counted and removed at each water renewal. Abnormalities in condition (including male sex) or presence of winter eggs were checked and recorded. At study termination, length of the adults without significant mortality was measured by digital photography and image analysis and compared with the equally measured control animals. The following endpoints were evaluated quantitatively: Immobility of parental generation daphnids, Time to the first brood, Cumulative number of live offspring at day 21 (study end), Individual length of adults

TEST CONCENTRATIONS
- Spacing factor for test concentrations: 3.16
- Range finding study: yes
- Test concentrations range finding study: three nominal test concentrations of 0.1, 1.0 and 10 µg Ag/L
- Results used to determine the conditions for the definitive study: There was no effect on the observed endpoints of Daphnia magna in any of the tested concentrations in the range-finding test.
- Test concentrations definitive study: nominal test concentrations of 3.16, 10, 31.6, 100 and 316 µg Ag/L

Reference substance (positive control):

yes

Remarks:

An acute immobilization tests over 24 h was performed with K2Cr2O7 (separate study, January 2017).

Duration:

21 d

Dose descriptor:

EC10

Effect conc.:

50.68 µg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

element

Basis for effect:

reproduction

Remarks on result:

other: Results based on total Ag. 95% CL: 26.07 - 67.88 ug/L

Duration:

21 d

Dose descriptor:

EC50

Effect conc.:

106.51 µg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

element

Basis for effect:

reproduction

Remarks on result:

other: Results based on total Ag; 95% CL: 83.54 - 139.08 ug/L

Duration:

21 d

Dose descriptor:

NOEC

Effect conc.:

23 µg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

element

Basis for effect:

reproduction

Remarks on result:

other: Results based on total Ag

Duration:

21 d

Dose descriptor:

LOEC

Effect conc.:

70.1 µg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

element

Basis for effect:

reproduction

Remarks on result:

other: Results based on total Ag

Duration:

21 d

Dose descriptor:

EC10

Effect conc.:

33.39 µg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

element

Basis for effect:

reproduction

Remarks on result:

other: Results based on conventional dissolved Ag; 95% CL: 18.40 - 43.06 ug/L

Duration:

21 d

Dose descriptor:

EC50

Effect conc.:

62.54 µg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

element

Basis for effect:

reproduction

Remarks on result:

other: Results based on conventional dissolved Ag; 95% CL: 50.39 - 79.44 ug/L

Duration:

21 d

Dose descriptor:

LOEC

Effect conc.:

43.6 µg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

element

Basis for effect:

reproduction

Remarks on result:

other: Results based on conventional dissolved Ag

Duration:

21 d

Dose descriptor:

NOEC

Effect conc.:

12.8 µg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

element

Basis for effect:

reproduction

Remarks on result:

other: Results based on conventional dissolved Ag

Duration:

21 d

Dose descriptor:

EC10

Effect conc.:

0.292 µg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

element

Basis for effect:

reproduction

Remarks on result:

other: Based on truly dissolved Ag; 95 % CL : 0.143 - 0.396 ug/L

Duration:

21 d

Dose descriptor:

EC50

Effect conc.:

0.616 µg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

element

Basis for effect:

reproduction

Remarks on result:

other: Based on truly dissolved Ag; 95 % CL :0.475 - 0.821 ug/L

Duration:

21 d

Dose descriptor:

LOEC

Effect conc.:

0.4 µg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

element

Basis for effect:

reproduction

Remarks on result:

other: Based on truly dissolved Ag;

Duration:

21 d

Dose descriptor:

NOEC

Effect conc.:

0.061 µg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

element

Basis for effect:

reproduction

Remarks on result:

other: Based on truly dissolved Ag;

Details on results:

- Observations on body length and weight: Adult body length showed significant differences between control and the two highest treatments.
- Other biological observations: There was a significant difference in survival of individuals between control and the two highest treatments with silver nanoparticles. No other clinical signs were observed in any replicate at any concentration tested. Neither any physical nor pathological symptoms were obtained. Age of the first reproduction was not significantly affected.
- Mortality of control: No mortality in control observed.
- Particle size and zeta potential: The measurements of size and zeta potential of fresh samples of the two highest test concentrations showed a lower PDI in the range of 0.3 – 0.5 than in the lower test concentration where the PDI mainly was between 0.8 and 1.0. In the aged samples the PDI increased up to 1.0 indicating the sedimentation and agglomeration of nanoparticles. In addition, a higher count rate than in the measurements of the lower test concentrations indicated that nanoparticles were present in the freshly prepared media. However, the test dispersions still were polydisperse and no precise measurement of nanoparticle size was possible. The measurement at the three lower test concentrations produced no sufficient results. This revealed that the chosen concentrations, which represented an optimal range for the reproduction test, were not sufficient for the size and zeta potential analysis. Especially the count rate of the size measurements indicated that the test concentrations are too low.

Results with reference substance (positive control):

24 h EC50 value (immobilization): 1.088 mg/L (95% CL: 0.632 – 1.841 mg/L)

Reported statistics and error estimates:

A LOEC was calculated by using ANOVA followed by Student-t test, Fisher`s Exact Binomial test, Dunnett’s or Williams’ test or an appropriate non-parametric test. If no effect was observed up to the highest tested concentration the NOEC values were determined using appropriate statistical methods.
Since the test results of the reproduction test show a concentration-response relationship they were statistically analysed to determine an EC50, EC20 and EC10 value together with 95 % confidence intervals using Probit-analysis assuming log-normal distribution of the values.
Statistical calculations were done with the computer program ToxRat® Pro 3.2.1.

 

Table1: Percent reduction of survival, length and reproduction compared to controls after 21 days based on mean measured total Ag concentration.

Mean measured total Ag [µg Ag/L]	% reduction of survival (immobility)	% reduction of reproduction per introduced parent	% minimum detectable difference of age of first production	% decrease in length	% inhibition of intrinsic rate r
Control	0	0	0	0	0
2.29	0 (-)	-6.4 (-)	7.7 (-)	1.5 (-)	-6.1 (-)
7.34	0 (-)	4.0 (-)	9.9 (-)	3.0 (-)	4.1 (-)
23.0	0 (-)	7.5 (-)	11.1 (-)	1.3 (-)	5.0 (-)
70.1	16.7 (+)	22.9 (+)	10.6 (-)	5.5 (+)	3.4 (-)
193	33.3 (+)	85.1 (+)	9.7 (-)	15.4 (+)	35.8 (+)

(+) statistically significant difference between controls / (-) no significant difference between controls and treatments; Cochran-Armitage (survival), Welsh t-test with Bonferroni Adjustment (length and age of first reproduction) and Step-down Jonckheere-Terpstra Test (reproduction and intrinsic rate r); significance level 0.05, one-sided smaller.

Table 2: Percent reduction of survival, length and reproduction compared to controls after 21 days based on mean measured conventional dissolved Ag.

Mean measured conventional dissolved Ag [µg Ag/L]	% reduction of survival (immobility)	% reduction of reproduction per introduced parent	% minimum detectable difference of age of first production	% decrease in length	% inhibition of intrinsic rate r
Control	0	0	0	0	0
1.09	0 (-)	-6.4 (-)	7.7 (-)	1.5 (-)	-6.1 (-)
3.93	0 (-)	4.0 (-)	9.9 (-)	3.0 (-)	4.1 (-)
12.8	0 (-)	7.5 (-)	11.1 (-)	1.3 (-)	5.0 (-)
43.6	16.7 (+)	22.9 (+)	10.6 (-)	5.5 (+)	3.4 (-)
104	33.3 (+)	85.1 (+)	9.7 (-)	15.4 (+)	35.8 (+)

(+) statistically significant difference between controls / (-) no significant difference between controls and treatments. Cochran-Armitage (survival), Welsh t-test with Bonferroni Adjustment (length and age of first reproduction) and Step-down Jonckheere-Terpstra Test (reproduction and intrinsic rate r); significance level 0.05, one-sided smaller.

Table 3: Percent reduction of survival, length and reproduction compared to controls after 21 days based on mean measured truly dissolved Ag.

Mean measured truly dissolved Ag [µg Ag/L]	% reduction of survival (immobility)	% reduction of reproduction per introduced parent	% minimum detectable difference of age of first production	% decrease in length	% inhibition of intrinsic rate r
Control	0	0	0	0	0
0.001	0 (-)	-6.4 (-)	7.7 (-)	1.5 (-)	-6.1 (-)
0.010	0 (-)	4.0 (-)	9.9 (-)	3.0 (-)	4.1 (-)
0.061	0 (-)	7.5 (-)	11.1 (-)	1.3 (-)	5.0 (-)
0.400	16.7 (+)	22.9 (+)	10.6 (-)	5.5 (+)	3.4 (-)
1.130	33.3 (+)	85.1 (+)	9.7 (-)	15.4 (+)	35.8 (+)

(+) statistically significant difference between controls / (-) no significant difference between controls and treatments. Cochran-Armitage (survival), Welsh t-test with Bonferroni Adjustment (length and age of first reproduction) and Step-down Jonckheere-Terpstra Test (reproduction and intrinsic rate r); significance level 0.05, one-sided smaller.

 

Table 4: Effective concentrations based on mean measured total Ag concentrations for the exposure of Daphnia magna for 21 days.

Mean measured test item concentrations [µg/L] - total Ag
Parameter^a		EC10	EC20	EC50	LOEC	NOEC
Immobility	Value	63.47	106.87	289.59	70.1	23.0
	95 %-cl lower	1.56	29.18	145.45
	95 %-cl upper	121.29	361.69	n.d.
Reproduction	Value	50.68	65.39	106.51	70.1	23.0
	95 %-cl lower	26.07	40.48	83.54
	95 %-cl upper	67.88	83.40	139.08
Age of first reproduction	Value	0.0	0.0	0.0	> 193	≥ 193
	95 %-cl lower	0.0	0.0	0.0
	95 %-cl upper	0.0	0.0	0.0
Length	Value	118.68	278.31	1421.2	193	70.1^a
	95 %-cl lower	30.92	178.74	492.92
	95 %-cl upper	187.49	4983.56	n.d.
Intrinsic rate r	Value	100.94	137.75	249.70	193	70.1
	95 %-cl lower	n.d.	n.d.	n.d.
	95 %-cl upper	n.d.	n.d.	n.d.

n.d.: not determined
^a: The NOEC for length was calculated to be at 23.0 µg Ag/L based on arithmetic mean measured concentrations of total Ag. However, due to the low inhibition of 5.5% at a concentration of 70.1 µg Ag/L, the NOEC was set to 70.1 µg Ag/L, since effects below 10 % compared to control are generally not considered to be ecotoxicologically relevant and it is generally recommended by OECD and EFSA to use the EC10 approach in preference to the NOEC approach for the environmental risk assessment [8],[9],[10].

Table 5: Effective concentrations based on mean measured conventional dissolved Ag concentrations for the exposure of Daphnia magna for 21 days.

Mean measured test item concentrations [µg/L] – conventional dissolved Ag
Parameter^a		EC10	EC20	EC50	LOEC	NOEC
Immobility	Value	37.80	61.05	152.75	43.6	12.8
	95 %-cl lower	0.051	9.57	81.24
	95 %-cl upper	68.62	218.18	n.d.
Reproduction	Value	33.39	41.42	62.54	43.6	12.8
	95 %-cl lower	18.40	26.96	50.39
	95 %-cl upper	43.06	51.25	79.44
Age of first reproduction	Value	0.0	0.0	0.0	104	43.6
	95 %-cl lower	0.0	0.0	0.0
	95 %-cl upper	0.0	0.0	0.0
Length	Value	68.47	142.04	573.62	104	43.6^a
	95 %-cl lower	3.13	95.69	219.77
	95 %-cl upper	104.92	n.d.	n.d.
Intrinsic rate r	Value	60.03	78.13	129.33	104	43.6
	95 %-cl lower	n.d.	n.d.	n.d.
	95 %-cl upper	n.d.	n.d.	n.d.

n.d.: not determined
^a:The NOEC for length was calculated to be at 12.8 µg Ag/L based on arithmetic mean measured concentrations of conventional dissolved Ag. However, due to the low inhibition of 5.5% at a concentration of 43.6 µg Ag/L, the NOEC was set to 43.6 µg Ag/L, since effects below 10 % compared to control are generally not considered to be ecotoxicologically relevant and it is generally recommended by OECD and EFSA to use the EC10 approach in preference to the NOEC approach for the environmental risk assessment [8],[9],[10].

Table 6: Effective concentrations based on mean measured truly dissolved Ag concentrations for the exposure of Daphnia magnafor 21 days.

Mean measured test item concentrations [µg/L] – truly dissolved Ag
Parameter^a		EC10	EC20	EC50	LOEC	NOEC
Immobility	Value	0.312	0.579	1.893	0.40	0.061
	95 %-cl lower	n.d.	n.d.	n.d.
	95 %-cl upper	n.d.	n.d.	n.d.
Reproduction	Value	0.292	0.377	0.616	0.400	0.061
	95 %-cl lower	0.143	0.225	0.475
	95 %-cl upper	0.396	0.487	0.821
Age of first reproduction	Value	0.0	0.0	0.0	1.130	0.400
	95 %-cl lower	0.0	0.0	0.0
	95 %-cl upper	0.0	0.0	0.0
Length	Value	0.692	1.614	8.159	1.130	0.400^a
	95 %-cl lower	n.d.	n.d.	n.d.
	95 %-cl upper	n.d.	n.d.	n.d.
Intrinsic rate r	Value	0.451	0.714	1.719	1.130	0.400
	95 %-cl lower	0.016	0.203	1.062
	95 %-cl upper	0.710	1.234	n.d.

n.d.: not determined
^a:The NOEC for length was calculated to be at 0.061 µg Ag/L based on arithmetic mean measured concentrations of truly dissolved Ag. However, due to the low inhibition of 5.5% at a concentration of 0.400 µg Ag/L, the NOEC was set to 0.400 µg Ag/L, since effects below 10 % compared to control are generally not considered to be ecotoxicologically relevant and it is generally recommended by OECD and EFSA to use the EC10 approach in preference to the NOEC approach for the environmental risk assessment [8],[9],[10].

Validity criteria fulfilled:

yes

Remarks:

Mortality in controls was 0%; mean number of offspring in the control was 101/ female; dissolved O2 was > 3 mg/L; pH was within the range 6 - 9, and did not vary by more than 1.5 units; CV for the mean number of control offspring was 16.7 %.

Conclusions:

Based on mean measured total Ag concentrations the EC10 was 50.68 and 63.47 µg Ag/L for reproduction and immobility, respectively. For reproduction, the NOEC was 23.0 µg Ag/L and LOEC was 70.1 µg Ag/L.
Based on mean measured conventional dissolved Ag concentrations the EC10 was 33.39 and 37.80 µg Ag/L for reproduction and immobility, respectively. For reproduction, the NOEC was 12.8 µg Ag/L and LOEC was 43.6 µg Ag/L.
Based on mean measured truly dissolved Ag concentrations the EC10 was 0.292 and 0.312 µg Ag/L for reproduction and immobility, respectively. For reproduction, the NOEC was 0.061 µg Ag/L and LOEC was 0.40 µg Ag/L.

Executive summary:

The 21 day toxicity of silver nanoparticles to the test organism Daphnia magna was determined in a study according to the OECD 211 guideline. Daphnids were exposed to nominal concentrations of 3.16, 10.0, 31.6, 100 and 316 µg Ag/L in a semi-static system with renewal of test medium three times per week. The nominal test concentrations were prepared in modified Elendt M7 medium. The concentrations of the test item in the test media were determined by chemical analysis in the aqueous phase of all treatment levels by ICP-MS. The results were based on the arithmetic mean measured concentrations of total Ag, conventional dissolved Ag and truly dissolved Ag.

This is a guideline, GLP- study and considered suitable for use as a key study for this endpoint.

Additional information

Summary of available data for uncoated and coated nanosilver

Reliable and relevant data on the long-term toxicity of uncoated and coated nanosilver to invertebrates are available from three studies (Zhao and Wang 2011, Blinova et al. 2012, Schlich et al. 2017c,d).

Zhao and Wang (2011) report the effects on Daphnia magna growth and reproduction after exposure to uncoated nanosilver particles (20 nm) in a standard OECD 21 day reproduction test. The authors report that adult growth was significantly different from the control at the lowest concentration of nanosilver particles tested (5 µg/L). According to REACH guidance a NOEC of 2.5 µg/L can be derived from the LOEC (LOEC/2) as the effect observed at the LOEC was below 20%, relative to control response. Daphnia reproduction was unaffected at 5 µg/L but was significantly reduced at 50 µg/L. A parallel exposure to ionic silver (as silver nitrate) was conducted alongside the nanosilver experiments. However, no significant effects on survival, time to first brood, growth or reproduction were observed at the highest concentration tested (1.6 µg/L).

Blinova et al. (2013) also report the effects of exposure to nanosilver on Daphnia magna in an OECD standard reproduction test, but using natural test media rather than artificial water. NOECs of between 50 and 200 µg/L for two types of coated nanosilver: PVP and Collargol (protein) are reported. Aggregation of nanosilver particles was observed to occur in natural media, with protein coated particles aggregating to a lesser extent than PVP coated particles. Coating type was considered to influence aggregation behaviour to a greater extent than water physico-chemistry. In fact, the media effect on the particle size of aggregates was reported by the authors to be “unremarkable”.

The NOEC for Daphnia magna derived for nanosilver from Zhao and Wang (2011) is in the same order of magnitude as the EC10 of 2.14 µg/L for ionic silver reported for Daphnia magna in the REACH CSR (Bianchini and Wood 2008). The NOECs reported by Blinova et al. (2013) are an order of magnitude less sensitive than the EC10 for ionic silver used in the CSR.

Schlich et al. (2017 c,d) undertook comparative studies for nanosilver and silver nitrate with Daphnia magna. Schlich et al. (2017c) describes the long-term effects on Daphnia magna exposed to silver nitrate, while Schlich et al. (2017d) describes the long-term effects on Daphnia magna exposed to nanosilver. Both of these studies were conducted to OECD guideline 211 according to the principles of GLP, Both studies were conducted insynthetic, modified Elendt M7 medium, in which the trace elements and macro nutrients containing Cl^-were replaced by suitable elements containing nitrate (NO₃^-). EDTA was completely excluded from the test media, and to which the Daphnia were acclimated for at least 3 weeks prior to the studies. The nanosilver material used was a powder in aqueous suspension with the following particle size distribution: D25 = 7 nm, D50 = 8 nm, D75 = 9 nm (see section 4.5 of IUCLID), and contained 37% silver. The most sensitive end point in both tests was reproduction after 21 days. The EC10 values for reproduction for nanosilver were 33.4 (<0.45 µm) and 0.292 (<3kDa) µg/L measured dissolved silver. The comparable EC10 values for silver nitrate were 3.49 (<0.45 µm) and 0.059 (<3kDa) µg/L measured dissolved silver. Thus, the directly comparable EC10 values reported by Schlich et al (2017 c,d) were an order of magnitude more toxic for ionic silver than for nanosilver.

There is insufficient chronic data for aquatic invertebrates to investigate the influence of particle size of coating on the observed toxicity.