Silver chloride

Toxicity to soil microorganisms

Description of key information

The 28 day NOEC was 0.13 mg Ag/kg dw and the EC10 was 0.30 mg Ag/kg dw.  

Additional information

The microbial nitrogen (N) transformation test (AECOM 2011d) was conducted to comply with the OECD Guidelines for the Testing of Chemicals Method 216: Soil Microorganisms – Nitrogen Transformation Test. The test media that was used comprised of equal proportions of garden soil, silica medium sand and silt clay and had a pH of 7.3, organic carbon content of 1.3% and consisted of 81% sand, 8% silt and 11% clay.

The test design used six treatments of Ag plus a control and all Ag concentrations were confirmed by digestion and analysis. The confirmed measured concentrations of Ag in the treatments ranged from 0.04 to 31 mg Ag/kg dry soil. The microbial N transformation test was conducted over 28 days and involved the addition of alfalfa to the samples at the test initiation (day 0). The levels of nitrate-N were measured in the soil at day 0 and at the completion of the test (day 28). The difference in nitrate-N between day 0 and day 28 was then used to represent the N transformation rate in the soil. The results indicated that there was a 10% inhibition in N transformation at a concentration of 0.30 mg Ag/kg dry soil and a 50% inhibition at 8.7 mg Ag/kg dry soil (IC10 and IC50, respectively).

The review of this report indicates that the methodology and experimental design were compliant with the relevant OECD Guidelines for the Testing of Chemicals 216. All deviations from the Guideline have been reported and justified. These deviations from protocol are unlikely to have an impact on the outcome and results of the test. There are however several issues and concerns in addition to those discussed in the overal terrestrial endpoint summary (single soil, no aging after spiking) relating to the design and methodology of the experimental work which may impact on the interpretation and robustness of the results. These are summarised below:

• The pre-incubation of the soils was conducted at a soil hydration level considerably lower than that used during the test. This is not recommended by the OECD Guidelines for the Testing of Chemicals 216, which states that the pre-incubation soil moisture should be the same as that of the test. The hydration level was then increased on the day prior to test initiation as part of the soil spiking protocol. This low level of hydration for the preincubation period may have resulted in the lack of establishment of the soil microbial communities, which may in turn have an influence on the outcome of the test.

• There is no indication that any additional blanks and controls were used throughout the test. As a component of the experimental quality control and quality assurance (QA/QC) a blank comprising of the 0.1 M KCl extraction solution and an alfalfa blank, comprising of the alfalfa and no soil would assist in the validation of the results.

• The concentrations of Ag in each of the test treatments were measured by digestion of the samples, however, there is no indication that these measurements were replicated within each treatment. As spiking was conducted with a bulk soil sample (approximately 4 kg soil), confirmation of the spike should have involved multiple samples/replicates from each test container. This approach would also give an indication of the homogeneity of the spiked Ag throughout the test treatments. In addition, no recovery information is provided to validate the digestion and analysis of the soil samples.

• The report indicates that one test container was used per Ag treatment and the replicates for N analysis consisted of multiple grabs from each of the test containers. This method for the collection of replicates results in samples that are not independent and raises concerns relating to the validity of statistical analyses.

• To determine the N conversion in the soils over the 28 days of the test the average N concentration at day 0 was subtracted from each of the replicated day 28 values. This approach allowed for the calculation of an average N conversion over the duration of the experiment from the three replicates, along with an error value, therefore, the data could be used for hypothesis testing. By calculating the values in this manner, any error observed at day 0 has been removed because only the average was used to determine N transformation. This may have implications on the validity of the statistical analysis conducted on the dataset.

A number of other studies with silver are included as supporting studies in the dossier.However, they are considered less reliable than AECOM 2011d as they do not follow a standard method.

A testing proposal was submitted to ECHA for the testing of soil microorganisms in eight soils, after leaching and aging of the spiked soil.There were no objections to the testing proposal by ECHA or the Member State’s Competent Authorities and data will be added to this dossier once results are available.