Biphenyl

Endpoint summary

Additional information

Toxicity to fish

Brosier (1975) yielded key information on the acute toxicity of biphenyl to fish. A 96-h EC50 of 3.0 mg/L was obtained in a flow-through study with fathead minnow (Pimephales promelas). A single study is available for saltwater fish (Dill and Emmittee, 1980), showing a 96 hr. LC50 of 4.6 mg/L.

Mayes et al. (1988) was considered the key study for long-term toxicity of biphenyl to fish. This study reports the results of an early-life stage test started with newly fertilized rainbow trout eggs (Salmo gairdneri). After a total of 87 days of exposure a NOEC value of 0.229 mg/L was determined based on length measurements of surviving larvae. No reliable studies were identified on saltwater species.

Toxicity to aquatic invertebrates

For acute toxicity to freshwater invertebrates, the study of Gersich et al. (1988) was considered the key study, yielding a 48-h LC50 of 0.36 mg/L obtained in a flow-through study with Daphnia magna. The study of Buhl and Neff (1988) yielded key information for acute toxicity of biphenyl to marine invertebrates, with a 96-h EC50 > 0.269 obtained in a flow-through oyster shell deposition test with Eastern oyster (Crassostrea virginica). However, because this is an unbound effect concentration, this EC50 cannot be used for PNEC derivation and can therefore only be used as supporting information.

Gersich et al. (1988) is the key study for long-term toxicity of biphenyl to aquatic invertebrates, reporting a 21-d NOEC of 0.17 mg/L for both reproduction and mortality of Daphnia magna. No reliable studies were identified on long-term toxicity of biphenyl to saltwater species.

Overall, aquatic invertebrates appear to be the most sensitive group and therefore the NOEC of 0.17 mg/L will be the driver for derivation of the PNEC for both the freshwater and the marine aquatic compartment.

Toxicity to aquatic plants (preferably algae)

Because no reliable experimental data were identified for toxicity of biphenyl to algae, non-testing data (QSAR predictions) as well as testing data for diphenyl oxide and Therminol VP1 (an eutectic mixture of diphenyl oxide and biphenyl) were added to the dossier. All available effect concentrations were more or less in the same range and therefore, based on a weight-of-evidence approach, the 96-h EC50 and chronic value of 1.772 and 0.945 mg/L, respectively, predicted by the latest version of US EPA's ECOSAR (v.1.00, 2009) were considered as the key effect concentrations to be used for the aquatic hazard assessment. Because no chronic values are used under REACH, the NOEC was estimated to be 0.66 mg/L. These values indicate that algae are not the most sensitive group and therefore further testing is not considered necessary.

Toxicity to aquatic microorganisms

No reliable information was identified for this endpoint. However, based on a series of studies with ciliated protozoans, which were all assigned a Klimisch 3 score mainly because exposure concentrations were not verified, a weight-of-evidence approach resulted in a reasonable worst case LOEC of 5.6 mg/L (Dive et al., 1980). No NOEC was reported by Dive et al. (1980), however, based on the information in the publication, it can be assumed that the test concentration series 1.8 – 3.2 – 5.6 – 10.0 mg/L was used and therefore the NOEC can be assumed to be 3.2 mg/L. Nonetheless, the PNECstp is derived from results of the OECD 301C ready biodegradability test, where toxicity to the inoculum was not observed at 100 mg/L biphenyl, with is several-fold higher than the water solubility limit.