Registration Dossier

Ecotoxicological information

Endpoint summary

Administrative data

Description of key information

Orange oil is a Natural Complex Substance (NCS) with D-Limonene as a significant constituent, but it also contains other constituents of highly different water solubility in variable concentrations. The water solubility of the constituents ranges from < 1 up to 1560 mg/L. The short-term aquatic toxicity of Orange oil was tested as a WAF (Water Accommodated Fraction) as recommended for test items with partially soluble constituents (OECD Series on testing and assessment No 23, OECD 2000).

The constituents in the NCS interact with each other and influence each other’s solubility in water, so their concentrations ratios in water differ from those in Orange oil. Upon addition to water, equilibrium will be reached between the dissolved and undissolved fractions which will be characteristic of the loading rate of the substance. Their presence and interaction in the water phase determines the aquatic toxicity of Orange oil. Therefore the short-term aquatic toxicity of Orange oil was tested in water accommodated fractions.

For Orange oil, valid GLP studies on the aquatic toxicity are available for algae, Daphnia and fish. All studies were assessed on validity, especially on measures to prevent the loss of the volatile constituents and whether the exposure concentrations remained constant during the test.

Fresenius (Lebertz 2007) studies on acute toxicity of Orange oil for Daphnia and Algae:

Both Fresenius studies for Orange oil (Orange terpenes, Limonene content 93-95% W/W) are conducted with WAFs. The test report mentions the use of a solvent and 'evaporation of the solvent'. The laboratory was contacted to check the procedure. A solvent was used only during the 'Enlarged screening test' to dose very small amounts of material to the solutions. During the Main test the WAFs were prepared according to the formal procedure.


The test vessels for the Daphnia test are glass vessels with glass stoppers that were completely filled ('no gas phase') to prevent any loss of the test substance during the test.

The measured concentrations in the test vessels are considerably below the nominal concentrations (that is related to the WAF procedure), circa 10% was solved as determined by a sum parameter for hydrocarbons (specific peak of the IR at 2930 + 10 cm-1). During the test from t=0 to t=48h the concentrations decrease but remain > 80% of the start concentrations, showing that during the test the concentrations were stable. The result is expressed as a loading rate, the EL50. This value can be used for the classification and labelling of Orange oil (rel. 2).


For the alga test the test design requires exchange of air to ensure replenishment of CO2 during the test, so these vessels were not closed. Yet also in the alga test the concentrations remained stable during the test. However, we don't agree to the identification of the EC50 value of 210 mg/l: Due to the small growth rate effect (6%) at 12.5 mg/l (not detected in the biomass production), whereas at 25 and 50 mg/l no effects were observed, the EC50 is above the highest tested concentration (see graphs in report). Exclusion of the 6% would do justice to the two observed partial effect observations and decrease the EC50 to circa 150 mg/l. Thus for classification, loading rate EL50 ~ 150 mg/l. With this correction, the results for the algae test could still be classified as reliability 2.

The value of circa 150 mg/L is somehow out of range as compared to the results in other WAF studies. This may be be related to loss of test material during preparation of the WAFs. There is, however, no indication that algae would be more sensitive than daphnids and fish.


IBACON (Vinken and Wydra 2007) study on acute toxicity of Orange oil for Daphnia:

Study conducted with Orange oil (cold pressed 1-fold, Limonene content 95.1%), under GLP, GC-MS analytics included. The highest loading rates (down to 1 mg/L) were prepared as a true WAF, whereas the loading rates below 1 mg/L were diluted from there (in view of practical difficulties of preparing low level WAFs). However, the effects occurred in the range of the true WAF. Measures to prevent loss of the test material included the use of closed containers and precautions during the preparation of test media. The stability of the exposure was assessed by analyses (GC) of the concentration of limonene (rel. 2).

Fresenius (Lebertz 2011) studies on acute toxicity of Orange oil for Fish:

The toxicity of Orange oil (Orange terpenes, Limonene content 98.39 area%) to fish was assessed using Danio rerio according to OECD guideline 203 under GLP. Fish were exposed to WAFs between 1.0 and 10.0 mg/l in a semi-static setup for 96 hours. The concentration of hydrocarbons was measured. Validity criteria for the test were met. The acute toxicity (96h-EL50) of Orange oil towards Danio rerio with a WAF test is 5.65 mg/l (rel. 1).


Supporting evidence from QSAR studies

The toxicity of Orange oil to algae, daphnids and fish has also been investigated using the QSAR calculation method iSafeRat® that replaces an OECD 201 / 202 / 203 studies and considers the recommendations in the OECD Series on Testing and Assessment no 23 (i.e. WAF conditions). This method takes into account the bioavailable fraction per constituent in the WAF. The EC50s of each constituent are already known from literature or calculated using the iSafeRat QSAR model.

The results determined by the QSAR-WAF calculation method (range 2.8 – 4.8 mg Orange oil/L, rel. 2) are of the same order of magnitude to those determined in WAF laboratory experiments (all >1 and ranging between 1.1 and 8.6 mg/L for daphnids and fish). The predicted sensitivity for algae is in the same range. As this is a significantly lower result than the experimental WAF study (EL50 = circa 150 mg/L), the QSAR-WAF for algae (rel. 2) was identified as a key study for algae.

An overview of all valid results is presented in the table below:



Result (WAF, loading rates in mg/L)

Algae, Desmodesmus subspicatus

OECD 201

Lebertz 2007


72h-NOELr = 50 mg/L

72h-ErL50 ~150 mg/L

Algae, QSAR prediction

Thomas 2015


72h-EL50 = 4.3 – 4.8 mg/L

Daphnia magna

OECD 202

Lebertz 2007

48h-EL50 = 8.6 mg/L

Daphnia magna

OECD 202

Vinken & Wydra 2007


48h-EL50 = 1.1 mg/L

Daphnia magna, QSAR prediction

Thomas 2015

48h-EL50 = 3.9 – 4.3 mg/L

Fish, Danio rerio

Lebertz 2011


96h-EL50 = 5.65 mg/L

 Fish, QSAR prediction  Thomas 2015  96h-LL50 = 2.8 - 3.1 mg


Of fish, algae and Daphnia, the latter is the most sensitive to Orange oil based on an 48h-EL50 of 1.1 mg/l.

Additional information