Hydrocarbons, C18-C24, iso-alkanes <2% aromatics

Long-term toxicity to fish

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

fish short-term toxicity test on embryo and sac-fry stages

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

The study was conducted between 14 October 2009 and 18 November 2009.

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

other: see 'Remark'

Remarks:

Study conducted in compliance with agreed protocols, with no or minor deviations from standard test guidelines and/or minor methodological deficiencies, which do not affect the quality of the relevant results. The study report was conclusive, done to a valid guideline and the study was conducted under GLP conditions. Read-across is considered to be reliability 2.

Qualifier:

according to guideline

Guideline:

OECD Guideline 210 (Fish, Early-Life Stage Toxicity Test)

Deviations:

no

Qualifier:

according to guideline

Guideline:

other: US Code of Federal Regulations, Title 40, Part 797, Section 1600

Deviations:

no

Qualifier:

according to guideline

Guideline:

EPA OPPTS 850.1400 (Fish Early-life Stage Toxicity Test)

Deviations:

no

Principles of method if other than guideline:

Information provided by the Sponsor gave an EL/LL50 of greater than 100 mg/l. Therefore, as toxicity was not expected, the number of test concentrations was reduced from five, as recommended in the Test Guideline, to three (10, 32 and 100 mg/l loading rate WAF). This was considered not to affect the validity of the study and reducing the number of test concentrations also reduced the number of fish used which was consistent with the Sponsors Animal Welfare policy.

Furthermore In view of the difficulties associated with the evaluation of aquatic toxicity of poorly water soluble test materials, a modification of the standard method for the preparation of aqueous media was performed. An approach endorsed by several important regulatory authorities in the EU and elsewhere (ECETOC 1996 and OECD 2000), is to expose organisms to a Water Accommodated Fraction (WAF) of the test material in cases where the test material is a complex mixture and is poorly soluble in water and in the permitted auxiliary solvents and surfactants. Using this approach, aqueous media are prepared by mixing the test material with water for a prolonged period. Pre-study work showed that a preparation period of 24 hours was sufficient to ensure equilibration between the test material and water phase. At the completion of mixing, the test material phase is separated by siphon and the test organisms exposed to the aqueous phase or WAF (which may contain dissolved test material and/or leachates from the test material). Exposures are expressed in terms of the original concentration of test material in water at the start of the mixing period (loading rate) irrespective of the actual concentration of test material in the WAF.

GLP compliance:

yes (incl. QA statement)

Analytical monitoring:

yes

Vehicle:

no

Test organisms (species):

Pimephales promelas

Test type:

semi-static

Water media type:

freshwater

Limit test:

no

Total exposure duration:

33 d

Duration:

33 d

Dose descriptor:

NOELR

Effect conc.:

100 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: survival and growth

Remarks on result:

other: Water-accommodated fraction

Duration:

33 d

Dose descriptor:

other: Lowest Observed Effect Loading Rate” (LOEL)

Effect conc.:

> 100 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: There were no significant reductions (P ≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.

Remarks on result:

other: Water-accommodated fraction

Duration:

33 d

Dose descriptor:

other: No Observed Effect Loading Rate (NOEL)

Effect conc.:

>= 100 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: There were no significant reductions (P≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.

Remarks on result:

other: Water-accommodated fraction

RESULTS

Validation of Mixing Period

Pre-study investigational work (see attached Appendix 2) indicated that there was no significant increase in the amount of total organic carbon by extending the preparation period for longer than 24 hours. Therefore the test was conducted using a 23-Hour stirring period followed by a 1-Hour settlement period.

Definitive Test

Observations

The number of dead eggs observed during the definitive test are given in the attached Table 1. 

The number of dead larvae and hatched (live) larvae observed during the definitive test are given in

the attached Tables 2 and 3 respectively.

The number of hatched larvae (non-cumulative) per day are given in the attached Table 4.

The number of dead eggs and larvae were observed to be low throughout the duration of the test with no concentration dependent effects being observed. The mean hatching success rate for the control group was 92% thereby satisfying the validation criterion of greater than 75% hatching rate. The mean hatching rate for the test loading rates was 87% to 90%.

The mean survival rate of the larvae for the control group was 95% thereby satisfying the validation criterion for post-hatch survival success rate of greater than 70%. The mean survival rates for the test loading rates ranged between 81% and 98% (see attached Table 5).

The start of egg hatching was observed to be on Day 4 of the test and completion of hatching was observed on Day 5 of the test.

There were no significant mortalities or sub-lethal effects of exposure observed in any of the test concentrations.

Sub-lethal Effects

There were no sub-lethal effects observed in the test.

Length and dry weight data.

Statistical analysis of the length data by analysis of variance (see attached Appendix 4) showed the 10, 32 and 100 mg/l loading rate WAFs not to be significantly different (P≥0.05) from the control group. 

Statistical analysis of the dry weight data showed no significant differences (P≥0.05) between the control group and 32 and 100 mg/l loading rate WAF test groups. However, the 10 mg/l loading rate WAF test group was significantly different (P<0.05) from the control group.

The observed significant differences between the control group and the 10 mg/l loading rate WAF test group were considered to be due to the presence of a small number of larger fish in the 10 mg/l loading rate WAF test group. This was considered to be due to the normal variation observed in a natural population of organisms. The larval dry weight of the 10 mg/l loading rate WAF test group was not significantly reduced compared to the control group therefore this weight difference was considered not to affect the outcome of the study.

Given this information and data assessment above it was considered that no effect on survival or growth attributable to the test material was observed. 

Lowest observed effect loading rate.

The “Lowest Observed Effect Loading Rate” (LOEL), based on nominal loading rates, was considered to be greater than 100 mg/l loading rate WAF on the basis that there were no significant reductions (P≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.

No observed effect loading rate.

The “No Observed Effect Loading Rate” (NOEL), based on nominal loading rates, was considered to be 100 mg/l loading rate WAF on the basis that there were no significant reductions (P≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.

Vortex depth measurements.

The vortex depth was recorded at the start and end of each mixing period and was observed to be a dimple at the water surface on each occasion (see attached Table 6).

Observations on test material solubility.

Observations on the test media were carried out during the mixing and testing of the WAFs.

At the start and end of each mixing period, and after the 1-Hour settlement period the 10, 32 and 100 mg/l loading rates were observed to be clear, colourless water columns with an oily slick of test material at the surface. After siphoning and for the duration of the test, the 10, 32 and 100 mg/l loading rates were observed to be clear, colourless solutions. Microscopic inspection of the WAFs showed no micro-dispersions or undissolved test material to be present.

Physico-chemical measurements.

The results of the physico-chemical measurements are given in the attached Appendix 6. Temperature was maintained at 25ºC ± 2°C throughout the test, while there were no treatment related differences for oxygen concentration or pH. The temperature measurements recorded in Control Replicate R₁throughout the test by a Testo temperature logger are presented in the attached Figure 1.

The oxygen concentration in some of the test vessels was observed to have an air saturation value (ASV) in excess of 100%. This was considered to be due to the presence of microscopic air bubbles in the media super-saturating the diluent and was considered not to have had an impact on the outcome or integrity of the test as no adverse effects were observed.

The oxygen concentration for Day 12 old media control R₁, R₂and 10 mg/l LR WAF R₁were measured to be below the 60% Air Saturation Value given in the Protocol. However given that no adverse effects were observed throughout the test it was considered that this deviation had no impact on the outcome or validity of the test.

The water hardness values were observed to range from 138 to 142 mg/l as CaCO₃ at the start of the test and from 150 to 156 mg/l as CaCO₃ at termination of the test (see attached Appendix 7).

Total organic carbon analysis.

Total Organic Carbon (TOC) analysis (see attached Appendix 3) of the test preparations showed no significant differences in the amount of carbon present within the 10, 32 and 100 mg/l loading rate WAF test vessels when compared to the control vessels. This was expected due to the low aqueous solubility of the test material. Given the background level of carbon in the control vessels and also the low level of carbon in the test vessels, it was considered that the majority of the results were around the limit of quantitation of the analytical method.

The dissolved test material may have been one or several components of the test material. Given that toxicity cannot be attributed to a single component or mixture of components but to the test material as a whole, the results were based on nominal loading rates only.

Validity criteria fulfilled:

yes

Conclusions:

The application of the test material to newly laid eggs of fathead minnows was considered to have no effect on the survival or growth of the larvae.
The No Observed Effect Loading Rate was 100 mg/l loading rate WAF.

Executive summary:

Introduction.

A study was performed to assess the effects of the test material ‘Distillates (Fischer-Tropsch), heavy, C18-50 - branched, cyclic and linear’ on freshl

y hatched larvae of the fathead minnow (Pimephales promelas). The method followed that described in the OECD Guidelines for Testing of Chemicals (1992) No 210, "Fish, Early-Life Stage Toxicity Test”, US Code of Federal Regulations, Title 40, Part 797, Section 1600 and the US EPA Draft Ecological Effects Test Guideline OPPTS 850.1400.

Methods.

Based on data supplied by the Sponsor, newly laid eggs were exposed toWater Accommodated Fractions (WAFs) of the test material over a range of nominal loading rates of 10, 32 and 100 mg/lfor a period of 33 days at a temperature of
25 ºC ± 2°C under semi-static test conditions. 

The number of mortalities or any sub-lethal effects of exposure in each test and control vessel were recorded daily until termination of the test (28 days post-hatch). At test termination the length and dry weight of the surviving fish were measured.

Results. 

Over the duration of the test there were no significant mortalities or sub-lethal effects of exposure resulting from the exposure of fathead minnow (Pimephales promelas) larvae to nominal loading rates of 10, 32 and 100 mg/l loading rate WAF.

The mean hatching rate ranged from 87% to 92% and the mean survival rate ranged from 81% to 98%. The fish length and dry weight data obtained at termination of the test are summarised as follows:

	Nominal Loading Rate (mg/l)
	Control	10	32	100
Body Length (mean ± standard deviation, mm)	15.05 ± 1.16	15.37 ± 1.75	15.41 ± 1.06	15.34 ± 0.99
Dry Weight (mean ± standard deviation, mg)	11.7 ± 2.8	13.6 ± 4.7	12.4 ± 2.8	12.2 ± 2.8

Statistical analysis of these data showed there were no significant reductions (P≥0.05) between the control and all the test groups in terms of fish length or dry weight.

Conclusion.

Given the above results and information it was considered that the test material had no effect on the survival or growth of newly laid eggs of fathead minnows.

The “Lowest Observed Effect Loading Rate” (LOEL), based on nominal loading rates, was considered to be greater than 100 mg/l loading rate WAF on the basis that there were no significant reductions (P≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.

The “No Observed Effect Loading Rate” (NOEL), based on nominal loading rates, was considered to be 100 mg/l loading rate WAF on the basis that there were no significant reductions (P≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.

Total Organic Carbon (TOC) analysis of the test preparations showed no significant differences in the amount of carbon present within the 10, 32 and 100 mg/l loading rate WAF test vessels when compared to the control vessels. This was expected due to the low aqueous solubility of the test material. Given the background level of carbon in the control vessels and also the low level of carbon in the test vessels, it was considered that the majority of the results were around the limit of quantitation of the analytical method.

The dissolved test material may have been one or several components of the test material. Given that toxicity cannot be attributed to a single component or mixture of components but to the test material as a whole, the results were based on nominal loading rates only.

Reference 2

Endpoint:

fish early-life stage toxicity

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

2009-10-16 - 2009-12-16

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

OECD Guideline 210 (Fish, Early-Life Stage Toxicity Test)

Deviations:

no

Qualifier:

according to guideline

Guideline:

EPA OPPTS 850.1400 (Fish Early-life Stage Toxicity Test)

Deviations:

no

GLP compliance:

yes (incl. QA statement)

Remarks:

Date of GLP inspection: 15 September2009 Date if signature on GLP certificate: 26 November 2009

Analytical monitoring:

no

Details on sampling:

Chemical analysis of test loading rates
The test material concentrations in the test preparations were not determined by compound specific analysis at the request of the Sponsor.

Total organic carbon analysis
Analysis of the WAFs was carried out by Total Organic Carbon (TOC) analysis. Water samples were taken from the control and each test group (Replicates R1 and R2 pooled) on Days 0 (fresh media), 4, 7, 11, 14, 18, 21, 25, 28, 32 (old and fresh media) and 33 (old media). Duplicate samples were taken and stored at approximately -20 ºC for further analysis if necessary.

A semi-static test regime was employed in the test involving renewal of the test preparations twice weekly to ensure that the concentrations of the test material remained near nominal and to prevent the build-up of nitrogenous waste products.

Vehicle:

no

Details on test solutions:

Validation of mixing period
Pre-study investigational work was carried out to determine whether stirring for a prolonged period produced significantly higher levels of total organic carbon, as an indicator of soluble organic substances, in the WAF.
A WAF of nominal loading rate of 100 mg/L was prepared, in duplicate, in dechlorinated tap water. One loading rate was stirred for a period of 23 hours and the other for a period of 95 hours. After a 1-Hour standing period the mixtures were then removed by siphon and samples taken for Total Organic Carbon analysis.

Definitive test
Information provided by the Sponsor gave an EL/LL50 of greater than 100 mg/L. Therefore, as toxicity was not expected, the number of test concentrations was reduced from five, as recommended in the Test Guideline, to three (10, 32 and 100 mg/L loading rate WAF). This was considered not to affect the validity of the study and reducing the number of test concentrations also reduced the number of fish used which was consistent with the Sponsors Animal Welfare policy.

Experimental Preparation
Due to the low aqueous solubility and complex nature of the test material for the purposes of the definitive test the test material was prepared as a Water Accommodated Fraction (WAF).
Amounts of test material (110, 352 and 1100 mg) were each separately added to the surface of 11 L of dechlorinated tap water to give the 10, 32 and 100 mg/L loading rates respectively. The stirring vessels were sealed with minimal headspace to reduce losses due to the possible volatile nature of the test material. After the addition of the test material, the dechlorinated tap water was stirred by magnetic stirrer using a stirring rate such that a vortex was formed to give a dimple at the water surface. The stirring was stopped after 71 hours and the mixtures allowed to stand for 1 hour. A wide bore glass tube, covered at one end with Nescofilm was submerged into the vessel, sealed end down, to a depth of approximately 5 cm from the bottom of the vessel. A length of Tygon tubing was inserted into the glass tube and pushed through the Nescofilm seal. The aqueous phase or WAF was removed by mid-depth siphoning (the first 75-100 mL discarded) to give the 10, 32 and 100 mg/L loading rate WAFs. Microscopic inspection of the WAFs showed no micro-dispersions or undissolved test material to be present.
Following siphoning the WAFs were dispensed to two test vessels to give replicates R1 and R2.
The control group was maintained under identical conditions but not exposed to the test material.
Total Organic Carbon (TOC) analysis was performed on the fresh media on Day 0, old and fresh media on Days 4, 7, 11, 14, 18, 21, 25, 28, 32 and old media on Day 33.

Vortex depth measurements
The vortex depth was recorded at the start and end of each mixing period.

Total organic carbon analysis
Analysis of the WAFs was carried out by Total Organic Carbon (TOC) analysis. Water samples were taken from the control and each test group (Replicates R1 and R2 pooled) on Days 0 (fresh media), 4, 7, 11, 14, 18, 21, 25, 28, 32 (old and fresh media) and 33 (old media). Duplicate samples were taken and stored at approximately -20 ºC for further analysis if necessary.


Exposure conditions
Glass exposure vessels containing approximately 300 mL of test media were used for each control and test vessel. The test volume was increased to 2 L on Day 14 of the test to accommodate the growing larvae. The test vessels were filled with minimal headspace and covered to reduce losses through volatility. At the start of the test 30 eggs were placed in each test vessel. Egg baskets were not used due to the small volumes used and to minimise bacterial build up.
The number of fertilised eggs introduced per concentration was 60 ie. 30 per replicate.
The control group was maintained under identical conditions but not exposed to the test material.
A semi-static test regime was employed in the test involving renewal of the test preparations twice weekly to ensure that the concentrations of the test material remained near nominal and to prevent the build-up of nitrogenous waste products.
The test vessels were maintained at 25ºC ± 2°C with a photoperiod of 16 hours light and 8 hours darkness with 20 minute dawn and dusk transition periods throughout the duration of the test. The eggs and larvae were not individually identified.
The start of hatching was observed on Day 4 of the test and completion of hatching on Day 5. The larvae were fed protozoan (Paramecia micronucleatum) only from Day 5 to Day 7 as the larvae were too small at this time to feed on brine shrimp nauplii. The protocol stated that the larvae would be fed both protozoan (Paramecia micronucleatum) and brine shrimp nauplii on one day to avoid a sudden change from one food source to another. This was not carried out in error. This was considered not to affect the test as the larvae were considered sufficiently large to eat brine shrimp nauplii and no adverse effects were observed. On Day 8 and throughout the remainder of the test the larvae were fed brine shrimp nauplii only.
The number of dead eggs (up to completion of hatching), dead and live larvae and sub-lethal effects of exposure were recorded daily. The criteria of death for eggs were marked loss of translucency and change in coloration leading to a white opaque appearance. The criteria of death for larvae and juvenile fish were one or more of the following: immobility, absence of respiratory movement, absence of heart beat, white opaque coloration and lack of reaction to mechanical stimulus.

Test organisms (species):

Pimephales promelas

Details on test organisms:

Test Species
The test was carried out using freshly laid eggs of fathead minnows (Pimephales promelas). The in­house breeding stock fish were from a batch bred on 30 June 2009 and were maintained in dechlorinated tap water in glass tanks with an activated carbon and biological filtration system.
The lighting cycle was controlled to give a 16 hours light and 8 hours darkness cycle with 20 minute dawn and dusk transition periods. The water temperature was controlled at approximately 25 ºC with a dissolved oxygen content of greater than or equal to 8.0 mg O2/L. The breeding stock fish were fed ZM 400 flake food daily.
Each breeding tank was supplied with inverted plastic guttering for the fish to lay eggs on and be fertilised. Fertilised eggs were collected from the breeding tanks on 12 November 2009 and used for the definitive test. The eggs were less than 24 hours old on introduction into the test system.
The diet and diluent water are considered not to contain any contaminant that would affect the integrity and outcome of the study.

Test type:

semi-static

Water media type:

freshwater

Limit test:

yes

Total exposure duration:

33 d

Hardness:

The water hardness in each vessel was measured at the start and on termination of the test and was determined using the methods described in Fields and On-Site Methods for Analysis of Water (British Standard Institution 1993).
The water hardness values were observed to range from 134 to 144 mg/L as CaCO3 at the start of the test and from 132 to 140 mg/L as CaCO3 at termination of the test.

Test temperature:

Temperature was maintained at 25 ± 2 °C throughout the test.

pH:

There were no treatment related differences for pH.
The pH and the dissolved oxygen concentration were measured using a WTW pH/Oxi 340I pH and dissolved oxygen meter.

Dissolved oxygen:

There were no treatment related differences for oxygen concentration.
The pH and the dissolved oxygen concentration were measured using a WTW pH/Oxi 340I pH and dissolved oxygen meter.

Salinity:

Freshwater used.

Nominal and measured concentrations:

Nominal loading rates of 10, 32 and 100 mg/L.
The test material concentrations in the test preparations were not determined by compound specific analysis at the request of the Sponsor.

Total organic carbon analysis
Analysis of the WAFs was carried out by Total Organic Carbon (TOC) analysis. Water samples were taken from the control and each test group (Replicates R1 and R2 pooled) on Days 0 (fresh media), 4, 7, 11, 14, 18, 21, 25, 28, 32 (old and fresh media) and 33 (old media). Duplicate samples were taken and stored at approximately -20 ºC for further analysis if necessary.

Details on test conditions:

Physico-chemical measurements
The water temperature, pH, light intensity and dissolved oxygen concentrations were recorded daily throughout the test. The measurements on Day 0, and after each test media renewal, represent those of the freshly prepared test concentrations while the measurements taken prior to each test media renewal and on termination of the test represent those of the used or old test preparations. The pH and the dissolved oxygen concentration were measured using a WTW pH/Oxi 340I pH and dissolved oxygen meter and the temperature was measured using a Hanna Instruments HI 93510 digital thermometer. The temperature was also monitored approximately every hour in Control Replicate R1 using a Testo temperature logger. The light intensity was measured using a digital luxmeter.
The water hardness in each vessel was measured at the start and on termination of the test and was determined using the methods described in Fields and On-Site Methods for Analysis of Water (British Standard Institution 1993).

The test was carried out using freshly laid eggs of fathead minnows (Pimephales promelas). The in­house breeding stock fish were from a batch bred on 30 June 2009 and were maintained in dechlorinated tap water in glass tanks with an activated carbon and biological filtration system.

Exposure conditions
Glass exposure vessels containing approximately 300 mL of test media were used for each control and test vessel. The test volume was increased to 2 L on Day 14 of the test to accommodate the growing larvae. The test vessels were filled with minimal headspace and covered to reduce losses through volatility. At the start of the test 30 eggs were placed in each test vessel. Egg baskets were not used due to the small volumes used and to minimise bacterial build up.
The number of fertilised eggs introduced per concentration was 60 ie. 30 per replicate.
The control group was maintained under identical conditions but not exposed to the test material.
A semi-static test regime was employed in the test involving renewal of the test preparations twice weekly to ensure that the concentrations of the test material remained near nominal and to prevent the build-up of nitrogenous waste products.
The test vessels were maintained at 25ºC ± 2°C with a photoperiod of 16 hours light and 8 hours darkness with 20 minute dawn and dusk transition periods throughout the duration of the test. The eggs and larvae were not individually identified.
The start of hatching was observed on Day 4 of the test and completion of hatching on Day 5. The larvae were fed protozoan (Paramecia micronucleatum) only from Day 5 to Day 7 as the larvae were too small at this time to feed on brine shrimp nauplii. The protocol stated that the larvae would be fed both protozoan (Paramecia micronucleatum) and brine shrimp nauplii on one day to avoid a sudden change from one food source to another. This was not carried out in error. This was considered not to affect the test as the larvae were considered sufficiently large to eat brine shrimp nauplii and no adverse effects were observed. On Day 8 and throughout the remainder of the test the larvae were fed brine shrimp nauplii only.
The number of dead eggs (up to completion of hatching), dead and live larvae and sub-lethal effects of exposure were recorded daily. The criteria of death for eggs were marked loss of translucency and change in coloration leading to a white opaque appearance. The criteria of death for larvae and juvenile fish were one or more of the following: immobility, absence of respiratory movement, absence of heart beat, white opaque coloration and lack of reaction to mechanical stimulus.

Test Water
The test water used for the definitive test was laboratory tap water dechlorinated by passage through an activated carbon filter (Elga AC1) and partly softened (Elga Nimbus 1248D Duplex Water Softener) giving water with a total hardness of approximately 140 mg/l as CaCO3. After dechlorination and softening the water was passed through a series of computer controlled plate heat exchangers to achieve the required temperature.

Reference substance (positive control):

no

Duration:

33 d

Dose descriptor:

NOELR

Effect conc.:

100 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: There were no significant reductions (P≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.

Remarks on result:

other: Water-accommodated fraction

Duration:

33 d

Dose descriptor:

LOELR

Effect conc.:

> 100 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: There were no significant reductions (P≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.

Remarks on result:

other: Water-accommodated fraction

Details on results:

The “Lowest Observed Effect Loading Rate” (LOEL), based on nominal loading rates, was considered to be greater than 100 mg/L loading rate WAF on the basis that there were no significant reductions (P≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.
The “No Observed Effect Loading Rate” (NOEL), based on nominal loading rates, was considered to be 100 mg/L loading rate WAF on the basis that there were no significant reductions (P≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.

Validation of Mixing Period
Pre-study investigational work indicated that there was an increase in the amount of total organic carbon by extending the preparation period from 24 to 96 hours. At the request of the Sponsor a preparation period of 72 hours was used to maintain consistency with other studies conducted on the test material.

Definitive Test

Observations
The number of dead eggs and larvae were observed to be low throughout the duration of the test with no concentration dependent effects being observed. The mean hatching success rate for the control group was 94% thereby satisfying the validation criterion of greater than 75% hatching rate. The mean hatching rate for the test loading rates was 92% to 95%.
The mean survival rate of the larvae for the control group was 98% thereby satisfying the validation criterion for post-hatch survival success rate of greater than 70%. The mean survival rates for the test loading rates ranged between 93% and 97% .
The start of egg hatching was observed to be on Day 4 of the test and completion of hatching was observed on Day 5 of the test.
There were no significant mortalities or sub-lethal effects of exposure observed in any of the test concentrations.

Sub-lethal Effects
There were no sub-lethal effects observed in the test.

Length and dry weight data
Statistical analysis of the length and dry weight data by analysis of variance showed the 32 and 100 mg/L loading rate WAFs not to be significantly different (P≥0.05) from the control group.
However, the 10 mg/L loading rate WAF test group was significantly different (P<0.05) from the control group in that the larvae were greater in length and dry weight compared to the control.
The observed significant differences between the control group and the 10 mg/L loading rate WAF test group were considered to be due to the presence of a small number of larger fish in the 10 mg/L loading rate WAF test group. This was considered to be due to the normal variation observed in a natural population of organisms. The larval length and dry weight of the 10 mg/L loading rate WAF test group was not significantly reduced compared to the control group therefore these differences were considered not to affect the outcome of the study.
Given this information and data assessment above it was considered that no effect on survival or growth attributable to the test material was observed.

Lowest observed effect loading rate
The “Lowest Observed Effect Loading Rate” (LOEL), based on nominal loading rates, was considered to be greater than 100 mg/L loading rate WAF on the basis that there were no significant reductions (P≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.

No observed effect loading rate
The “No Observed Effect Loading Rate” (NOEL), based on nominal loading rates, was considered to be 100 mg/L loading rate WAF on the basis that there were no significant reductions (P≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.

Vortex depth measurements
The vortex depth was recorded at the start and end of each mixing period and was observed to be a dimple at the water surface on each occasion.

Observations on test material solubility
Observations on the test media were carried out during the mixing and testing of the WAFs.
At the start of each mixing period the 10, 32 and 100 mg/L loading rates were observed to be clear, colourless water columns with test material floating on the surface. At the end of each mixing period, and after the 1-Hour settlement period the 10, 32 and 100 mg/L loading rates were observed to be clear, colourless water columns with a thin layer of test material at the surface. After siphoning and for the duration of the test, the 10, 32 and 100 mg/L loading rates were observed to be clear, colourless solutions. Microscopic inspection of the WAFs showed no micro-dispersions or undissolved test material to be present.

Physico-chemical measurements
Temperature was maintained at 25ºC ± 2°C throughout the test, while there were no treatment related differences for oxygen concentration or pH.
The oxygen concentration in some of the test vessels was observed to have an air saturation value (ASV) in excess of 100%. This was considered to be due to the presence of microscopic air bubbles in the media super-saturating the diluent and was considered not to have had an impact on the outcome or integrity of the test as no adverse effects were observed.
The water hardness values were observed to range from 134 to 144 mg/L as CaCO3 at the start of the test and from 132 to 140 mg/L as CaCO3 at termination of the test.

Total organic carbon analysis
Total Organic Carbon (TOC) analysis of the test preparations showed no significant differences in the amount of carbon present within the 10, 32 and 100 mg/L loading rate WAF test vessels when compared to the control vessels. This was expected due to the low aqueous solubility of the test material. Given the background level of carbon in the control vessels and also the low level of carbon in the test vessels, it was considered that the majority of the results were around the limit of quantitation of the analytical method.
The dissolved test material may have been one or several components of the test material. Given that toxicity cannot be attributed to a single component or mixture of components but to the test material as a whole, the results were based on nominal loading rates only.

Reported statistics and error estimates:

Evaluation of data
For the estimation of the “Lowest Observed Effect Loading Rate” (LOEL) and the “No Observed Effect Loading Rate” (NOEL) the length and dry weight data obtained on termination of the test for the control and each test group were compared using one way analysis of variance incorporating Bartlett’s test for homogeneity of variance (Sokal and Rohlf 1981) and Dunnett’s multiple comparison procedure for comparing several treatments with a control (Dunnett, 1955). All statistical analyses were performed using the SAS computer software package (SAS, 1999 - 2001).

Please see Attachment section for:

Table 1          Number of Dead Eggs in the Definitive Test                                                

Table 2          Number of Dead Larvae in the Definitive Test                                             

Table 3          Cumulative Number of Hatched (Live) Larvae in the Definitive Test      

Table 4          Number of Hatched (Live) Larvae (Non-cumulative) in the Definitive Test

Table 5          Hatching and Survival Rates in the Definitive Test                                    

Table 6          Vortex Depth Measurements at the Start and End of Each Mixing Period  

Figure 1        Temperature Measurements Recorded Over the Duration of the Test Using a Using a Testo Data Logger.

                                                                                               

Appendix 1   Typical Water Quality Characteristics                                                            

Appendix 2   Validation of Mixing Period                                                                                           

Appendix 4   Statistical Analysis                                                                                            

Appendix 5   Fish Length and Weight Values                                                                     

Appendix 6   Physico-Chemical Measurements                                                                 

Appendix 7   Water Hardness Values                                                                                   

Appendix 3     Total Organic Carbon Analysis

RESULTS

Samples	Nominal Loading Rate (mg/L)	Concentration of TOC (mg C/L)	Concentration of TOC Corrected for Control (mg C/L)
Day 0 (Fresh Media)	Control	2.03	-
	10	2.67	0.64
	32	2.93	0.90
	100	2.27	0.24
Day 4 (Old Media)	Control	3.55	-
	10	3.81	0.26
	32	2.87	<Control
	100	3.40	<Control
Day 4 (Fresh Media)	Control	1.74	-
	10	1.61	<Control
	32	2.10	0.36
	100	2.44	0.70
Day 7 (Old Media)	Control	2.32	-
	10	2.59	0.27
	32	2.23	<Control
	100	2.75	0.43
Day 7 (Fresh Media)	Control	1.37	-
	10	1.89	0.52
	32	1.72	0.35
	100	2.43	1.06

 

 Appendix 3 (continued)  Total Organic Carbon Analysis

Samples	Nominal Loading Rate (mg/L)	Concentration of TOC (mg C/L)	Concentration of TOC Corrected for Control (mg C/L)
Day 11 (Old Media)	Control	6.21	-
	10	5.79	<Control
	32	6.33	0.12
	100	5.83	<Control
Day 11 (Fresh Media)	Control	1.17	-
	10	1.71	0.54
	32	1.34	0.17
	100	2.69	1.52
Day 14 (Old Media)	Control	8.87	-
	10	8.34	<Control
	32	7.95	<Control
	100	7.86	<Control
Day 14 (Fresh Media)	Control	3.32	-
	10	6.65	3.33
	32	5.61	2.29
	100	5.89	2.57
Day 18 (Old Media)	Control	7.24	-
	10	8.07	0.83
	32	8.15	0.91
	100	6.92	<Control
Day 18 (Fresh Media)	Control	2.95	-
	10	2.45	<Control
	32	3.66	0.71
	100	4.18	1.23
Day 21 (Old Media)	Control	5.74	-
	10	5.30	<Control
	32	5.97	0.23
	100	6.25	0.51

 

Appendix 3 (continued)   Total Organic Carbon Analysis

Samples	Nominal Loading Rate (mg/L)	Concentration of TOC (mg C/L)	Concentration of TOC Corrected for Control (mg C/L)
Day 21 (Fresh Media)	Control	3.00	-
	10	24.26	21.26[1]
	32	7.47	4.47
	100	8.05	5.05
Day 25 (Old Media)	Control	9.18	-
	10	7.86	<Control
	32	7.43	<Control
	100	8.43	<Control
Day 25 (Fresh Media)	Control	3.18	-
	10	7.36	4.18
	32	8.23	5.05
	100	5.27	2.09
Day 28 (Old Media)	Control	7.83	-
	10	7.11	<Control
	32	8.10	0.27
	100	8.28	0.45
Day 28 (Fresh Media)	Control	3.64	-
	10	3.99	0.35
	32	2.26**	<Control
	100	3.46	<Control
Day 32 (Old Media)	Control	6.73	-
	10	6.62	<Control
	32	5.74	<Control
	100	7.04	0.31
Day 32 (Fresh Media)	Control	2.99**	-
	10	2.80	<Control
	32	2.78	<Control
	100	3.77	0.78
Day 33 (Old Media)	Control	4.39	-
	10	3.92	<Control
	32	3.70	<Control
	100	3.57	<Control

[1]High value considered to be due to contamination from an unknown source. Frozen duplicate sample gave a similar result.

**Duplicate sample analysis as original sample result was deemed to be anomalous.

   

Validity criteria fulfilled:

yes

Conclusions:

The application of the test material to newly laid eggs of fathead minnows was considered to have no effect on the survival or growth of the larvae. The No Observed Effect Loading Rate was 100 mg/L loading rate WAF.

Executive summary:

Introduction.

A study was performed to assess the effects of the test material on freshly hatched larvae of the fathead minnow (Pimephales promelas). The method followed that described in the OECD Guidelines for Testing of Chemicals (1992) No 210, "Fish, Early-Life Stage Toxicity Test”, US Code of Federal Regulations, Title 40, Part 797, Section 1600 and the US EPA Draft Ecological Effects Test Guideline OPPTS 850.1400.

Methods.

Based on data supplied by the Sponsor, newly laid eggs were exposed to Water Accommodated Fractions (WAFs) of the test material over a range of nominal loading rates of 10, 32 and 100 mg/L for a period of 33 days at a temperature of 25ºC±2°C under semi-static test conditions. Due to the possible volatile nature of the test material, the test was conducted using completely filled test vessels which were then covered. The number of mortalities or any sub-lethal effects of exposure in each test and control vessel were recorded daily until termination of the test (28 days post-hatch). At test termination the length and dry weight of the surviving fish were measured.

Results.

Over the duration of the test there were no significant mortalities or sub-lethal effects of exposure resulting from the exposure of fathead minnow (Pimephales promelas) larvae to nominal loading rates of 10, 32 and 100 mg/L loading rate WAF.

The mean hatching rate ranged from 92% to 95% and the mean survival rate ranged from 93% to 98%. The fish length and dry weight data obtained at termination of the test are summarised as follows:

	Nominal Loading Rate (mg/L)
	Control	10	32	100
Body Length (mean ± standard deviation, mm)	17.13 ± 1.29	17.90 ± 1.17	17.47 ± 1.05	17.55 ± 1.29
Dry Weight (mean ± standard deviation, mg)	18.0 ± 4.4	20.4 ± 4.7	19.9 ± 4.4	19.6 ± 5.1

Statistical analysis of these data showed there were no significant reductions (P≥0.05) between the control and all the test groups in terms of fish length or dry weight.

Conclusion.

Given the above results and information it was considered that the test material had no effect on the survival or growth of newly laid eggs of fathead minnows.

The “Lowest Observed Effect Loading Rate” (LOEL), based on nominal loading rates, was considered to be greater than 100 mg/L loading rate WAF on the basis that there were no significant reductions (P≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.

The “No Observed Effect Loading Rate” (NOEL), based on nominal loading rates, was considered to be 100 mg/l loading rate WAF on the basis that there were no significant reductions (P≥0.05) in terms of fish length and dry weight when compared to the control at the end of the test.

Total Organic Carbon (TOC) analysis of the test preparations showed no significant differences in the amount of carbon present within the 10, 32 and 100 mg/L loading rate WAF test vessels when compared to the control vessels. This was expected given the extremely low aqueous solubility of the test material. Given the background level of carbon in the control vessels and also the low level of carbon in the test vessels, it was considered that the majority of the results were around the limit of quantitation of the analytical method.

The dissolved test material may have been one or several components of the test material. Given that toxicity cannot be attributed to a single component or mixture of components but to the test material as a whole, the results were based on nominal loading rates only.

The application of the test material to newly laid eggs of fathead minnows was considered to have no effect on the survival or growth of the larvae. The No Observed Effect Loading Rate was 100 mg/L loading rate WAF.

Reference 3

Endpoint:

fish early-life stage toxicity

Type of information:

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

Adequacy of study:

weight of evidence

Justification for type of information:

1. Hypothesis for the analogue approach:
The hypothesis for the analogue approach is that the test substance GTL Base Oil Distillates (C18-C50 branched, cyclic and linear hydrocarbons – Distillates / 848301-69-9 / 482-220-0) is produced in the same Fischer-Tropsch process as GTL Gasoil which is the starting material from which the registration substance is produced by fractional distillation. The source substance contains constituents of the target substance, Hydrocarbons, C18-C24, iso-alkanes, <2% aromatics, although it covers a wider carbon number distribution. The substances therefore have qualitatively similar properties (RAAF Scenario 2 applies). See Endpoint Summary (CSR Section 7.1.1.2 for additional information).
2. Source and target chemical(s)
The source substance GTL Base Oil Distillates (Distillates (Fischer-Tropsch), heavy, C18-50 - branched, cyclic and linear) is composed of linear, branched and cyclic hydrocarbons of chain length C18-C50.
The source substance GTL Gasoil (C8-C26) (C8-C26 branched and linear hydrocarbons – Distillates) is the substance from which the registration substance is produced. GTL Gasoil is composed of linear, branched and cyclic hydrocarbons of chain length C8-C26.
The target substance, Hydrocarbons, C18-C24, iso-alkanes, <2% aromatics, is composed of linear, branched and cyclic hydrocarbons of chain length C18-24.
3. Analogue approach justification
The constituents of the source and target substances are all hydrocarbons. Identical constituents have identical toxicity profiles. There is no evidence of toxicity in the studies on the source substance, and no evidence of toxicity in studies on hydrocarbons with shorter carbon chain length and a narrower range of carbon number, therefore there is no evidence for mixture effects or for increased toxicity with increased concentration.

Reason / purpose for cross-reference:

read-across source

Reason / purpose for cross-reference:

read-across source

Reason / purpose for cross-reference:

read-across: supporting information

Duration:

33 d

Dose descriptor:

NOELR

Effect conc.:

100 mg/L

Description of key information

GTL Gasoil; (33-d) NOELR for freshwater fish (Pimephales promelas): ≥100 mg/l P. promelas. [OECD 210; test mat. (WAFs) (nominal)

GTL Base Oil Distillates; (33-d) NOELR for freshwater fish (Pimephales promelas): ≥100 mg/l P. promelas. [OECD 210; test mat. (WAFs) (nominal)

Key value for chemical safety assessment

Fresh water fish

Fresh water fish

Effect concentration:

100 mg/L

Additional information

GTL Gasoil

A Fish Early Life Stage study has been conducted for with GTL Gasoil using the recommended freshwater species Pimephales promelas (Fathead minnow) for GTL Gasoil (Priestly, 2010a). The test was conducted in accordance with OECD 210 (1992) and in compliance with GLP.

Newly laid eggs from laboratory cultures were exposed for a period of 33 days to water accommodated fractions (WAFs) of the test material over a range of nominal loading rates of 10, 32 and 100 mg/l at a temperature of 25°C ± 2°C under semi-static test conditions. As the test material is volatile, the test was conducted using completely full test vessels which were covered. Exposure to the WAFs confirmed using total organic carbon (TOC) analysis.

Throughout the duration of the study, no significant (P ≥0.05) effects on mortality or sub-lethal effects (fish length, dry weight) as a result of exposure of fathead minnow to any of the nominal loading rate WAFs compared to the control were observed. The NOELR, based on nominal loading rate was determined to be 100 mg/L.  

A supporting Fish Early Life Stage study, specifically for Chinese notification, was conducted with GTL Gasoil, in accordance with the OECD 210 (2013) test guideline (Yang, 2015a). Newly laid eggs from laboratory cultures were exposed for a period of 33-days to a limit test loading rate WAF of 100 mg/L. The study was conducted at the specified temperature conditions for this species (25°C ± 2°C) and photoperiod of 12:12 hours light:dark, under semi-static test conditions as per the OECD 210 guidance. The study was compliant with the Chinese Environmental Protection Industry Standard ‘Guidelines of chemical testing good laboratory practices’ (HJ/T 153-2004, SEPA).  

Throughout the duration of the study, no significant (P >=0.05) mortality or sub-lethal effects (fish length, dry weight) as a result of exposure of zebra fish (Danio rerio) larvae to the nominal loading rate WAFs of 100 mg/L was observed.

Justification for Key Study Chronic Toxicity Testing for Fish

The key study (Priestly, 2010) and the supporting study (Yang, 2015a) are considered reliable as they were both performed using the OECD testing guidelines (OECD 210 TG) and did not deviate from the study criteria. The Priestly (2009) study is considered key because it was a fully GLP study and is given a reliability score corresponding to Klimisch 1. The supporting study, which was conducted specifically for Chinese registration, was only compliant with the Chinese State Environmental Protection Administration (SEPA 2004)  requirements rather than GLP.

The Yang (2015a) study provides evidence to support the key study and is given a reliability score corresponding to Klimisch 2. The data from both studies are sufficient and well documented to be scientifically acceptable, the Priestly (2010) study is considered key, and the Yang (2015) study as supporting.  

Discussion on methodology

In undertaking chronic toxicity testing the best use of all available data should be made. However, the assessment of ecotoxicity data for many petroleum products is complicated as products are UVBCs containing many individual substances with a range of solubilities. Chemical analyses of the aqueous concentrations of all constituents are not possible due to the complexity of the composition.

The concentration of each constituent dissolved in the water phase at any particular ‘loading’ should be maximized. The maximum possible water concentration of each component is typically achieved through prolonged stirring of the water-petroleum substance mixture to produce a Water Accommodated Fraction (WAF). WAFs are prepared individually and not by serial dilution of a single stock solution.  In addition, a sealed system approach may be appropriate for more volatile petroleum substances.

GTL Gasoil is a poorly water soluble UVBC and poses specific challenges when preparing aqueous solutions for toxicity testing.  This substance contains constituents with a range of physio-chemical properties (e.g. volatility, water solubility) and fall under the OECD guidance document 23 (2nd Edition) description and definition of “difficult to test” substances (Organisation for Economic Co-operation and Development [OECD] 2019). All long-term aquatic studies with GTL Gas oil followed the recommendations laid out for such substances.

Experimental and predicted (Target Lipid Model) data for constituents of GTL Gasoil are discussed in the PBT report attached in Section 13 of IUCLID. The conclusions based on this data are set out in Section 8.1.1.1.3 (Toxicity assessment) of this CSR.

GTL Base Oil Distillates

A Fish Early Life Stage study has been conducted with GTL Base Oil Distillates using the freshwater species Pimephales promelas (fathead minnow) (Priestly, 2010b). The test was conducted in accordance with OECD 201 and was compliant with GLP. TOC content was analysed in the test media for stability and exposure concentration.

Newly laid eggs were exposed to water accommodated fractions (WAFs) of the test material over a range of nominal loading rates of 10, 32 and 100 mg/l for a period of 33 days at a temperature of 25°C ± 2°C under semi-static test conditions. 

There were no significant mortalities or sub-lethal effects at any loading rate. The mean hatching rate ranged from 87 to 92% and the mean survival rate ranged from 81 to 98%. There were no statistically significant reductions in mean fish length or dry weight between the control group and the 32 and 100 mg/l WAFs loading rate. However the observations at the 10 mg/l loading rate WAF were significantly different from the control. This was considered by the authors to be due to normal biological variation. The report concluded that there were no effects on survival or growth of newly laid eggs of fathead minnows when exposure to loading rates up to 100 mg/l GTL Distillates. The No Observed Effect Loading Rate (NOELR) was therefore determined to be at least 100 mg/l.

Total Organic Carbon (TOC) analysis of the freshly prepared test preparations showed amount of carbon present within the 10, 32 and 100 mg/l loading rate WAF test vessels to range from less than the limit of quantitation (LOQ) for the method (1.0 mg C/l) to 11.64 mg C/l. TOC analysis of the old test media showed amount of carbon present to range from less than the LOQ to 2.75 mg C/l There was no significant difference in TOC content between control and exposure vessels. This was expected due to the low water solubility of the test substance. 

Conclusion

Long-term fish toxicity tests are available for two GTL-derived substances in the relevant carbon number range for the registered substance. In the absence of adverse effects in these studies it can be concluded that, based on weight of evidence, the NOELR(freshwater fish) for Hydrocarbons, C18-C24, isoalkanes, <2% aromatics is ≥100 mg/l.