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

Endpoint summary

Description of key information

A logistic function QSAR model has been used to predict BCF values for the constituents of Hydrocarbons, C18-C24, isoalkanes, <2% aromatics (Peter Fisk Associates, 2021b). Use of the logistic function QSAR model, predicts a worst-case scenario for high log Kow constituents as it assumes a plateau at log BCF 3.3 (BCF 2000) for constituents with log Kow >4.3 (equivalent to carbon chain length ~9). 

Substances in the QSAR training set have measured log BCF values between 1.85 and 3.77, predicted log BCF values between 1.8 and 3.3, and log K_ow values (measured and predicted) between 3.44 and 8.25. Measured log BCF values for compounds with log K_ow >6 are systematically lower than the predicted value. The fit for BCF values derived from measured BMF values in this region is a lot better.

Additional information

As shown in Table 4.7, the logistic function QSAR model indicates that all of the constituents may reach the criteria for B. There are no vB constituents.

Predictions for cyclic structures and those with quaternary carbons at carbon numbers are not presented. Higher cyclic structures and those with quaternary carbons with a carbon number higher than 8 to 10 all have predicted log BCF values of 3.32.

Further details are available in the attached QMRF report (Peter Fisk Associates, 2021a) and QPRF report (Peter Fisk Associates, 2021b), and in the supporting technical basis report attached in Section 13 (Peter Fisk Associates, 2021.

Table 4.7 Predicted BCF values and assessment of bioaccumulation (B) for constituents of Hydrocarbons, C18-C24, isoalkanes, <2% aromatics

Carbon No	Linear/Branched	log Kow	log BCF (predicted – logistic QSAR)	BCF (predicted – logistic QSAR) (l/kg)	B/vB
18	Linear	9.18	3.32	2090	B
18	1-branched	9.11	3.32	2090	B
18	2-branched	9.03	3.32	2090	B
18	3-branched	8.96	3.32	2090	B
18	4-branched	8.89	3.32	2090	B
19	Linear	9.67	3.32	2090	B
19	1-branched	9.6	3.32	2090	B
19	2-branched	9.53	3.32	2090	B
19	3-branched	9.45	3.32	2090	B
19	4-branched	9.38	3.32	2090	B
20	Linear	10.16	3.32	2090	B
20	1-branched	10.09	3.32	2090	B
20	2-branched	10.02	3.32	2090	B
20	3-branched	9.94	3.32	2090	B
20	4-branched	9.87	3.32	2090	B
21	Linear	10.65	3.32	2090	B
21	1-branched	10.58	3.32	2090	B
21	2-branched	10.51	3.32	2090	B
21	3-branched	10.43	3.32	2090	B
21	4-branched	10.36	3.32	2090	B
22	Linear	11.15	3.32	2090	B
22	1-branched	11.07	3.32	2090	B
22	2-branched	11.00	3.32	2090	B
22	3-branched	10.93	3.32	2090	B
22	4-branched	10.85	3.32	2090	B
23	Linear	11.64	3.32	2090	B
23	1-branched	11.56	3.32	2090	B
23	2-branched	11.49	3.32	2090	B
23	3-branched	11.42	3.32	2090	B
23	4-branched	11.34	3.32	2090	B
24	Linear	12.13	3.32	2090	B
24	1-branched	12.05	3.32	2090	B
24	2-branched	11.98	3.32	2090	B
24	3-branched	11.91	3.32	2090	B
24	4-branched	11.83	3.32	2090	B

Consideration of bioconcentration testing

The option to undertake a fish bioaccumulation study of GTL Gasoil (C8-C26) has been assessed for its feasibility. The considerations and conclusions also apply to Hydrocarbons, C18-C24, isoalkanes, <2% aromatics, which is derived from GTL Gasoil.

An aquatic bioconcentration study (e.g. OECD 305) is only possible if the concentration of the dissolved phase chemical can be measured in water which is an issue for GTL Gasoil (and GS310) because the individual hydrocarbon constituents of GTL Gasoil are very poorly soluble in water. 

The measured and predicted water solubility data available for alkanes (C8 to C26) indicate that the water solubility of the majority of the constituents of GTL Gasoil would be considerably less than 1 mg/l.

At these solubility levels there would be considerable difficulties/challenges in detecting these individual hydrocarbon constituents using standard analytical procedures in water and fish tissue. For an individual chemical this problem is usually overcome by the use of radio-labelling techniques. However, owing to the large number of constituents it is not feasible to use radio-labelling methods which would be relevant/representative for GTL Gasoil as a whole substance. Further difficulties would be encountered in deciding which of the large number of potential individual constituent hydrocarbons would be used to monitor bioaccumulation to ensure the test is representative and therefore valid.

An investigation was conducted (Williams, 2011) of the feasibility of the recovery of GTL Gasoil from whole rainbow trout at concentrations that would be appropriate for a bioconcentration study. Method development and validation was performed. Method development was conducted to investigate different approaches to the extraction and analysis process.

These data may be used to draw final conclusions about the feasibility of conducting the OECD 305 study.

The detection method was found to have acceptable linearity. The analytical procedure had acceptable recoveries of test item from fish tissue. The method of analysis was validated and proven to be suitable for the measurement of the test item in whole rainbow trout at concentration of 50, 100 and 500 mg/kg. The method of analysis was validated and proven to be suitable for use down to a LOQ of 13.2 mg/kg of test item in fish tissue.

The LOQ is also estimated in terms of each individual n-alkane (omitting n-pentadecane and n-heptadecane as they were found in the rainbow trout control tissues), by reference to the percentage composition data. Based on the individual n-alkane LOQs in fish tissue and the known or estimated solubility values for the constituents, the bioconcentration factor (BCF) value range which could be detected have been estimated for each constituent. Based on the relative concentration of the individual constituents present in GTL Gasoil and the bioconcentration factor (BCF) value range which could be detected, it is suggested that such an OECD 305 study could be conducted using n-tetradecane (n-C14), n-hexadecane (n-C16) and n-octoadecance (n-C18)(providing their aqueous exposure concentrations could be continuously maintained and determined throughout the duration of the uptake phase of the BCF study).

Given the interferences arising from n-pentadecane (n-C15) and n-heptadecane (n-C17) in naturally occurring fish tissue reported in this study, measurement of all constituents present in GTL Gasoil would not be feasible in a BCF study using a non-radiolabelled test substance, but such a study could be conducted using selected n-alkanes as marker compounds for GTL Gasoil.

Other available data on bioaccumulation

Some information is available regarding bioaccumulation, from a solid phase micro extraction (SPME) study (See EPSR Harris, 2013). SPME is a biomimetic extraction system that tries to mimic the way organisms extract chemicals from water.

A biomimetic extraction technique using SPME in conjunction with gas chromatography (GC) was employed to assess the level of water soluble hydrocarbons and hence the potential toxicity of water accommodated fractions (WAFs) for a range of GTL products. In addition, the toxicity of WAFs was also monitored using an in vitro MicrotoxTM assay, a biosensor-based measurement system based on Vibro fischeri bacteria. In this study the majority of GTL samples (including GTL Gasoil, CAS 848301-67-7 from which the registered substance is derived) analysed using SPME-GC indicated a low total peak area for soluble hydrocarbons (on average less than 12 000), with no detectable toxicity using the MicrotoxTM assay. The low total peak area for soluble hydrocarbons for these substances is indicative of a low potential to bioaccumulate.

The related C8-16 (GTL Kerosine) and C4-10 (GTL Naphtha) samples, consisting of shorter chain hydrocarbons, produced a more soluble hydrocarbon fraction, and GTL Naphtha resulted in a moderate level of toxicity detected with the MicrotoxTM assay. The level of toxicity seen with the GTL Naphtha is, however, considerably less than that seen with the crude oil gasoline.

Consideration of bioconcentration testing of constituents

GTL Gasoil and Hydrocarbons, C18-C24, isoalkanes, <2% aromatics, which is distilled from GTL Gasoil have been demonstrated to be readily biodegradable in aerobic studies in aquatic media (see Section 4.1). The degradation/removal in soil has been measured for GTL Gasoil in two separate screening studies (refer to CSR Section 4.1.2.2). The main objective of these studies was to identify any constituents of GTL Gasoil that may potentially be persistent in the terrestrial environment, and thus warranting further investigation for their bioaccumulation potential either in a fish bioconcentration or dietary study, as requested in ECHA decision number TPE-D-0000001392-80-03/F.

In an initial OECD 307 soil degradation study the GTL Gasoil was applied to a single soil and although it was feasible to monitor loss of GTL Gasoil from soil through chemical extraction and analysis, the test was unable to assess whether the losses were attributed to biodegradation or a combination of physical loss and biodegradation. In the second study the GTL Gasoil was once again added to a single biotic soil with an additional set of sterile (abiotic) soil systems to assess the physical loss of the substance from soil. Three individual n-alkanes (dodecane, hexadecane and eicosane) and a C15 iso-alkane were also subjected to a series of biotic and abiotic OECD 307 studies. These alkanes were applied separately to a single soil type (biotic and abiotic) to monitor their respective degradation rates. The individual n-alkanes were added to the soil at a concentration comparable to their respective concentrations in the GTL Gasoil sample used in the soil degradation studies. 

It should be noted that as a complex substance there had to be several deviations from the OECD 307 test guidelines which were originally designed to assess the degradation rates of single (typically crop protection) chemicals.  For example, the soil was spiked at much higher concentrations than those predicted to occur from normal use because it was only feasible to meet the recovery and precision criteria for the OECD 307 tests when GTL Gasoil was dosed at 1000 mg/kg. This is several orders of magnitude greater than concentrations which would be used for single substances. Furthermore, because abiotic loss factors could not be ruled out the results of the studies were reported as disappearance rates rather than biodegradation rates.

In both the GTL Gasoil soil studies it could be seen that, based on total petroleum hydrocarbon (TPH) measurements, the GTL Gasoil rapidly disappeared from the soil. 

Half-lives (DT₅₀) and disappearance rates for 75 and 90% (DT₇₅and DT₉₀values respectively) for the studies are summarised in Table 4.8. 

Table 4.8 Calculated DT values for the GTL Fuel and alkanes*

DisappearanceRate (days)		GTL Fuel		Hexadecane		Eicosane		C15 iso-alkane
	1st study	2nd	study
	Aerobic	Aerobic	Sterile	Aerobic	Sterile	Aerobic	Sterile	Aerobic	Sterile
DT50	11	22	83	1.5	33	11	120	0.9	1.0
DT75	23	43	170	5.4	86	26	260	10	25
DT90	39	71	280	11	160	46	440	ND	62

Notes: * Dodecane data excluded because >50% disappeared during soil preparation. DT values >56 days (i.e. the test duration) have been extrapolated. ND = Not Determined

The results of the studies undertaken indicate that, although sterile controls can provide an indication of physical losses, the OECD 307 test will ultimately determine ‘disappearance’ as opposed to biodegradation of constituents of a complex substance like GTL Gasoil.

In conclusion, based on the OECD 307 studies, it is not possible to identify any constituents of GTL Gasoil as P or vP and thus warranting further investigation for their bioaccumulation potential either in a fish bioconcentration or dietary study. This conclusion also applies to Hydrocarbons, C18-C24, isoalkanes, <2% aromatics.

Conclusion on testing for bioaccumulation

 As described in the attached endpoint study record (See EPSR Williams, 2011 in the IUCLID dataset), it is not feasible to undertake an OECD 305 study on whole substance.    

In addition, based on OECD 307 (Aerobic and Anaerobic Transformation in Soil) studies (refer to Section 5.2.3), it is not possible to identify any constituents of GTL Gasoil as P or vP, and thus warranting further investigation for their bioaccumulation potential either in a fish bioconcentration or dietary study.  

Furthermore, as discussed above, it can be demonstrated by QSAR that none of the constituents of GTL Gasoil meets the criteria for vB, and bioconcentration factors can be adequately predicted.  

The available information is adequate for the purposes of both PBT/vPvB assessment and environmental exposure assessment.  

In keeping with the requirement of REACH to minimise in vivo testing on vertebrates, it is therefore inappropriate to conduct a fish bioconcentration or dietary study on the grounds of animal welfare.