CSTEE Opinion on the Risk Assessment of Short Chain Length Chlorinated Paraffins
Introduction
The CSTEE has been asked to give its opinion on the risk assessment on short chain length chlorinated paraffins produced within the EU Programme for Existing Chemicals. Chlorinated paraffin products are based on polychlorinated alkanes and they are divided into three categories depending on the carbon chain length, short (C10-C13), medium (C14-C17) and long (C20-C30), and the reviewed document deals with the first of these categories. A working group within CSTEE was established and it has reviewed the Final draft (dated May 1998) of the risk assessment.
General substance information
This chapter deals with the identity and properties of the chemicals. The commercial products are extremely complex mixtures of mainly n-alkanes with different chlorine content. The many different compounds in the products make the risk assessment difficult in several ways, something which is obvious already from the section on physico-chemical properties. In the text the chlorinated paraffins are described as hydrolytically stable, in a table it is mentioned that the water solubility is 0.15-0.47 mg/l "with partial hydrolysis" with a reference to the IUCLID database. In the CD-ROM version of this base, however, the data for water solubility are in the range 0 - 0.01 mg/l. Other published data range from 0.00049 to 1.26 mg/l. Water solubility is an essential parameter in this risk assessment and the uncertainty in the data base will influence the final result.
Large variation can also be found for other parameters, such as vapour pressure and partition coefficients. Different compounds in the product may thus be distributed differently in the environment and it is not possible to find the same mixture everywhere. This means that the test results for different effects of the original product will not necessarily be valid for the composition present in a specific environmental compartment. An important parameter for these substances is the chlorine content. The commercial short chain length products can contain between 49 and 71% chlorine as an average, which makes the heterogeneity of this group still more pronounced. These difficulties could have been more clearly pointed out in the risk assessment document.
The report approaches these problems by a straightforward application of the protocols stated in the Technical Guidance Document, in the same way that should be applied to a pure chemical substance. Taking into account the available information this is the only possible solution, but clearly the high level of uncertainty produced by this approach and/or the specific application of the Precautionary Principle (i.e. when taking out a value from the range of physical-chemical properties) should be clearly identified.
General information on exposure
Less than 15000 tonnes were, according to the industry, produced in 1995 at two sites within the EU. Nothing is said in the assessment about export from or import to the EU of these chemicals. The applications of short chain length chlorinated paraffins are described and the metal working sector is using more than 70% of these products. Other important applications are in rubber, paints, sealants, leather and textiles.
Some qualifying statements in the text, such as "is reported to be low", would have gained strength if their sources had been referenced.
Environmental exposure assessment
Emission factors for production of short length chain chlorinated paraffins have been taken from the Technical Guidance Document. The values used are: 0 to air, 0.003 to wastewater and 0.0001 to soil. A release factor of 0.0001 to wastewater, referring (indirectly) to unpublished data from Gesellschaft Deutscher Chemiker, is also used in the following calculations. Information, not available to CSTEE, from the two manufacturers in the EU indicate, however, that the present emissions to waste water are much lower than these figures and estimated to be less than 9.9-26.7 kg/year.
In the estimation of emissions from metalworking, the fate of short chain length chlorinated paraffins on the swarf is a special problem. About 90% of these are re-used and it is claimed that the additives are destroyed in this process. This is not necessarily true as the swarf often is pre-heated with hot air before added to the melting process. These sites may thus be point sources for chlorinated paraffins in the air. There is also a possibility for the formation of more toxic compounds from the chlorinated paraffins in this heating process. Formation of PCN and PCB have been reported in pyrolysis experiments, a formation that also may take place in the metal working operations.
The emissions from paint is, with reference to a Canadian report, expected to be minimal. However, in a recent report, which was not available to the risk assessor, the emission from surfaces with a paint containing short chain length chlorinated paraffins was measured. Extrapolation of the results to the 1150 tonnes used annually in paint within the EU, gives a continental emission of about 9 tonnes. There are probably also contributions from the surfaces painted last year and so on, so this source may be significant.
Due to the many assumptions that have to be made, the release estimates are subject to a large uncertainty. The dominating source is not unexpectedly the use in metal working with emissions of about 1700 of the total 1800 tonnes per year in the EU.
The chemical analysis of the complex chloroparaffin products is difficult and most of the levels reported on environmental concentrations are determined with a low precision. Recently an elaborated method using HRGC-ECNI-HRMS was described. This would be able to give more exact results if pure reference compounds were available.
In spite of the analytical difficulties some levels of chlorinated paraffins have been reported and in the following table some of these are compared to the regional levels predicted by the EUSES model. It can be seen that the predictions after all are close to or within the reported data sets.
| Medium | PEC (EUSES) regional | Measured concentration | Unit |
| Air | 12 | No data | ng/m3 |
| Surface water | 0.33 | 0.05-4.0 | µg/L |
| Sediment | 1.2 | 0.05-15 | mg/kg ww or dw |
| Soil | 11 | 0.10 | mg/kg ww or dw |
| Fish | 2.6 | 0.025-3.3 | mg/kg ww |
| Meat | 0.15 | 0.032-0.088 | mg/kg ww |
Recently high levels of short chain chlorinated paraffins were reported in biological samples from the Arctic indicating that these chemicals are effectively transported over long distances.
The EUSES model was also used to predict local scenarios for the concentrations of short chain length chlorinated paraffins in different matrices. The outcome indicates high concentrations in almost all matrices, especially plant roots, which are expected to contain more than g/kg concentrations. It is obvious that this is wrong and it is also explained as being the effect of unrealistic partition coefficients being used for the soil-root distribution. It is obvious that EUSES has a problem with the highly lipophilic compounds in these calculations.
Environmental effects assessment
The effect assessment clearly reflects the lack of information for essential environmental compartments such as sediments and soils, as well as the difficulties in the toxicological interpretation of some toxicological studies.
The toxicity values selected in the report and used for the risk assessment are considered appropriate. Most (all) reported acute toxicity values for aquatic organisms are above the expected water solubility limit (clear values for this essential parameter are not available), however, these values are not used in the final assessment.
The NOAEL of 166 ppm selected from the bird reproduction study and its comparison with mammalian toxicological information is also well justified. However, the value is restricted to a single mixture with a C10-12, 58% Cl chlorinated paraffin. The avian reproduction toxicity of other mixtures in unknown.
Environmental risk characterisation
The methodology for PNEC derivation and the application of uncertainty factors are in agreement with the protocols stated in the TGD for chemical substances.
Taking into account all mentioned problems, the use of equilibrium partitioning for sediment- and soil-dwelling organisms is clearly not recommended, but as stated in the report there is no other possibility with the present level of information.
Human exposure assessment
No measured data for occupational health exposures was found for the short chain length chlorinated paraffins and therefore these were modelled by the use of EASE. The outcome seems reasonable, but it is surprising that people working with manufacturing and formulation are exposed to ten times higher dermal exposure than those working in the metal working sector.
Also consumer exposure had to be estimated. It is claimed that chlorinated paraffins are used infrequently in leather treatment as they are relatively expensive. This may be argued, as a recent investigation showed that the emission from three different shoes all contained chlorinated paraffins. The rest of the estimates seem to be worst case types and the most predominant sources found to be leather clothing and metal working fluids.
The EUSES model outcome was used to predict indirect exposure via the environment. The extremely high levels predicted for root crops indicate that this exposure route corresponds to about 99% of the human intake. There are no measured data for this matrix so the assessor had to eliminate it by declaring it as unrealistic. The CSTEE agrees with this assumption, realising that we then accept that this route does not contribute at all which is unsatisfactory. The few measured data seem to support the rest of the model outcome to some extent.
Human health effects assessment
The report gives an acceptable presentation of the available documentation on animal and human toxicity of short-chain chlorinated paraffins. These compounds are of very low acute toxicity. Treatment with the chlorinated paraffins may lead to skin irritation upon repeated application, whereas they may cause mild eye irritation. The paraffins do not appear to induce skin sensitisation.
The liver and thyroid are target organs in repeated dose studies with rats and mice. The liver damage is associated with peroxisome proliferation, whereas the thyroid effects are correlated to altered thyroid hormone status and glucuronyl transferase induction. Humans would be much less sensitive to peroxisome proliferation and thyroid hormone perturbation than rats and mice.
In rats, neoplastic effects are seen in the liver, the thyroid (females) and the kidneys (males) as well as mononuclear cell leukaemia. There was poor survival in this experiment. In order to elucidate the mechanism underlying the renal tumorigenic effect, there were indications that a 58% chlorinated paraffin caused hyaline droplet formation in male rats, but this could not be confirmed by immunocytochemical techniques. In mice, neoplastic effects were again seen in the liver and the thyroid (females). In male mice there was a dose-related increase (control 0%, low dose 6%, high dose 12%) in alveolar/bronchiolar carcinomas. The report states that no significance for human health can be read into this pattern of results. This statement is overly strong, it presumably takes into account that the control animals in this experiment did not show any evidence of lung tumours, whereas the historical control incidence was 5.8%.
There were no studies specifically investigating effects on fertility. In rats, signs of developmental toxicity were only seen at maternally toxic doses.
It could have been mentioned that chlorinated paraffins induce gap junctional intercellular communication between cells in culture.
Human risk characterisation
The report presents exposure estimates to short-chain chlorinated paraffins for workers (manufacture and use) and consumers (inhalation and dermal exposure from use of metal working fluids and leather and textile clothing) as well as indirect exposure of man via the environment. The CSTEE judges the estimates presented in the report to be reasonable.
The report suggests that the dose-related increases in the incidence of adenomas and carcinomas observed in the liver, thyroid and kidney are not relevant to human health, considering the likely underlying mechanisms for these tumours. The CSTEE agrees that peroxisome proliferation may be a plausible mechanism for the liver tumours and that altered thyroid hormone homeostasis may be a plausible mechanism for the thyroid tumours, respectively. One cannot totally discount the possibility that such tumours could arise in humans, but there would be very large differences in carcinogenic sensitivity towards the short-chain chlorinated paraffins between rats and mice vs. humans with respect to induction of such neoplasms. There is a possibility that the kidney adenomas which were observed, may be linked to the male rat-specific phenomenon associated with a 2u-microglobin accumulation in hyaline droplets. However, hyaline droplet nephropathy was not clearly demonstrated. Thus, the CSTEE agrees with the view of the Specialised Experts that the consequences of the finding of kidney adenomas in male mice cannot be ruled out. The CSTEE is in agreement with the use of a NOAEL of 100 mg/kg/day for kidney carcinogenicity.
It is the opinion of the CSTEE that the finding of alveolar/bronchiolar carcinomas in male mice should not totally be discounted, as has been done in this report. A dose-related trend was seen with the lowest effective dose of 125 mg/kg/day. The underlying mechanism for this finding is not known, although it should be noted that the chlorinated paraffins are not genotoxic. It is the view of CSTEE, although the finding of lung tumours in male mice may be of importance for humans, that it would not substantially alter the outcome of the risk characterisation.
CSTEE conclusions
The report, as a whole, is a good document that tries to summarise the available information on these substances/mixtures, applying the principles set out by the Technical Guidance Document to assess the environmental risks associated to the life cycle of the short chain length chlorinated paraffins.
The CSTEE realises that the risk assessment of the short chain length chlorinated paraffins is a difficult task. The complex nature of these products makes both exposure and effect assessments complicated. The amount and quality of data on the environmental levels and effects of this group of compounds are limited. This puts the models in focus and the outcome of their estimates important. Some shortcomings of EUSES are obvious and it may be discussed if the risk assessment should have been supported by a few measurements in critical media.
In the exposure assessment the emissions from goods containing short chain length chlorinated paraffins are only discussed very briefly. It has recently been realised that several other persistent organic compounds such as PCB and DEHP are emitted from a number of different matrices. Even if the information for chlorinated paraffins is very limited, it may have been possible to extrapolate some knowledge from other compounds.
The risk assessment can be considered as the best possible solution for the environmental effects assessment and risk characterisation of these mixtures, taking into account the available information. The uncertainty in the process is higher than that expected for pure chemicals, not only in relation with the lack of ecotoxicological information, but also with the problems associated with the lack of characterisation and comprehensive understanding of physical-chemical vs. toxicological relationships.
The generation of additional information is considered essential to increase the scientific basis of this assessment and to reduce the level of uncertainty. Nevertheless, the conclusion of potential unacceptable environmental risks associated to the life cycle of these chlorinated paraffins is considered scientifically sound and in agreement with an acceptable use of the Precautionary Principle.
The CSTEE agrees with the conclusions of the human risk characterisation that with respect to workers there is at present no need for further information and/or testing or for risk reduction measures beyond those which are being applied already. Furthermore, the use of short-chain chlorinated paraffins poses no significant risk for consumers or for man exposed via the environment.
The main comments of the CSTEE can be summarised in the following points:
- Products based on chlorinated paraffins are very complex and their properties (physical, chemical and biological) will cover wide ranges which makes an evaluation very difficult which could have been more clearly explained in the report
- Due to difficult chemical analysis, exposure data are scarce
- TGD has been followed in most instances but a lower release factor (provided by industry) was used for production
- Emissions from goods (e.g. shoes and paint) may have been underestimated
- The EUSES model seems to give acceptable results for several compartments, but the uptake in e.g. plant roots is probably a dramatic overestimation resulting in the assumption that root crops is responsible for the major (99%) human intake of chlorinated paraffins
- The risks identified in some local aquatic environments is judged to be an appropriate result of the risk assessment
- No measured data for occupational health was found and the outcome of the EASE model surprisingly predict a ten times higher exposure in the production of the compounds than in their use in metal workshops
- The report gives an acceptable presentation of the available documentation on animal and human toxicity, although it could have been mentioned that these substances induce gap junctional intercellular communication between cells in culture
- The finding of lung tumours in male mice may be of importance for humans, but it would not alter the outcome of the risk characterisation