1. Background

Certain mineral oils may be used for several different purposes and taxed at different rates of duty. In the case of diesel fuel for road vehicles, it may also be used as heating gas oil and in this case is normally sold a much lower prices. Gas oil for agricultural and marine use may also benefit from reduced tax rates. This being the case, there is a financial incentive to misuse heating fuel as road fuel and most Member States combat this by adding a fiscal mark to fuels released at reduced tax rates.

This fiscal mark consists of two parts. Firstly a tracer chemical is used which can be detected by a simple roadside test using a reagent. In addition, a dye is also used to give an immediate visual indication that the fuel has not borne the full rate of tax.

Council Directive 95/60/EC provides the legal base for the introduction of a standard marking system for use throughout the Community. It was negotiated under the auspices of Article 9 of Council Directive 92/81/EEC (the mineral oils structures Directive) and was adopted in December 1995. A call for expressions of interest, which listed a number of criteria that should be met, including health and safety provisions, was then issued and a number of products were submitted for consideration.

Extensive evaluation and testing was conducted by IRC Ispra and JRMM Geel assisted by the Customs laboratories of the Member States. Combustion testing was also carried out by Stazione Sperimentale per I Combustibili in Milan and this showed no detectable differences between emissions from marked and unmarked fuel. Following this testing, the Solvent Yellow 124 (N-ethyl-N-[2-[1-(2-methylpropoxy)ethoxyl]ethyl]-4-phenylazo]-benzeneamine] system was selected as that which most closely met the required criteria.

This product is marketed by BASF as Sudan 455 marker and is also marketed by several other manufacturers. It should be noted that the selection of mineral oil marking systems covers only the standardisation of the tracer chemical because the type and colour of dye remains a matter for individual Member States. However, two Member States have concerns about this product, The UK which has a problem with illicit removal ('laundering') and feels this is too easily achieved with this product and Denmark which has health and safety concerns.

The Danish concerns stem from the chemical structures as an azo dye and from the problems in Sweden that were initially suspected to have been caused by the introduction of SY124 but these were eventually attributed to a change in fuel specification following a Swedish study. There were plans to introduce SY124 in Denmark at the end of 1997 but these were abandoned at the last minute due to a press campaign claiming that the product posed an additional health risk compared to unmarked products. Denmark is now seeking confirmation that the marked fuel is no more harmful than an unmarked product.

The product has been in use for some 10 years in a number of Member States. Because only some testing has been carried out by the manufacturer as well as during the Swedish investigation, Denmark still has reservations. These centre on the reaction of the SY124 system and the potential reaction with the dyes used.

The CSTEE has been asked to examine this problem and if possible to conclude that the risks to human health from the marked product are no greater than those posed by the unmarked fuel. The impact on the environment has not been addressed.

Originally the following background material has been provided to the CSTEE:

1. Preliminary evaluation of the product by IRMM Geel (CSTEE/99/3-Add.1)

2. Results of the SSC combustion tests (CSTEE/99/3-Add.2)

3. Excise Committee working paper CED 273 (CSTEE/99/3-Add.3)

4. Correspondence from Danish Environment Service (CSTEE/99/3-Add.4)

5. Summary of Swedish Grön diesel report (CSTEE/99/3-Add.5)

6. BASF-study report on acute toxicity, skin irritation and eye irritation (CSTEE/99/3-Add.7/C)

7. BASF-study report on mutagenicity test using micro-organisms (CSTEE/99/3-Add.7/D)

8. BASF-study report on Metaphase analysis of chromosomal aberration in cultured mammalian cells (CSTEE/99/3-Add.7/E)

9. Publication of Fischer and Bjarnson (1996) (CSTEE/99/3-Add.7/F)

Additional information received subsequently is given in the text. Information on analytical procedures also provided (CSTEE/99/3-Add.8-10) are not used in this report.

In October 1993, a marking system was introduced in Sweden for diesel oils used in working machines and for the purpose of house heating. The added dyes are:

Solvent Yellow 124 (Somalia Yellow, T10 Yellow LBN) CAS 34432-92-3

Solvent Blue 79 (Sudan Blue) CAS 64553-79-3, 90170-70-0

Solvent Blue 98 (Automate Blue) CAS 74499-36-8

After the introduction of the coloured oil products, reports of complaints and symptoms started to appear from people occupationally exposed, especially among workers using wood harvesters. Some of the symptoms reported were headache and skin and airway irritations. This raised concern about the health effects of the added colours, and led the Swedish Government to appoint a commission to evaluate the marker system. The commission initiated several studies and published a report (Grön diesel 1995).

The Swedish commission considered it unlikely that the health problems reported in conjunction with exposure to coloured diesel oil could be attributed to the markers and dyes used. The inherent mutagenic and environmentally hazardous properties of the dyes as such, however, do mean that alternative markers and dyes would be desirable. None of the alternative substances used for similar purposes in other countries appeared to offer any advantages. The Swedish commission emphasised that when better alternatives are available in terms of mutagenicity and environmental impact, they must be introduced according to the substitution principle. The low concentrations of the substances up to 6 ppm in diesel oil and the practically total combustion do mean, however, that the short-term and long-term risks, seen in terms of occupational health, may be considered extremely small when the markers and dyes are used for the intended purpose.

The CSTEE used this data and other found in the literature or provided by BASF to evaluate the toxicology of the dye Solvent Yellow 124 and its contribution to the overall toxicology of diesel oils.

2. Toxicology of Solvent Yellow 124:

Solvent Yellow is an azo-dye and a structural analogue of the carcinogenic p-dimethylaminoazobenzene. The other dyes Solvent Blue 79 and Solvent Blue 98 are antraquinone derivatives. All compounds are genotoxic in in vitro test systems.

Acute Toxicity

The LD50 in rats after oral application is approximately 7.5 ml/kg (CSTEE/99/3 - Add.7/C).

Skin Irritation

A patch test, which has been performed according to FDA with 6 rabbits showed slightly irritating properties. 0.5 ml of the compound has been applied to the shaved dorsal skin in part intact, in part scarified. The macroscopic investigations were performed 24 and 72 hrs after application. No erythema or scabs occurred in the intact and scarified skin, whereas oedema was found in the scarified skin (CSTEE/99/3 - Add.7/C).

Eye Irritation

The test has been performed according to FDA and showed a slightly irritating effect. 0.1 ml have been applied to the right eyes of 6 albino rabbits which thereafter have been observed up to 7 days. The maximal effect has been seen at 24 hrs (CSTEE/99/3 - Add.7/C).

In vitro Genotoxicity Tests

The chemical has been tested using S. typhimurium strains TA 98, 100, 1535, and 1537 as well as E.coli WP2uvrA- with and without metabolic activation. The compound has been applied at 50, 100, 500, 1000, and 5000 m g/plate. A dose dependent increase in the number of reverting colonies from about 30 to about 170 per plate was seen in TA 98 strains in the presence of metabolic activation (CSTEE/99/3 - Add.7/D).

In a fibroblastic cell line derived from Chinese hamster lung (CHL/IU) cells the compound did not induce chromosome aberrations. Maximum concentrations have been 625 m g/ml during continuous incubation for 24 and 48 hrs without S-9 metabolising system and for 18 hrs in the presence of S-9 (The Centre of Japan Biological Chemistry 1988) (CSTEE/99/3 - Add.7/E).

Skin-sensitisation and Irritation of 3 diesel oil classes and three indicator dyes (Fischer and Bjarnason 1996):

Three dyes, Solvent Yellow 124, Solvent Blue 79 and 89, have been tested as individual compounds or in the 3 diesel fuels MK1, MK2, and MK3 (environmental classes I, II and III).

The individual dyes in petrolatum up to 10 % (v/v) did not induce irritation in 15 volunteers. Three individuals were tested with the dyes as delivered and two of these showed discrete erythematous reactions to Solvent Blue 98 (60 % in naphtha). The other dyes did not induce reactions at full strength. In experiments with 15 volunteers and 15 workers who had previously reported skin diseases from green diesel, the Diesel oils of environmental classes I and II proved to be strong irritants after 24 h exposure and have been more irritating than Diesel oil class III. The latter is an elder formulation with higher content of aromatic hydrocarbons and higher viscosity. Presence of one, two or three dyes at concentrations up to 6 ppm (Solvent Yellow 124, 3-5 ppm Solvent Blue 79 and 98) did not change the irritant properties of the Diesel fuels. Urticarial reactions have not been observed with the single dyes at 10 v/v % in 4 volunteers, nor in 15 volunteers exposed to Diesel fuels with and without dyes.

In the 15 men with previous skin disease the same differences of irritancy of the Diesel oils occurred as in the control persons and no contact allergy or urticaria was observed due to the dyes or dyed diesel oils using the different conditions.

It is concluded that the skin diseases which have been preferentially observed in unprotected workers which were directly exposed to "green Diesel" are unlikely to be due to the dyes. It has also been reported that these effects have not been seen when the workers handled the Diesel using protective measures as recommended. The authors also conclude that

the irritation reported from "green diesel" is probably due to the change from "heavier" to "lighter" oil, which may be more irritating to the skin, at the same time that dyes were added to distinguish oils meant to be used for different purposes.

3. Combustion products (Grön Diesel report)

Health problems were also reported concerning eye and upper respiratory tract irritation, headaches and tiredness after exposure to exhausts from the combustion of coloured oil. A questionnaire was sent out to 281 drivers of wood harvesters in a northern part of Sweden. 83 of those reported some health problems in connection with coloured diesel oil. Eventually, 16 machine drivers along with 16 controls took part in an inhalation study. They were exposed during 1 h to diluted exhausts (0,25 %) from the combustion of coloured and uncoloured diesel oil of environmental class I and class II.

There were no significant effects on lung function, airway reactivity or performance in psychometric tests from the exposure to diluted diesel exhaust. A slight increase in inflammatory markers in nasal lavage was noted in the machine drivers group, only after the exposure to exhausts from environmentally class II non-coloured diesel. The machine drivers group experienced a higher rate of symptoms such as headache, cough, and eye, nose and throat irritation after exposure to diesel exhausts compared to clean air, and the control persons noted the smell of diesel exhaust. However, there was no difference in symptoms between coloured and non-coloured fuel (Nielsson et al 1994).

Chemical analysis and mutagenicity tests on exhausts from coloured and non-coloured environmental class I diesel showed no significant differences. The fraction of unburnt Solvent Yellow 124 in the emissions from a diesel engine and from a domestic boiler was shown to be less than 0.01 % and 0.0001 %, respectively (Grägg et al 1995).

4. Toxicology of diesel fuels and heating oils (CONCAVE 1996)

Gas oil products comprise among others automotive fuels, heating oils and marine fuel. These oils are complex and contain variable mixtures of hydrocarbons, predominantly of carbon range C11 to C25 and boiling over a temperature interval 150 to 450° C. Automotive fuels for diesel engines are market as automotive gas oil, automotive diesel fuel, diesel fuel No. 2 and rail road engine gas oil.

Gas oil products contain straight and branched chain alkanes (paraffins), cycloalkanes (naphthenes), aromatic hydrocarbons and mixed aromatic cycloalkanes (cycloalkanoaromatics).

Most commercial gas oil products contain polycyclic aromatic hydrocarbons (PAH). In straight-run gas oil components these are mainly 2 and 3-ring compounds, with relatively low concentrations of 4 to 6-ring PAHs. The use of heavier atmospheric, vacuum or cracked gas oil components is likely to result in an increase in the content of 4 to 6-ring PAHs, some of which are known to be carcinogenic.

Commercially available gas oil products may contain low concentrations of additives such as flow improvers, corrosion inhibitors, defoamers, dyes/markers, anti-oxidants, stability improvers, cetane improvers, detergents and anti-static additives.

There are 68 substances listed in EINECS, which describe the gas oils which may be used for the vehicle and heating fuels. Many of them are in the list of dangerous substances (Annex I of the Classification and Labelling Directive - 67/548/EEC), some of them are classified as carcinogens.

The toxicity of a number of gas oils and marketed fuel has been investigated by API, DGMK, CONCAVE and others (see CONCAVE 4.1 Toxicity).

Animal Experiments

AcuteToxicity

The oral LD50 values in rats are several g/kg with clinical signs including hypoactivity, ataxia, incontinence, hair loss and gastric haemorrhage/irritation.

The dermal LD50 in rabbits were above 2g/kg with marked irritation in the treatment area.

The LC50 upon inhalation was about 5 mg/l with marked inflammation of the respiratory tract and the lungs.

Irritancy and sensitisation

The several gas oil products tested induced moderate to severe skin irritation but no eye irritation. In a modified Buehler Test no sensitisation has been found.

Subacute/subchronic studies

Gas oil streams have been investigated in dermal studies over periods of 4 or 13 weeks and inhalation studies for 4 weeks.

In the dermal studies, significant severe skin irritation at all doses (approx. 2-8 g/kg/day) occurred; reduced body weights were seen at 125 and 500 mg/kg/day.

Upon inhalation up to 4 weeks inflammatory changes in the nasal tissues and lungs of rats at about 20 mg/m³ were seen.

Carcinogenicity

In a series of mouse skin painting studies with different gas oils applied 2 x per week up to 2 years all materials which have been tested caused the development of benign and malignant skin tumours. A common feature was the occurrence of severe skin irritation and a long latency period before tumours developed (e.g. 90 weeks).

Genotoxicity

In-vitro bacterial mutagenicity assays indicate that the hydrotreated or hydrodesulphurized straight run gas oil components range in activity from inactive to weakly positive. The mouse lymphoma assays which have been conducted on the straight run gas oils without subsequent hydrodesulphurization have yielded positive results in the presence of S9 metabolic activation. The 3 components which have been subjected to hydrodesulphurization were only weakly positive or positive in the presence of S9 and negative without S9. The in-vivo bone marrow cytogenetics assays and sister chromatid exchange assay were negative for straight run components, with or without hydrodesulphurization.

Developmental Toxicity

A sample of commercially available diesel fuel was examined in a teratology study in rats. Pregnant rats were exposed to 102 or 402 ppm of diesel fuel vapour on days 5 through 15 of pregnancy. The only sign of maternal toxicity was a reduction in food consumption at the highest exposure concentration. No foetotoxic or teratogenic effects were observed in the study.

Developmental toxicity studies have been carried out in rats on six gas oils applied to the skin daily on days 0 through 19 of gestation. The dose levels varied for each gas oil but ranged from 8 up to 1000 mg/kg/day.

With the exception of the coker light gas oil, all other materials tested caused foetotoxicity (increased resorptions, reduced litter weight, reduced litter size) at doses which also caused maternal toxicity (mainly reduced weight gain, but also increased liver weight, reduced thymus weight).

Human Experience

Occupational exposure to gas oil products results in irritation of the skin, mucous membranes and central nervous system depression. Irritation of mucous membranes of the upper respiratory tract was seen at concentrations well above 5 mg/m³. Accidental eye contact with liquid gas oil may cause mild, transient stinging or redness, as well as exposure to high concentrations of mist or vapour.

Excessive exposure under conditions of poor personal hygiene may lead to irritation, dermatitis and also oil acne and folliculitis, leading to warty growth which may become malignant.

5. Conclusions

The CSTEE has evaluated the available information on the toxicology of diesel oil and of Solvent Yellow 124 used as an additive in diesel oil and the reports on the symptoms observed in workers exposed to coloured diesel oil and its combustion products.

The CSTEE considers it unlikely that the health problems reported in conjunction with exposure to coloured diesel oil products could be attributed to the dye used at concentrations up to 6 ppm. The inherent toxicity of diesel oil itself explains the symptoms reported by exposed persons. The occupational health risks of the low concentrations of 6ppm Solvent Yellow 124 in diesel oil are considered small.

However, the toxicological and ecotoxicological databases of Solvent Yellow 124 are poor and do not meet the requirements of a large volume product. Data on the stability of the azo dye should be provided. The database must be improved before an appropriate risk assessment can be made.

6. References

CONCAVE: Gas oils (diesel fuels/heating oils). Product dossier no. 95/107, Brussels, September 1996.

Fischer T and Bjarnason B: Sensitising and irritant properties of 3 environmental classes of diesel oil and their indicator dyes. Contact Dermatitis 34, 309-315, 1996.

Grön Diesel - Environmental and Health Risks: Report from the Commission on the marker system for certain oil products. Swedish Official Report Series 1995:3, Ministry of Finance, Stockholm, Sweden (Swedish with English Summary. See CSTEE/99/3-Add.5).

Grägg K et al: Chemical analyses and biological tests in combustion emissions from coloured and non-coloured diesel fuel (Swedish) 1995. Annex 8 of Grön diesel report.

Nilsson C-A et al: Acute effects of inhalation of combustion emissions from coloured and non-coloured diesel fuel (Swedish) 1994. Annex 3 of Grön diesel report.

Siemiatycki J. et al: Associations between several sites of cancer and twelve petroleum-derived liquids: Results from a case-referent study in Montreal. Scand J Work Envir Health 13, 493-504, 1987.

List of documents made available to the Scientific Committee on Toxicity, Ecotoxicity and the Environment via its Secretariat to help it reach the opinion requested by the services of the Commission on the subject of "Mineral oil marking system", (Euromarker) - safety of the preferred candidate

CSTEE/99/3

Selection of a Community wide mineral oil marking system - Safety of the preferred product.

Note from Mike Bott, Head of Unit DG XXI/C/4 (27/4/99)

CSTEE/99/3 - Add. 1

Preliminary evaluation of proposals submitted after the call for expression of interest 96/C45/10 and recommendation to the Excise Sub- Committee of DG XXI. Preliminary evaluation of the product (Sudan Marker 455 liquid - C.I. Solvent Yellow 124) by IRMM Geel.

CSTEE/99/3 - Add. 2

Presentation and results of the SSC combustion tests (subcommittee meeting of 1 July 1998).

CSTEE/99/3 - Add. 3

Excise Committee - Selection of a fiscal marker for gas oil and Kerosene (EUROMARKER) - Recommendations for the Euromarker sub-Committee)

Working document CED N° 273 - XXI/1273/98-EN.

CSTEE/99/3 - Add. 4

Correspondence from Danish Environment Service - Colouring of gas and diesel oil - Euromarker (22 December 1998).

CSTEE/99/3 - Add. 5

Summary of Swedish study.

CSTEE/99/3 - Add. 6

Sudan Marker 455 Liquid.

Safety data sheet (version 4 - revised 20.03.1996) according to 91/155/EEC from BASF Aktiensgesellschaft

Fax from CCR Ispra - 4/5/99.

CSTEE/99/3 - Add. 7/A

Note from Mr. Mike Bott, acting Head of Unit DG XXI, to Mr. Costa-David concerning "Selection of a Euromarker - Safety of the preferred product" - Health and safety aspects of Solvent Yellow 124 - 21/05/99.

CSTEE/99/3 - Add. 7/B

Fax from Dr. Christos Vamvakaris, BASF AG, EFM/SAM, to Mrs. Brigitte Hiller, Joint Research Centre - Ispra - 20/04/98 - concerning Sudan Marker 455 liq.

CSTEE/99/3 - Add. 7/C

Somalia Yellow LBN Liquid.

Compound : T 10 Yellow LBN (xylene solution) - Solvent Yellow 124.

CSTEE/99/3 - Add. 7/D

Report - "Mutagenicity test using Micro-organisms" made by Research Laboratory "The Centre of Japan Biological Chemistry Co., LTD" on Solvent Yellow 124 - 20/07/1988.

CSTEE/99/3 - Add. 7/E

Report - "Metaphase Analysis of Chromosomal Aberration in Cultured Mammalian Cells" made by Research Laboratory "The Centre of Japan Biological Chemistry Co., LTD" on Solvent Yellow 124 - 25/11/1988.

CSTEE/99/3 - Add. 7/F

Report from T. Fischer & B. Bjarnason (National Institute for Working Life, Solna, Sweden) on "Sensitising and irritant properties of 3 environmental classes of diesel oil and their indicator dyes" - 4/10/1995.

CSTEE/99/3 - Add. 8

Journal article -Sundberg, N. et al.: "HPLC analysis of Solvent Yellow 124 -The marker in Diesel Oil" (Journal of Forensic Sciences, Vol. 41, n. 2, March 1996, pp. 300-303) - 3/96

CSTEE/99/3 -Add. 9

Journal article -Barbieri, M. and Mascherpa, A.: "Determinazione mediante HPLC dei denaturanti nei prodotti petroliferi destinati alla carburazione" (La Rivista dei Combustibili, vol. 49, fasc. 11-12, novembre-dicembre 1995) -12/95

CSTEE/99/3 -Add. 10

Journal article -Henricsson, S. and Westerholm, R.: "Liquid chromatographic method for analysing the colour marker Solvent Yellow 124, N-ethyl-N[2-(1-isobutoxyethoxy)ethyl](4-phenylazophenyl)amine, in diesel fuels" (Journal of Chromatography A, vol. 723, 1996, pp. 395-398 -1996.

CSTEE/99/3 -Add. 11/A

Report from Morton International Limited (Hounslow, Middlesex, UK): "Safety Data Sheet. Automate Blue 8GHF" -15/11/96

CSTEE/99/3 -Add. 11/B

Report from Morton International Limited (Hounslow, Middlesex, UK): "Safety Data Sheet. Automate Blue 8G" -8/8/95

CSTEE/99/3 -Add. 12

Colour Index International Pigments and Solvent Dyes (from ETAD), pp. 208 (Solvent Yellow 124), 255 (Solvent Blue 79) and 256 (Solvent Blue 98).

CSTEE/99/3 -Add. 13

Report "Experiences from introducing a mineral oil marking system in Sweden", by Prof. K. Victorin -23/8/99

CSTEE/99/3 -Add. 14

Draft text by Prof. H.Greim -23/8/99

CSTEE/99/3 -Add. 15

E-mail from Mss. Birgit Sokull-Kluettgen (JRC Ispra) to Jorge Costa-David (EC, Health and consumer protection Directorate-General) regarding the absence of information on the three chemicals under study in the ECB's databases -19/8/99

CSTEE/99/3 -Add. 16

List of searches for documentation carried out between 20/7/99 and mid August 99 with the respective results -30/8/99

CSTEE/99/3 -Add. 17

2 tests :

- CI Solvent Yellow 124 (CAS-No. 34432-92-3)

- CI Solvent Blue 79 (CAS-No. 90170-70-0)

BASF AG - 29/7/1999.