| Front page | | Contents | | Previous | | Next |

Environmental and Health Assessment of Substances in Household Detergents and Cosmetic Detergent Products

12. Fragrances

Perfume may be made up by hundreds of constituents. Single chemical substances or simple herbal extracts may also be used to impart fragrance to the product. The purpose of the perfume may be to mask unpleasant odours from other constituents, or to leave a fragrance trace where the product has been used. Perfumes have frequently received attention because of their potential hazard to health, and, consequently, the health hazard assessment constitutes the main part of this Chapter. The environmental hazard assessment has received far less attention and, hence, available data describing the environmental fate and effects of some fragrance constituents are included.

12.1 Potential hazard to health

Contact allergy

The main hazard to health of perfume is hypersensitivity, i.e. contact allergy or intolerance. Contact allergy to perfume occurs with a relatively high incidence. Thus, in an assessment of an unselected population of 567 Danes, 1.1% were found to be allergic to Balsam of Peru and 1.1% were allergic to fragrance mix (Nielsen and Menné 1992). Both Balsam of Peru and fragrance mix are markers for perfume allergy. The incidence of perfume allergy is only surpassed by nickel allergy, which had an incidence of 6.7% in the same population. There is no cure for perfume allergy. Once a person is sensitized exposure to even minute amounts give rise to eruptions and eczema. Eruptions and eczema may be alleviated with steroid creams, although this treatment is not without side effects if used extensively and frequently. The best prophylaxis is avoidance of perfumes, which is not easy as the use of perfume in various household products is widespread. Research is being carried out in order to establish safe concentration below which reactions cannot be elicited.

Most manufacturers who use perfume in their formulations refer to the IFRA Code of Practice when considering type and concentration of perfume in the products. The recommendations in IFRA Code of Practice should be used with caution and evaluated critically. Many of the references are given as "private communication to IFRA" with neither date nor source.

Intolerance by inhalation

Intolerance to perfumes by inhalation is another debated hazard. Symptoms may vary from feeling ill, over coughing, phlegm, wheeze, chest tightness, headache, exertional dyspnea, acute respiratory illnesses, hay fever, child respiratory trouble, to physician confirmed asthma. Symptoms of hyperreactivity of the respiratory tract and asthma without IgE-mediated allergy or demonstrable bronchial obstruction can be induced by perfume. This was shown by placebo-controlled challenges of nine patients with perfume. The same patients were also subjected to perfume provocation with or without a carbon filter mask to ascertain whether breathing through a filter with active carbon would prevent the symptoms. The patients breathed through the mouth during the provocations, as they used a nasal clamp to prevent any smell of perfume. The patient’s earlier symptoms could be verified by the perfume provocation. Breathing through the carbon filter had no protective effect. The symptoms were not transmitted via the olfactory nerve, since the patients could not smell the perfume, but they may have been induced by a trigeminal reflex via the respiratory tract or by the eyes (Millqvist and Lowhagen 1996). Cases of occupational asthma induced by perfume substances such as isoamyl acetate, limonene, cinnamaldehyde and benzaldehyde tend to give persisting symptoms even though the exposure is below the occupational exposure limits (Jensen and Petersen 1991).

Inhalation intolerance has also been reproduced in animals. The emissions of five fragrance products for 1 hour produced various combinations of sensory irritation, pulmonary irritation, decreases in expiratory airflow velocity as well as alterations of the functional observational battery indicative of neurotoxicity in mice. Neurotoxicity was found to be more severe after mice were repeatedly exposed to the fragrance products, being four brands of cologne and one brand of toilet water (Anderson and Anderson 1998).

Potency ranking for sensitization

According to a Japanese study (Nakayama 1998) perfume constituents may be classified in four classes, A, B, C and D, where A is common and primary sensitizers, B is rare sensitizers, C is virtually non-sensitizing fragrances and D is considered as non-sensitizers. The classification was the result of patch test trials on cosmetic dermatitis patients and controls. During the trials a number of fragrances were discovered to produce no positive reactions on either the patients or the controls even at high concentrations of 5-10%. In Japan, the recommendation of using class C and D fragrances rather than A and B fragrances in cosmetic products has produced significantly lower reaction rates in Japan than in the United States and Europe.

Table 12.1
Potency ranking for sensitization of fragrances

The following sections describe the potential hazards to the environment and health of some of the most frequently used fragrances in detergent and cleaning products.

12.2 Polycyclic musks

AHTN (7-acetyl-1,1,3,4,4,6-hexamethyl-1,2,3,4-tetrahydronaphthalene) (CAS No. 1506-02-1; 21145-77-7) and HHCB (1,3,4,6,7,8-hexahydro-4,6,6,7,8,8-hexamethylcyclopenta-[gamma]-2-benzopyran) (CAS No. 1222-05-5) are used as fragrances in cosmetics and detergents, fabric softeners, household cleaning products, air fresheners, etc. Both substances are high volume chemicals with a use volume of 585 and 1,482 tonnes in Europe in 1995, respectively. AHTN and HHCB represent about 95% of the market for polycyclic musks in EU (Plassche and Balk 1997). The following survey of the environmental properties of AHTN and HHCB is based entirely on the risk assessment by Plassche and Balk (1997).

Occurrence in the environment

Both AHTN and HHCB have been found in the environment, e.g., in river water and fish and in samples of human fat and milk. Emissions of AHTN and HHCB take place by private use, and the total volume of these substances is expected to be discharged via wastewater treatment plants. A part of the AHTN and HHCB is released into the aquatic environment when the effluent is discharged into the recipient. Another part will enter the terrestrial environment after sorption to wastewater sludge and application of the sludge on agricultural land. The highest measured influent concentrations in different wastewater treatment plants were 0.0044 and 0.0029 mg/l for AHTN and HHCB, respectively, whereas the highest measured effluent concentrations were 0.0031 an 0.0025 mg/l, respectively. The presence of polycyclic musks in the aquatic environment has been reported for rivers in Germany, the Netherlands, Sweden, and Japan. E.g., concentrations of up to 0.4 and 0.26 m g/l were reported for German rivers. Concentrations in suspended organic matter have been found to be in the range of 0.06-1.2 mg/kg dry weight for AHTN and 0.05-0.58 mg/kg dry weight for HHCB.

Aerobic biodegradability

The ultimate aerobic biodegradability of AHTN and HHCB has been determined in various standard screening tests. All the available results indicate a low level of ultimate biodegradation of both AHTN and HHCB under screening test conditions (Table 12.2). However, it has been shown that primary biodegradation of AHTN and HHCB may occur by different soil-born fungi.

Bioaccumulation

Both AHTN and HHCB have high log K_ow values (>> 3.0) and have the potential to bioconcentrate in aquatic organisms.

Table 12.2
Ultimate aerobic biodegradability of polycyclic musks.

Aguatic toxicity. Algae

The toxicity towards algae has been determined for both AHTN and HHCB according to OECD Test Guideline 201 with the green alga Pseudokirchneriella subcapitata (formerly Selenastrum capricornutum). The 72 h-EC50 for the growth rate were > 0.80 mg/l for AHTN (NOEC, 0.44 mg/l) and > 0.85 mg/l for HHCB (NOEC, 0.47 mg/l).

Invertebrates

The chronic toxicity of AHTN and HHCB has been determined in a Daphnia magna 21-day test according to OECD Test Guideline 202. For AHTN the 21 d-EC50 for the immobilisation of the parent generation was 0.34 mg/l (95% confidence interval, 0.24-0.43 mg/l). The 21 d-EC50 on reproduction was 0.24 mg AHTN per litre (95% confidence interval, 0.24-0.25 mg/l), the NOEC for reproduction was 0.20 mg/l, and reproduction was almost completely inhibited at 0.40 mg/l. For HHCB 21 d-EC50 for immobilisation was 0.29 mg/l (95% confidence interval, 0.20-0.42 mg/l). The 21 d-EC50 on reproduction was 0.28 mg HHCB per litre (95% confidence interval, 0.24-0.25 mg/l), and the NOEC for reproduction was 0.11 mg/l.

Fish

A 21-d prolonged toxicity test has been carried out with bluegill sunfish (Lepomis macrochirus) according to the OECD Test Guideline 204. Concentrations of AHTN up to 0.18 mg/l did not affect the survival of the fish. The 21 d-LC50 for AHTN was determined to 0.31 mg/l and fish growth was significantly reduced at 0.18 mg/l. For HHCB the 21 d-LC50 was 0.45 mg/l and fish growth was significantly reduced at 0.39 mg/l. The overall NOEC of the test was 0.093 mg/l as determined by the onset of clinical signs (Table 12.3).

The toxicity to early life stages of fathead minnow (Pimephales promelas) was examined according to the OECD Test Guideline 210. The hatchability of eggs was not significantly affected by AHTN in any of the test concentrations. Larvae survival after 32 days of exposure was not affected at 0.067 mg/l and below, while larvae growth was not affected at 0.035 mg/l. For HHCB hatchability was not significantly affected in any of the test concentrations. Larvae survival and larvae growth was not affected at 0.68 mg/l and below, after 32 days of exposure.

Table 12.3
Effects of AHTN and HHCB to fish.

Soil organisms

Toxicity tests with earthworms (ISO 11268) showed no mortality or growth inhibition of adult earthworms after 4 weeks of exposure with AHTN at 250 mg/kg, whereas reproduction was not affected at 105 mg/kg. For HHCB, survival of adult earthworms was not affected at 250 mg/kg, whereas the growth rate and reproduction were inhibited at 250 mg/kg and 105 mg/kg, respectively.

Effects on human health

AHTN and HHCB have been under evaluation by The EU Scientific Committee on Cosmetic and Non-Food Products according to the record of their 3^rd plenary meeting in Brussels, 20 May 1998. AHTN and HHCB have been tested in a rat two-generation study. The oral doses producing levels of AHTN and HHCB in milk of the lactating rat being approximately 1,000 times higher than the levels reported in human milk were determined. Groups of 28 time-mated rats were then dosed at that level and multiples of that level starting in the third week of pregnancy. This dosing was then continued post-partum until weaning. From the litters, randomly selected off-spring (24 males and females/group) were retained to maturity and assessed for general health and development as well as for behavioural effects and reproductive capability. The F2 generation was maintained until 21 days post-partum at which time all F1 and F2 animals were sacrificed. No effects were seen even at the highest doses. This study was performed by the Research Institute for Fragrance Materials (Ford and Bottomley 1997).

HHCB was negative in two genotoxicity (mutagenicity) tests: the micronucleus test with human lymfocytes and with the human hepatoma cell line Hep G2, in doses up to cytotoxicity (Kevekorde et al. 1997), and in the SOS chromotest (Mersch-Sundermann 1998). HHCB acts as a moderate irritant on rabbit skin (RTECS 2000). The acute toxicity for both AHTN and HHCB is relatively low as the lowest toxic doses exceed 4,500 mg/kg/day administered over a few days (RTECS 2000). No data on allergenicity were found.

12.3 Camphene

Ecotoxicology

Camphene (CAS No. 79-92-5) is a natural component in essential oils and a terpene found in camphor. In a modified MITI (I) test (OECD 301C) only 1-4% of camphene was degraded in 28 days (IUCLID 2000). Camphene is thus not readily biodegradable. The log K_ow is 4.1 and camphene is therefore potentially bioaccumulative in aquatic organisms. BCF values of 432-922 and 606-1290 were determined at exposure concentrations of 15 and 1.5 m g/l, respectively, in a 56-day bioaccumulation test with carp (Cyprino carpio).

The toxicity of camphene towards algae is low with EC50 values > 1,000 mg/l. For Daphnia magna an EC50 value of 22 mg/l has been determined (IUCLID 2000). The highest toxicity of camphene has been found in tests with fish. E.g., the LC50 were 0.72 mg/l (96 hours, flow-through) for zebra fish (Danio rerio), 1.9 mg/l (96 hours, static) for sheepshead minnow (Cyprinodon variegatus), and 2.0 mg/l (48 hours, static) for ricefish (Oryzias latipes) (IUCLID 2000).

Effects on humman health

When tested at 4% in petrolatum, camphene produced no irritation in a 48-hour closed patch test on human subjects. In a study of the sensitizing properties of 17 terpenes and related compounds found in essential oils, camphene was found not to be a sensitizer for human skin. Camphene is absorbed through the skin (HSDB 1998).

12.4 2-Pinene

Ecotoxicology

2-Pinene (CAS No. 80-56-8) is a main component of turpentine. Biotransformation has been examined in experiments confirming that the bacterium Pseudomonas maltophilia is able to grow on alpha-pinene with formation of the following metabolites: Limonen, borneol, campher, 2-(4-methyl-3-cyclohexeneyliden) propionic acid, and perill-acid. The log K_ow of 4.83 indicates that 2-pinene has the potential to bioaccumulate in aquatic organisms. The aquatic toxicity of 2-pinene to crustaceans has been examined in tests with two different species. The 48 h-EC50 was 41 mg/l towards Daphnia magna, whereas the LC50 ranged between 1 and 1.5 mg/l for Chaetogammarus marinus (48-96 hours) (IUCLID 2000).

The highest acute toxicity to aquatic organisms has been found in tests with fish, as an 96 h-LC50 of 0.28 mg/l was determined in a static test with fathead minnow (Pimephales promelas) (IUCLID 2000).

Human health

Contact sensitization is uncommon (De Groot et al. 1994). Application of pure pinene on human skin gives severe irritation. The oral rat LD₅₀ is 3,700 mg/kg. Pinene is absorbed through the skin and lungs (HSDB 1998). The Danish occupational exposure limit is 25 ppm (Arbejdstilsynet 2000).

12.5 d-Limonene

Human health

d-Limonene (CAS No. 5989-27-5) itself has a low sensitizing capacity. However, it is easily oxidized at air exposure and the oxidation products formed are strong sensitizers. The frequency of allergic reactions to d-limonene containing oxidation products is comparable to that of common allergens such as formaldehyde (Karlberg 1998). The oxidation of d-limonene may be counteracted by addition of antioxidants. The effects of such antioxidants, however, wear off with time, whereupon formation of oxidation products starts. Furthermore, the antioxidant BHT, which is commonly used in hand soaps, may constitute a health hazard, since it has been shown to promote skin cancer in mice after induction with benzo[a]pyrene (Taffe and Kensler 1988; Danish Toxicology Centre 1995). The Danish occupational exposure limit is 75 ppm (tentative, dipentene).

Classification

d-Limonene is included in Annex 1 of list of dangerous substances of Council Directive 67/548/EEC and classified as flammable with the risk phrase R10 (Flammable) and Irritant (Xi) with the risk phrases R38 (Irritating to skin) and R43 (May cause sensitisation by skin contact).

12.6 Camphor

Human health

Camphor (CAS No. 76-22-2) is moderately toxic with an LD50 of 1.31 g/kg. A fatal dose for a 1-year old child is 1 g of camphor. Cases of collapse have been reported after local application of camphor in the nostrils. The substance can be transferred to the fetus through placenta. Dust and vapors are very irritating to skin and mucous membranes. Camphor is quickly absorbed through skin. It is irritating to skin and eyes. Sensitization to camphor is rare.

12.7 Coumarin

Human health

Contact sensitization due to exposure with coumarin (CAS No. 91-64-5) may occur (De Groot et al. 1994). Oral rat LD50 is 293 mg/kg. Recent experiments have shown clear evidence of carcinogenic activity of coumarin in female B6C3F1 mice by oral administration, while there is some evidence in male F344/N rats and male B6C3F1 mice.

12.8 Terpineol

Human health

Contact sensitization due to exposure with terpineol is uncommon (De Groot et al. 1994). Oral rat LD50 is 4,300 mg/kg (RTECS 1998).

12.9 a -hexylcinnamaldehyde

Human health

Contact sensitization due to exposure with a -hexylcinnamaldehyde (CAS No. 101-86-0) is rare (De Groot et al. 1994). Oral rat LD50 is 3,100 mg/kg (RTECS 1998).

12.10 Eucalyptus oils

Human health

Contact sensitization is rare, but has been seen at concentrations as low as 2%. Oral rat LD50 is 2,480 mg/kg (RTECS 1998). Eucalyptus oils (CAS No. 8000-48-4) are moderately irritating for skin and eyes.

| Front page | | Contents | | Previous | | Next | | Top |