[Front page] [Contents]

Alternatives to brominated flame retardants

Summary

The Danish Environmental Protection Agency has initiated several projects on flame retardants. The present project is a screening of the reviews, handbooks and readily available literature and databases for information on environmental and health properties for a number of alternatives to brominated flame retardants.

The selected compounds have previously been identified in the project "Brominated Flame Retardants", Danish Environmental Protection Agency, 1999.

Table 2.1

Covered compounds and availability of environmental and health data. Poor, medium and good refers to a (subjective) assessment of the availability of data. It is not an evaluation of the quality of the data, nor whether sufficient data is available for a complete health and environmental assessment.

Trivial name

CAS no.

Physical-chemical

Health

Environment

Triphenyl Phosphate

115-86-6

Good

Good

Good

Tricresyl Phosphate

1330-78-5

Good

Poor (formulation data)

Poor (formulation data

Resorcinol bis(diphenylphosphate)

57583-54-7

Poor

Poor

Poor

Phosphonic acid (dimethyl ester)

20120-33-6

Poor

Poor

Poor

Aluminium Trihydroxide

21645-51-2

Medium

Poor (partly data on aluminium)

Poor (partly data on aluminium)

Magnesium Hydroxide

1309-42-8

Medium

Poor

Poor

Ammonium Polyphosphates

14728-39-9 and 68333-79-9

Poor

Poor

Poor (formulation data)

Red Phosphorus

7723-14-0

Medium

Medium (different allotropic forms)

Medium (different allotropic forms)

Zinc Borate

1332-07-6

Poor

Poor (data for boric acid and zinc)

Poor (data for sodium borate and zinc)

Melamine

108-78-1

Good

Medium

Medium

Antimontrioxide

1309-64-4

Good

Good

Good

Quinidincarbonate

Not available

Poor

Poor

Poor (data from quinidine sulphate)

The data availability is very variable among the suggested alternatives for brominated flame retardants. In the screening project information is collected based in the name or CAS number of the suggested compound. Therefore, a precise match of name and number is required and as shown in the table above a poor availability of data is not uncommon. However, if the compound is a slight modification of another compound or belongs to a family of related compounds it is possible that useful information can be obtained by searching for information on such compounds in a more comprehensive project (e.g. in the case of quinidine carbonate).

Several of the inorganic compounds are salts of metals and may dissociate in the hydrosphere. To some extent the lack of data on the selected compounds may be ameliorated by using data on the parent metal. In the case of zinc borate this is, however, somewhat complicated since both zinc and boric acid may contribute to the combined toxicity.

The type of data that are missing varies between compound. Typically missing data on the environment side are biodegradation data and bioaccumulation data. On the health side a less clear pattern is observed.

The available data indicate that the triphenyl and tricresyl phosphates may have low impact on health, but are quite toxic in the environment. Poor data availability for the structurally related resorcinol prohibits conclusions regarding the effect pattern.

For this compound only very few data was identified. The phosphonic acid (dimethyl ester) appears acutely toxic at 13 mg/kg bodyweight in rats and mutagenic effects has been reported. A formulation of the compound was lethal to fish (LC50) at approx. 1 ml/l (density unknown).

The data sets on these compounds are relatively limited. It appears that limited toxic effects can be induced in mammals after exposure to high doses. Aluminium trihydroxide is generally not toxic in the available tests. Both metal-ions play a metabolic role in mammals, but the data for the metal-ions indicates acute toxic levels for Al to fish and crustaceans at <1-10 mg/l and approx. 65 mg/l for crustaceans exposed to Mg.

Red phosphorus data are limited and conclusions are unclear. The yellow phosphorus is reportedly acutely toxic to humans (fatal dose 1 mg/kg), but the red allotropic form is described as less toxic. Acute toxic concentrations (LC50 or EC50) of unspecified allotropic form in the aquatic environment occurs at 0.009 – 0.012 mg/l for fish and crustaceans.

There is practically no data on the compound. Based on comparison with sodium borate and boric acid the possible main effects in humans are expected to be irritation of skin, eyes and throat, and harm to the unborn child. In the environment zinc-ion is very toxic to crustaceans.

Melamine seems to be only mildly toxic when ingested by animals. The available data does not show evidence of cancer induction by melamine. One experiment indicates that melamine may be harmful to crustaceans, but otherwise the reviewed toxicity data show little aquatic toxicity.

The bioaccumulation of this compound is presumably low in the natural pH range (pH 6-8). The available biodegradation data indicates that this compound is persistent both under aerobic and anaerobic conditions.

Antimony trioxide is in the EU classified as "Harmful (Xn)" and must be labelled with the risk-phrase "Possible risk of irreversible effects" (R40) due to possible carcinogenicity. The substance is reported as teratogenic. The effects in ecotoxicological test are primarily on algae (ranging from very toxic to harmful), but toxicity in crustaceans or fish is very low.

No data was identified on quinidine carbonate for health or environmental properties. The toxicity of quinidine carbonate estimated from the toxicity of quinidine sulfate indicates that quinidine carbonate could be harmful to crustaceans, but not to fish.

[Front page] [Contents] [Top]