Registration Dossier

Environmental fate & pathways

Endpoint summary

Currently viewing:

Administrative data

Description of key information

Additional information

Abiotic processes

Since trimellitic anhydride (TMA) has been shown to be readily biodegradable (Point 5.2.1), a study of its hydrolysis under a range of pH conditions is not strictly necessary.

However, a study of the hydrolysis of the closely related substance trimellitic anhydride acid chloride has been performed in the context of the requirements for substances that may come into contact with food via packaging material. This study examined the behaviour of trimellitic anhydride acid chloride in unbuffered distilled water as a simple food simulant. Trimellitic anhydride acid chloride was found to hydrolyse completely at 40 °C in less than 0.5 hour to trimellitic acid and hydrogen chloride. Hydrolysis of the monomer was virtually instantaneous. Based on its close structural similarity, rapid hydrolysis is anticipated for TMA, which would also be expected to undergo almost instantaneous conversion to trimellitic acid (TMLA) on contact with aqueous matrices in the environment and in vivo in exposed humans and other organisms.

This behaviour defines TMLA (and its salts) rather than the anhydride as the entity relevant to exposure in the environment and also justifies the consideration of toxicology data provided by studies in which animals or test systems were administered TMLA (bypassing the_in vivo_hydrolysis step) rather than the parent TMA polyester monomer.

The stability of TMA in the atmosphere was calculated using the software AOPWIN (v1.92). It is predicted that the substance will be degraded in the atmosphere by reaction with photochemically-produced hydroxyl radicals, with a half-life ranged from 161.0 hours to 482.9 hours, at different test conditions.

The overall degradation rate constant with OH radicals is estimated to be equal to 6.89E-08 cm3/(molecule-day).

The stability of TMLA in the atmosphere was calculated using the software AOPWIN (v1.92). It is predicted that the substance will be degraded in the atmosphere by reaction with photochemically-produced hydroxyl radicals, with a half-life ranged from 78.6 hours to 235.8 hours, at different test conditions.

The overall degradation rate constant with OH radicals is estimated to be equal to 1.41E-07 cm3/(molecule-day).

In the presence of moisture, trimellitic anhydride (TMA) is rapidly hydrolysed to trimellitic acid (TMLA) and this entity is therefore more relevant to various environmental compartments than the submission substance.

The most conservative half-life value of 235.8 hours, obtained by applying the recommended northern hemisphere settings that are considered relevant in a European context, was then selected.


Two screening tests of the "ready" biodegradability of trimellitic anhydride (TMA) are available.

In the first (Lebertz, 1991a) TMA was tested for ready biodegradability according to the 1984 OECD 301B (Sturm Test) procedure, at concentrations of approximately 10 and 20 mg/L. The measured CO2 yield from TMA exceeded 60% of theoretical at both concentrations and the 60% threshold was crossed within the "10-day window", i.e. within 10 days of CO2 production reaching 10% of theoretical.

In the second (CITI, 1988), TMA applied at 100 mg/L was tested for biodegradability by the Chemicals Inspection and Testing Institute of Japan to fulfil the requirements of the Japanese Chemical Substances Control Law. A composite inoculum (applied at 30 mg suspended solids/L) originating from ten specified locations around Japan, not deliberately adapted to the test substance, fed with peptone and glucose prior to use and renewed at regular intervals (see OECD Guideline 301C 1984 and 1992 for details) was employed as standard practice at CITI for these investigations. An automated respirometer was used to make continuous measurements of biochemical oxygen demand (BOD) and recorded BOD was compared to the theoretical oxygen demand (ThOD) calculated assuming the complete mineralisation of TMA to its terminal oxidation products. This comparison provides a measure of ultimate biodegradation. Measured BOD expressed as %ThOD reached 96% within 28 days in this study. Confirmatory indications are provided by total organic carbon (TOC) analyses - this non-specific technique showed 99% loss of the tested substance (ultimate degradation) and are consistent with the figure of 96% for ultimate biodegradation based on BOD measurement.

Both studies demonstrate that trimellitic anhydride is readily biodegradable and this result signifies that TMA will degrade rapidly and completely, without the formation of stable metabolites, under aerobic conditions in a variety environmental compartments (aquatic and terrestrial) and that extensive biodegradation may be anticipated in aerobic biological wastewater treatment processes. This (in addition to exposure considerations) obviates the need for studies of the degradation of TMA in water/sediment systems or in soil.

Trimellitic anhydride is not persistent (not P).


The threshold that triggers the need to investigate a potential bioconcentration/bioaccumulation tendency experimentally is a log10 Kow value greater than or equal to 3.0.

The log Kow of trimellitic anhydride (TMA), determined experimentally according to the OECD 117 (HPLC method) is 0.06.

The log Kow of trimellitic acid (TMLA), determined experimentally according to OECD 107 (Shake Flask method) is 0.61.

The US EPA's KOWWIN model predicts a log Kow of 1.95 for TMA and a log Kow of 0.95 for its hydrolysis product TMLA.

Considering all the charge-state forms of trimellitic acid dissolved in water and in octanol, a LogD (distribution coefficient) was estimated taking account for the pH dependence of the substance in aqueous solution.

For the environmental and PBT assessments, trimellitic acid was therefore considered more significant, with logD= -4.83.

TMLA is not expected to remain stable in the form of the free acid under environmental conditions. Aquatic ecotoxicology studies have been conducted with TMA after converting it to the sodium salt(s) of TMLA .This is considered representative of the likely behaviour of TMA (TMLA) in the environment.

The log Kow of ionised forms of TMLA relevant to considerations of environmental exposure will be lower still. All these log10 Kow values lie below the trigger of 3.0 and both TMA and TMLA are therefore expected not to exhibit significant bioconcentration or bioaccumulation tendencies.

It may be concluded that both trimellitic anhydride and acid are not bioaccumulative (not B).

Transport and distribution

Adsorption / desorption

(Q)SAR-modelled adsorption coefficient (Koc) values for trimellitic anhydride (TMA) and trimellitic acid (TMLA) obtained with the KOCWIN v2.00 model of the US EPA range from ca. 0 to 387 L/kg. Based on these values, TMA and TMLA are classed as moderately to very mobile and are expected to have a low tendency to adsorb to soils and sediments. Koc may be influenced by and vary significantly in response to pH. Under environmental conditions where TMA is expected to have undergone complete hydrolysis, TMLA will rapidly be converted to salts whose Koc may be expected to be lower (mobility higher) than that of the free acid.

The low Koc value modelled for TMLA also implies a low tendency to associate with sludge solids during the primary settlement and secondary biological stages of waste water treatment. The majority of the TMLA load contained in a treatment plant influent load may therefore be expected partition to the aqueous phase and to be routed toward aerobic biological treatment. Since process effluents discharged to treatment facilities are typically neutralised to protect both the plant hardware (concrete and metalwork) from corrosion and the biological treatment process from pH-shock effects, TMLA is likely to be discharged in the form of salts that are more highly water soluble and have a correspondingly lower Koc than the parent acid. Salts formed by the pre-treatment neutralisation step are likely to have an even lower tendency than that of free TMLA to bind to sludge solids.

Henry's Law constants

The HLC value of 1.29E-05 Pa m³/mol at 25 deg C indicates that TMA is essentially not volatile from surface water

The HLC limit value of 1.01E-07 Pa m³/mol at 25 deg C indicates that TMLA is essentially not volatile from surface water.

Distribution modelling

- Mackay, Level III Fugacity modelling estimates distribution in which the soil is the main environmental compartment for the anhydride and both soil and water to be the significant environmental compartments for the acid degradation product should environmental exposure occur.