Trimelitic Acid
Trimelitic Acid
UNEP PUBLICATIONS 1
FOREWORD INTRODUCTION
TRIMELLITIC ANHYDRIDE &TRIMELLITIC ACID
CAS N: 552-30-7; 528-44-9
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
2
SIDS Initial Assessment Report
For
15th SIAM
(Boston, USA, 22-25 October 2002)
Chemical Name: Trimellitic Anhydride (TMA) and Trimellitic Acid
(TMLA)
CAS No: 552-30-7, 528-44-9
Sponsor Country: U.S.A and ICCA
National SIDS Contact Point in Sponsor Country:
Mr. Oscar Hernandez Industry: Mr. Dave Dutton
U.S. EPA BP Amoco
Risk Assessment Division (7403M) 150 W. Warrenville Road
1200 Pennsylvania Ave, NW Naperville, Illinois 60563
Washington, DC 20460 Tel: (630) 420-5079
Tel: 202-564-7641 [email protected]
[email protected]
HISTORY:
There was no additional testing needed to complete the SIDS endpoints. Collection of data and
preparation of the documents was performed by industry with a separate review process from panel
members. Documents were then submitted to the sponsor country. The sponsor country performed two
independent reviews and prepared comments to industry for finalisation of documents to be considered
at SIAM 14. Both industry and the sponsor country worked jointly to finalise documents for SIAM 14.
The follow data sources were reviewed in the preparation of this document: Hazardous Substance Data
Base (HSDB), ChemSystems, 2000, SRI 2000, SRC PhysProp Database, Registry of Toxic Effects of
Chemical Substances (RTECS), IUCLID, International Chemical Safety Cards, NIOSH Summary,
OSHA Health Guidelines, International Occupational Safety and Health Information Centre, NTP
Chemical Repository, ACGIH TLV Documentation. Literature searches were conducted on Medline,
Publine and Toxline.
COMMENTS:
Deadline for circulation: February 1, 2002
Date of Circulation:
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
3
SIDS INITIAL ASSESSMENT PROFILE
CAS No. 552-30-7
528-44-9
Chemical Name Trimellitic Anhydride (TMA)
Trimellitic Acid (TMLA)
Structural Formula
OH
O
O
O
O
O
O H
O H
O OH
O
SUMMARY CONCLUSIONS OF THE SIAR
Category/Analogue Rationale
Trimellitic anhydride (TMA) and trimellitic acid (TMLA) are considered to be structural analogues. In addition, in
aqueous environments TMA is readily converted to TMLA. TMA rapidly forms TMLA under the conditions used
to test its toxicity, the toxicities of TMA and TMLA are qualitatively believed to be the similar for systemic toxicity
with these two chemicals being presented as analogues. The only difference being sensitization and potentially local
effects/reactions at the initial point of contact (skin, eye, and respiratory irritation). The sensitization potential of
TMA, may be directly attributed to the formation of haptens following a reaction with proteins. TMLA does not
react with proteins to form haptens, and therefore does not share this mode of action for sensitization.
Human Health
TMA exhibits low acute toxicity by the oral, dermal, and inhalation routes. The oral LD
50
has been reported to
range from 2,030 to 3,340 mg/kg in male and female rats, with stomach lesions appearing as the most consistent
lesion upon necropsy. In rats, the inhalation LC
50
value was reported to exceed a concentration of 2,330 mg/m
3
,
with lung lesions appearing as the most consistent lesion upon necropsy. The LC
50
for TMLA was reported to be
>3,750 mg/m
3
, with necropsy findings considered within normal limits. A dermal LD
50
value of 5,600 mg/kg was
reported for TMA. Because TMA rapidly converted to TMLA in the body, the acute toxicity of TMLA is expected
to be similar to that of TMA. Both chemicals are considered to have mild skin and severe eye irritation potential.
Studies on TMA suggest that these materials may also be respiratory sensory irritants. TMA but not TMLA should
be considered a dermal sensitizer.
In repeated dose inhalation studies, the principal effects of TMA are on the immune system and the lung. In a 13-
week inhalation repeat dose study, elevated antibody levels and lung foci were observed in rats following exposures
to relatively low concentrations of TMA (0.002 0.054 mg/m
3
), however a NOAEL was not identified. Elevated
antibody levels, asthma, allergic rhinitis, and a late respiratory systemic syndrome (LRSS) are associated with
occupational exposures to TMA in some workers. The toxicity of TMA following repeated oral exposures is low,
based on NOAELs of approximately 500 mg/kg-day identified for both rats and dogs. In a 13 week inhalation
study, immunological and pulmonary effects were not associated with repeated exposures to TMLA; the NOAEL
was determined to be 300 g/m
3
(the highest dose tested). In vivo genotoxicity data are not available however, three
in vitro assays with TMA were negative. Although a reproductive toxicity test has not been conducted for TMA,
histopathological changes to reproductive tissues have not been observed in rats following subchronic exposures,
and it has been found to be neither teratogenic nor fetotoxic in developmental toxicity studies.
TMA: C9H4O5
TMLA: C
9
H
6
O
6
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
4
Environment
TMA has a melting point of 165C, a boiling point of 390C, a vapor pressure of 7.6 x 10
-5
Pa @ 25C, and
assuming no hydrolysis a log K
ow
of 1.95 and a water solubility of 1,036 mg/L. TMLA has a melting point of
219C, an unknown boiling point, a vapor pressure of 3.8 x 10
-6
Pa @ 25C, a log K
ow
of 0.95 and a water solubility
of 21,000 mg/L. The half-life of TMA and TMLA in air is estimated to be 13.4 and 6.6 days, respectively, due to
direct reactions with photochemically generated hydroxyl radicals. In the presence of water, TMA rapidly
hydrolyzes (within 10 minutes) to form TMLA. Based on both chemicals physical chemical properties, TMA and
TMLA are likely to partition to the water compartments in the environment. Acute toxicity testing in fish,
invertebrates, and algae indicate a very low order of toxicity with measured No-Observed-Effect -Concentrations
(NOECs) of 896, 792 and 739 mg/L, respectively. TMA and TMLA are readily biodegraded under aerobic
conditions in sewage sludge, and are expected to biodegrade in soil and water as well. TMA and TMLA are not
expected to bioaccumulate in food webs based on a BCF of 3.2.
Exposure
Approximately 100,000 metric tonnes/year TMA are produced worldwide, the majority of which (65,000 metric
tonnes/year) are produced in the U.S. Most of the TMA produced (65%) is used in the synthesis of plasticizers for
PVC resins, while smaller amounts (30%) are used as a reactant in wire and cable insulation enamels and polyester
resins for powder coatings. When TMA is processed into the above materials, it is fully consumed and therefore, is
not readily available for releases to the environment. All TMLA produced is used to make TMA. Occupational
exposures to TMA and TMLA are likely to occur by the inhalation and dermal routes in settings where TMA is
produced or used. Historical monitoring data have revealed mean concentrations ranging from 0.00051 to 0.77
mg/m
3
. Because TMA is rapidly hydrolyzed to form TMLA in the presence of water, consumer and environmental
exposures to TMA are not anticipated. Data regarding these potential exposures to TMLA are largely lacking, but
exposures are expected to be low outside of the workplace.
RECOMMENDATION
The chemical is currently of low priority for further work.
RATIONALE FOR THE RECOMMENDATION AND
NATURE OF FURTHER WORK RECOMMENDED
Although a reproductive toxicity study and an in vivo genotoxicity are not available for TMA or TMLA, sufficient
data are available to address these endpoints. Therefore, no additional studies are recommended to meet the SIDS
data set.
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
5
Full SIDS Summary
CAS No.: 552-30-7 SPECIES PROTOCOL RESULTS
PHYSICAL CHEMISTRY
2.1 Melting Point 165 C
161-168 C
2.2 Boiling Point 390 C
240-245 C @ 19 x 10
2
Pa
2.3 Density 1.54
2.4 Vapor Pressure EPIWIN Suite,
1997
7.6 x 10
-5
Pa
1.3 x 10
-3
Pa
1.4 x 10
-5
Pa
2.5 Partition Coefficient
KOWWIN v1.66
KOWWIN v1.66
CLOGP
ALOGP
XLOGP
After hydrolysis to TMLA
0.95
No hydrolysis
1.95
1.61
0.80
1.14
2.6 Water Solubility
Measured
WSKOW V1.40
WSKOW v1.4
Int Act
AnalALOGSP
After hydrolysis to TMLA
21,000 mg/L
Assuming no Hydrolysis occurs
1,036 mg/L
1,211 mg/L
860 mg/L
2,777 mg/L
Hydrolysis Measured Hydrolysis complete within 10 minutes.
ENVIRONMENTAL FATE AND PATHWAYS
3.1.1 Photodegradation Estimate
AOPWIN
Half-life: 13.4 days
3.2 Monitoring Data Occupational 0.00051 0.77 mg/m
3
3.3 Environmental fate &
distribution
Estimate v 2.2
Level I
Level II
Level III
Assumes hydrolysis to TMLA
Air 7.68 x 10
-7
%
Water 99.2%
Soil 0.78%
Sediment 0.02%
Air 7.68 x 10
-7
%
Water 99.2%
Soil 0.78%
Sediment 0.02%
Air 3.46 x 10
-6
%
Water 50.6%
Soil 49.3%
Sediment 0.02%
3.5 Biodegradation Modified Sturm
(OECD 301B)
>60% within 7 days
89-101% within 28 days
3.7 Bioaccumulation Calculated
BCFWIN v2.14
BCF = 3.2
ECOTOXICOLOGICAL DATA
4.1 Acute Fish Leuciscus idus
melanotus
OECD 203 96-hour NOEC > 896 mg/L
4.2 Acute Daphnid Daphnia
magna
OECD 202 48-hour EC
0
> 792 mg/L
4.3 Acute Aquatic Plant Scenedesmus
subspicatus
OECD 201 96-hour NOEC > 739 mg/L
4.4 Toxicity to Bacteria Activated
sludge
OECD 209 EC
50
- >100, <500 mg/L
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
6
CAS No.: 552-30-7 SPECIES PROTOCOL RESULTS
TOXICOLOGICAL DATA
5.1.1 Acute Oral Rat LD50 = 2,730 mg/kg
5.1.2 Acute Inhalation Rat
Mouse
LC50 > 2,330 mg/m
3
LOEL = 21.5 mg/m
3
5.1.3 Acute Dermal Rabbit
Rat
LD50 > 2000 mg/kg
LD50 = 5600 mg/kg
5.2.1 Skin Irritation Rabbit PDIS = 1.7/8
5.2.2 Eye Irritation Rabbit Draize score 110/110
5.3 Sensitization Guinea Pig
Guinea Pig
Mouse
Mouse
Mouse
Rat
Dermal
Dermal
Dermal
Dermal
Dermal
Dermal
Positive (in acetone)
Negative (applied neat)
Positive (in acetone/olive oil)
Positive (in acetone/olive oil)
Positive (in acetone/olive oil)
Positive (in acetone/olive oil)
5.4 Repeated Dose Rat
Rat
Rat
Rat
Rat
Mouse
Rat
Rat Respiratory
Sensitization
13-week inhalation study with 0, 3, and 38 week
recovery: Slight increase in lung weight and volume,
slight pulmonary pneumonia. Pulmonary physiology
parameters unaffected Antibody levels elevated in a
dose-dependent manner. Minimal effects in the 3
and 38 week recovery group.
LOEL = 0.002 mg/m
3
2-week inhalation study with 0-12 day recovery:
NOAEL >0.3 mg/m
3
6.5-week inhalation study: Increased lung weight,
and volume, external hemorrhagic foci,
inflammatory cell infiltration, and bronchoalveolar
pneumonia. Effects were more severe than in the 13-
week study Antibody levels and lung foci were
elevated in a dose-dependent manner. LOEL = 0.002
mg/m
3
2-week inhalation study: Increased hemorrhagic foci
of the lung, increased lung weight, and TMA
specific antibodies were observed. Effects greater in
males than females. Estrogen reduced foci in both
males and females. Testosterone had no effect.
LOEL = 0.5 mg/m
3
2,6, or 10 day inhalation study: No effect after 2
days, minimal lung injury after 6 days marked after
10 days .
LOEL = 0.1 mg/m
3
5-day inhalation study, 14-day recovery: Decreased
time of inspiration and expiration, increased length
of apneic periods.
LOEL 0.01 mg/m
3
1-2-week inhalation study, 0-12 day recovery:
Antibody response elevated in a dose-dependent
manner at ten and 22 days but not at five days. Lung
foci completely resolved after 12 days of recovery,
but reappeared following a single challenge
exposure.
LOEL = 0.01 mg/m
3
Animals exposed on day 1,5, and 10, challenged on
day 22 or 29 inhalation study: Elevated antibodies
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
7
Rat
Rat
Rat
Rat
Dog
correlated to the number of lung foci, lung weight
and lung displacement volume.
LOEL = 0.33 mg/m
3
1-10 day inhalation study: Lung injury and antibody
levels increased on day 7-10.
LOEL0.5 mg/m
3
2,6, or 10 day inhalation study: Increase antibody
levels.
LOEL 0.1 mg/m
3
13-week oral feeding study: No effect on
appearance, behavior, pathology, or urine values.
Dose dependent increase in leukocyte count.
LOEL = 1000 ppm in the diet or approximately 50
mg/kg
90-day oral feeding study: No effects on appearance,
behavior, pathology, urine values or leukocyte
count.
NOEL = 10,000 ppm in the diet or approximately
500 mg/kg.
13-week oral feeding study: No effects observed on
appearance, behavior, pathology, serum chemistry,
or urine values.
NOEL 20,000 ppm in the diet or approximately 500
mg/kg
5.5 Genetic Toxicity In Vitro
A Bacterial Salmonella
typhimurium
Salmonella
typhimurium
OECD 471
OECD 471
Negative with and without metabolic activation.
Negative with and without metabolic activation
B Non-Bacterial Chinese
Hamster
Ovary Cells
Chinese
Hamster
Ovary Cells
OECD 476
(HGPRT
mutation assay)
OECD 473
(chromosome
aberration )
Negative with and without metabolic activation
Negative with and without metabolic activation
5.6 Genetic Toxicity In vivo
5.7 Carcinogenicity
5.8 Reproductive Toxicity Repeat dose No effect on reproductive organs in two rat and one
dog sub-chronic feeding studies.
NOEL approximately 500 mg/kg
No effect of reproductive organs in sub-chronic rat
inhalation study
NOEL 0.054 mg/m
3
5.9 Developmental
Toxicity/Terotogenicity
Rat, Guinea
Pig
Developmental No fetotoxicity or developmental toxicity at
concentrations up to 0.5 mg/m
3
. No maternal
toxicity other than an increase in hemorrhagic lung
foci.
NOEL: for developmental and terotogenic effects
0.5 mg/m
3
5.10 Toxicokinetics Rat Tmax =<3 houurs. Elimination rate constanst ranged
from 0.0015-0.214, biological half-life ranged from
3-46 days.
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
8
Full SIDS Summary
CAS No.: 528-44-9 SPECIES PROTOCOL RESULTS
PHYSICAL CHEMISTRY
2.1 Melting Point Estimate
MPBPWIN v1.4
219 C
2.2 Boiling Point Converts to anhydride prior to boiling
2.3 Density
2.4 Vapor Pressure EPIWIN
MPBPWIN v1.4
3.8 x 10
-6
Pa
2.5 Partition Coefficient KOWWIN
v1.66
CLOGP
Int Anal Prog
ALOGP
XLOGP
0.95
0.57
0.81
0.78
0.87
2.6 Water Solubility 21,000 mg/L @ 25 C
ENVIRONMENTAL FATE AND PATHWAYS
3.1.1 Photodegradation Estimate
AOPWIN
Half-life: 6.55 days
3.2 Monitoring Data
3.3 Environmental fate &
distribution
Estimate v2.2
Level I
Level II
Level III
Air 7.68 x 10
-7
%
Water 99.2%
Soil 0.78%
Sediment 0.02%
Air 7.68 x 10
-7
%
Water 99.2%
Soil 0.78%
Sediment 0.02%
Air 3.46 x 10
-6
%
Water 50.6%
Soil 49.3%
Sediment 0.02%
3.5 Biodegradation Modified Sturm
(OECD 301B)
>60% within 7 days (test material
TMA)
89-101% within 28 days (Test
material TMA)
3.6 COD
3.7 Bioaccumulation
ECOTOXICOLOGICAL DATA
4.1 Acute Fish Leuciscus idus
melanotus
OECD 203 96-hour NOEC > 896 mg/L (Test
material TMA)
4.2 Acute Daphnid Daphnia
magna
OECD 202 48-hour EC0 > 792 mg/L (Test
material TMA)
4.3 Acute Aquatic Plant Scenedesmus
subspicatus
OECD 201 96-hour NOEC > 739 mg/L (Test
material TMA)
4.4 Toxicity to Bacteria Activated
sludge
OECD 209 EC5 - EC50 - >100, <500mg/L (Test
material TMA)
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
9
CAS No.: 528-44-9 SPECIES PROTOCOL RESULTS
TOXICOLOGICAL DATA
5.1.1 Acute Oral Rat LD50 = 2,730 mg/kg Test material
TMA)
5.1.2 Acute Inhalation Rat LC50 > 3,750 mg/m
3
5.1.3 Acute Dermal Rabbit
Rat
LD50 > 2000 mg/kg (Test material
TMA)
LD50 = 5600 mg/kg (Test material
TMA)
5.2.1 Skin Irritation Rabbit PDIS = 0.7/8
5.2.2 Eye Irritation Rabbit Draize score = 59.7/110
5.3 Sensitization Rat Inhalation Negative
5.4 Repeated Dose Rat
Rat
Inhalation
(OECD 407)
13-week inhalation study, NOEL: 0.3
mg/m
3
4-week oral gavage study: Abnormal
findings were restricted to diarrhea at
the highest dose.
NOEL: 300 mg/kg
5.5 Genetic Toxicity In vitro
A Bacterial Salmonella
typhimurium
Salmonella
typhimurium
OECD 471
OECD 471
Negative with and without metabolic
activation. (Test material TMA)
Negative with and without metabolic
activation (Test material TMA)
B Non-Bacterial Chinese
Hamster
Ovary Cells
Chinese
Hamster
Ovary Cells
OECD 476
OECD 473
Negative with and without metabolic
activation. (Test material TMA)
Negative with and without metabolic
activation. (Test material TMA)
5.6 Genetic Toxicity In vivo
5.7 Carcinogenicity
5.8 Reproductive Toxicity Repeat dose No effect on reproductive organs in
two rat and one dog sub-chronic
feeding studies.
NOEL approximately 500 mg/kg
No effect of reproductive organs in
sub-chronic rat inhalation study
NOEL 0.054 mg/m
3
. (Test material
TMA)
5.9 Developmental
Toxicity/Teratogenicity
No fetotoxicity or developmental
toxicity at concentrations up to 0.5
mg/m
3
. No maternal toxicity other
than an increase in hemorrhagic lung
foci.
NOEL: for developmental and
terotogenic effects 0.5 mg/m
3
. (Test
material TMA)
5.10 Toxicokinetics Rat Tmax =<3 hours. Elimination rate
constant ranged from 0.0015-0.214,
biological half-life ranged from 3-46
days. (Test material TMA)
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
10
SIDS Initial Assessment Report
Analog Justification
Because TMA and TMLA are structurally similar, and because TMA is readily converted to TMLA in
aqueous environments, information on these two chemicals is presented together in a single SIAP and
SIAR. Studies on stability of TMA in water suggest complete hydrolysis occurs in less than ten minutes.
Since TMA rapidly forms TMLA under the conditions used to test its toxicity, the toxicities of TMA and
TMLA are believed to be the same with a few exceptions. Specifically, the allergic symptoms (both
respiratory and dermal sensitization) of TMA are directly attributable to the reaction of TMA with free
amines within proteins to form haptens, which when present at sufficiently high levels in tissues can
produce sensitization. TMLA does not react with proteins to form haptens, and therefore does not share
this mode of action for sensitization. In addition to the immunological differences, there may be some
slight quantitative differences in reactions at the site of contact. While both materials cause slight
irritation of the skin and are severe eye irritants, there appears to be a difference in the magnitude of
response. This difference in the magnitude of irritation may be attributable to the heat of hydrolysis of
the anhydride. With the exception of the immunological and site of contact effects, all other endpoints
are expected to be the same.
1.0 IDENTITY
In the presence of water, trimellitic anhydride (TMA: CASRN 552-30-7) is readily and completely
converted to trimellitic acid (TMLA: CASRN 528-44-9). The chemical properties of TMA and
TMLA are summarized in the table below.
Property Value
Chemical Formula TMA: C
9
H
4
O
5
TMLA: C9H6O6
TMA
OH
O
O
O
O
TMLA
O
O H
O H
O OH
O
Molecular Weight TMA: 192.12
TMLA: 210.14
Purity TMA: 98%
TMLA: >98%
Impurities TMA: TMLA
TMLA: --
Physical form: TMA: Solid white flake
TMLA: Solid white crystal
Melting Point TMA: 165
o
C
TMLA: 219
o
C
Boiling Point TMA: 390
o
C
TMLA: --
Density TMA: 1.54 g/mL at 20
o
C
TMLA: --
Vapor Pressure TMA: 7.6 x 10
-5
Pa at 25
o
C
TMLA: 3.8x10
-6
Pa at 25
o
C
Partition Coefficient (Log K
ow
)* TMA: 1.95 (assumes no hydrolysis or reaction with the alcohol)
TMLA: 0.95 (non-ionized form)
Water Solubility* TMA: 1,036 mg/L (assumes no hydrolysis)
TMLA: 21,000 mg/L
Hydrolysis Rate TMA: Complete hydrolyse within 10 minutes.
Odor Threshold --
Synonyms TMA: 1,2,4-benzenetricarboxylic acid, cyclic 1,2-
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
11
anhydride
anhydrotrimellitic acid
trimellitic acid anhydride
1,2,4-benzenetricarboxylic acid anhydride
1,3-dioxo-5-phthalancarboxylic acid
4-carboxyphthalic anhydride
TMLA: 1,2,4-benzenetricarboxylic acid
*The water solubility and partition coefficient for trimellitic anhydride are listed above for completeness.
However, the most environmentally relevant value must reflect hydrolysis of the anhydride to the acid.
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
12
2.0 GENERAL INFORMATION ON EXPOSURE
Manufacturing and Processing
In the U.S. trimellitic anhydride (TMA) is produced in a batch process. Pseudocumene and air, the
raw materials, are mixed with solvent and catalyst in an enclosed reactor. The reaction is run to
completion and to produce trimellitic acid (TMLA). TMLA, a reaction intermediate, is separated
from the solvent and routed through a series of purification steps. It is then dehydrated to form
trimellitic anhydride (TMA) and water as the byproduct. TMA is further treated to remove impurities,
then flaked and stored in silos. TMA product is packaged for sale in bags of various sizes, most
commonly, 1 and 0.5 metric ton, 50 Kg and 25 Kg. TMLA has no separate commercial value. Because
the manufacturing system is a closed process, very little TMA is released to the environment during
the production process. Solvent is recovered and recycled. Spent catalyst containing some impurities
is routed to a catalyst recovery furnace. Catalyst is reclaimed. For the one and only manufacturing
site in the U.S., waste streams are routed to an on-site wastewater treatment plant that utilizes both
anaerobic and aerobic treatment processes.
Estimated National Production or Import Volume
Currently, TMA capacity worldwide is about 100,000 metric tonnes/year, which may be broken down
to approximately 65,000 metric tonnes/year produced in the U.S. and the remainder produced outside
the U.S (ChemSystems, 2000; SRI, 2000). In 1990, TMA production worldwide was estimated to be
50,000 metric tonnes/year (IPCS, 1992). Production in the 1970s was estimated to exceed 2.3 metric
tonnes/year (HSDB, 2001). These data suggest that TMA production is generally increasing over
time.
Uses and Functions
TMA is a highly reactive chemical and is a starting material for a variety of organic chemical
products. Approximately 65% of the TMA produced in the U.S. is used in the synthesis of plasticizers
for polyvinyl chloride (PVC) resins. These plasticizors have applications in wire and cable insulation,
automotive parts and medical equipment. Approximately 30% of the TMA produced in the U.S. is
used as a reactant in wire and cable insulation enamels and polyester resins in powder coatings. The
remaining 5% of U.S. production is used for a variety of purposes including as an epoxy curing agent,
textile sizing agent, rubber curing accelerator, electrostatic toner binder, and vinyl cross-link agent
(ChemSystems, 2000; SRI, 2000). TMA is fully consumed in these uses and is therefore not available.
100% of the TMLA produced in the U.S. is used to make TMA.
Form of Marketed Product
TMA is used in the synthesis of plasticizers that are in turn compounded with PVC to make flexible
plastic products such as automotive dashboards and coatings for electrical wire and cable. TMA is
also used in polyester resin in products used in military, industrial and aerospace applications
(ChemSystems, 2000; SRI, 2000). Many of the epoxy resin and surface coating systems may contain
2-10% reacted TMA within the polymer (OSHA, 1992). Coatings on the inside of tin cans used for
foodstuffs contain up to 0.04% reacted TMA within the polymer.
No TMLA is marketed to consumers per se since all is used for the production of TMA.
2.1 Environmental Exposure and Fate
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
13
2.1.1 Photodegradation
Given their relatively low vapor pressures, TMA or TMLA that may be present in the atmosphere is
expected to be associated predominantly with the particulate phase, which may be removed by both
wet and dry deposition processes. Half-lives of 13.4 and 6.6 days have been estimated for TMA and
TMLA, respectively, using the AOPWIN software (SRC, 2001) based on reaction with photo-
chemically derived hydroxyl radicals.
2.1.2 Stability in Water
TMA is expected to rapidly hydrolyze to form TMLA in water (complete hydrolysis in 10 minutes in
water at 27-32
o
C) (Horan , 1962). Given their relatively low vapor pressures, volatilization of TMA
or TMLA from surface water is also not expected to be significant. Based on studies using inoculated
sewage sludge in which more than 50% of TMA/TMLA was degraded within 5 days (Letz et al.,
1987; Lebertz, 1991a), biodegradation of TMLA may occur in water under aerobic conditions.
2.1.3 Stability in Soil
TMA is expected to hydrolyze to form TMLA in moist soils. Given their relatively low vapor
pressures, volatilization of either chemical from surface soils is not expected to be significant. Based
on studies using inoculated sewage sludge in which more than 50% of TMA/TMLA was degraded
within 5 days (Letz et al., 1987; Lebertz, 1991), biodegradation of TMLA may occur in soils under
aerobic conditions.
2.1.4 Environmental Transport and Distribution
Using default release estimates, predictions based on Levels 1 and 2 fugacity-based fate and transport
models (Trent University, 1999) suggest that the majority of the TMA or TMLA released to the
environmental will partition primarily to the water (99.2%) compartment, with a smaller amount
found in the soil compartment (0.78%), and negligible amounts in the air and sediment (<0.1%)
compartments. Although no specific information was located regarding the amount or mechanism of
TMA release to the environment, a Level 3 fugacity model was also used. Based on a Level 3
fugacity model (Trent University, 1999), the majority (50.6%) of TMA/TMLA released is predicted to
partition to the water compartment, with a slightly smaller amount in the soil compartment (49.3%),
and negligible amounts in the sediment (<0.1%) and air (<0.1%) compartments. A larger percentage
is predicted for soil by the Level 3 model, since this level allows for continuous release to soil.
However, specific data regarding the direct release of TMA to soil were not located.
2.1.5 Biodegradation
TMA was readily degraded in screening tests using sewage sludge. Under aerobic conditions, 97%
and 77% of the theoretical CO2 was generated within 28 days when TMA was tested at concentrations
of 10mg/L and 20 mg/L respectively (Lebertz, 1991a). In another study 89-101% of the TMA was
removed over a 4-week period (Letz et al. 1987). Given the rapid hydrolysis of TMA to TMLA in
aqueous systems, results most likely reflect biodegradation of TMLA. TMA and TMLA are readily
biodegradable under aerobic conditions in sewage sludge, are expected to biodegrade in water and soil
as well and are not expected to bioaccumulate.
2.1.6 Bioaccumulation
Using v2.14 of BCFWIN a Bioconcentration Factor (BCF) of 3.2 was calculated for TMA and TMLA
which suggests that they are not expected to bioconcentrate in aquatic organisms.
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
14
2.2 Human Exposure
2.2.1 Occupational Exposure
Occupational exposures to TMA or TMLA would most likely occur via the inhalation and dermal
routes. Although little information is available to quantify potential dermal exposures, a number of
studies have reported TMA concentrations in air associated with occupational exposures. It is assumed
that workers are predominantly exposed to TMA. Average airborne TMA dust concentrations ranged
from 0.006-2.1 mg/m
3
for production workers from three different job categories (Bernstein et al.,
1983). After engineering improvements were made, TMA concentrations decreased to approximately
0.01 mg/m
3
. Industrial hygiene data from a TMA production plant in 1989 revealed exposure
concentrations ranging from 0.003 to 0.77 mg/m
3
(Grammer et al., 1991). In the U.S. at the
production site, TMA concentrations were measured over a 14-year period and determined to range
from <0.001-2.1 mg/m
3
(in air) for workers belonging to several different job categories (Grammer et
al., 1992). The highest arithmetic mean TMA concentration detected in a resin factory was reported
to be 0.0193 mg/m
3
(van Tongeren et al., 1995). Geometric mean exposure concentrations calculated
from personal monitoring data ranged from <0.00053 to 0.17 mg/m
3
for various exposure classes of
workers at a large manufacturing complex producing TMA (Zeiss et al., 1992; Grammer et al., 1999).
Average concentrations of TMA for a full shift were reported to range from 0.0005 to 0.0193 mg/m
3
for four facilities using TMA (Barker et al., 1998).
Monitoring data collected for a TMA-manufacturing plant during 1988-1999, reported mean 8-hour
TWA concentrations ranging from 0.002 to 0.43 mg/m
3
and STEL concentrations ranging from 0.045
to 0.70 mg/m
3
for workers from four different job categories (BP Amoco personal communication,
2001).
Occupational exposure limits (OELs) for TMA are listed below for several countries.
Exposure Limit (Country) (mg/m
3
) (ppm)
TWA (Australia) 0.04 0.005
TWA (Canada) 0.04 0.005
MAK (Czechoslovakia) dust
fumes
0.04
0.005
0.005
0.0006
MAK (Germany) 0.04 0.005
MAK (Netherlands) 0.04 0.005
OES (United Kingdom) 0.04 0.005
NIOSH REL (U.S.) 0.04 0.005
ACGIH TLV (U.S.) 0.04 0.005
STEL (Germany) 0.08 0.01
In 1978, NIOSH estimated that approximately 20,000 workers in the U.S. had potential for exposure
to TMA in various applications and processes (OSHA, 1992).
2.2.2 Consumer Exposure
Because TMA rapidly hydrolyzes to form TMLA, consumer exposure to TMA is not expected to
occur. TMA present in consumer products is generally reacted to form polymers and therefore is
contained within the matrix of the polymer. As a result, potential exposures to consumers is
negligible
2.2.3 Indirect Exposure via the Environment
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
15
Based on manufacturing and processing procedures in the U.S., releases to the environment are
anticipated to be negligible. In addition, TMA rapidly hydrolyzes to form TMLA in the presence of
water therefore, significant environmental exposures to TMA are not expected to occur.
Data regarding potential environmental exposures to TMLA were not located. As previously
discussed, results from fugacity modeling indicates that should releases of the chemicals occur, the
primary partitioning compartment would be the water.
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
16
3.0 HUMAN HEALTH HAZARDS
Analog Justification
Because TMA and TMLA are structurally similar, and because TMA is readily converted to TMLA in
aqueous environments, information on these two chemicals is presented together in a single SIAP and
SIAR. Since TMA rapidly forms TMLA under the conditions used to test its toxicity, the toxicities of
TMA and TMLA are believed to be the same with perhaps two exceptions. Specifically, the allergic
symptoms of TMA are directly attributable to the reaction of TMA with amino acids to form haptens,
which when present at sufficiently high levels in tissues can produce sensitization. TMLA does not
react with amino acids to form haptens, and therefore does not share this mode of action for
sensitization. . In addition to the immunological differences, there may be some slight differences in
reactions at the site of contact in terms of the magnitude of response. With the exception of the
immunological and site of contact effects, all other endpoints are expected to be the same.
3.1. Toxicokinetics and Metabolism
Tissue concentration time-course data were collected for rats exposed to 0.95 mg/m
3
(TMA) for 45
minutes (IITRI, 1988a). Animals were sacrificed 3 hours, 1, 2, 4, 8, 16, and 32 days following
exposure. In general, the highest tissue concentrations were obtained at the first time point (Tmax<3
hours). A second Tmax of eight days was reported for lung lymph nodes in male rats, suggesting a
possible role in the gender differences observed for lung toxicity. Biological half-lives ranging from 3
to 46 days were estimated from the data (corresponding first order elimination constants of 0.015-
0.214 /hour). Specific half-lives for TMA in the lungs were estimated to be 21 days in male rats and
16 days in female rats. Similarly, in lung associated lymph nodes, half-lives of 13 and 33 days were
estimated for male and female rats, respectively. Because TMA is rapidly hydrolyzed to TMLA in the
body, these data also likely reflect the kinetics of TMLA. Although one might anticipate that the half-
lives of TMLA to be of shorter duration because unlike TMA, TMLA lacks the protein-reactive
anhydride moity.
3.2 Acute Toxicity
Data available from laboratory animals exposed to TMA indicate that its acute toxicity is relatively low,
regardless of the route of exposure.
Oral In rats, the acute oral LD
50
value derived for female animals (2,030 mg/kg) was slightly
lower than the value calculated for male animals (3,340 mg/kg) (IITRI, 1991a). For both sexes
combined, an oral LD
50
value of 2,730 mg/kg was derived. Upon necropsy of the animals that
died, a number of stomach lesions (e.g., wall thinning, ulcerations, hemorrhage, necrosis) were
observed.
Inhalation In rats, 3/10 animals died following a four-hour exposure to 2,330 mg/m
3
TMA,
indicating that the acute LC
50
value is likely to exceed this concentration (IITRI, 1992a). No
rats died following a four-hour exposure to concentrations as high as 3,750 mg/m
3
TMLA
(IITRI, 1988b). Gross necropsy of the animals from the TMA study revealed a number of
effects on the lung (e.g., red foci, mottled, fluid-filled). Gross necropsy of the animals from
the TMLA study revealed five rats with no gross lesions, three rats with lung foci, two with
red areas on the lung and one with a distended bladder. The findings for the TMLA study were
considered of a minor nature and within normal limits.
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
17
Altered breathing patterns (e.g., decreased time of inspiration and expiration, increased length of
apneic periods) were noted in mice exposed to 21.5-150 mg/m
3
for 30 minutes (Schaper and Brost,
1991).
Dermal In New Zealand albino rabbits, no deaths were observed following a dermal dose of
2,000 mg/kg (IITRI, 1991b). Dermal irritation was observed in all animals. However, no
treatment-related lesions were noted upon necropsy. In rats, a dermal LD
50
value of 5,600
mg/kg was reported (Rom, 1992).
Results of toxicities studies on TMA and TMLA suggest that the acute toxicity is relatively low
independent of the route of exposure and that most effects noted were consistent with an irritation
effect.
3.3 Irritation/Corrosiveness
3.3.1 Dermal
In rabbits, mild irritation (score=1.7/8.0) was observed following a 500 mg dermal dose of TMA
applied to a 240 cm
2
patch of pre-moistened skin for 4 hours (IITRI, 1991c). Signs of irritation were
generally resolved by the end of the observation period (14 days).
Dermal application of 500 mg TMLA produced irritation (0.7/8.0) in rabbits (IITRI, 1988d). The
irritation was greatest during the first 60 minutes and was generally reversible by 48-72 hours. No
signs of corrosivity were observed.
Results of studies suggest that both TMA and TMLA are slightly irritating to skin.
3.3.2 Eye
Similarly, ocular administration of TMA in rabbits produced signs of irritation reached a maximum
(Draize score of 110.0/110.0) at 24-hours following exposure (Hatoum and Johnson, 1991).
Rabbits receiving an ocular dose of 100 mg TMLA reached a maximum eye irritation score of
59.7/110.0 at 24 hours. Lackluster pitting and pannus formation were observed.
Results of studies evaluating the potential of TMA and TMLA to cause eye irritation suggest that both
materials should be considered severe eye irritants.
Results from skin and eye irritation studies imply that TMA may be slightly more irritating than
TMLA. One possible explanation for this apparent difference in the magnitude of response may be the
heat of hydrolysis of trimellitic anhydride. However, one should be cautious in deriving quantitative
conclusions based on skin and eye irritations studies as these types studies tend to give qualitative
rather than quantitative results.
3.4 Sensitization
Skin - Although dermal exposure to a 30% solution of TMA in dimethyl sulfoxide (induction)
and 5% TMA in acetone (challenge) (0.3 mL) produced dermal sensitization in guinea pigs
(IITRI, 1987), dermal sensitization was not elicited in guinea pigs treated with 300 mg TMA
powder (IITRI, 1993). In mice, dermal sensitization was elicited using 10-50% solutions of
TMA in acetone/olive oil (0.025-0.050 mL) (Dearman et al., 1992, 1996). In rats, dermal
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
18
sensitization was produced using 25-50% solutions of TMA in acetone/corn oil (0.15 mL) (Art
et al., 1998).
Dermal sensitization studies indicate that the presence of a solvent increases the dermal sensitization
potential of TMA, perhaps by increasing absorption. While under normal conditions of manufacture
and use TMA would be encounter as a powder and would not be used in a solvent it is prudent to
consider TMA a potential dermal sensitizer. Studies on the potential dermal sensitization of TMLA
were not available. However, TMLA is not likely a dermal sensitizer as it lacks the protein-reactive
anhydride moity and furthermore TMLA was negative in respiratory sensitization potential studies.
Respiratory - In a respiratory sensitization study rats were exposed to 50 ug/m
3
TMLA, six
hours per day for five days (ITRI 1989b). Following a three-week rest period, animals were
challenged with a single inhalation exposure to TMLA (50 ug/m
3
), TMA (50 ug/m
3
) or filtered
air. There were no statistically significant effects on lung weight, volume, foci or serum IgG
antibody levels. Results suggest that TMLA does not induce respiratory sensitization nor does
it have cross-reactivity with TMA.
3.5 Repeated Dose
Data regarding the toxicity of TMA following repeated exposures are summarized below.
Oral No adverse effects have been observed in rats following dietary exposures to 1,000-
10,000 ppm (50-500 mg/kg/day) TMA for 90 days (Hill Top, 1969; IBT, 1970). A dose-
dependent increase in leukocyte count (NOEL 50 mg/kg/day) was observed in rats from one
study (Hill Top, 1969), but was not observed in the second study (IBT, 1970). The eleveated
leukocyte count reported in the Hill Top study may have been due to increased incidence of
bronchitis, peribronchitis, and/or focal pneumonia reported in control and treated groups.
Although a slight increase in adrenal weight was noted in dogs following dietary exposure to
1,000-20,000 ppm (25-500 mg/kg/day) TMA for 13 weeks, the number of animals tested per
dose (two of each sex) was insufficient to assess the statistical significance of this increase.
No adverse effects (histopathology) were observed in any treated animals.
Inhalation No adverse effects were observed in rats exposed to 0.3 mg/m
3
for six hours/day,
five days/week for two weeks (IITRI, 1985). In rats exposed to 0.1 mg/m
3
TMA six
hours/day, lung injury was absent after two days of exposure, minimal after six days of
exposure, and marked after ten days of exposure (Zeiss et al., 1988). A dose-dependent
increase in antibody levels and lung foci was observed in rats exposed to TMA concentrations
of 0.010, 0.030, 0.10 or 0.30 mg/m
3
for six hours/day, five days/week for one or two weeks
(Zeiss et al., 1987; Leach et al., 1987). The lung foci completely resolved within 12 days after
the last exposure, but reappeared following exposure to a single challenge concentration.
Exposure to 0.5 mg/m
3
TMA produced hemorrhagic foci of the lung and increased antibody
levels in rats treated for six hours/day, five days/week for two weeks (IITRI, 1992). Estrogen
treatment reduced the number of lung foci in both male and female rats, while testosterone
treatment had no effect. A dose-dependent increase in lung lesions (hemorrhagic foci,
inflammatory cell infiltration, bronchoalveoloar pneumonia) and antibody levels was observed
in rats exposed to 0.002, 0.015, or 0.054 mg/m
3
for six hours/day, five days/week for up to 13
weeks (Leach et al., 1989). These effects were more pronounced in rats following 6.5 weeks
of exposure than observed in animals following 13 weeks of exposure, suggesting some degree
of adaptation. A NOEL was not identified. Mechanistic studies demonstrate that when the
immune system of rats is suppressed, TMA exposure does not produce lung lesions (Leach et.
al., 1989).
In mice, exposure to 0.010, 0.070, or 0.150 mg/m
3
for 30 minutes/day for five days produced
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
19
altered breathing patterns (decreased time of inspiration and expiration, increased length of
apneic periods) (Schaper and Brost, 1991). However, no histopathological changes were
evident in the lungs of treated animals. Results are consistent with a sensory irritation effect.
Data available from laboratory animals exposed to TMLA indicate that its subchronic toxicity is
also relatively low, regardless of the route of exposure.
Oral Gastrointestinal symptoms (e.g., diarrhea, watery cecal contents, cecal distortion) were
apparent in rats exposed to 1,000 mg/kg-day TMLA by oral gavage five days/week for four
weeks. However, no treatment-related effects were observed in rats exposed to TMLA doses
of 300 mg/kg-day or less (Hankinson and Sakal, 1991).
Inhalation No treatment-related effects were observed in rats exposed to concentrations of
0.05, 0.1 or 0.3 mg/m
3
TMLA for six hours/day, five days/week for 13 weeks (IITRI, 1989a).
In contrast to results obtained with TMA, exposure to TMLA failed to produce lung lesions or
increased antibody levels in treated animals.
Repeat dose studies on TMA and TMLA suggest that the subchronic toxicity is relatively low
regardless of the exposure route with the most notably effects being irritation at the site of application
(TMA and TMLA) and an immunological response (TMA only).
3.6 Genetic Toxicity (In Vitro)
In vitro studies of the potential genotoxicity of TMA have consistently reported negative results. In
Chinese hamster ovary cells, TMA concentrations as high as 2,000 mg/L failed to produce an increase
in either HGPRT mutations or chromosomal aberrations in the presence and absence of a metabolic
activation system (rat liver S9) (Bigger and Sigler, 1991; Putnam and Morris, 1991). Similarly,
negative results were obtained for a mutagenic response in several strains of Salmonella typhimurium
(TA98, TA100, TA1535, TA1537) in the presence and absence of a metabolic activation system (rat
liver S9) (San and Wagner, 1991). Because TMA is rapidly hydrolyzed to TMLA in aqueous
solutions, these data likely reflect the genotoxicity of TMLA as well. Tests results suggest that the
potential for genotoxicity is low for both TMA and TMLA.
3.7 Genetic Toxicity (In Vivo)
Although no in vivo genotoxicity studies were located for TMA or TMLA, the consistent negative
results observed for these chemicals from in vitro studies suggests that the potential for significant
genotoxicity is low.
3.8 Carcinogenicity
No data regarding the carcinogenicity of TMA or TMLA were located.
3.9 Reproductive Toxicity
Although a multigenerational reproductive toxicity test was not located for TMA or TMLA, data
available from other studies suggest that the potential for significant toxicity to reproduction from
exposures to these chemicals is low. For example, subchronic inhalation exposures of male and
female rats to TMA concentrations up to 0.054 mg/m
3
, or to TMLA concentrations up to 0.30 mg/m
3
did not result in any histopathological effects to reproductive tissues (IITRI, 1989a). Similarly, no
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
20
histopathological effects of reproductive tissues were observed in rats exposed to concentrations as
high as 10,000 ppm TMA in feed (approximately 500 mg/kg-day) for 90 days (IBT, 1970; Hill Top,
1969), or in dogs exposed to concentrations as high as 20,000 ppm TMA in feed (approximately 500
mg/kg-day) for 13 weeks (Hill Top, 1969). Additionally, reproductive performance was not affected
in female rats and guinea pigs following exposure to TMA concentrations of 0.5 mg/m
3
on days 6
through 15 of gestation (Ryan, 1988). Because TMA is likely hydrolyzed to form TMLA in tissues,
these studies also provide information about TMLA. Data suggest a low potential for adverse
reproductive effects.
3.10 Developmental Toxicity/Teratogenicity
Inhalation exposures to 0.5 mg/m
3
TMA for six hours/day on days 6-15 of gestation did not produce
any signs of fetotoxicity or teratogenicity in guinea pigs (Ryan, 1988). In similarly treated pregnant
rats, lung foci and increased antibody levels were observed (Ryan, 1988). Although no signs of
fetotoxicity or teratogenicity were observed in the offspring, increased antibody levels were noted in
neonatal rats. Following a challenge exposure, lung foci were only observed in the offspring whose
mothers had not completely recovered from the original TMA exposure. Lung foci were not observed
in adult offspring. With the exception of the effects on antibody levels and lung foci, these data likely
reflect the toxicity of TMLA because TMA is rapidly hydrolyzed to TMLA in the body. Results are
consistent with a low potential for developmental effects.
3.11 Human Experience
Eighteen workers were exposed to average concentrations of TMA ranging from 0.0006 to 2.1 mg/m
3
for three job categories (Bernstein et al., 1983). Five of the workers were found to have elevated
antibody levels to TMA, one worker had a late onset respiratory systemic syndrome (LRSS)
associated with TMA, and another worker had allergic rhinitis. LRSS can be characterized by a series
of immunological symptoms that are delayed, typically four to eight hours after exposure has ended
and may include, coughing, wheezing, breathlessness, congestion, fever, chills, fatigue and
generalized aching. Recovery is complete six to twelve hours after onset of symptoms.
An eleven-year study was conducted on 46 workers exposed to TMA. Seven workers had elevated
antibody levels, one of which also had rhinitis and another of which had TMA asthma/rhinitis.
Fourteen workers were found to have a positive antibody response to TMA-human serum albumen
(TMA-HSA). Industrial hygiene data from a single year (1989) revealed concentrations ranging from
<0.003 to 0.77 mg/m
3
(Grammer et al., 1991, 1992).
A group of 119 workers who had the potential for exposure to TMA for at least one year were
followed for a period of five years (Grammer et al., 1998). Of the 16 workers with elevated levels of
immunoglogin E (IgE), three had asthma, and an additional six developed asthma during the follow-
up period. Of the 44 workers with elevated levels of immunoglobin G (IgG), six had asthma, and an
additional two workers developed asthma during the follow-up period. Of the 102 workers without
elevated IgE levels, none had asthma, and only a single worker developed asthma during the follow-
up period. These data indicate that workers with elevated IgE or IgG levels are at increased risk of
developing asthmatic allergic sensitivity to TMA.
In a group of 474 workers exposed to mean concentrations ranging from <0.00053-00.17 mg/m
3
TMA, 6.8% had a TMA immunologic syndrome, 31.6% had an irritant response, and 61.6% were
asymptomatic (Zeiss et al., 1992). An exposure-response relationship was apparent with increased
antibody levels in this same group of workers when they were grouped into one of five exposure
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
21
groups (percent affected workers indicated in parentheses): 0.13 mg/m
3
(29%); 0.036 mg/m
3
(4%);
0.002 mg/m
3
(5%); 0.00051 mg/m
3
(0%); and <0.00053 mg/m
3
(0%) (Grammer et al., 1999).
In a study of nine workers exposed to TMA-containing paint powder, one worker exhibited obvious
illness and two worker had evidence of TMA-related pulmonary dysfunction and immunological
response (Letz et al., 1987). Monitoring data indicated that the workers had been exposed to
concentrations of TMA in air that were more than 100-times higher than the occupation exposure limit
of 0.04 mg/m
3
.
A group of 196 workers were exposed to TMA over a 12-year period (Zeiss et al., 1990). Seventeen
of the workers were found with IgE-mediated asthma/rhinitis, seven were found with LRSS, and four
were found with both conditions.
Although all of the human exposures summarized above likely included some exposure to TMLA, no
specific cases of human exposures to TMLA were located, and animal studies indicate that TMLA
lacks the potential to sensitize and elicit an immune reaction.
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
22
4.0 HAZARDS TO THE ENVIRONMENT
4.1 Acute Aquatic Toxicity
Because TMA rapidly hydrolyzes to form TMLA in water, following pH adjustment TMLA and
sodium trimellitate salts are the actual form of the chemical evaluated in aquatic toxicity tests.
Analytical methods used measured the concentration of TMLA and its salt. Although information
regarding the chronic toxicity of TMA/TMLA in aquatic species was not located, data regarding the
acute toxicity of TMA/TMLA in aquatic species are summarized below.
Fish No signs of toxicity were observed in Leuciscus idus melanotus (Golden Orfe) exposed
to TMA/TMLA nominal concentrations of 130, 220, 350, 600, or 1,000 mg/L for 96 hours
under static conditions (Knacker et al., 1993).Based on measured concentrations the 96 hour
NOEC > 896 mg/l.
Invertebrates - No signs of toxicity were observed in Daphnia magna (water flea) exposed to
TMA/TMLA nominalconcentrations of 130, 220, 350, 600, or 1,000 mg/L for 48 hours under
static conditions (Knacker et al., 1992). Based on measured concentrations, the 96h NOEC
was > 792 mg/l.
Plants - No signs of toxicity were observed in Scenedesmes subspicatus (green algae) exposed
to TMA/TMLA nominal concentrations of 62.5, 125, 250, 500, or 1,000 mg/L for 96 hours
under static conditions (Knacker et al., 1992). Based on measured concentrations, the NOEC
was > 739 mg/l.
Bacteria In activated sludge, bacterial respiration was inhibited by TMA/TMLA (Lebertz,
1991). An EC
50
value of 5.7 mg/L was calculated from the definitive portion of the
respiration inhibition study. Preliminary tests in bacterial inhibition study tested 1, 10, and 100
mg/L and found minimal effects. The definitive portion of the study tested 500 to 4000 mg/L
and found complete inhibition at all concentrations.
The calculated EC50 for respiration inhibition is not fully representative of the test results, that found
100 mg/L had minimal inhibition (approximately 6%) while 500 mg/L had nearly complete
inhibition. The test report concluded that the EC50 must be in the range of between 100 and 500
mg/L. Consequently, the biodegradation test, conducted at 10 and 20 mg/L, would not be likely to
reflect inhibitory effects of the test material.
Results suggest that TMA and TMLA have a low potential to cause significant acute aquatic toxicity.
4.2 Chronic Aquatic Toxicty
No data was available.
4.3 Terrestrial Effects
Although no data were located regarding the toxicity of TMA or TMLA in terrestrial mammals, the
low toxicity in laboratory animals suggests that their toxicity to terrestrial mammals in general would
also be low.
4.4 Other Environmental Effects
No additional data regarding other potential environmental effects of TMA or TMLA were located.
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
23
5.0 CONCLUSIONS AND RECOMMENDATIONS
TMA is currently of low priority for further work.
Approximately 100,000 metric tonnes/year TMA are currently produced world-wide, the majority of
which (65,000 metric tonnes/year) are produced in the U.S. Most of the TMA produced (65%) is used
in the synthesis of a plasticizer for PVC resins, while smaller amounts (30%) are used as a reactant in
wire and cable insulation enamels and polyester resins in powder coatings. All of the TMLA
produced in the world is used to produce TMA.
TMA exhibits low acute toxicity by the oral, dermal, and inhalation routes. The oral LD
50
has been
reported to range from 2,030 to 3,340 mg/kg in male and female rats (average = 2,730 mg/kg), with
stomach lesions appearing as the most consistent lesion upon necropsy. In rats, the inhalation LC
50
value was reported to exceed a concentration of 2,330 mg/m
3
, with lung lesions appearing as the most
consistent lesion upon necropsy. A dermal LD50 value of 5,600 mg/kg was reported. The inhalation
LC
50
for TMLA was reported to exceed 3,750 mg/m
3
. For other routes, TMLA is expected to have
similar or lower toxicity, based on the rapid hydrolysis of TMA to TMLA in tissues.
In repeated dose studies, the principle effects of TMA are on the immune system and the lung.
Elevated antibody levels and lung foci have been observed in rats following subchronic exposures to
relatively low concentrations (0.002 0.054 mg/m
3
). Elevated antibody levels, asthma, allergic
rhinitis, and LRSS are associated with occupational exposures to TMA in humans. However, humans
appear to be less sensitive to these effects than are rats. The toxicity of TMA following repeated oral
exposures appears to be low, with NOAELs of approximately 500 mg/kg-day identified for both rats
and dogs. Immunological and pulmonary effects are not associated with repeated exposures to
TMLA. Genotoxicity (In vitro) and developmental studies of TMA have generated negative results.
Negative results for the genotoxicity and developmental toxicity of TMLA are inferred from the rapid
hydrolysis of TMA to TMLA in tissues.
Based upon their chemical-physical properties, TMA and TMLA are not persistent in the environment
and are not expected to bioaccumulate in food webs. In the presence of water, TMA rapidly
hydrolyzes to form TMLA (complete hydrolysis in less than ten minutes). The half-life of TMA and
TMLA in air is estimated to be 13.4 and 6.6 days, respectively, due to direct reactions with
photochemically generated hydroxyl radicals. TMA and TMLA would be readily biodegraded under
aerobic conditions. Biodegradation results likely reflect biodegradation of TMLA because of the rapid
hydrolysis of TMA. Under environmentally relevant conditions TMLA is likely to be available as a
salt. Human exposures to TMA and TMLA are likely to be limited to occupation settings where TMA
is produced or used. Information regarding potential consumer or environmental exposure are
generally lacking, but are expected to be low. Fugacity-based fate and transport modeling efforts
suggest that TMA and TMLA are likely to partition to soil and water compartments in the
environment. Acute toxicity testing in fish, invertebrates, and algae indicate a very low order of
toxicity with no effect concentrations greater than 896, 792, and 739 mg/L respectively. The actual
test material was likely trimellitic acid and its sodium salt..
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
24
6.0 REFERENCES
Barker, R.D., van Tongeren, MJA, Harris, J.M., et.al. 1998. Risk factors for sensitization and
respiratory symptoms among workers exposed to acid anhydrides: a cohort study. Occupational and
Environmental Medicine Vol 55 (10) 684-691.
Bernstein, D.I., Roach, D.E., McGrath, K.G., et al. 1983. The relationship of airborne trimellitic
anhydride concentrations to trimellitic anhydride-induced symptoms and immune responses. J. Allergy
Clin. Immunol. 72 (6): 709-713.
Bigger, C.A. and Sigler, C.I. 1991. CHO/HGPRT mutation assay with confirmation. Final Report.
Microbiological Associates, Inc. Laboratory Study Number TA039.332001.
ChemSystems, 2000. Trimellitic Anhydride/Pyromellitic Dianhydride, July 2000.
Dearman, R.J., Hegarty, J.M., and Kimber, I. 1991. Inhalation exposure of mice to trimellitic anhydride
induces both IgG and IgE anti-hapten antibody. Int. Arch. Allergy Appl. Immunol. 95: 70-76.
Dearman, R.J., Mitchel, J.A., Basketter, D.A., and Kimber, I. 1992. Differential ability of occupational
chemical contact and respiratory allergens to cause immediate and delayed dermal hypersensitivity
reactions in mice. Int. Arch. Allergy Immmunol. 97: 315-321.
Grammer, L.C., Shaughnessy, M.A., Kenamore, B.D., and Yarnold, P.R. 1999 A clinical and
immunologic study to assess risk of TMA-induced lung disease as related to exposure. JOEM. Volume
41 (12): 1048-1051.
Grammer, L.C., Shaughnessy, M., and Kenamore, B. 1998. Utility of antibody in identifying
individuals who have or will develop anhydride-induced respiratory disease. CHEST. 114: 1199-1202.
Grammer, L.C., Harris, K.E., Sonenthal, K.R., et al. 1992. A cross-sectional survey of 46 employees
exposed to trimellitic anhydride. Allerg. Proc. 13(3): 139-142.
Grammer, L.C., Harris, K.E., Sonethal, K.R., et al. 1991. Identification and control of trimellitic
anhydride immunologic lung disease: an 11-year experience. Clinical Research. Vol. 39 (3): 703A.
Hatoum, N. and Johnson, W. 1991. Primary eye irritation study of trimellitic anhydride in rabbits.
IITRI Study No. 1693, Test Article No. 128H.
Hankinson, G.J. and Sakal, L. 1991. 1,2,4-benzenetricarboxylic acid (trimellitic acid) four-week oral
toxicity study in the rat. Final Report. HAEL NO. 86-0149G1.
Hill Top Research. 1969. Thirteen week dietary administration of trimellitic anhydride to rats.
Miamiville, OH. S-192
Hill Top Research. 1969. Dietary administration of trimellitic anhydride to dogs for 13 weeks.
Miamiville, OH. S-260
Horan, J.K. 1962. Aqueous hyrolysis of TMA memorandum. Amoco Chemicals Corporation,
ACM62-9/File 2570.9/2610190.
HSDB, 2002. Trimelltic Anhydride. Hazardous Substance Database. Agency for Toxic Substances and
Disease Registry, Public Health Service. Dept of Health and Human Services. Atlanta,GA.
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
25
IBT. 1970. Ninety-day subacute oral toxicity of LM 3813 (TMA) in albino rats. Northbrook, IL:
Industrial Bio-Test Laboratories. No. B7989
IITRI. 1993. Dermal sensitization study of trimellitic anhydride in guinea pigs using the modified
Buehler Method. Final Report. IITRI Project No. L08429, Study No. 1.
IITRI. 1992a. Acute inhalation toxicity study of trimellitic anhydride (TMA) in rats. Final Report.
IITRI Project No. L08100, Study No. 1749, Test Article No. 128L.
IITRI. 1992b. Inhalation study of sex-related differences associated with the respiratory sensitization
response of trimellitic anhydride (TMA): a research project. Final Report. IITRI Project No. L08100,
Study No. 1537, Test Article No. 128G.
IITRI. 1991a. Acute oral toxicity study of trimellitic anhydride in rats. Final Report. IITRI Project
NO. L08100, Study No. 1699, Test Article No. 128I.
IITRI. 1991b. Acute dermal toxicity study of trimellitic anhydride in rabbits. Final Report. IITRI
Project No. L08100, Study No. 1700, Test Article No. 128I.
IITRI. 1991c. Acute dermal irritancy/corrosivity study of trimellitic anhydride in rabbits. Final Report.
IITRI Project No. L08100, Study No. 1694, Test Article No. 128H.
IITRI. 1989a. Thirteen-week inhalation toxicity study of trimellitic acid (TMLA) in rats. Final Report.
IIT Project No. L8100, Study No. 1424, Test Article No. 228C.
IITRI. 1989b. Respiratory sensitization screen of trimellitic acid (TMLA) in rats. Final Report. IITRI
Project No. L8100, Study No. 1422, Test Article No. 228C.
IITRI. 1988a. Distribution and kinetic study of 14C-trimellitic anhydride in rats. Final Report. IITRI
Project No. L8100, Study No. 1021, Test Article No. 422.
IITRI. 1988b. Acute inhalation toxicity study of trimellitic acid in rats. Final Report. IITRI Project
No. L8100, Study No. 1423, Test Article No. 228C.
IITRI. 1988c. Abbreviated primary eye irritation study of trimellitic acid in rabbits. Study No. 1425,
Test Article No. 228C.
IITRI. 1988d. Abbreviated acute dermal irritancy/corrosivity study of trimellitic acid in rabbits. Study
No. 1426, Test Article No. 228C.
IITRI. 1987. Dermal sensitization study of trimellitic anhydride (TMA) in guinea pigs. Final Report.
IITRI Project No. L8100, Study No. 1196, Test Article No. 128.
IITRI. 1985. Two-week inhalation toxicity study of trimellitic acid in rats. Final Report. IITRI Project
No. L8100, Study No. 698, Test Article No. 228.
IPCS (Intenationational Programme on Chemical Safety), 1992. Health and Safety Guide No. 71
Trimellitic Anhydride World Health Organization for the International Programme on Chemical Safety
(a collaborative programme of the United Nations Environment Programme, the International Labour
Organisation, and the World Health Organization), Geneva Switzerland.
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
26
Knacker, T., Zietz, E., Schallnass, H., and Diehl, T. 1993. A study of the acute toxicity to fish
(Leuciscus idus melanotus) of trimellitic anhydride. Final Report. Battelle Europe Study Number: BE-
EA-128-91-01-F3A-1.
Knacker, T., Brodesser, J., SchallnaB, H., and Reifenberg, P. 1993. A study of the toxicity to algae
(Scenedesmus subspicatus) of trimellitic anhydride. Final Report. Battelle Europe Study Number: BE-
EA-128-91-02-ALG-1.
Knacker, T., Zietz, H. Schallnass, H., and Reifenber, P. 1992. A study of the acute immobilisation to
DAPHNIA of trimellitic anhydride. Final Report. Battele Europe Study Number: BE-EA-128-91-02-
DAK-1.
Leach, C.L., Hatoum, N.S., Zeiss, C.R., and Garvin, P.J. 1989. Immunologic tolerance in rats during 13
weeks of inhalation exposure to trimellitic anhydride. Fundamental and Applied Toxicology. 12: 519-
529.
Leach, C.L., Hatoum, N.S., Ratajczak, H.V., Zeiss, C.R., and Garvin, P.J. 1988. Evidence of
immunologic control of lung injury induced by trimellitic anhydride. Am. Rev. Respir. Dis. 137: 186-
191.
Leach, C.L., Hatoum, N.S., Ratajczak, H.V., Zeiss, C.R., Roger JC, and Garvin, P.J. 1987. The
pathologic and Immunologic Repsonse to Inhaled Trimellitic Anhydride in Rats. Toxicol Appl
Pharmacol. 87(1):67-80
Lebertz, H. 1991a. Study on the ready biodegradability (modified Sturm Test) of trimellitic
anhydride. Final Report. Battelle Europe Study No.: BE-EA-128-91-01-STT-01.
Lebertz, H. 1991b. Study on the toxicity of trimellitic anhydride towards bacteria. Final Report.
Battelle Europe Study-No.: BE-EA-128-91-01-BHT-01.
Letz, G., Wugofski, L., Cone, J.E., et al. 1987. Trimellitic anhydride exposure in a 55-gallon drum
manufacturing plant: clinical, immunologic, and industrial hygiene evaluation. American Journal of
Industrial Medicine. 12: 407-417.
Liu, R., Bargatze, R.F. and Katz, D.H. 1980. Induction of immunologic tolerance to the trimellitate
haptenic group in mice: model for a therapeutic approach to trimellitic anhydride-induced
hypersensitivity syndromes in humans? J. Allergy Clin. Immunol. 66 (4): 322-326.
Putnam, D.L. and Morris, M.J. 1991. Chromosome aberrations in chinese hamster ovary (CHO) cells.
Final Report. Microbiological Associates, Inc. Laboratory Study Number TA039.337100.
Ryan, B.M. 1988. Teratological evaluation of trimellitic anhydride (TMA) in rats and guinea pigs.
Submitted in partial fulfilment of the requirements for the degree of Master of Science in Biology in the
School of Advanced Studies of Illinois Institute of Technology, Chicago, Illinois.
Rom WN, 1992. Environmental and Occupational Mediceine. 2
nd
Edition. Little, Brown and Co.,
Boston, MA.
San, R.H. and Wagner, V.O. 1991. Salmonella/mammalian-microsome plate incorporation
mutagenicity assay (Ames Test) with a confirmatory assay. Microbiological Associates, Inc. Laboratory
Study Number TA039.501014.
OECD SIDS TRIMELLITIC ANHYDRIDE AND TRIMELLITIC ACID
UNEP PUBLICATIONS
27
Schaffhausen, J.G. 1991. Partition coefficient for TMA. Letter to W.H. Frye/MS E-1A Naperville.
Schaper M and Brost MA, 1991. Respiratory Effects of Trimellitic Anhydride Aerosols in Mice. Arch
Toxicol. 65(8): 671-677.
Schmitt, S. L. Technical dossier for TMA. CAS No. 552-30-7/PM/REF NO. 25550. Submitted by:
Amoco Corporation.
SRI, 2000. "Polyimides", Chemical Economics Handbook, November 2000
Trent University 1999. Fugacity-based Environmental Equilibrium Partitioning Model. Version 2.2
Environmental Modeling Center, Trent University, Peterborough, Ontario.
van Tongeren, M.J., Barker, R.D., Gardiner, K., et al. 1995. Exposure to acid anhydrides in three resin
and one cushioned flooring manufacturing plants. Ann. Occup. Hyg., Vol. 39. No. 5: 559-571.
Warbrick, E.V., Dearman, R.J., Basketter, D.A., Gerberick, G.F., Ryan, C.A., and Kimber, I. 1998.
Analysis of cytokine mRNA expression following repeated exposure of mice to chemical contact and
respiratory allergens. Journal of Applied Toxicology. 18: 205-213.
Zeiss, C.R., Mitchell, J.H., Van Peenen, P.F., et al. 1992. A clinical and immunologic study of
employees in a facility manufacturing trimellitic anhydride. Allergy Proc. Vol. 13 (4): 193-198.
Zeiss, C.R., Mitchell, J.H., Van Peenen, P.F., Harris, J., and Levitz, D. 1990. A twelve-year clinical
and immunologic evaluation of workers involved in the manufacture of trimellitic anhydride (TMA).
Allergy Proc. 11 (2): 71-77.
Zeiss CR, Leach CL, Smith LJ, Levitz D, Hatoum NS, Garvin PJ, Patterson R, 1988. A Serial
Immunologic and Histopathologic Study of Lung Injury Induced by Trimellitic Anhydride. Am Rev
Respir Dis. 137(1): 191-196.
Zeiss CR, Levitz D, Leach CL, Hatoum NS, Ratajczak HV, Chandler MJ, Roger JC, Garvin PJ 1987.
Model of Immunologic Lung Injury Induced by Trimellitic Anhydride Inhalation: Antibody Response. J
Allergy Clin Immunol. 79(1):59-63.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
28
SIDS DOSSIER: TRIMELLITIC ANHYDRIDE (TMA)
CAS No. 552-30-7
Sponsor Country: U.S.A.
DATE: January, 2002
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
29
1. GENERAL INFORMATION
1.01 SUBSTANCE INFORMATION
A. CAS-Number: 552-30-7
B. Name (IUPAC name) : Trimellitic anhydride
C. Name (OECD name): Trimellitic anhydride
D. CAS Descriptor
E. EINECS-Number: 209-008-0
F. Molecular Formula: C9H4O5
G. Structural Formula
H. Substance Group
I. Substance Remark
J. Molecular Weight: 192.12
1.02 OECD INFORMATION
A. Sponsor Country: U.S.A.
B. Lead Organisation:
Name of Lead Organisation: BP-Amoco Chemicals
Contact person: David Dutton
Address:
U.S.A.
Tel:
Fax:
1.1 GENERAL SUBSTANCE INFORMATION
A. Type of Substance element [ ]; inorganic [ ]; natural substance [ ]; organic [ X ];
organometallic [ ]; petroleum product [ ]
B. Physical State (at 20C and 1.013 hPa)
gaseous [ ]; liquid [ ]; solid [ X ]
C. Purity (indicate the percentage by weight/weight)
98% purity, typical
1.2 SYNONYMS: 1,2,4-benzenetricarboxylic acid, cyclic 1,2-anhydride
anhydrotrimellitic acid
trimellitic acid anhydride
1,2,4-benzenetricarboxylic acid anhydride
1,3-dioxo-5-phthalancarboxylic acid
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
30
4-carboxyphthalic anhydride
1.3 IMPURITIES : trimellitic acid (TMLA)
methyl di-basic acids
1.4 ADDITIVES
1.5 QUANTITY
65,000 metric tonnes/year produced in U.S.
30,000 metric tonnes/year outside U.S.
Reference: SRI, 2000; ChemSystems, 2000
50,000 tonnes per annum in 1990
Reference IPCS, 1992
>2.27x106 g/year in the 1970s
Reference: HSDB, 2001
1.6 LABELLING AND CLASSIFICATION
Labelling
Type:
Specific limits:
Symbols:
Note:
R-phrases:
S-phrases:
Text of S-phrases:
Remarks:
Classification
Type:
Category of danger:
R-phrases:
Remarks:
1.7 USE PATTERN
A. General
65%: Synthesis of plasticizer in PVC resins
30%: Wire and cable insulation enamels and polyester resins in powder coatings
5%: Other epoxy curing agent, textile sizing agent, rubber curing accelerator,
electrostatic toner binder, vinyl cross-link agent
Type of Use: Category: Wide dispersive
Industrial Vinyl chloride plasticizers,
Various polymers and polyesters,
Dyes and pigments,
Paints and coatings,
Pharmaceuticals,
Surface active agents,
Specialty chemicals,
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
31
Agricultural Agricultural chemicals.
Type of Use: Category: Non dispersive
Industrial Curing agent for epoxy and other resins
Numerous modifiers and intermediates
Remarks: A number of epoxy resin and surface coating systems, containing between 2-10%
TMA, are available as dry powder formulations and are intended for application
either by electrostatic dry powder spraying or by dipping pre-heated articles into
fluidized beds
Reference: OSHA, 1992
B. Uses in Consumer Products
Function Amount Present Physical State
Remarks: Non-stick coatings on utensils maximum of 10% solid
and equipment in household
and food industry
Remarks: Coatings (inside) of tin ~0.04% solid
Cans for foodstuff
Remarks: Epoxy-resin based 10% solid
Surface coatings
Reference: Amoco Corporation, 1991
1.8 OCCUPATIONAL EXPOSURE LIMIT VALUE
Exposure limit value
Value: TWA 0.04 mg/m
3
(0.005 ppm)
Remarks: Australia
Reference: IPCS, 1992
Value: TWA 0.04 mg/m
3
(0.005 ppm)
Remarks: Canada
Reference: IPCS, 1992
Value: TWA 0.04 mg/m
3
(0.005 ppm) dust; 0.005 mg/m3 (0.0006 ppm) fumes
Remarks: Czechoslovakia (MAK)
Reference: IPCS, 1992
Value: TWA 0.04 mg/m
3
(0.005 ppm)
Remarks: Germany (MAK)
Reference: WHO, 1992
Value: TWA 0.04 mg/m
3
(0.005 ppm)
Remarks: Netherlands (MAK)
Reference: WHO, 1992
Value: TWA 0.04 mg/m
3
(0.005 ppm)
Remarks: United Kingdom (OES)
Reference: WHO, 1992
Value: TWA 0.04 mg/m
3
(0.005 ppm)
Remarks: U.S. (NIOSH REL, ACGIH TLV)
Reference: WHO, 1992
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
32
Short term exposure limit value
Value: 0.08 mg/m
3
(0.01 ppm)
Remarks: STEL - Germany
Reference: WHO, 1992
1.9 SOURCES OF EXPOSURE
(a)
Media of release: Airborne dust and fumes
Source:
Remarks: In 1978, NIOSH estimated that approximately 20,000 U.S. workers were at
risk of exposure to trimellitic anhydride in its various applications. The
NIOSH 1972 National Occupational Hazard Survey found 475 workers out of
3515 employed in nonmetallic mineral products and engine electrical
equipment industries were exposed to TMA. The NIOSH 1982 survey found
97 workers of the total payroll of 269 in the printing ink industry were
exposed to TMA.
Reference: OSHA, 1992
1.10 ADDITIONAL REMARKS
A. Options for disposal
Remarks:
Reference:
B. Other remarks
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
33
2. PHYSICAL-CHEMICAL DATA
2.1 MELTING POINT
(a)
Value: 165C (330F)
Decomposition: Yes [ ] No [ ] Ambiguous [ ]
Sublimation: Yes [ ] No [ ] Ambiguous [ ]
Method: Other
GLP: Yes [ ] No [ ] ? [x]
Remarks:
Reference: Amoco Corporation, 1997
(b)
Value: 161-168
o
C
Decomposition: Yes [ ] No [ ] Ambiguous [ ]
Sublimation: Yes [ ] No [ ] Ambiguous [ ]
Method:
GLP: Yes [ ] No [ ] ? [x]
Remarks:
Reference: Amoco Corporation, 1991
2.2 BOILING POINT
(a)
Value: 390
o
C
Pressure:
Decomposition: Yes [ ] No [ ] Ambiguous [ ]
Method:
GLP: Yes [ ] No [ ] ? [x]
Remarks:
Reference: Amoco Corporation, 1991
(b)
Value: 240-245
o
C
Pressure: 19 x 10
2
Pa
Decomposition: Yes [ ] No [ ] Ambiguous [ ]
Method:
GLP: Yes [ ] No [ ] ? [x]
Remarks:
Reference: Amoco Corporation, 1991
2.3 DENSITY
(a)
Type: Bulk density [ ]; Density [X]; Relative Density [ ]
Value: 1.54
Temperature: 20
o
C
Method:
GLP: Yes [ ] No [ ] ? [X]
Remarks: specific density
Reference: WHO, 1992
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
34
2.4 VAPOUR PRESSURE
(a)
Value: 7.6 x 10
-5
Pa, (5.69x 10
-7
mmHg)
Temperature: 25
o
C
Method: calculated [ X]; measured [ ] Year:
GLP: Yes [ ] No [ ] ? [x]
Remarks:
Reference: US EPA, EPWIN Suite, 1997
(http://www.epa.gov/oppt/exposure/docs/episuitedi.htm).
(b)
Value: 9.86 x 10
-6
mm Hg
Temperature: 25
o
C
Method: calculated [X ]; measured [ ] Year:
GLP: Yes [ ] No [ ] ? [x]
Remarks:
Reference: Daubert, T. E., and R. P. Danner (1989)
(c)
Value: <1.1 x 10
-7
mm Hg
Temperature: 25
o
C
Method: calculated [X ]; measured [ ] Year:
GLP: Yes [ ] No [ ] ? [x]
Remarks:
Reference: Amoco Corporation, 1997
(d)
Value: 5.69 x 10
-7
mm Hg
Temperature: 25
o
C
Method: calculated [X ]; measured [ ] Year:
GLP: Yes [ ] No [ ] ? [x]
Remarks:
Reference: MPBPWIN v1.40 in EPIWIN Suite
2.5 PARTITION COEFFICIENT log
10
P
ow
(a)
Log Pow: 0.95 trimellitic acid (TMLA)
Temperature: room temperature
Method: calculated [ X ]; measured [ ]
GLP: Yes [ ] No [ ] ? [ ]
Remarks: TMA would have only transitory existence in an octanol/water mixture.
Hydrolysis of TMA in aqueous alcohol is extremely rapid at room
temperature. Consequently, TMLA would be formed upon dissolving TMA
in this solvent system. Furthermore, small amounts of the diacid-octyl ester
may form when octyl alcohol reacts with the anhydride moiety of TMA,
though the hydrolysis reaction is more prevalent.
Test Substance TMA (hydrolysis to TMLA)
Reference: KOWWIN Version 1.66.
(b)
Log Pow: 1.95 (assumes no hydrolysis)
Temper ature: 25 C
Method: calculated [X], measured [ ]
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
35
GLP: Yes [], No [], ?[X]
Remarks: If log Pow is estimated without considering hydrolysis of TMA to TMLA.
Slightly larger estimates are obtained. However, the most environmentally
relevant value must reflect the hydrolysis of the anhydride to the acid.
Test Substance: Trimellitic anhydride
Reference: KOWWIN (EPIWIN Suite)
(c)
Log Pow: 1.61 (assumes no hydrolysis)
Temperature: 25 C
Method: calculated [X], measured [ ]
GLP: Yes [ ], No [ ], ?[X]
Remarks: If log Pow is estimated without considering hydrolysis of TMA to TMLA.
Slightly larger estimates are obtained. However, the most environmentally
relevant value must reflect the hydrolysis of the anhydride to the acid.
Test Substance: Trimellitic anhydride
Reference: CLOGP Program
(d)
Log Pow: 1.61 (assumes no hydrolysis)
Temperature: 25 C
Method: calculated [X], measured [ ]
GLP: Yes [ ], No [X], ?[ ]
Remarks: If log Pow is estimated without considering hydrolysis of TMA to TMLA.
Slightly larger estimates are obtained. However, the most environmentally
relevant value must reflect the hydrolysis of the anhydride to the acid.
Test Substance: Trimellitic anhydride
Reference: Interactive Analysis Program
(e)
Log Pow: 0.80 (assumes no hydrolysis)
Temperature: 25 C
Method: calculated [X], measured [ ]
GLP: Yes [ ], No [X], ?[ ]
Remarks: If log Pow is estimated without considering hydrolysis of TMA to TMLA.
Slightly larger estimates are obtained. However, the most environmentally
relevant value must reflect the hydrolysis of the anhydride to the acid.
Test Substance: Trimellitic anhydride
Reference: ALOG Program
(f)
Log Pow: 1.14 (assumes no hydrolysis)
Temperature: 25 C
Method: calculated [X], measured [ ]
GLP: Yes [ ], No [X], ?[ ]
Remarks: If log Pow is estimated without considering hydrolysis of TMA to TMLA.
Slightly larger estimates are obtained. However, the most environmentally
relevant value must reflect the hydrolysis of the anhydride to the acid.
Test Substance: Trimellitic anhydride
Reference: XLOGP Program
2.6 WATER SOLUBILITY
A. Solubility
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
36
(a)
Value: 1,036 mg/L (assumes no hydrolysis)
Temperature:
Description: Miscible[ ]; Of very high solubility [ ];
Of high solubility [ ]; Soluble [X ]; Slightly soluble [ ];
Of low solubility [ ]; Of very low solubility [ ]; Not soluble [ ]
Method: Other
GLP: Yes [ ] No [ ] ? [ ]
Remarks: If water solubility estimated without considering hydrolysis to the acid, then
lower values will be calculated.
Reference: SRC, 2001
(b)
Value: 21,000 mg/L (after hydrolysis to trimellitic acid)
Temperature 25 C
Description: Miscible [ ], Of very high solubility [ ], Of high solubility [ ], Soluble [X ],
Of low solubility [ ], Of very low solubility [ ], Not soluble [ ]
Method:
GLP: Yes [ ], No [ ], ? [ ]
Remarks: Upon contact with water, trimellitic anhydride rapidly hydrolyzes to
trimellitic acid.
Reference SRC, 2001
(c)
Value: 1,211 mg/L (assumes no hydrolysis)
Temperature:
Description: Miscible[ ]; Of very high solubility [ ];
Of high solubility [ ]; Soluble [X ]; Slightly soluble [ ];
Of low solubility [ ]; Of very low solubility [ ]; Not soluble [ ]
Method: Other
GLP: Yes [ ] No [ ] ? [ ]
Remarks: If water solubility estimated without considering hydrolysis to the acid, then
lower values will be calculated.
Reference: WSKOW v1.40 in EPWIN Suite
(d)
Value: 860 mg/L (assumes no hydrolysis)
Temperature:
Description: Miscible[ ]; Of very high solubility [ ];
Of high solubility [ ]; Soluble [X ]; Slightly soluble [ ];
Of low solubility [ ]; Of very low solubility [ ]; Not soluble [ ]
Method: Other
GLP: Yes [ ] No [ ] ? [ ]
Remarks: If water solubility estimated without considering hydrolysis to the acid, then
lower values will be calculated.
Reference: Interactive Analysis Program
(e)
Value: 2777 mg/L (assumes no hydrolysis)
Temperature:
Description: Miscible[ ]; Of very high solubility [ ];
Of high solubility [ ]; Soluble [X ]; Slightly soluble [ ];
Of low solubility [ ]; Of very low solubility [ ]; Not soluble [ ]
Method: Other
GLP: Yes [ ] No [ ] ? [ ]
Remarks: If water solubility estimated without considering hydrolysis to the acid, then
lower values will be calculated.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
37
Reference: ALOGS Program
2.7 FLASH POINT (liquids)
(a)
Value: 227
o
C
Type of test: Closed cup [ ]; Open cup [x]; Other [ ]
Method: Other
GLP: Yes [ ] No [ ] ? [x]
Remarks:
Reference: Amoco Corporation, 1991
2.8 AUTO FLAMMABILITY (solid/gases)
No data available
2.9 FLAMMABILITY
(a)
Results: Extremely flammable [ ]; Extremely flammable - liquefied gas [ ];
Highly Flammable [ ]; Flammable [ ]; Non flammable [ ];
Spontaneously flammable in air [ ]; Contact with water liberates highly
flammable gases [ ]; Other [ ]
Method:
GLP: Yes [ ] No [ ] ? [ ]
Remarks: The National Fire Protection Association has not assigned a flammability
rating to trimellitic anhydride. Other sources rate trimellitic anhydride as
combustible when this substance is exposed to heat or open flame.
Reference: OSHA, 1996
2.10 EXPLOSIVE PROPERTIES
No data available
2.11 OXIDIZING PROPERTIES
No data available
2.12 ADDITIONAL REMARKS
Remarks: No additional remarks
2.13 ADDITIONAL DATA
No additional data
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
38
3. ENVIRONMENTAL FATE AND PATHWAYS
3.1 STABILITY
3.1.1 PHOTODEGRADATION
(a)
Type: Air [ X]; Water [ ]; Soil [ ]; Other [ ]
Light source: Sun light [ ]; Xenon lamp [ ]; Other [ ]
Light spectrum:
Relative intensity:
Concentration of Substance:
Temperature:
Direct photolysis:
Half life: 13.4 days
Degradation:
Quantum yield:
Method: calculated [X]; measured [ ]
Other
GLP: Yes [ ] No [ X] ? [ ]
Test substance: Trimellitic anhydride
Remarks: Reaction rate with photo-chemically produced hydroxyl radicals estimated
(0.797x10
-12
cm
3
/mol-s)
Result: Degrades on exposure to light or heat to high molecular weight addition
products.
Reference: AOPWIN (SRC, 2001); Horan, 1962; Amoco Corporation, 1991
3.1.2 STABILITY IN WATER
(a)
Type: Aqueous hydrolysis
Half life:
Degradation: hydrolyzed within 10 minutes by stirring in water at 27-32C (80 to
90
o
F).
GLP: Yes [ ] No [X ] ? [ ]
Test substance: Trimellitic anhydride
Remarks: Degrades on exposure to light or heat to high molecular weight addition
products.
Reference: Horan, 1962; Amoco Corporation, 1991
3.1.3 STABILITY IN SOIL
No data available
3.2 MONITORING DATA (ENVIRONMENT)
No data available
3.3 TRANSPORT AND DISTRIBUTION BETWEEN ENVIRONMENTAL COMPARTMENTS
INCLUDING ESTIMATED ENVIRONMENTAL CONCENTRATIONS AND
DISTRIBUTION PATHWAYS
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
39
3.3.1 TRANSPORT BETWEEN ENVIRONMENTAL COMPARTMENTS
No data available
3.3.2 THEORETICAL DISTRIBUTION (FUGACITY CALCULATION)
(a)
Media: Air-biota [ ]; Air-biota-sediment-soil-water [ X]; Soil-biota [ ];
Water-air [ ]; Water-biota [ ]; Water-soil [ ]; Other [ ]
Method: Fugacity level I [X]; Fugacity level II [X]; Fugacity level III [X];
Fugacity level IV [ ]; Other (calculation) [ ]; Other (measurement)[ ]
Results: Assumes no hydrolysis to the acid.
Level I Level II Level III
Air 2.25E-4% 2.25E-6% 7.9E-4%
Water 92.5% 92.5% 36.5%
Soil 7.3% 7.3% 63.5%
Sediment 0.16% 0.16% 0.028%
Remarks: Estimates are based on the assumption that no hydrolysis occurs. Default
release estimates assumed
Reference: Trent University, 1999
(b)
Media: Air-biota [ ]; Air-biota-sediment-soil-water [ X]; Soil-biota [ ];
Water-air [ ]; Water-biota [ ]; Water-soil [ ]; Other [ ]
Method: Fugacity level I [X]; Fugacity level II [X]; Fugacity level III [X];
Fugacity level IV [ ]; Other (calculation) [ ]; Other (measurement)[ ]
Results: Estimated distribution and media concentrations reflecting trimellitic
anhydride hydrolysis to trimellitic acid.
Level I Level II Level III
Air 7.68E-7% 7.68E-7% 3.46E-6%
Water 99.2% 99.2% 50.6%
Soil 0.78% 0.78% 49.3%
Sediment 0.02% 0.02% 0.026%
Remarks: Trimellitic anhydride hydrolyses in water and under humid conditions to
trimelltic acid (TMLA). Therefore, models using the physical chemical
parameter for TMLA are thought to be more environmentally relevant.
Reference: Trent University, 1999
3.4 IDENTIFICATION OF MAIN MODE OF DEGRADABILITY IN ACTUAL USE
Results:
Remarks:
Reference:
3.5 BIODEGRADATION
(a)
Type: aerobic [ X ]; anaerobic [ ]
Inoculum: adapted [ ]; non-adapted [ ]; ? [ ]; sewage [ X ]
Concentration: 10.19 mg/l related to COD [ ]; DOC [ X ]; Test substance [ ];
Medium: water [ ]; water-sediment [ ]; soil [ ]; sewage treatment [ X ]
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
40
Degradation: >60% within 7 days
Results: Readily biodeg. [ X ]; Inherently biodeg. [ ]; under test condition no
biodegradation observed [ ], Other [ ]
Method: OECD Guideline 301 B, Modified Sturm-Test
GLP: Yes [ X ] No [ ] ? [ ]
Test substance: Trimellitic anhydride
Remarks: Sewage microorganisms from a sewage plant working with predominantly
domestic sewage used as the inoculum.
Reference: Lebertz, 1991a
(b)
Type: aerobic [ X ]; anaerobic [ ]
Inoculum: adapted [ ]; non-adapted [ ]; ? [ ]; sewage [X]
Concentration: 100 ppm related to COD [ ]; DOC [ ]; Test substance [X];
Medium: water [ ]; water-sediment [ ]; soil [ ]; sewage treatment [X]
Degradation: 89-101% over 4 weeks
Results: Readily biodeg. [ X ]; Inherently biodeg. [ ]; under test condition no
biodegradation observed [ ], Other [ ]
Method:
GLP: Yes [ ] No [ ] ? [X]
Test substance: Trimellitic anhydride
Remarks: Since TMA rapidly hydrolyzes, this study assesses biodegradation of TMLA.
This study was not considered key because full report was not available and
specific protocol was not stated, though results were consistent with the key
study (lebertz 1991a)
Reference: Letz et al., 1987
3.6 BOD
5
,COD OR RATIO BOD
5
/COD
No data available
3.7 BIOACCUMULATION
Method: Calculated
Type of test: Bioconcentration Factor
GLP: Yes [], No [X], ? []
Test substance: Trimellitic anhydride
BCF: 3.2
Remarks:
Reference: BCFWIN v2.14
3.8 ADDITIONAL REMARKS
No additional remarks
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
41
4. ECOTOXICOLOGICAL DATA
4.1 ACUTE/PROLONGED TOXICITY TO FISH
(a)
Type of test: static [x]; semi-static [ ]; flow-through [ ]; other [ ]; open-system [ ];
closed-system [ ]
Species: Leuciscus idus melanotus (Golden orfe)
Exposure period: 96 hr.
Results: LC
0
(96 hr): > 1000 mg/L
LC
50
(96 hr): could not be determined.
NOEC (96 hr): = 1000 mg/L based on nominal concentrations
NOEC (96 hr): >896 mg/L based on the measured average concentration of
the highest concentration level tested.
Analytical monitoring: Yes [x] No [ ] ? [ ]
Method: OECD Guideline for Testing of Chemicals No. 203 Fish, Acute Toxicity
Test, adopted April 4, 1984 and the German Water Endangerment
Classification Scheme, DIN 38 412, Part 15 adopted June 1982.
GLP: Yes [x] No [ ] ? [ ]
Test substance: TMA hydrolyzes to TMLA in water. Test solutions were neutralized using
sodium hydroxide. Therefore, the test material was trimellitic acid and its
sodium salt.
Remarks: The highest concentration causing no mortality within the period of the
range-finding test was 1000 mg/L. The lowest concentration causing 100%
mortality within the period of the range-finding test was >1000 mg/L
Reference: Knacker et al., 1993.
4.2 ACUTE TOXICITY TO AQUATIC INVERTEBRATES
A. Daphnia
(a)
Type of test: static [x]; semi-static [ ]; flow-through [ ]; other [ ]; open-system [ ];
closed-system [ ]
Species: Daphnia magna (Straus)
Exposure period: 48 hr.
Results: EC0: >1000 mg/L
EC50: could not be determined.
EC0: >792 mg/L (based on the measured average concentration of t he
highest concentration level tested).
Analytical monitoring: Yes [ ] No [ ] ? [ ]
Method: OECD Guideline No. 202, Part I Daphnia sp., Acute Immobilisation Test
and Reproduction Test adopted April 4, 1984.
GLP: Yes [x] No [ ] ? [ ]
Test substance: TMA hydrolyzes to TMLA in water. Test solutions were neutralized using
sodium hydroxide. Therefore, the test material was trimellitic acid and its
sodium salt.
Remarks: Highest concentration causing no immobilization within the period of the
range-finding test: 100 mg/L. The lowest test concentration causing 100%
immobilization within the period of the range-finding test: > 100 mg/L.
Reference: Knacker, et al., 1992.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
42
4.3 TOXICITY TO AQUATIC PLANTS e.g. Algae
(a)
Species: Scenedesmus subspicatus (green algae)
End-point: Biomass [ ]; Growth rate [x]; Other [ ]
Exposure period: 96 hr.
Results: NOEC = 1000 mg/L based on nominal concentrations;
NOEC = 739 mg/L based on the measured average concentration of the
highest concentration level tested.
Analytical monitoring: Yes [x] No [ ] ? [ ]
Method: OECD Guideline 201, 1984.
GLP: Yes [x] No [ ] ? [ ]
Test substance: It is thought that trimellitic anhydride was hydrolysed under test conditions.
As a result it is believed that under test conditions and after pH adjustments
to the required physiological value trimellitic acid and trimellitic sodium salt,
respectively, were the test materials investigated in this study.
Remarks: TMA hydrolyzes to TMLA in water. Test solutions were neutralized using
sodium hydroxide. Therefore, the test material was trimellitic acid and its
sodium salt.
Reference: Knacker et al., 1993
4.4 TOXICITY TO BACTERIA
(a)
Type: Aquatic [ ]; Field [ ]; Soil [ ]; Other [x]
Species: activated sludge
Exposure Period: 3 hr.
Results: The range-finding study tested 1, 10, 100 mg/L and found no or minimal
inhibition (6% at 100 mg/L). The definitive portion of the study tested 500 to
4000 mg/L and found complete inhibition at all concentrations tested. The
following EC values were extrapolated from data derived from the definitive
portion of the study only: EC
5
: 0.095 mg/L
EC
25
: 1.1 mg/L
EC
50
: 5.7 mg/L
EC
75
: 30.4 mg/L
EC95: 340 mg/L
However, data obtained from the two studies combined suggest that the
actual EC50 falls in the range between 100 and 500 mg/L.
Analytical monitoring: Yes [ ] No [ ] ? [ X ]
Method: OECD-Test Guideline 209 Activated Sludge, Respiration Inhibition Test
GLP: Yes [ X ] No [ ] ? [ ]
Test substance: Trimellitic anhydride
Test Condition: Activated sludge was added to the test solution and was aerated with
compressed air for 3 hr. After the contact time, the solutions were poured
into an oxygen-bottle and oxygen consumption was recorded for 10 minutes
to determine respiration rates. TMA was likely hydrolyzed to TMLA under
the conditions of this assay.
Reference: Lebertz, 1991b
4.5 CHRONIC TOXICITY TO AQUATIC ORGANISMS
No data available, methods to extrapolate acute toxicity data to chronic exposures are readily
available.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
43
4.6 TOXICITY TO TERRESTRIAL ORGANISMS
No data available
4.7 BIOLOGICAL EFFECTS MONITORING (INCLUDING BIOMAGNIFICATION)
No data available
4.8 BIOTRANSFORMATION AND KINETICS
No data available
4.9 ADDITIONAL REMARKS
No additional remarks
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
44
5. TOXICITY
5.1 ACUTE TOXICITY
5.1.1 ACUTE ORAL TOXICITY
(a)
Type: LD
0
[ ]; LD
100
[ ]; LD
50
[ X ]; LDL
0
[ ]; Other [ ]
Species/strain: Rat/Sprague-Dawley
Value: 2,730 mg/kg
Method:
GLP: Yes [ X] No [ ] ? [ ]
Test substance: Trimellitic anhydride administered 50% (w/v) suspension in corn oil
Remarks: Groups of ten male and ten female rats were administered 0, 2000, 3500, or
5000 mg/kg TMA via gavage. Animals were observed for 14 days
following exposure. A 95% confidence limit of 1,730-4,290 mg/kg was
reported for both sexes combined, with slightly lower values reported for
females (2,030 mg/kg: CL=700-5,890 mg/kg) than for males (3,340 mg/kg:
CL=1,740-6,410 mg/kg). Deaths generally occurred within 1-48 hours after
exposure. Stomach lesions (thinning, ulcerations, hemorrhage, necrosis)
were noted.
Reference: IITRI, 1991a
5.1.2 ACUTE INHALATION TOXICITY
(a)
Type: LC
0
[ ]; LC
100
[ ]; LC
50
[ ]; LCL
0
[X ]; Other [ ]
Species/strain: Rat/Sprague-Dawley
Exposure time: 4 hours
Value: 2,330 mg/m
3
Method: Particulate
GLP: Yes [X] No [ ] ? [ ]
Test substance: Trimellitic anhydride - particle size = 4.4 microns (SD=2.3 microns)
Remarks: Ten rats (five males; five females) were exposed to TMA for 4 hours. Three
rats (two males; one female) died during the study. The acute inhalation
LC50 value was therefore concluded to exceed 2,330 mg/m
3
. During
exposure rats exhibited labored breathing. Body weights were increased
during the study. Gross necropsy revealed ef fects on the lung (red foci,
mottled, fluid filled).
Reference: IITRI, 1992a
(b)
Type: LC0 [ ]; LC100 [ ]; LC50 [ ]; LCL0 [ ]; Other [ X ]
Species/strain: Mouse/Swiss-Webster
Exposure time: 30 min
Value: LOEL = 21.5 mg/m
3
Method: Aerosol dissolved in acetone
GLP: Yes [] No [ ] ? [X ]
Test substance: Trimellitic anhydride - 21.5, 72, 150 mg/m
3
, 75% particles < 0.65 um in
diameter
Remarks: Alterations in breathing patterns (decreased time of inspiration and
expiration, increased length of apneic periods). No histopathological
changes were evident. Authors concluded that respiration effects may be
attributable to stimulation of vagal nerve endings in deep lung.
Reference: Schaper and Brost, 1991
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
45
5.1.3 ACUTE DERMAL TOXICITY
(a)
Type: LD0 [ X ]; LD100 [ ]; LD 50 [ ]; LDL0 [ ]; Other [ ]
Species/strain: Rabbit/New Zealand albino
Value: 2000 mg/kg
Method: Single dose applied to 240 cm
2
patch
GLP: Yes [X] No [ ] ? [ ]
Test substance: Undiluted trimellitic anhydride
Remarks: Five male and five female rabbits received a single dermal dose of 2,000
mg/kg, applied for 24 hours. Animals were observed for 14 days following
exposure. No deaths were observed. The authors concluded that the acute
dermal LD
50
value for TMA exceeds 2,000 mg/kg. Dermal irritation
(erythema, edema) was observed in all animals immediately following the
exposure, however, all animals recovered during the observation period.
Body weights were slightly increased in females but unchanged in males.
No treatment-related lesions were noted upon necropsy.
Reference: IITRI, 1991b
(b)
Type: LD0 [ ]; LD100 [ ]; LD 50 [ ]; LDL0 [ ]; Other [ X ]
Species/strain: Mice/female BALB/c
Value:
Method:
GLP: Yes [ ] No [ ] ? [ X ]
Test substance: Undiluted trimellitic anhydride.
Remarks: The ability of TMA and dinitrochlorobenzene to elicit immediate and
delayed cutaneous hypersensitivity reactions in mice were compared.
Topical exposure to both chemicals resulted in delayed hypersensitivity.
Only TMA induced an immediate (1 hr) dermal reaction following local
challenge. The study demonstrated that different classes of occupational
chemical allergen exhibit a variable potential to elicit immediate and delayed
dermal hypersensitivity reactions in mice, and provide a novel approach to
the classification and characterization of human allergens.
Reference: Dearman et al., 1992.
(c)
Type: LD0 [ ]; LD100 [ ]; LD 50 [ X ]; LDL0 [ ]; Other [ ]
Species/strain: Rat
Value: 5,600 mg/kg
Method:
GLP: Yes [ ] No [ ] ? [ X ]
Test substance: TMA
Remarks: Study demonstrates a dermal LD50 of 5,600 mg/kg
Reference: Rom, 1992.
5.1.4 ACUTE TOXICITY, OTHER ROUTES OF ADMINISTRATION
No data available
5.2 CORROSIVENESS/IRRITATION
5.2.1 SKIN IRRITATION/CORROSION
(a)
Species/strain: Rabbit/New Zealand White
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
46
Results: Highly corrosive [ ]; Corrosive [ ]; Highly irritating [ ];
Irritating [ ]; Moderate irritating [ ]; Slightly irritating [ X ];
Not irritating [ ]
Classification: Highly corrosive (causes severe burns) [ ];
Corrosive (caused burns) [ ]; Irritating [ X ]; Not irritating [ ]
Method: 4-hours application of 0.5 g to a 240 cm
2
moistened skin patch
GLP: Yes [X ] No [ ] ? [ ]
Test substance: Undiluted trimellitic anhydride
Remarks: Three male and three female rabbits were administered a single dermal TMA
dose of 0.5 g to a 240 cm
2
patch of pre-moistened skin for four hours
(excess chemical removed with light mineral oil). Animals were monitored
for 14 days following exposure. A primary dermal irritation score of 1.7
(maximum of 8) was reported, however, effects generally reversed by the
end of the observation period.
Reference: IITRI, 1991c
5.2.2 EYE IRRITATION/CORROSION
(a)
Species/strain: Rabbit
Results: Highly corrosive [ ]; Corrosive [ ]; Highly irritating [x];
Irritating [ ]; Moderate irritating [ ]; Slightly irritating [ ];
Not irritating [ ]
Classification: Irritating [x]; Not irritating [ ]; Risk of serious damage to eyes [ ]
Method: Other
GLP: Yes [ ] No [ ] ? [x]
Test substance: Trimellitic anhydride
Remarks: Signs of ocular irritation were maximum (i.e., Draize score = 110.0./110.0) at
the 24-hour examination.
Reference: Hatoum and Johnson, 1991.
5.3 SENSITISATION
(a)
Type: Dermal Sensitization
Species/strain: Guinea Pig/Hartley
Results: Sensitizing [ X ]; Not sensitizing [ ]; ambiguous [ ]
Classification: Sensitizing [ X ]; Not sensitizing [ ]
Method:
GLP: Yes [ ] No [ ] ? [x]
Test substance: Trimellitic anhydride - 30% solution in dimethyl sulfoxide for induction; 5%
solution in acetone for challenge.
Remarks: Ten adult male Hartley guinea pigs were administered 0.3 mL of the
induction solution, once a week for three weeks. Two weeks after the last
induction dose, the induced animals and ten control animals were
administered 0.3 mL of the challenge solution once a week for two weeks.
Positive erythema reactions were observed in 7/10 animals after the first
challenge compared to 0/10 in controls; however, a majority of both treated
and controls exhibited positive erythema reactions after the second
challenge.
Reference: IITRI, 1987
(b)
Type: Dermal Sensitization
Species/strain: Guinea Pig/Hartley
Results: Sensitizing [ ]; Not sensitizing [ X ]; ambiguous [ ]
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
47
Classification: Sensitizing [ ]; Not sensitizing [ X ]
Method:
GLP: Yes [ X ] No [ ] ? []
Test substance: Undiluted trimellitic anhydride
Remarks: Ten adult male Hartley guinea pigs were administered 0.3 g TMA once a
week for three weeks. Two weeks after the last induction dose, the induced
animals and ten control animals were administered a 0.3 g challenge dose,
and another challenge dose after an additional 13 days. Unlike the earlier
study in which TMA was dissolved in dimethyl sulfoxide and acetone (see
separate summary for IITRI, 1987), positive erythema reactions were not
observed. The authors concluded that TMA was not sensitizing under the
conditions of this study.
Reference: IITRI, 1993
(c)
Type: Dermal Sensitization
Species/strain: Mouse/BALB/c
Results: Sensitizing [X]; Not sensitizing [ ]; ambiguous [ ]
Classification: Sensitizing [X]; Not sensitizing [ ]
Method:
GLP: Yes [ ] No [ ] ? [X]
Test substance: 10% trimellitic anhydride solution in acetone/olive oil (4:1)
Remarks: Groups of ten female mice received 50 L of the test solution bilaterally on
each shaved flank, followed by a second treatment after five days. Five days
after the second treatment the mice had 25 uL of the test solution applied to
the backs of both ears. Exposure produced an increase in hapten-specific
immunoglobin E (IgE) and total IgE in serum. Cytokine secretion from
lymph nodes cells displayed profiles characteristic of Th1 and Th2-type cell
stimulation. Other chemicals were also tested in this assay.
Reference: Dearman et al., 1996
(d)
Type: Dermal Sensitization
Species/strain: Mouse/BALB/c
Results: Sensitizing [X]; Not sensitizing [ ]; ambiguous [ ]
Classification: Sensitizing [X]; Not sensitizing [ ]
Method:
GLP: Yes [ ] No [ ] ? [X]
Test substance: Trimellitic anhydride - 50% solution in acetone/olive oil (4:1)
Remarks: Groups of four female mice received 25 uL of the test solution on the backs
of both ears. Three days following exposure the mice were injected with
radiolabeled thymidine and sacrificed. Exposure to TMA produced an
increase in serum IgE and a lymphocyte proliferation response. Other
chemicals were also tested.
Reference: Dearman et al., 1992
(e)
Type: Dermal Sensitization
Species/strain: Mouse/BALB/c
Results: Sensitizing [X]; Not sensitizing [ ]; ambiguous [ ]
Classification: Sensitizing [X]; Not sensitizing [ ]
Method:
GLP: Yes [ ] No [ ] ? [X]
Test substance: Trimellitic anhydride - 10% solution in acetone/olive oil (4:1)
Remarks: Groups of ten female mice received 50 uL of the test solution bilaterally to
each shaved flank once, followed by a repeat dose five days later.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
48
Five days after the second treatment, animals received 25 uL of the test
solution on the backs of both ears daily for three days. Exposure to TMA
produced an increased expression of Th2 cytokines. Other chemicals were
also tested.
Reference: Dearman et al., 1996
(f)
Type: Dermal Sensitization
Species/strain: Rat/Norway and Wistar
Results: Sensitizing [X]; Not sensitizing [ ]; ambiguous [ ]
Classification: Sensitizing [X]; Not sensitizing [ ]
Method:
GLP: Yes [ ] No [ ] ? [X]
Test substance: 50% - 25% TMA solutions in acetone/olive oil (4:1)
Remarks: Groups of six female rats received 150 uL of a 50% solution bilater ally to
each shaved flank once. Seven days after the initial treatment, animals
received 75 uL of a 25% solution on the backs of both ears. Two weeks
following dermal exposure, animals were administered one or two challenge
exposures to TMA via inhalation at concentrations ranging from 16-52
mg/m
3
. Inhalation exposure to TMA produced dermal effects (encrustation,
erythema, scaliness) to both flanks and ears within one day of exposure,
persisting for two to five days. Serum IgE levels were elevated in Norway
rats but not Wistar rats. However, both species responded to the respiratory
challenge with altered breathing rates and histopathological changes to the
larynx and lungs. Other chemicals were also tested.
Reference: Arts et al., 1998
5.4 REPEATED DOSE TOXICITY
(a)
Species/strain: Rat/Sprague-Dawley
Sex: Female [ ]; Male [ ]; Male/Female [ X ]; No data [ ]
Route of Administration: Inhalation
Exposure period: 6 hrs/day
Frequency of treatment: 5 d/wk; 13 wks
Post exposure observation period: up to 38 wks
Dose: 0, 2, 15, 54 ug/m
3
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X ]; Concurrent vehicle [ ]; Historical [ ]
NOEL: --
LOEL: 2 ug/m
3
Method:
GLP: Yes [X ] No [ ] ? [ ]
Test substance: Trimellitic anhydride
Remark: Three groups consisting of ten male and ten female Sprague Dawley rats
each were exposed to TMA via inhalation for 13 weeks. Recovery times
of 0, 3, and 38 weeks were evaluated. No treatment related deaths were
observed. The lung appeared to be the only tissue affected, resulting in
minimal treatment-related effects observed (slight increase in lung
weight and volume and a small degree of pulmonary pneumonia).
Pulmonary physiology parameter s were unaffected. These results are in
contrast to the more severe effects observed following 6.5 weeks of
exposure under the same conditions (see separate summary), suggesting
some degree of adaptation (immunologic tolerance). Antibody levels
were elevated in a dose-dependent manner, beginning at the lowest dose
tested. Lung foci were also increased in a dose-dependent manner,
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
49
beginning at the mid-dose in females and low-dose in males; however,
statistical significance was achieved only in high-dose males due to
large variability. Minimal effects were observed in the 3 and 38-week
recovery groups.
Reference: Leach, 1989
(b)
Species/strain: Rat/Sprague-Dawley
Sex: Female [ ]; Male [X]; Male/Female [ ]; No data [ ]
Route of Administration: Inhalation
Exposure period: 6 hrs/day
Frequency of treatment: 5 d/wk; 2 wks
Post exposure observation period: Up to 12 days
Dose: 0, 30, 300 ug/m
3
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X ]; Concurrent vehicle [ ]; Historical [ ]
NOEL: 300 ug/m
3
LOEL: --
Method:
GLP: Yes [ x ] No [ ] ? [ ]
Test substance: TMA
Remark: Groups of 30 male Sprague Dawley rats were each exposed to 0, 30, or
300 mg/m
3
TMA via inhalation for two weeks. Recovery times of 0 and
12 days were evaluated. No deaths occurred during the study, no
alteration in serum antibody levels, and no significant clinical signs were
observed in the test article-treated groups. There were no statistically
significant effects of treatment on body weights or body weight gains,
organ weights, clinical chemistry or hematology parameters or serum
antibody levels. There were no gross or histopathological lesions
attributable to exposure to the test article. There were no effects seen on
repeat (challenge) exposure to the test article.
Reference: IITRI, 1985
(c)
Species/strain: Rat/Sprague-Dawley
Sex: Female [ ]; Male [ X ]; Male/Female [ ]; No data [ ]
Route of Administration: Inhalation
Exposure period: 6 hrs/day
Frequency of treatment: 5 d/wk; 6.5 wks
Post exposure observation period:
Dose: 0, 2, 15, 54 ug/m
3
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X ]; Concurrent vehicle [ ]; Historical [ ]
NOEL: --
LOEL: 2 ug/m
3
Method:
GLP: Yes [X ] No [ ] ? [ ]
Test substance: Trimellitic anhydride
Remark: Groups of ten male Sprague Dawley rats each were exposed to TMA via
inhalation for 6.5 weeks. No treatment related deaths were observed.
The lung appeared to be the only tissue affected, resulting in treatment-
related effects (increased lung weight and volume, external hemorrhagic
foci, inflammatory cell infiltration, and bronchoalveolar pneumonia) that
were more severe than observed in similarly treated rats exposed for 13
weeks (see separate summary). Antibody levels and lung foci were
elevated in a dose-dependent manner, beginning at the lowest dose
tested.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
50
Reference: Leach, 1989
(d)
Species/strain: Rat/Sprague-Dawley
Sex: Female [ ]; Male [ ]; Male/Female [ X]; No data [ ]
Route of Administration: Inhalation
Exposure period: 6 hrs/day
Frequency of treatment: 5 d/wk; 2 wks
Post exposure observation period:
Dose: 0, 500 ug/m
3
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X ]; Concurrent vehicle [ ]; Historical [ ]
NOEL: --
LOEL: 500 ug/m
3
Method:
GLP: Yes [ X] No [ ] ? [ ]
Test substance: Trimellitic anhydride
Remark: Groups of ten male and ten female Sprague Dawley rats were exposed to
500 ug/m
3
TMA or filtered air via inhalation for two weeks. Group 1
animals received filtered air and served as controls. Groups 2 and 3
were gonadectomized. Group 3 animals were cross-treated with sex
hormones (females given testosterone; males given estrogen). Group 4
served as the surgery control, and Group 5 only received TMA
exposure. Changes in body weight were attributable to hormone
exposure only. Hemorrhagic foci of the lung, increased lung weight,
and TMA-specific IgG antibodies were observed in all animals treated
with TMA. These effects were more pronounced in males than in
females. Estrogen treatment reduced the number of foci in both males
and females. Testosterone treatment did not have a significant effect.
The precise mechanism by which estrogen alters TMA toxicity is not
known, but suggests an interaction between the immune and endocrine
systems.
Reference: IITRI, 1992
(e)
Species/strain: Rat/Sprague-Dawley
Sex: Female [ ]; Male [ X ]; Male/Female [ ]; No data [ ]
Route of Administration: Inhalation
Exposure period: 6 hrs/day
Frequency of treatment: 2, 6, or 10 days
Post exposure observation period:
Dose: 0, 100 ug/m
3
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X ]; Concurrent vehicle [ ]; Historical [ ]
NOEL: --
LOEL: 100 ug/m
3
Method:
GLP: Yes [ ] No [ ] ? [ X ]
Test substance: Trimellitic anhydride (micronized powder)
Remark: Groups of 15 male Sprague Dawley rats were exposed to TMA via
inhalation for two, six, or ten days. No effects were noted after two
days of exposure. Exposure to TMA produced minimal lung injury and
ultrastructural changes at six days, becoming marked at ten days.
Similarly antibody levels in bronchoalveolar lavage (BAL) and serum
were elevated in a duration-dependent manner at six and ten days.
Antibody levels in BAL and serum were correlated with lung injury.
Reference: Zeiss et al., 1988; Pien et al., 1988
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
51
(f)
Species/strain: Mouse/Swiss-Webster
Sex: Female [ ]; Male [ X ]; Male/Female [ ]; No data [ ]
Route of Administration: Inhalation
Exposure period: 30 min/day
Frequency of treatment: 5 days
Post exposure observation period: 14 days
Dose: 0, 10, 70, 150 ug/m
3
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X ]; Concurrent vehicle [ ]; Historical [ ]
NOEL: --
LOEL: 10 ug/m
3
Method:
GLP: Yes [ ] No [ ] ? [ X ]
Test substance: Aerosol, dissolved in acetone
Remark: Alterations in breathing patterns (decreased time of inspiration and
expiration, increased length of apneic periods) were observed. No
histopathological changes were evident. Authors concluded that
respiration effects may be attributable to stimulation of vagal nerve
endings in deep lung.
Reference: Schaper and Brost, 1991
(g)
Species/strain: Rat/Sprague-Dawley
Sex: Female [ ]; Male [ ]; Male/Female [ X ]; No data [ ]
Route of Administration: Inhalation
Exposure period: 6 hrs/day
Frequency of treatment: 5 days/week, 1-2 weeks
Post exposure observation period: 0, 12 days
Dose: 0, 10, 30, 100, 300 ug/m
3
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X ]; Concurrent vehicle [ ]; Historical [ ]
NOEL: --
LOEL: 10 ug/m
3
Method:
GLP: Yes [ ] No [ ] ? [ X ]
Test substance: Trimellitic anhydride (micronized powder)
Remark: Groups of 60 rats (40 male and 20 female) were exposed to TMA via
inhalation for one or two weeks. Animals were sacrificed either after
the last exposure or following a 12-day recovery period. Antibody
response was elevated in a concentration-dependent manner beginning at
the lowest concentration at ten and 22 days, but not at five days. A
statistically significant correlation was reported for antibody levels and
hemorrhagic lung foci after ten days. Lung foci completely resolved
after 12 days of recovery, but reappeared following exposure to a single
challenge concentration.
Reference: Zeiss et al., 1987; Leach et al., 1987
(h)
Species/strain: Rat/Sprague-Dawley
Sex: Female [ ]; Male [ X ]; Male/Female [ ]; No data [ ]
Route of Administration: Inhalation
Exposure period: 6 hrs/day
Frequency of treatment: days 1, 5, 10 (challenge on day 22 or 29)
Post exposure observation period: 1 day
Dose: 330, 500 ug/m
3
Control group: Yes [ ]; No [ X ]; No data [ ];
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
52
Concurrent no treatment [ ]; Concurrent vehicle [ ]; Historical [ ]
NOEL: --
LOEL: 330 ug/m
3
Method:
GLP: Yes [ ] No [ ] ? [ X ]
Test substance: Trimellitic anhydride (micronized powder)
Remark: In the first study, six male rats were exposed to 500 ug/m
3
TMA for 6
hours/day on days one, five, and ten. Serum samples were collected
every second day beginning on day 1 through 26. A six-hour challenge
exposure of 540 ug/m
3
was administered on day 29 and animals were
sacrificed on day 30. Serum antibody levels were increased beginning at
day 5-7, reaching a peak one day that then declined for IgM, but
plateaued for IgG. IgG levels were 10-fold higher than either IgA or
IgM levels. In study 2, 18 male rats were exposed to 330 ug/m
3
TMA
for 6 hours/day on days one, five, and ten. A six-hour challenge
exposure of 300 ug/m
3
was administered on day 22 and animals were
sacrificed on day 23. Antibody levels were highly correlated with the
number of lung hemorrhagic foci, lung weights, and lung displacement
volumes. In a third study, eight male rats were exposed to 500 ug/m
3
TMA for six hours/day on days one and 5. A six-hour challenge
exposure of 500 ug/m
3
was administered on day 29 and animals were
sacrificed on day 30. Again, antibody levels were highly correlated with
the number of lung hemorrhagic foci, lung weights, and lung
displacement volumes, even with only two exposures.
Reference: Zeiss et al., 1989
(i)
Species/strain: Rat/Sprague-Dawley
Sex: Female [ ]; Male [ X ]; Male/Female [ ]; No data [ ]
Route of Administration: Inhalation
Exposure period: 4 hrs/day
Frequency of treatment: 1-10 days
Post exposure observation period: 1 day
Dose: 0, 500 ug/m
3
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X ]; Concurrent vehicle [ ]; Historical [ ]
NOEL: --
LOEL: 500 ug/m
3
Method:
GLP: Yes [ ] No [ ] ? [ X ]
Test substance: Trimellitic anhydride (micronized powder)
Remark: Groups of five male rats were exposed to 500 ug/m
3
TMA for four
hours/day for one to ten days. Lung injury was markedly increased on
days 7-10. Antibody response in serum and lung lavage fluid correlated
with lung injury.
Reference: Zeiss et al., 1992
(j)
Species/strain: Rat/Sprague-Dawley
Sex: Female [ ]; Male [ X ]; Male/Female [ ]; No data [ ]
Route of Administration: Inhalation
Exposure period: 6 hrs/day
Frequency of treatment: 2, 6, or 10 days
Post exposure observation period:
Dose: 0, 100 ug/m
3
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X ]; Concurrent vehicle [ ]; Historical [ ]
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
53
NOEL: --
LOEL: 100 ug/m
3
Method:
GLP: Yes [ ] No [ ] ? [ X ]
Test substance: Trimellitic anhydride (micronized powder)
Remark: Groups of ten male Sprague Dawley rats were exposed to TMA via
inhalation for two, six, or ten days. Antibody levels were elevated in
serum and bronchoalveolar lavage (BAL). Antibody levels in BAL
were ~15 times higher than in matched serum pair.
Reference: Chandler et al., 1987
(k)
Species/strain: Rat/albino
Sex: Female [ ]; Male [ ]; Male/Female [ X ]; No data [ ]
Route of Administration: Diet
Exposure period: 13 weeks
Frequency of treatment: ad libitum
Post exposure observation period: none
Dose: 0, 1,000, 5,000, 10,000 ppm
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X ]; Concurrent vehicle [ ]; Historical [ ]
NOEL: --
LOEL: 1,000 ppm (assuming a food intake of 0.05 kg/kg-day, dose=50 mg/kg-
day)
Method:
GLP: Yes [ ] No [ ] ? [ X ]
Test substance: TMA
Remark: Groups of ten male and ten female rats were exposed to TMA in the diet
for 13 weeks. No effects were observed with respect to appearance and
behavior, pathology, or urine values. A dose-dependent increase in
leukocyte counts was observed in both males and females. However,
this effect was not observed in a second study conducted in rats (see
summary for IBT, 1970).
Reference: Hill Top, 1969a
(l)
Species/strain: Rat/albino
Sex: Female [ ]; Male [ X ]; Male/Female [ ]; No data [ ]
Route of Administration: diet
Exposure period: 90 days
Frequency of treatment: ad libitum
Post exposure observation period: none
Dose: 0, 10,000 ppm
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X ]; Concurrent vehicle [ ]; Historical [ ]
NOEL: 10,000 ppm (assuming a food intake of 0.05 kg/kg-day, dose=500
mg/kg-day)
LOEL: --
Method:
GLP: Yes [ ] No [ ] ? [ X ]
Test substance: TMA
Remark: Groups of ten male and ten female rats were exposed to TMA in the diet
for 90 days. No effects were observed with respect to appearance and
behavior, pathology, or urine values. Unlike an earlier study conducted
in rats (see summary for Hill Top, 1969) no treatment related effects
were observed on leukocyte counts.
Reference: IBT. 1970
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
54
(m)
Species/strain: Dog/beagle
Sex: Female [ ]; Male [ ]; Male/Female [ X ]; No data [ ]
Route of Administration: Diet
Exposure period: 13 weeks
Frequency of treatment: ad libitum
Post exposure observation period:
Dose: 0, 1,000, 10,000, 20,000 ppm
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X ]; Concurrent vehicle [ ]; Historical [ ]
NOEL: 20,000 ppm (assuming a food intake of 0.025 kg/kg-day, dose=500
mg/kg-day)
LOEL: --
Method:
GLP: Yes [ ] No [ ] ? [ X ]
Test substance: TTMA
Remark: Groups of two male and two female dogs were exposed to TMA in the
diet for 13 weeks. No treatment-related effects were observed with
respect to appearance and behavior, pathology, serum chemistry, or
urine values. Adrenal weights were slightly increased in treated
animals, but could not be assessed statistically due to the small number
of animals tested.
Reference: Hill Top. 1969b
5.5 GENETIC TOXICITY IN VITRO
A. BACTERIAL IN VITRO TEST
(a)
Type: Mutagenicity
System of testing: Salmonella TA98, TA100, TA1535, TA1537
Concentration: 33, 100, 333, 1000, 3333, 10000 ug/plate
Metabolic activation: With [ ]; Without [ ]; With and Without [ X]; No data [ ]
Results:
Cytotoxicity conc: 1000
Precipitation conc: --
Genotoxic effects: + ? --
With metabolic activation: [ ] [ ] [ X]
Without metabolic activation: [ ] [ ] [ X]
Method: OECD 471
GLP: Yes [X ] No [ ] ? [ ]
Test substance: TMA
Remarks: In the dose range-finding study, toxicity, but no precipitation, was reported at
concentrations of 1,000 ug/plate or more. TMA did not produce a positive
mutagenic response under the conditions of this assay.
Reference: San and Wagner, 1991
(b)
Type: Mutagenicity
System of testing: Salmonella TA97, TA98, TA100, TA1535, TA1537
Concentration: 100; 333; 1,000; 3,333; 10,000 ug/plate
Metabolic activation: With [ ]; Without [ ]; With and Without [ X]; No data [ ]
Results:
Cytotoxicity conc: 10,000 ug/plate
Precipitation conc: --
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
55
Genotoxic effects: + ? --
With metabolic activation: [ ] [ ] [ X]
Without metabolic activation: [ ] [ ] [ X]
Method: Ames assay
GLP: Yes [ ] No [ ] ? [ X]
Test substance: TMA
Remarks: TMA was not mutagenic under the conditions of this assay
Reference: Mortelmans et al., 1986
B. NON-BACTERIAL IN VITRO TEST
(a)
Type: HGPRT mutations
System of testing: Chinese hamster ovary cells
Concentration: 500; 750; 1,000; 1,500; 2,000 ug/mL
Metabolic activation: With [ ]; Without [ ]; With and Without [ X]; No data [ ]
Results:
Cytotoxicity conc: --
Precipitation conc: --
Genotoxic effects: + ? --
Without metabolic activation: [ ] [ ] [ X ]
Method:
GLP: Yes [ X] No [ ] ? [ ]
Test substance: TMA dissolved in dimethylsulfoxide
Remarks: The mutagenicity of TMA was evaluated using the CHO/HGPRT assay with
and without liver S-9 from Aroclor induced rats. Results were negative
under the conditions of this assay.
Reference: Bigger and Sigler, 1991
(b)
Type: Chromosomal aberrations
System of testing: Chinese hamster ovary cells
Concentration: 260, 520, 1040, 2080 ug/mL
Metabolic activation: With [ ]; Without [ ]; With and Without [ X]; No data [ ]
Results:
Cytotoxicity conc: Mitotic inhibition (41%) at highest concentration w/o activation
Precipitation conc: --
Genotoxic effects: + ? --
Without metabolic activation: [ ] [ ] [ X ]
Method:
GLP: Yes [ X] No [ ] ? []
Test substance: TMA dissolved in dimethylsulfoxide
Remarks: The cytogenicity of TMA was evaluated using the CHO cells with and
without liver S-9 from Aroclor induced rats. Toxicity, as indicated by
mitotic inhibition, was noted at the highest concentration without activation.
Results for chromosomal aberrations were negative under the conditions of
this assay.
Reference: Putman and Morris, 1991
5.6 GENETIC TOXICITY IN VIVO
Although no in vivo genotoxicity studies were located for TMA or TMLA, the consistent negative
results observed for these chemicals from in vitro studies suggests that the potential for significant
genotoxicity is low.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
56
5.7 CARCINOGENICITY
No data available
5.8 TOXICITY TO REPRODUCTION
Although a multigenerational reproductive toxicity test was not located for TMA or TMLA, data
available from other studies suggest that the potential for significant toxicity to reproduction from
exposures to these chemicals is low. For example, subchronic inhalation exposures of male and
female rats to TMA concentrations up to 0.054 mg/m3, or to TMLA concentrations up to 0.30
mg/m3 did not result in any histopathological effects to reproductive tissues (IITRI, 1988, 1989).
Similarly, no histopathological effects of reproductive tissues were observed in rats exposed to
concentrations as high as 10,000 ppm TMA in feed (approximately 500 mg/kg-day) for 90 days
(IBT, 1970; Hill Top, 1969), or in dogs exposed to concentrations as high as 20,000 ppm TMA in
feed (approximately 500 mg/kg-day) for 13 weeks (Hill Top, 1969). Additionally, reproductive
performance was not affected in female rats and guinea pigs following exposure to TMA
concentrations of 0.5 mg/m3 on days 6 through 15 of gestation (Ryan, 1988). Because TMA is
likely hydrolyzed to form TMLA in tissues, these studies also provide information about TMLA.
5.9 DEVELOPMENTAL TOXICITY/TERATOGENICITY
(a)
Species/strain: Rat/Sprague-Dawley
Sex: Female [ X ]; Male [ ]; Male/Female [ ]; No data [ ]
Route of Administration: Inhalation
Duration of the test: 6 hrs/day
Exposure period: gestation day 6-15
Frequency of treatment: Daily
Doses: 0, 500 ug/m
3
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X]; Concurrent vehicle [ ]; Historical [ ]
NOEL Maternal Toxicity: --
NOEL Fetotoxicity: --
NOEL Teratogenicity 500 ug/m
3
Results: Lung foci and TMA-specific antibody were observed in exposed dams.
TMA-specific antibody was also noted in neonatal rats. Lung foci were only
observed in the challenged offspring whose mothers had not completely
recovered from the original TMA exposure. Lung foci were not observed in
adult rat offspring.
Method:
GLP: Yes [ ] No [ ] ? [X]
Test substance:
Remarks: No teratogenic effects or fetal deaths were observed.
Reference: Ryan, 1988
(b)
Species/strain: Guinea Pig/Hartley
Sex: Female [X]; Male [ ]; Male/Female [ ]; No data [ ]
Route of Administration: Inhalation
Duration of the test: 6 hrs/day
Exposure period: gd 6-15
Frequency of treatment: Daily
Doses: 0, 500 ug/m
3
Control group: Yes [ X ]; No [ ]; No data [ ];
Concurrent no treatment [ X]; Concurrent vehicle [ ]; Historical [ ]
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
57
NOEL Maternal Toxicity: --
NOEL Fetotoxicity: 500 ug/m
3
NOEL Teratogenicity 500 ug/m
3
Results: Lung foci and TMA-specific antibody were observed in exposed dams.
TMA-specific antibody was also noted in serum of guinea pig fetuses, but
not in neonatal guinea pigs. Unlike rats (see separate summary above), lung
foci were not observed in neonatal or adult guinea pigs.
Method:
GLP: Yes [ ] No [ ] ? [X]
Test substance:
Remarks: No teratogenic effects or fetal deaths were observed.
Reference: Ryan, 1988
5.10 OTHER RELEVANT INFORMATION
A. Specific toxicities
(a)
Type: Mechanistic study on lung foci in rats
Species/Strain Rat/Sprague-Dawley
Results: Mechanistic studies on the lung lesions of rats exposed to TMA via
inhalation indicated that (1) if rats were immunosuppressed, then TMA did
not cause lung lesions; (2) serum from nontolerant, TMA-sensitized rats
contained antibody which when passively transferred into nave recipient rats
resulted in TMA-induced lung lesions following a single TMA challenge
exposure; (3) enzyme, protein and cellular analyses of lung lavage fluid
indicated that TMA produces pulmonary inflammation and resultant
hemorrhage into the lung, but had no effect on macrophage function.
Remarks: The formation of lung lesions in rats following inhalation exposure to TMA
was consistent with immune complex injury syndrome.
References: IITRI, 1988
(b)
Type: Mechanistic study on IgG binding
Species/Strain Human, Rat/Sprague-Dawley
Results: In vitro study of the inhibition of human and rat IgG binding by trimellitic rat
serum albumin (TM-RSA) and trimellitic human serum albumin (TM-HSA).
Rat IgG binding was inhibited by both TM-RSA and TM-HSA, while human
IgG binding was inhibited only by TMA-HSA.
Remarks: Human IgG appears to be more specific than rat IgG.
References: Chandler et al., 1987
(c)
Type: Mechanistic study on inhalation exposure
Species/Strain Mouse/female BALB/c
Results: Atmospheres containing low molecular weight respiratory allergens can
initiate specific IgE responses in mice and inhaled chemicals may differ in
their ability to induce IgE antibody.
Remarks: Inhalation exposureto TMA resulted in the appearance of both serum IgG
and IgE anti-hapten anitbody. Under the same exposure conditions, 2,4-
dinitrochlorobenzene (a contact allergen that lacks the capacity for
respiratory sensitization) failed to elicit detectable quantities of DNP
antibody.
References: Dearman et al., 1991.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
58
(d)
Type: Mechanistic study on IgE and IgG antibody responses to the trimellitic (TM)
hapten group
Species/Strain Mouse/ (BALB/c x A/J)F1 hybrids
Results: The administration of TM-D-GL effectively abolished the ongoing IgE and
IgG responses in mice previously immunized with TM-protein conjugate.
Remarks: This study provides evidence for the potential clinical application of the D-
GL immunotherapeutic approach to TM sensitivity.
References: Liu et al., 1980.
(e)
Type: Mechanistic study by inhalation
Species/Strain Rat/Sprague-Dawley
Results: Results confirmed that cyclophosphamide-treated rats showed very little if
any blastogenic response to TMA. However, the saline-treated rats gave the
normal immune response. Cyclophosphamide eliminated T- and B-cell
function, which prevented the occurrence of TMA lesions.
Remarks: Rats were exposed to 95 g/m
3
TMA for six h/day, five days/wk for two
weeks. The rats received daily injections of the immunosuppressant
cyclophosphamide or saline.
References: Leach et al., 1988.
(f)
Type: Mechanistic study by dermal absorption
Species/Strain Mouse/BALB/c strain, female
Results: Mice were dermally exposed to TMA or to 2,4-dinitrochlorobenzene
(DNCB) by repeated applications. An elevation in the expression of mRNA
for interleukin 4 and interleukin 10 by lymph node cell from both the TMA
and DNCB-treated mice was observed within six hours of culture and
reaching maximum levels after 72 hours. Changes in cytokine mRNA in
allergen-activated lymph node cells preceded protein production; however,
the kinetic profiles were similar.
Remarks: This study suggests that the divergent cytokine secretion profiles shown by
mice treated by repeated dermal exposure to contact and respiratory allergens
are primarily controlled at the level of transcription.
References: Warbrick et al., 1998.
B. Toxicodynamics, toxicokinetics
(a)
Type: Distribution and Kinetic Study
Species/Strain Rat/Sprague-Dawley
Results: Tmax = <3 hours
Elimination rate constants ranged from 0.015 0.214
Biological half-lives ranged from 3-46 days
Remarks: Fourteen male and 14 female Sprague-Dawley rats were exposed to 950
ug/m
3 14
C-radiolabeled TMA (radiolabel was in the 2-carbonyl position) via
whole body inhalation for 45 minutes. Particle sizing analysis was not
performed because the test atmosphere was radioactive and therefore, the
fraction of respirable particles was determined. Animals were sacrificed 3
hrs, 1, 2, 4, 8, 16, and 32 days post-exposure for tissue analysis. The highest
concentrations were generally observed at the first time point (T
max
<3 hours).
A second T
max
of eight-days was reported for lung lymph nodes in male rats,
suggesting a potential role in gender lung toxicity in male rats as reported in
a previous study. Sex differences in half-lives were reported for popliteal
and lung lymph nodes, bone marrow, and heart.
References: IITRI, 1988a
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
59
5.11 EXPERIENCE WITH HUMAN EXPOSURE
(a)
Results: Highest arithmetic mean TMA exposures of 19.3 ug/m
3
were found in a resin
factory (approximately half the eight-hour TWA OES of 40 ug/m
3
).
Remarks: A retrospective cohort study was carried out in four factories (three alkyd
resin factories and one cushioned flooring factory) to investigate the nature of
exposure-response relationships for sensitization to TMA and other
anhydrides. Arithmetic mean exposure levels to TMA were below their
respective OESs in the resin factories. Relatively high full-shift exposures to
TMA occurred in the cushioned floor facility, although no high peak
exposures were detected.
Reference: van Tongeren et al. 1995
(b)
Results: Seven employees had positive IgE antibody levels against TM-human serum
albumin, one of these employees had rhinitis, and another possibly had TMA
asthma/rhinitis. Positive IgG antibody levels against TM-HAS was reported
in fourteen employees (although only three had titers high enough to cause
disease).
Remarks: This was an eleven year study of employees exposed to TMA that included
periodic serum antibody studies and health questionnaires. Industrial
hygiene data from the plant in 1989 reported exposures ranging from <0.003
mg/m
3
to 0.77 mg/m
3
. Respirators with a protection factor of 100 were worn
routinely in areas where TMA was used.
Reference: Grammer et al., 1991.
(c)
Results: Personal monitoring data results included geometric means for different
exposure classes: 1: 0.17 mg/m
3
; 2: 0.087 mg/m
3
; 3: <0.00055 mg/m
3
; 4:
<0.00041 mg/m
3
; 5: <0.00053 mg/m
3
. Nearly 7% of the employees had a
TMA immunologic syndrome, 31.6% displayed an irritant response, and the
remaining employees (61.6%) had no symptoms.
Remarks: Employees of a large chemical manufacturing complex were (n=474) studied
to relate TMA exposure to serologic and clinical outcomes. All employees
were assigned to a TMA exposure class from 1 (highest) to 5 (lowest).
Reference: Zeiss et al., 1992.
(d)
Results: Mean personal monitoring data results were presented by an exposure class:
1 (0.13 mg/m
3
); 2 (0.036 mg/m
3
); 3 (0.002 mg/m
3
); 4 (0.00051 mg/m
3
); and 5
(<0.00053 mg/m
3
). Of the 28 employees that were assigned to exposure class
1, 29% developed disease (related to IgG or IgE titer). In exposure class 2,
4% of the employees developed disease, 5% of the employees in class 3
developed disease, and no employees developed disease in exposure classes 4
and 5.
Remarks: A three-year study of 286 employees of a TMA manufacturing facility was
performed. Employees were assigned exposure classifications ranging from 1
(highest) to 5 (lowest), and immunological response was related to exposure.
Reference: Grammer et al., 1999
(e)
Results: Mean full-shift TMA exposure in four facilities ranged from 0.5-19.3 ug/m
-3
Remarks: Workers exposed to TMA were studied to determine the relation between
exposure to TMA (and other acid anhydrides, AA) and the risk of developing
skin prick test responses to AA-HSA.
Reference: Barker et al., 1998.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
60
(f)
Results: This study investigated nine workers who were exposed to a paint powder
that contained TMA at a 55-gallon drum manufacturing plant.
Environmental monitoring showed airborne TMA levels to be over 100 times
the OHSA PEL of 0.04 mg/m
3
.
Remarks: One employee exhibited obvious illness and two of the workers had definite
evidence of TMA-related pulmonary dysfunction and immunologic response.
Three workers showed IgG antibody against TM-HSA significantly higher
than control serum. One worker showed IgE antibody against TM-HSA.
Reference: Letz et al., 1987
(g)
Results: Forty-six employees exposed to TMA were investigated using periodic serum
antibody studies and questionnaires.
Remarks: Seven employees had positive IgE antibody against trimellityl-human serum
albumin (TM-HSA), one had TMA rhinitis, and another potentially had TMA
asthma/rhinitis. Positive IgG antibody against TM-HSA was observed in
fourteen employees, although only three had titers high enouth to cause
disease (none of them had symptoms associated with late respiratory systemic
syndrome (LRSS) or pulmonary disease anemia (PDA)). TMA exposure
concentrations for two different job categories ranged from <0.001-2.1 mg/m
3
and 0.005-0.32 mg/m
3
over a 14-year period.
Reference: Grammer, et al., 1992
(h)
Results: Average airborne TMA dust concentrations ranged from 0.006 - 2.1 mg/m
3
for three different job categories. Five workers had antibody against TM-
HSA, of these, three were diagnosed with the LRSS and one with TMA-
induced allergic rhinitis. After local exhaust ventilation had been improved,
average airborne dust concentrations decreased to approximately 0.01 mg/m
3
and the symptomatic improvement was noted in the individuals with the
LRSS.
Remarks: Eighteen workers exposed to TMA powder were evaluated. Annual clinical
evaluations and serum radioimmunoassays for total antibody binding and
specific IgE binding to TM-HSA were performed.
Reference: Bernstein et al., 1983
(i)
Results: A total of 119 subjects exposed to TMA for at least one year were identified
from a previous cross-sectional study. These individuals were studied for the
next five years to determine if they would develop an immunologic
respiratory disease due to TMA exposure. In 1990, 16 individuals showed
IgE against TMA conjugated to human serum albumin. Of these, three had
immediate asthma and six developed asthma during the five-year follow-up.
Of those without IgE against TM-HSA, none had immediate asthma in 1990
and only 1 out of 102 developed asthma after five years. Of those with IgG
against TM-HSA (44), six had immunologic respiratory disease in 1990 and
two more developed it in the following 5 years.
Remarks: Development of antibody (both IgE and IgG) against TM-HSA is predictive
of subjects who have or will develop immunlogically mediated respiratory
disease based on TMA exposure. The authors also concluded that the absence
of antibody is a potent negative predictor.
Reference: Grammer et al., 1998.
(j)
Results: Workers (n=196 individuals) involved in the manufacture of TMA were
studied for 12 years. Seventeen workers had IgE-mediated asthma/rhinitis
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
61
with a positive prick test to TM-HSA (with IgE antibody of 0.8-57 ng TM-
HSA bound/ml). Seven individuals had a late respiratory systemic syndrome
(LRSS) and four workers had both syndromes. Three workers had late onset
asthma, one had marked arthralgia and myalgia occurring hours after
exposure to TMA.
Remarks: The authors reported a reduction in the number of workers exhibiting an
immunologic syndrome during 1982-1987 in spite of the increased TMA
production. This finding paralleled environmental control and worker
education efforts.
Reference: Zeiss et al., 1990.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
62
6. REFERENCES
Barker RD, Van Tongeren MJ, Harris KE, Grammer K, Venables KM, Taylor AJ 1998. Risk Factors for
Sensitization and Respiratory Symptoms Among Workers Exposed to Acids Anhydrides: A Cohort Study.
Occup Environ Med. 55(10): 684-691
Bernstein, D.I., Roach, D.E., McGrath, K.G., et al. 1983. The relationship of airborne trimellitic anhydride
concentrations to trimellitic anhydride-induced symptoms and immune responses. J. Allergy Clin. Immunol.
72 (6): 709-713.
Bigger, C.A. and Sigler, C.I. 1991. CHO/HGPRT mutation assay with confirmation. Final Report.
Microbiological Associates, Inc. Laboratory Study Number TA039.332001.
ChemSystems, 2000. Trimellitic Anhydride/Pyromellitic Dianhydride, July 2000.
Chandler MJ, Zeiss CR, Leach CL, Hatoum NS, Levitz D, Garvin PJ, Patterson R, 1987. Levels and Specificity
of Antibody in Bronchoalvaolar Lavage (BAL) and Serum i n an Animal Model of Trimellitic Anhydride-
Induced Lung Injury. J Allergy Clin Immunol, 80(2): 223-229.
Dearman, R.J., Hegarty, J.M., and Kimber, I. 1991. Inhalation exposure of mice to trimellitic anhydride
induces both IgG and IgE anti-hapten antibody. Int. Arch. Allergy Appl. Immunol. 95: 70-76.
Dearman, R.J., Mitchel, J.A., Basketter, D.A., and Kimber, I. 1992. Differential ability of occupational
chemical contact and respiratory allergens to cause immediate and delayed dermal hypersensitivity reactions in
mice. Int. Arch. Allergy Immmunol. 97: 315-321.
Grammer, L.C., Shaughnessy, M.A., Kenamore, B.D., and Yarnold, P.R. 1999 A clinical and immunologic
study to assess risk of TMA-induced lung disease as related to exposure. JOEM. Volume 41 (12): 1048-1051.
Grammer, L.C., Shaughnessy, M., and Kenamore, B. 1998. Utility of antibody in identifying individuals who
have or will develop anhydride-induced respiratory disease. CHEST. 114: 1199-1202.
Grammer, L.C., Harris, K.E., Sonenthal, K.R., et al. 1992. A cross-sectional survey of 46 employees exposed
to trimellitic anhydride. Allerg. Proc. 13(3): 139-142.
Grammer, L.C., Harris, K.E., Sonethal, K.R., et al. 1991. Identification and control of trimellitic anhydride
immunologic lung disease: an 11-year experience. Clinical Research. Vol. 39 (3): 703A.
Hatoum, N. and Johnson, W. 1991. Primary eye irritation study of trimellitic anhydride in rabbits. IITRI
Study No. 1693, Test Article No. 128H.
Hill Top Research. 1969a. Thirteen week dietary administration of trimellitic anhydride to rats. Miamiville,
OH. S-192
Hill Top Research. 1969b. Dietary administration of trimellitic anhydride to dogs for 13 weeks. Miamiville,
OH. S-260
Horan, J.K. 1962. Aqueous hydrolysis of TMA memorandum. Amoco Chemicals Corporation, ACM62-
9/File 2570.9/2610190.
HSDB, 2002. Trimelltic Anhydride. Hazardous Substance Database. Agency for Toxic Substances and
Disease Registry, Public Health Service. Dept of Health and Human Services. Atlanta,GA.IBT. 1970.
Ninety-day subacute oral toxicity of LM 3813 (TMA) in albino rats. Northbrook, IL: Industrial Bio-Test
Laboratories. No. B7989
IITRI. 1993. Dermal sensitization study of trimellitic anhydride in guinea pigs using the modified Buehler
Method. Final Report. IITRI Project No. L08429, Study No. 1.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
63
IITRI. 1992a. Acute inhalation toxicity study of trimellitic anhydride (TMA) in rats. Final Report. IITRI
Project No. L08100, Study No. 1749, Test Article No. 128L.
IITRI. 1992b. Inhalation study of sex-related differences associated with the respiratory sensitization response
of trimellitic anhydride (TMA): a research project. Final Report. IITRI Project No. L08100, Study No. 1537,
Test Article No. 128G.
IITRI. 1991a. Acute oral toxicity study of trimellitic anhydride in rats. Final Report. IITRI Project NO.
L08100, Study No. 1699, Test Article No. 128I.
IITRI. 1991b. Acute dermal toxicity study of trimellitic anhydride in rabbits. Final Report. IITRI Project No.
L08100, Study No. 1700, Test Article No. 128I.
IITRI. 1991c. Acute dermal irritancy/corrosivity study of trimellitic anhydride in rabbits. Final Report. IITRI
Project No. L08100, Study No. 1694, Test Article No. 128H.
IITRI. 1989. Respiratory sensitization screen of trimellitic acid (TMLA) in rats. Final Report. IITRI Project
No. L8100, Study No. 1422, Test Article No. 228C.
IITRI. 1988a. Distribution and kinetic study of 14C-trimellitic anhydride in rats. Final Report. IITRI Project
No. L8100, Study No. 1021, Test Article No. 422.
IITRI. 1987. Dermal sensitization study of trimellitic anhydride (TMA) in guinea pigs. Final Report. IITRI
Project No. L8100, Study No. 1196, Test Article No. 128.
IITRI 1985. Two-Week Inhalation Toxicity Study of Trimellitic Anhydride in Rats. Final Report. IITRI Project
No. L8100, Study No. 674, Test Article No., 128C
IPCS (Intenationational Programme on Chemical Safety), 1992. Health and Safety Guide No. 71 Trimellitic
Anhydride World Health Organization for the International Programme on Chemical Safety (a collaborative
programme of the United Nations Environment Programme, the International Labour Organisation, and the
World Health Organization), Geneva Switzerland.
Knacker, T., Zietz, E., SchallnaB, H., and Diehl, T. 1993. A study of the acute toxicity to fish (Leuciscus idus
melanotus) of trimellitic anhydride. Final Report. Battelle Europe Study Number: BE-EA-128-91-01-F3A-1.
Knacker, T., Brodesser, J., SchallnaB, H., and Reifenberg, P. 1993. A study of the toxicity to algae
(Scenedesmus subspicatus) of trimellitic anhydride. Final Report. Battelle Europe Study Number: BE-EA-
128-91-02-ALG-1.
Knacker, T., Zietz, H. SchallnaB, H., and Reifenber, P. 1992. A study of the acute immobilisation to
DAPHNIA of trimellitic anhydride. Final Report. Battele Europe Study Number: BE-EA-128-91-02-DAK-1.
Leach, C.L., Hatoum, N.S., Zeiss, C.R., and Garvin, P.J. 1989. Immunologic tolerance in rats during 13 weeks
of inhalation exposure to trimellitic anhydride. Fundamental and Applied Toxicology. 12: 519-529.
Leach, C.L., Hatoum, N.S., Ratajczak, H.V., Zeiss, C.R., and Garvin, P.J. 1988. Evidence of immunologic
control of lung injury induced by trimellitic anhydride. Am. Rev. Respir. Dis. 137: 186-191.
Leach, C.L., Hatoum, N.S., Ratajczak, H.V., Zeiss, C.R., Roger JC, and Garvin, P.J. 1987. The pathologic and
Immunologic Repsonse to Inhaled Trimellitic Anhydride in Rats. Toxicol Appl Pharmacol. 87(1):67-80
Lebertz, H. 1991a. Study on the ready biodegradability (modified Sturm Test) of trimellitic anhydride. Final
Report. Battelle Europe Study No.: BE-EA-128-91-01-STT-01.
Lebertz, H. 1991b. Study on the toxicity of trimellitic anhydride towards bacteria. Final Report. Battelle
Europe Study-No. : BE-EA-128-91-01-BHT-01.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
64
Letz, G., Wugofski, L., Cone, J.E., et al. 1987. Trimellitic anhydride exposure in a 55-gallon drum
manufacturing plant: clinical, immunologic, and industrial hygiene evaluation. American Journal of Industrial
Medicine. 12: 407-417.
Liu, R., Bargatze, R.F. and Katz, D.H. 1980. Induction of immunologic tolerance to the trimellitate haptenic
group in mice: model for a therapeutic approach to trimellitic anhydride-induced hypersensitivity syndromes in
humans? J. Allergy Clin. Immunol. 66 (4): 322-326.
Mortelmans K, Haworth S, Lawlor T, Speck W, Trainer B, Zeiger E, 1986. Salmonella mutagenicity Tests: II.
Results from Testing of 270 Chemicals Environ Mutagen. 8(Suppl 7): 1-119
Putnam, D.L. and Morris, M.J. 1991. Chromosome aberrations in chinese hamster ovary (CHO) cells. Final
Report. Microbiological Associates, Inc. Laboratory Study Number TA039.337100.
Rom WN, 1992. Environmental and Occupational Mediceine. 2
nd
Edition. Little, Brown and Co., Boston, MA.
Ryan, B.M. 1988. Teratological evaluation of trimellitic anhydride (TMA) in rats and guinea pigs. Submitted
in partial fulfilment of the requirements for the degree of Master of Science in Biology in the School of
Advanced Studies of Illinois Institute of Technology, Chicago, Illinois.
San, R.H. and Wagner, V.O. 1991. Salmonella/mammalian-microsome plate incorporation mutagenicity assay
(Ames Test) with a confirmatory assay. Microbiological Associates, Inc. Laboratory Study Number
TA039.501014.
Schaffhausen, J.G. 1991. Partition coefficient for TMA. Letter to W.H. Frye/MS E-1A Naperville.
Schaper M and Brost MA, 1991. Respiratory Effects of Trimellitic Anhydride Aerosols in Mice. Arch Toxicol.
65(8): 671-677.
Schmitt, S. L. Technical dossier for TMA. CAS No. 552-30-7/PM/REF NO. 25550. Submitted by: Amoco
Corporation.
SRI, 2000. "Polyimides", Chemical Economics Handbook, November 2000
Trent University 1999. Fugacity-based Environmental Equilibrium Partitioning Model. Version 2.2
Environmental Modeling Center, Trent University, Peterborough, Ontario.
van Tongeren, M.J., Barker, R.D., Gardiner, K., et al. 1995. Exposure to acid anhydrides in three resin and
one cushioned flooring manufacturing plants. Ann. Occup. Hyg., Vol. 39. No. 5: 559-571.
Warbrick, E.V., Dearman, R.J., Basketter, D.A., Gerberick, G.F., Ryan, C.A., and Kimber, I. 1998. Analysis
of cytokine mRNA expression following repeated exposure of mice to chemical contact and respiratory
allergens. Journal of Applied Toxicology. 18: 205-213.
Zeiss, C.R., Mitchell, J.H., Van Peenen, P.F., et al. 1992. A clinical and immunologic study of employees in a
facility manufacturing trimellitic anhydride. Allergy Proc. Vol. 13 (4): 193-198.
Zeiss, C.R., Mitchell, J.H., Van Peenen, P.F., Harris, J., and Levitz, D. 1990. A twelve-year clinical and
immunologic evaluation of workers involved in the manufacture of trimellitic anhydride (TMA). Allergy Proc.
11 (2): 71-77.
Zeiss CR, Leach CL, Levitz D, Hatoum NS, Garvin PJ, Patterson R, 1989. Lung Injury Induced by Short-Term
Intermittent Trimellitic Anhydride (TMA) Inhalation. J Allergy Clin Immunol. 84(2):219-223.
Zeiss CR, Leach CL, Smith LJ, Levitz D, Hatoum NS, Garvin PJ, Patterson R, 1988. A Serial Immunologic
and Histopathologic Study of Lung Injury Induced by Trimellitic Anhydride. Am Rev Respir Dis. 137(1): 191-
196.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
65
Zeiss CR, Levitz D, Leach CL, Hatoum NS, Ratajczak HV, Chandler MJ, Roger JC, Garvin PJ 1987. Model of
Immunologic Lung Injury Induced by Trimellitic Anhydride Inhalation: Antibody Response. J Allergy Clin
Immunol. 79(1):59-63.
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
66
Robust Summari es for Tri mel l i ti c Anhydri de
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
67
PHYSICAL/CHEMICAL ELEMENTS
MELTING POINT
TEST SUBSTANCE
- Trimellitic Anhydride (TMA)
METHOD
- Method/guideline:
- GLP: ?
- Year (study performed):
- Remarks:
RESULTS
- Melting point: 165C (330
o
F)
- Decomposition:
- Sublimation:
- Remarks:
CONCLUSIONS
- The melting point for TMA is 165C
DATA QUALITY
REFERENCES
- Amoco Corporation. 1997. Material Safety Data Sheet
(www.vetmed.ucdavis.edu/msds/mf/Amoco/files/01260000.html)
OTHER
- Values ranging from 161-168
o
C have been reported for TMA (Amoco Corporation, 1991).
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
68
BOILING POINT
TEST SUBSTANCE
- Trimellitic Anhydride (TMA)
METHOD
- Method:
- GLP:
- Year (study performed):
- Remarks:
RESULTS
- Boiling point: 390C (730
o
F)
- Pressure:
- Pressure unit:
- Decomposition (yes/no/ambiguous)
- Remarks:
CONCLUSIONS
- The boiling point for TMA is 390 C
DATA QUALITY
REFERENCES
- Amoco Corporation. 1997. Material Safety Data Sheet
(www.vetmed.ucdavis.edu/msds/mf/Amoco/files/01260000.html)
OTHER
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
69
VAPOR PRESSURE
TEST SUBSTANCE
- Trimellitic Anhydride (TMA)
- Remarks:
METHOD
- Method: Calculated
- GLP:
- Year (study performed): 2002
- Remarks: Used SMILES notation of O=C(OC(=O)c1ccc(C(=O)O)c2)c12
RESULTS
- Vapor Pressure: 7.6 x 10
-5
Pa (5.69 x 10
-7
mm HG)
- Temperature: 25 C
- Decomposition:
- Remarks:
CONCLUSIONS
- The vapor pressure for TMA is 7.6x10
-5
Pa
DATA QUALITY
- Reliability: Klimisch Code 2 Reliable with restrictions, values is an estimate using an
accepted method.
REFERENCES
- US EPA, EPIWIN Suite, 1997 (http://www.epa.gov/oppt/exposure/docs/episuitedl.htm).
OTHER
- Vapor Pressure: 9.86x10
-6
mm Hg (Experimental), Daubert, T.E., and R.P. Danner (1989)
(Data from MPBPWIN v1.40 database in EPIWIN Suite.)
- Vapor Pressure: 5.69 x 10
-7
mm Hg (Calculated) (MPBPWIN v1.40 in EPIWIN Suite)
- Vapor Pressure: <1.1x10
-7
mm Hg, Amoco Corporation. 1997. Material Safety Data Sheet
(www.vetmed.ucdavis.edu/msds/mf/Amoco/files/01260000.html)
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
70
PARTITION COEFFICIENT
TEST SUBSTANCE
- Trimellitic Anhydride (TMA)
METHOD
- Method: calculated
- GLP: No
- Year (study performed): 2002
- Remarks: Used SMILES notation of O=C(OC(=O)c1ccc(C(=O)O)c2)c12
RESULTS
- Log Pow: 0.95
- Temperature:
- Remarks: TMA would only have transitory existence in an octanol/water mixture.
Hydrolysis of TMA in aqueous alcohol is extremely rapid at room temperature.
Consequently, TMLA would be formed upon dissolving TMA in this solvent system.
Furthermore, it is expected that small amounts of the diacid-octyl ester will form when octyl
alcohol reacts with the anhydride moiety of TMA, though the hydrolysis reaction is more
prevalent.
CONCLUSIONS
- The effective Log P
ow
value for TMA is 0.95.
REMARKS
- The model predicts the Log Pow of the unionized form though it is expected that at least
some of the trimeelitic acid would be ionized under environmentally relevant pH.
DATA QUALITY
- Reliability: Klimisch Code= 2 Reliable with restrictions. Value is an estimate using an
accepted method.
REFERENCES
- KOWWIN Version 1.66. (http://www.epa.gov/oppt/exposure/docs/episuitedl.htm)
OTHER
- If Log P
ow
is estimated without considering hydrolysis of TMA to TMLA, slightly larger
estimates are obtained (see calculated values below). However, the most environmentally
relevant value must reflect the hydrolysis of the anhydride to the acid. Consequently, the
recommended value is the same as for TMLA.
- Estimated Log Pow: 1.95 KOWWIN (EPIWIN Suite)
- Estimated Log Pow: 1.61. CLOGP Program (http://www.daylight.com)
- Estimated Log P
ow
: 1.61. Interactive Analysis Program (http://www.logp.com)
- Estimated Log P
ow
: 0.80. ALOGP Program (http://www.lhn.unil.ch/Appl/cchem2.html)
- Estimated Log P
ow
: 1.14. XLOGP Program (ftp2.ipc.pku.edu.cn)
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
71
WATER SOLUBILITY
TEST SUBSTANCE
- Identity: Trimellitic Anhydride (TMA)
- Remarks:
METHOD
- Method:
- GLP: ?
- Year (study performed):
- Remarks:
RESULTS
- Value: 21,000 mg/L
- Description of solubility:
- pH value and concentration at temperature C:
- pKa value at 25 C:
- Remarks: Moderate solubility after TMA hydrolysis to TMLA.
CONCLUSIONS
- TMA has moderate solubility in water after hydrolysis to TMLA. The recommended value
is for TMLA.
DATA QUALITY
REFERENCES
- SRC PhysProp Database. 2001. (http://esc.syrres.com/interkow/webprop.exe)
CAS=552-30-7.
- Amoco Corporation. 1997. Material Safety Data Sheet
(www.vetmed.ucdavis.edu/msds/mf/Amoco/files/01260000.html)
- Trimellitic anhydride fact sheet, http://ull.chemistry.uakron.edu/erd/chemicals/2501-
3000/2996.html
OTHER
- If water solubility is estimated without considering hydrolysis to acid, then lower values will
be estimated, as in following:
- Water solubility: 1211 mg/L (Estimate reflecting melting point) (WSKOW v1.40 in
EPIWIN Suite)
- Water solubility: 1036 mg/L. WSKOW v1.40 (in EPIWIN Suite)
- Water solubility: 860 mg/L. Interactive Analysis Program (http://ww.logp.com)
- Water solubility: 2777 mg/L. ALOGS Program, (http://www.lhn.unil.ch/Appl/cchem2.html)
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
72
ENVIRONMENTAL FATE ELEMENTS AND PATHWAYS
PHOTODEGRADATION
TEST SUBSTANCE
- Trimellitic Anhydride (TMA)
METHOD
- Method/guideline: Estimated - AOPWIN
- Type (test type): Estimated
- GLP:
- Year (study performed): 2001
- The atmospheric hydroxyl radical concentration of 1.5 x 10
6
molecule/cm
3
was used as a
standard default in the program.
RESULTS
- Direct photolysis:
- Half-life t : 13.4 days (321.6 hours)
- Remarks: Overall OH Rate Constant: 0.80E-12 cm
3
/molecule-sec. When exposed to humid
air, TMA hydrolyzes to TMLA at an apparent linear rate. Because the estimated photolysis
half-life of TMLA is shorter, the actual half-life of TMA in the air may be less than
estimated.
CONCLUSIONS
- The direct photolysis half-life in air is estimated to be 322 hours. Actual half-life may be
reduced if TMA hydrolyzes to TMLA in humid air.
DATA QUALITY
Reliability: Klimisch Code= 2 Reliable with restrictions. Value estimated using an accepted
method.
REFERENCES
- SRC. 2001. Atmospheric Oxidation Program for Microsoft Windows (AOPWIN).
Syracuse Research Center.
OTHER
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
73
STABILITY IN WATER
TEST SUBSTANCE
- Trimellitic Anhydride (TMA)
METHOD
- Method/guideline: Aqueous hydrolysis
- Type (test type):
- GLP: No
- Year (study performed): 1962
- Remarks: 50 grams of TMA was added to 200 ml of distilled water at 80 F (27C). A
temperature rise of 20 to 25 degrees was noted. Disappearance of trimellitic anhydride
flakes took from 8 to 9 minutes. When flakes were dissolved the sample was filtered and an
infra-red spectra was obtained. No absorption was noted at 5.35 or 5.60 microns, indicating
complete hydrolysis of the anhydride,
- Duration: less than 20 minutes
- Positive Controls:
- Negative Controls:
- Analytical procedures: Infra-red spectrophotometry
RESULTS
- Measured value:
- Degradation: TMA hydrolyzed to form TMLA acid within 10 minutes by stirring with
water at 80-90
o
F (27-32C).
- Breakdown products: Trimellitic acid
- Remarks: While temperature was not held constant, the range was within environmentally
relevant temperatures. The pH was not reported though one can assume the pH of distilled
water to be between 6 and 7 depending on the amount of dissolved gas and upon addition of
TMA the pH would drop to approximately 4.0.
CONCLUSIONS
- TMA was hydrolyzed to acid within 10 minutes in water at 80-90
o
F.
DATA QUALITY
- Reliability: Klimisch Code= 2 Reliable with restrictions. Temperature and pH were not
controlled.
REFERENCES
- Horan, 1962; Amoco Corporation, 1991.
OTHER
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
74
TRANSPORT BETWEEN ENVIRONMENTAL COMPARTMENTS (FUGACITY)
TEST SUBSTANCE
- Trimellitic Anhydride (TMA)
- Remarks:
METHOD
- Test (test type): Calculated
- Method: Levels I, II, and III
- Year (study performed): 2002
- Remarks: Half-lives in water, soil and sediment estimated using EPIWIN.
- Chemical Assumptions: Molecular weight 192; water solubility 21,000 g/m3; Vapor
pressure 7.6 x 10
-5
Pa; Log P
ow
0.95; Melting point 165
o
C; half-life in air 321.8
hours; half-life in water 360 hours; half life in soil 360 hours; half-life in sediment
1440 hours; all other parameters were default values. Level III model assumed continuous
1000 kg/hr releases to each compartment (air, water, and soil).
RESULTS
- Media: Air, soil, water and sediment concentrations were estimated.
- Estimated Distribution and Media Concentration reflecting TMA hydrolysis to TMLA:
Level I Level II Level III
Air 7.7E-7% 7.6E-7% 3.4E-6%
Water 99.2% 99.2% 50.6%
Soil 0.78% 0.78% 49.3%
Sediment 0.02% 0.02% 0.02%
- Remarks: Since default assumptions for release estimates were used, resulting
environmental concentrations are not provided.
- A biodegradation study showed that biodegradation of TMA/TMLA was more rapid than
value used in this model: 65% biodegradation was reported within 5 days (i.e., t
1/2
in water
<120 hours). Using a smaller value in the level III model reduces the concentration in water
compartment (and reduces the percentage estimate).
CONCLUSIONS
- These results indicated that TMA, hydrolyzed to TMLA in water and under humid
conditions, will partition primarily to water. Virtually no TMLA will partition to air. Soil
and sediment concentrations will be minimal at equilibrium. The Level III model suggests
soil may contain a significant percentage of TMLA, reflecting the assumed pattern of
chemical release (equal loading of water, soil and air)..
DATA QUALITY
- Reliability: Klimisch Code= 2 Reliable with restrictions. Value is an estimate using an
accepted method.
REFERENCES
- Trent University. 1999. Fugacity-based Environmental Equilibrium Partitioning Model.
Version 2.2. Environmental Modeling Centre, Trent University, Peterborough, Ontario.
(Available at http://www.trentu.ca/envmodel.)
- US EPA EPIWIN Suite. (Estimates of half-lives in water, soil, sediment from QSAR0
OECD SIDS TRIMELLITIC ANHYDRIDE
UNEP PUBLICATIONS
75
OTHER
- Using the EPIWIN software without modification to reflect hydrolysis of TMA to TMLA
results in different estimates. For completeness, these are shown below. However, because
of the relatively rapid and complete hydrolysis of TMA to TMLA in water or in humid
conditions, the reasonably expected behavior of TMA will be that observed for TMLA
- Chemical Assumptions if TMA itself is modeled (i.e., ignoring hydrolysis): Molecular
weight 192; water solubility 1211 g/m3; Vapor pressure 7.6 x 10
-5
Pa; Log P
ow