Brandt RAB/ECAD May 2, 1991 RAB PR£ RM1 PRESENTATION FOR C9 AROMATIC HYDROCARBON FRACTION SUMMARY: There are no data on exposures to the C9 aromatic hydrocarbon fraction. However, based on exposure modeling using 1,2,4-Trimethylbenzene (1,2,4-TMB), a major component of the C9 fraction, there may be health risks to consumers using solvent based furniture polish /varnishes composed of 50% C9 fraction. Workers may also be at risk during manufacturing, processing and use of the C9 fraction. Based on TRI data for 1,2,4-TMB there may be a risk to aquatic organisms from releases from one manufacturing site. The major human health hazard concerns are reproductive and developmental toxicity effects. I. CHEMICAL/PHYSICAL CHARACTERISTICS Structure: CS, R- CH. or R X-2 X 1,2,3- TMB (Hemimellitene) 1,3,5- TMB (Mesitylene) 1,2,4- TMB (Pseudocumene) TMB (mixed isomers) o-Ethyltoluene m-Ethyltoluen« p-Ethyltoluen« CAS Nos. 526-73-8 108-67-8 95-63-6 25551-13-7 611-14-3 620-14-4 622-96-8 vp (mm HG at 25°C) 1.55 1.55 2.1 2.5 2.99 3.01 Typical % by weight in C9 fraction 8.2 7.6 41.3 9.1 17.4 8.6 Ethyltoluene (mixed isomers)25550-14-5 (D. Bushman, 1991) The C9 aromatic fraction is obtained from the catalytic reforming of crude petroleum. The major components of the C9 fraction ar« th« mixed ethyltoluen« (ET) (ortho-, meta- and para- ET) and trim*thyIbenzene (1,2,3, 1,3,5- and 1,2,4-TMB) isomers which have the same molecular weight of 120.2, boil over a narrow range and are similar in chemical and biological properties. A small percentage of other compounds mak« up the rest of the fraction (Bushman, 1991, pp.1-8). ET and TMB are colorless liquids at room temperature, readily soluble in most organic solvents and relatively insoluble in water. II. BACKGROUND/SELECTION RATIONAL ------- The Tenth Report of the Interagency Testing committee (ITC) (47 FR 22585, May 25, 1982) designated ET and 1,2,4-TMB for priority consideration for environmental and health effects testing. In its Eleventh Report (47 FR 54624, December 3, 1982), the ITC recommended that the other TMB isomers (1,2,3-and l,3,rj-) also be considered for testing. The testing recommendations were based on the exposure potential of the chemicals and the lack of sufficient information on environmental and health effects. Subsequently, the Environmental Protection Agency (EPA) promulgated a final Phase I rule on May 17, 1985 (50 FR 20662) in 40 CFR 799.2175, requiring testing of the C9 aromatic hydrocarbon fraction (C9 fraction), which contains ET and TMB as its major components. On January 23, 1987 a Phase II rule (52 FR 2522) was promulgated issuing final test standards and reporting requirements. Modifica- tions to the rule were made on June 29, 1989 (54 FR 27352). EPA found (l) production and use of the C9 fraction was substantial as a solvent and as a component in motor fuels; (2) widespread exposure exists; (3) existing data are inadequate to predict the effects of exposure; and (4) testing was required to assess the potential of the C9 fraction to cause adverse health and environmental effects. Section 4 test data and evaluations of the C9 fraction were reviewed and commented on by the Health and Environmental Review Division (HERD) of OTS to assess the adequacy of the data for use in risk assessment. Subchronic/chronic toxicity tests, metabolism studies and environmental fate and effects tests were not required because sufficient data existed (FR 20663-20664, May 17, 1985; McCormack, 1990, p.2). III. FINDINGS AND CONCLUSIONS A. PRODUCTION/USB. The 1985 estimate of production volume for C9 fraction solvents was 379 million pounds/year (Rawie, 1991, pp. 1-2). The total C9 fraction manufactured is produced by fewer than 200 facilities (including 190 petroleum refineries), with an annual production of 77,729 million pounds. The primary manufac- turing process is petroleum refining (Pederson, 1991, p.l). The primary use of the C9 fraction, 99.448% of the production volume, is as a gasoline additive. C9 fraction solvents are used in coatings, cleaners, other chemicals and pesticides, printing and inks and other miscellaneous uses (Pederson, 1991, p.2). Data representing 80% of domestic production of C9 fraction solvents showed a median ET/TMB content of 80%, with a range of 75 -90% (FR 20673, May 17, 1985). A wide range of solvents also exists which are adequate substitutes for C9 solvents (Rawie, 1991, p.6). B. ENVIRONMENTAL RELEASES. The C9 fraction is released to the environment in substantial quantities through its use in motor ------- fuels and solvents. only one component (1,2,4-TMB) is a Toxic Release Inventory (TRI) chemical (Pederson, 1991, p.l). C. ENVIRONMENTAL FATE. Most released ET and 1MB partitions to the atmosphere. D. EXPOSURES, (a) Occupational; Exposure to the C9 fraction is from dermal contact with, or inhalation of, vapors. (1) Manufacturing - There are no data on worker exposure to the C9 fraction during manufacturing. However, using toluene in the petroleum refinery industry as a reference, less than 600 workers would potentially be exposed to the C9 fraction in the petroleum refinery industry. Petroleum refining is the primary manufacturing process of the C9 fraction (Pederson, 1991, p.l). Using OSHA monitoring data (4-84 to 4-83) on toluene exposures in certain manufacturing and processing industries, inferences can be made as to exposures to the C9 fraction in those same industries. In general, worker exposures to toluene, and by inference, to the C9 fraction, can be expected to be relatively low. [This is based on the OSHA data on toluene indicating that from 87-100% of the air samples taken from the following manufacturing industries: petroleum refining, plastics materials and resins, cyclic crudes and intermediates, and industrial organic chemicals, were less than % the toluene Permissible Exposure Level (PEL) of 200 ppm.] Exposures to the C9 fraction can be expected to be less than those from toluene due to the C9 fraction's lower volatility. (2) Processing - There are no data on worker exposure to the C9 fraction or to TMB during processing (Pederson, 1991, pp. 2,9). Using the same train of thinking as above, the worker exposures to the C9 fraction during processing can be expected to be relatively low. [This is based on the OSHA data on toluene indicating that from 87-97% of the air samples taken from the following processing industries — plastics materials and resins, paints and allied products, and cyclic crudes and intermediates — were less than S the PEL.] (3) Use - There are no data on worker exposure to the C9 fraction during industrial use. OSHA exposure monitoring data on TMB for workers using C9 fraction solvents are incomplete. There are no data on household worker exposure to C9 fraction solvents used for cleaning and polishing furniture, etc. (b) General Population. There are no data on general population exposure to the C9 fraction from either air or water release. Estimates of worst case inhalation exposure modeling to 1,2,4-TMB, based on the Toxic Release Inventory (TRI) data for populations near 10 air release sites, ranged from 1 to 14 tag/day, with a median of 3 mg/day. The two highest levels were 14 and 6 mg/day (Lynch, 1991, p.3). ------- Estimates of exposure to 1,2,4-TMB by ingestion of con- taminated drinking water (DW) and contaminated fish (FI) were made at 3 sites with the following results (Lynch, 1991, p.4): DW (ua/yr) FI fua/yr) 12.5 46.1 12 44.25 2.03 7.49 (c) Consumer. No data on levels of consumer exposure to the C9 fraction are available. Estimated exposures to 1,2,4-TMB have been calculated assuming that the chemical is used as a solvent or vehicle in consumer paints or solvent-based varnishes at a concen- tration of 50%. Exposures via dermal contact (D) and inhalation (I) are estimated below (Lynch, 1991, p. 5): Consumer Product Route of Exposure Exposure fmcr/yrl fm,g/per event) Solvent-based Varnish D 1.06 x 104 1330 Aerosol Spray Paint D 5.60 x 102 560 Aerosol Furniture Polish I 4.01 x 104 308 Solvent-based Varnish I 7.09 x 102 89 Aerosol Spray Paint I 1.50 x 10* 15 For purposes of risk assessment, dermal absorption is calculated at 10% of exposure; inhalation is calculated at 100%. (d) Environmental. The estimated K values of TMB are indicative of moderate mobility in soil (US EPA, 1987, p. 3). The concern has been in the release of TMB to landfills and possible leaching out of the material. A rule, however, banning releases of TMB to landfills was promulgated under RCRA, 1990 (personal communication with Dave Lynch, April 11, 1991) . Surface water concentrations resulting from the discharges and annual exposure to 1,2,4-TMB are estimated below for three sites (Lynch, 1991, p.4) : Mean flow Low flow fug/ 11 (ppm) fua/11 17.1 0.017 263 0.26 16.43 0.016 15.66 0.015 2.78 0.002 21.66 0.02 ------- B. HAZARDS. The toxicity of the C9 fraction is assessed here as a complete entity rather than by its component chemicals. The C9 fraction test substance was required to have a minimum ET content of 22%, a minimum TMB content of 15%, and minimum total ET/TMB content of 75%. (Page 1 shows the composition of a typical C9 fraction.) The health effects of concern are as follows (HERD, 1991, p.l): Reproductive LOEL: 100 mg/kg/day (100 ppm/inhalation/rat) Developmental LOEL: 210 mg/kg/day (100 ppm/inhalation/mouse) NOEL (Maternal Tox.): 210 mg/kg/day (100 ppm/inhal/mouse) LOEL (Maternal Tox.): 1100 mg/kg/day (500 ppm/inhal/mouse) Aquatic Acute IC50 (daphnid) : Acute EC50 (algae) : 16-day ECSO (daphnid) 1.44 mg/1 (ppm) 0.84 mg/1 (ppm) 1.08 mg/1 (ppm) 0.18 mg/1 (ppo) (TMB) (ET) (both TMB and ET) Human Health Observations - Of 27 workers who had worked for several years with a solvent (Fleet-X-DV99) containing 30% 1,3,5- TMB, 50% 1,2,4-TMB and other hydrocarbons (1,2,2-TMB and 1-methyl- 4-ethylbenzene) some experienced nervousness, tension, anxiety ajid asthmatic bronchitis, as well as hematologic effects (ACGIH TLV's. 4th ED & Suppl. 1980, p. 416, IN HSDB, Pseudocumene; US EPA, 1987). Developmental/Reproductive Toxicity. A C9 fraction inhalation developmental toxicity study in mice established a LOEL of 100 ppm (210 mg/kg/day) (Campbell, 1989, Conclusions). No NOEL was estab- lished. The maternal toxicity LOEL is 500 ppm (1100 mg/kg/day); a probable maternal toxicity NOEL is 100 ppm (210 mg/kg/day) (Camp- bell, 1989 IN HERD, 1991, p.3). A 3-generation C9 fraction inhalation reproductive study in rats showed evidence of parental and developmental toxicity at all doses (100, 500 and 1500 mg/kg/day). The LOEL is 100 mg/kg/day; no NOEL was established (Seed, 1989 IN HERD, 1991, p.2). Neurotoxieity. Adult male rats exposed by inhalation to C9 fraction concentrations of 0, 100, 500 or 1500 ppm (100, 500 or 1500 mg/kg/day) for 13 weeks showed a decrease in body weight by 13%, after the first week, observed in the high-dose group only. No signs of neurotoxicity were evident, as evaluated by motor activity, startle response, forelimb and hindlimb grip strength, hindlimb splay and thermal responses (McCormack, 1990, p.11; Rees, 1989 IN HERD, 1991, p.4). A variety of neur©behavioral effects were observed in the developmental study with mice following C9 fraction inhalation of ------- 1500 ppm (3200 mg/kg/day). The LOEL for neurobehavioral effects was established at 1500 ppm. The NOEL equaled 500 ppm (Rees, 1989 IN HERD, 1991, p.4) . Mutagenicity. The C9 fraction does not induce gene mutations in prokaryotes or in mammalian cells in culture. It does not cause chromosomal aberrations in mammalian cells in vitro or in vivo or DNA effects in mammalian cells (Cimino, 1988 IN HERD, 1991, p.5) . These negative results of the mutagenicity studies did not trigger any additional mutagenicity or carcinogenicity testing,, as specified in the C9 test rule (Cimino, 1988). There are a lack of carcinogenicity test data on the C9 fraction (US EPA, 1986 IN US EPA, 1987, p.7). Acute, subchronic and other Toxicity. Animal Data - No acute toxicity data are available for tha C9 fraction as a whole. The para-ET (p-ET) and 1,2,4-TMB isomers are the best characterized constituent chemicals. Both have low acute toxicity. - Four out of 16 rats died of respiratory failure during a single continuous 24-hour exposure to 2400 ppm 1,3,5-TMB (ACGIH, TLV's 3rd Ed and Suppl 1971-1979, p.269, IN HSDB, Mesitylene). Inhalation exposure to rats of 979 ppm p-BT, 6 hr/day 5 days/wk for 13 weeks resulted in liver enlargement. Gavage administration of 400 or 800 mg/kg/day for 14 days also resulted in liver enlargement (McCormack, 1990, p.4). Rats exposed by inhalation to tha TMB (3-isomer) mixture at 1700 ppra for 10-21 days had no adverse effects in rats. Exposure for 4 months to the same concentration caused diminished weight gain and a progressive increase in lymphopenia and neutrophil with a marked depression of the central nervous system (OP. cit. HSDB Pseudocumene). Ecotoxicity. The calculated log P values on TMB and ET are 4.1 and 4.0, respectively. Both values suggest a high potential for accumulation of these substances in aquatic organisms. Calculated acute aquatic toxicities include (HERD, 1991, p.6): ------- rma/1) 96-hr LC50 (fish): 10.9 (both ET and TMB isomers) 48-hr IC50 (daphnid) : 1.44 (both ET and TMB isomers) 96-hr EC50 (algae): 0.84 (TMB) 1.08 (ET-) 16-day EC.0 (daphnid): 0.18 (both ET and TMB isomers) Chronic algal toxic ity: 0.24 (TMB) 0.34 ppm (ET) Calculated acute daphnid and acute algal values indicate a moderate to high acute toxicity for both groups of isomers. Acute testing using daphnids and algae would be necessary to confirm these estimates of toxicity. Calculated chronic values suggest a low chronic toxicity for both groups of isomers (HERD, 1991, p. 6). F. RISKS. Possible risks to consumers from use of aerosol furniture polish. Possible risks to workers- during manufacturing, process- ing, and during use of C9 as an industrial solvent and as a household end product. Possible risks to aquatic organisms from one release site. IV. REGULATION The current PEL for individual isomers or mixtures of TMB is 25 ppm (125 mg/m3) , with a STEL of 35 ppm (170 mg/m3). V. DISCUSSION (a) Consumer. Health risks to consumers may result from the inhalation of vapors and/or from dermal absorption of the C9 fraction solvent end products. Using a worst case exposure scenario of consumer inhalation exposure to vapors from aerosol furniture polish, the MOEs from the LOELs for reproductive toxicity (100 mg/kg/da) and for developmental toxicity (210 mg/kg/da) from animal studies would be 17 and 35, respectively. Risks to consumers from combined dermal and inhalation exposures to furniture varnish would also give low MOE's (22, and 48 respective- ly) . Since the modeling is based on worst case assumptions, including polishing furniture every 3 days, the risks would not necessarily be to the average consumer. However, using these assumptions there may be risks to persons who clean households as a profession, (see occupational section). Another possible risk to consumers may be from exposure to the C9 fraction when pumping gasoline. This concern is addressed in "SECTION 4 ISSUES" (Section VI., below). ------- (b) occupational! There are no worker exposure data on the C9 fraction. However, exposure levels to toluene in similar indus- tries (including petroleum refining) indicate that most worker exposures during manufacturing and processing would be to concen- trations less than k the PEL. This comparison with toluene is only a gross one since the PEL for toluene is 200 ppm, and that for TMB, the major component of the C9 fraction, is 25 ppm. However, since toluene is more volatile than the C9 fraction, a gross comparison is possible. If C9 fraction exposures were similar to those of toluene, i.e., most of the exposures were < ** PEL for TMB (< % of 25 ppm), workers would be exposed below the LOELs for reproductive and developmental toxicities from animal studies (100 mg/kg/da, and 210 mg/kg/da, respectively). However, there would still be a concern for reproductive and developmental toxicities risk since the margin of exposure from these LOELs would be 8 and 16, respectively. [210 mg/kg/da (Developmental LOEL) * 13 mg/kg/da; and 100 mg/kg/da (Reproductive LOEL) + 13 mg/kg/da.] (*Sea calculations below.) OSHA data on worker exposure to TMB during use as an industri- al solvent is incomplete. It indicates that about 8% of samples taken were less than the detection level. OSHA data on worker exposures to toluene to the same Standard Industrial Codes (SICs) for industrial use indicate that some exposures are > ^ PEL. Therefore, there may also be a concern for reproductive and developmental toxicity risk due to industrial solvent exposures. There is no data on exposure to the C 9 fraction as a solvent in end products during use by professional household cleaners. There may be a health risk to these persons who usa these products on a daily basis (see discussion in consumer section). (c) General Population. There are no data on C9 friction releases to air. However, an estimate of the C 9 fraction exposures can be made using TRI data on 1,2,4-TMB. The TRI data does not tell us if the C9 fraction was released from these sites, and only 1,2,4-TMB was measured, or if only 1,2,4-TMB was released. However, it is known that 1,2,4-TMB is used primarily as a component of the C9 fraction. If the 1,2,4-TMB data can be used as a measure of exposure expected from the C9 fraction, then the margin of exposures (MOE) from the LOEL (lOOmg/kg/da) for reproduc- tive toxicity from an animal study would be 357 and 833, respec- tively for the two sites with greatest releases of 1,2,4-TMB. These MOEs indicate that the expected exposures would be well below the LOELs from the animal study. Given that the modeling of the TRI data uses worst case assumptions, there does not appear to be a health risk to the general population from air releases. Exposure to contaminated drinking water and contaminated fish present no risk, since the estimated levels are below the level of concern. 8 ------- (d) Environmental. There is a concern for potential risk to aquatic organisms during low flow conditions at a Lancaster PA site. Surface water low flow concentrations based on TRI data on 1,2,4-TMB were above a chronic effect level for daphnids. *[TMB PEL=» 25 ppm or I25mg/m3; I25mg/m3 X I,25m3/hr X 8 hr day = 1250 mg/da exposure. Divide by a 50 kg person - 25 rag/kg/da; \ the TMB PEL - 13 mg/kg/da] VI. SECTION 4 ISSUES Although EPA recognized that there may be substantial human exposure to gasoline and its component hydrocarbons, including the C9 fraction, the Agency did not consider exposure to the C9 fraction in gasoline as part of its basis for finding substantial human exposure to the C9 fraction (50 FR 20664, May 17, 1985). EPA justifies its decision by indicating that the C9 fraction was among approximately 300 chemical species in gasoline and the concentrations in a typical motor gasoline are relatively low (approximately 3%) . Also, since existing data showed unleaded gasoline to be carcinogenic in laboratory animal inhalation studies, exposure controls for gasoline were expected to be based on these data or on additional testing of gasoline aimed at characterizing its overall toxicity as a complex product. Data on the C9 fraction alone would be of minimal relevance to that overall determination (50 FR 20666, May 17, 1985). VII. OTHER ISSUES EPA currently has a Petroleum Cluster Policy group. An RM1 meeting is scheduled for May 1, 1991. VIII. RECOMMENDATIONS 1. Notify CPSC of the possible risks to consumers from C9 fraction solvent end products and the need for exposure informa- tion. 2. Bring information to the attention of OSHA, NIOSH, and ACGIH indicating the need for exposure information to verify a possible risk to workers from the C9 fraction. 3. Give information to appropriate EPA Regions to communicate to the industrial site a possible aquatic toxicity concern. 4. Send copy of RM1 information to Petroleum Cluster Policy Group. ------- IX. REFERENCES Bushman, Daniel R. January, 1991. RMl Industrial Chemicals Branch (ICB) Form. Cimino, M. April 1, 1988. Review of Mutagenicity Testing Results on C9 Aromatic Hydrocarbon Fraction. Memorandum from P. Fenner-Crisp, Health and Environmental Review Division (HERD) to R. Troast, Test Rules Development Branch, Existing Chemical Assessment Division (ECAD). Washington, DC: OTS, US EPA. HERD. February 1991. HERD (Health and Environmental Review Division) Profile Working Draft, C9 Aromatic Hydrocarbons. HSDS. Mesitylene. Hazardous Substance Databank. -Pseudocumene. Lynch, David G. April 2, 1991. Preliminary Exposure Assessment for C-9 Hydrocarbons. Memorandum from David G. Lynch, Exposure Assessment Branch (EAB) , to Ethe], Brandt, Existing Chemical Assessment Division (ECAD). McCormack, K., P. Wirdzek, R. Troast, R. Nelson, N. Chandhari and Merrifield. 9/6/90. Summary of C9 Aromatic Hydrocarbon Fraction: Ethyltoluenes, and Trimethylbenzenes. Pederson, Mark E. March 7, 1991. RMl Initial Chemical Engineering Branch (CEB) Summary, Aromatic C9 Hydrocarbons. Rawie, Carol. Feb. 20, 1991. RMl Economic Analysis for C-9 Hydrocarbons. Memorandum from Carol Rawie, Regulatory Impacts Branch (RIB), Economics and Technology Division (ETD), to Ethel Brandt, Existing Chemical Assessment Division (ECAD): Rees, D. Cooper. 1989. Section 4 Testing of C9 Aromatic Hydrocar- bons: Neurotoxicity data. Memorandum from C. Rees, Toxic Effects Branch (TEB), to C. Auer, Health and Environmental Review Division (HERD). Washington, DC: OTS: USEPA. U.S. Environmental Protection Agency. June, 1987. Office of Research and Development (ORD), Office of Health and Environmental Assessment (OHEA), Environmental Criteria and Assessment Office (ECAO), "Health Effects Assessment for Trimethylbenzenes11. 10 ------- RM1 Meeting Summary Final 8/28/91/CG Date: May 22, 1991 Subject: C-9 Aromatics Chairman: Jim Willis Coordinator: Ethel Brandt Supporting Documents: Briefing paper Background Information: The C9 aromatic hydrocarbon fraction is obtained from the catalytic reforming of crude petroleum. The Interagency. Testing Committee recommended that C9 fraction be tested under Section 4 of TSCA, based on the exposure potential and the lack of sufficient information on environmental and health effects. 77,729 million pounds/yr of C9 fraction are produced. The primary use of this volume is as a gasoline additive. Other uses include use as a solvent in coatings, cleaners, other chemical and pesticides, printing and inks and other miscellaneous uses. A wide range of substitutes exists for C9 solvents. Releases of C9 fraction to the environment occur to the atmosphere (via stack and fugitive emissions, and evaporation from spills), to water, and to soil (via landfill waste). There are no data on worker exposure to C9 fractions during manufacture, processing, or use, nor is there any data available on household worker exposure, general population exposure, or consumer exposure to C9 fractions. Worker exposures can be estimated using toluene as a reference. The estimated exposures during manufacture and processing are low. Surface water concentrations resulting from the discharges and annual exposure to 1,2,4-Trimethylbenzene have been estimated, and are listed in the briefing paper. The toxicity of the C9 fraction as a complete entity was assessed by HERD, and the concerns are listed in the briefing paper, along with LOELs and NOELs. In general, concerns were for reproductive and developmental toxicity, as well as a moderate to high acute toxicity to aquatic organisms, and low chronic toxicity to aquatic organisms. There is a possible risk to consumers from the use of aerosol furniture polish. Workers could be at risk during manufacturing, processing and use of C9 as an industrial solvent and as a household product. Risks to aquatic organisms are possible from one release site. ------- Page 2 C-9 Aromatics 8/28/91 Discussion: There has been no verification that C-9 fraction is actually found in any consumer products. It was therefore decided that such verification should be the first order of business. However, it does not seem, prudent to spend a great deal of resources on such verification. ETD will perform a limited survey of manufacturers. If no information can be easily obtained, the search will be discontinued. If it can be verified that the C-9 fraction is in consumer products, then a letter of concern will be written to producers, as well as to CPSC, based on the ethylhexanoic acid model. This letter will notify producers and CPSC of the possible risk to consumers from C9 fraction solvent end products and the need for exposure information. The information will also be given to the appropriate EPA Region for communication of a possible concern for aquatic toxicity for the industrial site. The RM1 information packet will be sent to the Petroleum Cluster Policy Group, as well as the Indoor Air Cluster Group. With the exception of the above mentioned activities and referrals, C9 Aromatic Hydrocarbons will be dropped from further RM1 review. ------- |