United States Environmental Protection Agency Air and Radiation Research Triangle Park, NC 22711 (MD-13) January 1993 EPA 453/N-93-001 Natich ewsletter National Air Toxics Information Clearinghouse In This Issue... Update on EPA's RIHRA Program 5 Region VII Evaluates Open-Path FUR Systems for Air Toxics 6 Frequently Used Acronyms CAA- Clean Air Act Amend- ments of 1990 GACT- Generally Available Control Technology HAPs- Hazardous Air Pollutants HON- Hazardous Organic NESHAP MACT- Maximum Achievable Control Technology NESHAF - National Emission Stan- dards for Hazardous Air Pollutants OAQPS - Office of Air Quality Planning and Standards ~ Rftcyciftd/Recyciable Printed with Soy/CanoJa Ink on paper that contain# at least 50% post-consumer recycled fiber 1 Clean Air Act Activities: EPA Issues New Standards* Recent EPA activity in air toxics includes the promulgation of the Early Reductions rule on December 29, 1992 (57 FR 61970), and two proposed National Emission Standards for Hazardous Air Pollutants (NESHAP) during December 1992. The proposed standards will reduce emissions from the Synthetic Organic- Chemical Manufacturing Industry (SOCMI) and coke ovens. On September 24,1992, EPA also published the schedule for setting standards for source categories. Early Reductions Rule Promulgated The Early Reductions Program** offers facilities a six-year compliance extension from maximum achievable control technology (MACT) or generally available control technology (GACT) standards required by the Clean Air Act (CAA) if they achieve early reductions in emissions of hazardous air pollutants (HAPs). Regulations to implement this rule were proposed on June 13, 1991, and finalized on December 29, 1992. One significant change was made following proposal, based on comments received during the public comment period: the pollutants on the high-risk list were in- creased from 35 to 47; 17 pollutants were added and 5 were deleted (see Table 1). (continued on page 2) Ohio Studies Air Toxics Risk by Paul Koval, David Nuber, and Phil Downey, Ohio Environmental Protection Agency, Division of Air Pollution Control the work. With the technology and methodology available from EPA's UATMP and a rising concern for in- creased cancer risks from multipollutant, multisource interactions in urban areas, Ohio EPA, Division of Air Pollution Con- trol (DAPC), initiated a preliminary sampling program for VOCs throughout the State. (continued on page 3) Produced in STAPIPA / ALAIPCS© conjunction State and Territorial Air Pollution Program Administrators with Association of Local Air Polution Control Officials Ohio EPA has conducted several studies on noncriteria pollutants and their potential impact on human health and the environment. These studies were triggered by the Urban Air Toxics Monitoring Program (UATMP). But until recently, attempts to assess volatile organic compound (VOC) concentrations in ambient air have been hampered by a lack of appropriate equipment to perform f 1 L o '001 6 vy t j v# ------- New Standards (continued from page 1) The primary reason for adding specific pollutants was that EPA revised the methodology used to select the pollutants (described in the preamble to the final rule). Five pollutants were deleted because of revisions to health effects data since the time the rule was proposed [e.g., a change in the reference concentration (RfC,*** see Table 2)]. In addition, the weighting factors for five pollutants were increased to account for bioaccumulation or persistence of the pollutants. The Early Reductions Program cur- rently has 70 enforceable commitments under review, which translates into over 30 million pounds of HAP reductions. For additional information, contact**** David Beck, (919) 541-5421, or Rick Colyer, (919) 541-5262. Hazardous Organic NESHAP (HON) Proposed The Hazardous Organic NESHAP (HON) was proposed on December 31, 1992 (57 FR 62608), and will regulate HAP emissions from the SOCMI source category as well as from seven non- SOCMI processes. Because the SOCMI source category covers so many sources, the HON is expected to result in a larger reduction in emissions than achieved with any other single NESHAP. The HON is expected to reduce HAP emis- sions by more than 500,000 tons each year after it is fully implemented. Major provisions of the HON include specific MACT requirements for five kinds of emission points, parameter monitoring, and emissions trading. Background information on the HON and emissions trading can be found in the July 1992 issue of the NATICH Newsletter. The parameter monitoring provision will be used to demonstrate compliance with the HON's operating requirements. Sources would be required to monitor operating parameters of emission control devices and report periods during which the measured parameter values are out- side certain site-specific ranges. EPA is proposing to allow sources to set site- specific ranges for each control to account for variation in emission point (continued on page 3) Table 1. List of High-Risk Pollutants3 CAS No. Chemical Weighting Factor 53963 2-Acetylaminofluorene 100 107028 Acrolein 100b 79061 Acrylamide 10 79107 Acrylic acid 10 107131 Acrylonitrile 10 0 Arsenic compounds 100 1332214 Asbestos 100 71432 Benzene 10 92875 Benzidine 1,000 0 Beryllium compounds 10 542881 Bis (chloromethyl) ether 1,000 106990 1,3-Butadiene 10 0 Cadmium compounds 10 57749 Chlordane 100h 532274 2-Chloroacetophenone 100 0 Chromium compounds 100 107302 Chloromethyl methyl ether 10 0 Coke oven emissions 10 334883 Diazomethane 10 132649 Dibenzofurans 10 96128 1,2-Dibromo-3-chloropropane 10 111444 Dichloroethyl ether 1° 79447 Dimethyl carbamoyl chloride 100 ™ 122667 1,2-Diphenylhydrazine 10 106934 Ethylene dibromide (Dibromoethane) 10 151564 Ethylenimine (Aziridine) 100 75128 Ethylene oxide 10 76448 Heptachlor 100b 118741 Hexachlorobenzene .a o o 77474 Hexachlorocyclopentadiene 10 302012 Hydrazine 100 O Manganese compounds 10 0 Mercury compounds .e o o 101688 Methylene diphenyl diisocyanate (MDI) 10 60344 Methyl hydrazine 10 624839 Methyl isocyanate 10 0 Nickel compounds 10 62759 N-Nitrosodimethylamine 100 684935 N-Nitroso-N-methylurea 1,000 56382 Parathion 10 75445 Phosgene 10 7803512 Phosphine 10 7723140 Phosphorus 10 75558 1,2-Propylenimine (2-Methyl aziridine) 100 1746016 2,3,7,8-Tetrachlorodibenzo-p-dioxin 100,000 8001352 Toxaphene (Chlorinated camphene) 100 75014 Vinyl chloride 10 ^ 8 Pollutants in boldface were added to this list. b Weighting factor was adjusted from 10 to 100. 2 ------- New Standards (continued from page 2) Table 2. Pollutants Deleted from the High-Risk List CAS No. Chemical Weighting Factor 98077 Benzotrichloride 10 126998 Chloroprene 10 79345 1, 1,2,2-Tetraehloroethane 10 584849 2,4-Toluene diisocyanate 10 75354 Vinylidene chloride (1,1-Dichloroethylene) 10 racteristics and control device igns because available data indicated that establishing minimum or maximum parameter values applicable to all con- trols would be difficult. This provision was included in the HON because of developments in a related program, the operating permits program. For this program, the preamble to the final rule directs EPA to include monitoring in all future regulations to assure continuous compliance with emis- sion standards. This type of monitoring is considered "enhanced monitoring" because it provides data to determine compliance. For the HON, compliance with the numerical emission limits is determined by performance tests, but continuous parameter monitoring is used to determine compliance with the operating requirements. The monitoring data will also allow source owners or operators to certify whether compliance was continuous or intermittent through- out the reporting period as required by the compliance certification provisions of • operating permits program rule. In addition, EPA is developing a separate rule for all sources subject to MACT that will detail the process that sources must follow to develop enhanced monitoring and compliance certifications. Future MACT standards will include enhanced monitoring and compliance certification provisions as appropriate for the types of controls being required. For further information on the HON proposal, contact Daphne McMurrer, (919) 541-0248. For additional informa- tion on enhanced monitoring or com- pliance certifications, contact Jane Engert (703) 308-8677, Stationary Source Compliance Division, Mail Code EN-341W, U.S. EPA, 401 M Street, S.W., Washington, D.C. 20460. Coke Oven Standard Proposed Another MACT standard was pro- posed on December 4, 1992 (57 FR 57534), which is expected to reduce coke oven emissions by approximately 1500- «0 tons/year when implemented. At current level of control, coke oven batteries release approximately 1826 tons/year of coke oven emissions. These emissions, which contain polycyclic organic matter, benzene, and other chemicals than can cause cancer, are among the most toxic of all air pollutants. The coke oven standard was developed through regulatory negotia- tion, in which representatives from State and local agencies, industry, environ- mental groups, and other interested par- ties participate in EPA's decision-making process. For further information, contact Amanda Agnew, (919) 541-5268. Source Category Schedule Proposed The source category schedule for emission standards was proposed on September 24, 1992 (57 FR 44147). This schedule dictates when standards are due for each category on the initial list of Ohio's Approach Summarized In the summer of 1990, DAPC con- ducted field trials of its VOC canister sampling equipment. By the summer of 1991, the initial program had been expanded to include an increased number of samples collected at existing sites and sampling locations; this sampling was conducted in a complementary manner to the 1990 sampling. The 1990 and 1991 data were combined and used in risk assessments for sites with a sufficient sampling size. A total of 70 SUMMA® canister samples were collected in 14 counties during several different 24-hour time periods. The sampling locations ranged from areas of light industry to urban source categories. A 30-day public com- ment period ended on October 26, 1992, and 18 comments were received. For additional information, see the Federal Register notice or contact Chuck French, (919) 541-0467. *See related articles in the July 1992 issue. * *For background information, see also the July 1991 issue of the Newsletter. * * *See related article in this issue. *** *All EPA contacts for the standards discussed in this article can be reached at the following address: U.S. EPA, Office of Air Quality Planning and Standards (OAQPS), MD-13, Research Triangle Park, North Carolina 27711. land use. The samples were analyzed for 41 nonpolar VOC target compounds. Of the 41 target compounds, 29 were identified and measured in the samples collected (see Table 1). Only two com- pounds, benzene and toluene, were iden- tified in every sample. Three compounds, 1,1-dichloroethene, 1,2-dichloroethane, and benzyl chloride, were only found once in the 70 samples. Risk Appears Higher for Two Cities For each site, the criterion for con- ducting a risk assessment was the availability of monitoring data for at least 5 days of a 365-day period. Sufficient data were available for three of the sampling (continued on page 4) Ohio Studies Air Toxics (continued from page 1) 3 ------- Ohio Studies Air Toxics (continued from page 3) locations: Cleveland, Columbus, and Cin- cinnati. For the assessment, only the inhalation route of exposure to the com- pounds was evaluated. Carcinogenic com- pounds were evaluated using the U.S. EPA guidelines for carcinogenic risk assess- ment. Noncarcinogenic compounds were evaluated based upon the U.S. EPA inhala- tion reference concentration (RfC) * meth- odology for those compounds for which RfCs were available. For the exposure assessment component of the risk assess- ment, the average ambient air concentra- tions measured at each site were used. The aggregate carcinogenic risk for the monitoring locations in Cleveland and Columbus fell within the acceptable range of 10' and 10''. The aggregate car- cinogenic risk for the Cincinnati monitor- ing site appears to be at the upperlimit of the acceptable range, although high con- centrations of one particular compound tended to drive the results of the risk assessment. The aggregate percentages of the RfC, a measure of the noncancer risk, for Cincinnati and Colum- bus were below 100 percent. In Cleve- land, where concentrations of p-dichloro- benzene drove the percentage of the RfC above 100 percent, DAPC is continuing monitoring to determine why the concen- trations for this compound are so high. Although the ability to increase the number of actual sampling sites will be determined by budgetary constraints, DAPC plans to continue the existing monitoring program. Ohio EPA plans to conduct a more comprehensive ambient air VOC monitoring program and more accurate and detailed urban air toxics risk assessments. For a copy of the full report of this study with details on sampling, analysis, and results, please write David Nuber, Toxicologist, Ohio EPA, Division of Air Pollution Control, 1800 Watermark Drive, Columbus, Ohio 43215. The length of the report necessitates that Ohio EPA charge a $15 fee. Please make checks payable to Treasurer, State of Ohio. (It is available to Federal, State, and local agencies at no charge.) For more information, call David Nuber at (614) 644-2270. *See related article in this issue. Table 1. Compounds That Have Been Measured in Ambient Air Samples Collected in Ohio" Frequency Maximum Minimum Observed in Concentration Concentration Compound 70 Samples Measured (jAg/m3) Detected (n.g/m3) Benzene 70 145.59 0.52 Toluene 70 73.92 2.20 Trichlorofluoromethane 69 38.37 0.48 m + p-Xylene 69 57.55 0.45 1,1,1-Trichloroethane 68 49.91 0.63 Dichlorodifluoromethane 67 206.51 0.63 1,2,4-Trimethylbenzene (i5 18.22 0.62 o-Xylene 65 21.14 0.49 Ethylbenzene 63 15.84 0.48 Dichloromethane 50 77.68 0.42 Tetrachloroethene 49 72.90 0.80 Styrene 46 39.44 0.42 Carbon tetrachloride 44 0.93 0.62 4-Ethyl toluene 43 7.86 0.53 Methyl chloride 43 56.50 0.35 1,1,2-Trichloro- 1,2,2-trifluoroethane 41 14.70 0.77 1,3,5-Trimethylbenzen<: 40 15.20 0.52 p-I)iehlorobenzene 29 26.80 0.61 Trichloroethene 20 17.03 0.55 ^ 3-Chloropropene 14 1.28 0.30 T richloromethane 14 12.00 0.49 Chlorobenzene 10 6.19 0.46 m-Dichlorobenzene 5 3.79 1.56 1,2,4-Trichlorobenzene 5 21.30 0.89 o-Dichlorobenzene 3 2.27 0.64 Methyl bromide 2 62.50 57.50 Benzyl chloride 1 0.83 0.83 1,2-DichIoroethane 1 1.09 1.09 1,1-Dichloroethene 1 0.94 0.94 "Target compounds not detected were the following: vinyl chloride, ethyl chloride, 1,1-dichlor- oethane, 1,1,2,2-tetrachloroethane, 1,2-dichIoro-l, 1,2,2-tetrafluoroethane, cis-l,2-dichloroethene, 1,2-dichloropropane, cis-l,3-dichloropropene, trans-l,3-dichloropropene, 1,1,2-trichloroethane, 1,2-dibromomethane, and hexachlorobutadiene. Readers Take Note The correct OSHA exposure limit for lead is 50 \ig/m3 (8-hour time- weighted average) The correct EPA Reference Air Concentration is 0.09 pg/m3 (quarterly average), The values appearing in the last footnote to the , "Lead Contamination Hazard" article on page 3 in the November 1992 * issue were incorrect. The Newsletter staff apologizes for any inconvenience. 4 ------- Update on EPA's RIHRA Program* Risk assessment plays a fundamen- icu role at EPA, both in establishing priorities for standards required by the 1990 Clean Air Act Amendments (CAA) and in developing regulations to protect public health. Risk assessment, as defined by the National Academy of Sciences, is the scientific characterization of the potential adverse effects of environmen- tal hazards on human health. The CAA requires that the technology-based stan- dards now being developed for hazard- ous air pollutants (HAPs) be followed by an evaluation of the residual health risks remaining after compliance with these standards eight or nine years after pro- mulgation. Deferring risk assessments for HAPs allows not only evaluation of the risk assessment process, but also a window of opportunity to perform critical research to address the major uncertain- ties affecting risk assessments. In 1988, KPA's Office of Research and Development (ORD) established the Research to Improve Health Risk Assessments (RIHRA) Program. The «HRA mandate is to conduct systematic, egrated, long-term research, rather than to conduct short-term research in response to near-term regulatory pro- gram needs. The RIHRA program is targeted to improve the process of risk assessment while complementing other EPA research programs that respond directly to media-specific program needs (e.g., air and water). The intent of the program is to identify and conduct research leading to improved risk assessment procedures, primarily through development of generic or model compound-based risk assessment models. RIHRA research focuses on the major factors affecting health risk assessment: estimating human ex- posures to a pollutant; estimating the concentration and persistence of that pollutant at critical target sites in the body (e.g., pharmacokinetic modeling); linking pharmacokinetic and mechanistic data to generate scientifically sound risk •essment models (e.g., biologically ed dose-response modeling); and characterizing uncertainty. This article, the first in a series, focuses on current activities supported by RIHRA and other studies affecting the methodology for deriving reference doses (RfDs) and concentrations (RfCs). Subsequent articles will report on RIHRA research on benchmark-dose modeling, which affects use of the no-observed-adverse-effect level (NOAEL) in calculating RfDs; research leading to an improved scientific founda- tion for physiologically based pharmaco- kinetic modeling; and research on derivation of parameters of the Moolgavkar-Knudson cancer incidence equation, which may significantly affect cancer risk estimation at low doses. These activities will advance the scien- tific basis for risk assessment and lead to improvements in risk assessments for all regulatory programs. Many EPA health effects studies, including research supported through RIHRA and other ORD research pro- grams, in pharmacokinetics and dosimetry, neurotoxicity, reproductive toxicity, developmental toxicity, pulmonary toxicity, and in statistical analyses affect or will affect RfD and inhalation RfC methodologies. Eor health endpoints other than cancer, the Agency currently uses an RfD for oral exposures and an RfC for inhalation exposures. In either approach, critical health data are evaluated and NOAELs or lowest- observed-adverse-effect levels (LOAELs) are identified. The RfC methods also dosimetrically adjust these levels for interspecies differences. These levels are divided by one or more uncertainty fac- tors to derive an RfC or RfD from experi- mental data. These uncertainty factors are intended to account for: 1. the variation in sensitivity among members of the human population; 2. the uncertainty in extrapolating animal data to the case of humans; 3. the uncertainty in extrapolating from data obtained in a study that is of less-than-lifetime exposure; 4. the uncertainty in using LOAEL data rather than NOAEL data; and 5. the inability of any single study to adequately address all possible adverse outcomes in humans. Historically, most uncertainty fac- tors have been set at ten. Current research to support or modify these uncertainty factors will directly affect the estimation of RfCs and chemical-specific risk assessments. Ultimately, this research will affect risk management decisions. RIHRA studies supplement and build upon results from a host of prior studies designed to improve RfD/RfC methodology. In general, all of these studies seek to improve or replace the uncertainty factors used when deriving RfCs and RfDs. Current research is underway, for example, to study regional dosimetry in the pulmonary tract of humans, laboratory rodents, rhesus monkeys, and in physical models. As a result of these critical studies of dosimetry, the EPA's RfC Workgroup has modified the interspecies uncertainty factor from 10 to 3 when dosimetric adjustments are applied. Additional RIHRA-supported activities are designed to improve understanding of the toxicological data upon which RfCs and RfDs are based, thereby reducing or improving the data base uncertainty factor. Other critical factors in RfC/RfD methodology under study include evaluation of less-than- lifetime exposures and variability in human and laboratory populations. Another important activity is the iden- tification of critical health studies to improve the foundation of RfC/RfD methodology. RIHRA-supported research leading to improvements on all these fronts supports related research activities, including the development and evalua- tion of other methods to obtain data on risks above the RfD/RfC. These methods include categorical regression and benchmark-dose modeling approaches. Future articles will provide addi- tional details on several of these activities. For further information on RIHRA, contact John Vandenberg, Ph.D., RIHRA Director, EPA Health Effects Research Laboratory, MD-51, Research Triangle Park, North Carolina 27711, or call (919) 541-4527. *See related article in the July 1991 Newsletter. 5 ------- Region VII Evaluates Open-Path FTIR Systems for Air Toxics* by Jody Hudson, Region VII Table 1. OP-FTIR Qualitative Performance Summary Results As Percent of Tune VOCs Were Correctly Identified Chemical Specie Cone. Range (ppb)a No. of Releases OP-FTIR Systems Tested System A <*) System B (%) System C (%) Dichloromethane 130 - 230 6 100 100 100 1,1,1-T richloroethane 33 - 93 6 100 100 100 Trichloroethylene 255 - 330 3 100 100 100 Tetrachloroethylene 65 - 97 3 100 100 100 Freon 113 46-76 3 100 100 100 Chlorobenzene 29-78 3 100 100 67 Toluene 34 - 100 3 0 33 67 IsoOctane 34 - 110 ti 0 100 0 '"'Average concentration across 100-meter path as determined from canister data. EPA Region VII and University of Kansas researchers have completed a field study to assess and document the analytical capabilities of several open- path Fourier transform infrared (OP- FTIR) spectrometer systems. The study, which was largely funded by the EPA's Air/Superfund Coordination Program, was conducted to answer questions from the air monitoring community about the application of OP-FTIR as an emerging technology for measuring toxic air pollutants in ambient air. Because of its many advantages over traditional "point monitoring" techniques, there is signifi- cant, widespread interest both within and outside the EPA in the use of OP- FTIR as an air toxics monitoring tool. However, many basic questions con- cerning analytical capabilities have not been answered or supported by well- documented data, so many would-be OP-FTIR users are reluctant to apply this technology. These questions, related pri- marily to the systems' analytical qualitative and quantitative performance, include: 1. how well can the OP-FTIR systems identify specific pollutants; 2. how accurate are the concentration measurements; and 3. how reproducible are the measure- ments? The study was designed to answer these and other questions regarding OP-FTIR performance. The study design involved conduct ing a series of 15 controlled blind releases (specific compounds and concentrations unknown to operators) of a range of tox- ic air pollutants. As these compounds were released to the air, three different OP-FTIR systems were operated side by side to conduct concurrent measurements through the resulting plume. The OP-FTIR measurements were taken approximately 50 meters downwind of the release source with a straight line path-length of 100 meters through the plume. Concurrent canister samples were collected with EPA Method TO-14** to provide a reference for the OP-FTIR data. Nine stainless-steel canisters were deployed every 10 meters across the path to characterize the average concentration across the path- length for each controlled release. The network schematic is shown in Figure 1. The study results (Table 1) showed that all three OP-FTIR systems demon- strated excellent qualitative performance for the chlorinated VOCs released. Quantitative performance was measu^ in terms of both accuracy and precision relative to the canister results. (continued on page 7) Figure 1. OP-FTIR Performance Study Network Design 6 ------- Region VII (continued from page 6) Figure 2. OP-FTIR Quantitative Performance Summary for Accuracy Mean Accuracy as Percent of Reference Value 250- 225 200- 175- 150- 125- 100- 75 50 25 0 [ j System A [] System B ! System C As shown in Figures 2 and 3, accuracy was best for the halogenated volatile organic compounds (VOCs), which ranged from 73 to 120 percent (expressed as percentage of canister value). The preci- sion was also good with most results being 10 to 15 percent of the relative standard deviation. A statistical data analysis showed no difference at the 0.05 significance level (0.05a) between the measurement data generated by the OP-FTIR systems and Method TO-14 for the aliphatic chlorinated VOCs released. A more detailed discussion of this study was presented at the June Air & Waste Management Association's national meeting in Kansas City. A copy of this report can be obtained from Jody Hud- son or Mark Thomas at (913) 551-5000. *See related article in the September 1991 Newsletter. * *EPA Method TO-14 uses SUMMA® passivated canisters for ambient-level VOC collection and is followed by cryogenic preconcentration and gas chromatography/ mass spectrometry analysis. Figure 3. OP-FTIR Quantitative Performance Summary for Precision Precision as RSD% \sSS, [] System A ^System B 1! System C v\VV \ 's \ \ Attention Readers The NATICH Bulletin Board System is now available on the QAQFS Technology Transfer Net- work (TTN). The number to access the TTN is (919)541-5742. Necessary Communications Settings: 8 Data bits No parity 1 Stop bit Full duplex Terminal emulation - VT100, VT102, or ANSI For more information, contact Vasu Kilaru, (919)541-5332. 7 ------- Tlic NAT1CH Newsletter is published six times a year by the National Air Toxics Information Clearinghouse. The Newsletter is prepared by Radian Corporation under KI'A Contract Number 68-1 >1-0125, Work Assignment 2-11. The KPA Editor is Carol Jones, EPA Office of Air Quality Planning and Standards. Research Triangle Park, North Carolina 27711, Telephone: (910) fvl 1-fxVll. 'Hie Radian Project Director is Linda Cooper, Radian Corporation, P. (). Box 115000, Research Triangle Park, North Carolina 27709, (919)541-9100. 'Hie Newsletter is distributed tree of charge. To report address changes, write Meredith Haley, Radian Corporation, P. O. Box 13000, Research Triangle Park, North Carolina 27709. The views expressed in the NATICH Newsletter do not necessarily reflect the views and policies of the Environmental Pro- tection Agency. Mention of trade names or commercial products does not constitute any endorsement or recommendation lor use by KI'A. United States Environmental Protection Agency Pollutant Assessment Branch Research Triangle Park, NC 27711 Official Business Penalty for Private Use $300 BULK RATE Postage and Fees Paid P.P. A. EPA 453/N-93-001 January 1993 ------- |