April 1985 DIOXIN EMISSIONS FROM INDUSTRIAL BOILERS BURNING HAZARDOUS MATERIALS by C. Castaldini Acurex Corporation Energy & Environmental Division 555 Clyde Avenue P.O. Box 7555 Mountain View, California 94039 EPA Contract No. 68-03-3241 Project Officer: Robert A. Olexsey Research Program Manager Alternative Technologies Division Hazardous Waste Engineering Research Laboratory Cincinnati, Ohio 45268 for Hazardous Waste Engineering Research Laboratory US. ENVIRONMENTAL PROTECTION AGENCY Office of Research and Development Cincinnati, OH 45268 ------- April 1985 DIOXIN EMISSIONS FROM INDUSTRIAL BOILERS BURNING HAZARDOUS MATERIALS by C. Castaldini Acurex Corporation Energy & Environmental Division 555 Clyde Avenue P.O. Box 7555 Mountain View, CA 94039 EPA Contract No. 68—03—3241 Project Officer: Robert A. Olexsey Research Program Manager Alternative Technologies Division Hazardous Waste Engineering Research Laboratory Cincinnati, Ohio 45268 for Hazardous Waste Engineering Research Laboratory U.S. ENVIRONMENTAL PROTECTION AGENCY Office of Research and Development Cincinnati, Ohio 45268 ------- NOT ICE This document has been reviewed In accordance with U.S. Environmental Protection Agency policy and approved for publication. Mention of trade names or comercial products does not constitute endorsement or recomendation for use. ------- FOREWORD Today’s rapidly developing and changing technologies and industrial products and practices frequently carry with them the increased generation of solid and hazardous wastes. These materials, if improperly dealt with, can threaten both public health and the environment. Abandoned waste sites and accidental releases of toxic and hazardous substances to the environment also have important environmental and public health implications. The Hazardous Waste Engineering Research Laboratory assists in providing an authoritative and defensible engineering basis for assessing and solving these problems. Its products support the policies, programs, and regulations of the Environmental Protection Agency, the permitting and other responsibilities of State and local governments and the need of both large and small businesses in handling their wastes responsibly and economically. This report describes the results obtained from laboratory analyses of flue gas samples from industrial boilers burning hazardous waste materials. The purpose of the analyses was to determine if detectable quantities of polychlorinated dibenzo—p—dioxlns (PCDD) and polychiorinated dibenzo—furans (PCDF) are emitted from such boilers when hazardous waste materials are burned. This information will be useful to regulatory officials who are responsible for setting standards, to permitting officials who may encounter questions about dioxin emissions during a permitting process and to members of the technical community who are seeking safe methods for hazardous waste disposal. For further information, please contact the Alternative Technologies Division of the Hazardous Waste Engineering Research Laboratory. David G. Stephan, Director Hazardous Waste Engineering Research Laboratory iii ------- ABSTRACT Laboratory analyses for polychiorinated dibenzo—p—diOXiflS (PCDD) and polychlorinated dibenzo—furans (PCDF) were performed on waste fuels and stack gas emission samples from five Industrial boiler test sites cofiring liquid hazardous waste fuels. Analytical results Indicate that, apart from creosote sludge, chlorinated wastes were void of PCDD and PCDF homologs at detection limits in the range of 0.045 to 4.7 ppb. Creosote sludge cofired with wood waste In a stoker boiler was found to contain 7.4 ppm of total dioxins, primarily hepta and octa homologs. Stack gas concentrations of PCDD were highest for the creosote wood-fired stoker at about 75 ng/m 3 . Other PCDD results indicate concentrations ranging from below detection levels (0.0022 to 0.019 ng/m 3 ) to a maximum of 1.1 ng/m 3 . PCDF concentrations were generally higher with total furan levels up to 5.5 nglm 3 . No 2,3,7,8 —tetrachlorodibeflZOi,diOXin was detected in any waste fuel. While the 2,37,8—TCDD isomer was detected in the flue gas emissions from one boiler, the measured value for the emissions of 23,7,8—TCDD was equal to the detection limit for the compound in the particular flue gas (0.002 ng/m 3 ). Iv ------- Forward . . . . . . . . . . . . . . . . Abstract . . . . . . . . Tables . . . . . . . . . . . . . Introduction . . . . . . 1.1 Program Objective Test Program Description 2.1 Test Description . . . . . . . . 2.2 Analysis Protocol Results Interpretation and Conclusion . . . . References . . . . . . . . . Appendix A —— Battelle Columbus Laboratory Analytical Report . . . • . . . • . . I I 1I iv • I I I I • • • I vi • . I I 1 • 1 . 6 • . . . . . . • • 6 8 10 16 19 . • . . 20 CONTENTS 1 2 3 4 V ------- TABLES Number Page 1 Summary of Test Site Boilers . . 2 2 Summary of Test With Waste Fuel Firing 3 3 Summary of Test Average DRE’s for Volatile POHC’s 4 4 Test Sites Selected for PCDD and PCDF Analyses . . . . . . . 7 5 Dioxins Concentrations in Waste Fuels . . 11 6 Furan Concentrations in Waste Fuels . . . . . 12 7 Flue Gas Concentrations —— Dioxins . . . . . . . . 13 8 Flue Gas Concentrations —— Furans 14 9 QA/QC Results for Flue Gas Sample Extracts . . 15 10 Site A Dioxin DRE’s . . . 17 11 Mass Flow Rates of Total Dioxins and Furans . 18 vi ------- SECTION 1 INTRODUCTION In 1982, the EPA ’s Hazardous Waste Engineering Research Laboratory in Cincinnati initiated a series of field test programs on industrial boilers cofiring hazardous wastes. A principal objective of these tests was to develop a data base on the destruction and removal efficiencies (DRE) of industrial boilers co—firing hazardous organic components in waste fuels. This data base was required by the EPA’s Office of Solid Waste and Emergency Response to evaluate hazards and potential control measures in preparation for recommending regulations controlling ongoing industry cofire practices. Eleven boilers were tested under typical industrial operating conditions. The boiler specifications for these 11 tests are listed in Table 1. Waste specifications and additional boiler specifications are listed in Table 2. The majority of the tests were on gas— and oil—fired watertube boilers which correspond to the predominant cofiring application in industry. The wastes at Sites E through K were spiked with carbon tetrachloride and, usually, chlorobenzene and trichloroethylene. This was done to extend the range of volatile POHC’s to be monitored and to provide a common basis for intra—boiler comparisons. Table 3 summarizes the ORE results for volatile principal organic hazardous compounds (POHC’s). With the exception of Site F, site—average ORE’s exceeded 99.99 percent. The mass—weighted volatile organics DRE for all test sites was 99.998 percent. Additional results for the Site A stoker, however, indicated DRE’s lower than 99.99 percent for several semivolatile POHC’s including pentachlorophenol and polynuclear aromatics in the creosote sludge. Generally, lower DRE’s were attributed to nonsteady or transient combustion conditions. Recently, additional tests were performed to measure volatile POHC ORE’s and volatile PlC emissions during nonsteady and transient boiler operations typical for these combustion devices. These tests, documented in the Site L report (Ref. 2), indicate generally good destruction efficiency (ORE average 99.998 percent) under most nonsteady and off—specification combustion conditions investigated. 1.1 PROGRAM OBJECTIVE The objective of this program was to perform laboratory analyses of boiler flue gas samples to determine concentrations of polychlorinated I ------- TABLE 1. SUMMARY OF TEST SITE BOILERS (Source: Ref. 1) 9 Coe , 0rt,d ,0tsrbe 0 bdi?l ,l psct. d 71,, U 9 1.14 ..cbd ,eo oetl.llp flood 30 t o r tubS I PucospI 30br 1 3 1101 17_I 1613) 1% (1.141) 1.1 (0.07 I I (317 SO (53) 20 (230) 3 ? ? 11 1.030) (70 (.530) 11.6 (10) St ((.130) I SO (1.110) i cO . .oub iIud . 61usd wIth o.od I lIlpi 30u b , I I , utowltid oil qus • P$seolic ousou be or be lbs. .towIz.d bso.iu I btbo.1 sod be of Ifs bI , 0 , 0 30085 i. O.U 18 ,0 p11 11 chiusiusUd .1.11.4 borO... orgs .lcl I bthyl tbcrpIsb TO. ibem obeIrW bpp ,.dwct usuthe ouib ‘l to borus, tIwo.t I Pilot solusosti L ? lbs• stowliod oil burus ’ I IIØlp t61 .rI30Ud Sooflubl• S I, .1.1 red oil 11 co.l. boWl susbW be or be 6 oIl ,0Ibu fu,l 58 .1164 eli 0 bIll.. ouub hiqb ((If.. o.r or 1. sitrul, 8q i .l(i io. It,.. •to. 18d berth? — I ll?lci.I3p ArsHibi . oil bb,. d foul. berow, Uo 6 .51 1 Sl,. d oust, slib hoed with bus,, light oil .51 Opolt. 131 (97 ip.,.iIo. bliicyclouw !l.cbstI..q 1 3 04$. c $tIua ii , end .0,18 75,4 Low boll.. bowl b. 1 oulbu firs rub .9 4? .1!. (40 qpef fsr to. loud sills reduo.d .r 66 bsusro. Ul ii •IC4$ l sir. bus. bout 00 D5v’9wt C40.cit7 with 3 or I b ,r .s ,u I. , 0r.lc, be. Port loud .45* .1 7 10 .7?, (240 qph) o.ib 91.1.., rut, for 10 54$ sbe30 SO busu Purl loud with ) 130 . 1 1 1 (I S O 30h) usut, fioln Pit, Is.’ losdo show. SO porc , .t ier s.u P. I l d 018 ,8P 30 I. 05555$ Oso.11 IIi. 1097 cClo?l , 0 to 00 ...cowt o ouu8 7,07. SOP St boll. ’ tspscltp with o..i 460 .1/, (410 ,ehl oult, firing ‘It, bus SIaqed c u,tlo. for I. 10, with . ,f (JO •th ((20 qoh) slut, flow Yvette? , .i •Ir of I? or,’c,.t bus Typici I 70110 porcoot buo ,p sad llqht oIl .1. . 318 bilIetyp. £ btu’tub . slob., I Puctipud fire S ft.ld-.r.ctid . 5 8 ,1.6 , o Fl.id.i,cb4 cusoortod ou#r itat. , £ PIC6 Ip.d .u bes Obi. 0gb copucity. 1IO 1Sf’,’) Forest, .gl . . (ft ) .“ if F 5,us9 b,o.It b.p ’us.rl iy’fucq. Ilniocti . . . ‘ IIt I Priusry lou) ports) 7751141 3018 fouls IsiJ ocito . outbuoto. bed oust, btm’.l pus b8,Sl U I or oil 50. 6 oil 7.1 (60 ) 31 (3.310) 130 I I 6 IllS) 67 (1.450) 664 (7.160) 30. 6 oil 11. 1001 So. 6 .13. or 7,305,0 0.0 (*0) 6.1 (tIll 20 (270) Utuse 3? (3001 020 1 1 5 .4th) 011 7.5 1571 J Pichipud Pl , , . I Puctapud 308?bes 1.030) P ,lw.,i,,d to sI 3510) b8..1 pus Ii (1.630) 76 1.3 (IC) 7.0 II I) LI (PIP 3.6 (60) 60 (2.270) 67 13301 ------- TABLE 2. SUMMARY OF TEST WITH WASTE FUEL FIRING (Source: Ref. 1) CA) Site Ilunt er of waste—fuel— fired tests Volumetric heat release rate, kW/m 3 (10 Btu/hr—ft 3 ) Surface heat release rate, kW/m 2 (10 Btu/hr—ft 2 ) Bulk furnacea temperature, C ( F) Bulk furnacea residence time, sec Waste fuel heatinq value, NJ/kg (10 Btu/lb) Waste heat input, percent of total POHC’s of interest A 4 300 (29) 48 (16) 1.370 (2,500) 1.2 39 (17) 40 Phenol, pentachlorophenol, naphthalene fluorene. 2-4-Dirnethyiphenol B 3 745 (72) 106 (34) 1.320 (2.400) 0.8 0.03-0.18 (0.013—0.017) (1 Toluene C 3 78 (1.5) 150 (48) 1.320 (2.400) 2.0 39 (17) 38 Phenol 0 3 400 (39) 180 (57) 1,430 (2,600) 1.1 21 (8.8) 18 Tetrachloroethylene 3 230 (22) 100 (33) 1.370 (2.500) 1.3 42 (18) 48 BIs(2—chloroethyl)ether, toluene 1 580 (55) 37 (11) 1,550 (2,800) 0.7 27 (12) 22 Piethylmethacrylate, ohydroxy methyl I sobuty-ra to and a-hydroxy I sobutyrate methyl ether 6 380—710 (37—14) 24—49 (7.6—15) 1,480-1.590 (2,700—2.900) 0.5—1.0 25-27 (11—12) 19—43 Above plus carbon tetrachioride, chlorobenzene and trichloroethylene 1 420 (40) 26 (8.1) 1,480 (2,700) 1.1 37 (16) 56 Toluene, methylmethacrylate F 3 114 (11) 104 (34) 1,370 (2.500) 2.0 33 (14) 9.0 Carbon tetrachloride, chlorobenzene, trichloroethylene, toluene G 3 820 (79) 262 (81) 1.350 (2.450) 0.4 21 (9.0) 100 Carbon tetrachioride, epichiorohydrin, bi s( 2-chiorol sopropyl )ether H 3 180 (17) 183 (58) 1,310 (2,500) 2.0 17 (7.0) 2.4—4.3 Carbon tetrachloride, chlorobenzene. 1 • 1, 1—Trichioroethane I 2 340 (33) 180 (57) 1,430 (2,600) 1.8 25 (11) 8.2 Carbon tetrachloride, chlorobenzene, trichloroethylene, toluene, aniline, benzene. ni trobenzene J 6 690—1.750 (65—170) 118-300 (37-95) 1.310-1.370 (2,400-2,500) 0.3—0.7 42 (18) 100 Carbon tetrochloride, chlorobenzene, trichloroethylene, toluene K 1 270 (26) 370 (120) 1,370 (2,500) 1.8 40 (17) 65 Carbon tetrachloride, chlorobenzene. trichloroethylene, toluene, benzene 5 Not measured values. Estinmtes approsimate. of bulk gas temperature in the furnace were used to calculate bulk furnace residence time. Values to be considered ------- TABLE 3. SUMMARY OF TEST AVERAGE DRE t S FOR VOLATILE POHC’sa (Source: Ref. 1) ---- - m - - Weigti ted POHC Site B Site 0 Site E Site F Site G Site H Site I Site J Site K Range average Carbon tetrachiorlde 99.9990— 99.98— 99.995- 99.97— 99.9990— 99.997— 99.91— 99.9992 99.9998 99.9990 99.9990 99.9994 99.9993 99.9998 99.9998 (99.9996)b (99.995) (99.998) (99.98) (99.9993) (99.9990) 99.9998 Tr1ch oroethy1ene 99.994— 99.98- 99.99990- 99.998— 99.98 99.9994 99.9995 99.998 99.99992 99.99993 99.99993 (99.998) (99.996) (99.99991) (99.9996) 99.99990 1 1,1—TrIch1oroethane 99.97— 99.97 99.994 99.9996 99.9996 (99.994) Chlorobenzene 99.995- 99.96— 99.990- 99.997 99.8- 99.9 99.992 99.99990 99.992 99.997 99.9990 99.97 99.99992 (99.998) (99.98) (99.992) (99.990) (99.95) 99.99992 Benzene 99.91— 99.91— 99.990 99.98 99.996 (99.97) 99.996 Toluene 99.9992— 99.90— 99.9990— 99.90— 99.998 99.99990 99.97 99.9997 99.99996 99.991 (99.9996) 99.997 199.95) 99.998 (99.9990) 99.99996 Tetrachioroethylene 99.994- 99.994— 99.998 99.9992 99.9992 (99.998) Methylmethacrylate 99.95— 99.95— 99.991 99.997 99.995 (99.991) Mass weighted average 99.991 99.994— 99.95— 99.90— 99.995- 99.97— 99.97— 99.8- 99.996- 99.8— 99.998 99.99990 99.99990 99.9990 99.9990 99.9996 99.99992 99.99993 99.99996 99.99996 (99.9993) (99.995) (99.98) (99.998) (99.991) (99.998) (99.9990) (99.9991) aEach test average DRE Is general )y based on the weighted average of tr (p 1cate measurements. bNu,,ters In parentheses represent the site—specifIc POHC average ORE. ------- dibenzo—p—dioxins (PCDD) and polychiorinated dibenzo—furans (PCDF) including the 2,3,7,8 tetra isomer for selected boiler sites. In addition, analyses of waste fuels cofired at these selected test sites were also performed to determine whether PCDD and PCDF emissions could be attributed to waste fuels or to products of combustion of waste fuels. 5 ------- SECTION 2 TEST PROGRAM DESCRIPTION 2.1 TEST DESCRIPTION Table 4 lIsts test sites selected for PCDD and PCDF analyses. At Site A, four cofire tests were performed by burning a mixture of wood waste (chips, bark, and sawdust) with creosote sludge. The sludge contained phenolic coupounds including pentachiorophenol, generally identified as a precursor to dioxin emissions during combustion. Heat Input attributed to the creosote sludge was estimated at 40 percent for these tests. Operation of the stoker at Site A was erratic with large and frequent fluctuations in excess air and CO emissions. This operation is typical of batch—feed wood—fired stokers. Two test series were performed at Site D, each consisting of three cofired tests. A methanol waste stream containing tetrachloroethylene was burned with residual oil in a multiburner boiler. Heat input contribution due to the methanol waste ranged between 18 and 30 percent. A toluene waste stream containing bis(2—chloroethyl)ether was cofired in the second test series contributing about 47 percent of the total heat input to the boiler. Boiler operation during the initial test series (Dl) was characterized by several burner upsets caused by waste feed problems. Boiler heat input, and thus) furnace temperature, was higher for Dl tests than for the D2 test series. A nethylmethacrylate waste stream containing spiked amounts of carbon tetrachioride, chlorobenzene and trichloroethylene was cofired at Site E during a test series involving six individual cofire tests. With the exception of one, all tests were conducted with residual oil as the primary fuel. Natural gas was substituted for residual oil during one of the tests. Three different boiler heat input rates were Investigated. The waste stream contributed from 20 to 43 percent of the total heat input. Boiler operation was relatively steady for each test condition. The pulverized coal—fired boiler at Site H was cofired with methyl acetate waste spiked with carbon tetrachioride, chlorobenzene and l1,1—trichloroethane. The boiler was fired at capacity with the waste contributing only 2.4 to 4.3 percent of the total heat input. Boiler operation was steady during all three cofired tests. 6 ------- TABLE 4. TEST SITES SELECTED FOR PCDD AND PCDF ANALYSES Site PriiMry Waste fuel POHCs I i) Boiler type fuel Waste fuel concentration (mg/el) Nun er of tests A Watertube stoker Wood chips Creosote sludge Phenol 0.6 to 1.3 Four cofire tests • sawdust Pentachlorophenol 2.2 to 6.0 • Naphthalene 5.4 to 19 a Fluorene 4.4 to 7.6 2.4 Din thy1phenol 0.3 to 1.3 D l Field erected Ho. 6 fuel Methanol waste Tetrochloroethylene 46 to 310 converted oil watertube stoker D2 No. 6 fuel Toluene waste Bis(2—chloroethyl)ether 39 to 42 oil Toluene 950 E Packaged watertube No. 6 fuel Methylmethacrylate Methylmethacrylate 33 to 52 olla waste artificially Carbon tetrachloride 27 to 34 spiked Chlorobenzene 16 to 20 Trichloroethy lene 27 to 32 Pulverized Methyl acetate Carbon tetrachloride 24 to 51 coal waste artificially Chlorobenzene 26 to 53 spiked 1.1.1—Trichloroethane 20 to 41 Natural Methylinethacrylate Carbon tetrachloride 6.8 gas waste artificially Chlorobenzene 15 spiked aone of the six tests was performed with natural gas as the primery fuel H Field erected watertube I Packaged watertube Streams analyzed for PCDO and PCOF Creosote waste Flue gas Flyash from najiticlone Bottom ash Waste fuel Flue gas Waste fuel Flue gas Waste fuel Flue gas Three cof Ire tests Three cofire tests Six cofire tests . . . . . Three cofire tests S Waste • Flue fuel gas One test • Waste • Fluegas fuel ------- Sampling for semivolatile and nonvolatile organics at Site L was performed only during one cofire test. Carbon tetrachloride and ctilorobenzene spiked methylmethacrylate waste was burned with natural gas. Heat input from the waste was approximately 30 percent with steady boiler operations. Although several other nonsteady tests were performed at Site L, none of the measurements included protocols for sernivolatile and nonvolatile organic compounds. 2.2 ANALYSIS PROTOCOL Analysis for PCDD and PCDF including the isomers 2 ,3 ,7,8—TCDD and —TCDF were performed by Battelle Columbus Laboratory (BCL) on both waste fuels and flue gas samples corresponding to the test series listed in Table 4. The analytical method by BCL was based on high—resolution gas chromatographyf high—resolution mass spectrometry (HRGC/HRMS). Detection limits were on the order of 0.04 to 0.13 ng/ml (ppb) for tetra isomers and 1 to 4 ppb for octa isomers in the waste fuels. Detection limits for flue gas samples ranged between 0.005 to 0.08 ngfm 3 of gas (about 0.0003 to 0.0005 ppt(v)). Quality assurance procedures Included recovery analysis of spiked internal standards and analytical method blanks. Appendix A details the analytical procedures and results obtained. Soxhlet extraction of flue gas samples collected using the modified EPA Method 5 train were performed by the Acurex Laboratory using the procedure depicted In Figure 1. These extract samples were in turn cornposited and analyzed by BCL. Analyses focused on both waste fuels and flue gas samples. However, for Site A, selected ash samples collected from the stoker bottom hopper and from the nulticlone hopper were also analyzed. PCDD analysis results of ash streams documented In Reference 3 are not reported here. for each of the Site D, E, and H test series listed in Table 4, waste fuels were coniposited into one sample and analyzed. Similarly, organic extracts of flue gas samples for each test series were coniposited and analyzed. 8 ------- 8302-034 8302-0 34 r Probe and nozzle wash SC/MS as BNA’s Semivolatile PP’s and other predominant chromatograPhic peaks (—10 non-PP peaks) Report results in jg/train Back half rinse and impinger contents Figure 1. Flow scheme for extraction and analysis of organic samples from MM5 trains. 8302-031 XAD resin 8302 -040 r Filter Soxhiet extract Parti cul ate we 1 ght Dry and weigh Combine solids as Extract BNA s Measure vol u Part, cul ate weight Combine all extracts Combine all extracts Soxhlet extract Fluorene-d 10 Terphenyl-d 14 (2-nItrophenol—d4) d 5 -ni trobenzene) 9 ------- SECTION 3 RESULTS Tables 5 and 6 summarize dioxins and furans concentrations, respectively, in the waste fuels at each of the test sites selected for these analyses. Site A results are based on an individual sample of the creosote collected during test 2. As indicated, all waste fuels, with the exception of Site A, were void of PCDD’s or PCDF’s at detection limits indicated in the tables’ footnotes. Site A creosote contained about 7.4 ppm total PCDD’s, primarily hepta and octa isomers. No PCDF analysis was performed on the Site A creosote sludge since the sample was exhausted in the PCDD analysis. The highly toxic 2,3,7,8-TCDD was not detected in any of the waste fuels. Tables 7 and 8 summarize flue gas concentrations of PCDD and PCDF homologs, respectively. Site A emissions measured during two tests showed total PCDD of about 75 ng/m 3 , by far the highest level of any of the other test sites. Total PCDD at Site 0 measured 0.80 and 0.64 ng/m 3 for the two test series, respectively. Minute levels of 2,3,7,8—TCDD were detected in the sample for the 02 test series when cofiring with toluene and bis(2—chloroethyl)ether. However, the concentration was at the detection limit. Both Site E and H emissions Indicate no PCDD emissions. However, flue gas samples obtained at Site L showed 1.1 ng/m 3 total PCOD, primarily tetra isomers. PCDF emissions were generally higher than PCDD emissions. Emissions were lowest for the test series at Site E and highest for the test series Dl and 1. No PCDF data are available for Site A. In general. flue gas concentrations of total PCDD and PCDF compounds ranged between non—detected (below 0.08 ng/m 3 ) to 5.5 ng/m 3 with the exception of the wood/creosote stoker. These concentrations correspond to a maximum of about 0.3 ppt(v). Table 9 summarizes the results of XAD—2 sorbent analytical blank extracts and the results of analysis of a field XAD blank from the Site L test program. Results indicate below—detection levels of PCDD and PCDF homologs In all samples except the Site L field blank which showed 3.77 ny of TCDD. This compound was tentatively identified as a l2,3,4—TCDD isomer. The presence of the l,2,3,4—TCDD in the field blank sample could not be explained and thus is to be considered an artifact of the analytical procedure. 10 ------- TABLE 5. DIOXIN CONCENTRATIONS IN WASTE FUELS (ppb) Site identification Sample description Tetra Penta Hexa Hepta Octa Total PCDD 2,3,78—TCDD A Test 2 creosote sludge ND 2.1 360 3.600 3,400 7,400 ND Dl Composited of methanol/ PCE for tests 2, 3, and 4 ND ND ND ND ND ND ND D2 Composited of toluene/ BCEE for tests 5, 6, and 7 ND ND ND ND ND tID ND E Composited of MMA CC1 4 , TCE, and Cl for tests 3, 4, 5, 6, 7, and 8 ND ND ND ND ND ND ND H Composited of methyl acetate, Cd 4 , C14 and 1,l,1—TCA for tests 2, 3, and 4 ND ND ND ND ND ND ND L Test 2 MMA, Cd 4 and Cl q ND ND ND ND ND ND ND Analytical method blank ND ND ND ND ND ND ND Note ND —— Not detected. Below detection limits. Detection limits fn the range of 0.045 to 4.17 ppb. Detection limits are summarized in Appendix A, Table 4, Page 30 ------- TABLE 6. FURAN CONCENTRATIONS IN WASTE FUELS (ppb) Site identification Sample description Tetra Penta Kexa Hepta Octa Total PCDF 2,37,8—TCDF Dl Composites of methanol! PCE for tests 2, 3. and 4 ND ND lID ND ND ND ND 02 ComposItes of toluenef BCEE for tests 5, 6, and 7 ND ND HO ND ND ND ND E Composited of MMA, CC1 4 TCE and CL$ for tests 3, 4, 5, 6, 7, and 8 lID ND ND ND ND lID ND H Composited of methyl acetate Cd 4 , CI and 1,ll-TCA for tests 2, 3, and 4 ND ND ND ND MD NO lID 1 Test 2 MMA, CC1 4 and CI ND NO NO ND ND ND ND Analytical blank ND ND lID lID ND ND ND Note ND —- Not detected. Below detection limits. Detection limits in the range of 0.045 to 2.0 ppb. Detection limits are summarized in Appendix A, Table 3, Page 29 ------- Site identification Sample description Flue gas sample (m 3 ) Tetra Penta Ilexa Hepta Octa Total PCDD 23,7,8—TCDD A Test 2 FINS extract 2.34 43 14 7.5 5.5 4.6 75 ND Test 4 MM5 extract 2.50 38 24 12 1.6 0.8 76 ND D l Cornposited of FINS extracts for test 2, 3, and 4 27.8a 0.12 0.074 0.10 0.25 0.26 0.80 ND 02 Composited of rIMS extracts for tests 5, 6, and 7 18.4 0.027 0.032 0.052 0.12 0.41 0.64 0.002b E Composited of rIMS extracts for tests 3. 4, 5, 6, 7, and 8 51.5a ND ND ND ND ND ND ND H Composited of FINS extracts for tests 2, 3, and 4 26.Oa ND ND ND ND ND ND ND I Test 2 FINS extract 8.7 0.57 0.010 0.066 0.12 0.29 1.1 ND acompositod sample volumes bMeasured value was equal to the detection limit Note TlF — Not detected. Detection limit 0.005 to 0.08 ng/m 3 To convert ng/m 3 to ppt multiply table entries by 0.068 for tetra, 0.062 for penta, 0.056 for hexa, 0.052 for hepta and 0.048 for octa TABLE 7. FLUE GAS CONCENTRATIONS -- DIOXINS (ng/m 3 ) I - . ------- TABLE 8. FLUE GAS CONCENTRATIONS —— FURANS (ng/m 3 ) — --—————.——————- — Site Flue gas Total identification Sample description sample (m 3 ) Tetra Penta Hexa }lepta Octa PCDF 2 ,3 ,7 ,8-TCDF D l Con osited of FQ45 27.8 2.1 1.4 0.77 0.94 0.27 5.5 0.24 extracts for tests 2, 3, and 4 02 Composited of MM5 18.4a 0.10 0.035 ND 0.034 0.07 0.24 0.13 extracts for tests 5, 6, and 7 E Composited of ? l5 51.5a 0.12 0.019 ND ND ND 0.14 0.014 extracts for tests 3, 4, 5, 6, 1, and 8 H Composited of 1’ 4S 26.Oa ND 0.14 0.39 0.071 0.20 0.81 MD extracts for tests 2, 3, and 4 L Test 2 I l5 extract 8.7 0.61 0.53 0.62 0.38 0.34 2.5 0.033 aComposi ted sample voluees Note N1T — Not detected. Detection limit 0.005 to 0.037 ng/m 3 To convert nglm 3 to ppt , tiultiply table entries by 0.072 for tetra, 0.065 for penta,0.059 for hexa, 0.054 for hepta, and 0.050 for octa ------- TABLE 9. QA/QC RESULTS FOR FLUE GAS SAMPLE EXTRACTS (ng) Tetra Penta Hexa Hepta Octa 23,7,8—Tetra Method blank PCDD ND ND ND ND ND ND PCDF ND ND ND ND ND ND Site L field PCDD 3.77 ND ND ND ND ND blank PCDF ND ND ND ND ND ND Note ND —— Not detected; detection limit In the range of 0.040 to 1.50 ng in each extract sample 15 ------- SECTION 4 INTERPRETATION AND CONCLUSION Site A creosote sludge was the only waste fuel found to contain dioxins tn detectable quantities. Table 10 summarizes the DRE of each dioxin homolog and for the 2,3 ,7,8—TCDD at Site A. In the creosote waste, dioxin tended to predominate in the higher homologs —— hepta and octa. In the flue gas, the reverse is evident; dioxin concentrations tended toward the lower homologs —— tetra and penta. A plausible Interpretation is that during combustion the higher homologs In the waste were reduced to lower homologs which were emitted in the flue gas. This would account for the low DRE (negative in the case of tetra—CDD) for the lower homologs. Table 11 summarizes the mass flow rates for total dioxins and total furans for the test series Dl, 02, E, H, and 1. The calculated input rates are based on the combined waste firing rate for the test series and the sums of the detection limits for total PCDD and total PCDF. The flue gas emission rate was calculated using the combined flue gas emission rate for each test series since sample extracts were composited and analyzed as one sample. The table indicates that Input rates of dioxins and furans were not necessarily below the measured flue gas emission rate. Therefore, although no dioxins and furans were detected in each waste stream, it cannot be stated with certainty that measured dioxins and furans in the flue gas were products of combustion. 16 ------- TABLE 10. SITE A DIOXIN DRE’S 1 -lomol og/ Isomer Flowrates ( i g/sec) Creosote Flue gas DRE (percent) 2 ,3,7,8—TCDD ND ND NA Tetra—CDD ND 0.14 Penta—CDD 0.1 0.044 56.0 Hexa—CDD 18 0.024 99.87 Hepta—CDD 180 0.017 99.99 Octa—CDD 170 0.014 99.99 Total—CDD 370 0.24 99.94 Note NA —— not applicable, DRE cannot be computed because concentration in both input and flue gas streams were below the detection limit ND —— not detected; below the detection limits of 0.0032 for creosote and 0.0047 for the flue gas 17 ------- TABLE 11. MASS FLOW RATES OF TOTAL DIOXINS AND FURANS Total PCDD Total PCDF Site Input ratea (ng/s) Emission ratea (ng/s) Input ratea (rig/s) Emission ratea (ng/s) D 1 ND (2,400) 22 ND (2,500) 150 02 ND (3,600) 12 ND (1,800) 44 E ND (5,100) ND (4.2) ND (3,300) 7.4 H ND ( 20) ND (2.7) ND ( 17) 27 L ND ( 560) 7.4 ND C 570) 17 aBased on firing rate of waste and sum of detection limits for total PCDD and total PCDF listed in Appendix A, Table 3 and 4. 18 ------- REFERENCES 1. Castaldinl, et al., “Engineering Assessment Report Hazardous Waste Cofiring in Industrial Boilers,” Acurex Technical Report TR—84—159/EED, June 1984. 2. DeRosier, R., et al., “Emission Testing of Industrial Boilers Cofiring Hazardous Wastes —— Site L,” Acurex Technical Report, December 1984. 3. Castaldinl, C., et al., “Emission Testing of Industrial Boilers Cofiring Hazardous Wastes —— Site A,” Acurex Technical Report TR—83—112/EED, January 1983. 19 ------- APPENDIX A BATTELLE COLUMBUS LABORATORY ANALYTICAL REPORT 20 ------- 0 BalteIle Ci ,Iumbus L. h r li rie r() King Avtnw CoIurnhu , Oliii, 4 32( I 26 TItphcinc ( 14 424 -h4 24 TiIcx 24-c4 4 January 25, 1985 Mr. Carlo Castaldini Project Engineer Acurex Corporation 555 Clyde Avenue P.O. Box 7555 Mountain View, CA 94039 Dear Mr. Castaldini: We have completed the analyses of the fuel oils and XAD—2 resin extract samples. I am enclosing a description of the analytical methodology, the instrumentation, and the quality assurance measures that we employed. Copies of the computer generated selected ion chromatograms are included in the appendix. Please contact me at (614) 424—4247 if you have any questions. Si ncerely, ( a Q- . Fred 1. DeRoos, Ph.D. Principal Research Scientist Analytical Chemistry Section FLD/bsf 21 ------- Analytical Methodology Introduction This report describes the analytical procedures used to determine the levels of polychiorinated dibenzo-p-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF) in eleven composite samples from Acurex Corporation, Mountain View, California. The samples, consisting of fuel oils and XAD-2 extracts, were received by Battelle on December 14, 1984. Extraction The composited fuel oil samples were prepared by combining 5 ml of each of the fuel samples specified in your letter dated December 7, 1984. One milliliter of the composite fuel sample was removed for sample extraction and was spiked with three isotopically labelled internal standards. The internal standard spikes were 25 ng of 2,3,7,8—tetrachioro- dibenzofuran- 13 C 12 (2,3,7,8-TCDF- 13 C 12 ), 25 ng of 2,3,7,8-tetrachlorodi- benzo-p-dioxin- 13 C 12 (2,3,7,B-TCDD— 13 C 12 ), and 25 ng of octachlorodibenzo- -p-dioxin- 13 C 12 (OCDD- 13 C 12 ). The samples were diluted with 30 ml of hexane and washed multiple times with concentrated sulfuric acid. The hexane solution was then concentrated to 10 ml and eluted through a 5 g basic alumina column using 25 ml of hexane, 25 ml of 3% methylene chioride/hexarie, and 25 ml of 50% methylene chloride/hexane. The 50% methylene chioride/hexane fraction was collected, concentrated and exchanged into hexane. The sample was then passed through a multilayer silica gel column containing 1 g silica gel, 2 g 33% NaOH on silica gel, 1 g silica gel, 4 g 44% H 2 S0 4 on silica gel and 2 g silica gel. The eluate was collected, concentrated, solvent exchanged into hexane, and passed through a 1 g basic alumina column. The eluting solvents were 5 ml of hexane, 5 ml of 3% cnethylene chioride/hexane and 22 ml 22 ------- of 50% methylene chloride/hexane. The 50% methylene chioride/hexane fraction wa collected, concentrated to dryness and diluted with 20 i l of decane. Composite samples of the XAD-2 extracts were prepared by combining all of the specified samples. One of the 6 sets of samples that were combined, XE38, was prepared with methylene chloride since the samples to combine were dry. The composite XAD—2 samples were spiked with 25 ng 2,3,7,8-TCDF— 13 C 12 , 25 ng 2,3,7,8-TCDD- 13 C 12 and 25 ng OCDD- 13 C 12 and eluted through two silica gel columns. The first column contained 10 g of 44% concentrated H 2 S0 4 on silica gel and the second was a multilayer silica gel column as described above. The eluate was collected, concentrated to 5 ml, and passed through a 1 g basic alumina column. The elution solvents were 5 ml of hexane, 5 ml of 3% methylene chloride/hexane, and 22 ml of 50% methylene chloride/hexane. The 50% methylene chloride/hexane fraction was collected, concentrated to dryness and diluted with 20 l of decane. Analysis The extracts were analyzed and quantified for PCDD/PCDF using combined capillary column gas chromatography/high resolution mass spectrometry (HRGC/HRMS). The HRGC/HRMS consisted of a Carlo Erba Model 4160 gas chromatograph interfaced directly into the ion source of a VG Model 7070 mass spectrometer. The primary chromatographic column was a 3DM DB-5 fused silica column using helium carrier gas at a flow velocity of 25 cm/sec. The mass spectrometer was operated in the electron impact (El) ionization mode at a mass resolution of 9000-12000 (M/ M, 10% valley definition). The operating parameters of the HRGC/HRMS are sumarized in Table 1. The primary analyses, determination of total level of each PCDD/PCDF class, were carried out using two separate HRGC/HRMS runs. This approach was necessary due to chromatographic 23 ------- overlap of adjacent isomer classes and hardware limitations to the number of masses that could be sequentially monitored. The first run provided data for the PCDD and PCDF isomers having an odd number of chlorine substituents, while the second run determined the levels of the isomers containing an even number of chiorines. Although the DB-5 capillary column is excellent for PCDD/PCDF isomer class determinations, it does not provide the isomer specificity of the more polar phase columns. Since 2,3,7,8—TCDD and 2,3,7,8-TCDF are the most toxic PCDD/PCDF isomers, all samples that were found to contain these isomers, using the DB-5 column, were reanalyzed using 50M CP Sil-88 fused silica capillary column which resolves 2,3,7,8-TCDD from the other 21 TCDD isomers. While this column does not provide complete separation of 2,3,7,8-TCDF, the level of confidence is far greater than with the DB-5 column and is considered to be state of the art. All HRGC/HRMS data were acquired by multiple-ion—detection using a VG Model 2035 Data System. The exact masses that were monitored are shown in Table 2. Quality Assurance The operation of the HRGC/HRMS was evaluated each day by analyzing standard mixtures of PCDD/PCDF isomers. These consisted of 2,3,7,8-TCDF, 2,3,7,8-TCDD, 2,3,7,8-TCDF-’ 3 C12, and 2,3,7,8—TCDD- 13 C 12 mixtures to evaluate accuracy of quantification, mixtures of selected PCDD/PCDF isomers to evaluate the stability of the chromatographic elution windows, and TCDD isomer mixtures to evaluate isomer resolution. The mass accuracy of the MID unit was evaluated at least every four hours by focusing selected ion masses from perfluorokerosoene (PFK) and correcting the slope to account for minor variations. Mass focus stability was assured by the use of a reference PFK “lock mass” to correct for any mass focus drift. 24 ------- Quanti fi cation The PCDF/PCDD isomers were quantified by comparing the sum of the two ions monitored for each class to the sum of the two ions monitored for the corresponding internal standard. The 2,3,7,8-TCDF- 13 C 12 was used to quantify the TCDF isomers, the 2,3,7,8-TCDD— 13 C 12 was used to quantify the TCDD and the pentachioro through hexachloro isomers and the OCDD-’ 3 C 12 used for the heptachioro through octachioro isomers. Experimental relative response factors were obtained by analyzing a test mix which contained representatives of the tetra- through octa-dioxin classes. These response factors were included in all calculations used to quantify the data. The response factors for the furans were assumed to be the same as those for the corresponding dioxin class. The response factors were calculated using the sum of the two ions monitored for each class of isomers compared to the sum of the two ions monitored for the corresponding internal standard. The experimental response factors were: Tetra-COF 0.980 Tetra-CDD 1.001 Penta 0.672 Hexa 0.570 Hepta 2.498 Octa 1.132 The formula used for quantifying the isomers was: Conc ( / = Areas of Quant. Masses Quant. of Internal Standard g g Areas of mt. Std. Masses Resp. Factor x 1 g (for fuel oils) 25 ------- In the case.of the XAD-2 samples the level of PCDF/PCDD is reported in nanograms per total XAD-2 extract. The criteria that were used to identify PCDD and PCDF Isomers were: (1) Simultaneous responses at both masses (2) Isotope ratio within ± 15% of theoretical value (3) Retention times within windows determined by analyses of standards (4) Signal to noise ratio equal to or greater than 2.5 to 1 The 2,3,7,8-TCDF/TCDD isomers included the additional criterion that they coeluted within ± 1 second of their isotopically labelled analogs. A limit of detection was calculated for samples in which a particular chlorination class was not detected. The formula used was: Limit of Detection (n / ) = Hts. of Quant. Masses Quantity of mt. Std. x 2.5 g Rts. of mt. Stnd. Masses Wesp. Factor x 1 g (for fuel oils ) In the case of the XAD-2 samples the limit of detection is reported in nanograms per total XAD-2 extract. Re s u 1 t s The results from the PCDF/PCDD analyses are summarized In Tables 3 and 4 respectivelY. A detection limit is listed in parentheses for samples in which a particular chlorination class was not detected. The selected ion current traces for the sample and standard analyses are included in the appendix. 26 ------- TABLE 1. HRGC/HRMS OPERATING PARAMETERS Mass Resolution Electron Energy Accelerating Voltage Source Temperature Preamplifier Gain Electron Multiplier Gain Transfer Line Temperature Col urnn Injector Temperature Column Temp — Initial (3 mm) Column Temp - Program Column Temp - Final Carrier Gas Flow Velocity Injection Mode Injection Volume 9000-12000 (M/AM, 10% valley definition 70 eV 6000 volts 200°C volts/amp i 06 280°C DB—5 30M CP Sil-88 50M 300° C 160°C 30°C/mi n 250°C (CP Si1-88) 290°C (DB-5) Heli urn 25 cm/sec Spl itless 2 iti 27 ------- TABLE 2. EXACT MASSES USED FOR THE DETERMINATION OF PCDD AND PCDF Accurate Compound Mass 1 Mass Mass 2 Theoretical Mass Isotope Ratio 1/Mass 2 Monoch lorodibenzo-p-dioxins 218.0134 220.0105 3.09 Monochlorodlbenzofurans 202.0185 204.0156 3.09 Trichlorodiphenyl ethers 271.9562 273.9533 1.03 DicPilorodlbenzo—p—dioxlns 251.9745 253.9715 1.54 Dichlorod lbenzofurans 235.9796 237.9766 1.54 TetrachIorodiphenyl ethers 305.9173 307.9143 0.77 Trichlorodibenzo—p—dioxins 285.9355 287.9325 1.03 Trichlorodibenzofurans 269.9406 271 .9376 1.03 Pentachiorodiphenyl ethers 341.8753 343.8724 1.54 Tetrach lorodibenzo—p—dioxins 319.8965 321 .8936 0.77 Tetrachlorodibenzofurans 303.9016 305.8987 0.77 Hexachiorodiphenyl ethers 375.8364 377.8334 1.23 Pentach lodibenzo-p—diox lns 355.8546 357.8517 1.54 Pentach lorodibenzofurans 339.8597 341 .8567 1.54 Heptachlorodiphenyl ethers 409.7974 411.7944 1.03 Hexachlorodibenzo-p-diox lns 389.8156 391 .8127 1.23 Hexach lorodibenzofurans 373.8207 375.8178 1.23 Octachlorodiphenyl ethers 443.7584 445.7555 0.88 Heptach lorodibenzo-p-dioxins 423.7766 425.7737 1.03 I4 otach1orodibenzofurans 407.7817 409.7788 1.03 Nonachlorodiphenyl ethers 477.7194 479.7165 0.77 Octachlorodibenzo-p-dioxin 457.7377 459.7347 0.88 OctachIorodibenzofuran 441 .7428 443.7398 0.88 Decachlorodiphenyl ether 511.6805 513.6775 0.69 28 ------- TABLE 3. LEVELS OF PCDF (DETECTION LIMIT) WASTE FUELS (ppb) XAD—2 (ng/saniple) Sample 2,378— TCDF Tetra—CDF Penta—CDF Hexa—CDF Hepta—CDF Octa-CDF Dia (0.129) (0.129) (0.063) (0.094) (0.909) (1.999) 02 (0.073) (0.073) (0.075) (0.081) (1.757) (0.914) E38 (0.066) (0.081) (0.055) (0.080) (0.881) (1.113) H (0.050) (0.050) (0.074) (0.146) (0.667) (0.881) L2 (0.051) (0.051) (0.045) (0.073) (0.679) (1.734) MBb (0.065) (0.065) (0.082) (0.129) (1.021) (1.157) XD1C 6.69 57.31 38.78 21.44 26.17 7.55 XD2 0.23 1.87 0.64 (0.333) 0.62 1.28 XE38 0.72 6.34 0.98 (0.260) (2.586) (2.198) XH (0.968) (0.968) 3.56 10.24 1.99 5.25 XL2 0.29 5.34 4.36 5.43 3.26 2.98 XLB (0.120) (0.120) (0.040) (0.041) (0.095) (0.134) MB (0.182) (0.182) (0.087) (0.186) (0.389) (1.497) a 01 through L2 are waste fuel samples bLaboratory Method Blank CXD] through XLB are flue gas extract samples 29 ------- TABLE 4. LEVELS OF PCDF (DETECTION LIMIT) WASTE FUELS (ppb) XAD—2 (ng/sample) Sample 237,8— TCDF Tetra—CDF Penta—CDF Hexa—CDF Hepta—CDF Octa-CDF D1 (0.076) (0.076) (0.057) (0.076) (0.693) (2.063) D2 (0.072) (0.072) (0.075) (0.108) (1.755) (4.168) E38 (0.070) (0.068) (0.066) (0.094) (1.032) (2.106) H (0.075) (0.075) (0.084) (0.089) (0.702) (1.145) L2 (0.050) (0.050) (0.045) (0.074) (0.714) (1.648) MBb (0.090) (0.090) (0.064) (0.095) (0.823) (3.444) XD1C (0.106) 3.28 2.06 2.85 6.86 7.32 XD2 0.04 0.50 0.59 0.96 2.22 7.59 XE3B (0.174) (0.174) (0.296) (0.280) (2.471) (3.913) XH (0.0492 (0.492) (0.401) (0.898) (1.442) (3.024) XL2 (0.045) 4.92 0.09 0.57 1.06 2.52 XLB (0.064) 3.77 (0.046) (0.052) (0.108) (0.220) MB (0.198) (0.198) (0.089) (0.170) (0.738) (1.450) aD] through L2 are waste fuel samples bLaboratjjry Method Blank CXD1 through XLB are flue gas extract samples 30 ------- TECHNICAL REPORT DATA (Please read Inzrruc lions on the revene before completing) I REPORT NO 12 3 RECIPIENTS ACCESSION NO 4 TITLE AND SUBTITLE DIOXIN EMISSIONS FROM INDUSTRIAL BOILERS BURNING HAZARDOUS MATERIALS & REPORT DATE April 1985 PERFORMING ORGANIZATION CODE 7 AUTHOR(S) C. Castaldini 3 PERFORMING ORGANIZATION REPORT NO PERFORMING ORGANIZATION NAME AND ADDRESS Acurex Corporation 555 Clyde Avenue P. 0. Box 7555 Mountain View. California 94039 10 PROGRAM ELEMENT NO DIO9 11 CONTRACT/GRANTNO 68—03—3241 12 SPONSORING AGENCY NAME AND ADDRESS U. S. Environmental Protection Agency Hazardous Waste Engineering Research Laboratory 26 West St. Clair Street Cincinnati, OH 45268 13 TYPE OF REPORT AND PERIOD COVERED 14 SPONSORING AGENCY CODE 15 SUPPLEMENTARY NOTES lb b IM Laboratory analyses for polychiorinated dibenzo—p—dioxinS (PCDD) and polychlorinated dibenzo-furans (PCDF) were performed on waste fuels and stack gas emission samples from five industrial boiler test sites cofiring liquid hazardous wastes. Analytical results indicate that, apart from creosote sludge, chlorinated wastes were void of PCDD and PCDF compounds at detection limits in the range of 0.045 to 4.17 ppb. Creosote sludge cofired with wood waste in a stoker boiler was found to contain 7.4 ppr of total dioxins, primarily hepta and octa homologs. Stack gas concentrations of PCDD were highest for the creosote wood-fired stoker at about 75 ng/m . Other PCDD results indicate concentrations ranging from below detection levels (0.0022 ng/m 3 to 0.019 ng/m 3 ) to a maximum of 1.1 ng/m 3 . PCDF concentrations were generally higher with total furan levels up to 5.5 ng/rn 3 . No 2,3,7,8_tetrachloro -dibenZO-P-diOXifl (TCDD) was detected in any waste fuel. While the 2,3,7,8-TCDD isomer was detected in flue gas emissions from one boiler, the measured value for the emissions of 2,3,7,8-TCDD was equal to the detection limit for the compound in the particular flue gas (0.002 ng/m 3 ). 17 KEY WORDS AND DOCUMENT ANALYSIS DESCRIPTORS b IDENTIFIERS/OPEN ENDED TERMS C COSATI Field/Group Dioxin Boilers Hazardous Wastes Furans Hazardous Air Emissions RCRA Field Testing Engineering Chemistry 18 DISTRIBUTION STATEMENT 19 SECURITY CLASS (This Report) Unclassified 21 NO OF PAGES 32 20 SECURITY CLASS (This page) Unclassified 22 PRICE EPA Form 2220—1 (R.v. 4—77) PREVIOUS EDITION II OBSOLETE 31 ------- |