Document v6bxwvV5eoXoOnrNeE8DERmvE

T k /f0 Y ilM .' u p t i i k t . ' H i m .1 w l i ........................... ... i < i' very long; thus, equilibrium m:iy not l*e reached. An ex ample is the recent study of chlorobenzene (CR) convener uptake by Oliver aiul Niinii {]!) In this case, the half- times for uptake were approximately the following: tol- ru-CB, Ifi days; penta-CB, 50 days; in which the hexa-CB congener had not readied equilibrium ufler 120 days. This is the trend predicted by the equation. These authors correctly identified that the experimental BCF of HCB after 120 days was not an equilibrium value and showed that this could lead to erroneous prediction of fish con centrations in Lake Ontario. It should be emphasized that although the narcotic ef fect of PAB esters is well described by this equation, not all toxic effects are likely to be amenable to such simple analysis. It is apparent that by formulating the equations in the fugacity format, it becomes easier to manipulate the var iables and new insights are obtained into the pharma cokinetic processes. It is relatively easy to formulate and Lest equations describing uptake from food, to include metabolic degrading reactions, and thus to build up more comprehensive equations describing these pharmacokinetic processes. R e g istry No. 2,5-Dichlorobiphenyl, 348S3-39-1; 2,2',5-trichlorobiphenyl, 37680-65-2; 2,4',5-trichlorobiphenyl, 16606-02-3; 2,2',5,5'-tetrachlorobiphenyl, 35693-99-3; 2,3',4',5-tetrachlorobiphenyl, 32598-11-1; methyl p-aminobenzoate, 619-45-4; ethyl ( |j SpiiL'ii', A.; Il.iliiriink, *f I., fc.nrirmt I `>M<ul ( 'Itrtn, Hl>l2, j, :)y :cjo. (21 Mai'kiiy, I). Environ. S it. Tcrivud. 11171!, 13. 1218-1223. (3) Mai'kav, I P a t e r s o n , S. E nrtrmi, ,sYi. 7'rr/i/in/. 19KI, 16, 1onti- HIM. (-1) Markny, I.J.; I'tilcrson, I). Environ. Set. Technol, 11182, 1G, 654 A-6 6 0 A. (5) Bruggeman, W. A.; Marinin,' L. B. J. .; Kuoiman, D.; Hulzingcr, O. Chcmusphvre 1982, 10, 811-832, (6) Yalkowsky, S. H.; Carpenter, 0 . S.; Flynn, G. L.; Slunick, T. G. J. Pharm. S r i. 1973, 62. 19-19-1954. (7) Yalkowsky, S. H.; Slunick, 'I'. G.; Flynn, G. L. J. Pharm. Sri. 1974,*63, 691-695. (8) Mack ay, D. Environ. Sri. Technol. 1982, 16, 274--278. (9) Bruggeman, W. A.; Van Dor Steen, ,J.; Hulzingcr, O. J. Chronwtagr. 1982, 238, 335-316, (10) Woudburn, K. B. M. S. Thesis, University of Wisconsin, Madison, WI 1982. (11) RUgehuusen, H.; Gulli, J. A.; Esscr, H. O, Ecotoxicol. En viron. Saf. 1930, 13-1 157. (12) Southworth, G. R.; Beauchamp, J. J.; Schmeider, P. L. Environ. Sci. Technol. 197S, 2, 1062-1066. (13) Yalkowsky, S. H.; Valvani, S. C. J . I 'harm. Sri. 1980, 69. 912-922. (14) Oliver, B. G.; Niimi, A. J. Environ. Sri. Technol. 1983, 37, 287-291. Received for review Ju n e 6, 1983. Accepted December 7, 1983. This work was supported by the Ontario M inistry of Environ ment. Environmental Fate of Combustion-Generated Polychlorinated Dioxins and Furans dean M. Czuczwa and Ronald A. Hites* School of Public and Environmental Affairs and Department of Chemistry, Indiana University. Bloomington, Indiana 47405 Polychlorinated dioxins and furans were found in sed iments from the Saginaw River and Bay and from Lake Huron. The congener distributions of the dioxins and furans indicate that combustion may be the major source i of these compounds. The depth vs, concentration profiles in dated sediment cores showed that emission of dioxins and furans has increased greatly since 1940. This historical increase is similar to trends for the production, use, and disposal of chlorinated organic compounds and suggests that chlorinated precursors of dioxins and furans, present* in incinerator combustion fuels, may be the main source of the dioxins and furans found in these sediments. Introduction Polychlorinated dibenzodioxins (PCDD) and dibenzofurans (PCDF) are the subject of a recent, often heated debate because some of these compounds are very toxic. For example, 2,3,7,8-tetrachlorodibenzodioxin (2,3,7,8-TCDD) has been found to be acnegenic to humans (7), teratogenic to mice (2), carcinogenic to rats (3); and acutely toxic to guinea pigs (4). Other isomers of PCDD (75 total) show differing degrees of toxicity; isomers of PCDF (135 total) are generally as toxic as the corresponding PCDD. Initially, PCDD and PCDF were discovered as trace impurities in various chlorinated aromatic compounds. We will call these "industrially generated" dioxins and furans. PCDD and PCDF were found in chlorophenols (5-7), herbicides (8-10), and PCB's (11). Industrially generated PCDD and PCDF have entered the environment through accidental release during chlorophenol production (72), aerial application of phenoxy herbicides (75), and improper disposal of wastes (74). These events tend to be sporadic and localized. More recently, PCDD and PCDF have*been identified in effluents from combustion processes. In particular, dioxins and furans have been found in the fly ash and flue ga9 of municipal incinerators (15-18). The PCDD. and PCDF may be associated with small particulates, which have'Iong residence times in the atmosphere, and in this manner, combustion-generated dioxins and furans could become distributed over large areas. Thus, combustion may have made PCDD and PCDF ubiquitous in the en vironment. . The initial reports of PCDD and PCDF in municipal incinerator fly ash led to an investigation of a variety of combustion processes each of which was a possible source of PCDD and PCDF (79). PCDD and PCDF were mea sured in particulates from the combastion of municipal and chemical wastes and fossil fuels, and in some unusual samples such as cigarette smoke and charcoal-broiled steak. The researchers concluded that PCDD and PCDF.are ubiquitous products of the combustion of organic mate rials. In an interview (20), an author of this paper stated, 444 Environ. Sci. Technol.. Voi. 18, No. 6, 1984 0013-936X/84/0918-0444S01.50/0 / 1984 American Chemical Society sugosi ed (liiit ili ere may be fewer source;- M* i/otnlms- lion-generated dioxins than initially thought. A fundamental point in this debate renters on the mechanism of formation of dioxins and fiirans in com bustion sources. PCDD and PCDF may be formed by the cyclization of chlorinated precursors present in the fuel or by the reaction of organic compounds with inorganic chlorine both present in the fuel. It is likely that the first mechanism is operative; model pyrolysis experiments have shown that PCDD and/or PCDF are formed by pyrolvzing chlorinated precursors such as chlorobenzenes (23), chlorophenols (24), and PCB's (25). There is little experi mental evidence for the second mechanism, but it cannot yet be excluded. In any case, the first mechanism is likely to form PCDD and PCDF in higher yields than the second mechanism. Because of the toxicity of these compounds, it is im portant to know their environmental fate. We propose the following paradigm: Once emitted from a combustion source, the-particulates (carrying their load of dioxins and furans) can travel some distance, which is a strong function of the size of the particle. Larger particles will settle close to the source while small particles may have sufficient residence times in the atmosphere to be transported to remote locations. Thus, PCDD and PCDF may be carried by direct airborne transport to ultimate environmental sinks such as the oceans or lakes. After deposition in these aquatic systems, the dioxins and furans will settle to the bottom sediments. As earlier sediments become buried by materials deposited in subsequent years, an historical record to dioxin and furan inputs to the environment will be preserved. Some caveats should be stated regarding this general model for the environmental fate of PCDD and PCDF. First, the sources are highly variable. The quantity and isomeric distribution, of PCDD and PCDF emitted will depend on combustor design, operating conditions, com position of the fuel, and degree of emission control. Sec ond, environmental alterations in the air or water column due to photodecomposition, biodegradation, volatilization, or bioaccumulation may occur. Third, we have assumed i that these compounds are not subject to degradation once they are in the sediments. This is probably a good as sumption; a recent summary of the environmental chem istry of PCDD suggests that microbial degradation is negligible (26). Fourth, sediment mixing processes intro duce some, usually minor, uncertainty in interpreting the historical trends of compounds deposited in sediments. The goal of our study is to approach two of th present questions concerning PCDD. and PCDF: (a) Are the PCDD and PCDF which are present in the environment the result of industrial production or combustion? (b) Is there evidence regarding the historical input of these materials into the environment which could more clearly define the mechanism of their formation? To address these questions, we measured PCDD and PCDF in samples from combustion sources and from lacustrine sediments. Fly ashes from a municipal incinerator and from coalfired power plants were analyzed to study the dioxin and furan congener distributions typical of combuslion sam ples. These distributions can be used to determine if sources. Another was I 't'IJl) au! r< I ) i <nu <ni rut ion?, as ft function of distance fr o m anthropogenic activity. The most valuable information came from the analysis of sediment cores. Sections ol cores were analyzed for PCDD and PCDF and dated by radioisotopic techniques. Thus, the historical inpuL of dioxins and furans was obtained. We used these data to distinguish between anthropogenic and natural inputs of PCDD and PCDF. Experimental Section Fly Ash. Four fly ash samples were obtained from B. J. Kimble (Laboratory of Energy-Related Health Research, Davis, CA). Samples 1 and 2 were collected from the cyclone stage (cyclone ash) and electrostatic precipitator (ESP hopper ash), respectively, of a midwestern municipal incinerator. Samples 3 and 4 were from two coal-fired power plants burning western (low sulfur, low chlorine) coal. Approximately 10 g of fly ash was spiked with 100 ng of 37CI8-OCDD (KOR Isotopes, Cambridge, MA), allowed to'dry, and Soxhlet extracted with 200 mL of "distilled in glass" grade benzene (MCB Reagents, Cincinnati, OH) for 24 h. The resulting extract was concentrated by rotary evaporation to less than 1 mL and the solvent exchanged to hexane for alumina fractionation. Preextracted neutral alumina (Brockman Activity I, Fisher Scientific) was activated at 250 C for 2 h. It was then deactivated with 1% by weight distilled water and allowed to equilibrate for 16 h. A 0.5 X 6.5 cm microco lumn was packed and washed withhexane. The sample was introduced and eluted with 8 mL each of hexane, 2% methylene chloride in hexane, and 40% methylene chloride in hexane. Dioxins and furans eluted in the 40% fraction, which was concentrated to 100 /L by slowly passing a stream of purified N2 over the.sample. The sample was then ready for analysis by methane negative chemical ionization gas chromatographic mass spectrometry (NCI-GC/MS)/ Sedim ents. The sampling sites are shown in Figure 1. A sediment grab sample, nominal depth of 8 cm, was collected in 1981 from the Saginaw River (station 161) by C. P. Rice (Great Lakes Research Division, University of Michigan, Ann Arbor, MI). The location was close to the mouth of the Saginaw River (43 39'N, 83 51'W). A sediment core from Saginaw Bay (station 30A) (43 52'N, 83 40'W) and two companion sediment cores from Southern Lake Huron (SLH-75-46L, and H, 43 3iyN, 81 55'W, further referred to as cores 1 and 2, respectively) were collected by J. A. Robbins (National Ocean and At mospheric Administration, Great Lakes Environmental Research Laboratory, Ann Arbor, MI). We used only the top 1-cm section of the Saginaw Bay core. Two cores from southern Lake Huron were collected by'S. J. Eisenreich (Department of Civil and Mineral Engineering, University of Minnesota, Minneapolis, MN) and J. A. Robbins. These samples were collected in 1981 by using a box corer at coordinates 43 39'N, 81 59'W (core 3) and 43 59'N, 82 10'W (core 4). All cores were sectioned into intervals-of 1 cm Lu a d e p th of 10 cm and then into 2-5-cm segments, depending on depth. Approximately 50 g (wet weight) of sediment was placed in glass Soxhlet thimbles and spiked with between 2 ng Environ. Sci. Techno!., Voi. 18. No. 6, 1984 445 ilm work i<-r i ;i u u.2 Mi I 2 2.'< 17u :vi 210 achieved on a 30 m X 0.25 mm Dll-5 ['used silica column (J & W ScientuV, Rancho Cordova, CA) with helium carrier gas (splilless injection at 30 C, isothermal for 4 min, 4 0/m in to 280 C, isothermal for 20 min). The ion source temperature, was 250 C, and the pressure of methane was typically maintained at 0.7 torr in the ion source. To furl her increase sensitivity, selected ion mon itoring was used. Ions were monitored for tetrachloro- through octachlorodioxins and furans, including a con firming ion (M` or (M - Cl)", depending on the isomer]. The ions used for quantitation were the following: tetra- chlorodioxins (TCDD), m /e 322; penlachlorodioxins (PnCDD), m/e 356; hexachlorodioxins (HxCDD), m/e 355; heptachlorodioxins (HpCDD), m /e 389; octachlorodioxin (OCDD), m /e 423; letrachlorofurans (TCDF), m /e 306; pentachlorofurans (PnCDF), m /e 340; hexachlorofurans K (HxCDK), m/e 374; heptachlorofurans (HpCDF), m/e 408; octachlorofuran (OCDF), m /e 444. Tn addition, m /e 435 IE f Figure 1. Map showing the Lake Huron sample sites. was monitored, representing the M - Cl ion of the internal standard. Dioxins and furans were quantitated by ratioing the appropriate peak area to that of the internal standard and 200 pg of the 37Cla-OCDD standard, depending on the and correcting for relative response factors which were expected dioxin and furan levels. The samples were ex obtained from a standard mixture of PCDD and PCDF il tracted for the first 24 h with 200 mL of 2-propanol to (one isomer per congener class). Concentrations are re remove water, followed by extraction with 200 mL of ported in parts per billion (ppb = 10"s) or in parts per methylene chloride for an additional 24 h. The methylene trillion (ppt = 10"12). chloride and 2-propanol extracts were combined, then D ating of Sedim ent Cores. Sedimentation rates for f reduced to 2 mL by rotary evaporation, and subjected to : 'i a three-step chromatographic cleanup. the cores were determined by J. A. Robbins and K. A. Johansen (NOAA, Ann Arbor, MI) by using t h e 137Cs and Natural sediments sometimes contain significant 2,0Pb techniques of Robbins and Edgington (28). Sedi amounts of elemental sulfur which can interfere in the mentation rates varied from 0.15 to 0.41 cm/year and will analysis. Thus, sulfur was removed by an' activated copper be discussed below. column. Fifty grams of copper (purified electrolytic dust; Q uality A ssurance. The analytical work followed the Fisher Scientific) was activated with concentrated HCl. guidelines suggested by the ACS Committee on Environ-' A glass column (1 X 25 cm) was filled with the copper menLal Improvement (29). Experiments included repli slurry. The sediment extract was passed through the cates, procedural blanks, and recovery measurements. The column and eluted with 150 mL of methylene chloride. recovery averaged 75% even for the lowest level samples. The eluent was concentrated to 2 mL, and the solvent was The average reproducibility was better than 30%, the exchanged to hexane for fractionation on silica. least reproducible being the tetrachloro- and penta- Preextracted silica gel (Davidson Chemical, Baltimore, chloro-PCDD and -PCDF. The limit of detection was 20 MD) was activated at 160 C for 16 h, deactivated with pg for 1,2,3,4-TCDD and 0.2 pg for OCDD. Subsequently, 1% water, and loaded into a 1.5 X 25 cm column with it was found that lowering the GC/MS ion source tem hexane. The sample was introduced and eluted with 75 perature to 150 C resulted in increased sensitivity, low mL each of hexane, 15% methylene chloride in hexane, ering the limits of detection to 0.05 pg for 1,2,3,4-TCDD and methylene chloride. Dioxins and furans eluted in the and 0.1 pg for OCDD. second fraction. The solvent was again reduced and ex Method validation included an interlaboratory calibra changed to hexane. The final step was alumina chroma tion experiment. In Table I, dioxin concentrations mea tography as described above. sured by the above procedure in a sample.of St. Louis air NCI-GC/M S Analysis. Negative chemical ionization particulates (National Bureau of Standards, Standard (NCI) mass spectrometry is particularly sensitive to Reference Material 1648) are compared with those re molecules that have a high electron capture cross section ported by Bumb et al. (19). Although the extraction, due to electrophilic atoms such as chlorine (27). Thus, NCI cleanup, and mass spectrometric techniques are different, is on ideal technique for the analysis of PCDD and PCDF. the results agree within the measurement error of the The enhancement of sensitivity is especially pronounced procedures. for the more highly chlorinated dioxins and furans. In our laboratory, OCDD showed a lOG-fold increase in sensitivity Jiesults and Discussion over that of electron impact (El). Combustion Sources. Figure 2 silow's the results of the All the analyses were obtained on a Hewlett-Packard fly ash analyses; note that four different concentration 5985B GC/MS system. Chromatographic .separation was scales are used in this figure. Sample 1, collected at the 448 Environ. Scl. Techno!., Vol. 18, No. 6, 1984 I . L j s l El I J I . .E . 'C o r r;r...L> .< [ ' (.C u r o c r * icl-C i' c n(j - . r c n - , Cl'tv CC.no _G1_ i i ,, _e c o r i ..rc.i -c D rn p C tir o c o r ic o o PnC[)OH.coDBCOO o c o n rc o r I '/iO 'M .c p r M pcor ocar tcdd ppc o d c d o h pc o d ocdo I I 1. 1\l I un _HU ............ SITE 30A a ____ = . J 3 _! CQOxpCCOOCDIj ic in ' i-ir.o f i . - i n r - ( . c u 1' oenr i c n n c m .n o . t u o p e n o o c o u Tco r P n to r w .cD r m pcd f o c o r tcd d po c d d h . c o d m pcdo o co o Figure 2. Concentration profiles of PCDD and PCDF congeners in combustion paniculate samples from the following: 1, a midwestern municipal incinerator cyclone: 2. the electrostatic precipitator from the same incinerator; 3 and 4, two coal-fired power plants burning western coat. cyclone stage of a municipal incinerator, contained lower concentrations of PCDD and PCDF than the electrostatic precipitator (ESP) hopper ash (sample 2) from the same plant. This can be explained by the foilowing mechanism: At the cyclone stage, materials are close to the combustion chamber and are collected at a higher temperature than at the ESP. In the cyclone, the dioxins and furans could be primarily in the vapor phase and, therefore, not asso ciated with the particulates. As the effluents reach the ESP, the temperature has decreased, and PCDD and PCDF are now condensed on the particulate phase and, thus, are collected. It is useful to note some trends in the PCDD and PCDF congener distributions in the fly ash samples (see Figure 2; samples 2-4). First, a large number of isomers were detected for each PCDD and PCDF congener. Second, OCDD is the most abundant dioxin, and OCDF is present in much lower concentrations than OCDD. And third, the hexa- or heptachlorodibenzofurans are the most abundant furans. Similar dioxin congener profiles were also mea sured in fly ash samples from a powerhouse, a rotary kiln, and waste incinerating facilities in a recent study (79) and in municipal incinerator fly ash (75-78). These general trends will be compared in the profiles found in environ mental samples to distinguish among possible sources. It is important, however, to recognize that these samples represent only a few sampling occasions and, therefore, may not be representative of combustion processes in general. Furthermore, an extrapolation of these trends to those found in environmental samples does not take into account the probability that fly ash collected in an elec trostatic precipitator may not accurately reflect the PCDD and PCDF that are emitted. These data will simply be used to compare combustion processes with environmental samples. The coal fly ash samples (Figure 2, samples 3 and 4) differ significantly from the municipal incinerator ash samples. Although some PCDD and PCDF were detected, no tetrachloro- or pentachlorodioxins or -fumns were de tected, with limits of detection of 100 ppt (tetra) and 10 CORE 2 M r c o r PnCDf MiCOf HoCOf CCD^ TC30 * aCOO H*CDO *pCOD o co d *1 C O R E I 3 -- pq ..B3 TCOf P C D I' . c o r d p C D f o c o r tcdo ' PnCDO .C O D MpCDD OCDO Figure 3. PCDD and PCDF congener profiles in six surficial sediments (see Figure 1 tor site locations). ppt (penta). Dioxins and furans, when present, were in much lower concentrations than in Lhe municipal incin erator ash. For example, the levels of OCDD in the coal fly ash samples (2.2 and 3.S ppb in samples 3 and 4, re spectively) were at least 100 times lower than those found in the municipal incinerator ash (440 ppb). Although coal fly ash clearly contains Jess OCDD then municipal incinerator ash, it still could be a significant source of OCDD to the environment, since coal combustion is so prevalent. In 1974, the amount of particulates emitted from coal combustion was estimated to be 2.4 X 109kg (30). The amount of particulates emitted from solid waste in cineration In 1971 was estimated at 7 X 108kg (37). Thus, approximately 3 times more particulates are emitted from coal combustion than from solid waste combustion. If the OCDD concentration on coal particulates is 100 times lower, then total emission of OCDD from coal combustion would still be approximately 30 times lower than that from municipal waste incineration. Finally, we should address the ongoing debate regarding 2,3,7,8-TCDD in coal fly ash. No isomer of TCDD wus detected in these samples with a limit of detection of about 100 ppt. This reaffirms similar findings (27, 22) and suggests that coal combustion is nol a significant source of 2,3,7,8-TCDD to the environment. S u rfa c e Sedim ents. Figure 3 shows that the dioxin and furan congener profiles obtained from surficial sedi- *ftOJ aa-r DIOXINS TCDD __*__ PfCOO __Hicoo______ __ HpCDO . __ OCDD FURANS a <0 Ti i| lr 1 ss ?< 1 n 5 J: si !;1 s i'll S ' ** 1 ti J L L~ A : v ` , a_ _ v p;-- --t t ~ ,Y * " - TCDF__j__ PnCDf __ __ H,cor _ HpCDF 1 IH .7 ; __1__ a CD * 1 i1 l| Tl OCOF Figure 4. Mass chromatograms showing PCDD and PCDF In the surficial segment from core 1. These data have not been corrected for response factors. ments from the Saginaw River and Bay and from southern Lake Huron. PCDD and PCDF are ubiquitous in the samples studied, including the most remote locations. The profiles are similar to those shown in Figure 2. The con centrations of PCDD and PCDF are highest in those sediments collected closest to urban areas (161 and 30A) and lowest in the open lake cores. This indicates that the PCDD and PCDF found'in these samples are anthropo genic in origin. Figure 4 shows the mass chromatograms for the dioxins and furans in the surficial segment of core 1. These data illustrate the typical combustion "fingerprint" discussed earlier. A number of isomers for each dioxin and furan congener class are detected. One finds a predominance of OCDD and HpCDF and greater levels of TCDF and PnCDF than the corresponding dioxins. In general, the PCDD and PCDF isomer distributions, even in the most remote samples, are similar to each other and are indicative of combustion. We, therefore, conclude that combustion is probably the major source of PCDD and PCDF to these locations. These data emphasize the importance of de termining the entire PCDD/PCDF profile rather than just the 2,3,7,8-TCDD content. Future work will include iso mer-specific quantitation of all tetra- through octachlorodioxins and -furans, thus increasing our ability to distinguish among sources. Sedim ent Cores. As outlined above, we have obtained* data on sediment cores to determine the historical input of PCDD and PCDF. Obviously, the amount and the composition of fuels have changed with time. The effect ' of these changes on the input of dioxins and furans to the* environment should be reflected in the:sedimentary record. Similar work by Hites et al. (32) showed that sedimentary polycyclic aromatic hydrocarbons reflected the changing use of fossil fuels. The most abundant PCDD and PCDF in cores 1-4 were HpCDD, HpCDF, and OCDD. The depth vs. concentra tion profiles for these species are shown in Figure 5. The sedimentation rate for cores 1 and 2 (companion cores from the same location) was calculated from the radioisotopic data and was found tn he 0.15 cm/vear. This wn? uned fo r\ ! tipCnD 5 :*oo IPio ; I no m;? boo \'-- \NpCDf -tp C DO lt<0 ivms 1035i .L ir/l i/i0 ADO IppO M p C OF ' \* _ I : ...------1 rT ` V r r 1-- \ ----- j j HpCDO I 1920? - is 20 2* IB80 (ESO J HpCDO ^ r"v \j\ 1 S HpCDF S' I 12 (L hi o 1950F0 1 603 iiDl ieoe Cene. Ipoti 600 1200 1000 Cone. <ppu Figure 5. HpCDF. HpCDD, and OCDO concentrations in tour Lake Huron cores vs. depth (left axis) and depositional age (right' axis). estimate the year of deposition corresponding to each depth, and these data are also plotted in Figure 5 (top). The mixing depth was 2.8 cm and is an indicator of the degree of movement of materials after deposition. This indicates that there may be an averaging of inputs over as much as a 15-year interval. In both cores 1 and 2, there is an abrupt increase in PCDD and PCDF concentrations around 1940. Lake Huron core 3 had a sedimentation rate of 0.21 cm/year and a mixing depth of 6.4 cm. This core also showed that PCDD and PCDF inputs increased around 1940 (see Figure 5, bottom left). Core 4 had a substantially higher sedimentation rate (0.41 cm/year, mixing depth of 5.9 cm) and thus offered greater time resolution for this trend. From this core, it is apparent that PCDD and PCDF inputs increased slowly during the 1940s and early 1950s to the present levels (see Figure 5, bottom, right). In general, the concentrations of PCDD and PCDF in core sections corresponding to deposition before 1940 are low, representing a much lower input of these materials before this time. These changes cannot be accounted for by in situ deg radation of PCDD and PCDF in the sediment. The con gener profiles in all cores were similar to each other and were consistent along the depth of the core. This is best illustrated in Figure 6 which shows the congener distri butions for core 4 as a function of depth. Note the sim ilarity of the pattern with depth. Furthermore, isomer ratios were calculated at each depth in core 4 and were found to be constant. For example, the ratio of 1.2.3.4.6.7.8- HpCDF to 1,2,3,4,6,8,9-HpCDF was 0.71 0.27, the ratio of 1,2,3,4,6,7,9-HpCDD to 1,2,3,4,6,7,8HpCDD was 0.72 0.06, and the ratio of OCDD to 1.2.3.4.6.7.8- HpCDD was 2.8 0.36. There were no trends in the ratios with increasing depth for any of the heptaor oclachlorodioxins or -furans. Thus, there is no evidence of degradation of these PCDD or PCDF in these Lake Ht irnn m m I 0 -1 CM t c or p n c o r H . c o f H p c o f ocor t c d o PnCooM,coopcoo ocoa. 2 -3 CM i c d i * Pncor H>cor Hpcor ocor t c o d Pncoo h ^ c o o * pC o d o c o o 4 -5 CM TCDF PnCOF m .C D F M pCDF OCOF TCDO P n C O O M .C O O "pC C D OCOO 6 -7 CM CB ,,B8 un g i ea gg TCDF P nC D F m .C O F H p CDF OCOF TCOO PnC D D H .C O D M pC D D OCOD 8 -9 CM JSL TCOF P n C D F M .C O F Mpcor O CDF TCOO P /'C C O H C O O w p C D D ocoo 1 0 -1 2 CM _EH_ TCDF PnCOF H .C O F HpCDF OCDF TCOD PrvCOOMuCOOMpCOO OCOO Figure 6. PCDD and PCDF congener profiles in core 4 as a function o( depth. Clearly, the dioxin and furan inputs have changed con siderably over time; much more were deposited since 1940. It seems likely that a major source began in the 1940s and increased until the present time. What is such a source? The burning of coal has been a major combustion pro cess since the last century, and coal fly ash, as shown earlier, contains some PCDD and PCDF. The trend for U.S. coal consumption for the last century is shown in Figure 7 (top) (33). Coal use in the Great Lakes area parallels that of the nation. For example, in 1930, Illinois, Indiana, Ohio, Michigan, and Wisconsin accounted for 40% of the nation's coal consumption, in 1957, 30%, and in 1970, 30% (33). Therefore, we may safely compare U.S. coal consumption to sedimentary dioxins and furans in the Great Lakes. Figure 7 compares coal consumption with the total HpCDD, HpCDF, OCDD, and OCDF found in the four Lake Huron cores (see Figure 7, bottom). It is obvious that coal use cannot account for the increase in dioxin and furan concentration since 1940. Indeed, coal use has been relatively constant since 1910. The U.S. Tarriff Commission Reports of Production and Sales of Synthetic Organic Chemicals (34), published since 1918, reveal that the chemical industry grew greatly be ginning in 1940. Starting at this time, the production of chlorinated organic compounds such ns chlorobenzenes and chlorophenols increased substantially (see Figure 7, mid dle). These compounds are used in a variety of products, including building supplies, herbicides, and packaging. Much of these materials eventually become incorporated in solid wastes. The trend for ihe production of ehloro organic compounds is very similar to the sedimentary PCDD and PCDF profiles (compare Figure 7, middle and bottom). The agreement between these two trends is Hesnite the uncertainties introduced by sedi 1670 1060 1690 1900 1910 1930 1600 19<0 1950 I960 1970 I960 YEAR Figure 7. U.S. consumption of coal and production of synthetic chlorinated organics (includes ehloro- and dichlorobenzenes. 2,4-di- chloro- and 2.4,5-trlchlorophenoxyacetic acid, esters and salts, and pentachlorophenol) compared to the total PCDO and PCDF In the four Lake Huron cores as a function of time (all are plotted on a decade basis). ment mixing and the errors inherent in the dating and quantitation techniques. From these data, we conclude th at the input of dioxins and furans to the sedimentary environment is probably due to the combustion of chlorinated organic products present in various wastes. These wastes may be municipal wastes from Saginaw, Bay City, or other urban areas, or they may be industrial wastes from chemical manufac turing taking place in central Michigan. The direct dumping of chemical wastes (for example, from penta chlorophenol production) is an alternate, but unlikely, interpretation of our results. The agreement of the con gener and isomer profiles of PCDD and PCDF in the sediments with those in combustion effluents and in air particulates (see Table I) and the coincidence of the pro duction and concentration profiles (see Figure 7) are persuasive pieces of evidence that combustion is the major source. Direct dumping and coal or natural combustion may be real sources, but we believe them to be minor. In any case, it is clear th at the high levels of dioxins and furans found in presently accumulating sediments are not due to the "advent of fire." Acknowledgments We are grateful to B. J. Kimble for the fly ash samples, to, S. J. Eisenreich,- P. A. Meyers, C. P. Rice, and J. A* Robbins for the various sediment snmples, to' B. D. McVeety for instrumental assistance, and to S. L. Sikes for clerical support. R e g istry No. TCD D , 41903-57-5; PnCDD, 36088-22-9; HxCDl), 34465-46-8; H pC D lI, 37871-00-4; OCDD, 3268-87-9; TCDF, 55722-27-5; PnCDK, 30402-15-4; HxCDK, 55684*94-1; HpCDF, 38998-75-3; OCDK. 39001-02-0. Literature C ited (1) Kiminig, J,; Schulz, K. H. Dermalolo/;ica 1957,115, 540-546. (2) Schweiz, B. A.; Norris, J. M.; Sparscliu, G. L.; Rowe, V. K.; Gehring, P. J.; Emerson, J. L.; Gcrhig, C. G. Ado. Chem. Scr. 1973, No. 120. 55-69. (.1) Van Miller, J. I1.; 1niirli, J. .1.; Alien, .1. li. G/ieaias/Wicri* 1H77. f.`, 537- 541. (-11 fin pi*. l. N.; Vds, .J, G.; Muore, J. A.; Zinkl, J. C.; Ileill>t'k, fi. C. EH I` ilarintn. Hi'atth {`t'rspi'cl. 1973, 5, 1115-1-IO. (5) Blaser. W. W.; Brcriewcg, U. A.; Shadoff', J,. A.; Steli), li, H. Anni. Cheta. 11)75, -ffi, 984-986. (6) Buser, H. l. J . Chronwlngr. 11175, 107, 295-310. (7) Buser, JI. K.; BosshnrrU, H. P. J. Assnc. Off. Anal. Cileni. 11)75, 59, 502-509, (8) Pappe, C.; Buser, H. l.; Boashardt, H. P. C/iemosp7iere 197H, 7, 431-438. (9) Buser, H. R.; Bosshardt, H. P. J. Chrumalogr. 1974, 90, 71-77. (10) Yamagishi, T.; Mynzaki, T.; Akiyamrt, K.; Murila, M.; Nnkugmva, J.; Htsrii, S.; Kaneko, S. C/iernosp/icre 1981,10, 1137-114-1. (11) Vos, J. C\; Kocrnan, J. H.; Van der Mmis, H. L.; len Noe ver de Brnuw; de Vos, R:-H. Eovd Costavi. Toxicot. 1970, 8, 025-033. (12) Carreri, V. In "Dioxin: Toxicological and Chemical Aspects"; Cattaben, F,; Cavallaro, A.; Galli, G., Eds.; Spectrum Publications, Ine.: New York, 1978; pp 1-4. (13) Baughrnan, R. VV.; Meselson, M. Eiwiron. Health Perspect. 1973.5, 27-35. (14) Carter, C. D.; Kimbough, R. D.; Biddle, J. A.; Cline, R. E.; Zack, M. M.; Barthel, W. F.; Koehler. R. E.; Phillips, P. E. Science (W ashington, D.C.) 1975, 188, 738-740. (15) Olie, K.; Vermeulen, P. L.; Hutzinger, O. Chemosphere 1977, 6, 455-459. (16) Buser, H. R.; Bosshardt, H. P.; Rappe, C. Chemosphere 1978, 7, 165-172. (17) Eiceman, G. A.; Clement, R. E.; Karasek, F. W. Anal. Chem. 1979, 51, 2343-2350. (18) Caveliaro, A.; Bandi, G.; Invernizzi, G.; Luciani, L.; Mongni, E.; Gorni, A. Chemosphere 1980, 9, 611-621. t lin Hiitub. H. H.; Cniinnn-u. **''. B.; lu n e , S. S.; Glrdhil, ,1. H., H u m m e l. H. 11.; Kngi-l. li. (J.; Lampiirski, !.. L.; Luoma, K. Y.; Miller, 1). 1..; NV-m rii-k, T. !.; Sliadnl, I. A.; Ktehl, R. H,; Wt h<!, .1. S. .Sri i .'Jci' Washington, D.C.) 11180, 210, ;1S5 .j.'jn Chem. Eng. A'mi's 1979, 57 (7), 23-29. (211 Kimble, B. Gross, M. I*. Science (W ashington, D.C.) 198(1, 207, 59-61. (22) Junk, G, A.; Richard, -J. .1. Chemnsphcre 1981, 10, 1237-1241. (23) Buser, H. R. Chemosphere 1979, 8, 415-424. (2-1) Buser, fl. R. J. Chmmatagr. 1975, Ip, 95-108. (25) Buser, H. R.; Bosshardl, H. P.; Rappe, C.; Lindahl, R. Chemosphere 1978, 7, 419-429. (20) Kearney, P. C., presented al the 2nd International W ork shop on Chlorinated Dioxins and Related Compounds, Arlington, VA, cL 25-29, 1981. (27) Dougherty, U. C. liiumcd, Mass Spectrum. 1981,8, 283-292. (2S) Robbins, .J. A.; Edging Io, I). N. Cevchim. Cosmochim. Acta 1975, 39, 285-304. (29) ACS Committee on Environmental Im provement Anal. Chem. 1980, 52, 2242-2249. (30) Chrisp, C. E.; Fisher, G. L.; Lam mert, J. E. Science (Woshington, D.C.) 1977, 199, 73-75. (31) "Compilation of Air Pollution Emission Factors", 2nd ed.; U.S. EPA: W ashington, DC, 1973. (32) Hiles, R. A.; Laflamme, R. E.; Farrington, J. W. Science (W ashington, D.C.) 1977, 198, 829-831. (33) M iner. Ycarb., U.S. Bureau of Mines, 1870-1980. (34) "Production and Sales of Synthetic Organic Ghemicals"; U.S. T ariff Commission: W ashington, DC, 1919-1980. Received far review J u n e 13, 1983, Accepted October 28, 1983. This work was supported by the U.S. D epartment of Energy (iGrant 80EV-10449). Reduction and Dissolution of M an g an e se (III) and Manganese(IV) Oxides by Organics. 1. Reaction with Hydroquinone Alan T. Stone* and James J. Morgan W. M. Keck Laboratories of Environmental Engineering Science, California Institute of Technology, Pasadena, California 91125 The chemical processes by which manganese oxides are solubilized by reduction in anoxic waters are poorly un derstood. A study of the reduction and dissolution of manganese oxide suspensions by hydroquinone was un dertaken to determine the rate and mechanism of the solubilization reaction. Dissolution of the manganese(III,rV) oxide suspension by hydroquinone in the pH range 6.5 < pH < 8.5 is initially described by the following em pirical rate law: d[Mn2+] /d t = fcl|H+|(M6[QH2]10([MnO1]0 - [Mn!+]) where [Mn2+] is the dissolved manganese concentration, [QH2] is the hydroquinone concentration, and {MnOJj, is the amount of manganese oxide added. The apparent activation energy was found to be at +37 kJ/raol. Calcium and phosphate inhibited the reaction, by adsorbing on the oxide surface. A model is proposed for the observed rate dependence, according to which complex formation be tween hydroquinone and manganese oxide surface sites occurs prior to elecLron transfer. Introduction Within the pH range of natural waLers, Mn(III) and " To whom correspondt>nee should be addressed at the Depart ment erf Geography and Environmental Engineering, The Johns Hopkins University, Baltimore, MD 21213. Mn(IV) form sparingly soluble oxide/hydroxide solid phases, while Mn(II) is soluble. Kor this reason, dissolution reactions I and 2 greatly enhance the mobility of manga- MnOOH(s) + 3H+ + e" = Mn2+ + 2H20 (1) M n02(s) + 4H+ + 2e" = Mn2+ + 2H20 (2) nese in natural systems. Although oxygenation of Mn2+ has been studied extensively U-3), reduction and disso lution reactions are poorly understood. The purpose of this work is to systematically explore the factors th at in fluence how quickly manganese oxides are reduced and dissolved under natural conditions. Natural organic compounds have been found to reduce a variety of inorganic species and are the most readily available reductants in most natural systems. Soil fulvic acids have been shown to reduce Hg(II) to Hg(0), Fe(III) to Fe(II), and i2 to l' ). A number of studies have shown th a t organic compounds with structures similar to natural organics reduce and dissolve manganese oxides (5-7). Generation of radicals by reaction of manganese dioxide with hydroquinone was studied by Kukuzumi et al. (8, .0) and Ono et al. (JO). The rule of semiquinnne radical formation was found to be first order with respect to hy droquinone concentration and initial manganese dioxide loading. A mechanism was proposed that involves hy drogen atom abstraction from hydroquinone (0).