Determination of organic compounds leached from municipal

---

Anal. Chem. 1987, 59, 1027-1031

2. Cyt, cytosine 3. Thy, thymine 4. Ade, adenine

Nucleosides 5. Urd, uridine 6. Cyd, cytidine 7. dThd, deoxythymidine 8. Guo guanosine 9. Ado, adenosine 10. dpT, deoxypolythymidine

LITERATURE CITED Jorgenson, J. W.; Lukas, K. D. Science 1983, 222, 266-272. Hjerten, S.;Zhu, D. M. J . Chromatogr. 1985, 346, 265-270. Tsuda, T.; Nomua. K.; Nakagawa, G. J . Chromatogr. 1982, 248, 241-247. Terabe, S.;Otsuka, K.; Ichikawa, K.; Tsuchiya, A.; Ando, T. Anal. Chem. 1984, 56, 111-113. Lauer, H. H.; McManigill, D. Anal. Chem. 1988, 58, 166-170. Gassmann, E.: Kno, J. E.;Zare. R. N. Science 1985, 230, 813-815. Hjerten, S. J . Chromatogr. 1985, 347, 191-198. Aveyard, R.; Haydon, D. A. An Introduction to the Principles of Surface Chemistry; Cambridge University Press: Cambridge, 1973; Chapter 2. Everaerts, F. M.; Mikkers, F. E. p.; Verheggen, Th. p. E. M.; Vacik, J. In Chromatography,Part A ; Heftmann, E., Ed.; Eisevier: Amsterdam, 1983; Chapter 9.

1027

Giddings, J. C. S e p . Sci. 1969, 4 , 181-189. Dizaroglu, M.; Hermes, W. J . Chromatogr. 1979, 171, 321-330. Brown, P. R.; Grushka, E. Anal. Chem. 1980, 52, 1210-1215. Gel fiectrophoresis of Nucleic Acids: A Practical Approach; Rickwood, D., Hames, B. D., Eds.; IRL Press: Washington, DC, 1983. Tsuda, T.; Hakagawa, G.; Tato, M.; Yagi, K. J . Appi. Biochem. 1983, 5 , 330-336. Terabe, S.; Otsuka, K.; Ando, T. Anal. Chem. 1965, 5 7 , 834-841. Oko, M. U.; Venabie, R. L. J . Coiioid Interface Sci. 1971, 35, 53-59. Hunter, R. J. Zeta Potential in Colioid Science; Academic: London, 1981; Chapter 3. Scheller, K. H.; Hofstetter, F.; Mlchell, P. R.; Prijs, B.; Sigel, H. J . Am. Chem. SOC. 1981, 103, 247-260. Metal Ions in Bioiogicai Systems; Sigei, H., Ed.; Marcel Dekker: New York, 1979; Voi. 9. (20) Iler, R. K. The Chemistry of Silica ; Wiley: New York, 1978. (21) Wakatsuki, T.; Furukawa, H.; Kawaguchi, K. Soil Sci. Piant Nutr. (Tokyo) 1974, 2 0 , 353. (22) Burgess, J. Metal Ions in Solutions; Halsted Press: New York, 1978.

RECEIVED for review September 23,1986. Accepted December 18, 1986. B. L. Karger gratefully acknowledges support by the James L. Waters Chair in Analytical Chemistry, the National Science Foundation, and Dow Chemical Company. J. A- Smith gratefully acknowledges support by I'bXhst Aktiengesellschaft (FRG). This is Contribution No. 299 from ~and Materials ~ l the Barnett Institute of Chemical ~ ^. Science.

Determination of Organic Compounds Leached from Municipal Incinerator Fly Ash by Water at Different pH Levels F. W. Karasek,* G . M. Charbonneau, G . J. Reuel, and H. Y. Tong Department of Chemistry, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1

This study concerns the possiblilty of organlc compounds entering the environment through the leaching of municipal Incinerator fly ash wtth water. A Soxhlet extractlon of fly ash with water, followed by a benrene/water solvent extractlon was used to Isolate organlc compounds. The pH of the extracting llquld was varled (pH 4, 7, 10) and both the type and amount of compounds extracted dlffered. Many organlc compounds Including polychlorinated dibenzo-p -dloxlns, polychlorinated dlbenzofurans, polycyclic aromatic compounds, phenols, and hydrocarbons were found in the water extracts. Gas chromatographylfiame ionization detectton, gas chromatography/electron capture detectlon, and gas chromatography/mass spectrometry uslng electron Impact, posltlve and negative Ion chemical lonlzatlon techniques were used for compound identlficatlon and quantltatlon.

Fly ash, a fine particulate effluent from municipal incinerators, is the major byproduct produced from the burning of municipal waste. Approximately 35 000 tons of fly ash are produced for each million tons of waste incinerated (1). Considering the amount of municipal garbage incinerated in cities worldwide, the quantity of fly ash is significant. More than 600 organic compounds are known to be adsorbed on these particulates, 200 of which have been identified (2). Although electrostatic precipitators are used to remove the fly ash from the flue gas, approximately 1-2% of the produced

fly ash escapes to the atmosphere (I). Most studies have been carried out to determine the effect of the escaping fly ash (34). Little if any investigation has been done regarding the potential harmful effects of the disposal of the majority of the fly ash in landfill sites. Much of the fly ash buried is constantly exposed to rainwater, groundwater, and possibly wastewater as well. This exposure raises an important question as to whether organic pollutants in the fly ash could be removed by water and eventually contaminate the environment. Little if any investigation has been done regarding this possibility. Soxhlet extraction of the fly ash was used to resemble leaching conditions in which fly ash comes into contact with water and to provide a concentrated extract from which organic compounds could easily be isolated. Fresh solvent is passed through the fly ash during each cycle, and an exhaustive extraction would show the extent of leaching in the most serious case, where fly ash is constantly exposed to an aqueous environment. Since the pH of water greatly affects the type of compounds extracted in a typical solvent extraction, the p H of the extracting water was also altered. By the addition of formic acid and 2-methylethylamine, extractions were carried out a t a pH of 4 and 10 in addition to pH 7. Organic compounds were then transferred t o an organic solvent. A gas chromatographic analytical method consisting of five detection techniques was used for the analysis of organic compounds in the water extracts; flame ionization detection (GC/FID); electron capture detection (GC/ECD); mass

0003-2700/87/0359-1027$01.50/0 0 1987 American Chemical Society

~

1028

ANALYTICAL CHEMISTRY, VOL. 59, NO. 7, APRIL 1, 1987

spectrometry with electron impact ionization (GC/EIMS); mass spectrometry with positive ion chemical ionization (GC/PICIMS); mass spectrometry with negative ion chemical ionization (GC/NICIMS). Complementary data from these techniques facilitated the identification of a number of organic compounds in the water extracts.

EXPERIMENTAL SECTION Solvents, Standards,and Fly Ash Sample. All solvents used were "distilled in glass" grade, supplied by Caledon Laboratories (Georgetown, Ontario, Canada), with the exception of the water, which was deionized tap water. The standards of 1,2,3,4-tetrachlorodibenzo-p-dioxin (1234-TCDD), 1,2,3,4,7&hexachlorodibenzo-p-dioxin (123478-H6CDD), 1,2,3,4,6,7&heptachlorodibenzo-p-dioxin (1234678-H7CDD),and octachlorcdibenzo-p-dioxin (OCDD) were all purchased from Ultra Scientific (Hope, RI). The standard of 1,2,3,4,7-pentachlorodibenzo-p-dioxin (12347-p5CDD) was obtained from Cambridge Isotope Laboratories, Inc. (Woburn, MA). The normal alkane standards were purchased from Polyscience co. (Niles, IL), and their purities were greater than 98%. The formic acid, sulfuric acid, and sodium hydroxide were purchased from J. T. Baker Chemical Co. (Phillipsburg, NJ) and 2-methoxyethylamine was purchased from Aldrich Chemical Co. (Milwaukee, WI). The fly ash sample was collected from the electrostatic precipitator of a municipal incinerator in n Toronto, Ontario, Canada, and was provided by the Ontario Ministry of the Environment. Soxhlet Extraction of Fly Ash with Water. Approximately 300 mL of benzene was run through the Soxhlet apparatus for 1h prior to use, followed by 300 mL of water run for several hours to ensure the removal of contaminants. To prevent bumping during the Soxhlet extraction, a magnetic stirring apparatus using a Teflon stir bar was used to mix the water throughout the extraction. A 50-g sample of fly ash was placed in a coarse-porosity fritted glass extraction thimble and extracted smoothly for 48 h with 500 mL of water in the Soxhlet apparatus. During the extraction, the condensor temperature was maintained lower than 16 "C. No fly ash escaped through the fritted thimble during the extraction. The glass surfaces of the Soxhlet extractor and the outside of the thimble were then rinsed with a 50-mL volume of benzene, which was transferred to a separatory funnel. A coarse-porosity thimble was then used to filter the water extract to remove precipitated salts and the extract was then transferred to the separatory funnel containing the benzene. Several water rinses of the round-bottom flask were also filtered through the thimble and 25 mL of benzene was forced through the thimble by applying pressure. The round-bottom flask now free of precipitated salt was rinsed several times with a total of 25 mL of benzene, which was added to the separatory funnel. For the extraction at pH 7, deionized tap water only was used. The extraction at pH 4 was accomplished by adding formic acid prior to extraction and that a t pH 10 by adding 2-methoxyethylamine. After Soxhlet extraction, the p H 4 and 10 water extracts were neutralized by adding sodium hydroxide and sulfuric acid, respectively, prior to benzene solvent extraction. The extraction was repeated at each pH level and a blank was also obtained at each level. The extraction of fly ash was repeated 7 times at neutral pH to determine the levels of polychlorinated dibenzo-p-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF) leached into water. Benzene-Water Partitioning. After the Soxhlet extraction, enough benzene was added to the separatory funnel to bring the total volume of benzene to 150 mL, and this volume was used as the first of three 150-mL aliquots of benzene used for solvent extraction. The three aliquots of benzene extract were combined and the volume was reduced to approximately 15 mL by rotary evaporation under reduced pressure. The extract was then transferred to a cone-bottom flask using a Pasteur pipet, followed by several benzene rinses of the round-bottom flask, and the volume was further reduced to approximately 5 mL in the same manner. The organic phase in the cone (water was still present) along with several benzene rinses was transferred with a Pasteur pipet to a 50-mL pear-shaped flask. The volume in the flask was reduced to approximately 1mL by using rotary evaporation under reduced pressure and the extract transferred with several benzene

rinses to a calibrated sample vial. The volume was further reduced to 50 yL under a gentle stream of high-purity nitrogen and the vial was sealed and stored at -15 "C prior to analysis. The recovery studies of benzene-water partitioning were performed on a PCDD standard mixture containing 1234-TCDD, 12347-PSCDD,12347&H&DD, 1234678-H7CDD,and OCDD. The standard solution was diluted to various concentration levels in order to study the recoveries a t different concentrations. When the limit of quantitation for PCDD using GC/ECD was reached, the concentration of standard solution spiked in a 250-mL volume of water was approximately 2 pptr (parts per trillion). For the recovery the flask was shaken and allowed to sit for 30 min; then the solvent extraction procedure was performed as described except that the last two benzene aliquots were 50 mL each instead of 150 mL. A final volume of 20 yL was used for GC/ECD analysis. Gas Chromatographic Analysis. GC analyses were carried out on a Hewlett-Packard HP5880A gas chromatograph equipped with a flame ionization detector (FID). A cool on-column injector and a 30 m X 0.32 mm i.d. DB-5 fused silica capillary column (FSCC) (J&W) Scientific, Rancho Cordova, CA) were used for the analyses. The GC conditions were as follows: initial oven temperature, 80 "C, held for 1min, programmed to 300 "C at 5.0 "C/min and held for 10 min. The FID detector temperature was 350 "C and the ECD detector temperature was 325 "C. A helium carrier flow rate of 3 mL/min a t room temperature was used. QuantitativeAnalysis. Quantification was based on a method described by Tong and Karasek, in which average response factors (RF) using flame ionization detection (FID) are calculated for various compound classes (7). Gas Chromatograph/Mass Spectrometric Analysis. GC/MS analyses were performed with Hewlett-Packard HP5987A GC/MS system with an HPlOOO data system and an HP7914 Winchester disk drive. The instrument was operated with electron impact ionization (EI), positive ion chemical ionization (PICI), and negative ion chemical ionization (NICI) modes. A direct interface kept a t 300 "C linked the HP5880A GC to the mass spectrometer. In linear scan mode, ranges from 50 to 500 amu and from 60 to 500 amu at a scan speed of 500 amu/s were used for GC/EIMS and GC/CIMS analyses, respectively. The column and oven conditions were the same as those used for the GC analysis. For the E1 mode, an ionization voltage of 70 eV and a source temperature of 250 "C were used. In the CI mode, the reagent gas was ultra-high-purity methane (Linde Specialty Gas, Union Carbide, Canada). The electron energy was set a t 230 eV and ion source pressure at 1.6 torr. The source temperatures were 200 "C for the PICI mode 100 "C for the NICI mode. Selected ion monitoring (SIM) was used for the PCDD and PCDF analysw using E1 and NCI. Dwell times of 50 ys were used for each mass. The identification of PCDD and PCDF was confirmed by retention and by the ratios of the (Mt'-) and (M + 2 ) + / - mass chromatograms.

RESULTS AND DISCUSSION Soxhlet Extraction of Fly Ash with Water. Many organic compounds exhibit very low solubilities in water and are strongly adsorbed onto the surface of solid material. In order to simulate the leaching that may take place in a landfill site over an extended period of time, fly ash must be exposed to large volumes of water while maintaining efficient contact between the water and the fly ash. In the past most leaching studies of fly ash have focused on inorganic compounds. Since fly ash is composed of 70-95% inorganic material and many of these inorganic components exhibit considerable solubilities in water, conventional batch and column techniques are sufficient t o collect detectable quantities of inorganic material. Organic compounds, on the other hand, due to their low solubilities, require much larger volumes of water, making conventional batch and column techniques inadequate in many cases. In leaching studies, the level of organic compounds extracted into the water is limited by solubility and the adsorption energy t o the solid phase. For this reason large

ANALYTICAL CHEMISTRY, VOL. 59,

NO. 7,APRIL 1, 1987

TCDD

Table I. Number of Water Extraction Samples in Which PCDD Were Found

1029

TCDF /I

I

no. of samples in which PCDD were pH 4

7 10

no. of extracts

GC/ECD

5 7 2

2 5 1

detected GC/EIMS

GC/NICIMS

1 2

1 2

ND"

ND"

-14001 - I

20

" ND, nondetectable at the instrument detection limit.

-r--r---

24

16

32

36

20

24

26

32

38

20

24

26

32

16

26

32

36

1360

-

->

- - I 20

m 24

Flgure 2. GCIEIMSISIM data Illustrating the identification of TCDD and TCDF in a water extract of fly ash.

+A ,oo

200

b

300

IEMP1.C)

y-

Flgure 1. GC/ECD chromatograms of (a) water extract of fly ash and (b) water extract blank. For chromatographlc conditions, see Experimental Section.

volumes of water are required to assess the levels of leached compounds making batch and column techniques laborious and time-consuming. The water in a Soxhlet extractor on the other hand is constantly recycled and a small volume of water can be used. Organic compounds with lower water solubilities can in this manner be leached and isolated in the flask where accumulation occurs. During the extraction only 500 mL of water is used and the extract is concentrated and easily handled. In our experiments one cycle took 20 min and approximately 150 cycles took place during the extraction. PCDD/PCDF in Water Extracts. A total of 14 Soxhlet extractions were carried out to assess the leaching of PCDD and PCDF from fly ash. In these 14 water extracts, PCDD were detected in seven samples by GC/ECD and positively identified in three samples by GC/EIMS and GC/NICIMS analyses. The results are listed in Table I. Among the analytical techniques used, GC/ECD provided the highest sensitivity, a subpicogram detection limit for PCDD and PCDF. However, because of the presence of interfering compounds in the water extract, TCDD and P&DD could not be identified on the GC/ECD chromatograms. A typical pattern of H,CDD, H,CDD, and OCDD, which was usually observed in a benzene extract of fly ash, could be easily recognized on the chromatogram. Figure 1 shows a typical GC/ECD chromatogram of a water extraction sample containing PCDD. A corresponding GC/ECD chromatogram of an extraction blank is also shown on the figure. No PCDD impurities were found in this blank. The presence of PCDD and PCDF in the water extracts wm further confirmed by GC/EIMS/SIM and GC/NICIMS/SIM analyses. The GC/MS analyses had a lower sensitivity for PCDD and PCDF determination compared with GC/ECD and could provide unambiguous identification of PCDD and PCDF for two samples. All congener groups from TCDD to OCDD as well as the corresponding PCDF were found in the

water extract. The GC/EIMS/SIM data in Figure 2 illustrate the identification of TCDD and TCDF found in a water extract sample. No PCDD or PCDF were found in the water blank analyzed by GC/MS/SIM. A large variation exists in the concentrations of PCDD and PCDF found among these seven extraction samples. This variation is most likely caused by the instability of trace amounts of PCDD and PCDF in water and also by the problems encountered in trace analysis at the parts-per-trillion level. On the basis of the limited results obtained in the GC/ECD, GC/EIMS/SIM, and GC/NICIMS/SIM analyses, the concentration of PCDD leached into water is estimated t o be in a range of sub part per trillion to 400 pptr for each congener group from TCDD and OCDD as well as the corresponding PCDF. The recoveries of the PCDD standards from the benzenewater partitioning at the 2 pptr concentration level were 90% for H,CDD, 66% for H,CDD, and 76.9% for OCDD, with relative deviations from the mean of 6%, 26%, and 8%, respectively. Because of interferences, the recovery study was limited to the above standards. Due to the small deviations of the mean, the erratic recoveries must be due to the Soxhlet extraction procedure. PCDD and PCDF are leached from fly ash by water a t p H 4,7, and 10. The precise quantity of PCDD and PCDF is difficult to determine because the levels of PCDD and PCDF are very low, and a t such low levels, their tendency to adsorb onto glassware can present a problem for a reproducible quantification. Compound Identification. Many organic compounds were identified in the water extracts. The data obtained by GC/FID and GC/ECD were mainly used for compound quantification, but through the use of a hydrocarbon standard they also provided retention indexes for qualitative analyses. The identification of most compounds was primarily based on the data obtained from the three GC/MS analyses. Since three GC/MS analyses of a given sample were achieved on the same instrument and under the same GC oven conditions, the corresponding peaks of an unknown compound on the three total ion current (TIC) chromatograms were easily located. Figure 3 illustrates the three TIC chromatograms of one sample analyzed by GC/MS with the three ionization techniques. The combination of GC/EIMS, GC/PICIMs, and GC/NICIMS analyses generated a wealth of complementary information for compound identification. The electron impact mass spectrum was analyzed by use of an extensive library search system based on probability based matching against 70 000 reference mass spectra. Compounds Found in Water Extracts at Different pH Levels. The leaching of fly ash with water was studied under

1030

ANALYTICAL CHEMISTRY, VOL. 59, NO. 7, APRIL 1, 1987

Table 111. Compounds Identified in pH 7 Water Extract retention time, min

concn, ng/g of fly ash

dichloroethenylmethylbenzene

7.70 9.25 9.67 9.90 10.51 11.13 12.12 12.90 13.54 14.31 14.70 15.72

1.7 2.9 21.0 31.0 6.9 35.0 3.5 1.0 12.0 0.16 23.0 0.78

dichlorobenzamide dimethylbiphenyl

17.26

4.4

20.15 21.23

1.1 6.4

26.12 27.65 27.94 29.70 31.42 33.02 34.58 34.73 36.09 37.54

0.14 0.62 0.24 0.19 0.76 0.64 0.86 2.4 0.43 0.40

40.32 41.64 42.93 44.16 45.39 46.72

0.44 0.50 0.66 0.55 0.49 0.44

compound

j1

dichloropropenamide octene ketone ethylbenzene tetrachloropentene trichlorophenol trichlorophenol trichlorophenol methylbiphenylamine

I

GC/PICIMS

120000:

hydroxymethylbenzaldehyde ethenyltetrachlorobenzene

trichlorocresol dimethoxybenzaldehyde dimethyl phthalate tetrachlorophenol

I

12ccs0 1

methylbenzenedicarboxylic acid

GC/NICIMS

CDOCC

R

I2

15

23

24

2S

32

3E

4C

44

18

51

Figure 3. Total ion current (TIC) chromatograms of water extract of

fly ash analyzed by GCIEIMS, G U P I C I M S , and GCINICIMS.

Table 11. Compounds Identified in pH 4 Water Extract retention time, min

concn, ng/g of fly ash

4.25

1.9

6.09 6.16 6.30 6.47 6.84

1.1 0.82 0.56 1.2 0.88

9.58 9.89

3.4 a.5

trichlorocresol acenaphthene dimethoxybenzaldehyde pentachlorobenzene tetrachlorophenol diethyl phthalate dichlorobenzamide

12.90 13.20 13.47

0.72 1.8 6.3

methylbenzenedicarboxylic acid

16.87 20.12 21.19 22.21

0.74 0.41 6.0 0.16

38.95 40.32 41.65 42.92 44.16 45.38 46.70

0.23 0.28 0.44 0.47 0.47 0.43 0.36

compound dimethylphenol dichloropropenamide dimethylhexenone octene alcohol ethylbenzene dichlorobutadiene octene ketone tetrachloropentene bromobutanol trichlorophenol Cs-phenol 2-ethylbiphenyl ethenyltetrachlorobenzene

octadecane caffeine nonadecane dimethoxyphenanthrene nonacosane triacontane hentriacontane dotriacontane tritriacontane tetratriacontane pentatriacontane

15.02 15.68

12.0 1.8

acidic (pH 4), neutral (pH 7), and basic (pH 10) conditions. These p H levels cover the normal range found in waste disposal sites. T h e pH modifiers, formic acid and 2-methoxyethylamine were chosen because of their solubility in water and their boiling points (100.7 and 95 "C, respectively). These boiling points are close t o that of water (100 "C), and therefore

pentachlorophenol octadecane caffeine nonadecane eicosane heneicosane dimethoxyphenanthrene docosane tricosane tetracosane pentacosane hexacosane pentachlorochrysene heptacosane octacosane nonacosane triacontane hentriacontane dotriacontane tritriacontane tetratriacontane Dentatriacontane

these chemicals will affect the condensed water pH during a Soxhlet extraction. After several extraction cycles, the pH values of water in the thimble and in the round-bottom flask were similar. T h e extracts were neutralized prior to benzene-water partitioning t o promote the extraction of ionic species from the aqueous phase. Figure 4 shows the GC/FID chromatograms of the organic extracts obtained from water at pH 4,7, and 10. These three traces show the trend toward more organic material being removed from fly ash as the p H level of the water increases. T h e organic compounds identified in the different extracts differ somewhat as well. Tables 11,111,and IV list the compounds identified and quantified in pH 4,7 , and 10 extracts, respectively. T h e levels are reported in nanograms of compound leached per gram fly ash. Those compounds that are not quantified did not show distinguishable peaks on the GC/FID chromatogram and were identified by GC/NCI. A number of compounds are present in all three extracts. For example, a tetrachlorophenol isomer is present at 12 ng/g of fly ash in the p H 4 extract, 23 ng/g at p H 7, and 43 ng/g at p H 10. The increase in tetrachlorophenol concentration with increasing p H reflects the general trend of increasing amounts of organic compounds present as p H increases. There are similarities in the composition of the three extracts, such as the presence of low concentrations of n-alkanes. T h e low alkane concentrations are not unexpected due to their low solubilities in water. Some compounds were unique t o one particular extract; however, most compound classes present are represented in all three extracts. Blank extractions done

ANALYTICAL CHEMISTRY, VOL. 59, NO. 7, APRIL 1, 1987

1031

Table IV Compounds Identified in pH 10 Water Extracts compound methoxymethylbenzonitrile hydroxybiphenyl

trichlorophenol methylbiphenylamine hydroxymethoxybenzaldehyde ethenyltetrachlorobenzene

diphenylmethane trichlorocresol dimethoxybenzaldehyde methylnaphthalene tetrachlorophenol tetrachlorophenol dichloroethylmethylbenzene

retention time, min

concn, ng J g

6.48 9.75 9.85 10.54 11.04

14.0 4.8 17.0 13.0 1.7

12.21 12.88

1.0 1.6

14.41

42.0

14.99 15.69

43.0 0.88

19.30

3.8

20.10 21.30 22.02

0.93 10.00 1.3

27.93 30.44 34.55 36.07 37.52 38.94 40.30 41.63 42.90 44.15 45.37 46.69 48.23

4.4 0.31 0.31 0.46 0.43 0.44 0.63 0.70 0.65 0.59 0.48 0.35

of fly ash

dichlorobenzamide pentachlorophenol

methylphenylindole octadecane caffeine phenylfluorene dimethoxyphenanthrene

docosane diphenylpyridine hexacosane heptacosane octacosane nonacosane

triacontane hentriacontane dotriacontane

tritriacontane tetratriacontane pentatriacontane

hexatriacontane

1.7

a t each p H are very similar to each other in appearance and have a few peaks, and the compounds identified in the extraction blanks were not included in Tables 1-111. An important result of this study was the identification of polychlorinated compounds and toxic PAH in all three extracts. The levels of these compounds in the water were in the parts-per-trillion to low parh-per-billion range, which are several orders of magnitude lower than the levels found in the benzene extracts of similar fly ash from the incinerator. There are many differences between the leaching occurring in a Soxhlet extractor and that occurring in a landfill site. First, the leaching in the extractor occurs a t a much higher temperature than that underground. Many compounds may have been removed simply due to the heat, and the hot water is chemically different than water a t lower temperatures. Second, the effect of the surrounding matrix in a landfill site may attenuate by adsorption the compounds leached. The Soxhlet extraction allows for a rapid and systematic study of leached compounds but more study must be determined to see how these results correspond to actual leaching occurring in a landfill site.

CONCLUSION A Soxhlet extraction of fly ash with water was used to represent the leaching of organic compounds using water and

Flgure 4. GC/FID chromatograms of water extract of fly ash using water with three different pH levels.

may simulate the leaching of compounds in a landfill site with water. Water a t three different p H levels was used to investigate the effect of p H on the types and amounts of organic compounds extracted. A number of chlorinated and toxic compounds were identified in the extracts, and the pH of the water affects the type and amount of organic compounds removed. Although PCDD and PCDF were found in some extracts, their levels were very low. The method allows for the investigation of compounds that may be leached from fly ash in a landfill site.

LITERATURE CITED Karasek, F. W.; Onuska, F. I. Anal. Chem. 1982, 5 4 , 309A. Tong, H. Y.; Shore, D. L.; Karasek, F. W. J . Chromatogr. 1984, 285, 423. Rubel. F. N. Incineration of SolM Wastes, Pollut;on Technology Review No. 13; Noges Data Corp.: Park Ridge, NJ, 1974. Freeman, H. Environ, Sci. Techno/. 1978, 12, 1252. Esposito, M. P.; Tiernan, T. 0.; Dryden, F. E. "Dioxins"; U S . Environmental Protection Agency Report, EPA-600/2-80-197, 1980; p 122. Chlorinated Dloxins and Dibenzofurans in the Total Environment ; Chondhary, G., Keith, L. H., Rappe, C., Eds.; Butterworth: Toronto, 1983; p 125. Tong, H. Y.; Karasek, F. W. Anal. Chem. 1984, 5 6 , 2124.

RECEIVED for review January 31, 1986. Resubmitted November 11, 1986. Accepted December 8, 1986. This study was supported by a grant from the Ontario Ministry of the Environment, Canada.