EGN 104 CI CI a i Y i CI CI iii: eni I soil, clay, and caustic soda effects on solubility, sorption, and mobility of hexachlorocyclopentadiene S.F. Joseph Chou Robert A. Griffin :¥:¥:¥:¥:¥:¥:¥:¥:¥:¥:¥:¥:¥:¥:¥:¥:¥x::¥:: lliliiililii Prepared in cooperation with Municipal Environmental Research Laboratory U.S. Environmental Protection Agency ENVIRONMENTAL GEOLOGY NOTES 104 linois Department of Energy and Natural Resources STATE GEOLOGICAL SURVEY DIVISION 1983 Cover design: Sandra Stecyk Chou, S. F. Joseph Soil, clay, and caustic soda effects on solubility, sorption, and mobility of hexachlorocyclopentadiene/S. F. Joseph Chou and Robert A. Griffin. — Champaign, III. : Illinois State Geo- logical Survey, 1983. 53 p. ; 28 cm. — (Illinois— Geological Survey. Environmental geology notes; 104) 1. Solubility. 2. Adsorption. 3. Absorption. 4. Hexachlorocyclo- pentadiene. I. Title. II. Griffin, Robert A. III. Series. Printed by authority of the State of Illinois /1983/2000 d caustic soda effects on Solubility, sorption, and mobility of hexachlorocyclopentadiene S. F. Joseph Chou Robert A. Griffin Prepared in cooperation with Municipal Environmental Research Laboratory U.S. Environmental Protection Agency Cincinnati, OH 45268 ENVIRONMENTAL GEOLOGY NOTES 104 1983 ILLINOIS STATE GEOLOGICAL SURVEY Robert E. Bergstrom, Chief Natural Resources Building 615 East Peabody Drive Champaign, IL 61820 ACKNOWLEDGMENTS The authors gratefully acknowledge the U.S. Environmental Protection Agency, Municipal Environmental Research Laboratory, Solid and Hazardous Waste Research Division, Cincinnati, Ohio, for partial support of the study under Grant R-806335-01-1 ; Dr. Michael H. Roulier, Project Officer. The authors also thank, the Rohm and Haas Company, Philadelphia, Pennsylvania, for supplying the carbonaceous sorbent, Ambersorb XE-348; American Colloid Company for supplying the montmorillonite, International Mineral and Chemical Corporation for supplying the bentonite, and the Ottawa Silica Company for supplying the sand used in this study. The authors also wish to thank various members of the Illinois State Geological Survey (ISGS) for their assistance; Dr. N. F. Shimp, Chemical Group Head; the ISGS Analytical Chemistry Section, under the direction of Dr. R. R. Ruch; B. W. Fisher, Dr. M. I. M. Chou, and Dr. D. R. Dickerson of the Geochemistry Section. The authors also thank Dr. R. A. Larson of the University of Illinois and Dr. L. I. H. Wei of the Illinois Natural History Survey for performing mass-spectral analysis. DISCLAIMER The information in this document has been obtained through research funded in part by the United States Environmental Protection Agency under Grant Number R-806335 to the Illinois State Geological Survey, has been subject to the Agency's peer and administrative review, and has been approved for publication. The contents reflect the views and policies of the Agency. Mention of trade names or commercial products does not constitute endorsement or recommendation for use by either of the respective agencies. FOREWORD 1 ABSTRACT 2 INTRODUCTION 3 Chemistry of C-56 3 Manufacture and industrial uses 4 Environmental effects 4 Disposal and soil attenuation 5 CONCLUSIONS 6 RECOMMENDATIONS 7 METHODS AND MATERIALS 8 Experimental materials 8 Solubility studies 10 Photolysis and hydrolysis studies 1 1 Sorption studies 14 Mobility studies: determination by soil TLC and soil column leaching 15 RESULTS AND DISCUSSION 16 SOLUBILITY OF C-56 IN WATERS. SOIL EXTRACTS, AND LEACHATES 16 Solubility in waters and leachates 16 Effect of dissolved salts on the solubility 18 AQUEOUS CHEMISTRY OF C-56 21 Photolysis and hydrolysis 21 Identification of degradation products 25 SORPTION OF C-56 BY SOIL MATERIALS, ACTIVATED CARBON, AND AMBERSORB CARBONACEOUS SORBENT 41 Sorption of C-56 by soil materials 41 Effect of earth material TOC on sorption 45 Effect of soluble salts on sorption 46 MOBILITY OF C-56 IN SOIL MATERIALS: DETERMINATION BY SOIL TLC AND SOIL COLUMN LEACHING 47 Mobility of C-56 in soils 47 Soil column leaching study 49 REFERENCES 51 FIGURES 1. Solubility of C-56 in tap water as a function of time 16 2. GC chromatograms (A) deionized water soluble C-56 (B) tap water soluble C-56 17 3. Effect of dissolved salts on the water solubility of C-56 at 22°C— plot according to the Setschenow equation 20 4. First-order degradation plot of tap water soluble C-56 exposed to long- and short-wave UV light 21 5. Packed column GC chromatograms of variation in C-56 and degradation products in tap water at various pHs adjusted with either HCI or NaOH 22 6. GC chromatograms of C-56 and ketones separated using silica gel chromatography 23 7. GC chromatograms of C-56 in distilled water after exposing to sunlight 25 8. Mass spectrum of compounds II and III 27 9. Mass spectrum of compound IV 27 10. GC chromatograms of C-56 extracted from distilled water after exposure to sunlight 29 11. Mass spectrum of compound V 29 12. Mass spectrum of compounds VI and VII 30 13. Mass spectrum of compound VIII 30 14. GC chromatograms of C-56 in hexane solution exposed to sunlight 32 15. GC chromatograms of C-56 in methanol solution explosed to long-wave UV light 33 16. GC chromatograms of C-56 in methanol exposed to long-wave UV light 34 17. Mass spectrum of compound X 35 18. Mass spectrum of compound XI 35 19. Mass spectrum of compound XII 36 20. Proposed pathway for the degradation of C-56 in aqueous solution 37 21. GC chromatograms of C-56 in 0.5 N NaOH solution exposed to sunlight, then acidified with HCI 38 22. GC chromatogram of C-56 in distilled water extract 39 23. Mass spectrum of C 5 H 2 Cl4 40 24. Mass spectrum of C5H3CI3 40 25. Sorption of C-56 by 1.00 g and 3.50 g Catlin soil from tap water as a function of time 41 26. Freundlich adsorption isotherms of C-56 on seven soil materials from tap water 43 27. Freundlich adsorption isotherms of C-56 on four clays from tap water 44 28. Freundlich adsorption isotherms of C-56 on activated carbon and Ambersorb XE-348 from tap water 44 29. K p vs TOC for C-56 sorption by fifteen soil materials 45 30. Freundlich adsorption isotherms of C-56 on Flanagan soil from tap water, soil solutions, and caustic brine 46 31. Freundlich adsorption isotherms of C-56 from tap water, salt solutions, and caustic brine 46 32. Mobility of C-56 in two soils leached with DuPage leachate and acetone/ water mixtures (I) Avasoil (II) Catlin soil 48 33. GC chromatograms of standard C-56 and soil column leachate (A) C-56 standard (B) soil column leached with 64 inches of tap water 50 34. GC chromatograms of soil column leachate leached with (C) 128 inches of tap water (D) 192 inches of tap water 50 TABLES 1. Total organic carbon (TOC) analysis of waters used in C-56 solubility study 8 2. Sorbents used in hexachlorocyclopentadiene sorption studies 9 3. Conditions for gas chromatographic analysis 11 4. Salts and concentrations used in C-56 solubility studies in tap water 1 1 5. Solubility of C-56 at 22 C C after 15 hours equilibrium in waters and landfill leachates using contrifugation technique 18 6. Effect of dissolved salts on the water solubility of C-56 at 22°C and Setschenow parameter for C-56 19 7. Solubility of C-56 in caustic brine solution at 22°C 20 8. First-order rate constant, half-life, and correlation coefficient for the photodegradation of C-56 in distilled water, methanol, and hexane exposed to sunlight, long-wave UV light, and short-wave UV light 21 9. Photolysis of C-56 in hexane solution under laboratory lighting conditions 24 10. Percentage distribution of C-56 and impurities in hexane and water extracts under different pHs 24 11. Freundlich K N, correlation coefficient, and molar K for sorption of C-56 by various sorbents from tap water aqueous solution 42 12. Freundlich K,, N, correlation coefficient, and molar K for sorption of C-56 by Catlin soil and Flanagan soil from tap water, salt solutions, and caustic brine 47 13. Mobility of hexachlorocyclopentadiene in several soil materials leached with various solvents as measured by soil TLC 49 FOREWORD The Environmental Protection Agency was created because of increasing public and governmental concern about the dangers of pollution to the health and welfare of the American people. Numerous examples of noxious air, foul water, and spoiled land are tragic testimony to the deterioration of our natural environment. The complexity of that environment and the interplay of its components require a concentrated and integrated attack on the problem. This report summarizes the results of research funded in part by the Solid and Hazardous Waste Research Division, Municipal Environmental Research Laboratory, Cincinnati, Ohio and conducted by the Illinois State Geological Survey. The Municipal Environmental Research Laboratory is concerned with research to develop new and improved technology and systems to prevent, treat, and manage wastewater and other pollutant discharges from municipal and community sources, to preserve and treat drinking water supplies, and to minimize the adverse economic, social, health, and aesthetic effects of pollution. This publication is one of the products of that research. We view it as a vital communications link between the researcher and the user community. This publication describes the attenuation mechanisms and capacity of selected clay minerals and soils for hexachlorocyclopentadiene (C-56), the effect of caustic-soda brine on the attenuation and solubility of C-56, the aqueous chemistry of C-56, and a chemical model to predict C-56 migration through soil materials. Francis T. Mayo Director, Municipal Environmental Research Laboratory S. F. Joseph Chou Associate Organic Chemist Illinois State Geological Survey Robert A. Griffin Head, Geochemistry Section Illinois State Geological Survey ABSTRACT The aqueous chemistry, sorption, and mobility of hexachlorocyclo- pentadiene (C-56) in soil materials were studied in the laboratory. The solubility of C-56 in water, soil extracts, and sanitary landfill leachates ranged from 1.03 to 1.25 ppm. Sodium hydroxide (caustic soda), and sodium chloride decreased the solubility of C-56 in water as the con- centration of the salt increased; sodium hypochlorite slightly increased the solubility of C-56; and sodium perchlorate had no significant effects due to increasing salt concentration on the solubility of C-56. A caustic brine composed of a mixture of the salts was intermediate in decreasing solubility. The effect of individual salts upon C-56 solubility was approximately additive. The half-life of C-56 in water was approximately 3 months at both 22°C and 35°C, indicating little temperature dependence. When C-56 was exposed to sunlight and long-wave UV-light, the half -life of C-56 was less than 4 minutes in aqueous solution and less than 1.6 minutes in hexane or methanol solution. Although pH did not significantly affect the C-56 hydrolysis rate in aqueous solution, the presence of iron caused an increase in the hydrolysis rate. At least 12 products of photolysis and hydrolysis were identified. Pentachlorocyclopentenone and hexachloro- cyclopentenone were the major products found in distilled water. Cis- and trans-pentachlorobutadiene, tetrachlorobutenyne, and pentachloro- pentadienoic acid were the major products identified in mineralized water. Pentachlorocylopentenone, unstable under high temperature conditions in hexane solution and in water, was dimerized and then converted to hexachloroindone. Freundlich adsorption isotherm plots of C-56 sorption on soils, clay minerals, activated carbon, and Ambersorb XE-348 generally yielded linear regression lines with very high coefficients of correlation between the amount of C-56 adsorbed by a unit of adsorbent and the equilibrium concentration of C-56. The presence of salts in solution dramatically affected the sorption of C-56: brine, NaCl, and NaOH caused an increase in sorption while NaOCl caused a decrease in sorption. The increase in sorption was inversely related to the Setschenow parameter, indicating that the change in sorption was directly related to the effect of the salts on the solubility of C-56. The sorption capacity of C-56 was highly correlated with the total organic carbon (TOC) content of soil materials (r 2 = 0.97), which was the dominant soil characterization parameter; sorption appears to be predictable from the TOC content of soils. The mobility of C-56 in six soils was measured with several leaching solvents using soil thin-layer chromatography (TLC) and column leaching studies. C-56 remained immobile in the soils when leached with water, landfill leachate, or caustic brine but was highly mobile when leached with organic solvents. A further conclusion was that several degradation products of C-56 migrated through soils faster than C-56 itself and that the degradation products of C-56 warrant further study. INTRODUCTION Recent environmental legislation (the Resource Conservation and Recovery Act [RCRA] and others) has created a demand for information regarding the land disposal of toxic organic waste substances such as hexachlorocyclo- pentadiene (called HCCPD, Hex, PCL, or C-56). Small amounts of C-56 are occasionally present as impurities in pesticides and some has undoubtedly entered the environment in this way. The most likely route of entry into the environment arises from disposal of residues from the manufacture of products using C-56. Currently, major environmental concern is the disposal of the large quantities of industrial wastes containing C-56. Although C-56 has been used in the chemical industries for decades, little is known about its environmental impact on aquatic or soil systems. The purpose of this research was to determine the attenuation mechanisms and capacity of selected clay minerals and soils for C-56, to determine the effects of caustic-soda brine on the attenuation and solubility of C-56, to study the aqueous chemistry of C-56, and to develop a chemical model to predict C-56 migration through soil materials. Chemistry of C-56 Hexachlorocyclopentadiene is a dense, oily, pale to greenish-yellow liquid. Its molecular formula is C 5 C1 & ; its structure formula is: Hexachlorocyclopentadiene (C-56) C-56 has a molecular weight of 272.77; a vapor pressure of 1 mm Hg at 78-79°C, 0.1 mm Hg at 25°C (Atallah et al., 1980), and a density of 1.7119 (20°/4°C) (Ungnade and McBee, 1958). The solubility of C-56 in water has been variably reported as about 0.8 ppm (Lu et al . , 1975), 1.8 ppm (Zepp et al., 1979), and 2 ppm (Atallah et al., 1980). C-56 is extremely volatile and photoreactive in sunlight with near-surface half-lives in aquatic systems of less than 10 minutes (Zepp et al., 1979). C-56 shows an unexpected tendency to undergo the Diels-Alder reaction with many dienophiles at temperatures between 20° and 200°C. It condenses even with simple olefins, which normally do not react with dienes, and with polynuclear aromatic hydrocarbons, such as naphthalene and anthracene (Ungnade and McBee, 1958). Thus, the apparent disappearance of C-56 from the environment in routine multi-residue analyses should not be construed to mean that it is always degraded to smaller molecules, even though stable bioaccumulated metabolites of C-56 have been found (Lu et al . , 1975). Manufacture and industrial uses In the United States only two chemical companies have produced C-56: Hooker Chemical Company at Montague, MI, and Niagara Falls, NY; and Velsicol Chemical Corporation at Memphis, TN (SRI, 1978). Hooker sold hexachlorocyclopentadiene as C-56®; Velsicol's trade name is PCL; it is occasionally referred to in the literature as "hex" or as "hex waste". The total production of C-56 in the United States was approximately 50 million pounds in 1962 (Whetstone, 1964); Lu et al. (1975) estimated that the same amount was produced in 1972, and a report prepared for Hooker cited an annual production figure of about 50.5 million pounds (Equitable Environmental Health Inc., 1976). On the basis of these reports, the total production of C-56 in the 13-year period from 1962 to 1975 is estimated to be about 650 million pounds. The major uses of C-56 have been in the preparation of the chlorinated cyclodiene insecticides aldrin, dieldrin, endrin, chlordane, heptachlor, endosulfan, Kepone®, and mirex; the f ire-retardant monomers chlorendic acid (CA) and chlorendic anhydride (CAN), used primarily in polyester resins and to a lesser extent in alkyd resin coatings; and the f ire- retardant plastic additives known as Dechloranes® (Kirk-Othmer, 1964; U.S. EPA, 1977; Sanders, 1978). Although there are recently enacted governmental bans on the use of certain chlorinated pesticides, there is no documentation that the total production of C-56 has significantly subsided. The reason may be that C-56 continues to be used as a raw material in the production of fire retardants and of pesticides for export (Bell, Ewing, and Lutz, 1978 J National Academy of Sciences, 1978). Environmental effects The toxicological hazards, persistence, and widespread environmental contamination of some of the C-56-derived insecticides are well known (IARC, 1974); however, very little is known about the health and environmental effects of C-56, and the available information on CA, CAN, and the dechlorane materials is very meager. C-56 produces systemic toxicity of unknown mechanism in mammals via ingestion, inhalation, and dermal exposure. Degenerative changes in the brain, heart, adrenals, liver, kidneys, and lungs were observed in severely poisoned animals (U.S. EPA 1977). In a model ecosystem study, Lu et al. (1975) reported that C-56 has considerable environmental stability and moderate ecological magnification potential in aquatic organisms. Spehar et al. (1977) reported in a laboratory study that the bioconcentration factor was 11 for whole-body fathead minnows after 30 days exposure. C-56 showed adverse effects to the fathead minnows at concentrations as low as 3.7 ppb. Significant decreases in survival occurred at 7.3 ppb. C-56 was also toxic to mammals (Treon, Cleveland, and Cappel, 1955). An incident at a sewage treatment plant in Louisville, Kentucky — in which scores of workers experienced a variety of toxic symptoms following the improper disposal of C-56 manufacturing wastes — created a great demand for information on the effects of C-56 exposure on humans (Carter, 1977). C-56 has been identified and quantified in waste water, receiving streams, rivers, fish, soil, sediment, and air surrounding pesticide plants (Spehar et al., 1977; Swanson, 1976; Carter, 1977). C-56 also has proven to be a potent irritant. Industrial and laboratory workers have experienced irritation of the eye, irritation of the upper airway passages and head- aches upon exposure to C-56 vapors and burns upon contact of skin with the liquid (Ingle, 1953; Carter, 1977; personal experiences of the authors). Disposal and soil attenuation A literature search of Chemical Abstracts and Pollution Abstracts from 1970 to present yielded not one publication dealing specifically with the disposal, attenuation, or mobility of C-56 or "hex" waste in soil or soil materials. In unpublished literature, Weber (1979) reported that C-56 was strongly adsorbed by soil colloids and Rieck (1977a, 1977b) demonstrated poor extractabilty of C-56 from soil treated with C-56, which provides indirect evidence of strong adsorption. A major source of C-56 wastes is the manufacture of insecticides such as chlordane and heptachlor. The largest chlordane manufacturing plant In the U.S. is located in east-central Illinois. The liquid waste stream from the plant (considered typical waste for the industry) consisted of about 400,000 gallons per day that varied in chemical composition daily. Approximate organic composition consisted of a saturated suspension of hexachlorocyclopentadiene and small amounts of chlordane, hexachloro- benzene, heptaclor, and heptaclor epoxide. The mineral content averaged 12 percent sodium chloride, 4 percent sodium hypochlorite, and 3 percent sodium hydroxide (caustic-soda). The waste water pH was typically 11.9 to 12.0 (Gibb et al., 1978; Illinois EPA records; and personal communications). This mixture Is referred to as "hex waste". The "hex waste" is not only generated during chlordane manufacture, but also during production of resins, adhesives, pesticides, fire retardants, and other related products. The study of C-56, therefore, has broad applicability to many industries. There is a dearth of information regarding the attenuation and mobility of C-56 or "hex" wastes in soil or soil materials and of the effects of caustic-soda brines on the solubility and mobilty of this compound. There are virtually no data with which to assess the migration of "hex" wastes through soil or soil materials such as clay liners of waste impoundments or landfills. There are also complicating factors due to the effects of caustic brines on attenuation and solubility of C-56. For example, a short-term study commissioned by the plant showed that activated carbon was not feasible for removal of C-56 from the waste water because of interference from the sodium hypochlorite. Hypochlorite concentrations had to be less than 100 ppm to obtain adequate adsorption of C-56 by the carbon; plant waste waters usually contain 30,000 to 40,000 ppm of hypochlorite. Caustic brines may also change the solubility of the C-56. The phenomenon of salting-out of a nonelectrolyte from an aqueous solution by addition of salt is well known. An empirical equation that yields a good first approximation of salting-out data that would be suitable for use in predicting environmental contamination is that of Setchenow (1892): S log — = Km in which S is the solubility in pure water, S is the solubility in a salt solution of molality m, and K is a constant sometimes called the "salting coefficient". Determination of the salting coefficient experimentally for the "hex" waste would then allow prediction of the solubility of C-56 from a knowledge of brine concentration. Since sorption of C-56 by soil or clay is a function of concentration, knowledge of how the solubility changes with brine concentration is essential if accurate predictions of migration and mobility are to be made. CONCLUSIONS Studies of the aqueous chemistry, sorption, and mobility of C-56 have led to the following conclusions: . C-56 reached saturation concentration in water in less than 15 hours. . The solubility of C-56 in tap water, Sugar Creek water, soil extracts, and landfill leachates ranged from 1.03 to 1.25 ppm. The relatively small increases in the solubility of C-56 were related to the increasing total organic carbon (TOC) content of the respective waters. . Increasing concentrations of sodium hydroxide and sodium chloride decreased the solubility of C-56 in water, while increasing concentrations of sodium hypochlorite slightly increased the solubility of C-56 in water. A caustic brine composed of a mixture of the above salts was intermediate in decreasing solubility. Sodium perchlorate was inert, and increasing its concentration did not significantly affect the solubility of C-56. The effect of the various soluble salts upon C-56 solubility was approximately additive and a function of the anion associated with the salt. . The rate of hydrolysis of C-56 in water was not significantly affected by pH; however, additions of iron did increase the rate of the hydrolysis reaction. The half-life of C-56 in water in a sealed glass container protected from light was approximately 3 months at both 22°C and 35°C, thus indicating that the hydrolysis rate was not temperature dependent over this temperature range. . The rate of photolysis of C-56 in aqueous solution and organic solvents followed a first-order reaction. The half-life of C-56 in water or organic solvents exposed to long-wave UV light or sunlight was less than 3.5 minutes; C-56 photolyzed much faster under long-wave UV light than under short-wave UV light. . At least 12 products of photolysis and hydrolysis were identified. Hexachlorocyclopentenone and pentachlorocyclopentenone were the major products found in distilled water. Cis- and trans-pentachloro- butadiene, tetrachlorobutenyne, and pentachloropentadienoic acid were the major products identified in mineralized water. Pentachlorocyclopentenone was not stable under high-temperature conditions — it was dimerized and then converted to hexachloroindone. . The sorption of C-56 by soils, clays, activated carbon, and Ambersorb XE-348 from aqueous solutions could be described by the Freundlich adsorption equation. All correlation coefficients (r 2 ) were high except for low temperature ashed Catlin and Ava soil. . The presence of salts in solution dramatically affected the sorption of C-56. Brine, NaCl, and NaOH caused an increase in sorption and NaOCl caused a decrease in sorption. The increase in sorption was inversely related to the Setschenow parameter, indicating that the change in sorption was directly related to the effect of the salts on the solubility of C-56. . A high direct linear correlation was found between the total organic carbon (TOC) content (0.19 to 32.17%) of the soils studied and the amounts of C-56 sorbed (r 2 = 0.97). . C-56 remained immobile in soils when leached with water, landfill leachates, and caustic brine solution, but was highly mobile when leached with organic solvents such as methanol, dioxane, acetone, and/or acetone/water mixture (1:1 v/v). Mobility of C-56 in soil was proportional to the soil organic carbon content and the solubility of the compound in the leaching solvent. Some degradation products of C-56 were more mobile than C-56 itself during soil column leaching studies . RECOMMENDATIONS The results of this study showed that C-56 was very photoreactive and subject to hydrolysis and sorption reactions. However, the disappearance of C-56 from the environment should not be construed to mean that it is always degraded to smaller molecules or sorbed to soil particles. C-56 has also been shown to have the potential to condense into compounds of larger molecular weight. To determine the full environmental impacts of C-56 on aquatic systems, further study of the toxicity and mobility of C- 56 breakdown and/or condensation products is essential. Almost no attenuation of C-56 by soils occurred when soil contaminated with C-56 was leached with organic solvents. To decrease the risk, of potential migration of C-56 from a landfill, C-56 wastes and organic solvents should not be disposed of in the same landfill location and C-56 waste should not come in contact with leaching organic solvents or highly organic leachates. Additional research is also needed to determine the chemical transformations of C-56 in soils in real environmental settings, especially in landfills where buried C-56 wastes may come in contact with organic solvents, acids, or iron drums. The results and conclusions derived from this study deal specifically with attenuation and mobilty of C-56 in the liquid phase. Vapor phase transport through soil pores was ignored; for compounds such as C-56, which have an appreciable vapor pressure, this means of migration may be a significant mechanism for redistribution. More information is needed to assess the magnitude of this means of migration for all organic wastes, including C-56. METHODS AND MATERIALS Experimental materials A reagent grade C-56 was obtained from Pfaltz and Bauer, Inc., Stamford, CT, and was used without further purification. An analytical standard of C-56 (lot #0213) was also obtained from US-EPA, Research Triangle Park, NC. Both materials were essentially identical and the purity was approximately 98 percent. Waters, soil extracts, and landfill leachates of varying content of dissolved organic carbon (TOC) were collected and analyzed. The samples used in studies of C-56 solubility and their respective TOC values are given in table 1 . The water samples contained a range in TOC values varying from 0.32 to 271 ppm; this allowed us to determine if differences in solubility occurred as a result of dissolved organic matter in water. The sorbents used in the sorption and mobility studies are given in table 2, along with some results from characterization measurements performed on the sorbents. The organic matter content of some sorbents was varied by using the low- temperature ashing (LTA) techniques for bituminous coals described by TABLE 1. Total organic carbon (TOC) analysis of waters used in C-56 solubility study Waters TOC (ppm) Deionized water 0.50 Distilled water 0.32 Tap water 1.31 Sugar Creek water 7.62 Bloomfield soil extract 16.00 (soil: tap water = 1:3) Catlin soil extract 49.80 (soil: tap water = 1:3) Blackwell sanitary landfill 235 leachate DuPage sanitary landfill 271 leachate u o H O T3 C CO CU o u 3 o CO r-l 4J CM 60 O r^ CO o -* a\ st CM CO «* m CO vO f— 1 vO o ■^ • • • • o CM 00 o\ CM ^— o CO vO CM CM 1^- vO m m m 00 g m m CN en CM CO CM CM CM CM CM v— ' co a) j-i r-N cO 00 vO co — rm m v£> r^ r^ vO r^ — • o CM CO co 0) Csl CM m o st -3- — CM CM r^ CM oo CO 00 o g CM m m CN en CM ~— "-■ CM .—1 CO ^x u 3 en 0) c CD t-l o 60 • O t-\ o ' — ^ m en vO en ~— m CM CO O m >> >s CM o en > /— \ r^ vO • o • • m • m st m • CO 6^8 o — • o o o a> 4J •H G •H iH O CO T3 C CO co CO ej CO io •u 4-1 O CO st vO o • CM • st o m MO vO o • st • • CM 00 m \o • CO i-l o u • M c_> CU • I— 1 *v c p- r-l • • A "O •H M rH r-4 o >% • w a o .-1 o 4-1 o CJ #\ rH c l-l r-l - c CO 3 r-l CO i-4 CiJ £ A o • O o c CO C O >> •H i-H CJ o o o •H 4-1 4-> r-l r-l •H •H a 43 C •H M E c 4-J s 4-1 3 CO CO CO CO 0) co a) O *•> 43 CJ c 42 > 43 CJ CO CO 60 •rH (-1 a U co 5 3 G l-l CU CU N CU CO CO • H 22 •u T3 C -H 4*: 4J 4-» 14-1 1 c c •H r-4 •H 4-) 4-1 U-l r-l CO a w o< O O W CO r-l a CO 3 CO CO ON CO < H rJ r>> 4-1 c o CJ 13 o CO 3 T3 C CO co CO O S o o i—i PQ C 3 O CJ G 60 •H CO P. g CO 43 CJ 4-1 co E CO C O CO r-4 60 CO >. G co CO i— i r-l CJ vO c 3 O U G 60 •H CO P- 43 CJ G •H i-H 4-1 CO CJ CO >N CO T3 CM CM CO CJ < H rJ 4J c 3 O CJ >% rJ G QJ S O CO m C 3 O CJ co •r4 O 3 cr o cu CJ ?*> u P0 00 cy» m m m o m 00 m o • m G 3 O CJ >% i-l c CU as CJ a 00 v£> PQ 44 O i G O 4-1 43 60 3 O X CO CO CO rH VO CO >n CO T3 00 CO PQ PQ PQ 44 CJ 3 g H rJ 44 O < H 44 O g H 44 CJ 3 g + C •H i-H 4-t CO U Gluskoter (1965) and Kuhn, Fiene , and Harvey (1978). An aliquot of Catlin soil and Ava soil and three aliquots of muck, soil were ashed, each for different periods of time. The changes in organic matter content and surface area due to ashing are shown in table 2. Surface areas were measured by three methods, each selected to provide different information about the surface characteristics of the sorbents. The BET gas adsorption method was used as described by Thomas and Frost (1971). Both N and CO were used as adsorbates. The third method was ethylene glycol adsorption as described by Bower and Gschwend (1952). The carbonaceous sorbents, Ambersorb XE-348 (Rohm and Haas) and activated bone carbon (Fisher Scientific) were selected for study because of their potentially high sorption capacity for organics from aqueous solution. Solubility studies Aliquots of 350 mL of water, aqueous soil extract, or landfill leachate were placed into 500-mL amber-glass bottles. C-56 (0.5 mL) was added to each bottle, and the bottle was capped with a teflon-lined screw cap. Preliminary studies indicated that a shaking time of 12 hours in a Lab- Line Model No. 358/water bath of moderate speed was sufficient for the sample to reach equilibrium conditions. Therefore, all solubility studies were shaken overnight (15 hours) to ensure equilibrium. The water and C- 56 mixtures were shaken in a constant temperature water bath (22° ± 1°C) at moderate speed for 15 hours and then left quiescent for 3 hours. Aliquots of 25 mL of the aqueous solutions were placed into 50-mL stain- less steel centrifuge tubes and centrifuged in a constant temperature cen- trifuge (Beckman Model J-21B) at 22° ± 1°C at 22,700 x gravity for 30 minutes; then 10-mL aliquots of the supernatant were carefully withdrawn and placed into 22-mL screw-capped vials. All samples were extracted three times with 5 mL portions of water-saturated hexane . Each extraction was obtained by shaking the vial on a vortex mixer at fast speed for 1 minute; the hexane extract was then transferred to a graduated test tube with a Pasteur pipette. Results of preliminary studies showed that the extraction technique used in this study offered improved efficiency over conventional separatory funnel extraction techniques. The extraction efficiency was 95 percent or better in samples spiked with C-56 at three different concentrations. Approximately 2 grams of anhydrous sodium sulfate was added into each tube to remove trace amounts of water. The extracts were then analyzed on a Hewlett-Packard (HP 5840A) gas chromatograph or a Perkin-Elmer Sigma I gas chromatograph (GC). Tribromobenzene was used as the internal standard* Both gas chromatographs were equipped with Ni 63 electron capture detectors. A glass capillary column was used to facilitate the separation of the degradation products of C-56 from "pure" C-56 in distilled water and deionized water extracts. The conditions for the gas chromatographic analysis for C-56 are shown in table 3. To determine the effect of caustic brines and soluble salts on the aqueous solubility of C-56, varying amounts of soluble salts were added to water and the change of solubility measured. The concentrations and salts used are listed in table 4. TABLE 3. Conditions for gas chromatographic analysis Conditions HP 5840A and P-E Sigma I GC Sigma I GC Column Injector temperature Column temperature Detector temperature Carrier gas 6 ft x 2 mm I.D. glass column, 5% OV-17 on 100/120 mesh chromosorb WHP 250°C 150°C 300°C methane/argon, flow 32 mL/min. 28 m x 0.25 mm I.D. grade AA glass capillary column (SE-30), split ratio: 100 : 1 250°C 145°C 280°C methane/argon, velocity: 28 cm/sec. Photolysis and hydrolysis studies In the photolysis study, 35 mL portions of C-56-saturated water solutions were placed into 50-mL stainless steel centrifuge tubes and centrifuged in a constant temperature centrifuge (22° ± 1°C) at 22,700 x gravity for 30 minutes; 20-mL aliquots of the supernatant were then carefully withdrawn from each centrifuge tube and placed into a 250-mL low-actinic Erlenmeyer flask. Aliquots of 10 mL of the supernatant of the C-56-saturated distilled water were placed into 22-mL screw-cap closed vials and exposed to either outdoor sunlight, long-wave, or short--wave UV-light (Chromato- Vue Cabinet, Model CC-20, Ultra-Violet Products Inc., San Gabriel, CA) for 0, 1, 3, 5, and 7 minutes. The sunlight intensity was measured by using a YSI-Kettering Model 65A Radiometer (Yellow Springs Instrument Co., Inc., TABLE 4. Salts and concentrations used in C-56 solubility studies in tap water Salts Concentrations (mole/L) NaCl 0* 0.513 1.026 1.465 2.051 NaOCl 0.135 0.270 0.405 0.541 NaOH 0.188 0.375 0.563 0.750 NaClO^ 0.250 0.500 0.750 1.000 NaCl + NaOCl 0.648 1.296 1.870 2.592 NaCl + NaOH 0.701 1.401 2.028 2.801 NaOCl + NaOH 0.323 0.645 0.968 1.291 NaCl + NaOCl + NaOH 0.836 1.671 2.433 3.342 *0 means no salt added 11 Yellow Springs, OH). The average sunlight intensity was 740 joules/m 2 (or watts-sec/m 2 ) outside the vial. To free C-56 from the ketone reaction products reported by Chou, Fisher, and Griffin (1981), 0.5 mL of 1 N_ NaOH solution was added into each vial before extraction for analysis. Analyses were conducted by gas chromatography (GC) using the methods previously described. To identify the photodegradation products in water, higher doses of C-56 (204.5 and 17,000 vig of C-56) were injected into 125-mL serum bottles con- taining 100 mL of distilled water. The serum bottles were sealed with teflon-faced septa and then shaken for 1 hour. The C-56 suspended samples were exposed to sunlight for 0, 12, and 60 minutes. In a separate study, 204.5 Mg portions of C-56 were injected into 22-mL screw-cap closed vials containing 10 mL of 0.5 N NaOH solution and exposed to sunlight for and 10 minutes. All samples were extracted with two parts of 12.5 and 5 mL water-saturated hexane, respectively. For further identification, some extracts were condensed to 1 mL under a N stream. The GC analytical con- ditions were as noted in table 3, except that the column temperature was programmed from 120° to 250°C at 6°C/min. The initial column temperature was held for 3 minutes. To identify the hydrolysis acids, some of the photolyzed water samples were made alkaline with 10 N_ NaOH solution and then acidified with concentrated HC1 before being extracted with hexane. In a separate study, 5 mL of C-56 (2 ppm) in hexane and methanol was individually placed in 15-mL vials capped with teflon-lined caps; the vials were then exposed to either sunlight or long-wave UV-light for 0, 1, 3, 5, and 7 minutes. In another study, 5 mL of 0.71 ppm C-56 in hexane was pipetted into 15-mL screw-capped vials. One set of vials was covered with aluminum foil and the other set was uncovered. All samples were set on a bench top and exposed to the ceiling fluorescent lights (48" tubes) in the laboratory during working hours (8:00 a.m. to 5:00 p.m.). Two replicate samples were analyzed at 0, 2, and 8 days. To identify photodegradation products in hexane and methanol, 30 ppm and 4500 ppm of C-56 solutions were exposed under sunlight or long-wave UV-light for 10, 22, and 60 minutes. The photolyzed organic solutions were directly analyzed by gas chromatography. The photolysis rate and the half-life were calculated using the log form of the empirical first-order reaction equation: i r -Kt . _ , 0.693 log c = 7303 + log V where h = —T~ C - concentration of component at time zero C = concentration of component at time t K = first-order reaction rate constant Several different approaches were used in conducting the hydrolysis studies. First, 0.5 mL of C-56 was individually pipetted into 500-mL amber glass bottles containing 350 mL of pH adjusted tap water. The pH values were 2.51, 5.80, 8.45 (tap), and 11.73. Mixtures of C-56 and water were shaken at constant-temperature (22° ± 1°C) for 18 hours and then left 12 quiescent for at least 3 hours. Aliquots of 50 mL of the C-56-water solutions were filtered through 0.22 pm Millipore® cellulose acetate membranes that had been soaked in C-56-saturated water prior to filtration. The membranes were then washed with deionized water and 200 mL of the C-56-saturated tap water were flushed through the membrane prior to filtration of the 50 mL aliquot. In the second study, 50-mL aliquots of the filtered aqueous C-56 were placed in 125-mL amber serum bottles and sealed with teflon-faced septa and aluminum crimp cap seals. All samples were kept in darkness and two replicate samples were analyzed at 0, 1, 2, 4, 7, and 11 days. The extraction was accomplished by placing 10-raL aliquots of the filtrates into 15-mL -vials and then extracting as previously described. In another phase of this study, the 125-mL amber serum bottles were filled with C-56 saturated tap water and distilled water, respectively. The serum bottles were then sealed and kept in a constant-temperature water bath (22° ± 1 C) under darkness. Two replicate samples of each water were analyzed at 0, 4, 10, 16, and 24 days. To free C-56 from the ketone reaction products, 0.5 mL of IN NaOH solution was again added into each vial before extraction for analysis. In a third study, 100 mL of distilled water was placed into 125-mL amber glass serum bottles and 6yL (122.7 yg) of 20,450 ppm C-56 dissolved in acetonitrile was individually injected into each bottle. The serum bottles were sealed and kept in a constant-temperature water bath (35°C) under darkness. Two replicate samples were analyzed at 0, 1, 4, 8, 16, and 32 days. Aliquots of 20 mL of water-saturated hexane were withdrawn by hypodermic syringe and directly injected into the serum bottle through the teflon-faced septum. All samples were set on a rotary shaker and shaken for 30 minutes at 300 rpm. The extracts were withdrawn with syringe and diluted for further analysis. The hydrolysis rates were calculated as previously described. In a fourth study, 4250 yg of C-56 was individually injected into 125-mL amber glass serum bottles containing 100 mL phosphate buffer solutions of different pH (pH values were 2.6, 5.0, 7.0, 9.24, and 12.1). All samples were vigorously shaken for 1 hour and then set in a constant-temperature water bath (22° ± 1°C) for 5 days. After 5 days the water samples were extracted and diluted for analysis. The percentage of C-56 and all impurities were calculated on the basis of total peak areas from the gas chromatograms. The same amounts of C-56 added to distilled water and hexane were used as controls. In a fifth study, the iron catalyzed hydrolysis of C-56 was measured. Freshly cut iron wire (approximately 2 mm of SWG 30 wire) or iron powder (<100 mesh) (1 g) was placed into 125-mL amber glass serum bottles. Aliquots of 100 mL of 0.5 N_ HC1 were added into the serum bottle containing the iron wire, and 100 mL of distilled water was added individually to a second group of serum bottles; 4250 yg of C-56 was injected into both groups of serum bottles and then sealed. All samples were shaken for 1 hour and then set at room temperature under darkness for and 2 weeks. Two replicate samples of each treatment were analyzed at the two indicated periods. 13 The major photolysis and hydrolysis products were identified by mass spectrometry by using a Hewlett-Packard 5985 GC mass-spectrometer and data system operated by the Institute for Environmental Studies Research Laboratory, University of Illinois. A Varian CH-7 mass-spectrometer coupled to a Varian 1900 GC and support computer operated by the Illinois Natural History Survey was used to analyze some of the samples. A 30-meter fused silica flexible capillary column (SP-2100) was used on the HP 5985 GC/MS and the column temperature was programmed from 4°C to 240°C at 3°C increments per minute. A 6-ft glass column containing 5 percent OV-17 Chromosorb WHP (100/120 mesh) was used with the Varian 1900 GC and the column temperature was programmed from 120°C to 250°C at 3°C increment per minute. Sorption studies The batch equilibrium method was adopted for the sorption studies. Various soil types and clays (table 2) were selected as sorbents. The carbon-aceous sorbents, Ambersorb XE-348 and activated carbon, were also selected because of their potentially high sorption capacity for organics from aqueous solution. In the sorption studies, known volumes and concentrations of C-56 solution were shaken with varying weights of sorbents at a constant temperature of 22° ± 1°C in 125-mL amber glass serum bottles. Amber glass was used to prevent photolysis of the C-56, and each bottle was sealed with a teflon- faced septum and an aluminum crimp-cap to prevent volatilization. The initial concentration of C-56 was kept at approximately 70 to 80 percent of its maximum solubility in the aqueous solution to be tested to prevent a possible precipitation or capillary condensation of C-56 that may confound the interpretation of the adsorption results. A preliminary kinetic study was conducted. Two doses of the Catlin soil were used; 1.0 g and 3.5 g on a dry basis were placed into 125-mL amber glass serum bottles. Aliquots of 50 mL of the filtered C-56 solution were pipetted Into the bottles. The serum bottles were sealed and shaken at constant temperature. Two replicate bottles of each dose of soil were removed from the shaker at 0.5, 1, 2, 4, 6, 8, 10, and 18 hours and placed directly into a Model JS-7.5 rotor and centrifuged at a constant temperature of 22° ± 1°C. In a preliminary study, the centrifugation time and the final temperature of the aqueous solutions in the bottles were examined. The results showed that 15 minutes of centrifugation at 4000 rpm was sufficient to separate the clays and aqueous phases; the aqueous temperatures in the bottles remained at approximately 22° ± 1°C. Therefore, all the samples were centrifuged at 4000 rpm for 15 minutes; the seals were then broken and 10-mL aliquots of the clear supernatants were pipetted into 22-mL screw-capped vials. All aqueous samples were extracted and analyzed as previously described. In the kinetic study, the rate of sorption of C-56 by soil materials was relatively rapid; a shaking time of 2 hours was sufficient for reaching constant concentrations of C- 56 in solution. Therefore, samples were shaken 4 hours in all later studies to ensure equilibrium. 14 Blanks were carried through all experiments to determine the degree of sorption of C-56 onto the surface of the bottles. Approximately 17 to 24 percent of the C-56 was sorbed onto the surface of the serum bottles. The amount of sorption in each case was determined from the difference between the initial concentration and the equilibrium concentration multiplied by the volume of solution. A blank, was subtracted, and the amount sorbed by each test material was computed on a unit basis by dividing by the dry weight of the sorbent. Catlin and Flanagan soils were selected for study of the effects of caustic brines of various dilutions on the sorption of C-56. Solutions of C-56-saturated NaCl, NaOCl, NaOH, and the three salt mixtures were filtered through 0.22 urn Millipore® membranes. Part of the C-56 was volatilized during filtration; the concentration of C-56 in those filtrates was never higher than its maximum solubility; therefore, all filtrates were directly used. The same procedures previously described were followed in these sorption studies. Mobility studies: determination by soil TLC and soil column leaching The method used to prepare soil TLC plates has been reported previously by Griffin and his co-workers (1977). The C-56 was spotted 2 cm from the base of the plate and eluted a distance of 10.5 cm with tap water, landfill leachate, caustic brines, acetone/water mixtures, methanol, acetone, or dioxane. The plates were immersed in 0.5 cm of the solvent (eluent solution) in a closed glass chamber and were removed when the wetting front reached the 10.5-cm line. The soil plates were removed and the soil was immediately scraped off in 1-cm increments, starting 1.5 cm above the base of the plate. The soil was placed in glass tubes and extracted with hexane or other suitable organic solvents. The concen- tration of C-56 in the extracts was measured by GC analysis. In a soil column leaching study, 700 grams of Bloomfield loamy sand was placed into a chromatographic column (5 cm I.D. x 50 cm long), and 5 mL of C-56 was pipetted on the soil surface in the column. After being spiked with C-56, the soil column was covered with 100 grams of Bloomfield loamy sand to reduce volatilization and was then continuously leached with 3 L of tap water (192 inches of water at 24 inches/day, representing approximately 5 to 6 times the annual rainfall in Illinois). After each 1 L of water had percolated through the soil column, 250 mL of leachate was extracted with three 30 mL portions of hexane. The hexane extract was then analyzed in the same manner as described in previous sections. 15 RESULTS AND DISCUSSION SOLUBILITY OF C-56 IN WATERS, SOIL EXTRACTS, AND LEACH ATES Solubility in waters and leachates The data from the solubility-kinetic study are shown in figure 1. The solubility of C-56 reached equilibrium after 12 hours of shaking under the test conditions. o o c o c o o 1.2- - ^JL • m 1.0- /* • • # 0.8- / 0.6- _ > 0.4- 0.2- U ™| i i i i 1 1 1 i I 1 1 1 1 1 4 8 12 16 20 24 28 32 36 4044 48 52 56 60 Shaking time (hrs) isgs 1983 Figure 1. Solubility of C-56 in tap water as a function of time at 22°C The solubility of C-56 in distilled water, deionized water, tap water, Sugar Creek water, soil extracts, and landfill leachates varied from 1.03 ppm in Sugar Creek water to 1.25 ppm in DuPage landfill leachate; the results are given in table 5. An increase in T0C content in the tested waters and leachates was correlated to the slight increase observed in the solubility of C-56. The slight decrease in solubility of C-56 in tap water and Sugar Creek water (as compared to distilled and deionized water) may be due to the presence of soluble salts. The solubility of C-56 in water has previously been reported as about 0.8 ppm (Lu et al., 1975), 1.8 ppm (Zepp et al., 1979), and 2 ppm (Atallah et al., 1980). In preliminary studies, the concentration of C-56 in pure water could not be accurately quantified because of a degradation product that had a GC retention time almost identical with that of the C-56. However, after a series of studies, the problem was overcome. To free C-56 from the degradation product, 0.5 mL of 1 J^ NaOH was added in the supernatant before extraction for analysis. Resolution of GC peaks for the degradation product and for C-56 are illustrated with representative capillary column GC chromatograms of tap water and deionized water-soluble C-56 in figure 2. There is a possibility that previous researchers, using packed column GC techniques, may have measured the degradation product or a combination of the degradation product and C-56 rather than "pure" C- 56. This may have led to the conclusion that greater concentrations of C- 56 were present in water samples than was actually the case. 16 (A) (B) 3.87 4.17 3.87 \ 4.18 10.79 10.96 C-56 10.97 C-56 -A. Figure 2. GC chromatograms (A) deionized water soluble C-56 (B) tap water soluble C-56 (28 m x 22 mm glass capillary column) 17 TABLE 5. Solubility of C-56 at 22°C after 15 hours equilibrium in waters and landfill leachates using contrifugation technique Waters and leachates Concentration (mg/L)* Distilled water 1 . ] ] Deionized water 1.14 Tap water 1 .08 Sugar Creek water 1.03 Bloomfield soil extract 1.06 (soilrtap water = 1:3) Catlin soil extract 1.20 (soil:tap water = 1:3) Blackwell leachate 1.19 DuPage leachate 1.25 *Each value is a mean of 2 replicates except tap water which is 20 replicates. Effect of dissolved salts on the solubility The effect of several soluble salts on the solubility of C-56 was treated by fitting the solubility data to the Setschenow equation (1892): S log _£ = Km ( 1 ) The results are given in table 6; listed are the observed solubility ratios, S /S — where S is the solubility (ppm) of C-56 in tap water and S its solubility (ppm) in a salt solution of concentration m (mole/L). Table 6 also includes the Setschenow parameter, K. The best values of K were determined by linear regression analysis of log S /S as a function of m; the coefficient of correlation (r 2 ) between log S /S and salt concentration is also given in Table 6. Representative data for C-56, plotted according to the Setschenow equation, are shown in figure 3. Of the three salts studied, sodium hydroxide and sodium chloride decreased the solubility of C-56 in tap water (K = positive) and sodium hypochlorite slightly increased its solubility (K = negative). The three salt mixtures (brine) were intermediate in their depression of solubility of C-56. Sodium perchlorate did not significantly affect the solubility of C-56. The results indicated that there was an anion effect, and that the effect of the individual salts upon C-56 solubility was approximately additive. A similar additive effect of sea water upon hydrocarbon solubility has also been reported (Gordon and Thorne, 1967a, 1967b). The Setschenow K value for combined salt solutions can be calculated by knowing the values 18 TABLE 6. Effect of dissolved salts on the water solubility of C-56 at 22°C and Setschenow parameter for C-56 Setschenow parameter Salts K in (mole/L) So/S NaClOi, NaCl NaOCl NaOH NaCl + NaOCl NaCl + NaOH NaOCl + NaOH +0.020 +0.161 -0.058 +0.266 0.200 +0.248 +0.340 NaCl + NaOCl + NaOH 0.147 0.706 0.970 0.889 0.996 0.877 0.995 0.993 0.993 0.250 1.003 0.500 0.997 0.750 1.010 1.000 1.038 0.513 1.306 1.026 1.622 1.465 1.683 2.051 2.239 0.135 0.998 0.270 0.956 0.405 0.942 0.54 1 0.940 0.188 1.099 0.375 1.239 0.563 1.396 0.750 1.563 0.648 1.507 1.296 2.303 1.870 2.291 2.592 3.540 0.701 1.629 1.401 2.296 2.028 3.614 2.801 4.966 0.323 1.358 0.645 1.738 0.968 2.089 1.291 3.020 0.836 1.212 1.671 1.563 2.433 2.168 3.342 2.767 19 NaCI 1 1 1 1 1 1 1 1 T 0.4 0.8 1.2 1.6 2.0 2.4 2.8 3.2 3.6 Salt concentration (mole/L) isgs 1983 Figure 3. Effect of dissolved salts on the water solubility of C-56 at 22°C-plot according to the Setschenow equation of K for the individual component salts and using the following equation presented by Gordon and Thorne (1967a, 1967b): log S i = log S i - n E i = 1 N, K, C lis (2) where S^ is the solubility of C-56 in the salt solution, S^ is the solubility of C-56 in pure water, R, is the mole fraction of the 1th salt component in the combined salt, K^ is the Setschenow parameter of the Jjth salt component in the combined salt solution, C is the sum of the individual salt molarities, and n is the number of salts in the solution. Agreement between the observed K value (0.147, L/mole) and the calculated K value (0.149, L/mole) using equation (2) for the caustic brine solution data reported in table 6 was quite good. From equation (2) we can also roughly predict C-56 solubility in salt solutions of varying concentration. Values of S^ calculated from equation (2) and the observed Sj values for the brine are shown in table 7. The agreement between observed S^ values and calculated S^ values was quite satisfactory for many environmental applications. TABLE 7. Solubility of C-56 in caustic brine solution (NaCI + NaOCl + NaOH) at 22° C C s (mole/L) S± observ. (mole/L x 10 6 ) S^ calcd. (mole/L x 10 6 ) 0.000 0.836 1.671 2.433 3.342 4.00 3.30 2.56 1.84 1.44 4.00 3.00 2.25 1.74 1.27 *Solubility of C-56 in tap water was 4.00 mole/L x 10 6 (1.08 ppm) 20 0.01 tap water short-wave UV tap water long-wave UV t — i — i — i — i — i — i — i — i — i — I — I — r 2 4 6 8 10 12 14 Time (min) ISGS 1983 Figure 4. First -order degradation plot of tap water soluble C-56 exposed to long- and short-wave UV light AQUEOUS CHEMISTRY OF C-56 Photolysis and hydrolysis In studies of photolysis, C-56 was found to be very photoreactive. The rate of photolysis in aqueous solution and organic solvents followed a first-order reaction. Values of the first-order rate constant (K), the half-life (t, ), and correlation coefficient (r 2 ) between concentrations and times for the photolysis of C-56 in solutions exposed to sunlight and long-wave UV light are given in table 8. C-56 photolyzed much faster in hexane than in tap water or methanol. The half-life due to photolysis of C-56 in water was less than 3.5 minutes. A similar result has also been reported elsewhere (Zepp et al., 1979). In a separate study, it was also observed that C-56 was very photosensitive to the laboratory ceiling fluorescent light; the results are summarized in table 9. The results clearly indicate that more than 65 percent of the C-56 had disappeared in the hexane solution after exposure for 8 days. It was also found that C- 56 was much more photoreactive under long-wave UV light than under short- wave UV light (fig. 4). The photolysis rate of C-56 in water was at least 15 times higher when exposed to long-wave UV light than when exposed to short-wave UV light . TABLE 8. First -order rate constant (K), half-life (t,,), and correlation coefficient (r 1 ) for the photodegradation of C-56 72 in distilled water, methanol, and hexane exposed to sunlight, long-wave UV light (L-UV), and short-wave UV light (S-UV) Treatment Sun K (min *) L-UV S-UV Sun ti (min) L-UV S-UV Sun L-UV S-UV C-56 in water 0.24 0.30 0.019 3.47 2.28 36.47 1.00 1.00 1.00 C-56 in methanol 0.30 0.32 0.006 2.32 2.17 115.50 1.00 0.99 1.00 C-56 in hexane 0.45 0.45 0.013 1.55 1.55 53.31 1.00 1.00 0.99 21 00 O in ci ~2 X T3 a. S3 i/t -n 3 5 m > 3 c ed 6 / H j ,, ,1 ' l) M ,. ■ ii l.li|l li i n i| M" iiii|iiii »r ti 1 1 i n -i rr i p i ii - i m [■ i m 60 80 100 H li Mh lil.il il n ilili'ilti i| n ii'» n i|iiii i h i|iit r iiii|i ii r iiii] 120 140 16C nil I it 1 1 1 II i lit ||l h iiii|i -ftfr 1 1 1 1 1 1 ii h |1ii r i M i] n il )i m |iiii n ii|iii i 1 1 1 1 1 1 ill M ii|ii h iiii|iiii i n i|iiii iiii p iil iiii|i n i 1 1 1 1 1 1 » 1 1 rffTTTTn rmpm n ii|ii n rrm 180 200 220 240 260 280 300 320 (M/e) amu Figure 8. Mass spectrum of compounds II and III 100 50 - 1 1 I i i i i i I i i i i ' i i i I I i i i I i I i i I I ill | I I I l I I I I I I I I I I l MM 30 80 1 30 I I I I I I M I I I I I III I I I I I I I I I I I I | 180 230 £ 100 -i 50 - I I I I | ' I I I | I I I I | I ' i I | I I I I | I I I I | I I I I'l T l'l I I I I I I I I I I I I I | I I I I 1 1 I I I I I I I I I I I I I I I | | I I I I I I | | I I I | I I I | I I I I I I I I 230 280 330 380 430 (M/e) amu Figure 9. Mass spectrum of compound IV 27 In acidified photolyzed aqueous extracts, one GC peak. (fig. 10, peak d) also matched a standard, pentachloropentadienoic acid (V). This was confirmed by GC/MS (fig. 11) where the major key ions showed m/e, 268 (parent peak); m/e, 251 (-0H); m/e, 233 (-C1); m/e, 223 (-C00H); and m/e, 188 (-C00H, -CI). A reasonable explanation of the formation of compound V is the dissociation of the carbonyl-carbon bond of compound II or III by hydroxyl radical formed under photolysis conditions. CI CI CI (III) .CI CI OH CI / O M^OH CI ^ \r^i *CI \ CI 7~ \ i* V COOH CI V / ci CI CI (V) During the same study, addition of strong base into the photolyzed aqueous solution and subsequent acidification of the sample caused the acid (compound V) to be substantially increased, thus giving further credence to the proposed reaction sequence. The GC chromatogram of the basified and acidified aqueous extract (figure 10B) shows that peaks (a), (b), and (c) disappeared and that three new peaks were observed. GC/MS analysis results indicated that peak (f) was unresolved and contained two isomers of C^HClg (VI and VII) and that peak (g) was C Cl^ (VIII). The mass- spectral data for compounds VI and VII (fig. 12) show m/e, 224 (parent peak); m/e, 189 (-C1); and m/e, 153 (-2C1, -H), for compound VIII (fig. 13) show m/e, 188 (parent peak); m/e, 153 (-C1); and m/e, 118 (-2C1). The structure was postulated to be: (VIII) 28 a b + C-56 (O in c i 4.50 CO CO Q (A) Figure 10. GC chromatograms of C-56 extracted from distilled water after exposure to sunlight (A) acidified aqueous extract (B) basefied, then acidified aqueous extract 100 -i 50 - 8 1 E - g ■ I IT I IN lliliiil'illli nH M iti n t, I'llilliiil'l n llllll 30 | M it'i M I| n |t ■ n t|i|ti l iiii||tli iiii|illl' H il|i n l'liilll|li II M ili n iiI M I H i 1 ! h liiiil iiiiiIiIi'IIii m iII'IIIIiiiIi i n iiii i i ' 80 130 180 100 -i 50 180 llf'll lllllllll IIIIM|/|'|IM||i\I'IIII|Imi lllllllll Mlllllll llll llll Mllll|l| IIIUMM lllllllll 1 1 1 1 1 1 1 1 I 111111111 1 1 1 ■ I 230 280 330 (M/e) amu Figure 11. Mass spectrum of compound V 29 100 — . 80- 60- 40- S 20- ' l l n it u i ( M f l** f . M i f . ill' If. ., ,f .1 »f«i. | ,« m J ,l|) MC ,lll| l llli i M | H it nm lill *1 160 60 80 r 100 120 iiniiii 140 ■B loo-i tfr llll llil|llll lllllllll II ll | MM flu | III I iiii|iin iiii||||| 1 1 1 f f I f r i 1 1 1 1 1 1 1 1 1 iiiiiiiii t 1 1 1 j 1 1 1 1 iiii|iiii 1 1 1 r p 1 1 1 n iliiiii n iiiiiii ii n iiiil m il 180 200 220 240 260 280 300 320 (M/e) amu Figure 12. Mass spectrum of compounds VI and VII 160 r ■ ■ ■ • r lllllllll lllljllll llll|llll lllllllll 1 1 1 1 [ 1 1 1 1 llll|HII lllljllll I 1 1 1 1 p 1 1 1 1 1 1 1 1 • 1 1 1 1 1 ii r i j iii i i r 1 1 j 1 1 1 1 inn 180 200 220 260 240 (M/e) amu Figure 13. Mass spectrum of compound VIII 280 300 320 30 These results indicated that (1) ketones (compounds I, II, and III) can be disassociated under basic conditions and that (2) compounds VI and VII could be derived from decarboxylation of compound V and then, after dehydrochlorination, could form compound VIII. A similar result, cleavage of the dibromotetrachlorocyclopentenone to bromotetra-chloropentadienoic acid with alkali has been reported by Newcomer and McBee (1949). Products from photolyzed C-56 in hexane and methanol were also examined. The GC chromatograms of photolyzed C-56 in hexane and methanol are shown in figures 14 and 15, respectively. The data clearly indicate that compound I was the major photolysis product of C-56 in both hexane and methanol solutions exposed under sunlight or long-wave UV light. However, the major photolysis product was not stable in hexane or methanol solution; it gradually dimerized and finally converted into hexachloroindone (figs. 14B-C). A similar result was also observed in methanol solution, but the final conversion product was not identified (fig. 16). We postulated that the unidentified component could be C g HCl 7 or C g ClgO as described by Butz, Yu, and Atallah (1982). The secondary photolysis products in hexane were C H CI (IX), and hexachloroindone (IV); the secondary products in methanol were 1,2,3,4,- tetrachloro-5 , 5-dimethoxy 1-cyclopentadiene (X) , 1-methylol-pentachloro- cyclodiene (XI), and unknown components. Trace amounts of pentachloro- cyclopentadiene (XII) were also identified in the photolysis studies of C-56 in hexane and methanol solutions. The mass-spectra of compounds X, XI, and XII are shown in figures 17, 18, and 19, respectively. (IX) C6H13 CI \ CI OCH, OCH ? (X) CI (XI) CH 2 OH CI CI (XII) Study results on the photolysis of C-56 indicated that two types of reactions were involved: one was dechlorination and the other was UV light-initiated oxygen oxidation. The C-Cl bond dissociation of C-56 formed product IX in hexane solution and gave products X and XI in methanol solution. Replacement of C-Cl bonds by C-H is well known in environmental photo- chemistry (Mosier, Guenzi, and Miller 1969). The replacement product, pentachlorocyclopentadiene has been identified by GC/MS in samples exposed to light in this study; this product is thought to be an intermediary compound that could be oxidized to form compound I. Compound I was not only found during C-56 photolysis in water and methanol, but was also found in hexane. Thus we speculate that air, rather than the solvent, was the source of oxygen involved in the photo-oxidation. This interpetation is supported by Molotsky and Ballweber (1957), who showed that bubbling through a C-56 solution produced compounds II and III. 31 7.14 cu, c ci Civ ^ \^CI CI' Pci o in CI' —1 i^ CI JV_ 6.32 ISGS 1983 Figure 14. GC chromatograms of C-56 in hexane solution (30 ppm) exposed to sunlight (A) min (B) 10 min (C) sample B set at room temperature overnight 32 Figure 15. GC chromatograms of C-56 in methanol solution exposed to long-wave UV light (A) 30 ppm C-56 standard (B) 12 min (C) 22 min 33 0.40 ISGS 1983 Figure 16. GC chromatograms of C-56 in methanol exposed to long-wave UV light (A) 30 ppm C-56 exposed to UV light (B) sample A set at room temperature overnight 34 100 — | 80- 60 - 40- - 20- rU ' iI.I m i. N I h i |i m , M ii| M i rn i HMH , "» | " i 80 100 I lllll'l lM i U i'illlll. 120 140 160 I ' il il [M I'llil | h i i mffrm mm | h iI H li|i n i mrjrm m i n i m i u iii m i nn| iiii i nn 200 220 240 (M/e) amu 260 280 300 320 Figure 17. Mass spectrum of compound X 100 80- 60 - 40- 160 Mi l' | H 1 |l n I i u l|lili m i| i iii ) i 1i p il V 11ll|llil 1 ih |llll n il | i i nm li pm n ii|ii u i n i|iiii lili]i) n iiii|i m i |||i|iiii i n iiiiil iiii|ii u i nn 180 200 220 240 260 280 300 320 (M/e) amu Figure 18. Mass spectrum of compound XI 35 100 — , 80 60 - 40- iii i| n il Mm |ii n i n l|T 100 jiui mi rfnrt iiiiiIiii imimmi iiiii 100-1 60 80 120 140 160 Mil I I T I | 1 1 1 I IIIIIIIII 1 1 1 1 | 1 1 1 I ■ ■ 1 ■ |1 1 1 1 T 1 1 1 I 1 1 I I llll|llll 1111)11 r I I II 1 1 1 II I I 1 1 1 1 1 1 1 1 llll|llll 1 1 1 1 1 1 1 1 I lllljllll IIIIIIIM 1 1 | II II I I 1 1 1 1 1 1 1 1 1 Hill 180 200 220 240 260 280 300 320 (M/e) amu Figure 19. Mass spectrum of compound XII From the mass-spectral data presented here, the laboratory data, and the limited information in the scientific literature, we postulated the degradation pathway of C-56 shown in figure 20. Experimental verification for the proposed sequence comes from several sources. Russian researchers (Simonov et al., 1975) noted C0_ evolution in their studies as predicted in the step from pentachloropentadienoic acid to the pentachlorobutadiene isomers. These compounds and the formation of tetrachlorobutenyne were consistent with the mass-spectral data. These compounds were found only in the mineralized water samples; only ketones were found in distilled water. In addition, the postulated degradation pathway predicts that two isomers will form and that one isomer (compound VII) will be preferred because of steric considerations. This could possibly account for the uneven doublet peaks that occur in GC traces of samples from mineralized water samples. To test this proposed degradation pathway, 204 ug of C-56 in lOmL of 0.5 N_ NaOH solution were exposed to sunlight for 10 minutes; a control was kept under darkness. The usual transformation products formed (fig. 21). The GC trace after acid addition showed that petachlorocyclopentenone (peak d), hexachlorocyclopentenone (peak f), pentachloropentadienoic acid (peak h), pentachlorobutadiene isomers (peaks b and c), and tetrachlorobutenyne (peak a) were present. Peak (g) was not identified, but was presumed to be tetrachloropentadienoic acid. 36 AQUEOUS DEGRADATION PATHWAY C-56 PENTACHLORO-2 HEXACHLOROINDONE CYCLOPENTENONE CI CI CI CI + H,0 HEXACHLOR03 cfjJ/^Y-. CYCLOPENTENONE \ / CI CI Jh.o ci-^^COOH V CI /> TETRACHLORO PENTADIENOIC ACID ci ci HEXACHLORO-2- ty CYCLOPENTENONE PENTACHLORO- C PENTADIENOIC ACID CIS-PENTACHLORO BUTADIENE CI ^^Cl CI -^^COOH h CI CI^^CI CI-^H CIv^^CI c-^ci H TRAN& TETRACHLORO PENTACHLORO BUTENYNE BUTADIENE ISGS 1982 Figure 20. Proposed pathway for the degradation of C-56 in aqueous solution In studies of catalytic properties of iron on C-56 hydrolysis, two unknown hydrolysis products were observed in both 0.5 N_ HC1 solution with freshly cut iron wire and in distilled water with iron powder. A representative GC chromatogram of C-56 extracted from a mixture of iron powder and distilled water is shown in figure 22. Peaks (a) and (b) were significantly increased with the addition of iron. GC/MS analysis results indicated that peak (a) contained four chlorines. The mass-spectral data show m/e, 202 (parent peak) and m/e, 167 (-C1) (fig. 23). The structure was postulated to be tetrachloropentadiene (C 5 H 2 C1 1+ ). Two of the early eluted peaks (c and d) have a molecular weight of 168 and appear to be trichloropentadiene isomers (C 5 H_C1_, fig. 24). Newcomer and McBee (1949) reported that dechlorination of C-56 occurred when C-56 solution was refluxed with Zn dust. Peaks (a) and (b) were also observed in methanol soluble C-56 solutions. Compounds that produced peak (a) and small amounts of the early eluted peaks were also found in a soil-column leaching study to be reported in a later section. From this study it was clear that iron in soils could potentially catalyze the formation of secondary products of C-56 that might generate problems in natural environments because of their higher aqueous solubility. It is possible that similar hydrolysis products might be present in landfills containing C-56 wastes. In order to determine the chemical transformations of C-56 in soil, further study is necessary. 37 6.94 o" c CO o o (A) ID CM JLA o co _yv> CM (0 CD CD CM 10 co ■* 9 IT) <9- co n . h O) »- "" 00 r» CM 1 - 1 CM « k f\K CO CD O o ^^ (B) ISGS 1983 Figure 21. GC chromatograms of C-56 in 0.5 N NaOH solution exposed to sunlight, then acidified with HC1 (A) min (B) 10 min 38 6.97 m b \ JlL (A) in O A. o CD CM en *■ f' to CM (0 'V A a 5.10 7.05 CM 6 O en ^ CM CM HJ^/^-A^ J (B) U ,_ Ifl 0) CO CO 1 1 m | r M f Mfmui iiiiiiiii iim|imi minimi iiii|iiii !iii|iin iiiijiiii minimi Minim iiiiiiiii miiiiiii ini| 180 200 220 240 260 280 300 320 (M/e) amu Figure 23. Mass spectrum of C5H2CI4 100 80 60- 40 - 20 - Mil MIIMIII IIIIJIIII IMIjIIII Mil MM I ln|l I II llll| Mil IIMIMll II 1 1| II II ll 1 1 1 II 1 1 lllllllll IIIIJIIII 1 1 1 1 1 1 1 ! I MM [I 1 1 1 I 1 1 1 1 1 1 1 I 1 1 1 1 1 1 1 [ I III 20 40 60 80 100 120 140 160 S 100 -i 80 - 60 - 40 20- ifn iiiiiiiii iiiijiiii iiiijiiii iiiiimii iiiijiiii tii 1 1 1 1 1 1 miiiiiii iiiijiiM MiijiiM iiii|iiii miiiiiii iiiiiiiii imimiii r 1 1 1 j i r 1 1 niij 180 200 220 240 260 280 300 320 {M/e} amu Figure 24. Mass spectrum of C s H 3 Cl3 40 E a ji c o CO in O 0.75- 0.60- 0.45- 0.30- 0.15- 1.00g/50mL A -m- 3.50g/50mL —T- 14 16 10 1^ 12 18 Time (hrs) ISGS 1963 Figure 25. Sorption of C-56 by 1.00 g and 3.50 g Catlin soil from tap water as a function of time at 22° ± 1°C SORPTION OF C-56 BY SOIL MATERIALS, ACTIVATED CARBON, AND AMBERSORB® CARBONACEOUS SORBENT Sorption of C-56 by soil materials The effect of reaction-time on sorption of C-56 by Catlin soil is shown in figure 25. It is clear that 2 hours of shaking is sufficient to reach equilibrium conditions. The results for C-56 sorption by various soil materials, activated carbon, and Ambersorb XE-348 are shown in figures 26, 27, and 28. All data were fitted by linear regression to the log form of the empirical Freundlich adsorption equation: log - = log K £ + N log C m r in which x = ug of compound sorbed; m = weight of sorbent (g); C = equilibrium concentration of the solution (ug/mL); Kj and N are constants. Values of K^ and n were obtained from the regression equations as the intercepts at a concentration of 1 ppm and the slope, respectively. These Freundlich parameters and the correlation coefficient (r 2 ) between the amount of C-56 adsorbed by a unit of adsorbent and the equilibrium concentration of C-56 are shown in table 11. The molar K (Kp) (table 11) was calculated from mass K, (Kf) by using the equation described by Osgerby (1970). 41 TABLE 11. Ireundlich Kp N, correlation coefficient (r 2 ), and molar K,. for sorption of C-56 by various sorbents from tap water aqueous solution at 22° C Freundlich parameters Sorbents , (1-n) n -1 (yg mL g ) ( (1-n) n -1 (nmole mL g ) CARBON SORBENTS Activated carbon Ambersorb 16,854 2,706 0.60 0.9982 0.63 0.9736 28,371 4,381 SOIL MATERIALS Houghton Muck B 1,045 0.80 0.9822 1,356 Muck B ashed 1 day 910 0.76 0.9759 1,235.8 Muck B ashed 6 days 637 0.71 0.9641 924.5 Muck B ashed 38 days 64 0.68 0.9506 97 Bryce silty loam 385 0.86 0.9946 462 Muck A + Catlin silt 284 0.76 0.9971 388 loam (1:1) //5 soil 97 0.66 0.9725 151 Catlin silt loam 54 0.93 0.9982 59 Catlin ashed 22 days 4. 74* 0.0140 4.7* Flanagan silt clay loam 30 0.95 0.9943 32 Bloomfield loamy sand 14 0.55 0.9085 25 Ava silt clay loam 9 0.94 0.9148 10 Ava ashed 19 days 1. 18* 0.0392 1.2* Ca-bentonite 32 0.63 0.9838 52 Illite 24 0.55 0.9991 43.1 Montmorillonite 18 0.60 0.9928 30.3 Kaolinite 4. 4 0.58 0.9924 7.6 *Average amount of C-56 sorbed by unit of sorbent (yg/g) (omitted from regres- sion in figure 29) . 42 400 muck No. 5 soil Catlin Flanagan loomfield 1 1 1 — r 0.01 0.02 0.04 0.08 ~ I 1 1 — I — i — 0.2 0.4 0.6 1.0 — r 2.0 Equilibrium concentration of C-56 (ppm) Figure 26. Freundlich adsportion isotherms of C-56 in seven soil materials from tap water 22° t 1°C 1C, (molar) = K f (mass) x Mol wt Mol wt in which Mol wt is molecular weight of the compound, Kj (mass) is the Freundlich Kr, and n is the slope of the isotherm plotted according to the Freundlich equation. The sorption of C-56 on these seven soils followed the series: muck > Bryce > muck + Catlin (1:1) > #5 soil > Catlin > Flanagan > Bloomfield > Ava. This suggests a relationship between the organic matter content of these soils and their sorption capacity for C-56. The sorption of C-56 on four clays followed the series: Ca-bentonite > illite > montmorillonite > kaolinite. C-56 was strongly sorbed by activated carbon and Ambersorb XE-348. The sorption constants (K^) for activated carbon and Ambersorb XE-348 were 16,854, and 2,706, respectively. The results indicated that activated carbon and Ambersorb XE-348 were much more effective in removing C-56 from water than were most earth materials. 43 0.01 Ca-bentonite iliite montmorillonite kaolinite — i 1 1 — r— r— 0.02 0.04 0.06 0.1 — I - 0.2 — i — i — r 0.4 0.6 1.0 Equilibrium concentration of C-56 (ppm) ISGS 1983 Figure 27. Freundlich adsportion isotherms of C-56 on four clays from tap water at 22° ± 1°C 8000 6000 4000- 2000 - ^1000 ^ 800 x 600 H 400 - 200 - 100 Activated carbon K f = 16,854 Ambersorb XE-348 K f = 2,706 —1 — r 0.6 0.8 ISGS 1983 0.001 1 0.002 1 1 1 1 0.004 0.006 0.008 0.01 0.02 — i 1 — 1 — r 0.04 0.06 0.08 0.1 0.2 I 0.4 Equilibrium concentration of C-56 (ppm) Figure 28. Freundlich adsorption isotherms of C-56 on activated carbon and Ambersorb XE-348 from tap water at 22° ± 1°C 44 Effect of earth material TOC on sorption The relationship between total organic carbon (TOC) content of the 17 sorbents and sorption of C-56 by the earth materials was investigated. The molar K (Kp), plotted as a function of TOC (percentages), is shown in figure 29. A high correlation (r 2 ) between K p and TOC (%) was found with a linear regression relation of: K p = 42.65 TOC(%) (r 2 = 0.972) K oc = 4 > 265 Similar results were also found for PCB, PBB, and HCB on similar sorbents (Lee et al., 1979; Griffin and Chou, 1980). The slope of the line from figure 29 yields a new sorption constant (K ) normalized for the organic carbon content of the soils. The K determined for C-56 is approximately 4,265. The results clearly showed that the sorption capacity of the low- temperature ashed muck, decreased with decreasing TOC content (for increased ashing time, see table 11). Very low measurable sorption of C- 56 was found on low-temperature ashed Catlin and Ava soils. The results indicated that the sorption properties of soil materials for C-56 can be predicted relatively accurately when the TOC content of the involved earth materials is known. However, low temperature ashing may have altered the physical and chemical nature of the native soil organic matter. This is evidenced by the relatively low adsorption by the ashed Catlin and Ava soils and the very sharp drop in surface area (table 1) and sorption of C-56 by the 38 day-ashed muck sample. Extended ashing time appears to have converted the native humus organic carbon to a more inert form relative to C-56 sorption. A limited number of soil materials was used to develop the equation and the relationship between sorption and types of organic matter is unknown. Serious discrepancies between predicted and actual sorption for a given soil may occur, even though the overall relationship is highly significant. Therefore, caution should be used in interpreting these results. u. 800 K = 42.65 TOC r 1 = 0.97 K oc = 4,265 10 20 TOC (%) Figure 29. K,. vs TOC for C-56 sorption by fifteen soil materials 45 Effect of soluble salts on sorption The results for C-56 sorption by Catlin and Flanagan soil from tap water, salt solutions, and caustic brine are shown in figures 30 and 31. All data were fitted by linear regression to the log form of the empirical Freundlich adsorption equation. The Freundlich parameters, correlation coefficient (r 2 ), and molar K (K p ) values are shown in table 12. All the regression lines generated had coefficients of at least 0.98, which again indicated a good fit of the data by the Freundlich equation. 200 100 60 40 5 20 a 3 E "x 10 6 4 % NaOH Tap water NaOCI 0.01 0.02 0.04 0.06 0.1 0.2 0.4 0.6 1.0 Equilibrium concentration of C-56 (ppm) ISGS 1982 Figure 30. Freundlich adsorption isotherms of C-56 on Flanagan soil from tap water, soil solutions, and caustic brine at 22°C 200 3 E 100 - 60 40 20 H 10 6 4 2- aOC! 1 1 1 1 1 1 1 0.01 0.02 0.04 0.06 0.1 0.2 0.4 0.6 1.0 Equilibrium concentration of C-56 (ppm) ISGS 1983 Figure 31. Freundlich adsorption isotherms of C-56 from tap water, salt solutions, and caustic brine at 22°C 46 TABLE 12. Fieundlich K-, N, correlation coefficient (i 2 ), and molar K for sorption of C-56 by Catlin soil and Flanagan soil from tap water, salt solutions, and caustic brine at 22° C Freundlich parameters 1 (nmole (1-n) n -1 mL g ) Solutions Catlin Flanagan Catlin Flanagan Catlin Flanagan Catlin Fl anagan Tap water 54 28 0.93 0.92 1.00 0.99 59 31 4% NaOCl 27 22 0.60 0.83 1.00 1.00 45 27 12% NaCL 186 153 1.01 1.10 1.00 1.00 183 134 3% NaOH 261 136 0.28 0.29 0.98 0.99 670 344 4% NaOCl 247 127 1.07 1.19 0.98 1.00 225 99 + 12% NaCl + 3% NaOH The results shown in figures 30 and 31 clearly indicate that all the salts and brine dramatically affected sorption. The change in sorption followed the inverse trend of the Setschenow parameters reported in table 6. The salts causing the greatest depression in solubility caused the largest increase in sorption and vice versa. The effect of soluble salts on the solubility of C-56 appears to be directly related to its sorption by soil materials. MOBILITY OF C-56 IN SOIL MATERIALS: DETERMINATION BY SOIL TLC AND SOIL COLUMN LEACHING Mobility of C-56 in soils The mobility of C-56 in soils, expressed as R^ values, is summarized in table 13. The results indicated that under the conditions tested, C-56 stayed immobile in all soils when leached with tap water, caustic brines, or landfill leachate. However, C-56 was highly mobile when leached with organic fluids such as acetone/water mixtures, acetone, methanol, or dioxane. The mobility of C-56 increased when leached with acetone/water mixtures as the percentage of acetone increased in water; the results are shown in figure 32. Mobility of C-56 in soil was proportional to the solubility of C-56 in the solvent and to the soil organic content. C-56 was significantly more mobile in sandy soil than in muck soil. The recovery of C-56 from soil TLC plates in these studies was extremely low (ranging from 1.25 to 4.5%), which presented some difficulties. However, the relative mobility of C-56 was easily measurable, particularly when increased amounts of C-56 were applied on each spot and the analysis was done rapidly. In later studies, we found that the recovery of C-56 from soil TLC plates was higher (ranging from 20 to 78%) when the analysis was done immediately after the soil plates were leached. The low recovery of C-56 was probably due to losses from volatilization. It is obvious from our recent experience with polybrominated biphenyl (PBB) and hexachlorobenzene (HCB) that C-56 is much more volatile than PBB or HCB. 47 100 80 - 60- 40 20- i ^ i — i — i 1 — i 1 1 — i r o° 5? 80- 60- 40- 20 - _- 0- I j ( f ( , iiii^ :: jT yi ip Lijii] I 100 80- 60- 40- 20- 100 80- 60- 40- 20 i I r ..,,. i , r ,, > , ) ,, r ,, ( ,, i ,. ( -I 1 1 -^~ ..,..*. r . n. v . 123456789 10 -i — i — i — i — i — i — i -■■(■■-■ -f "*> r '*" t ' 01 23456789 10 (I) (ID ISGS 1983 Figure 32. Mobility of C-56 in two soils leached with DuPage leachate and acetone/water mixtures (I) Ava soil (II) Catlin soil (A) DuPage (B) acetone/water (50:50, v/v) (C) acetone/water (80:20, v/v) (D) acetone 48 Soil Caustic Materials 12-4-3 Muck soil ND*** Catlin silt 0.000 loam Flanagan silt ND clay loam Ava silt clay 0.000 loam Bloomf ield 0.000 limestone Ottawa sand ND TABLE 13. Mobility of hexachlorocyclopentadiene in several soil materials leached with various solvents as measured by soil TLC R f Value* Brine (%)** Tap Landfill Acetone/ 3-1-0.75 water leachate water(l:l) Methanol Acetone Dioxane ND 0.002 0.005 0.093 0.053 0.137 0.640 0.000 0.002 0.002 0.793 0.879 0.972 0.989 ND 0.001 0.002 0.841 0.893 0.994 0.993 0.000 0.002 0.003 0.871 0.891 0.990 0.956 0.000 0.002 0.004 0.869 0.972 0.981 0.992 ND ND** 0.003 0.903 ND ND 0.978 *Computed from statistical peak analysis of data by using values of 1st moment for grouped data. **NaCl - NaOCl - NaOH (in tap water). ***ND = Not determined. The above findings have significance in C-56 waste disposal. To decrease the risk of potential migration of C-56 from a landfill, C-56 wastes and organic solvents should not be disposed of in the same landfill location and C-56 waste should not come in contact with leaching organic solvents or highly organic leachates. The results of this study also suggest that migration of C-56 through soil in aqueous solutions may not be as important as vapor transport mechanisms. Soil column leaching study In a soil column leaching study, a sample of Bloomfield loamy sand that had been heavily spiked with C-56 to simulate a spill was leached with 192 inches of tap water. We found 0.0005 percent of the compound was leached from the soil column. This loss suggests that C-56 will not be readily leached from even highly contaminated soils by percolating water. However, some hydrolysis products apparently have much higher solubility in water than C-56 and are much more mobile in soil. This result was observed in the soil column leaching studies (note representative GC chromatograms of the leachate extracts in figs. 33 and 34). The figures clearly show that hydrolysis products were found in the first liter of leachate and that several extra peaks appeared and were substantially increased when additional water was percolated through the soil column. From this study, it is concluded that the secondary products of C-56, rather than C-56 itself, might migrate and generate environmental problems . 49 C-56 A. C-56 Standard B. 64 inches tap water Figure 33. GC chromatograms of standard C-56 and soil column leachate (A) C-56 standard (B) soil column leached with 64 inches of tap water (IS = internal standard) C. 128 inches tap water D. 192 inches tap water Figure 34. GC chromatograms of soil column leachate leached with (C) 1 28 inches of tap water (D) 192 inches of tap water (IS = internal standard) 50 REFERENCES Atallah, Y. H., D. M. Whitacre, and R. G. Butz, 1980, Fate of hexachloro- cyclopentadiene in the environment: American Chemical Society preprints, 180th National Meeting, San Francisco, CA, Aug. 24-29. Bell, M. A., R. A. Ewing, and G. A. Lutz, 1978, Review of the environ- mental effects of mirex and kepone: Prepared for the U.S. Environ- mental Protection Agency, Office of Research and Development, by Battelle Laboratory, Columbus, OH, EPA 600/1-78-013. Bower, C. A. and F. B. Gschwend, 1952, Ethylene glycol retention by soils as a measure of surface area and interlaying swelling: Soil Science Society of America, Proceedings, v. 16, p. 342-345. Butz, R. G., C. C. Yu, and Y. H. Atallah, 1982, Photolysis of hexachlorocyclopentadiene in water: Ecotoxicology and Environmental Safety, v. 6, p. 347-357. Carter, M. R. , 1977, The Louisville incident: Internal report, Surveillance and Analytical Division, Region IV, U.S. Environmental Protection Agency, Athens, GA. Chou, S. F. J., B. W. Fisher, and R. A. Griffin, 1981, Aqueous chemistry and adsorption of hexachlorocyclopentadiene by earth materials. Land Disposal: Hazardous Waste — Proceedings 7th Annual Symposium, U.S. Environmental Protection Agency, EPA-600/9-81-002b, p. 29. Dietsche, W. H., 1966, Diels-Alder Reaktionen Mit Tetrachlorocyclo- pentadienon: Tetrahedron Letters, no. 2, p. 201-204. Equitable Environmental Health, Inc., 1976, Literature review of the health and ecological effects of exposure to C-56: report prepared for Hooker Chemical and Plastics Corporation. Gibb, J. P., Keros Cartwright, D. E. Lindorf, and Arnold Hartley, 1978, Field verification of hazardous wastes migration from land disposal sites: Final report for Contract No. R-803216-01-3, U.S. Environ- mental Protection Agency, Cincinnati, OH. Gluskoter, H. J., 1965, Electronic low temperature ashing of bituminous coal: Fuel, v. 44, p. 285-291. Gordon, J. E., and R. L. Thorne, 1967a, Salt effects on the activity coefficient of naphthalene in mixed aqueous electrolyte solutions. I. Mixture of two salts: The Journal of Physical Chemistry, v. 71, no. 13, p. 4390. Gordon, J. E., and R. L. Thorne, 1967b, Salt effects on non-lectrolyte activity coefficients in mixed aqueous electrolyte solutions-II . Artificial and natural sea water: Geochimica et Cosmochimica Acta, v. 31, p. 2433. 51 Griffin, R. A., F. B. DeWalle, E. S. K. Chian, J. H. Kim, and A. K. Au, 1977, Attenuation of PCBs by soil materials and char wastes, in Management of gas and leachate in landfills, S. K. Banerji, (ed.)> U.S. Environmental Protection Agency, Cincinnati, OH, EPA 600/9-77-026., p. 208-217. Griffin, R. A., and S. F. J. Chou, 1980, Disposal and removal of halo- genate hydrocarbons in soils: Proceedings, 6th Annual Solid and Hazardous Waste Research Symposium, U.S. Environmental Protection Agency, EPA 600/9-80-010, p. 82. IARC (International Agency for Research on Cancer), 1974, IARC monographs on the evaluation of carcinogenic risk, of chemicals to man: some organochlorine pesticides, v. 5. Ingle, L., 1953, Toxicity of chlordane vapors: Science, v. 118, p. 213-214. Kirk-Othmer Encyclopedia of Chemical Technology, 1964, 2nd ed.: New York: Interscience Publishers. Kuhn, J. K., F. Fiene, and R. Harvey, 1978, Geochemical evaluation and characterization of a Pittsburgh No. 8 and a Rosebud Seam Coal: Report for DOE Research Contract EY-77-X-21-2155, Morgantown Energy Technology Center, NITS No. METC/CR-78/8. Lee, M. C, R. A. Griffin, M. L. Miller, and E. S. K. Chian, 1979, Adsorption of water-soluble polychlorinated biphenyl Aroclor 1242 and used capacitor fluid by soil materials and coal chars: Journal of Environmental Science and Health A14, p. 415-442. Lu, P., R. L. Metcalf, A. S. Hirive, and J. W. Williams, 1975, Evaluation of environmental distribution and fate of hexachlorocyclopentadiene, chlordene, heptachlor, and heptachlorepoxide in a laboratory model ecosystem: Journal of Agricultural Food Chemistry, v. 23, no. 5, p. 967-973. Molotsky, H. M. , and E. G. Ballweber, U.S. Patent No. 2,795,608, June 11, 1957: to Velsicol Chemical Corporation, Chicago, IL, for preparation of hexachlorocyclopentenones from hexachlorocyclopentadiene. Mosier, A. R., W. D. Guenzi, and L. L. Miller, 1969, Photochemical decomposition of DDT by a free-radical mechanism: Science, v. 164, p. 1083. National Academy of Sciences, 1978, Kepone/mirex/hexachlorocyclo- pentadiene: an environmental assessment, 78 p. Newcomer, J. S., and E. T. McBee, 1949, The chemical behavior of hexa- chlorocyclopentadiene. I. Transformation to octachloro-3a,4,7,7a- tetrahydro-4,7-methanoindene-l ,8-dione: Journal of the American Chemical Society, v. 71, p. 946. 52 Osgerby, J. M., 1970, Sorption and transport processes in soils: S.C.I. Monograph no. 37, Society of Chemical Industry, London, S.W. 1, p. 63-78. Rieck, C. E., 1977a, Effect of hexachlorocyclopentadiene on soil microbe populations: report submitted to Velsicol Chemical Corporation, Chicago, IL. Rieck, C. E., 1977b, Soil metabolism of 11+ C-hexachlorocyclopentadiene: Report submitted to Velsicol Chemical Corporation, Chicago, IL. Sanders, H. J., 1978, Flame retardants: Chemical Engineering News, v. 22, April 14. Schmitzer, J., S. Gab, M. Bahadir, und F. Korte, 1979, Mineralisierung chlorierter alkane, alkene und aromaten durch heterogene photolyse: Naturwissenschaf ten, v. 66, p. 362-363. Setchenow, M., 1892, Action de l'acide carbonique sur les solutions des sels: Annales de chimie et de physique, v. 25, p. 226. Siraonov, V. D., et al., 1975, Hydrolysis of hexachlorocyclopentadiene: Vsesoiuznoe Khimicheskoe Obshehestvo Imeni D. I. Mendeleeva Zhurnal, v. 20, p. 477. Spehar, R. L., G. D. Veith, D. L. DeFoe, and B. V. Bergstedt, 1979, Toxicity and bioac cumulation of hexachlorocyclopentadiene, hexachloro- norbornadiene and heptachloronorbornene in larval and early juvenile fathead minnows, Pimephales promelas : Bulletin of Environmental Contamination and Toxicology, 21, p. 576-583. SRI (Stamford Research Institute), 1978, Directory of chemical producers, United States of America: Menlo Park, CA. Swanson, D., 1976, Discharges from Hooker Chemical Company: Internal staff report, Toxic Matter. Section, Water Quality Division, Environmental Protection Bureau, Michigan Department of Natural Resources . Thomas, J., Jr., and R. R. Frost, 1971, Versatile apparatus for studying reactions involving gas adsorption or evolution: Illinois Academy of Science Transactions, v. 64, p. 248-253. Treon, J. F., F. P. Cleveland, and J. Cappel, 1955, The toxicity of hexachlorocyclopentadiene: Archives of Industrial Health, v. 11, p. 459-472. Ungnade, H. E., and E. T. McBee, 1958, The chemistry of perchloro- cyclopentadienes and cyclopentadienes, Chemical Review, v. 58, p. 240. U.S. Environmental Protection Agency (EPA), 1977, Early warning report on hexachlorocyclopentadiene and its derivatives: Office of Toxic Substances, Washington, DC 20460. 53 Weber, J. B., 1979, Adsorption of Hex by a Cape Fear Loam Soil: North Carolina State University report submitted to Velsicol Chemical Corporation, Chicago, IL. Whetstone, R. R. , 1964, Chlorinated derivatives of cyclopentadiene in Kirk-Othmer Encyclopedia of Chemical Technology, 2nd edition: New York: Interscience, v. 5, p. 240-252. Wolfe, N. L., R. G. Zepp, P. Schlotzhauer, and M. Sink, 1982, Trans- formation pathways of hexachlorocyclopentadiene in the aquatic environment: Chemosphere, v. 11, no. 2, p. 91-101. Zepp, R. G., N. L. Wolfe, G. L. Baughman, P. F. Schlotzhaner , and J. N. MacAllister, 1979, Dynamics of processes influencing the behavior of hexachlorocyclopentadiene in the aquatic environment: American Chemical Society Preprints, 178th National Meeting, Washington, DC, Sept. 9-14. 54