N 107 R. L. LANGEnhEjM, JR. DEPT. GEOL. UNIV. ILLINOIS 254 N. H. B., 1301 W. GREEN ST URBANA, ILLINOIS 61801 GEOCHEMICAL PROPERTIES OF COAL WASTES AND THE TOXICOLOGICAL EFFECTS ON AQUATIC LIFE W.R. Roy, T.M. Skelly, I.G. Krapac, R.A. Griffin, D.R. Dickerson, R.M. Schuller, J.J. Suloway, N.F. Shimp Illinois Department of Energy and Natural Resources STATE GEOLOGICAL SURVEY DIVISION ENVIRONMENTAL GEOLOGY NOTES 107 1984 Roy, William R. Geochemical properties of coal wastes and the toxicological effects on aquatic life / William R. Roy . . . and others. — Champaign, IL : Illinois State Geological Survey, 1984. 124 p. ; 28 cm. — (Environmental geology notes ; 107) 1. Coal mine waste— Analysis. 2. Coal preparation— Wastes- Analysis. 3. Coal gasification— Wastes— Analysis. 4. Spoil banks- Leaching— Toxicology. I. Title. Series. Printed by authority of the State of Illinois /1984/2000 GEOCHEMICAL PROPERTIES OF COAL WASTES AND THE TOXICOLOGICAL EFFECTS ON AQUATIC LIFE William R. Roy, Ivan G. Krapac, Robert A. Griffin, Donald R. Dickerson, Rudolph M. Schuller, and Neil F. Shimp STATE GEOLOGICAL SURVEY Thomas M. Skelly and John J. Suloway STATE NATURAL HISTORY SURVEY NATURAL RESOURCES BUILDING CHAMPAIGN, IL 61820 ENVIRONMENTAL GEOLOGY NOTES 107 1984 -i- CONTENTS ABSTRACT 1 INTRODUCTION 2 METHODS AND MATERIALS 4 Sample Collection and Preparation 4 Mineralogical Procedures 5 Methods for Inorganic Analyses 6 Methods for Organic Analyses 6 MINERALOGICAL CHARACTERIZATION 10 INORGANIC CHEMICAL CHARACTERIZATION 12 Major and Minor Constituents 12 Trace Constituents 16 AQUEOUS SOLUBILITY OF THE SOLID WASTES 20 ASTM-A and ASTM-B 20 U.S. EPA Extraction Procedure 21 ERRATA Geochemical Properties of Coal Wastes and the Toxicological Effects on Aquatic Life, Roy et al . , Illinois State Geological Survey Environmental Geology Notes 107, 1984. Page 2, paragraph 1, lines 13 and 14 - For "These acutely toxic, sublethal concentrations " read "These acutely nontoxic, sublethal concentrations". Page 79, paragraph 2, line 2 - For "(an acutely toxic, sublethal concentration)" read "an acutely nontoxic, sublethal concentration)" . Page 85, paragraph 3, line 5 - For "acutely toxic, sublethal concentrations" read "acutely nontoxic, sublethal concentrations". Digitized by the Internet Archive in 2012 with funding from University of Illinois Urbana-Champaign http://archive.org/details/geochemicalprope107royw -i- CONTENTS ABSTRACT 1 INTRODUCTION 2 METHODS AND MATERIALS 4 Sample Collection and Preparation 4 Mineralogical Procedures 5 Methods for Inorganic Analyses 6 Methods for Organic Analyses 6 MINERALOGICAL CHARACTERIZATION 10 INORGANIC CHEMICAL CHARACTERIZATION 12 Major and Minor Constituents 12 Trace Constituents 16 AQUEOUS SOLUBILITY OF THE SOLID WASTES 20 ASTM-A and ASTM-B 20 U.S. EPA Extraction Procedure 21 Carbonic Acid Procedure 22 Long-Term Equilibration Procedure 22 Large-Volume Generation Procedure 22 Results 24 ORGANIC CHARACTERIZATION OF THE SOLID WASTES AND LABORATORY EXTRACTS 27 Methylene Chloride Extraction of Leachates 34 Organics from Coal Solid Wastes 37 Liquid Chromatography of the CH CI Extracts 39 EQUILIBRIUM SOLUBILITY MODELING OF LABORATORY EXTRACTS AND FIELD LEACHATES FROM COAL SOLID WASTES 43 ACUTE STATIC BIOASSAYS OF EXTRACTS GENERATED FROM SAMPLES OF COAL SOLID WASTES 56 Methods and Materials 56 Results 57 Acute toxicity of extracts 57 Relative tolerances of test organisms 59 Chemical characterization of the diluted extracts 62 Discussion 62 Nontoxic leachates 62 Toxic leachates 63 -ii- CHRONIC BIOASSAYS OF LEACHATES GENERATED FROM COAL CLEANING REFUSE SAMPLES 64 Methods and Materials • 64 Results 66 Acute toxicity of the leachates 66 Chronic toxicity of the leachates 67 Discussion 72 SUMMARY 79 RECOMMENDATIONS 85 REFERENCES 87 APPENDIX A. COMPILATION OF CHEMICAL ANALYSES OF ALL EXTRACTS 97 APPENDIX B. INTERELEMENT INTERFERENCES IN INDUCTIVIELY COUPLED ARGON PLASMA AS APPLIED TO HIGH-IRON SOLUTIONS 105 APPENDIX C. COMPILATION OF CHEMICAL ANALYSES OF ACTUE BIOASSAY TEST SOLUTIONS AND CORRELATIONS BETWEEN CONSTITUENT CONCENTRATION AND MORTALITY 109 -111- FIGURES 1. Organic analysis methodology 7 2. HPLC of a known mixture of phenols and polyaromatic hydrocarbons referenced to toluene 9 3. Cutaway side and top view of an Agro tank set up to generate a large volume of coal waste leachate 23 4. Changes in pH for all seven long-term equilibrations 28 5. Changes in total Fe concentrations for the four acidic long-term equilibrations 29 6. Changes in SO 2 concentrations for all seven long-term equilibrations 30 7. Changes in Al concentrations for the acidic long-term equilibrations 31 8. Changes in silica concentrations for all seven long-term equilibrations 32 9. Changes in electrical conductance for all seven long-term equilibrations 33 10. Infrared spectra of the organics extracted by methylene chloride from the extracts of WSSL 36 11. Infrared spectrum of the methylene chloride extraction of the OBG solid waste 39 12. Infrared spectra of the liquid chromatography fractions of the methylene chloride extract of OBG 42 13. Calcium sulfate equilibria of the acidic laboratory extracts and field leachate samples 47 14. Calcium sulfate equilibria of the non-acidic laboratory extracts 51 15. Calcium carbonate equilibria of the non-acidic laboratory extracts 52 16. Silicon dioxide equilibria of the laboratory extracts of all seven wastes and the field leachate samples 53 -iv- 17. Fe solubility equilibria as a function of pH for the acidic laboratory extracts and the field leachate samples.... 55 18. Percent mortality of Daphnia magna , Lepomis cyanellus , Physa anatina and Pimephales promelas resulting from exposure to undiluted extracts generated from seven coal solid wastes and buffered solutions of reconstituted water... 58 19. Relationships between the OBG extract and mortality for Pimephales promelas , Daphnia magna , Lepomis cyanellus , and Physa anatina 59 B-l. Relationship between the spectral bandpass, analytical cobalt emission line, and two interfering iron lines 107 -v- TABLES 1. Liquid chromatography elution sequence , 9 2. Mineralogical composition of the seven solid waste samples by X-ray diffraction 11 3. Major and minor constituents in refuse samples 13 4. Major and minor constituent composition of the spoil samples, the slurry sample and the Kilngas waste 14 5. Sulfur species and total sulfur content of the seven solid wastes 14 6. Trace constituent concentrations in refuse samples 17 7. Trace constituent concentrations in the slurry, mine spoil and Kilngas sample 19 8. Range in pH and percent extracted by the four short-term procedures- for the four different waste types 25 9. Chemical analysis of the long-term equilibrates (LTEs) after 147 days of equilibration 26 10. Total organic carbon content of generated leachates 27 11. Total chromatographable organics and gravimetric analysis of methylene chloride extractions of coal waste extracts 35 12. Total chromatographable organics and gravimetric analysis of the methylene chloride extraction of coal solid wastes 37 13. Elemental sulfur content of the methylene chloride extracts of the coal solid wastes 38 14. Gravimetric analysis of liquid chromatography fractions from methylene chloride extracts of the solid wastes 40 15. Compounds identified by gas chromatography /mass spectrometry in the liquid chromatography fractions of the methylene chloride extract of the Kilngas residue 44 16. Chemical composition of acid mine drainage from the Old Ben No. 9 mine and the Will Scarlett mine 46 17. The Debye-Huckel Limiting Law and variants 49 -vi- 18. The LC-50 values, their 95% confidence intervals and the amount of dilution necessary to eliminate mortality in static bioassays of extracts (WSOG, WSSL, WSNG and OBG) generated from the solid wastes 60 19. Differences in the acute toxicities of extracts generated from four coal solid wastes (WSNG, WSOG, OBG and WSSL) as determined by an analysis of covariance 61 20. Differences in sensitivities of Daphnia magna , Pimephales promelas , Lepomis cyanellus and Physa anatina to extracts generated from four coal solid wastes (WSNG, WSOG, OBG and WSSL) as determined by an analysis of covariance 61 21. Chemical constituents found to be significantly correlated to mortality and to exceed recommended water quality levels in test solutions prepared from extracts (OBG, WSNG, WSOG and WSSL) generated from coal solid wastes 63 22. Results of the acute bioassays of WSNG and OBG conducted by LTE and LVG procedures 67 23. Lengths, weights, mortality and hatching percentages for fathead minnows exposed to five concentrations of WSNG leachate and a control 68 24. Lengths, weights, mortality and hatching percentages for green sunfish exposed to five concentrations of WSNG leachate and a control 68 25. Lengths, mortality and reproductive success of Daphnia magna exposed to five concentrations of WSNG leachate and a control 69 26. Shell lengths, weights, mortality and egg mass production of Physa anatina exposed to five concentrations of WSNG leachate and a control 70 27. Lengths, weights, mortality and hatching percentages for fathead minnows exposed to five concentrations of OBG leachate and a control 71 28. Lengths, weights, mortality and hatching percengages for green sunfish exposed to five concentrations of OBG leachate and a control 71 29. Lengths, mortality and reproductive success of Daphnia magna exposed to five concentrations of OBG leachate and a control 72 -VI 1- 30. Shell lengths, weights, mortality and egg mass production of Physa anatina exposed to five concentrations of OBG leachate and a control 73 31. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of OBG during chronic bioassays for fathead minnow ( Pimephales promelas ) 74 32. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of OBG during chronic bioassays for green sunfish ( Lepomis cyanellus ) 75 33. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of OBG during chronic bioassays for the crustacean, Daphnia magna 76 34. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of OBG during chronic bioassays for the snail, Physa anatina 77 35. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of WSNG during chronic bioassays for fathead minnow ( Pimephales promelas ) 80 36. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of WSNG during chronic bioassays for green sunfish ( Lepomis cyanellus ) 81 37. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of WSNG during chronic bioassays for the crustacean, Daphnia magna 82 38. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of WSNG during chronic bioassays for the snail, (Physa anatina) 83 -viii- ACKNOWLEDGMENTS This study was partially funded by a grant (No. G51 15002) from the Illinois Mining and Mineral Resources Research Institute, Southern Illinois University at Carbondale, Illinois. We are indebted to C. C. Hinckley of Southern Illinois University for the Mossbauer spectroscopic investigation. We also thank the Araax Coal Company, the Peabody Coal Company, and the Old Ben Coal Company for their assistance in collecting tbe spoil, refuse, and coal slurry samples, and the Allis-Chalmers Corporation for the Kilngas gasification residue. The authors also thank the Analytical Chemistry Section of the Illinois State Geological Survey, W. J. Nelson for sample collection, R. D. Harvey for the mineralogical characterization, and Susan Kamp, Shirley Lowe, Chuck Stone, Lynn Dettman, Jim Roberts, Mary Martin, and Sherri Sandberg for their assistance with the toxicity studies. Mention of trade names does not constitute an endorsement, either specific or implied, by any of tbe respective agencies. DISCLAIMER The views and conclusions presented in this document are those of the authors and should not be interpreted as necessarily representing the official policies or recommendations of the U.S. Department of the Interior, specifically the Bureau of Mines or the Office of Surface Mining — or any other agency of the U. S. government. -1- ABSTRACT Leachates from solid wastes generated by coal mining, cleaning, and gasification are potentially harmful to the environment. The toxicolog- ical effects of leachates on aquatic organisms have not been adequately assessed. In this investigation, samples of seven coal-related wastes were characterized chemically and mineralogically . Laboratory extracts of each sample, obtained by a variety of extraction methods, were used in both acute and chronic bioassays. A coal slurry sample (coarse fraction) and two samples of coal cleaning refuse were chemically and mineralogically similar; they generated acidic water-waste systems, both in the laboratory and in the field. Acidic laboratory extracts and field samples of acid mine drainage, collected from the same area as the solid samples, tended to be in equilibrium with some type of calcium sulfate (CaSO. ) phase and with maghemite. The chemical composition of the extracts was sensitive to the method used to generate the extracts, but generally had the same prevalent chemical character as the field samples. Acidic wastes contain appreciable quantities of soluble Al, Mn, Ni, SO , and Zn, as indicated by both laboratory extraction and field data. Long-term equilibrations suggest that Al, Fe, and SO continued to be released by the acidic wastes after 22 weeks, although no associated change in pH was apparent. Two mine spoil samples, essentially shale, tended to generate water- waste systems that were neutral in pH; consequently, they released lower quantities of potential pollutants than the acidic wastes. The gasifi- cation residue was compositionally similar to coal-fired power plant slag and fly ash; it generated moderately alkaline extracts with low concentrations of soluble, potentially toxic constituents. All solid wastes contained a variety of organic compounds; many are on the EPA Priority Pollutant List; however, these compounds were essen- tially insoluble in aqueous media. Each extract was tested for acute toxicity with four species of fresh- water aquatic organisms; the green sunfish ( Lepomis cyanellus ), the fathead minnow ( Pimephales promelas ), a crustacean ( Daphnia magna) , and a snail ( Physa anatina ). The extracts of the two mine spoil samples and of the gasification residue were not toxic to any of the organisms; whereas the extracts from the acidic refuse and slurry samples were acutely toxic to all the organisms. For all cases, pH was significantly correlated to mortality; Al, Fe, Mn, Ni, and Zn were significantly correlated to mortality only in specific instances. Data from 96-hour LC-50 determinations showed that dilutions up to 1:1000 were necessary to eliminate mortality. The most toxic extracts were those with the lowest pH, which necessitated the largest amounts of dilution to eliminate mortality. -2- Chronic bioassays, using the same organisms, were investigated with leachate samples from a relatively unweathered refuse and an older, more weathered refuse. Leachate from the weathered refuse was not acutely toxic; organisms exposed to dilutions of this leachate showed no detri- mental chronic effects. The unweathered refuse generated an acutely toxic leachate. Low concentrations of the acutely toxic leachate lowered hatchability for the Lepomis cyanellus and Pimephales promelas eggs, reduced weight gain for Physa , and with chronic exposure, in- creased the death rate for Lepomis cyanellus , Daphnia , and Physa . Dilutions of a toxic, high-iron leachate resulted in the precipitation of ferric hydroxide that was implicated as the major factor accounting for the physical and physiological toxic effects exhibited by all test organisms during the chronic experiments when exposed to these acutely toxic, sublethal concentrations. The toxicity of this leachate demon- strates the detrimental effects that this type of waste could have on the aquatic ecosystem, if it entered a receiving water in a natural setting. INTRODUCTION During the late 1970s, shortages of natural gas, fuel oil, and gasoline dramatically demonstrated the need for electric utilities to increase their use of coal. Predictions vary, but the National Coal Association forecasted an increase in coal usage from 787 million metric tons in 1976 to approximately 1.5 billion metric tons by 1985. Increasing the use of coal increases the production of coal wastes. Past and current mining techniques have produced various types of coal solid wastes that are potentially harmful to the environment. Coal conversion via gasifi- cation and liquefaction also produce solid wastes that pose environmen- tal problems. The Energy and Research Development Administration (1977) concluded that coal solid wastes could create substantial leaching and revegetation problems. Abandoned mine spoil banks and coal preparation wastes pose similar problems (Beyer and Hutnik, 1969; Mason, 1978). The Southern Illinois University Wildlife Research Study Group of Klimstra and others has defined 6,500 acres of abandoned preparation wastes or "gob" and 4,300 acres of slurry in Illinois alone. The impact on the aquatic environment of leachates from surface-mine spoil banks and of solid wastes generated during coal cleaning and synthetic fuel produc- tion has not been adequately assessed. This investigation comprises interrelated physical, chemical, and bio- logical analyses for a broad-based evaluation. The analytical synthesis of the data from this investigation will aid control technology and reclamation development. It will also provide a basis for determining the potential degree-of-toxicity of these wastes as required by PL 95- 87, the Surface Mine Control and Reclamation Act (SMCRA of 1977). Much information has recently become available for developing fish and wildlife management of reclaimed surface mines (Cairns, 1978). However, -3- to maximize the benefits and minimize the adverse effects of mining and reclamation technologies, it is essential to test the toxicity of leachates produced from abandoned spoil banks. Unreclaimed spoil banks abandoned before the passage of the Surface Mining Control and Reclamation Act of 1977 (PL 95-87) present a potential pollution hazard to our water resources (Hood et al., 1979). Another reclamation problem associated with the increased use of coal is the proper disposal of solid wastes from coal conversion and coal cleaning processes. The solid wastes from coal liquefaction and coal gasification will probably be redeposited on site where the coal was mined or in landfills (Talty, 1978). The volume of solid wastes pro- duced from coal conversion is enormous. One estimate (Sather et al., 1975) is that a commercial gasification plant (250 million ft 3 of gas/day) will generate 2.3 million tons of ash and dry refuse per year. Additional estimates (Seay, 1972; Asbury and Hoglund, 1974) predict that a single gasification plant will generate 625 acre feet of residue per year. This huge volume of materials, like the thousands of acres of spoil banks and coal cleaning refuse piles, are subject to leaching; the effect of these leachates on the aquatic environment has not been adequately assessed. The potential environmental hazard of coal solid wastes lies in the relatively large concentrations of accessory elements and the possible formation of acid from pyritic minerals. It is well known that iron sulfates and acids are produced from the oxidation of pyritc minerals in coal solid wastes exposed to air (Singer and Stumm, 1969; Smith et al., 1968; and Jones and Ruggeri, 1969). Sun et al. (1978) and Stone and Kahle (1977) conducted leaching studies on wastes from f luidized-bed gasifiers, but most work was done with wastes other than those from coal conversion processes producing synthetic fuels. Wewerka et al. (1978) have done extensive leaching studies on bulk residue from the Illinois Basin. Griffin et al. (1978) found that 31 to 60 chemical constituents measured in Lurgi gasification and H-coal liquefaction wastes presented a potential hazard; 16 of these constituents exceeded federal water- quality standards in equilibrated solutions with pHs ranging from 3 to 8. In addition, other studies (Rubin et al., 1976; Herbes and Beauchamp, 1977; Hildebrand et al., 1977; Schultz et al., 1978; and Southworth et al . , 1978) have suggested that selected inorganic and organic compounds, such as resorcinol and 6-methyl quinoline associated with synthetic fuel effluents, are toxic. Other workers have found similar results with fly ashes from coal combustion (Vaugham et al., 1975; Cherry et al., 1976; and Birge, 1978). The 1977 Executive Summary for the Eastern Energy and Land Use Group identified spoil-bank leaching as having a major impact on the environment (Mason, 1978). Some environmental hazards related to surface mining have been well documented (Curry and Fowler, 1978); best known are acid production and sedimentation and their effects on aquatic organisms (Jones, 1964; Parsons, 1968; Roback and Richardson, 1969; -4- Archibald and Gentile, 1971; Dills and Rogers, 1974; Herricks and Cairns, 1974; Preston, 1978; Orciari and Hummon, 1975; Matler et al., 1978; Winger, 1978; and many others). Bell (1971) subjected typical aquatic stream insects to various acidic conditions and demonstrated the detrimental effects of low pH. Cairns et al . (1971) demonstrated that benthic communities were reduced when a small portion of a mountain stream was subjected to acute acidic stress. Reppert (1964) demonstrated with fish cultures how toxicity tests could delineate the severity of acid mine drainage. However, few studies have investigated the chemical properties of spoil-bank leachates (Beyer and Hutnik, 1969; and Barnhisel and Rotromel, 1974); no bioassays of spoil-bank leachates have been conducted; and little is known of the chronic effects from any of the various coal-waste leachates. Toxicity tests or bioassays represent one of the best approaches for predicting the effects of land disturbance, mining related wastes, or coal processing and utilization wastes on aquatic life (Preston, 1978). After it is determined with bioassays whether a leachate or laboratory extract is toxic, the physicochemical character of the leachate can be analyzed to identify the potential toxic components of the leachate and to aid the development of control technology and reclamation methodologies. METHODS AND MATERIALS Sample Collection and Preparation Seven waste samples produced by coal utilization were collected: two samples of overburden or mining spoils, three samples of coal cleaning wastes, one sample from a dry slurry lagoon, and one solid waste by- product from a coal gasification plant. The two samples of mining spoils were taken from a mine operated by the Amax Coal Company (Delta Mine) in Williamson County in southern Illinois. One sample, DMNS , is a composite of the Lawson Shale Member, a dark gray Pennsylvanian unit stratigraphically located above the Herrin (No. 6) Coal Member; this shale is the largest unit of the overburden at the mine. DMNS was collected from a vertical section at the strip mine. At the same location, a second spoil sample, DMOS, was taken from a roadcut made through an 18- to 22-year old spoil mound, composed largely of the Lawson Shale. Samples of coal cleaning wastes were taken from a strip mine operated by tbe Peabody Coal Company (Will Scarlett Mine) in Saline County in southern Illinois. One of these (WSNG) was a sample of fresh cleaning waste collected from an active preparation plant processing coal from the Davis-Dekoven seams at the time of sampling. A second refuse sample (WSOG) was collected at this mine from a waste pile estimated to be -5- about 20 years old. A sample of the material (WSSL) slurried to a settling pond was also collected from the same preparation plant that . generated WSNG and WSOG. A third refuse waste (OBG) was collected from a waste pile in Franklin County, Illinois. This pile was approximately 50 years old and was the result of mining by shaft methods, activities associated with the Herrin Coal (Old Ben Mine No. 9). The source pre- paration plant is no longer present. From a Kilngas (KG) deep-bed gasifier, 19 liters of solid waste was supplied by Allis-Chalmers from their pilot plant in Milwaukee, Wisconsin. Specifications for the fuel used when the waste was gener- ated could not be obtained from this proprietary operation. The Kilngas process is described elsewhere (Illinois Institute of Natural Resources, 1979). Collecting representative samples from coal refuse and spoil piles would not only have been impractical, but beyond the scope of the study. Moreover, procedures have not been established for the collection of representative samples from waste piles; therefore, an effort was made to collect a 25-kg sample from each waste pile, visually avoiding anomalous materials. To divide the collected bulk waste samples into representative subsamples was more important than to attempt to collect an initial sample representative of the entire waste pile. All the mining wastes were first air-dried and ground to pass through a 9.53-mm sieve. This sized material was then subdivided using a sample splitter to produce a series of representative subsamples for each waste. Twelve subsamples from each waste were used for the solubility studies. An additional subsample for each waste was then further ground to pass through a 0.25-mm standard sieve and split to obtain representa- tive samples for chemical and mineralogical analysis of the solid wastes. The Kilngas solid residue used for the solubility studies was in pellet form, as received from the pilot plant. The pellets ranged in size from about 0.1 mm to 1.5 mm. Before the solid was analyzed, it was ground to pass through a 0.25-mm sieve and split. Mineralogical Procedures The seven waste samples were analyzed by X-ray diffraction for mineral- ogical characterization and by Mossbauer spectroscopy for the determin- ation of iron species. Prior to X-ray diffraction analysis, subsamples of the solid wastes used were ashed in a low-temperature asher (LTA). Low-temperature ashing isolates the desired mineral residue from the organic matter in the sample, and it often facilitates the resolving capabilities of X-ray dif fractometry . The samples were ashed on a LFE corporation Model LTA-504 radio-frequency asher. Then the ashed samples were qualitatively analyzed by X-ray diffraction with a Phillips Norelco X-ray dif f ractometer using copper Ka radiation and a graphite monochro- mometer. The method is discussed further in Russell and Rimmer (1979). -6- Mossbauer spectra were obtained with an Austin Science Associates spectrometer using a linear acceleration motor to move the source ( 51 Co in Pd). A Nicolet 1070 N Signal Averager with 1024 channels was used to collect the spectra. Two spectra were simultaneously collected and combined to yield a spectrum that was recorded in 512 channels. The spectra were obtained at room temperature; spectra obtained at liquid- nitrogen temperatures were not significantly improved. Liquid-helium temperatures were not available, and there was no identification of superparamagnetic species. A more detailed discussion of the Mossbauer spectroscopy techniques used in the analysis of these samples is given in Saporoschenko et al. (1980) and Hinckley et al. (1980). Methods for Inorganic Analyses The samples were chemically analyzed for Si, Al, Mg, Ca, K, Fe, Ti, and P using X-ray fluorescence spectrometry (XRF). Arsenic, Ba, Br, Ce, Co, Cr, Cs, Eu, Ga, Hf, La, Lu, Ni , Rb, Sb, Sc, Se , Sm, Sr, Ta, Tb, Th, U, W, Yb, and Zn were determined by instrumental neutron activation analysis (NAA) . Mercury determinations were made by neutron activation with radiochemical separation. Boron, Cu, Ge, Li, Mo, Pb, Sn, and V were done by optical emission spectrochemical analyses (0ES). A detailed discussion of sample preparation, detection limits, and procedures for these techniques can be found in Harvey et al. (1983). The sulfur determinations were carried out by ISO method 609-197 5E. Most chemical analyses of the supernatant solutions were determined by inductively coupled argon plasma spectrometry (ICAP) with a Jarrell-Ash Model 975 Plasma AtomComp. The constituents determined by ICAP were Al, As, B, Ba, Be, Ca, Cd, Cr, Cu, Fe , K, Mg, Mn, Mo, Na, Ni, Pb, Sb, Se , Si, Sn, V, and Zn. The procedures and techniques of this specific instrument are discussed by the Jarrell-Ash Division (1978). Sulfate was done turbidimetrically (American Public Health Association, 1975). Alkalinity was determined by titration with dilute sulfuric acid and oxidation-reduction potential (Eh), pH, and electrical conductance were measured by electrodes (U.S. EPA, 1979). Analytical Methods for Organic Analyses The solid samples were air-dried and ground to pass a 0.25-mm screen prior to organic analysis. The solid waste samples and laboratory extract samples were analyzed by the methods set forth in the IERL-RTP Procedures Manual: Level 1 Environmental Assessment (U.S. EPA, 1978). An overview of the methodology is shown in figure 1. The solid samples were extracted with 1 L of methylene chloride (CH Cl„) in a large (70-mm x 300-mm body) soxhlet extraction apparatus. The extracts were found to contain elemental sulfur. Free sulfur was removed from the extract by passing the solution over activated copper (Blumer, 1957). -7- Organic Extract or Neat Organic Liquid Concentrate Extract Infrared Analysis TCO Analysis Gravimetric Analysis Repeat TCO Analysis if Necessary Aliquot Containing 15-100 mg Solvent Exchange Liquid Chromatographic Separation Seven Fractions: Low Resolution Mass Spectra Analysis TCO and Gravimetric Analysis Figure 1 Organic analysis methodology. -8- Five liters of aqueous solutions (5 L) were extracted two separate times, at pH 2 and pH 12, using a 50— mL volume of CH CI per liter of solution for each extraction step. The extracts were combined to give a total volume of 1000 mL. A total chromatographable organics (TCO) determination was run on each of the extraction solutions at concentrations of 1000 mL, 100 mL, and 10 mL; Kuderna-Danish evaporators were used for the concentrations. The TCO analysis is designed and calibrated to quantify the weight of organics that have boiling points between 100 and 300 C. The TCO were determined using a Perkin-Elmer Sigma I gas chromatographic system. A 1.8-m by 3-mm O.D. stainless steel column packed with 3 percent SP2100 on 100/120 Supelcoport was used for separation. The eluted organics were analyzed with a flame ionization detector. The solvent was then evaporated completely by carefully heating at 40 C under a stream of dry nitrogen. The organic residues were dried to constant weight in a dessicator. The final weight is designated the gravimetric analysis (GRAV). An infrared (IR) spectrum was obtained for each organic residue using a Perkin-Elmer 283B Infrared Spectrophotometer. Where there was sufficient material available, the IR spectrum was obtained using a neat smear of the sample between two NaCl plates. Where only smaller (approximately 5 mg) amounts were available, the sample was dissolved in carbon tetrachloride (CC1 ) for IR analysis. All sample extracts of adequate quantity (>15 mg) were subjected to liquid chromatographic (LC) separation. The LC column was 200 mm x 10.5 mm ID, glass, with Teflon stopcock and a water jacket. Davison Silica Gel, 60 to 200 mesh, Grade 950 (Fisher Scientific Company) was used as adsorbent. The adsorbent was sequentially extracted with methanol, methylene chloride, and pentane, then activated at 110 C for 2 hours just prior to use. Table 1 shows the sequence for the chromatographic elution. The column elution rate was maintained at 1 mL/min. A TCO analysis was run on each fraction from the LC separation. The solvent was evaporated by gentle heat and a stream of dry nitrogen. GRAV analysis and an IR spectrum was obtained for each fraction. Low-resolution mass spectrometry (LRMS) was obtained for only selected samples of organic material extracted from the KG sample. Information derived from the LRMS of the KG sample was incorporated into the char- acterization of the organics extracted from the other solid wastes. High-pressure liquid chromatography (HPLC) analyses were run using UV detection at a wavelength of 254 nm. An altex Ultrasphere ODS, 5 pm, -9- Table 1. Liquid-chromatography elution sequence. Fraction Solvent composition Volume (mL) LC-1 LC-2 LC-3 LC-4 Paraffins, olefins Small aromatics (benzenes, naphthalenes) Large aromatics Large aromatics, some low polar and S compounds Polar species (e.g. phenols, alcohols phthalates, amines,) with increasing polarity Weak acids Pentane 25 20% Methylene chloride in pentane 10 50% Methylene chloride in pentane 10 Methylene chloride 10 5% Methanol in methylene chloride 10 20% Methanol in methylene chloride 10 50% Methanol in methylene chloride 10 ISGS 1983 Figure 2 HPLC of a known mixture of phenols and polyaromatic hydrocarbons referenced to toluene. Peak identification: 1. phenol; 2. p-cresol; 3. 2, 3-dimethylphenol; 4. 2, 4, 5-trimethyphenol; 5. toluene; 6. phenanthrene; 7. pyrene; 8. chrysene; 9. benzo (a) pyrene. -10- 4.6 x 250 mm HPLC column (Beckman Instruments, Inc., Berkeley, CA) equipped with a guard column was used. The total extractable organics and the LC fractions were dissolved in methanol for HPLC analysis* Toluene was added as an internal standard. The elution solvent was 80:20, v:v, methanol: water with a 1 mL/min flow rate. A mixture containing four methyl-substituted phenols, toluene, and four poly- aromatic hydrocarbons was prepared and used as a reference standard (fig. 2). MINERALOGICAL CHARACTERIZATION W. R. Roy and R. M. Schuller It is reasonable to predict that the predominant mineral assemblages in mining spoils and preparation wastes will be much like those in the sed- imentary strata associated with the coal seam and the mineral matter within the coal. Coal preparation wastes consist primarily of shale clasts, fragments, and nodules; pyritic coal; pyrite lenses and nodules; and underclay fragments. The three refuse samples (WSNG, WSOG, and OBG) and the slurry sample (WSSL) were mineralogically similar (table 2). They were primarily com- posed of quartz, pyrite, illite, kaolinite, feldspars, and anhydrite. These results were comparable to those reported elsewhere: Wewerka et al. (1978) also identified quartz, pyrite, illite, and kaolinite in three refuse samples, and generalized that most refuse samples from Illinois Basin coals are composed largely of clay minerals with lesser amounts of quartz and iron sulfides. They also found marcasite, calcite, and gypsum as minor components in some samples. Bush et al. (1974, 1975) noted the presence of chlorite and muscovite in addition to the major minerals previously mentioned. While the major mineralogical assemblages are similar between studies, the discrepancies relative to minor minerals may stem from masking effects. The abundance of quartz and pyrite in the samples of this study made further characterization by X-ray diffraction difficult, due to tbe masking by these dominant minerals. There was some indication that coquimbite (Fe ? (S0 ) • 9H 0) was present in refuse samples WSNG and WSOG, and that hematite was present in refuse OBG. Neither mineral was identified by Bush et al . (1974, 1975) or Wewerka et al. (1978) in their refuse samples. Griffin et al. (1980) reported a similar masking that was overcome to a certain extent by the application of Mossbauer spectroscopy, a more definitive technique for differentiating iron species . The Mossbauer spectrum of refuse OBG suggested a mixture of ferrous and ferric iron clays (table 2). The presence of pyrite was anticipated but not clearly established. The amount of pyrite may have been below the resolving capabilities of this technique. -11- Table 2. Mineralogical composition of the seven solid waste samples by X-ray diffraction. WSNG WSOG OBG WSSL DMNS DMOS KG Nonferrous Minerals anhydrite tr X X X X X anker ite tr calcite X X chlorite X tr dolomite X feldspars X X X tr X X X gypsum tr illite X X X X X X kaolinite X X X X X X mixed layer clay X X X X X Mg-calcite ? mullite X muscovite X X quartz X X X X X X X siderite X Ferrous Minerals coquimbite x 2 x 2 Fe 2 clay x 1 x 1 x 1 X 1 x 1 x 1 Fe +3 clay x 1 x 1 x 1 x 1 x 1 x 1 hematite x 2 x 2 natrojarosite x pyrite X 1 x 1 x 2 X 1 tr tr 1 determined or varified by Mdssbauer spectra. 2 not found in MSssbauer spectra. The Mossbauer spectrum for the slurry sample was resolved into three absorptions, similar to the spectra of refuse WSNG and WSOG. The most prominent peak was attributed to pyrite, while the other two were identified as ferrous and ferric clays. The mineralogy of the two spoil samples largely reflects the composition of the Lawson Shale, a micaceous and calcareous gray shale. DMNS is en- tirely Lawson Shale; visual inspection of hand samples of DMOS revealed pre-mining soils, weathered and unweathered flakes of shale, and -12- carbonate fragments. The mlneralogical composition as indicated by X-ray was essentially quartz, illite, kaolinite, muscovite, and calcite. The iron species found in the two spoil samples by Mossbauer spectroscopy were characteristic of clays. For both samples, there were two ferric and two ferrous species. The ferric absorptions have param- eter values typical of trans OH and cis OH sites in illite and similar dioctahedral clays. The Mossbauer parameters for the cis OH species were within the range usually assigned to pyrite, which is often the case. X-ray diffraction indicated only traces of pyrite, so the assignment of a clay species was preferred. The presence or absence of iron sulfate species identified by X-ray diffraction in the two spoil samples could not be confirmed by Mossbauer spectroscopy. The absorp- tion values for the iron sulfates fell within ranges that were also characteristic of clays, so that positive identification was not possible. The Kilngas solid waste sample possessed a mineral assemblage that indicated high temperature and oxidizing conditions. X-ray diffraction data indicated the presence of quartz, mullite, and iron oxides that are commonly found in coal fly ash (Roy et al., 1981), another high- temperature residue of coal. Although feldspars appear to be rare in coal fly ash, they were present in the Kilngas waste sample. Griffin et al . (1980) found Lurgi ash samples were mineralogically similar to coal ashes, but they also contained feldspars. The Mossbauer spectrum was characteristic of a water-quenched bottom ash (slag); the absorptions were broad and unresolved. Most of the iron was identified as ferrous iron with one ferric species. The broad absorptions were typical of iron in silica glass. Although hematite was suggested by X-ray diffraction, it was not confirmed by the Mossbauer data. INORGANIC CHEMICAL CHARACTERIZATION W. R. Roy Major and Minor Constituents The initial step in assessing the potential pollution hazards of coal solid wastes is commonly a complete chemical characterization of the samples. Such analytical data will determine the maximum amount of a given constituent that may be released to the environment. Mineralogical data (table 2) indicated six of the seven wastes were composed of primarily aluminosilicates; thus, the major elements in the refuse samples (table 3), the spoil samples (table 4), and the Kilngas sample were silicon and aluminum. Reported as percent silica (SiO ) and alumina (Al Oj, the SiO content of the refuse samples WSNG, WSOG, and OBG ranged From 30.21 percent to 40.43 percent, and Al ranged from 7.88 percent to 9.66 percent. Comparisons between these three refuse -13- •H to C CO o u ■H C 4-1 0) rt 3 M 4-1 4-1 •H c 4.) Ol H c o •H 4-) C •H •rH m tf> (1) Ol o CO co 00 00 m ~~ m m CO •— o 01 CO cy> 00 co r-» vO \D cr> «— • o 00 o 1^ P* m 3 H y-t m o 00 m m O o co •— • a\ o CO o CO 0) m ■— « ~~ CM CO o ■rH t-i I 1 1 1 1 1 1 I i 1 1 1 O CO m 00 m m CN *— ■ ~~ -tf o ' — ' r^- c; C T— i O o 00 in •— ■ O ~~ 00 O 00 o CM •H ■H ."A • • • • • • • • • • • • ■—I o a\ o r~- CI o o o ~- O CO o ■ — ' <1) r- 1 o CN ■ — ' V Ml M G ol cd r^ m vO 00 ^o VO m m — o —• CM r^- o vO r~* 00 ■— « CT> o — • CO O o a PQ o o o\ S ■z. CO CO o u o •rH o C •rH 01 C on cfl S-l on o CO u c o o •rH CO -14- Table 4. Major and minor constituent composition of the two spoil samples, the slurry sample, and the Kilngas waste. Average coal fly ash DMNS DMOS WSSL KG composition* Slag** Si0 2 59.03 64.00 15.51 49.00 44 47.69 Ti0 2 0.09 0.07 0.03 0.07 1.3 0.74 A1 2 3 15.82 13.14 4.41 22.09 23 15.98 Fe 2 3 6.66 5.89 5.02 16.84 11 19.62 CaO 3.26 3.78 0.27 4.27 8.2 6.11 MgO 0.11 0.08 0.12 0.15 2.7 0.84 Na 2 0.73 0.57 0.14 0.41 1.8 0.80 K 2 4.18 3.54 0.96 3.10 2.0 2.43 P2O5 0.14 0.10 0.03 0.03 0.8 0.18 S0 3 2.07 1.20 10.56 1.50 1.60 0.28 Organic C 2.19 2.63 57.24 4.48 5 — Inorganic C 1.00 0.78 0.04 0.01 — — * Roy and Griffin, 1982. ** Griffin et al. 1980. Table 5: Sulfur species and total sulfur content of the seven solid wastes (percent weight). WSNG WSOG OBG WSSL DMNS DMOS KG Sulfate sulfur 0.30 1.79 3.22 0.56 0.06 0.04 0.05 Pyritic sulfur 9.56 5.56 0.25 2.84 0.76 0.44 0.12 Organic sulfur 0.44 1.42 0.17 0.83 0.01 0.00 0.43 Total sulfur 10.30 8.77 3.64 4.23 0.83 0.48 0.60 samples, two samples studied by Griffin et al. (1980), and the three samples studied by Wewerka et al . (1978) suggest that the SiO„ content of refuse from Illinois Basin coals varies from about 29 to 5o percent (table 3). The Fe ? 0- content of seven of the refuse samples was about 13 percent; SO varied from about 1 percent to nearly 34 percent. The iron and sulfur contents were partially related to the pyrite content, which was the principle reactant in the production of acidic effluent. Analytical data for sulfur species (table 5) indicated that nearly all (92.8%) of the sulfur in WSNG was pyritic; whereas only 0.30 percent was the sulfate form. In WSOG, which is about 20 years old, about 63.4 percent -15- occurred as pyritic sulfur. Also, WSOG contained 1.49 percent more sulfate sulfur than the younger WSNG sample. The oldest refuse sample, OBG (about 50 years old) contained only 6.87 percent pyritic sulfur; most of the sulfur, 88.5 percent, occurred as sulfate compounds. The apparent shift from pyritic forms of sulfur to sulfate forms with increasing age of the wastes may be due to the gradual oxidation of the pyrite in the refuse piles, with the formation of sulfate compounds as a by-product. A simplified reaction for the formation of sulfuric acid from pyritic wastes (Wewerka et al . , 1976) would account for this apparent trend: 2FeS 2 + 70 2 + 2H 2 + 2FeS0 tt + 2H 2 S0 4 The relative acidity of leachate produced by refuse may be buffered by the neutralizing effect of carbonates mixed within the wastes; however, the calcium content, reported as percent CaO (table 3) was low, ranging from about 0.15 percent to 6 percent. The CaO content of the three samples in this study appeared to be lower than those reported by Wewerka et al. (1978) and Griffin et al. (1980). Although the methods used to mine and clean the coal will ultimately influence the composition of the resulting refuse, the trends for iron concentration observed here agree with the conclusion by Gluskoter et al. (1977) that coals from the Illinois Basin tend to have higher Fe concentrations than those from the Appalachian Basin (table 3). Organic carbon was a major component in the three refuse wastes, WSNG, WSOG and OBG, ranging from about 14 to 30 percent by weight. The fact that organic carbon makes up much of the bulk composition of refuse is often overlooked in published studies. The major and minor constituents in the two spoil samples DMNS and DMOS are shown in table 4. The composition of these two samples is typical for sedimentary rocks (Krumbein and Sloss, 1963), with the exception of the higher sulfur concentrations. As with the unweathered refuse sample (WSNG), most of the sulfur (92%) occurred in pyrite forms (table 5). The spoil sample DMNS is a composite sample taken from mine outcrops of the Lawson Shale. We believe that the mineralogical and chemical data in this report may represent the first such published data on this shale. Studies of Paleozoic shales in Illinois by the Illinois State Geological Survey researchers include the work by Lamar (1948), Grim, Bradley, and White (1957), and Kosanke et al. (1960); but none of these studies contains analytical data for the Lawson Shale. The strati- graphic equivalent of the shale in Indiana has also not received attention from the Indiana Geological Survey (Shaver et al., 1970). However, it should be noted that there was little stratigraphic control with respect to sampling; thus the results may be suggestive but not decisive. -16- The major component of the slurry sample (WSSL) is organic carbon (table 4). Inspection of hand samples indicated that much of the material con- sisted of fine-grained coal particles, reflecting its mode of deposition as a slurry mixture pumped into a storage basin next to the source prep- aration plant. Unlike the other six wastes, WSSL had a low SiO content (15.51%) -relative to most geological materials; but a sulfur trioxide (SO ) concentration (10.56%) was within the range expected for SO con- centrations of coal preparation refuse from Illinois Basin coals (table 3). About two-thirds of the sulfur in WSSL occurred as pyritic compounds (table 5). The mineralogy of the Kilngas waste was described as being similar to coal fly ash and slag in many respects. The major elemental composition was also comparable to other high-temperature combustion residues (table 4). Comparisons between the composition of the KG sample (table 4), an empirical formula for coal fly ash from Roy and Griffin (1982), and a slag sample (Griffin et al., 1980) suggested the KG sample had an intermediate composition: the concentrations of some major and minor constituents fell between the ranges in composition established by fly ash and the slag sample. Trace Constituents The concentrations of trace constituents in the three refuse samples WSNG, WSOG and OBG are listed in table 6, along with data for other refuse samples for comparisons. As with most natural materials, the trace constituent concentrations of refuse are variable. Arsenic varied from 12 to 94 mg/kg in refuse samples, vanadium from 4 to 86 mg/kg, and zinc from 39 to 500 mg/kg. The central concept of coal cleaning is to concentrate the coal from the noncarbonaceous materials, e.g. silicates, sulfides, oxides, phosphates; therefore, trace constituents that generally occur with inorganic phases are concentrated in cleaning wastes relative to the concentrations of these same constituents in the material mined. For example, As, Co, Ni, Pb, and Sb commonly occur as sulfides (Gluskoter et al., 1977) and could be expected to be concentrated in refuse by present coal cleaning tech- niques. In fact, the arithmetic means of As, Co, Ni and Pb in the samples of Illinois Basin coal refuse were found to be more than twice the arithmetic means (Gluskoter et al., 1977) of the same constituents in Illinois Basin coals. The trace constituents of the two spoil samples, DMNS and DMOS (table 7) are very similar, with the exception of F and Mn. Barium, F, Mn, Sr, and Zn were the predominant trace constituents in these two samples; concentrations exceeded 200 mg/kg. The trace-constituent concentrations of the slurry sample were broadly similar to those of the refuse samples WSNG, WSOG and OBG, although As, Ce and F were lower than in Illinois Basin coal refuse. The Kilngas sample was broadly comparable to the other six wastes, with the exception of the relatively high levels of Cr -17- C Q) - U) • 3 rH > cd cfl O M u u CD 01 13 en o> CO en CO 3 en cu •H H O C i-( •rH Cfl ■-I o i-l o o 00 0) CN o\ 4-1 ■— ■ 05 • *^ CO O 3 c C 4J cfl CD i— I Ph cfl ^ u CD < 3 0) 4-1 3 c cfl PL. CN CO O CO I CN I CO in m m CN CM CN I I I co co cn m CN I cn o 1 1 PQ vO 00 vO ■— " vO vO ^ 3 cfl 0) VH oo cfl ■H 3 G O U Q fcd fa O O EC PC t-J J ►J S S -18- oi M C to Oi T3 01 3 C H XI H < m a) c > Q) o M 4-1 H • 0) 4-1 13 i— I H CO n) co ,sd o o m ra u B 3 o CU o *3» CO 0) ffl U) m 00 pa D •— ■ O^ in y-i t-^ co — m o o o O I -^■o"\inooo -rr m co — ■ o CN — co m co oo I I I CN CN -3" VO VO CTl CN C\ CO CO — CO m CN O vO CN — 4-) m m o W • m m o rd O CN £ c I I CNOcOcOOco 4-J u o a 01 co -19- Table 7. Trace constituent concentrations in the slurry, mine spoil, and Kilngas samples (mg/kg) . Sample WSSL DM0 S DMNS KG As 9 15 9 3 Ba 223 533 573 560 Br 7 <2 <2 <2 Cd <0.5 <1.9 <1.9 <1.9 Ce 38 119 98 119 Co 9 20 16 24 Cr 30 89 71 1437 Cs 2 5 5 17 Cu 16 25 25 205 Eu 1 1 2 2 F 150 460 363 75 Ga 7 17 22 25 Hf 3 10 9 6 Hg 0.12 0.11 0.17 0.07 La 17 43 57 47 Li 24 51 53 64 Lu 0.2 0.6 0.6 0.7 Mn 150 530 1075 424 Ni 16 33 29 210 Pb 40 31 33 25 Rb 17 43 57 47 Sb 3 2 2 1 Sc 5 15 13 19 Se 4 14 11 5 Sm 4 9 10 9 Sr 57 192 225 206 Ta 0.3 1 1 1 Tb 0.5 1 1 1 Th 5 17 14 18 V 4 7 7 20 W 0.6 <1 <1 1 Yb 1 4 3 4 Zn 58 106 92 113 Zr 95 252 278 130 -20- and Ni. Chromium levels of this magnitude (1531 mg/kg) are generally not associated with solid wastes from coal utilization (Wewerka et al., 1976; Griffin et al., 1980; Roy et al., 1981). The observed Cr concentration may have been an artifact of the gasification process. AQUEOUS SOLUBILITY OF THE SOLID WASTES W. R. Roy, I. G. Krapac, and R. M. Schuller Solubility studies by laboratory extraction or leaching procedures are currently considered the best available approach for assessing the potential environmental impact of le.achates from solid wastes. At present, there are more than 50 different extraction methods cited in the literature, although many are essentially variations of the "shake- test." Some procedures were designed to simulate a particular disposal scenario and may not apply to more than one type of waste or technique for its disposal. It would be expeditious if one extraction procedure could be developed to represent all possible disposal systems and waste materials. At present there is no universally accepted procedure. Moreover, such a universal procedure may be an unrealistic expectation. One objective of this investigation was to generate aqueous extracts from selected coal utilization wastes using five extraction procedures that are most widely used by the scientific community or have been pre- viously studied by the authors. The five extraction procedures that were used with each of the seven wastes included (1) the American Society for Testing and Materials Method-A (ASTM-A), (2) the ASTM-B method, (3) the U.S. E?k Extraction Procedure (EP), (4) a carbonic acid extraction procedure (CAP), and (5) a long-term equilibration procedure (LTE). Also, a large-volume generation (LVG) procedure was developed for the chronic bioassays. ASTM-A and ASTM-B The methods developed by the American Society for Testing and Materials (1979) are the Method-A (water shake extraction) and Method-B (acid shake extraction). These procedures involve (1) the shaking of a known weight of waste with either water of a specified composition or an acidic buffer solution, and (2) the separation of the aqueous phase for analysis. Both were designed to quickly obtain a solution for eval- uating extractable materials from wastes. The ASTM procedures recommend using the waste materials in the physical form in which they were disposed. For the seven wastes currently being studied, this recommendation was found to be impractical. The extreme variability of particle size (clay to boulder) would have made effective -21- subsampling impossible. Thus, the samples were air dried and ground to pass a 9.53-mm (3/8-inch) standard sieve. The ASTM-A procedure used is a short-term shake test and involved shaking a slurry composed of 700 ± 1 g of waste with a volume of distilled-deionized water equal to four times the weight of the sample. The slurries were formed in one-gallon, wide-mouth glass bottles and sealed with Teflon-lined plastic lids. Great effort was taken to avoid contact between the slurries and any surface that might cause sample contamination. Use of the Teflon liners prevented leaching of plastisizers from the plastic lids, which could have interfered with organic analysis of the samples. The slurries were shaken for 48 hours at 100 strokes per minute using a large capacity reciprocal shaker; then they were allowed to settle for one hour. Immediately after the bottles were opened, pH, oxidation-reduction potential (Eh), specific conduc- tance, and alkalinity measurements were made. Next, the slurries were filtered through Whatman Qualitative I filter paper, followed by filtering through Millipore 0.45-um pore size, cellulose acetate membrane filters. Subsamples were taken for cation, anion, and organic characterization. The ASTM-B procedure was conducted identically to the ASTM-A procedure; however, a sodium acetate-acetic acid buffer solution (pH 3.5 ± 0.1) was the extraction media. U.S. EPA Extraction Procedure The EP was also intended to serve as a quick test for identifying wastes that are potential pollution hazards (U.S. EPA, 1980). The EP required grinding the solid sample to pass through a 9.53-mm (3/8-inch) standard sieve. Wide-mouth, one-gallon glass bottles with Teflon-lined lids, identical to those for the ASTM procedures, were used for the EP. Also, the samples were shaken at the same rate using the same shaker. A total of 200 g of solid waste and 3200 mL of distilled water were placed in the gallon bottles and shaken for 24 hours ± 0.5 hour. During the 24- hour period, the slurries were periodically removed from the shaker and pH adjusted to 5.0 ± 0.1 pH units with 0.5 N acetic acid. The pH was adjusted at 15-, 30-, and 60-minute intervals, with advancement to the next longest interval when the change in pH did not exceed 0.5 pH units. Adjustment continued for a total of 6 hours. After the 24-hour period, the pH of the slurry was between 4.9 to 5.2. The total volume of acid added for pH adjustment could not exceed 4 mL per gram of solid waste. Adjustment was stopped when this volume was exceeded. At the end of the extraction period, the liquid and solid phases were separated. The volume of the liquid phase was then adjusted with distilled water so that its volume was 20 times that occupied by a quantity of water at 4 C, equal in weight to the initial solid sample. Sample filtration and preparation procedures were identical to those used for the ASTM methods. -22- Carbonic Acid Procedure The last short-term shake procedure was a carbonic acid procedure (CAP). This method involved shaking 700 g of sample with 2800 mL of deionized water that had been purged with CO gas. The CO -saturated water had a pH of 4.1. The extract was shaken for 48 hours, in a procedure similar to the ASTM methods. The rationale for using CO -saturated water as an extractant was to simulate the solubilization of the waste in contact with soil solu- tions. Soils often contain higher partial pressures of CO than the atmosphere above the soil, due to the aerobic metabolism or micro- organisms and the decay of soil organic matter. Also, carbonic acid was used as an acidic extract since carbonic acid itself was not toxic to the organisms used in the bioassays. It would have beeen difficult to assess the biological toxicity of the extracts due to solutes if the aqueous medium was also toxic as was the sodium acetate used in ASTM-B. Long-Term Equilibration Procedure The long-term equilibration procedure (LTE) used in this study was developed by Griffin et al. (1980) for samples of coal solid wastes. A variation of this procedure was also used by Suloway et al. (1983) for coal fly ash samples. This procedure involved mixing 1700 g of sample with 17 L of deionized water in a 19-L reaction vessel made of Pyrex glass. These mixtures were stirred for 30 minutes, twice a week, for 20 to 21 weeks in order to (as a first approximation) simulate slurry ponding environments. However, this extraction procedure is more specifically oriented toward generating a solution at chemical equilibrium with the solid wastes. This was done to produce a solution that may approximate the aqueous chemistry of pond effluent or leachate in settings where conditions of metastable chemical equilibrium develop. Samples of the extracts were periodically taken and analyzed over the 21 weeks to monitor the changes in chemical composition of the solutions. Large— Volume Generation Procedure A large-volume generation procedure (LVG) was designed to produce the large amount of acidic leachate that was initially anticipated for the chronic bioassays. Two 5678-L (1500 gal) Agro tanks were each filled with about 4.5 tons of coal refuse (fig. 3). Refuse from the OBG and WSNG sites was used in the LVG procedure. The selection of these two sites as source wastes was influenced by several factors including age, acute toxicity of extracts, and accessibility of large amounts of the material . -23- ■ °o. &° K 6/6; ^,-Wells-_ -Water level .J Coal refuse- V-! Inert gravel';! >zzflL pililifff Cutaway side view Figure 3 Corrugated perforated pipe Top view ISGS 1984 Cutaway side and top view of an Agro R tank set up to generate a large volume of coal waste leachate. A drainage system constructed of plastic corrugated and perforated pipe was installed in each tank (fig. 3), and four monitoring wells were evenly placed in each tank. The bottom of the tanks were each lined with chert gravel (the Meramac Gravel, a Mississippian-age gravel of relatively inert chemical composition) to ensure that the drainage system would remain permeable. Then each tank was filled with 6.53 kL (725 gal) of deionized water, completely saturating the solid waste (fig. 3). At the bottom of each tank a pump was connected to an outlet that was fed by the drainage system. The leachate was pumped at a rate of 454 L/hr (120 gal/hr) for 6 to 7 hours twice a week. The solution was recycled through the waste for 31 to 36 weeks, forming acidic leachates . The leachate was pumped through polyethylene tubing to a discharge point at the top of each tank, below the surface of the leachate. Samples of the leachate were periodically collected from the bottom drain, the surface of the solution, and the four monitoring wells to assess the compositional variability within the tanks. 1 By definition, solutions formed by shake procedures such as the ASTM-A and EP are extracts and not leachates, although the terms extracts and leachates are often used synonymously in the literature. Since the liquid phase was passed through the wastes in the LVGs, the acidic effluent was considered a true leachate. -24- Results Representative subsamples from the seven wastes were used in the generation of extracts by the four short-term shake procedures (ASTM-A, ASTM-B, CAP, EP) and the LTE procedure. The results of the short-term extractions, long-term equilibrations, and the two large volume applications are given in Appendix A. Although it is beyond the scope of this project, the triplicate data (not shown) for each waste for each of the short-term shake procedures did give some indication of the reproducibility of the methods. The limited amount of data does not permit statistical analysis and also, the reproducibility of the analytical methods used for each aqueous constituent would have to be evaluated and taken into account. In most cases, however, either the ranges of the triplicates fell within one standard deviation of the mean, or the means could be regarded as accurate to about 10 percent, depending on the constituent and analytical concentration. There have been several studies aimed at evaluating the nature of extraction methods (Lowenbach, 1978). Gulledge and Webster (1979) analyzed the precision and reproducibility of the ASTM-A, the ASTM-B, and the U.S. EPA's TEP or toxicant extraction procedure. Epler et al. (1980) conducted an evaluation of the EP using a wide variety of waste materials. Studies by Cox et al. (1977) and Chu et al. (1976) indicated that small-scale laboratory shake tests could not accurately simulate the field leaching characteristics of coal storage piles. Davis et al . (1981) further supported this conclusion for both short-term shake tests and column leaching methods. They further suggested that the data collected from laboratory procedures would have to be normalized with relation to the size of the storage pile and the duration and intensity of rainfall events before the in-situ leachate quality could be predicted from the laboratory results. On the other hand, Wewerka et al. (1978) concluded that the initial results from column leaching studies using coal refuse from the Illinois Basin were very represen- tative of in-situ leachates. This comparison was made with data from another study (Martin, 1974), however, not with leachates from the same piles from which the extracted material was collected. Further study of laboratory extraction procedures and their relationship to in-situ leachates is needed. It is difficult to generalize solubility trends from evaluating the short-term procedures in this study. Differences in the amount solute per gram of sample by each waste may stem principally from differences in pH and dissolution kinetics. It is not unreasonable to expect that solution concentrations will increase with decreasing pH where many aqueous metal ions are stable in solution and where the solid sample may dissolve. Thus, the acidic extractions such as EP (pH 5.0), CAP (pH 4.1), and ASTM-B (pH 3.1) would be expected to progressively extract more (depending on final pH) than the non-acidic ASTM-A procedure. The -25- Table 8. Soluble constituents: range in pH and percent extracted from four types of solid wastes, using four short- term procedures. Coal cleaning refuse (WSNG, WSOG, OBG) Mine spoil (DMNS, DMOS) Slurry (WSSL) KilnGas (KG) pH 1.62- 4.27 5.07- 7.57 2.43- 3.88 5.00- 8.10 Exti racted (%) Al 0.08- 3.7 _i 0.6 - 1.20 _i Ca 22.4 -80.0 1.6 -21„2 31.6 -38.4 1.2 - 4.4 Fe 0.04-28.0 _i 10.8 -19.2 _i Mg 22.0 -52.4 23.2 -70.0 42.0 -52.4 7.2 -84.0 Mn 10.7 -23.5 0.8 -28.2 36.0 -52.4 1.76- 6.64 Ni 9.2 -24.9 0.4 - 7.2 36.3 -52.2 4.72-12.0 SO^ 2.6 -31.3 6.92- 8.75 8.14-18.7 5.58- 9.45 Zn 2.7 -28.0 0.4 - 6.0 44.0 -53.2 0.32- 8.0 *all or most of the analytical concentrations were below detection limits. ranges in pH and the percent extracted for selected elements for each of the four types of wastes by the four short-term procedures are presented in table 8. Magnesium appears to be the only constituent that was essentially soluble in all seven wastes by all four short-term procedures. The slurry sample released substantial amounts (greater than 35% of the matric concentration) of Mn, Ni, and Zn independent of the extraction procedure. These anomalously high solubilites warrant further study. Unlike the short-term extracts, the long-term equilibrates (LTEs) permitted the study of changes in chemical composition with time. These LTEs were carried out for 20 to 21 weeks, at which time the experiments were terminated and the resulting extracts used in the toxicity characterizations. During the extraction interval, subsamples were periodically taken to monitor the chemical status of the solutions. The composition of the LTEs when the experiments were terminated is given in table 9. The pHs of the slurry and refuse samples were acidic presumably as a result of soluble Al salts, exchangeable Al, and the oxidation of sulfide minerals; whereas Kilngas samples generated an alkaline solution. The pH of many fly-ash water systems is alkaline and has been attributed to the hydrolysis of matrix oxides (Talbot, Anderson, and Adren, 1978; Shannon and Fine, 1974; Elseewi, Page, and Grimm, 1980). -26- Table 9. Chemical analysis of the long-term equilibrates (LTEs) after 147 days of equilibration (concentrations in mg/L) . WSNG WSOG OBG WSSL DMNS DM0 S KG pH 2.66 1.92 3.27 2.24 7.56 7.77 9.02 Eh (mv) 1 +686 +695 + 797 +855 +545 +536 +446 EC (dSnf 1 ) 3.4 10.9 2.2 4.6 0.8 0.57 0.6 Al 4 1.9 69.8 64.5 27.4 <0.07 <0.07 <0.07 As 0.57 Si 2 <0.14 0.58 <0.07 <0.07 <0.07 Ba <0.001 0.006 0.004 <0.01 0.02 0.04 0.004 Be 0.03 0.01 0.006 0.02 <0.01 <0.01 <0.01 Ca 262 123 458 60.4 144 67.0 40.8 Cd 0.48 1.26 0.02 0.64 <0.01 <0.01 <0.01 CI 6.5 2.90 15.8 <1.0 2.1 4.1 3.6 Cr 0.09 0.10 <0.02 <0.02 <0.02 <0.02 <0.02 Cu 0.03 0.51 0.01 0.58 <0.01 <0.01 <0.01 F - <0.50 <0.50 <0.50 <0.50 <0.50 <0.50 Fe 623 2552 0.21 971 <0.05 <0.05 <0.05 K 7.21 52.7 0.44 0.60 6.30 2.59 4.48 Mg 49.8 3.26 25.3 32.5 25.5 21.5 40.8 Mn 4.73 <0.05 2.35 6.42 0.15 <0.01 <0.01 Mo <0.02 <0.02 <0.02 <0.02 <0.02 0.05 0.12 Na <0.55 <0.55 3.60 <0.94 3.94 5.53 4.00 Ni 1.95 0.34 0.22 0.91 <0.05 <0.05 <0.05 Pb <0.01 1.67 <0.01 0.01 <0.01 <0.01 <0.01 Sb <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Se <0.06 <0.06 <0.06 <0.06 <0.06 <0.06 0.06 Si 143 103.1 51.2 55.2 3.45 7.25 7.06 Sn <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 S0 4 304 2 8721 1383 3001 427 245 314 V <0.02 <0.08 <0.06 <0.08 <0.06 <0.06 <0.06 Zn 5.43 Si 2 4.59 2.99 0.33 0.11 0.42 x relative to a normal hydrogen electrode. Concentration could not be determined due to spectral interferences (spe Appendix B). Roy and Griffin (1984)' incorporated this conclusion into a qualitative model of fly ash-water interactions that accounted for changes in solution chemistry as a function of time. Analogous investigations with coal conversion wastes are lacking. The pHs of the LTE extracts of the seven wastes (fig. A) were essentilaly constant after two weeks of equilibration. However, pH may not be a conclusive parameter to evaluate the development of chemical equilibrium or a meta-stable equilibrium in aqueous systems. The concentrations of iron in the four acidic extracts (OBG, WSNG, WSOG, and WSSL) varied to a greater extent than the corresponding pHs (fig. 5). Solution Fe was still increasing in the LTE extract of WSSL after 20 weeks. Sulfate concentrations (fig. 6) appeared to reach steady state conditions after about a week in the alkaline solutions, but continued -27- Table 10. Total organic-carbon content (rag/L) of generated leachates. Method Sample ASTM-A 1 ASTM-B 2 CAP 1 EP 1 LTE 1 WSOG 4.42 2717 20.59 136.0 4.73 WSNG 4.43 2837 7.19 1.05 1.08 OBG 2.42 2685 2.18 1.16 0.78 WSSL 34.25 2838 6.63 202.0 1.41 DMOS 13.1 2802 5.34 3 3.17 DMNS 0.87 2821 1.12 3 1.76 KG -0.45 3 2767 0.17 43.0 1.17 Blank 3.95 284 1 1.81 2.24 1 values corrected for blank. 2 values not corrected for blank. 3 value for sample less than blank. to increase in the acidic systems until reaching steady state conditions after about 11 weeks. Aluminum concentrations became essentially invariant in the WSOG and OBG extracts after about 10 weeks (fig. 7); but aluminum continued to go into solution in the LTE extracts of WSNG and WSSL throughout the extraction interval. The amount of solution silicon in the WSNG and WSSL extracts (fig. 8) paralleled aluminum in that the silicon concentrations had not reached invariant levels but were steadily increasing. In contrast, the amount of soluble silicon in the. alkaline samples decreased with time even though the pHs of these solutions remained alkaline. A plot of electrical conductance as a function of time (fig. 9) indicated that the acidic systems (OBG, WSOG, WSSL, and WSNG) had not reached equilibrium after 21 weeks; whereas the alkaline systems appeared to have reached steady state conditions with respect to the ionic composition of the solutions. ORGANIC CHARACTERIZATION OF THE SOLID WASTES AND LABORATORY EXTRACTS D. R. Dickerson and R. A. Griffin The total organic carbon (TOC) content of the laboratory extracts generated by the ASTM-A, ASTM-B, carbonic acid procedure (CAP), U.S. EPA Extraction Procedure (EP), and long-term equilibration (LTE) was determined for each of the seven solid wastes (table 10). The extracts generated by the ASTM-A and LTE methods presented no particular problem for routine TOC analysis. ■28- 10.0 8.0 6.0- c 3 I a 4.0 2.0 ^M ■ -o A OBG • DMNS ■ DMOS A WSSL O WSNG D WSOG J 8 *- A A A— KG —A 1 "^ O —A O— — O — A l i i i i D — t - ■■ - r — a 10 12 Weeks 14 16 18 20 22 Figure 4 Changes in pH for all seven long-term equilibrations. -29- 3 C _o o (U 0) m 3 c 05 -C LL CJ |-|/6lU) SUOUBJJU83UO0 9-j -30- OOM io l n/ 6lu ) suoijBJiuaouoo "OS 3 O" C o o o o o O O o o o r» ^r <- o c u * o GO c co C/> (1) O! m 3 c u. U (I/Bui) suojjejiuaouoD ''QS -31- _l C3 C3 C3 c/i 2 O m oo GO CO o 3 CD c o "D c o u < c 1^- CO ci> 0) m 3 c ca u_ U (~|/6iu) suoijejjuaouoo |v -32- (-|/6lU) SUOUBJ11J30UOD |S -33- O 2 O co 2 O O Q Q § 5 § * <]|i40ao (soqiuw) 33 -34- The CAP extracts were generated using water saturated with carbon dioxide, as previously discussed. To determine the TOC values for this set of solutions, it was necessary to first determine total carbon, then remove the inorganic carbon by an acid sparge, and follow with a second total carbon analysis. Thus, the second total carbon value is also the total organic-carbon value. The inorganic carbon is the difference between the first and second total carbon values. The EP necessitated the addition of organic matter via the buffering solution. Theoretically, it should have been possible to determine the TOC value of the leached organic matter by correcting for the amount of acetic acid added. This would appear to be valid for the KG extract, where a reasonable TOC was obtained. However, the TOC values determined for the DMOS and DMNS extracts were only about one-half the theoret- ically expected value calculated for the added acetic acid. The TOC values obtained for the WSOG and WSSL extracts (no acetic acid added) would be expected to be similar in value to those obtained for the ASTM- A extracts. However, for poorly understood reasons, they are about an order of magnitude greater. The ASTM-B extracts were also prepared using an acetate buffer system; thus the results should be interpreted as suggested for the EP extracts. The results, however, were somewhat erratic and showed the blank to have a slightly higher TOC value than any of the sample extracts. This may have been due in part to experimental error in attempting to find a small difference between two large numbers. It was more likely due to microbial utilization of some of the acetate buffer, which is an excellent substrate for microbial growth. Methylene Chloride Extraction of Leachates The results of the total chromatographable organics (TCO) and gravi- metric analyses of the methylene chloride (CH CI ) extracts of the solutions generated from the solid wastes are given in table 11. The results indicate that the TCO levels in the liquid phases were below the detection limits of the recommended EPA method and calibration range. The GRAV analysis yielded only very small amounts of extracted organics (table 11), consistent with the low TOC content of the extracts. The infrared (IR) spectra obtained for the CH CI extracts were all very similar, although five dissimilar extraction methods and seven different solid waste materials were used (fig. 10, for example). The main absorption bands were in the regions of 3400 cm -1 (H-bonded OH groups), 2900 cm -1 (aliphatic C-H stretching), 1735 cm" 1 (C=0 stretching of COOH, -C00R, or ketonic C-0), 1610 cm" 1 (aromatic C=C and/or H-bonded C=0), 1510 cm" 1 (aromatic C=C), 1460 cm" 1 (C-H deformation of CH or CH groups), 1365 cm" 1 (0-H deformation, CH bending, or C-0 stretching), 1250 cm" 1 (C-0 of ether linkage), 1100 cm" 1 (very broad and unresolved due to a mix of C-0, C-O-H linkages of alcohols, esters, and ethers), -35- Table 11. Total chromatographable organic s and gravimetric analysis of methylene c nloride (CH2CI2) extraction of coal waste extracts. Extraction Extract CH2CI2 TCO Gravimetric Sample method (D (L) (mg/L) (g/D ASTM-A Blank 3 0.6 <0.022 0.0001 WSOG 5 <0.022 0.0007 WSSL 5 <0.022 0.0003 WSNG 5 <0.022 0.0004 DMNS 5 <0.022 0.0007 DMOS 5 <0.022 0.0008 KG 5 <0.022 0.0004 OBG ASTM-B 5 <0.022 0.0004 Blank 2.5 0.5 <0.022 0.0003 WSOG 5 <0.022 0.0003 WSSL 5 <0.022 0.0003 WSNG 5 <0.022 0.0003 DMNS 5 <0.022 0.0005 DMOS 5 <0.022 0.0007 KG 5 <0.022 0.0004 OBG CAP 5 <0.022 0.0005 Blank 2.5 0.5 <0.022 0.0006 WSOG 5 <0.022 0.0008 WSSL 5 <0.022 0.0005 WSNG 5 <0.022 0.0006 DMNS 5 <0.022 0.0005 DMOS 5 <0.022 0.0007 KG 5 <0.022 0.0004 OBG EP 5 <0.022 0.0005 Blank 2.5 0.5 <0.022 0.0000 WSOG 5 <0.022 0.0003 WSSL 5 <0.022 0.0003 WSNG 5 <0.022 0.0003 DMNS 5 <0.022 0.0004 DMOS 5 <0.022 0.0006 KG 5 <0.022 0.0005 OBG LTE 5 <0.022 0.0003 WSOG 2.5 0.5 <0.022 0.0006 WSSL 2.5 0.5 <0.022 0.0005 WSNG 2.5 0.5 <0.022 0.0005 DMNS 2.5 0.5 <0.022 0.0006 DMOS 2.5 0.5 <0.022 0.0004 KG 2.5 0.5 <0.022 0.0006 OBG 2.5 0.5 <0.022 0.0003 -36- a I c I d e 4000 3000 2000 1600 1200 800 Cm ISGS 1984 Figure 10 Infrared spectra of the organics extracted by CH 2 CI 2 from extracts of WSSL generated by these methods: (a) LTE, (b) ASTM-A, (c) EP, (d) CAP, (e) ASTM-B. and 800 cm" 1 (aromatic substitution). The general shape of the conglomerate of peaks in the 3600 cm -1 to 2700 cm -1 regions indicated the presence of carboxylic acid groups, which should also be reflected in the 17 35 cm -1 peak. Peaks in the 1800 cm -1 to 1600 cm" 1 range may also be enhanced by overtone and summation peaks. From the functional groups indicated by the IR spectra, the organics contained in the extracts were of a highly polar nature, as expected in an aqueous extract. Since these organics are derived from coal solid wastes, it is likely that some may be included on the EPA priority -37- Table 12. Total chromatographable organic s and gravimetric analysis of the methylene chloride (CH2CI2) extraction of coal solid wastes. Values expressed for TCO are based on duplicate runs and computed from a standard curve. Error is estimated as ± 10%. Solid Sample (g) CH2CI2 (L) TCO (mg/kg) Gravimetric (g/kg) WSNG 150g 287 20.5 WSOG 150g 239 22.3 OBG 75g 3 153 7.23 WSSL 50g 3 372 18.9 DMNS 150g 27 0.73 DMOS 150g 33 0.22 KG 130 NA 1 0.91 Solvent ND 2 0.0001 *not analyzed. 2 less than detectable limit. 3 mixed with equal weight of organic free silica before extraction. pollutant list. They do not, however, appear to be in concentrations considered to be hazardous, although long-term accumulation may increase concentrations to hazardous levels. Characterization of Organics Extracted from the Coal Solid Wastes The soxhlet extraction of coal solid wastes WSOG, WSNG, WSSL, and OBG presented considerable difficulty. These four solid wastes each contained more than 20 percent organic carbon (table 3), which appar- ently brought about the problem. Only a small part of the organic matter was soluble in CH CI . The combination of another part of the organic material and the CH CI was a "tacky" semisolid. This latter combination tended to plug the extraction thimble and to render the bed of solid waste impervious to the bulk of the CH 2 C1 2 . Hence, during each cycle the majority of the solvent flowed over rather than through the thimble of solid waste. Three wastes (KG, DMOS, and DMNS), which contained less than 5 percent organic carbon (table 12) presented none of the above extraction problems. -38- Table 13. Elemental sulfur content of the methylene chloride (CH2CI2) extracts of the coal solid wastes. Sample Sulfur (wt %) WSOG 34 WSNG 4 OBG <50 WSSL DMOS £20 DMNS £17 KG <91 34.3 (elemental analysis) 4.2 (elemental analysis) (calculated by difference) Not determined (calculated by difference) (calculated by difference) (calculated by difference) In an attempt to circumvent this problem, a smaller quantity, 50 to 75 g, of coal solid waste (OBG and WSSL) was diluted with an equal weight of organic-free silica sand before extraction. This method worked very well for the OBG solid waste (about 14% organic C) , but less well for the WSSL waste (57% organic C). This fact indicates a need for a greater dilution (with an inert material) of wastes with a high organic carbon content. The CH ? C1_ extracts of the solid wastes were found to amounts of elemental sulfur (table 13). The presence sulfur was confirmed by a melting point-mixed melting The free sulfur was removed by passing the extract in separate column of activated copper (Blumer, 1957) or lower 5 cm of the liquid chromatography (LC) column wi copper. The quantity of elemental sulfur in the CH CI wastes DMOS, DMNS, OBG, and KG (table 13) were estimat percent extract recovered from the LC column. The su the CH 2 C1 2 extracts of WSOG and WSNG (table 13) was de ASTM method. contain varying of elemental point technique, solution through by packing the th activated extracts of ed from the lfur content of termined by the After extraction and, where necessary, filtration, analyses were continued as specified by the U.S. EPA (1978). The total chromato- graphable organic (TCO) was determined as prescribed in the Level I Manual; except that after many trials, it was determined that a 3 percent loading of SP 2100 (Supelco, Inc., Bellefante, PA) gave better resolution with the chromatograph than did either a 10 percent loading of SP 2100, or of OV-101 as recommended in the Level I procedures. -39- 4000 3000 2000 1600 1200 Cm" 1 800 ISGS 1984 Figure 11 Infrared spectrum of the methylene chloride (CH 2 Cl 2 ) extraction of the OBG solid waste. The results of the TCO analyses are given as milligram TCO per kilogram of solid waste in table 12. Also shown in table 12 are the grams of organic matter extracted per kilogram of solid waste. Except for the DMNS and DM0S (which are low in total organic carbon), the total chromatographable organics are 2 percent or less of the total extracted organic (gravimetric in table 12). The infrared spectra of the CH Cl ? extracts of the coal solid wastes did not generally show well resolved peaks due to the myriad of organic compounds with varying functional groups that can be extracted from a coal waste. The infrared spectrum of the CH CI extract of the OBG solid waste is depicted in figure 11. This spectrum is much better resolved than the spectra of the other wastes. Discernible in this spectrum are peaks at 3350 cm" 1 (H-bonded OH and/or NH 2 ), 3040-3020 cm" 1 (naphthalene-type aromatic C-H) 2950, 2925 and 2860 cm" 1 (aliphatic C-H stretching), 1800-1700 cm" 1 (C=0 stretching of COOH, -C00R, or ketonic C=0), 1600 and 1500 cm" 1 (aromatic C=C), 1460 cm" 1 (C-H deformation of CH„ or CH„ groups), 1356 cm (CH bending, or C-0 stretching), 1300 cm" 1 to 1000 cm" 1 (unresolved C-0, C-O-H linkages of alcohols, esters and ethers), 870, 810 and 740 cm -1 (aromatic substitution). The CH 2 C1 2 extracts of the other solid wastes are generally similar, but peaks were less well resolved. Liquid Chromatography of the Methylene Chloride (CH 2 C1 2 ) Extracts of the Coal Solid Wastes A portion of the CH Cl„ extractable matter from each solid waste was separated further into seven fractions, using liquid column chromato- graphy (LC). Each fraction was analyzed for TCO content and total gravimetric organic content. -40- Table 14. Gravimetric (GRAV) analysis of liquid chromatography fractions from methylene chloride (CH2CI2) extracts of the solid wastes. Fract: Lon (mg) WSNG WSOG OBG WSSL DMNS DMOS KG Total LC 92.7 100.2 102.9 100.0 112.0 90.4 41.2 LC 1 22.5 48.7 1.7 9.1 9.6 17.0 24.7 LC 2 4.4 13.6 1.0 5.5 1.8 2.4 3.6 LC 3 15.4 14.4 14.7 16.0 22.9 15.3 5.3 LC 4 16.4 7.8 8.6 7.9 15.8 ? .,| 2.2 LC 5 16.4 10.0 9.4 10.2 1.8 4.3 j LC 6 25.4 16.8 34.4 24.1 26.5 17.5 5.9 LC 7 -- 1.8 1.2 11.4 4.9 14.0 10.4 1.0 Recovered 102.3 112.5 80.2 77.7 92.3 72.0 42.7 Recovered (%) 110.4 112.3 77.2 77.7 82.4 79.6 103.6 The LC analyses were run on samples WSNG and WSOG before discovering that the extract was contaminated with large quantities of "free" sulfur. Some unforeseen effects of this sulfur may partially account for the greater than 100 percent recovery of the CH Cl 2 extractable matter from the LC columns for these two samples. For these two samples, the TCO quantity was below the detectable limits of the method and calibration range used. The free sulfur was removed from the WSSL, DMNS, DMOS, and OBG by passing the dissolved fraction over freshly prepared, reduced copper powder as part of the LC column packing. The recovery of material from the LC column for these samples was less than the amount loaded onto the column. The greater part of the amount lost was assumed to be due to the amount of free sulfur removed by the activated copper in the packing. The TCO was determined on all fractions from the LC runs. In only the first three LC fractions of each set was there any determined value for TCO that was above the lower detection limits of the calibration curve. Since these values, although positive, are so close to the lower detection limit, their validity must be questioned at this time. Also, the amount of TCO is relatively insignificant compared to the gravimetric values (table 14). In general, fractions 1, 3, and 6 contained the greater portions of the total LC fractions: LC fraction 1 -41- contained the paraffins, fraction 3 the large aromatics, and fraction 6 the polar compounds. These organic classes are in keeping with the type of solid waste being extracted. For all practical purposes, the organics found in the wastes originated from coal, and were quite similar. Thus, the LC analyses of only two wastes, OBG and KG, will be discussed. Also, it can be assumed that most compounds found in one waste will be found in the other wastes, al- though concentrations may vary. This was confirmed by the infrared (IR) spectra of the CH CI extracts, typified by the spectrum of the CH CI extract of the OBG solid waste (fig. 11). A gross interpretation of the spectrum indicated the presence of aliphatic (saturated and unsatur- ated), aromatic, and polyaromatic compounds with hydroxyl, amino, and carbonyl (ester, ketone, aldehyde, and carboxyl) functional groups. The IR spectra of the LC fractions of the CH CI extract of the OBG solid waste are presented in figure 12. They show a direct increase in complexity as the eluting solvent was made more polar. It would appear that the majority of the polar compounds contain oxygen functional groups. Peaks for nitrogen functional groups were no doubt present (shifting to slightly longer wavelength of the broad peak at 3450 cm" 1 to 3320 cm -1 ) but tended to be masked by other peaks. The spectrum of LC-1 had a weak peak at 3050 cm" 1 , indicating unsatur- ation rather than aromatic compounds. Also the peak at 1268 cm and at 745 cm" 1 may indicate t-butyl- and/or iso-propyl-groups. The latter may also be due to an aliphatic chain ~(CH) -, where n equals at least 4. The spectrum of the LC-6 fraction was almost identical to the spectrum of the total CH Cl~ extract (fig. 12). Approximately 40 percent of the total extract was round in LC-6. The CH CI _ extract of KG was subjected to analysis by high performance liquid chromatography (HPLC) and GC/MS. No nitrogen compounds were identified by gas chromatography /mass spectrometry (GC/MS) in the CH CI. extract of the Kilngas residue. The HPLC analysis of the total extract- able organics from KG was obtained with toluene as an internal marker. There were two prominent peaks, one preceding and another following toluene. The former is due to polar compounds such as phenols, and the latter to di- and polyaromatic hydrocarbons. The HPLC analysis of LC-2 showed essentially all the compounds eluting later than toluene in the range expected for naphthalene and alkyl substituted naphthalenes. The HPLC analysis of LC-3 showed two small peaks eluting ahead of toluene, indicating few or no polar compounds in this fraction; the larger concentration of peaks were in the elution range of napthalenes and phenanthrenes. Fractions LC-4 and LC-5 were combined before HPLC analysis; the chromatogram showed the majority of compounds eluting later than toluene. Since there was a relatively strong hydroxyl peak in the IR spectrum, one would expect these later peaks to be due to larger molecular weight phenols (>200) and to aryl esters, as well as -42- LC-2 LC-3 LC-5 LC-7 J I I I J L J L 4000 3000 2000 1600 1200 800 _ -1 ISGS 1984 Figure 12 Cm Infrared spectra of the liquid chromatography fractions of the methylene chloride (CH 2 CI 2 ) extract of OBG. -43- polyaromatics. The HPLC analysis of LC-6 showed a large envelope of early eluting peaks, as expected for highly polar compounds. Table 15 lists all compounds tentatively identified by GC/MS in the LC fractions of the CH CI extract of KG. The chromatogram of LC-1 (KG) showed a large envelope of both unresolved and well resolved peaks eluting in the C to C range; the major compounds identified were mostly n-paraffins (table 15). In the GC/MS chromatogram of LC-2, 36 peaks were analyzed; 19 were identified as alkyl substituted benzenes and biphenyls, or acenaphthene. In the GC/MS analysis of fraction ration LC-3, 14 major peaks were identified as fused ring aromatics with alkyl substitutions in most cases. The combined fractions LC-4 and LC-5 were not analyzed by GC/MS. Eleven of the compounds in fraction LC-6 were identified by GC/MS: five of the peaks were due to phthalate esters; two thiophenols were also identified. The aqueous extracts from the spoil banks and solid waste may contain some phenolic and other compounds (polyaromatic hydrocarbons) that are on the EPA priority pollutant list. None were identified directly in the organics extracted from the aqueous extract; however, they were identified in the organics extracted from the solid coal wastes by an organic solvent. By inference, those that are water soluble should be found in the aqueous leachate. Since not more than 0.8 mg/L total organic material was extracted, and since the TOC was also the same order of magnitude, it seems doubtful that any single priority pollutant would be present in the extracts in a quantity that would be hazardous. EQUILIBRIUM SOLUBILITY MODELING OF LABORATORY EXTRACTS AND FIELD LEACHATES FROM COAL SOLID WASTES W. R. Roy and R. A. Griffin The application of chemical equilibrium models can lead to useful insights into the chemistry of aqueous systems such as coal waste leachates or acid mine drainage. However, the results of such modeling must be interpreted cautiously. The lack of uniform experimental procedures can result in discrepancies in reported values for solubility constants for some mineral phases, making the determination of equilibrium controls difficult. Moreover, pure mineral compounds in distilled water are often used to determine thermodynamic parameters that might not be representative of these compounds when they are associated with complex multicomponent materials such as coal solid wastes. However, solubility models do provide a first approximation for understanding these complex solutions. As a step towards understanding coal-waste aqueous systems, the chemical analyses of laboratory extracts and field leachate samples were treated by the thermodynamic model WATEQ2 . This program uses equilibrium constants to estimate the theoretical activities of aqueous species and -44- Table 15 Compounds identified by gas chromatography /mass spectrometry (GC/MS) in the liquid chromatography fractions (LC) of the methylene chloride (CH2CI2) extract of the Kilngas residue. Fraction LC-1 Cg-Cyclohexane C 16 -Branched Alkane n-Hexadecane Cj Q-Cyclohexane n-He ptadecane Pristane (branched C 19 Hi t0 ) n-Octadecane n-Nondecane n-Eicosane n-Heneicosane n-Docosane n-Tricosane n-Tetracosane n-Pentacosane n-Hexacosane Fraction LC-2 2 C 1 -Naphthalenes 5 C 2 -Naphthalenes C 4 -tetrahydronaphthalene 5 C3-Naphthalenes 8 C^ -Naphthalenes C 7 -Benzene 2 Acenaphthenes or C 2 -Biphenyls 2 C3 _ Acenaphthenes or C 3 -Biphenyls Acenaphthene or Ci,-Biphenyl Fraction LC-3 2 C 2 -Naphthalenes 3 C3~Naphthalenes 3 Cij -Naphthalenes Phenathrene C 2 -Phenanthrene C3~Phenanthrene Acenaphthene or C^-Biphenyl Cj -Phenanthrene Cij-Phenanthrene Fraction LC-6 2,6-Di-t-butyl benzoquinone C 5 -Thiophenol C 6 -Thiophenol Diethyl phthalate Methoxyanthracene Substituted cychohexane 2,4 -bis- ( 1-Methyl butyl) phenol Phthalate (di-isobutyl?) Phthalate (di-butyl?) Phthalate (diethyl hexyl?) Phthalate calculates the saturation indicies for 309 minerals. The theory of WATEQ2 and its computer implementation are discussed elsewhere (Truesdell and Jones, 1974; Plummer et al., 1976; Ball et al., 1979). As previously discussed, subsamples of the LTE extracts were taken with time and analyzed. The analytical data from these samples were submitted to the WATEQ2 program. The results of the WATEQ2 program were -45- subsequently used as an aid in constructing solubility diagrams. In addition to the LTEs, the chemical composition of the ASTM-A extracts was also modeled using the final composition at the end of the 48-hour extraction interval. To provide a limited basis for comparison, field samples of waste streams at the Old Ben No. 9 mine site and the Will Scarlett mine site were also collected and modeled using WATEQ2 . Two samples were collected from the OBG site: one was a sample of acid mine drainage, flowing from the base of a refuse pile into a manmade basin. The second sample was taken from the ponded effluent. The chemical analysis of these samples (table 16) indicated that they were acidic and contained high concentrations of Fe, Al, and SO . The solid waste sample used to generate the laboratory extracts was taken at the same location from the refuse pile that generated the two effluent samples. Three field samples were taken at the Will Scarlett mine. One sample was of effluent from a ridge of coal cleaning refuse where the WSOG sample was collected. A second sample was collected from a slurry lagoon downstream from an active discharge pipe. The high water-to- waste ratio may account for the low iron and SO 2 concentrations (table 16) observed at the site, relative to the other samples. A sample of stagnant effluent from an inactive slurry basin (table 16) contained 68 times more iron in solution than the current basin and a lower pH. The inactive lagoon was estimated to be about 40 years old; the differences in chemical composition between the two sites may be a function of the water-to-waste ratios, parent waste variation, and duration of pyrite dissolution and acid formation. Neither site corresponds to the slurry sample taken for the laboratory extracts (WSSL) because that particular basin was dry at the time of sampling. All three slurry deposits were in proximity to one another. While the samples suggest the composition and variability of waste streams in the field, they were not collected in an attempt to characterize acid mine drainage in the two mining areas — an objective well beyond the scope of this investigation. These samples were collected to evaluate the solubility relations predicted by WATEQ2 for laboratory extracts, relative to some actual field waste systems. The acidic solutions, both laboratory and field samples, tended to be in equilibrium with a calcium sulfate (CaSO ) phase (fig. 13). This may be true for most solutions influenced by acidic coal solid wastes. Griffin et al. (1980) found that three Lurgi ashes, a medium-temperature char, and a liquefaction residue were generally in equilibrium with gypsum (CaSO »2H 0); whereas a fly ash sample (an acid Fersic by the classification system of Roy and Griffin, 1982) and a high-sulfur, cleaning refuse sample were in equilibrium with gypsum in long-term equilibrated extracts. Roy and Griffin (1984) also noted that both acidic and alkaline coal fly ash extracts from Illinois Basin coals equilibrated with anhydrite in long-term studies. The LTE of WSNG was -46- Table 16„ Chemical composition of acid mine drainage from the Old Ben No. 9 mine (OBG) , and the Will Scarlett mine (concentrations in mg/L) . OBG WSOG OBG acid mine acid mine WSSL WSSL pond drainage drainage discharge inactive pH 2.83 3.25 2.40 3.35 2.63 Eh (tnV) 1 +686 +493 +735 +747 +643 EC (dSm -1 ) 3.88 13.98 5.79 2.55 9.01 Al 190 1120 272 43.9 664 As <0.07 <0.07 <0.07 <0.07 <0.07 B <0.01 <0.01 <0.01 <0.01 <0.01 Ba 0.009 <0.002 <0.002 <0.002 0.02 Be 0.02 <0.001 <0.001 <0.001 0.01 Ca 288 445 218 695 471 Cd 0.17 <0.03 <0.03 0.07 0.04 Cr 0.03 <0.02 <0.02 <0.02 <0.02 Cu <0.02 <0.02 0.45 0.34 <0.02 Fe (total) 669 5625 1440 24.1 1645 K 2.63 <1.54 3.74 5.80 <1.54 Mg 94.5 502 54.0 96.8 694 Mn 12.9 65.3 14.0 12.3 184 Mo <0.01 <0.01 <0.01 <0.01 <0.01 Na 65.0 260 < 1 .79 20.2 < 1 . 79 Ni 0.55 <0.01 <0.01 <0.01 5.05 Pb <0.08 <0.08 <0.08 <0.08 <0.08 Sb <0.03 <0.03 <0.03 <0.03 <0.03 Se <0.08 <0.08 <0.08 <0.08 <0.08 Si 12.4 33.8 8.33 7.49 38.6 Sn 0.04 <0.04 <0.04 <0.04 <0.04 V 0.06 <0.06 <0.06 <0.06 <0.06 Zn 11.5 56.0 7.25 4.61 37.7 SO,, 2- 3841 20,397 6194 1805 12,183 relative to a standard hydrogen electrode. -47- o CO a 1.4- 1.6- 1.8- 2.0 2.2 2.4- 2.6- 2.8- Supersaturated © WSNG ▲ WSOG • OBG £_ WSSL Slurry (inactive) Slurry discharge ▲ 147 D 30 DA ASTM-A A O AMD 10D 24 H © ASTM-A • ASTM-A MOD ) 24 H <2> O10D 1 Q28D © 24 H Undersaturated 3.2 — i — 3.0 2.8 - 1 — 2.6 ~1— 2.4 — I — 2.2 — i 2.0 pCa 2 Figure 13 Calcium sulfate equilibria of the acidic laboratory extracts (ASTM-A and the long-term equilibration) and field leachate samples. -48- initially undersaturated with respect to gypsum, then came into equilibrium with this solid phase after 75 days. The ASTM-A extract of WSNG also was in near equilibrium with gypsum, suggesting that even though the two laboratory extraction methods are dissimilar with respect to slurry concentrations and extraction intervals, the solubility of gypsum exerted a dominant influence over calcium and sulfate activities in both systems. The LTE of the WSOG sample appeared to be under- saturated with respect to both gypsum and anhydrite during the entire extraction interval. This situation could be interpreted in three different ways: (1) The solution had not reached thermodynamic equilibrium, thus, the liquid phase would have equilibrated with a CaSO phase if the equilibration interval had been extended. (2) The solution was in equilibrium, but a calcium sulfate phase more soluble than gypsum or anhydrite was present. Anhydrite was detected in all of the seven waste samples with the exception of KG. (3) The WATEQ2 model could not yield accurate data due to the high ionic strength of the solution. The last interpretation is made in reference to the method by which this thermodynamic model calculated the activities of Ca 2 and SO. . The activity coefficient of a given constituent may be calculated using equations resulting from the Debye-Huckel Limiting Law (table 17). For dilute solutions (an ionic strength of <0.005), equation 1 is used. For solutions having an ionic strength greater than 0.01, the more reliable Extended Debye-Huckel equation 2 is used, which attempts to take into account the hydrated sizes of the ions in addition to the electrostatic interactions. However, this equation begins to break down in solutions containing ionic strengths greater than 0.1. The Davies equation (table 17, equations 5 and 6) can be used to estimate individual activity coefficients in solutions having an ionic strength up to 0.5 M. The WATEQ2 program uses both the Debye-Huckel equations and a Davies- Extended Debye-Huckel equation (7): but the single ion activity coefficients of Ca 2+ , Mg 2+ , Na + , K + , CI', SO^", C0 3 2 ", and HC0 3 " are always calculated by the Extended Debye-Huckel relationship, since WATEQ2 was written primarily for natural surface waters. Several acidic, high-salt extracts and leachates had ionic strengths in excess of 0.1; thus solubility interpretations involving Ca 2 and SO 2 must be made with extreme caution. The ASTM-A extract of WSOG appeared to be in equilibrium with anhydrite (fig. 13), which was found to be present by X-ray diffraction. Unlike the laboratory extracts, the field sample of acidic effluent at the WSOG site was in equilibrium with gypsum. The LTE extracts of WSSL were undersaturated with respect to CaSO phases during the equilibration interval. The ionic strength of the -49- Table 17. The Debye-Hiickel Limiting Law and variants. Debye-HUckel Limiting Law (1923) log yi = -AziVT (1) (2) Extended Debye-HUckel (1923-1934) log yi = ] ^Ba/I Giintelberg (1926) log yi = '^ fa + bI (3) Guggenheim and Schindler (1934) log yi = . ;, (4) Davies (1938) log vi = -Azi 2 , u -0.2 I (5) ries (1938) log yi = -Azi 2 j '- ^ -0.2 I Davies (1962) log yi = -Azi 2 + fa "0.30 I (6) Robinson and Stokes (1959) log yi = . ^ ;, + bl (7) yi = activity coefficient of ion i I = ionic strength Zi = charge of ion i A = 1.82 x 10 6 (eT)~ 3 / 2 (where e = dielectric constant) a = radius of hydrated ion „ 50.2916 x 10 8 d 2 ,.. 3 „ n A , , , . . ... B = ( 10 gH 2 0) (d = density of water) / J- (eT) 3 ' 2 moles 2 b = empirical constant liquid phase never exceeded 0.05; therefore, the model was not suspect. The electrical conductance of this system was still gradually increasing when the experiment was terminated, suggesting non- equilibrium conditions. The LTE extracts of OBG clustered between the gypsum-anhydrite boundaries and compared favorably to the pond sample taken at the OBG site. The sample of the acidic effluent (AMD) flowing from the OBG refuse pile appeared to be supersaturated with respect to both CaSO. phases; but again the high ionic strength (0.3866) rendered the activities calculated for Ca 2 and SO 2 ~ suspect due to the limitations of the Extended Debye-Huckel equation. The short-term ASTV- A extract of the solid waste collected from the OBG site appeared to be in equilibrium with anhydrite, which was detected in the solid sample by X-ray diffraction. The laboratory extracts of the slurry sample were also undersaturated with respect to gypsum; however, they were in equilibrium with anhydrite, as were both the field slurry samples. The -50- stagnant effluent (WSSL-inactive) sample had an ionic strength of 0.2313, making the calculated activities open to question. The LTE and ASTM-A extracts of the non-acidic materials, DMNS, DMOS, and KG, were all undersaturated with respect to CaSO^ (fig. 14). The LTE of DMNS was also initially undersaturated with respect to calcite (fig. 15), like the ASTM-A extract, hut later appeared to approach equilibrium. The LTE of DMOS was initially near equilibrium with calcite. Later it appeared to have equilibrated with aragonite, a metastable phase of calcite, some time after 75 days. The aragonite, which could have formed as a precipitate at the expense of the dissolution of calcite, eventually became the dominant CaCO _ phase controlling the aqueous solubility of Ca 2+ . Note that the ASTM-A extract was in equilibrium with calcite, possibly as a result of the comparatively short solubilization interval. The LTE of the Kilngas waste sample appeared to be near equilibrium with calcite, then became supersaturated with respect to both CaC0_ phases. Neither carbonate or sulfate minerals were identified in the KG sample, and the solubility of Ca 2 did not seem to be controlled by either in long-term experiments. The mineral phases influencing the solubility of Ca' 1 in KG are unknown but may possibly be CaO forms in the high-temperature residue and/or calcium feldspars. Carbonate phases may exercise some influence in short-term experiments as suggested by the ASTM-A extract of KG. The possible sources of solution silicon and aluminum in the laboratory extracts and field leachates include the feldspars and clay minerals detected by X-ray diffraction (table 2). However, all the acidic LTE, ASTM-A, and field samples were undersaturated with respect to albite (Na{AlSi 3 8 >), anorthite (Ca{Al Si g }), adularia (KUlSi^}) , analcime (Na{AlSi 2 6 >), halloysite (Al^Si^O }{0H} ), illite ( •» K, Mg, alumino- silicate hydroxide), kaolinite (Al Si {OH} ), and montmorillonite (=Ca, Na, K, Mg, Fe, aluminosilicate hydroxide). The LTE and ASTM-A extracts of non-acidic materials, DMNS, DMOS, and KG, were also under- saturated with all the feldspar minerals considered by WATEQ2. The LTEs of the alkaline samples were initially supersaturated with respect to illite, montmorillonite, and kaolinite, then became either slightly supersaturated or undersaturated with respect to these clay minerals during the latter part of the equilibration period. Figure 16 shows the silicon dioxide equilibria. All the samples fell within the range of silicon solubilities expected from either quartz, cryptocrystalline SiO~, or amorphous SiO_. To simplify the illus- tration, only the final subsamples of the LTEs are shown. Since quartz is a dominant mineral in the wastes, it can be expected that quartz would exert a maior control on the solubility of H.SiO,. This relationship may have been partially obscured because glass containers were used in all of the extraction studies, shifting the silicon equilibria towards the less stable, amorphous SiO. boundaries. However, all the field leachate samples also fell along the amorphous SiO boundaries. -51- o CO a 2.0-t 2.2- 2.4- 2.6- 2.8 3.0 3.2- 3.4- 3.6- Supersaturated Undersaturated A140D ASTM-A 75 D A ASTM-A 140 D 10D A75D 24 H A 28 • 28 D • 10 D • 24 H A 24 H A10D I DMNS • DMOS A KG — i — 3.4 3.2 — i 1 1 — 3.0 2.8 2.6 2.4 2.2 pCa : + Figure 14 Calcium sulfate equilibria of the non-acidic laboratory extracts (ASTM-A and the long-term equilibrations) -52- o o a. 4.0n 4.5 5.0- 5.5- 6.0- 6.5 A 75D A 28 D A 140 D Supersaturated A 10D ■ DMNS • DMOS A KG MOD ^ ASTM-A 28 D #10D 75 D Undersaturated ASTM-A 3.4 3.2 3.0 2.8 2.6 2.4 2.2 pCa* Figure 15 Calcium carbonate equilibria of the non-acidic laboratory extracts (ASTM-A and the long-term equilibrations). -53- .'■'.Amorphous Si0 2 LTE AMD .......„,,....& ,v,^ Supersaturated :'\v.lte. : :::amd-. - . ASTM-A^^ • • ASTM-A Cryptocrystalline Si0 2 A pond T V—ft- ■ ••■ ■■ . / /^^ — , ■■•■ ,. — i — LTE Quartz Si0 2 AMD 9=3" I LTE ASTM-A LTE Undersaturated 1 ■ih -//- ASTM-A pH A WSNG ■ DMNS A WSOG O DMOS • OBG KG O WSSL l~~l Slurry (inactive) \/ Slurry discharge Figure 16 Silicon dioxide equilibria of the laboratory extracts (ASTM-A and the long-term equilibrations) of all seven wastes and the field leachate samples. Griffin et al. (1980) found that Al 3 activities in long-term extracts of coal solid wastes were dominated by metastable aluminum phases, (amorphous A1(0H) and A100H) rather than aluminum containing minerals present in the solid wastes. In this study, all the acidic systems were undersaturated with respect to these particular metastable phases. In general, it appeared that most of the laboratory extracts and field samples were not in equilbrium with any known metastable phases of aluminum. Two field leachate samples (WSOG acid mine drainage and WSSL- discharge) were near equilibrium with a metastable A10HS0 phase. The Al equilibria for these acidic systems remains unresolved. Iron equilibria are often shown by constructing stability relationships of iron species by either holding Fe 2 or Fe 3 concentrations constant and varying Eh and pH. Griffin et al. (1980) constructed iron oxide and sulfide stability fields where Fe 2 aqueous = 10 _6 M and total sulfur = 10 3 M. Solution iron concentrations in this study varied from 0.21 mg/L Fe to 5625 mg/L Fe; thus, the classical iron-stability diagram based on an arbitrarily chosen Fe 2 or Fe 3 concentration may have little meaning. -54- Figure 17 shows the Fe 3+ equilibria as a function of pH. While the ratio of Fe 3+ to Fe 2+ is influenced by Eh, the equilibrium Fe 3 activity of any iron mineral is independent of Eh. Solubility relationships may be derived by comparing the Fe 3 activities calculated by WATEQ2. In general, the laboratory extracts and field leachates tended to plot near the maghemite boundary, a metastable phase of hematite (fig. 17). Consequently, all the aqueous samples were undersaturated with respect to amorphous ferric hydroxide. This disagrees with the work of Nordstrom et al . (1979) for field samples of acid mine waters in the West Shasta Mining District in northern California. Their work, also based on WATEQ2 modeling, indicated that iron equilibrium was influenced by Fe(OH) . Griffin et al. (1980) concluded that amorphous Fe(OH) was probably a control on iron concentrations in some extracts; whereas maghemite solubility controls were operative in other systems. Reaction kinetics may partially explain some of the discrepancies between the studies. Note that while the LTE of OBG was initially in equilibrium with maghemite, the extract equilibrium shifted to magnetite after 77 days. A magnetite boundary is shown for pFe 2 = 7.35, the pFe activity after 147 days, since the equilibria of magnetite is Eh dependent. The LTE of WSNG was near equilibrium with Fe(OH) after 28 days of equilibration, then appeared to have equilibrated with maghemite. The kinetic hypothesis was not confirmed, due to the fact that all the short-term ASTM-A extracts were near equilibrium with maghemite within only 48 hours; whereas the LTEs took months to reach the same phase boundary. The Fe(OH) -Fe question remains unresolved. The field sample from the WSOG site was near equilibrium with maghemite; but both the ASTM-A and LTE extract of the solid sample taken from the WSOG site appeared to be approaching equilibrium with the more stable goethite (FeOOH) phase. Natrojarosite, NaFe(S0 ) (OH) , was identified in OBG by X-ray diffraction (table 2). The LTE of OBG may have also equilibrated with natrojarosite (S.I. = -0.221 ) 1 ; while most of the other systems, including the field samples from the OBG area, were supersaturated with respect to natrojarosite. Nordstrom et al.(1979) also found that field samples of acid mine drainage were supersaturated with respect to natrojarosite. Coquimbite (Fe 2 (S0 ) »9H 0) was detected by Mossbauer in WSNG and WSOG, but there are no thermodynamic data available on this iron sulfate phase, which precludes solubiity analysis . l S.I. or saturation indice is defined as the log of the ratio of the experimentally derived activity product to the equilibrium activity product . -55- © WSNG ▲ WSOG • OBG WSSL 1 I Slurry (inactive) ^A Slurry discharge ■upersaturated Figure 17 Fe 3+ solubility equilibria as a function of pH for the acidic laboratory extracts (ASTM-A and the long-term equilibrations) and the field leachates. -56- ACUTE STATIC BIOASSAYS OF EXTRACTS GENERATED FROM SAMPLES OF COAL SOLID WASTES T. M. Skelly and J. J. Suloway The major objective of the biological investigations was to determine whether extracts generated from samples of spoil banks (DMNS and DMOS), coal cleaning wastes (WSNG, WSOG, WSSL, and OBG), and a coal conversion waste (KG) — were acutely toxic to aquatic organisms. Other objectives included determining (1) the relative toxicities of the extracts that were shown to be acutely toxic, (2) the relative sensitivities of the four test organisms to the toxic extracts, and (3) any significant correlations between the acute toxicity of a solution and its physico- chemical composition. Methods and Materials The LTE extracts, upon attaining steady-state concentrations, were decanted from the carboys and filtered through 0.45p. pore-size Millipore membranes. Samples were taken for physicochemical characterization and the remaining supernatant solution was used as the undiluted extract in the bioassays. Each extract sample was assayed for acute toxicity with four species of aquatic organisms: green sunfish ( Lepomis cyanellus ), fathead minnow ( Pimephales promelas ), a crustacean ( Daphnia magna ), and a snail ( Physa anatina ). The acute toxicity tests were divided into two phases: a screening procedure and the LC-50 determination (the concentration that is lethal for 50% of the organisms). During the screening procedure, the test organisms were exposed to the extract diluted with soft recon- stituted water (U.S. EPA, 1975). Graphic probit analysis (Litchfield and Wilcoxon, 1949) was used to determine the LC-50 values. In acute bioassay tests, 1- to 2-day old Daphnia , 1- to 2-day old minnow fry, juvenile snails (shell length = 5.0 ± 0.5 mm), and yoke-sac green sunfish fry were exposed to the extracts in an environmental chamber maintained at a constant temperature (20 ± 1 C) for a constant photoperiod (16L-8D). For each bioassay, 10 test organisms were placed in a 250-mL beaker containing 200 mL "full-strength" extract, diluted extract, or soft reconstituted water (control). Bioassays were conducted in quadruplicate for each of the four organisms; pH, temper- ature, and dissolved oxygen concentrations were monitored. Mortality data were collected at 24, 48, 72, and 96 hours, except for the Daphnia tests, which were terminated after 48 hours. After 96 hours, snails that appeared to be dead were placed in reconstituted water and checked within 24 hours to confirm mortality assessments. The test organisms were not fed and test solutions were not aerated during the bioassays. Four control solutions, each containing 200 mL of soft reconstituted water and 10 test organisms, were tested with each set of bioassays. -57- Regression analysis was used to determine differences between the mortality data sets by testing the hypothesis that the slopes of the lines describing each of the two data sets were equal and that the y- intercepts were also equal. Mortality data were regressed on physico- chemical data using the General Linear Model procedure of the Statistical Analysis System. The IBM 4341 computer at the University of Illinois was used to perform the analyses. Results Acute toxicity of extracts. The screening procedures were conducted to determine if the undiluted extracts were acutely toxic. The mean initial pH of the control solutions (n = 108) was 7.91 ± 0.18 (range = 7.69 to 8.85), and the mean final pH was 7.72 ± 0.14 (range = 7.30 to 7.91). The mean initial dissolved oxygen was 7.94 ± 0.37 mg/L (range = 6.93 to 8.86 mg/L), and the mean final dissolved oxygen concentration was 8.56 ± 0.64 mg/L (range = 6.67 to 9.90 mg/L). The mean hardness, as determined by titration, was 29.6 ± 5.37 mg/L CaCO (range = 15 to 35 mg/L). The mean survival of the test organisms exposed to the control solutions was 95.0, 97.0, 98.2 and 98.1 percent for Daphnia , Physa , Lepomis and Pimephales , respectively. To examine the relationship between pH and mortality, static bioassays were conducted with reconstituted water ranging in pH from 3.0 to 9.0. The results of the screening procedures and the pH experiments are depicted in figure 18. The results of the pH experiments were similar to those of the screening procedures. The undiluted extracts and recon- stituted water produced total mortality when the pH was below 6.0, 5.6, 5.4, and 3.5 for Pimephales , Daphnia , Lepomis and Physa . Undiluted extracts generated from the spoil bank (DMNS and DMOS) and Kilngas (KG) samples were not acutely toxic (mortality <5%) to Daphnia , Pimephales , Lepomis and Physa (fig. 18). Undiluted extracts produced by the LTE procedure from the coal cleaning waste samples (WSOG, WSNG, WSSL, and OBG) were acutely toxic to all four test organisms. These solutions were very acidic (pH <3.5). Some extracts produced total mortality within minutes after bioassays were initiated. All four test organisms were also sensitive to acidic reconstituted water. The order of sensi- tivity to low pH was Pimephales > Daphnia > Lepomis> Physa . The LTE solutions from the refuse pile samples (WSNG, WSOG and OBG) and the slurry sample (WSSL) were evaluated for their relative acute toxicities in the LC-50 determinations. In addition, results from these bioassays were used to estimate both the amount of dilution necessary to eliminate mortality during an acute bioassay and the relative sensi- tivities of the test organisms to the LTE solutions. The LC-50 value is reported as the volume of solution/ 100 mL necessary to cause 50 percent mortality. The LC-50 values, their 95-percent confidence intervals, and dilutions necessary to eliminate mortality are listed in table 18. The pH values -58- 100 60- 40- 20- WSOG WSNG OBG WSSL Daphnia magna A reconstituted water t r DMNS DMOS KG 1 1 W w WSOGXWSNG OBG WSSL Lepomis cyanellus DMNS -^ ' DMOS KG -i 1 1 r t 1 •* |» 1 100. WSOG\ WSNG \OBG 80- WSSL 60- 40- 20- • • • Physa anatina DMNS D M OS KG -——• • A- WSOG\ WSNG OBG WSSL Pimephales promelas 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 Figure 18 pH ISGS 1983 Percent mortality of Daphnia magna, Lepomis cyanellus, Physa anatina and Pimephales promelas resulting from exposure to undiluted extracts generated from seven coal solid wastes and buffered solutions of reconstituted water. listed are for the undiluted extracts before dilution with soft recon- stituted water. There is an inverse relationship between toxicity and the LC-50 value: an increase in the toxicity corresponds to a decrease in the LC-50 value. Of the four extracts shown to be acutely toxic by the screening procedures, WSOG produced the most toxic solution and OBG produced the least toxic solution. The extracts produced from WSNG, WSOG, and WSSL were very toxic to all four test organisms: LC-50 values were less than 3.0 mL/100 mL. Generally, acidic extracts or leachates (pH <5.0) were acutely toxic. With increasing acidity, there was an increase in the toxicity, a decrease in the LC-50 value, and an increase in the amount of dilution necessary to eliminate mortality. The most toxic extract, WSOG, re- quired a dilution of 1:1000 in order for Daphnia to survive a 48-hour bioassay. All four acutely toxic extracts required a relatively large amount of dilution (>1:20) to eliminate mortality, except for OBG when tested with Physa . Physa was the most tolerant test organism with the highest relative LC-50 values and the lowest dilutions to eliminate mortality. Daphnia and Pimephales were the most sensititve to the toxic effects of the extracts (fig. 19). As the concentration of the extract increased, there was an increase in mortality for all species. The -59- 100n 8 10 Fimirp 1Q Leachate (volume leachate /total volume) (%) Relationships between the OBG extract and mortality for Pimephales promelas, Daphnia magna, Lepomis cyanellus, and Physa anatina. mortality curve for Physa anathina is shifted to the right side of the figure, demonstrating the superior tolerance of snails to the solutions. Relative toxicities of the four extracts were significantly different (P <0.01) from one another in all cases for all four organisms (table 19) In 20 of 24 comparisons, the difference between the toxicity data sets were highly significantly different (P <0.001 or 0.0001). Generally, all four organisms were significantly different in their sensitivity to the extracts (table 20); however, Pimephales and Daphnia were similar in their tolerance to WSNG and OBG extracts, while Lepomis and Daphnia were similar in their tolerance to OBG. These results are supported by the 95-percent confidence intervals for the LC-50 values (table 18); for example, the 95-percent confidence intervals for the LC- 50 of OBG to Daphnia (7.3 to 7.5) overlaps that for Lepomis (7.2 to 12.4). Physa , the most tolerant test organisms, was very significantly different in its tolerance in 7 of 12 comparisons (table 20). Relative tolerances of test organisms. Daphnia and Pimephales were the species most sensitive to the toxic effects of the extracts. Lepomis cyanellus was less tolerant than Physa ; however, it was more tolerant -60- Table 18. The LC-50 values, their 95 percent confidence intervals, and the amount of dilution necessary to eliminate mortality in static bioassays of extracts (WSOG, WSSL, WSNG and OBG) generated from the solid wastes. Organism PH LC-50 Dilution for mL/100mL zero percent mortality WSOG Daphnia magna 1.67 0.36 ± .02 Pimephales promelas 1.71 0.42 ± .004 Lepomis cyanellus 1.68 0.60 ± .02 Physa anatina 1.65 0.98 ± WSSL .05 Daphnia magna 2.09 0.77 ± .02 Pimephales promelas 2.21 0.84 ± .02 Lepomis cyanellus 2.06 0.96 ± .05 Physa anatina 2.05 1.65 ± .05 Daphnia magna Pimephales promelas Lepomis cyanellus Physa anatina WSNG 2.44 1.70 ± .02 2.30 1.69 + .02 2.52 1.95 ± .08 2.51 2.90 ± .09 1:1000 1:263 1:200 1:154 1:143 1:143 1:118 1:91 1:57 1:65 1:57 1:44 OBG Daphnia magna Pimephales promelas Lepomis cyanellus Physa anatina 3.34 7.4 ± 0.1 3.33 6.8 ± 0.3 3.36 9.5 ± 2.9 3.37 14.2 ± 0.3 >1:50 1:20 1:33 1:9 than Daphnia or Pimephales promelas* Daphnia has been shown to be more sensitive to cations than fish (Anderson, 1948). There was a distinct threshold where the buffering capacity of the re- constituted water used for dilution was exhausted. Characteristically, above that threshold, the final pH of dilutions containing Lepomis cyanellus , Pimephales promelas , or Daphnia would increase 0.4 to 0.9 units above the initial pH. Below that point, pHs would remain constant or drop up to 0.5 units. Dilutions containing Physa anatina showed an increased level of buffering, which for WSOG, existed throughout the pH range tested. Solutions containing Physa increased in pH by 0.3 to 1.6 units from the initial values. It was theorized that the shell of the snail, composed of calcium carbonate, was being dissolved in these acidic dilutions, thereby adding to the buffering capacity of the test solutions and raising the pH. An experiment was performed with the equilibrated WSOG -61- Table 19. Differences in the acute toxicities of extracts generated from four coal solid wastes (WSNG, WSOG, OBG and WSSL) as determined by an analysis of covariance. The levels of significant difference are listed and were obtained by analyzing mortality data of 48- and 96- hour bioassays using Daphnia magna , Pimephales promelas , Lepomis cyanellus and Phy. sa anatina. WSNG WSNG WSNG WSOG WSOG OBG vs vs vs vs vs vs WSOG WSSL OBG WSSL OBG WSSL Pimephales 0.0001 0.0001 0.000! 0.01 0.01 0.001 Daphnia 0.0001 0.0001 0.001 0.0001 0.0001 0.0001 Physa 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 Lepomis 0.01 0.01 0.001 0.001 0.0001 0.0001 0.01 = significant difference- 0.001 = highly significant difference. 0.0001 = very highly significant difference- Table 20. Differences in sensitivities of Daphnia magna , Pimephales promelas , Lepomis cyanellus and Physa anatina to extracts generated from four coal solid wastes (WSNG, WSOG, OBG and WSSL) as determined by an analysis of covariance. The levels of significant difference are listed and were obtained by analyzing mortality data of 48- and 96- hour static bioassays. Pimephales Pimephales Pimephales Daphnia Daphnia Lepomis vs vs vs vs vs vs Daphnia Lepomis Physa Lepomis Physa Physa WSNG N.S. 0.001 0.0001 0.05 0.001 0.01 WSOG 0.001 0.01 0.0001 0.0001 0.0001 0.0001 OBG N.S. 0.05 0.0001 N.S. 0.001 0.001 WSSL 0.01 0.01 0.001 0.0001 0.0001 0.0001 N.S. = difference not statistically significant at 0.05 level, 0.05 = significant difference. 0.01 = significant difference. 0.001 = highly significant difference. 0.0001 = very highly significant difference. -62- extract using snail shells (with the animals removed) to determine if this could have been the cause of the observed increases in pH values. In all dilutions within the range of concentration toxic to Physa , the pH rose by 0.3 to 0.7 pH units in 96 hours due to shells alone. The pH of control solutions dropped by 0.1 to 0.2 units. These dilutions were all below the buffering cutoff point for WSOG. The snail sbell does contribute to the buffering capacity of the dilu- tion water. This increased buffering capacity is partially responsible for the relatively increased tolerance of the snails to the acute toxicity of tbe extracts generated from the coal solid waste samples. Chemical characterization of the diluted extracts. After the acute static bioassays were completed, subsamples of all test solutions were filtered, preserved with nitric acid, and stored for analysis. To determine which constituents were concentrated enough to be potentially hazardous, we analyzed samples that produced partial mortality as well samples containing the lowest concentration at which total mortality occurred and the highest concentration at which no mortality occurred. The mortality data were regressed on the physico chemical data using a multiple regression procedure (Nie et al., 1975). The results of the multiple and liuear regressions are included in tables C-l through C-15 in Appendix C. Ranges of concentration of each chemical constituent in the test solutions and the recommended water- quality level for each chemical constituent are also given. Subsamples of the test solutions of WSNG during the Daphnia bioassays were not analyzed. Table 21 summarizes the results listed in Appendix C. The chemical constituents listed were found to be significantly correlated to mortality (P <0.05) and present in concentrations exceeding recom- mended water quality levels (U.S. EPA, 1976). Calcium was not included in table 21 since it is not toxic at the concentrations measured in the test solutions. For all extracts and all test organisms, the pH of the test solutions is significantly correlated to mortality and exceeds the recommended water- quality level (pH <6.5). Only eight physicochemical parameters other than pH are listed in table 21. Aluminum, Fe, Mn, and Ni appeared to be significantly correlative to mortality. Zinc was highly correlated with mortality in test solutions of OBG. More physicochemical parameters were correlated to mortality in the bioassys using Lepomis cyanellus and Physa anatina than in those using Daphnia magna and Pimephales promelas . This difference was probably due to the fact that Lepomis and Physa exhibited higher levels of tolerance, thus the solutions were more concentrated in their bioassays. Discussion Nontoxic leachates. Several authors (Beyer and Hutnik, 1969; Barnhisel and Rotromel, 1974; Hood et al . , 1979; Klimstra and Jewell, 1973; Mason, -63- Table 21. Chemical constituents found to be significantly correlated to mortality and to exceed recommended water quality levels in test solutions prepared from extracts (OBG, WSNG, WSOG and WSSL) generated from coal solid wastes. Test Organism Daphnia P: Lmephales ; Lepomis Physa Extract magna P ] come las cyanellus anatina OBG pH pH pH PH Mn Mn Al Al Ni Ni Mn Mn Zn Zn Ni SOi* Ni Zn WSNG PH pH PH PH Mn Al Al Zn Fe Mn Ni Fe Mn Ni WSOG pH pH PH PH Fe Fe Pb Al Fe WSSL pH pH pH pH Cd Fe Al Ni Fe Mn Ni 1978; and many others) have cited spoil banks as sources of problem leachates. The laboratory extracts generated from the spoil bank samples (DMNS and DMOS) in this study were essentiallly neutral in pH, contained relatively low concentrations of accessory elements, and were not acutely toxic (mortality <5%). The steady state extract generated from the Kilngas solid waste (KG) was alkaline (pH 9.0), contained relatively low concentrations of accessory elements, and was not acutely toxic. Similar results were obtained in a study testing laboratory extracts generated from Lurgi gasification ashes (Suloway et al., 1981). Toxic leachates. The extracts generated from the coal cleaning refuse samples (OBG, WSNG, and WSOG) and the slurry sample (WSSL) were acidic (pH <3.5), contained relatively high concentrations of some accessory elements, and were acutely toxic to all four test organisms. Because of the complexity of the chemical composition of the systems and the unde- termined synergistic and antagonistic effects of the chemical constitu- ents composing the solutions, it was not possible to identify the chemical constituents directly responsible for the observed mortality; however, certain tentative conclusions can be drawn. -64- The mortality caused by the OBG extract can be attributed, in part, to the low pH of the test solutions, since the pH of some test solutions was below the tolerance levels established during the pH experiments with reconstituted water. The pHs of the most concentrated test solutions were approximately 5.0. Reconstituted water with a pH of 5.0 will cause total mortality in Daphnia, Pimephales, and Lepomis bioassays and 20 percent mortality in Physa bioassays. Also the concentration of Zn in some OBG test solutions exceeded the LC-50 values established for Daphnia by Biesinger and Christensen (1972), for Pimephales by Benoit and Holcombe (1978), and for Physa by Patrick (1968). Relatively high concentrations of aluminum were measured in the OBG test solutions; however, LC-50 values for aluminum have not been established for the test organisms. As was the case in the OBG test solutions, the pH values of the most concentrated test solutions for the WSNG waste exceeded the tolerance levels of all four test organisms. Also the concentration of Zn measured in the WSNG bioassays using Pimephales exceeded the LC-50 value for Zn established by Benoit and Holcombe (1978). In addition to toxic pH levels, the WSOG test solutions contained concentrations of Fe that exceed the LC-50 value for Daphnia (Biesinger and Christensen, 1972). The WSSL test solutions were characterized by toxic pH levels. CHRONIC BIOASSAYS OF LEACHATES GENERATED FROM COAL CLEANING REFUSE SAMPLES T. M. Skelly and J. J. Suloway The major objectives of these biological investigations were to determine (1) what concentrations of leachates generated from two coal cleaning refuse samples (OBG and WSNG) caused sublethal, toxic effects in chronic bioassays using Pimephales promelas , Lepomis cyanellus , Daphnia magna and Physa anatina ; (2) the relative sensitivities of the four test organisms to the leachates; (3) the relative sensitivities of the parameters measured (such as growth, reproduction, mortality) during the bioassays, and (4) any significant correlations between the chronic toxicity of a leachate and its physicochemical composition. Methods and Materials The composition of the leachates generated from samples of OBG and WSNG by the large volume generation (LVG) procedure was continuously monitored. This particular procedure was used to generate the leachate because of an anticipated need for large volumes of solution for the chronic bioassays. After about 1 year, the leachates were first assayed for acute toxicity with the four test organisms; these toxicity tests were divided into phases: (1) the screening procedure, and (2) the LC-50 -65- determination. Methodologies for these acute tests were identical to those in the discussion of acute static bioassays. The effects of chronic exposure to sublethal concentrations of leachates from OBG and WSNG were determined using a continuous flow mini-diluter exposure system modified from a design proposed by Benoit et al. (1981). The system delivered 12.5 mL/min of five concentrations and a control solution to each of four replicate chambers. During the bio- assays using fish and snails, the chambers contained 700 mL of test solution; while during the bioassays using Daphnia , the test chambers contained approximately 500 mL of test solution. The dilution water, which was a blend of dechlorinated tap and deionized water, was passed through an ultraviolet filter before use and heated to the test temperature at each diluter. Water temperature for all fish and snail bioassays was maintained at 25.0 C; Daphnia experiments were conducted at 22.5°C. All tests were held under a constant (16L-8D) photoperiod. The four species of test organisms were exposed to dilutions of the OBG leachate at concentrations of 10, 5, 2.5, 1.25, 0.63 and 0.0 (control) percent (by volume) of the undiluted OBG leachate, and 0.1, 0.05, 0.025, 0.0125, 0.00625 and 0.0 percent of the WSNG leachate. Fathead minnows were tested for 32 days (4 days as embryos and 28 days post-hatch). To initiate these embryo-larval chronic bioassays, 30 eggs were placed in screened egg cups in each chamber. The egg cups were mechanically, slowly rocked to facilitate oxygen transfer and cleaning of the fish eggs. When hatching was completed, the total number of hatched individuals was recorded, and the number of larval fish per chamber was reduced to 15. Fish were fed live, brine shrimp nauplii ( Artemia salina ) three times per day. Dead fish and debris were siphoned from each exposure chamber every morning before the first feeding. When the test was terminated, fish were killed with ice water and measured for total length and weight. Green sunfish were tested for 29 days (1 day as embryos and 28 days post-hatch) with WSNG. The green sunfish bioassay with OBG was terminated after 14 days because of insufficient leachate. Test procedures were the same as those used in the fathead minnows bioassays. Daphnia magna experiments for both OBG and WSNG were conducted for 28 days; 5 individual Daphnia were maintained separately within stainless steel cages in each exposure chamber. Tests were initiated with 2- to 26-hour-old Daphnia. Daphnia were fed a blended solution of trout chow and baker's yeast 3 times per day. When the Daphnia grew large and reproduced ( Daphnia are parthenogenic), the number of progeny were counted and removed. Total length of each adult was measured at test termination. The snail experiments with OBG and WSNG were conducted for 28 days. All snails were measured for shell length and weight prior to and after the exposure period. At test initiation, 15 snails (shell length = 5.0 ± -66- 0.5 mm) were placed into each of the 24 exposure chambers. Several behavioral parameters were monitored 3 times per day, 3 days per week. The number of snails out of the water, snails taking on air, and mating behavior were noted. Egg masses deposited by the snails were counted .and removed each day. Occasionally, an egg mass went undetected, in which case the juvenile snails were removed. Snails were fed a gelatinous plug of wheatgerm, dry milk, and cerophyll (Standon, 1951) once per day, every other day. During each chronic bioassay for all test species, mortality, water temperature, and the pH of the undiluted leachate were monitored daily. Diluter flow rates were measured twice daily and adjusted as necessary. Twice per week, water samples were taken from one-half of the exposure chambers. By sampling alternate replicates, A and C, then B and D, all exposure chambers were sampled once a week. Approximately 100 mL of test solution was siphoned for each sample. These samples were analyzed initially for pH and dissolved oxygen, then preserved for inorganic constituent analysis after filtering. Since an iron precipitate formed when the WSNG leachate was diluted, the capillary tubes in the dilution apparatus were cleaned every 4 hours. Also, the toxicant dilution water cell and mixing cell were replaced daily to help maintain proper flow rates. Other cells in the dilution apparatus were replaced as required. During replacement, the dilutor was shut down for approximately 45 minutes. Results Acute toxicity of the leachates. Acute toxicity tests were conducted in the second year to determine the range of concentrations of leacbates that should be used during the chronic toxicity tests. First, screening procedures were conducted to determine if the undiluted leachates were acutely toxic (table 22). Undiluted OBG leachate was not acutely toxic to any species of test organism; the pH range of undiluted OBG during the second year (pH 5.70 to 6.04) was much higher than that measured in the leachate generated during the first year of the study (pH 3.33 to 3.37). Undiluted WSNG generated during the second year of the study was acutely toxic to the three species of test organisms tested. The relative sensitivity of the test organisms was the same as that observed in the first year of the study; Pimephales and Daphnia were the most sensitive and Physa was the most tolerant species. The pH range of undiluted WSNG was higher in the second year (pH 3.49 to 3.98) than that measured in the first year (pH 2.30 to 2.51); however, the LC-50 values calculated for WSNG using data collected during Daphnia and Pimephales bioassays were considerably lower in the second year than those calculated in the first year. In contrast, LC-50 values calculated for WSNG using data collected during Physa bioassays in the first and second years were similar, despite the fact that the pH of the undiluted -67- Table 22. Results of the acute bioassays of WSNG and OBG conducted by the LTE and LVG procedures. Procedure LTE LVG LC-50 LC-50 Organism PH mL/100 mL PH mL/100 mL OBG Pimephales 3.33 6.8 + 0.3 5.74 >100 1 Daphnia 3.34 7.4 ± 0.1 5.62 >100 Lepomis 3.36 9.4 ± 2.9 6.04 >100 Physa 3.37 14.2 ± 0.3 WSNG 5.70 >100 Pimephales 2.30 1.69 ± 0.02 3.56 .240 ± 0.010 Daphnia 2.44 1.70 ± 0.02 3.49 .335 ± 0.018 Lepomis 2.52 1.95 ± 0.08 3.51 Physa 2.51 2.90 ± 0.09 3.98 3 .20 ± 0.20 1 undiluted OBG was not acutely toxic to any of the species of test organisms during acute bioassays. 2 green sunflsh larvae were not available for acute bioassays . leachate was considerably higher in the second year (pH 2.51 vs. pH 3.98). As was the case in the first year of the study, WSNG was a more toxic leachate than OBG. Since 0.1 percent was the highest conentration of WSNG at which no toxic effects were observed during the acute bioassays, this concentration was selected to be used as the highest concentration of WSNG for the chronic bioassays. Since even undiluted OBG was not acutely toxic, a relatively high concentration of OBG (10%) was selected as the highest concentration of OBG for the chronic bioassay. This concentration was selected primarily because a limited amount of un- diluted OBG was generated during the second year. Chronic toxicity of the leachates. The results of the chronic bioassays of WSNG using Pimephales promelas are summarized in table 23. One-way Analysis of Variance (ANOVA) was used to analyze the significance of differences in length and weight. Chi-square analysis was used to analyze the significance of differences in hatchability and survival. Growth as measured by length (F5 3Qg = 0.0003, P > 0.10) and weight (F 5 308 = 0.0039, P > 0.10) was not significantly affected by the dilutions of the WSNG leachate. The mean lengths ranged from a high of 21.6 mm in both the control and the 0.05-percent concentration, to 20.7 mm in the highest concentration, 0.1 percent. A similar trend was observed in the mean weights, with the lowest weight measured in the -68- Table 23. Lengths, weights, mortality, and hatching percentages for fathead minnows exposed to five concentrations of WSNG leachate and a control. Data were obtained during 32-day, embryo-larval bioassays conducted in flow-through mini-diluters. Concentration of Total length Weight (g) Mortality Hatchability leachate (%) N 1 X ± SD 2 (mm) X ± SD (%) (%) 0.10 54 20.7 ± 1.59 0.073 ± 0.017 8.3 84.0 0.05 47 21.6 ± 1.78 0.087 ± 0.023 8.3 91.7 0.025 57 21.2 ± 1.85 0.085 ± 0.022 4.8 98.3 0.0125 51 21.1 ± 1.86 0.085 ± 0.022 5.0 89.3 0.00625 49 21.4 ± 1.60 0.083 ± 0.018 11.5 95.7 0.00 (control) 56 21.6 ± 1.44 0.085 ± 0.018 5.0 99.2 number of data -mean ± one sample standard deviation Table 24. Lengths, weights, mortality, and hatching percentages for green sunfish exposed to five concentrations of WSNG leachate and a control. Data were obtained during 29-day, embryo-larval bioassays conducted in flow-through mini-diluters. Concentration of leachate (%) Total length X ± SD 2 (mm) Weight (g) X ± SD Mortality (%) Hatchability (%) 0.10 24 16.9 ± 1.45 0.060 ± 0.016 45.0 56.0 0.05 43 16.5 ± 1.44 0.059 ± 0.012 21.7 74.5 0.025 40 16.2 ± 1.16 0.054 + 0.012 26.7 78.1 0.0125 40 15.8 ± 1.80 0.049 ± 0.014 29.5 63.1 0.00625 44 16.2 ± 0.97 0.053 ± 0.010 21.7 76.0 0.00 (contrc ■1) 43 16.4 ± 0.79 0.056 ± 0.009 23.3 80.7 "-number of data -mean ± one sample standard deviation strongest concentration and the highest mean weight measured in the chambers with a 0.05-percent concentration of WSNG. There was no significant difference in the amount of mortality of fathead minnow larvae exposed to various concentrations of the WSNG leachate. The highest mortality (11.5%) occurred in the lowest concentration of WSNG. The lowest mortality (4.8%) was observed in the third highest concentration, 0.025 percent. The only parameter measured during the chronic bioassays of WSNG using fathead minnows in which a significant difference was observed was hatchability. The hatchability in the control (99.2%) was significantly higher than that measured in test solutions of 0.10 percent leachate (hatchability = 84.0%, P < 0.001), 0.05 percent leachate (hatchability = 91.7%, P < 0.025), and 0.015 percent leachate (hatchability = 89.3%, P < 0.005). -69- Table 25. Lengths, mortality, and reproductive success of Daphnia magna exposed to five concentrations of WSNG leachate and a control. Data were obtained during 28-day, chronic bioassays conducted in flow-through mini-diluters. Concentration of Progeny/Adult First day of Mortality Total length leachate (%) X ± SD 1 reproduction (%) X ± SD (mm) 0.10 0.0 ± 0.0 75.0 0.05 8.5 ± 4.4 18 75.0 3.5 ± 0.22 0.025 10.5 ± 7.1 12 40.0 3.7 ± 0.28 0.0125 16.0 ± 12.0 11 50.0 3.7 ± 0.37 0.00625 20.0 ± 13.9 13 30.0 3.7 ± 0.31 0.00 (control) 24.2 ± 12.5 11 15.0 4.0 ± 0.33 ^mean ± standard deviation The results of the chronic bioassyas of WSNG using Lepomis cyanellus are summarized in table 24. The same parameters were measured and the same statistical analyses were used as those reported for fathead minnows. Again the growth parameters, mean length (F^ 229 = 0*004 , P > 0.10) and mean weight (F^ 229 = 0.0046, P > 0.10) were not significantly different between treatments. In fact, the largest fish were those found in the chamber dosed with the highest concentration of WSNG. There was a significant increase in the amount of mortality in the highest concen- tration over that found in the control ( x 2 ■ 9.58, P < 0.005). The hatchability in the control (90.7%) was significantly higher than that measured in the test solutions of 0.10-percent leachate (hatchability = 56.0% P < 0.005) and 0.0125-percent leachate (hatchability = 63.1%, P < 0.005). The results of the chronic bioassays of WSNG using Daphnia magna are summarized in table 25. The leachate had several dramatically detri- mental effects on the Daphnia . The 75-percent mortality observed in chambers exposed to 0.10- and 0.05-percent leachate was significantly higher than the 15-percent mortality observed in the controls, as determined by chi-square contingency analysis. The mortality in the other treatments (30% to 50%) was also higher than that observed in the controls, but the difference was not statistically significant because of the small sample sizes. Reproduction, as measured by the mean number of progeny per adult, was significantly reduced in the three highest concentrations (0.10%, 0.05%, and 0.025% WSNG) from that observed in the control. The reproduction at doses of 0.0125 and 0.00625 percent WSNG was also reduced, but the difference was not statistically signifi- cant. The first day of reproduction also appeared to be delayed at the higher concentrations. The mean total length of Daphnia adults was not significantly reduced by exposure to different concentrations of WSNG (F4 47 = 0.0014, P > 0.10). -70- Table 26. Shell lengths, weights, mortality, and egg mass production of Physa anatina exposed to five concentrations of WSNG leachate and a control. Data were obtained during 28-day, flow-through bioassays. Concentration of S_hell length Weight (g) Mortality Egg mass production leachate (%) N 1 X ± SD 2 (mm) X ± SD (%) X ± SD 0.10 33 5.4 ± 0.40 0.018 ± .003 30.0 89.3 ± 27.0 0.05 49 5.4 ± 0.50 0.020 ± .005 13.3 133.3 ± 12.7 0.025 49 5.7 ± 0.73 0.022 ± .007 15.0 106.3 ± 17.3 0.0125 43 5.8 ± 0.48 0.023 ± .005 18.3 108.5 ± 24.2 0.00625 45 5.9 ± 0.80 0.026 ± .009 18.3 99.0 ± 14.1 0.00 (control) 46 5.9 ± 0.56 0.025 ± .006 8.3 119.3 ± 22.1 dumber of data 2 mean ± one sample standard deviation The results of the chronic bioassays of WSNG using Physa anatina are summarized in table 26. One-way ANOVA was used to analyze the signifi- cance of differences in length and weight between treatments. Although mean shell length increased with decreasing concentration of WSNG, and although the mean shell length of individuals exposed to the control was greater than those exposed to the higher concentrations of WSNG, the differences were not statistically significant (Fc 059 = 2.113, P > 0.05). However, the differences in mean weight were significant (Fc 229 = 8.89, P < 0.001). The mean weight data was further analyzed using'a Student-Newman-Keuls test. The results of this analysis demonstrated that the mean weight of the controls was significantly greater than that of those individuals exposed to 0.1-percent WSNG (P < 0.001), 0.05- percent WSNG (P < 0.001), and 0.025-percent WSNG (P < 0.05). The mortality observed in the different treatments was also significantly different ( x 2 = 13.07, df = 5, P < 0.025). Specifically, the mortality in the controls was significantly less than that observed in the 0.1- percent WSNG treatment ( x 2 = 8.80, df = 1 , P < 0.005). Reproduction as measured by egg mass production, was not significantly affected by exposure to WSNG (F 5 2 23 = 2.237, P > 0.05). Few significant differences were observed in the chronic bioassays of OBG. The results of the chronic bioassays of OBG using fathead minnows are summarized in table 27. Growth, as measured by mean length and weight, was lowest in the controls. Mean total length of those individuals exposed to the highest concentration of OBG (10%) was signficantly greater than the controls (P < 0.001). No significant differences were seen in mean weights of the snails exposed to the various concentrations of OBG. Mortality was low (<10%) and hatch- ability was nearly complete (>93%) in the chronic bioassyas of OBG using Pimephales promelas . No significant differences in mortality or hatch- ability were observed between the various concentrations of OBG and the control. -71- Table 27. Lengths, weights, mortality, and hatching percentages for fathead minnows exposed to five concentrations of OBG leachate and a control. Data were obtained during 32-day, embryo-larval bioassays conducted in flow-through mini-diluters. Concentration of Total length Weight (g) Mortality Hatchability leachate (%) N 1 X ± SD 2 (mm) X ± SD (%) (%) 10.0 61 24.2 ± 1.63 0.133 ± .031 0.0 93.2 5.0 49 23.5 ± 1.59 0.131 ± .029 8.3 98.3 2.5 51 23.3 ± 1.98 0.013 ± .037 6.7 93.9 1.25 57 23.7 ± 1.95 0.140 ± .037 3.3 98.3 0.625 53 23.1 ± 2.28 0.127 ± .037 10.0 98.3 0.00 (contro D 49 23.1 ± 1.37 0.126 ± .025 1.7 97.3 1 number of data ^ean ± one sample standard deviation Table 28. Lengths, weights, mortality and hatching percentages for green sunfish exposed to five concentrations of OBG leachate and a control. Data were obtained during 14-day, embryo-larval bioassays conducted in flow-through mini-diluters. Concentration of Total length Weight (g) Mortality Hatchability leachate (%) N 1 X ± SD 2 (mm) X ± : SD (%) (%) 10.0 33 8.4 ± 0.71 -3 18.3 96.6 5.0 39 8.3 ± 0.87 -3 8.3 98.3 2.5 30 8.4 ± 0.85 -3 30.0 99.1 1.25 39 8.2 ± 0.84 -3 13.3 98.3 0.625 39 8.2 ± 0.72 -3 10.0 96.7 0.00 (control) 41 8.2 ± 0.97 -3 10.0 99.2 number of data meaa ± one sample standard deviation not measured The green sunfish chronic bioassys of OBG were terminated after 14 days because of insufficient leachate. Only total length, mortality, and hatchability were recorded. The results of these chronic bioassays are summarized in table 28. Mean length did not differ significantly between treatments (F^ 2 05 = 0*5276, P > 0.10). Mortality showed no discernible trend. The only significant difference for green sunfish was the higher mortality for individuals exposed to a leachate concen- tration of 2.5-percent OBG than for those in the control solution (x 2 = 6.17, df = 1, P < 0.025). Hatchability for eggs in all treatments was nearly complete (>96.5%) during the bioassays of OBG using green sunfish; no significant differences in hatchability among treatments were observed. -72- Table 29. Lengths, mortality and reproductive success of Daphnia magna exposed to five concentrations of OBG leachate and a control. Data were obtained during 28-day, chronic bioassays conducted in flow-through mini-diluters. Concentration of leachate (%) Progeny /Adult X ± SD First day of reproduction Mortality (%) Total length X ± SD (mm) 10.0 5.0 2.5 1.25 0.625 0.00 (control) 56.6 ± 35.2 10 35.0 4.0 ± 0.38 66.3 ± 19.3 10 15.0 4.3 ± 0.17 56.5 ± 21.0 12 35.0 4.1 ± 0.27 52.5 ± 22.8 12 30.0 4.0 ± 0.20 71.5 ± 15.0 12 35.0 4.2 ± 0.06 70.0 ± 47.8 11 35.0 4.0 ± 0.32 Vean ± one sample standard deviation The results of the chronic bioassays of OBG using Daphnia magna are listed in table 29. Reproduction, measured as number of progeny produced per adult, was greatest in the controls and the lowest concen- tration of OBG; however, due to the large variability in production of progeny, this difference was not significant (Fc tj = 0.8699, P > 0.10). Also, the first day of reproduction did not vary among treat- ments (between day 10 and day 12). Mortality was high (15% to 35%) in all treatments; there were no significant differences between mortality observed in the controls and in the other treatments (x = 3.305, df = 5, P > 0.10). Mean total length of adult Daphnia showed no discernible trend. The only significant difference in total length was between the control and those individuals exposed to the concentration of 5.0- percent leachate (P <0.05). The results of the Physa bioassays of OBG are summarized in table 30. Growth of Physa , as measured by mean shell length (Fc 308 = 0.00067, P > 0.10) and mean weight (F^ 308 = 0.0022, P > 0.10), was not significantly affected by the OBG leachate. Observed mortality ( x 2 = 1.03, df= 5, P > 0.10) and egg mass production (F5 18 = 0.9890, P > 0.10) did not differ significantly among the treatments and control. Discussion The chronic bioassays were designed to supplement acute toxicity data by determining the sublethal effects — changes in growth, reproduction, and mortality — associated with chronic exposure to dilute concentrations of two toxic leachates generated from the refuse wastes samples OBG and WSNG. The full strength OBG leachate generated by the LVG procedure was not acutely toxic to any of the organisms tested, unlike the OBG LTE solution. The differences in the effects of the OBG solutions (table 22) were attributable to differences in the solution generation pro- cedures, which produced dissimilar solutions. The particle size of the OBG waste that was used in the LTE experiment was <9.53 mm; it was -73- Table 30. Shell length; s, weights, mortality, and egg mass produc tion of Physa anatina expo: sed to five concentrations of OBG leachate and a control. Data i were obtained during 28-day, fl ow-thr ough bioassays. Egg mass Concentration Shell length Weight (g) Mortality production of leachate (%) N 1 X ± SD 2 (mm) X ± SD (%) X ± SD 10.0 52 5.0 ± 0.56 0.012 .003 8.3 54.5 ± 5.7 5.0 55 5.2 ± 0.54 0.013 .003 6.7 62.5 ± 12.1 2.5 52 5.2 ± 0.58 0.014 .004 6.7 61.0 ± 7.7 1.25 51 5.2 ± 0.65 0.013 .005 5.0 75.5 ± 31.0 0.625 49 5.3 ± 0.62 0.014 .005 15.0 66.0 ± 5.5 0.00 (control) 55 5.2 ± 0.70 0.013 .005 5.0 55.8 ± 16.4 number of data mean ± one sample standard deviation necessary to sieve the sample in order to mix the solution without damaging the stirring motors. In contrast, the OBG sample used in the LVG procedure was not sorted, but loaded into the Agro tank in the same physical state as it was in the field. The unsorted sample contained boulder-sized clasts mixed with other coarse-grained particles. The rate of pyrite oxidation in the samples and the subsequent formation of H SO would increase with increasing surface area. This may be the major factor accounting for the lower pH and consequent toxicity of the LTE extract of OBG relative to that of the leachate from the LVG procedure. Refuse pile heterogeneity is also a likely factor. Because there was a limited quantity of OBG leachate, it was assayed as a 10-percent solution for the highest concentration for chronic exposure to determine if any long-term effects might be associated with exposure to this nontoxic leachate in acute experiments. The organisms exposed to the OBG leachate were generally not detri- mentally affected by dosages of this nontoxic solution. In fact, Pimephales promelas , the fish exposed to the highest concentration of OBG were larger than the controls, suggesting that this leachate pro- vided a favorable growth medium. The chemical composition of the test solutions (tables 31 to 34 ) was quite innocuous, which accounts for its relative lack of toxic effects. The WSNG leachate generated by the LVG procedure was quite toxic throughout the course of this investigation. Although the WSNG leachate generated changed in chemical composition with time, its toxicity was still very great. Its low pH of 3.49 to 3.98 was partially responsible for the toxicity during the screening procedures and LC-50 determin- ations. The pH of 7.8 to 8.5 during the chronic tests was not near the lethal limits for organisms, and yet mortality ( Lepomis , Daphnia , Physa ), lowered hatchability ( Lepomis cyanellus , Pimephales promelas ), reduced reproduction ( Daphnia ), and reduced weight gain ( Physa ) were all associated with chronic exposure to the WSNG leachate. -74- Table 31 . Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of OBG during chronic bioassays for fathead minnow ( Pimephales promelas ) leter OBG Concentrations (%) 0.625 Paran 10.0 5.0 2.5 1.25 0.0 pH(ur tits) 7.757-7.850 7.' 990-8.053 8.: 180-8.257 8.1 D90-8.338 8.: 324-8.415 8.. 425-8.481 Dissc lived 5.78-6.21 5 .83-6.13 5 .73-5.99 5 .71-6.01 5. .82-6.08 5 .51-6.21 oxygen Al 0.19-0.23 .20-0.26 .22-0.25 .21-0.26 0. .21-0.26 .24-0.27 B 0.48 0.46 0.45 .44-0.45 .44-0.45 .43-0.45 Ba 0.02 .02-0.03 .02-0.03 .02-0.03 0, .02-0.03 .02-0.03 Ca 70.7-88.2 35.9-37.7 25.3-26.6 19.1-20.4 16.1-17.2 13.0-13.5 K 1.8-2.0 : 1.7-2.1 : 1.6-2.2 : 1.7-2.3 : 1.8-1.9 : 1.9-2.3 Mg 12.8-13.2 11.3-11.5 10.6-10.8 10.2-10.4 10.2-10.2 < 9.9-10.0 Mn 0.27-0.34 .12-0.15 0, .06-0.08 .02-0.03 0.01 <0.01 Na 30.2-31.5 31.9-32.9 32.5-33.2 32.6-34.1 32.3-34.3 33.0-34.3 Si 3.94-4.12 3 .17-3.22 2. .78-3.00 2 .59-2.87 2. .53-2.76 2 .39-2.66 Zn 0.06-0.33 .03-0.06 0. .05-0.10 .02-0.04 .01-0.06 <0 .01-0.01 As <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 Be <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Cd <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 Co <0.004 <0.003 <0.003 <0.003 <0.003 <0.003 Cr <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Cu <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Fe <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Mo <0. 01-0. 01 <0.01 <0.01 <0.01 <0.01 <0.01 Ni <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Pb <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Sb <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Se <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Sn <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 V <0.12 <0.12 <0.12 <0.12 <0.12 <0.12 -75- Table 32. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of OBG during chronic bioassays for green sunfish ( Lepomis wanellus ) . OBG Concentrations (%) Parameter 10.0 5.0 2.5 1.25 0.625 0.0 pH (units) 7.745-7.902 7.976-8.149 8. 138-8.346 8. 224-8.467 8. 274-8.544 8.320-8.592 Dissolved 7.43-7.82 7.37-7.69 7 .53-8.35 7 .38-7.81 7 .26-7.76 7.52-7.98 oxygen Al 0.12-0.16 0.13-0.16 .12-0.15 .14-0.15 .13-0.17 0.15-0.17 B 0.39-0.79 0.36-0.81 .35-0.81 .34-0.81 .34-0.81 0.33-0.81 Ba 0.03 0.03 0.03 0.03 .03-0.10 0.03 Ca 58.2-73.4 32.6-34.6 22.9-24.4 17.2-18.5 13.8-15.4 11.0-12.3 K <2.14 <2.2-2.2 <2 .07-2.09 <2.14 <2 .09-2.64 <2. 66-2. 66 Mg 10.8-11.4 9.6-10.2 1 3.9-9.5 1 3.6-9.2 ! 3.4-9.1 8.3-8.8 Mn 0.23-0.24 0.12 .06-0.07 .03-0.04 .01-0.02 <0.01 Na 25.6-26.6 26.9-27.6 27.2-27.6 27.2-28.0 27.5-29.0 27.0-28.4 Si 3.83-4.56 3.09-3.90 2 .73-3.55 2 .58-3.36 2 .47-3.19 2.38-3.14 Zn <0.06 <0.05 <0 .04-0.05 <0 .04-0.07 <0.05 <0.04 As <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 Be <0.002 <0.002 <0.002 <0.002 <0.002 <0.002 Cd <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Co <0.008 <0.008 <0.008 <0.008 <0.008 <0.008 Cr <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Cu <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Fe <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Mo <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Ni <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Pb <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 Sb <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Se <0.06 <0.06 <0.06 <0.06 <0.06 <0.06 Sn <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 V <0.18 <0.18 <0.18 <0.18 <0.18 <0.18 -76- Table 33. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of OBG during chronic bioassays for the crustacean, Daphnia magna . OBG Concentrations (%) Parameter 10.0 5.0 2.5 1.25 0.625 0.0 pH (units) 7.722-7.874 7.932-8.100 8.110-8.272 8.327-8.425 8.252-8.481 8.447-8.505 Dissolved 6.95-7.54 7.02-7.25 7.00-7.44 6.88-7.38 6.83-7.22 6.77-7.35 oxygen Al <0.11-.14 <0. 11-0. 13 <0. 14-0. 14 0.14-0.16 0.14-0.17 0.16-0.18 B 0.38-0.58 0.36-0.56 0.35-0.56 0.34-0.56 0.34-0.55 0. 34-0.56 Ba 0.027-0.029 0.028-0.030 0.028-0.030 0.030-0.036 0.028-0.031 0.028-0.031 Ca 61.7-67.7 33.7-35.0 21.2-23.3 17.0-17.6 14.2-14.4 11.2-11.4 K <2.1 <2.2 <2.4 <2.4 <2.2 <2.0 Mg 10.8-11.0 9.5-9.6 8.8-8.9 8.6 8.4-8.5 8.3-8.4 Mn 0.17-0.20 <0. 07-0. 07 <0.03 <0. 01-0. 02 <0.01 <0.01 Na 25.2-25.6 25.5-26.9 26.5-27.0 26.4-27.7 27.0-27.8 27.2-28.0 Si 3.68-4.05 2.99-3.41 2.66-2.98 2.47-2.82 2.38-2.76 2.30-2.72 Zn <0.10 <0.12 <0.06 <0.08 <0.04 <0.06 As <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 Be <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Cd <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Co <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 Cr <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Cu <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Fe <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Mo <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Ni <0.01 <0...01 <0.01 <0.01 <0.01 <0.01 Pb <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 Sb <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Se <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 Sn <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 V <0. 15 <0.15 <0.15 <0.15 <0.15 <0.15 -77- Table 34. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of OBG during chronic bioassay's for the snail, Physa anatina . OBG Concentrations (%) Parameter 10.0 5.0 2.5 1.25 0.625 0.0 pH (units) 7.790-7.879 8.036-8. 212 8.220-8.406 8.360-8.534 8.392-8.608 8.450-8.672 Dissolved 7.44-7.70 7.41-7. 62 7.48-8.22 7.37-8.22 7.34-7.61 7.28-7.71 oxygen Al 0.130-0.152 0.142-0. 165 0.150-0.176 0.152-0.176 0.150-0.169 0.157-0.178 B 0.442-0.644 0.416-0. 638 0.417-0.631 0.411-0.630 0.407-0.633 0.405-0.633 Ba 0.018-0.026 0.020-0. 027 0.020-0.028 0.020-0.028 0.020-0.030 0.021-0.032 Ca 67.7-68.6 34.2-35 .2 24.2-25.3 18.4-19.0 15.7-16.1 12.4-13.0 Fe <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 K <0.20 <1.9-2.C 1 <2.0 <1 .9-2.3 <2.1 <2.2 Mg 12.0-12.1 10.6-10. 9 9.9-10.2 9.6-9.9 9.9-9.7 9.4-9.5 Na 28.4-28.9 29.4-29. 9 29.7-30.7 23.7-30.8 30.2-31.1 30.9-31.5 Si 4.03-4.44 3.28-3.74 2.96-3.36 2.74-3.15 2.66-3.12 2.57-3.00 As <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 Be <0.006 <0.006 <0.006 <0.006 <0.006 <0.006 Cd <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Co <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 Cr <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Cu <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Mo <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Ni <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Se <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 Sb <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Pb <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 Sn <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 V <0.14 <0.15 <0.15 <0.15 <0.15 <0.15 Zn <0. 07-0. 21 <0.13 <0. 08-0. 17 <0.32 <0.33 <0. 14-0. 14 -78- Although there may have been some synergistic effects of various chemical constituents in WSNG, iron appeared to be the only chemical present at toxic levels in chronic-test solutions (tables 35 to 38). Acid mine drainage often contains high levels of ferrous iron (Fe 2+ ) that oxidizes to ferric iron (Fe 3+ ) and precipitates as ferric hydroxide (Fe(OH) } as the pH is elevated and oxygenation takes place (Smith and Sykora, 1976), for example, when mine drainage enters a receiving water. Mixing WSNG with dilution water in these experiments raised the pH and formed a precipitate, predominantly a ferric hydroxide phase. The presence of this visible, yellow flocculent, also reflected by the turbidity values of solutions (tables 35 to 38), represented both a physical and physiological interference to the functional processes of the organisms. The lowered hatchability of both green sunfish and fathead minnow eggs was probably due to suffocation, as the precipitate would attach to eggs and block oxygen transfer. Small masses of ferric hydroxide suspensions are similar in size to the pores in the chorion of fathead minnow eggs. Consequently, clogging these pores by ferric hydroxide particles would reduce oxygen diffusion across the chorion and cause embryo mortality (Smith et al., 1973). Eggs in this study were suspended in egg cups and rocked in exposure chambers, which may have helped keep some eggs free of the precipitate. If eggs had rested on the chamber bottoms, hatchability would probably have been further reduced. The gills of fish that hatched and died were often covered with a layer of ferric hydroxide, suggesting interference with respiration. Growth of surviving fish appeared to be little affected by the different concentrations of WSNG; although the highest concentration of WSNG pro- duced high mortality of green sunfish. These results support previous evidence that different life stages and species can be differentially affected by chronic exposure to ferric hydroxide suspensions (Smith et al., 1973; Smith and Sykora, 1976; Sykora et al. 1972, 1975). Daphnia mortality and reproductive problems were common at the highest concentrations of WSNG, the problems were probably a result of consuming the ferric hydroxide precipitate, as these crustaceans are filter feeders on particulate matter in water. The gut of these Daphnia (easily visible in these translucent organisms) was often a yellow-gold color, similar to the color of ferric hydroxide precipitate. Consumption of precipitate may also have been responsible for reduced growth and increased mortality in the highest WSNG concentrations for Physa . These snails often consumed the precipitate as was evident from their orange fecal material. The WSNG concentrations of 0.1, 0.05, and 0.025 percent, which corres- ponded to total solution iron concentrations of 3.37 to 8.1 mg/L, 1.43 to 4.37 mg/L, and 0.79 to 1.40 mg/L, were toxic in some form to all -79- organisms tested. Significant differences in development were observed in both fish species in their exposure to 0.0125-percent WSNG relative to control solutions. These WSNG solutions were characterized by iron concentrations of 0.42 to 0.58 mg/L total solution iron. Concentration of total iron in solutions of 0.00625-percent WSNG (the "no effect" concentration) used in green sunfish and fathead minnow tests ranged from 0.26 to 0.29 mg/L. Add to these values an estimation of the amount of iron present as a precipitate (based on a mean value of 3881 mg/L total iron in the full strength WSNG leachate) and a safe concentration of total iron therefore would be between 0.29 and 0.64 mg Fe/L. These chronic bioassays have demonstrated that a 0.0125-percent solution of WSNG (an acutely toxic, sublethal concentration) will cause toxic effects in fathead minnows and green sunfish. These tests also show that other low concentrations of this leachate are toxic to various organisms in the aquatic food chain — specifically, to a crustacean, a snail, and two fish species. Disruption of a single element in the aquatic food chain could modify an entire aquatic ecosystem. The toxicity shown here for the WSNG leachate demonstrates the potential effects of these wastes should they ever enter a receiving water in a natural setting. SUMMARY • The coal slurry (coarse fraction) and refuse samples were chemically and mineralogically similar; aluminosilicates mixed with organic and inorganic sulfur and carbonaceous matter. The composition of the mine spoil samples was typical of shale, although higher in sulfur. The Kilngas sample resembled other high-temperature coal residues, such as slag and fly ash, except for higher Cr and Ni concentra- tions. Arsenic, Co, Ni, Pb, Si and Sb were concentrated in the refuse samples, presumably associated with sulfide minerals. • The slurry (coarse fraction) and the refuse in the mining areas studied have the capacity for producing very acidic effluent and do not appear to contain sufficient carbonate minerals to have any appreciable neutralizing influence. Laboratory extracts of the mine spoils tended to be neutral in pH; whereas the Kilngas sample generated alkaline systems. o Laboratory extractions of the slurry and refuse samples indicated that Al, Mn, Ni, SO , and Zn are in soluble forms and may leach out in appreciable quantities. Elevated levels of these same constit- uents were detected in field samples taken in the proximity of the solid sample. • The solid coal wastes contained various organic compounds; many are included on the EPA Priority Pollutant List. However, these com- pounds were found to be essentially insoluble in aqueous media. -80- Table 35. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of WSNG during chronic bioassays for fathead minnows, Pimephales promelas . WSNG Concentrations (%) Parameter 0.10 0.05 0.025 0.0125 0.00625 0.0 pH( units) 7.770-7.868 8.021-8.056 8.151-8.176 8.130-8.287 8.254-8.307 8. 313-8.367 Dissolved 7.53-7.85 7.44-7.77 7.50-7.79 6.98-7.80 7.40-7.62 7 .26-7.45 oxygen Turbidity 12.3-20.0 6.7-9.0 3.6-4.8 2.2-3.2 2.0-4.4 1.3-1.8 (JTU) Al 0.17-0.20 0.16-0.19 0.16-0.19 0.16-0.20 0.16-0.19 .16-0.17 B 0.34-0.46 0.34-0.46 0.34-0.45 0.34-0.45 0.34-0.46 .35-0.46 Ba 0.024-0.027 0.024-0.026 0.026 0.026-0.027 0.026-0.027 O.i 026-0.030 Ca 9.30-9.94 8.95-9.68 8.85-9.61 8.82-9.52 8.86-9.55 8 .89-9.51 Co 0.07-0.10 0.03-0.04 <0. 02-0. 03 <0. 02-0. 02 <0.02 <0.02 Fe 3.37-4.44 1.74-1.89 0.92-0.94 0.42-0.58 0.26-0.29 <0.02 K <2.20 <2.18 <2.10 <2.16 <2.05 <2.28 Mg 9.10-9.70 8.72-9.40 8.52-9.27 8.52-9.28 8.54-9.26 8 .58-9.16 Mn 0.040-0.059 0.025-0.030 0.010-0.014 0.008 <0.008 <0.008 Na 25.7-27.7 25.0-27.4 25. 3-27.1 25.9-31.6 25.3-29.4 25.6-27.2 Si 2.48-2.78 2.45-2.70 2.45-2.67 2.44-2.67 2.44-2.72 2 .45-2.69 As <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 Be <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 Cd <0.39 <0.39 <0.39 <0.39 <0.39 <0.39 Cr <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Cu <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Mo <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Ni <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Pb <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Sb <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 Se <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 Sn <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 V <0.08 <0.08 <0.08 <0.08 <0.08 <0.08 Zn <0. 02-0. 03 <0.02 <0.02 <0.02 <0.02 <0.02 -81- Table 36. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of WSNG during chronic bioassays for green sunfish, Leponis cyanellus . WSNG Concentrations (%) Parameter 0.10 0.05 0.025 0.0125 0.00625 0.0 pH (units) 7.801-7.874 8.042-8.159 8.254-8.284 8.319-8.377 8.396-8.426 8.444-8.496 Dissolved 7.74-7.90 7.35-7.79 7.56-7.90 7.32-7.54 7.28-7.69 7.39-7.62 oxygen Turbidity 15.0-18.0 7.4-12.1 4.5-13.7 2.5-3.1 2.2-2.4 1.1-2.5 (JTU) Al 0.17-0.28 0.16-0.18 0.16-0.19 0.16-0.18 0.16-0.17 0.17-0.17 B 0.36-0.46 0.36-0.45 0.35-0.46 0.35-0.46 0.34-0.45 0.34-0.46 Ba 0.026-0.027 0.024-0.026 0.026-0.027 0.026-0.027 0.025-0.027 0.026-0.027 Ca 9.26-10.06 9.01-9.85 8.77-9.62 8.64-9.64 8.70-9.57 8.74-9.48 Co <0. 07-0. 18 0.02-0.05 0.02-0.05 <0.02 <0.02 <0.02 Fe 3.54-6.60 1.43-2.13 0.79-0.98 0.42-0.52 0.26-0.27 <0.02 K <2.32 <2.14 <2.66 <2.71 <2.62 <2.14 Mg 9.18-9.74 8.84-9.43 8.02-9.32 8.48-9.21 8.48-9.18 8.53-9.04 Mn 0.041-0.067 0.023-0.031 0.010-0.017 <0. 006-0. 008 <0.010 <0.010 Na 25.9-27.0 25.8-27.2 25.5-27.3 25.5-29.7 26.0-27.6 25.0-27.6 Si 2.50-2.87 2.47-2.76 2.45-2.70 2.44-2.70 2.42-2.68 2.42-2.70 Zn 0.02-0.05 <0.02 <0.02 <0.02 <0.02 <0.02 As <0.08 <0.08 <0.08 <0.08 <0.08 <0.08 Be <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 Cd <0.08-<0.91 <0.08-<0.91 <0.08-<0.91 <0.08-<0.91 <0.08-<0.91 <0.08-<0.91 Cr <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Cu <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Mo <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Ni <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Pb <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 Sb <0-04 <0.04 <0.04 <0.04 <0.04 <0.04 Se <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 Sn <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 V <0.08 <0.08 <0.08 <0.08 <0.08 <0.08 -82- Table 37. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of WSNG during chronic bioassays for the crustacean, Daphnia magna . WSNG Concentrations (%) Parameter 0.10 0.05 0.025 0.0125 0.00625 0.0 pH (units) 7.760-7.888 8.027-8.074 8.174-8.203 8.275-8.293 8.314-8.328 8.382-8.449 Dissolved 7.72-8.42 7.72-8.66 7.83-8.20 7.82-8.70 7.83-8.51 7.96-8.68 oxygen Turbidity 21.7-51.4 6.7-10.8 3.8-4.4 2.6-2.8 1.9-2.0 1.1-1.2 (JTU) Al 0.24-0. 35 0.18-0.27 0.17-0.20 0.16-0.20 0.14-0.17 0.15-0.19 B 0. 34-0.46 0.35-0.46 0.34-0.45 0.34-0.44 0.34-0.47 0.35-0.46 Ba 0.026-0. 029 0.025-0.029 0.025-0.026 0.024-0.027 0.024-0.027 0.024-0.036 Ca 9.4-10.2 9.0-9.9 8.8-9.8 8.7-9.7 8.3-9.6 8.8-9.6 Co 0.11-0.31 0.02-0.12 <0. 03-0. 04 <0. 02-0. 02 <0.02 <0.01 Fe 5.64-8.16 1.81-4.37 1.10-1.40 0.57-0.74 <0. 25-0. 61 <0.04 K <2.54 <2.00 <2.46 <2.40 <2.00 <2.26 Mg 9.34-9.80 8.76-9.56 8.44-9.36 8.29-9.26 8.10-9.21 8.61-9.14 Mn 0.079-0.084 0.027-0.047 0.012-0.022 0.010-0.018 0.008-0.018 <0.008 Na 26.5-27.0 25.8-27.4 24.7-26.8 24.1-27.1 25.8-26.7 25.8-26.8 Si 2.68-2.78 2.52-2.70 2.44-2.66 2.45-2.60 2.43-2.64 2.44-2.65 As <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 Be <0.003 <0.003 <0.003 <0.003 <0.003 <0.003 Cd <0.38 <0.38 <0.38 <0.38 <0.38 <0.38 Cr <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Cu <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Mo <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Ni <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Pb <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 Sb <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 Se <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 Sn <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 V <0.08 <0.08 <0.08 <0.08 <0.08 <0.08 Zn <0.08 <0.05 <0.03 <0.03 <0.03 <0.03 -83- Table 38. Ranges of mean physicochemical parameter values (mg/L) for replicate concentrations of WSNG during chronic bioassays for the snail, Physa anatina . WSNG Concentrations (%) Parameter 0.10 0.05 0.025 0.0125 0.00625 0.0 pH(units) 7.776-7.828 8.036-8.056 8.213-8.228 8.328-8.356 8.366-8.406 8.390-8.462 Dissolved oxygen 7.70-8.29 7.50-8.00 7.68-8.12 7.51-8.07 7.57-8.22 7.78-8.00 Turbidity (JTU) 20.9-30.8 7.1-8.8 3.8-4.4 2.68-2.94 2.1-2.3 1.1-1.5 Al 0.19-0.22 0.17-0.20 0.17-0.19 0.14-0.18 0.16-0.19 0.14-0.18 B 0.36-0.47 0.36-0.47 0.36-0.47 0.36-0.46 0.36-0.46 0.36-0.46 Ba <0. 026-0. 027 0.026-0.027 0.026-0.027 0.026-0.029 0.026-0.027 0.026-0.029 Ca 9.3-9.9 9.0-9.6 8.9-9.5 8.9-9.5 8.9-9.7 8.8-9.4 Fe 4.34-5.54 1.72-2.14 0.82-0.96 0.38-0.54 0. 19-0.32 <0.07 K <2.1 <2.2 <2.5 <2.2 <2.4 <2.0 Mg 9.23-9.88 8.93-9.37 8.78-9.34 8.63-9.21 8.60-9.35 8.58-9.06 Mn 0.06-0.08 <0. 02-0. 03 <0.02 <0.01 <0.01 <0.01 Na 26.6-27.3 26.8-28.1 26.4-28.1 25.8-28.4 26.8-27.5 26.3-26.7 Si 2.60-2.84 2.56-2.78 2.57-2.76 2.56-2.75 2.49-2.74 2.53-2.74 As <0.06 <0.06 <0.06 <0.06 <0.06 <0.06 Be <0.005 <0.005 <0.005 <0.005 <0.005 <0.005 Cd <0.55 <0.55 <0.55 <0.55 <0.55 <0.55 Co <0. 10 <0.04 <0.03 <0.02 <0.02 <0.02 Cr <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Cu <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Mo <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 Ni <0.03 <0.03 <0.03 <0.03 <0.03 <0.03 Se <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 Sb <0.05 <0.05 <0.05 <0.05 <0.05 <0.05 Pb <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 Sn <0.04 <0.04 <0.04 <0.04 <0.04 <0.04 V <0. 10 <0.10 <0.10 <0.10 <0.10 <0. 10 Zn <0. 04 -0.04 <0.02 <0.40 <0.50 < 1 .60 <0.02 -84- • All the solid wastes contained elemental sulfur that was co-extracted with the organics by methylene chloride. • Acid mine drainage, whether field effluent or laboratory extracts of solid samples, tended to be in equilibrium with some type. of CaSO phase. • The concentrations of Al and Si in acid mine drainage do not appear to be controlled by the clay minerals or the feldspars in the refuse samples. The laboratory and field solutions tended to be in equilibrium with some type of SiO ? phase, while the solution chemistry of Al remains unresolved. • There is some evidence suggesting that acid mine drainage, whether field or laboratory extracts, may equilibrate with maghemite rather than an amorphous Fe(OH) phase. Fe(OH) solubility would over- predict Fe concentrations. ® Extracts generated from samples of spoil banks and a coal gasifica- tion waste were not acutely toxic to the crustacean, Daphnia magna , fathead minnows ( Pimephales promelas) , green sunfish ( Lepomis cyanellus ) , and the snail, Physa anatina whereas, the undiluted extracts produced by a long-term equilibration procedure from all of the coal cleaning wastes studied were acutely toxic to all four test organisms, requiring as much as a 1:1000 dilution to eliminate mortality. • The nontoxic leachates were essentially neutral in pH and contained low concentrations of accessory elements. The toxic extracts from the coal cleaning wastes were all acidic (pH <3.5) and contained concentrations of other elements that were both significantly correlated to mortality of the organisms and present at levels exceeding recommended water quality standards. ® The most toxic extracts were those with the lowest pH. The most acidic leachates, therefore, had the lowest LC-50 value and required the greatest amounts of dilution to eliminate mortality. • © Daphnia magna was generally considered to be the most sensitive organism in the acute toxicity studies, followed closely by Pimephales promelas , then Lepomis cyanellus and Physa anatina . The dissolution of the calcium carbonate shell of Physa lent additional buffering capacity to test solution waters, accounting partly, for its greater tolerance to some solutions toxic to other organisms. One coal cleaning refuse leachate produced by a large-volume genera- tion procedure for the chronic bioassays was acutely toxic to the test organisms. Another leachate produced by a different refuse sample using the long-term equilibration procedure was toxic, whereas the leachate produced by the large-volume procedure was not toxic, -85- reflecting the differences in the generation procedures, and possibly, sample heterogeneity. • Concentrations of the nontoxic refuse leachate produced by the large- volume procedure showed no detrimental chronic effects. • Low concentrations of the toxic refuse leachate produced by the large-volume procedure caused toxic effects: decreased hatchability for Lepomis cyanellus and Pimephales promelas eggs; increased mortality with chronic exposure for Lepomis cyanellus , Daphnia, and Physa; and reduced weight gain in Physa . • Soluble ferrous iron in the toxic leachate converted to insoluble ferric species when the leachate was diluted. Ferric hydroxide precipitate was the major factor implicated in the physical and physiological effects exhibited by all test organisms subjected to acutely toxic, sublethal concentrations of this leachate. • Low concentrations of the toxic leachate were shown to affect several life stages and species of aquatic organisms that represent several trophic levels in aquatic ecosystems. Given that the disruption of a single element in an aquatic food chain can have dramatic conse- quences on an aquatic system, the toxicity of this leachate is demonstrative of the potential effects that this type of waste can have on the aquatic ecosystem, if it enters a receiving water in a natural setting. RECOMMENDATIONS • A limitation of this type of project and the earlier work of Griffin et al. (1980), is that single grab samples were used; such samples may or may not be representative of the deposit under study. Recent work by Krapac, Smyth, and Griffin (1983) showed that two refuse piles in southwestern Illinois are heterogeneous with respect to water soluble constituents via laboratory extractions. Similar investigations need to be carried out with more thorough sampling of each type of waste (e.g. refuse, slurry, mine spoils) to gain a more accurate assessment of the environmental impact of coal utilization. • Laboratory extractions of coal solid wastes resemble field leachates in terms of their prevalent chemical character but may or may not duplicate the chemical composition of leachates quantitatively. Coal by-products and their associated leachates are extremely variable; therefore, it is not recommended that additional work be conducted to attempt to design extraction procedures applicable to all types of coal wastes where the objective is to produce quantitative facsimiles. -86- • The solubility relationships in both the field leachates and laboratory extracts remains only partially understood. Additional work is needed to characterize both the crystalline and X-ray amorphous phases in the solid phases, particularly at concentration levels below typical X-ray dif f ractometry detection limits. The computer-assisted model WATEQ2 should not be used indiscriminately in its present form with nondilute systems such as acid mine drainage. • Coal solid wastes do not appear to contain sufficient amounts of water-soluble organic compounds to pose serious environmental problems. A more complete and quantitative investigation would be useful in evaluating the validity of this conclusion. • Given the low pH and elemental composition of many coal cleaning waste leachates, these complex fluids should be characterized and treated as potentially toxic-forming substances. • Current EPA criteria state that 1.0 mg/L Fe is a safe concentration for freshwater aquatic life. These studies suggest a safe concentra- tion of total iron lies between 0.29 and 0.64 mg/L Fe. In light of this research and other studies (Smith et al., 1979) further inves- tigation into the mechanisms of iron toxicity related to formation of hydroxide precipitates is warranted. • The continuous-flow mini-diluter system for toxicity testing should not be used with complex solutions that precipitate compounds like ferric hydroxide; the coating effects of this precipitate necessi- tates a regimental cleaning schedule to maintain proper flow rates. The system is superb for testing non-iron-containing leachates. • These toxicity tests have been performed using extracts generated in the laboratory. Further tests should be initiated to study leachates collected from actual field locations. -87- REFERENCES American Public Health Association, 1975, Standard methods for the examination of water and wastewater (14th edition), American Public Health Association, Washington, D.C., 1193 p. American Society for Testing and Materials, 1979, Proposed methods for leaching of waste materials: Annual Book of ASTM-A Standards, Part 31, Water, Philadelphia, PA, p. 1258-1261. Anderson, B. G., 1948, The apparent thresholds of toxicity to Daphnia magna for chlorides of various metals when added to Lake Erie water: Transactions of the American Fisheries Society, v. 78, p. 96-113. Archibald, P. A., and E. Gentile, 1971, Distribution of phytoplankton in selected streams of the Slippery Rock Creek Watershed: Pennsylvanian Academy of Science, v. 45, p. 66-70. Asbury, J. G., and B. M. Hoglund, 1974, Facility siting in the state of Illinois, Part I: Siting regulation alternatives: Illinois Institute of Environmental Quality, IIEQ Doc. 74-4, Chicago, IL. Ball, J. W., and E. A. Jenne, 1979, WATEQ2 - A chemical model for trace and major element speciation and mineral equilibria of natural waters: American Chemical Society Symposium Series 93, Chemical Modeling in Aqueous Systems, p. 815-836. Barnhisel, R. I., and A. L. Rotromel, 1974, Weathering of clay minerals by simulated acid coal spoil-bank solutions: Soil Science, v. 118, p. 22-27. Bell, H. L., 1971, Effect of low pH on the survival and emergence of aquatic insects: Water Resources, v. 5, p. 213-219. Benoit, D. A., and G. W. Holcombe, 1978, Toxic effects of zinc on fathead minnows, Pimephales promelas : Journal of Fisheries Biology, v. 13, p. 701-708. Benoit, D. A., V. R. Mattson, and D. L. Olson, 1982, A continuous-flow mini-diluter system for toxicity testing: Water Research, v. 16, p. 457-464. Beyer, L. E., and R. J. Hutnik, 1969, Chemical properties of toxic strip-mine spoil banks in Pennsylvania: American Chemical Society Division of Fuel Chemistry v. 13, p. 120-123. -88- Biesinger, K. K. , and G. M. Christensen, 1972, Effects of various metals on survival, growth, reproduction, and metabolism of Daphnia magna : Journal of the Fisheries Research Board of Canada, v. 29, p. 1691-1700. Birge, W. J., 1978, Aquatic toxicology of trace elements of coal and fly ash, in_J. H. Thorp and J. W. Gibbons (eds.), Energy and environmen- tal stress in aquatic ecosystems: DOE Symposium Series, p. 219-240. Blumer, M., 1957, Removal of elemental sulfur from hydrocarbon fractions: Analytical Chemistry, v. 29, p. 1039-1041. Busch, R. A., R. R. Backer, and L. A. Atkins, 1974, Physical property data on coal waste embankment materials: U.S. Bureau of Mines, Report of Investigations 7964, 142 p. Busch, R. A., R. R. Backer, L. A. Atkins, and C. D. Kealy, 1975, Physical property data on fine coal refuse: U.S. Bureau of Mines, Report of Investigations 8062, 40 p. Cairns, J., 1978, Summary, recommendations and the future of reclaiming surface mines for fish and wildlife ^n_ D. E. Samuel, J. R. Stauffer, C. H. Hocutt, and W. T. Mason (eds.), Surface mining and fish/wildlife needs in the eastern United States: Fish and wildlife service and West Virginia University sponsored symposium. FWS/0BS- 78/81, U.S. Department of the Interior. Cairns, J., J. S. Crossman, K. L. Dickson, and E. E, Herricks, 1971, The recovery of damaged streams: Association Southeastern Biologists Bulletin, v. 13, p. 79-106. Cherry, D. S., R. K. Guthrie, J. H. Rodgers, J. Cairns, and K. L. Dickson, 1976, Responses of mosquito fish ( Gambusia affinis) to ash effluent and thermal stress: Transactions of the American Fisheries Society, v. 105, p. 686-694. Chu, T.-Y. J., P. A. Krenkel, and R. J. Ruane , 1976, Characterization and reuse of ash pond effluents in coal-fired power plants: 31st Annual Purdue Industrial Waste Conference, Purdue Univ., West Lafayette, IN; Ann Arbor Science, Ann Arbor, MI, p. 690-712. Cleland, J. G., and G. L. Kingsbury, 1977, Multimedia environmental goals for environmental assessment: U.S. EPA, v. 1, EPA-600/7-77- 136a and v. 2, EPA 600/7-77-136b, Research Triangle Park, NC. Cox, D. B., T.-Y. J. Chu, and R. J. Ruane, 1977, Quality and treatment of coal pile runoff, in_ Proceedings of the Third Symposium on Coal Preparation, Louisville, KY, National Coal Association and Bituminous Coal Research, p. 252-277. -89- Curry, J. A., and D. K. Fowler, 1978, Coal strip mining in Appalachia- fish and wildlife considerations, in_D. E. Samuel, J. R. Stauffer, C. H. Hocutt, and W. T. Mason (eds.), Surface mining and fish/wildlife needs in the eastern United States: Fish and Wildlife Service and West Virginia University sponsored sympsoium, FWS/OBS- 78/81, U.S. Department of Interior. Davies, C. W., 1962, Ion association, Butterworths and Co., Ltd., 190 p. Davies, C. W., 1938, The extent of dissociation of salts in water. Part VIII. An equation for the mean ionic activity coefficient of an electrolyte in water, and a revision of the dissociation constants of some sulphates: Journal of the American Chemical Society, v. 271, p. 2093-2098. Davis, E. C, W. J. Boegly, and T. Tamura, 1981, A laboratory and field analysis of factors affecting quality of leachate from coal storage piles: Environmental Sciences Division Publication 1668, Oak Ridge National Laboratory, TN, ORNL/TM-7337, 198 p. Debye, Von P., and E. Hu'ckel, 1923, Zur Theorie der Elektrolyte: Physikalische Zeitschrift, v. 24, p. 185-206. Dills, C, and D. T. Rogers, Jr., 1974, Macroinvertebrate community structure as an indicator of acid mine pollution: Environmental Pollution, v. 6, p. 239-262. Elseewi, A. A., A. L. Page, and S. R. Grimm, 1980, Chemical character- ization of fly ash aqueous systems: Journal of Environmental Quality, v. 9, p. 424-427. Epler, J. L., F. W. Lovimen, T. K. Rao, E. M. Burnett, W. H. Griest, M. R. Guerin, M. P. Meskerinec, D. A. Brown, N. T. Edwards, C. W. Gehrs, R. E. Millemann, B. P. Parkhurst, B. M. Ross-Todd, D. S. Shriner, and H. W. J. Nilson, 1980, Toxicity of leachates: Oak Ridge National Laboratory, Oak Ridge Tennessee, U.S. EPA-600/2-80- 057. Energy and Research Development Administration, 1977, Alternative fuels demonstration program: Final environmental impact statement, v. 1 and 2. Gluskoter, H. J., R. R. Ruch, W. G. Miller, R. A. Cahill, G. B. Dreher, and J. K. Kuhn, 1977, Trace elements in coal: Occurrence and distribution: Illinois State Geological Survey Circular 499. Griffin, R. A., R. M. Schuller, J. J. Suloway, S. T. Russell, W. F. Childers, and N. F. Shimp, 1978, Solubility and toxicity of potential pollutants in solid coal wastes, ln_ F. A. Ayer (ed.), Environmental Aspects of Fuel Conversion Technology III, U.S. EPA- 600/7-78-063. -90- Griffin, R. A., R. M. Schuller, J. J. Suloway, N. F. Shimp, W. F. Childers, and R. H. Shiley, 1980, Chemical and biological characterization of leachates from coal solid wastes: Illinois State Geological Survey Environmental Geology Notes 89, 99 p. Grim, R. E., W. F. Bradley, and W. A. White, 1957, Petrology of the Paleozoic Shales of Illinois: Illinois State Geological Survey Report of Investigations 203, 35 p. Gulledge, W. P., and W. C. Webster, 1979, ASTM leachate test evaluation program, Report FP-1183, Electric Research Institute, Palo Alto, CA. Guntelberg, E. , 1926, Untersuchungen uber Ioneninteraktion: Zeitschrift fur Physikalische Chemie, v. 123, p. 199-247. Harvey, R. D., R. A. Cahill, C. L. Chou, and J. D. Steele, 1983, Mineral matter and trace elements in the Herrin and Springfield Coals, Illinois Basin Coal Field: Illinois State Geological Survey Contract/Grant Report 1983-4, Champaign, IL, 162 p. Herbes, S. E., and J. J. Beauchamp, 1977, Toxic interaction of mixtures of two coal conversion effluents components resorcinol and 6 methyl quinoline to Daphnia magna : Bulletin of Environmental Contamination and Toxicology, v. 17, p. 25-32. Herricks, E. E. and J. J. Cairns, 1974, Rehabilitation of streams receiving acid mine drainage: Virginia Polytechnic Institute and State University Water Resources Center Bulletin 66, 284 p. Hildebrand, S. G., R. M. Cushman, and J. A. Carter, 1977, The potential toxicity and bioaccumulation in aquatic systems of trace elements present in aqueous coal conversion effluents, in Proceedings of the University of Missouri Annual Conference of Trace Substances of Environmental Health, v. 10, p. 305-313. Hinckley, C. C. , G. V. Smith, H. Twardowska, M. Saporoschenko, R. H. Shiley, and R. A. Griffin, 1980, Mossbauer studies of iron in Lurgi gasification ashes and power plant fly and bottom ash: Fuel, v. 59, p. 161-165. Hood, W. C, T. A. Delfino, and P. D. Robinson, 1979, Acid-leachable metals in stream sediments, in Elements in Coal and Potential Environmental Concerns Arising from these Elements: Mini-Symposium Series 79-06, SEM-AIME Annual Meeting, New Orleans, LA, p. 15-20. Illinois Institute of Natural Resources, Allis-Chalmers , and Illinois Power Company, 1979, Kilngas (low BTU gasification) demonstration plant: technical and economic review. -91- Jarrell-Ash Division, 1978, Jarrell-Ash Plasma AtomComp Direct-Reading Spectrometer System: Fisher Scientific Company, Bulletin 434A, 17 P« Jones, J. B., and S. Ruggeri, 1969, Abatement of pollution from abandoned coal mines by means of in situ precipitation techniques: American Chemical Society, Division Fuel Chemistry, v. 13, p. 116- 119. Jones, J. R. E., 1964, Fish and river pollution: Butterworths, London, 203 p. Klimstra, W. D., and S. Jewell, 1973, Stripmining: resources in conflict, impacts and reclamation efforts in Illinois: North American Wildlife and Natural Resources Conference, v. 38, p. 121- 132. Kosanke, R. M., J. A. Simon, H. R. Wanless, and H. B. Willman, 1960, Classification of the Pennsylvanian strata of Illinois: Illinois State Geological Survey Report of Investigations 214, 84 p. Krapac, I. G., C. A. Smyth, and R. A. Griffin, 1983, Variability of constituent concentrations in laboratory extracts from coal-cleaning refuse, in Sixth Annual Madison Conference of Applied Research and Practice on Municipal and Industrial Waste, Madison, WI, p. 500-514. Krumbein, W. C, and L. L. Sloss, 1963, Stratigraphy and sedimenta- tion: W. H. Freeman and Co., San Francisco, CA, 660 p. Lamar, J. E., 1948, Clay and shale resources of extreme southern Illinois: Illinois State Geological Survey Report of Investigations 128, 107 p. Litchfield, J. T., and F. Wilcoxon, 1949, A simplified method of evaluating dose-effect experiments: Journal of Pharmacology and Experimental Therapeutics, v. 96, p. 99-113. Lowenbach, W., 1978, Compilation and evaluation of leaching test methods: U.S. Environmental Protection Agency, EPA-600/2-78-095. Martin, J. F., 1974, Quality of effluents from coal refuse piles, rn_ Proceeding of the First Symposium on Mine and Preparation Plant Refuse Disposal, Louisville, KY, National Coal Association, p. 26- 37. Mason, W. T., 1978, Methods for the Assessment and Prediction of Mineral Mining Impacts on Aquatic Communities: A Review and Analysis, December 6-7, 1977, Harpers Ferry, West Virginia: Eastern Energy and Land Use Group, Office of Biological Services, Fish and Wildlife Service, U.S. Department of the Interior. -92- Matler, W. J., J. J. Ney, and 0. E. Maugham, 1978, Sustained impact of abandoned surface mines on fish and benthic invertebrates populations in headwater streams of southwestern Virginia, Jri_ D » E » Samuel, J. R. Stauffer, C. H. Hocutt and W. T. Mason (eds.), Surface mining and fish/wildlife needs in the eastern United States: Fish and Wildlife Service and West Virginia University sponsored symposium, FWS/OBS-78-81 . Nichols, L. E., Jr., and F. J. Bulow, 1973, Effects of acid mine drainage on the stream ecosystem of the East Fork of the Obey River, Tennessee: Journal of the Tennessee Academy of Science, v. 48, p. 30-39. Nie, D. H., C. H. Hull, J. G. Jenkins, K. Steinbrenner , and D. H. Bent, 1975, Statistical package for the social sciences: McGraw-Hill Book Company, New York, NY, 675 p. Nordstrom, D. K., E. A. Jenne , and J. W. Ball, 1979, Redox equilibria on iron in acid mine waters: American Chemical Society Symposium Series 93, Chemical Modeling in Aqueous Systems, p. 51-80. Office of Technology Assessment, 1977, Analysis of the proposed National Energy Plan, August, 1977, 242 p. Orciari, R. D., and W. 0. Hummon, 1975, A comparison of benthic oligochaete populations in acid and neutral benthic environments in southwestern Ohio: Ohio Journal of Science, v. 75, p. 44-45. Parsons, J. D., 1968, The effects of acid strip-mine effluents on the ecology of a stream: Archives of Hydrobiology, v. 65, p. 25-50. Patrick, R. , J. Cairns, A. Scheier, 1968, The relative sensitivity of diatoms, snails, and fish to twenty common constitutents of industrial wastes: The Progressive Fish Culturist, v. 30, p. 137- 140. Plummer, L. N., B. F. Jones, and A. H. Truesdell, 1976, WATEQF - A FORTRAN IV version of WATEQ, a computer program for calculating chemical equilibrium of natural waters: U.S. Geological Survey- Water Resources Investigations, 76-13, 61 p. Preston, H. R. , 1978, Bioassay Techniques, in Methods for the Assessment and Prediction of Mineral Mining Impacts on Aquatic Communities: A review and analysis. December 6-7, 1977, Harper's Ferry, West Virginia: Eastern Energy and Land Use Group, Office of Biological Services, Fish and Wildlife Service, U.S. Department of Interior. Reppert, R. T., 1964, Aquatic life and the acid reaction, in Proceedings of the Symposium on Industrial Waste Control, v. 5, p. 27-49. -93- Roback, S. S., and J. W. Richardson, 1969, The effects of acid mine drainage on aquatic insects: Proceedings of the Academy of Natural Science of Philadelphia, v. 121, p. 81-107. Robinson, R. A., and R. H. Stokes, 1959, Electrolyte solutions (second ed.), Butterworths, London. Roy, W. R. , and R. A. Griffin, 1982, A proposed classification system for coal fly ash in multidisciplinary research: Journal of Environmental Quality, v. 11, p. 563-568. Roy, W. R., and R. A. Griffin, 1984, Illinois Basin coal fly ashes. 2. Equilibria relationships and qualitative modeling of ash-water reactions: Environmental Science and Technology, v. 18, p. 739-742. Roy, W. R., R. G. Thiery, R. M. Schuller, and J. J. Suloway, 1981, Coal fly ash: a review of the literature and proposed classification system with emphasis on environmental impacts: Illinois State Geological Survey Environmental Geology Notes 96, 69 p. Rubin, I. B., M. R. Guerin, A. A. Hardigree, and J. L. Epler, 1976, Fractionation of synthetic crude oil from coal biological testing: Engineering Research, v. 12, p. 358-365. Russell, S. J., and S. M. Rimmer, 1979, Analysis of mineral matter in coal, coal gasification ash, and liquefaction residues by scanning electron microscopy and X-ray diffraction, in C. Karr (ed.), Analytical methods for coal and coal products, v. 3, p. 133-162. Saporoschenko, M., C. C. Hinckley, G. V. Smith, H. Twardowska, R. H. Shiley, R. A. Griffin, and S. J. Russell, 1980, Mossbauer spectroscopic studies of the mineralogical changes in coal as a ^unction of cleaning, pyrolysis, combustion and coal-conversion processes: Fuel, v. 59, p. 567-574. Sather, N. R. , W. M. Swift, J. R. Jones, J. L. Beckner, J. H. Addington, and R. L. Wilburn, 1975, Potential trace element emissions from the gasification of Illinois coals: Illinois Institute for Environmental Quality, IIEQ Doc. 75-08, Chicago, IL. Schultz, T. W., S. Davis, and J. N. Dumont , 1978, Toxicity of coal conversion gasifier condensate to the fathead minnow: Bulletin of Environmental Contamination and Toxicology, v. 2, p. 237-243. Seay, G. J., 1972, Evaluation of sites for an Illinois coal gasification industry: The Illinois Institute for Environmental Quality, Chicago, IL. -94- Shannon, D. G., and L. 0. Fine, 1974, Cation solubilities of lignite fly ashes: Environmental Science and Technology, v. 8, p. 1026-1028. Shaver, R. H. , A. M. Burger, G. R. Gates, H. H. Gray, H. C. Hutchison, S. J. Keller, J. B. Patton, C. B. Rexroad, N. M. Smith, W. J. Wayne, and C. E. Wier, 1970, Compendium of rock-unit stratigraphy in Indiana: Indiana Geological Survey Bulletin 43, 229 p. Singer, P. C. , and W. Stumm, 1969, Rate-determining step in the production of acid mine wastes: American Chemical Society Division Fuel Chemistry, v. 13, p. 80-87. Smith, E. E., K. Svanks, and K. Shumate, 1968, Sulfide to sulfate reduction studies, ±n_ Second Symposium on Coal Mine Drainage Research, Pittsburgh, PA, p. 1. Smith, E. J., and J. L. Sykora, 1976, Early developmental effects of lime-neutralized iron hydroxide suspensions on brook trout and coho salmon: Transactions of the American Fisheries Society, v. 105, p. 308-312. Smith, E. J., J. L. Sykora, and M. A. Shapiro, 1973, Effect of lime- neutralized iron hydroxide suspensions on survival, growth, and reproduction of fathead minnow ( Pimephales promelas ): Journal of the Fisheries Research Board of Canada, v. 30, p. 1147-1153. Smith, E. J., K. Y. Chen, P. V. Hodson, J. B. Pearce, and D. L. Swanson, 1979, Iron, in_ R. V. Thurston, R. C. Russo, C. M. Fetterolf, Jr., T. A. Edsall, and Y. M. Barber, Jr. (eds.), A Review of the EPA Red Book: Quality Criteria for Water: Water Quality Section, American Fisheries Society, Bethesda, MD, p. 121-125. Smith, G. V., J. Liu, M. Saporoschenko, and R. H. Shiley, 1978, Mossbauer spectroscopic investigation of iron species in coal: Fuel, v. 57, p. 41-45. Southworth, G. R. , J. J. Beauchamp, and P. K. Schmieder, 1978, Bio- accumulation potential and acute toxicity of synthetic fuel effluents in freshwater biota: azaarenes: Environmental Science and Technology, v. 12, p. 1062-1066. Standon, 0. D., 1951, Some observations upon the maintenance of Australorbis glabratus in the laboratory: Annals of Tropical Medicine and Parasitology, v. 45, p. 80-82. Stone, R., and R. Kahle, 1977, Environmental assessment of solid residues from f luidized-bed fuel processing: U.S. Environmental Protection Agency, EPA-600/7-77-139, Research Triangle Park, NC. -95- Suloway, J. J., W. R. Roy, T. M. Skelly, D. R. Dickerson, R. M. Schuller, and R. A. Griffin, 1983, Chemical and toxicological properties of coal fly ash: Illinois State Geological Survey Environmental Geology Notes 105, 70 p. Suloway, J. J., R. M. Schuller, and R. A. Griffin, 1981, Acute toxicity of leachates from coal gasification and liquefaction solid wastes to the fathead minnow, Pimephales promelas : Journal of Environmental Science and Health, v. 16, no. 4, p. 419-445. Sun, C. C., C. H. Peterson, R. A. Newby, W. G. Vaux, and D. L. Keairns, 1978, Disposal of solid residue from f luidized-bed combustion: Engineering and laboratory studies, U.S. Environmental Protection Agency, EPA-600/7-78-049, Research Triangle Park, NC. Sykora, J. L., E. J. Smith, and M. Synak, 1972, Effect of lime- neutralized iron hydroxide suspensions on juvenile brook trout ( Salvelinus fontinalis Mitchill): Water Research, v. 6, p. 935-950. Sykora, J. L., E. J. Smith, M. Synak, and M. A. Shapiro, 1975, Some ob- servations of spawing brook trout ( Salvelinus fontinalis Mitchill) in lime neutralized iron hydroxide suspensions: Water Research, v. 9, p. 451-458. Talbot, R. W. , M. A. Anderson, and A. W. Andren, 1978, Qualitative model of heterogeneous equilibria in a fly ash pond: Environmental Science and Technology, v. 12, p. 1056-1062. Talty, J. T., 1978, Assessing coal conversion processes: Environmental Science and Technology, v. 12, p. 890-894. Truesdell, A. H. and B. F. Jones, 1974, WATEQ, a computer program for calculating chemical equilibria of natural waters: U.S. Geological Survey Journal of Research, v. 2, p. 233-274. U.S. Environmental Protection Agency, 1972, Procedure for conducting embryo-larval bioassays: Bioassay Committee, National Water Quality Laboratory, E.P.S., Duluth, MN. U.S. Environmental Protection Agency, 1975, Methods for acute toxicity tests with fish, macroinvertebrates and amphibians: Committee on methods for toxicity tests with aquatic organisms, EPA-660/3-75-009. U.S. Environmental Protection Agency, 1976, Quality criteria for water: Superintendent of Documents, Washington, DC, 256 p. U.S. Environmental Protection Agency, 1978, Procedures manual: Level I environmental assessment, (2nd ed.), Industrial Environmental Research Laboratory, Research Triangle Park, NC, EPA-600/7-78-201 . -96- U.S. Environmental Protection Agency, 1978, RCRA Section 3001 Draft Regulations, "Criteria identification and listing of hazardous wastes." U.S. EPA Office of Water and Hazardous Materials. U.S. Environmental Protection Agency, ,1979, Methods for chemical analysis of water and wastes, Environmental Monitoring and Support Laboratory, Cincinnati, OH, EPA-600/4-79-020. U.S. Environmental Protection Agency, 1980, Appendix II - EP Toxicity Test Procedure: Federal Register, 5-19-80, v. 45, no. 98, Superintendent of Documents, Washington, DC, p. 33063-33285. Vaugham, B. E., K. H. Abel, D. A. Cataldo, J. M. Hales, C. E. Hane, C. A. Rancitelli, R. C. Routson, R. E. Wilding, and E. G. Wolf, 1975, Review of potential impact on health and environmental quality from metals entering the environment as a result of coal utilization: Pacific Northwest Lab., Batelle Memorial Institute, Richland, WA, 75 p. Wewerka, E. M. , J. M. Williams, and P. L. Wanek, 1976, Assessment and control of environmental contamination from trace elements in coal processing wastes, in Proceedings of the National Conference of Health, Environmental Effects and Control Technology of Energy Use, U.S. EPA 600/7-76-002. Wewerka, E. M. , J. M. Williams, N. E. Vanderborgh, A. W. Harmon, P. Wagner, P. L. Wanek, and J. D. Olsen, 1978, Trace element character- ization of coal wastes - second annual progress report: DOE LA- 7360-PR, U.S. EPA 600/7-78-028a. Winger, P. V., 1978, Fish and benthic populations of the New River, Tennessee, ln_ D. E. Samuel, J. R. Stauffer, C. H. Hocutt, and W. T. Mason (eds.), Surface mining and fish/wildlife needs in the eastern United States: Fish and Wildlife Service and West Virginia University sponsored symposium. FWS/0BS-7/81 . -97- APPENDIX-A COMPILATION OF CHEMICAL ANALYSES OF ALL EXTRACTS -98- Table A-l. Chemical composition of extracts generated from samples of WSNG (concentrations in mg/L) . Extraction procedures ASTM-A 2 ASTM-B 2 EP 2 CAP 2 LTE 3 LVG 1 * PH 3.15 4.25 3.24 2.92 2.66 3.29 Eh (mV) 5 +551 +439 +578 +500 +686 +500 EC (dSm _1 ) 3.0 5.5 1.2 3.5 3.4 9.4 Al 18.7 8.68 3.20 21.9 41.9 115 As <0.25 <0.07 0.13 0.35 0.57 10.7 Ba <0.01 Si 6 <0.01 <0.01 <0.001 <0.001 Be 0.03 0.02 0.01 0.03 0.03 0.2 Ca 264 258 48.9 231 262 416 Cd 0.06 0.05 0.07 0.39 0.48 1.4 CI 1.6 2.7 1.9 1.8 6.5 -l Cr <0.06 <0.01 <0.02 <0.02 0.09 <0.01 Cu <0.29 0.03 <0.003 <0.003 0.03 <0.004 F 3.6 1.3 0.8 3.6 -l -l Fe 629 419 151 684 623 3825 K <11.3 10.9 2.04 5.71 7.21 6.68 Mg 55.4 58.0 11.8 52.3 49.8 497 Mn 5.63 6.00 0.99 4.96 4.73 Si 6 Mo <0.07 <0.01 <0.03 <0.03 <0.02 <0.01 Na 27 1147 <0.94 <0.94 <0.55 <2.6 NHi* 1.9 1.4 -l -l -l -l Ni 2.48 2.37 0.44 2.31 1.95 13.4 Pb <0.37 <0.01 <0.01 <0.03 <0.01 <0.04 Sb <0.38 <0.06 <0.02 <0.02 <0.02 <0.03 Se 0.33 <0.05 <0.05 <0.05 <0.06 <0.08 Si 11.1 16.3 <0.12 4.13 143 51.0 Sn <0.10 <0.02 <0.06 <0.06 <0.05 <0.04 SOi» 2007 2067 897 3022 3042 10,436 V 0.06 0.02 <0.12 0.30 <0.02 Si 6 Zn 1.36 1.42 0.70 1.47 5.43 54.0 *not determined. 2 average of three replicates. 3 sample taken after 147 days of equilibration. ^sample taken after 133 days of leaching. 5 relative to normal hydrogen electrode. Concentration could not be determined due to spectral interferences -99- Table A-2. Chemical composition of extracts generated from sample of WSOG (concentrations in mg/L) . Extraction procedures ASTM-A 2 ASTM-B 2 EP 2 CAP 2 LTE 3 pH 1.75 2.18 2.12 1.62 1.92 Eh (mV) 1 * +655 +647 664 +655 695 EC (dSm -1 ) 17.7 11.9 6.3 17.4 10.9 Al 384 214 42.9 180 69.8 As <0.01 1.77 0.71 2.65 Si 5 Ba <0.01 Si 5 <0.01 <0.01 0.006 Be 0.01 0.01 <0.01 <0.006 0.01 Ca 287 193 50.2 249 123 Cd 0.15 Si 5 0.42 2.28 1.26 CI 6.5 2.5 <1.0 2.0 2.9 Cr Si 5 <0.01 <0.02 <0.02 0.10 Cu <0.29 <0.03 <0.53 Si 5 0.51 F <0.5 <0.5 <0.5 <0.5 <0.5 Fe 5483 3544 787 2434 2552 K 36.9 18.2 9.01 32.4 52.7 Mg 15.1 10.6 2.19 9.61 3.26 Mn Si 5 2.64 Si 5 Si 5 <0.05 Mo <0.07 <0.01 <0.03 <0.03 <0.02 Na <27 1038 <0.84 <0.94 <0.55 NH4 <1.0 <1.0 -l -l _i Ni Si 2 <0.01 <0.01 <0.01 0.34 Pb <0.01 <0.06 0.61 <0.01 1.67 Sb <0.38 <0.06 0.06 <0.02 <0.02 Se <0.33 <0.05 <0.05 <0.05 <0.06 Si 23.1 20.9 <0.08 11.8 103.1 Sn <0.10 <0.02 <0.05 <0.06 <0.05 SO,, 18,782 12,797 4115 17,910 8721 V <0.01 0.84 0.37 2.11 <0.08 Zn 0.38 1.22 0.10 0.47 Si 5 x not determined. 2 average of three replicates. sample taken after 21 weeks of equilibration, ''relative to normal hydrogen electrode. could not be determined due to spectral interferences -100- Table A-3. Chemical composition of extracts generated from samples of OBG (concentrations in mg/L) . Extraction procedures ASTM-A 2 ASTM-B 2 EP 2 CAP 2 LTE 3 LVG 1 * pH 3.12 4.27 3.49 3.06 3.27 4.68 Eh (mV) 5 +644 +566 +655 +638 +797 +591 EC (dSm -1 ) 3.0 5.7 1.6 3.0 2.2 2.3 Al 158 49.7 22.5 120 64.5 2.21 As <0.01 <0.07 0.08 0.31 <0. 14 <0.06 Ba 0.02 0.03 <0.01 0.01 0.004 <0.001 Be 0.01 <0.001 <0.01 <0.006 0.006 <0.001 Ca 754 632 246 546 458 567 Cd 0.05 0.03 0.01 0.04 0.02 <0.02 CI 12.7 11.9 1.4 7.1 15.8 -l Cr <0.06 0.02 <0.02 <0.02 <0.02 <0.01 Cu <0.29 <0.03 0.01 0.02 0.01 <0.004 F 3.7 2.1 0.7 2.9 -l -l Fe 12.5 4.46 2.23 13.2 0.21 0.08 K <1.1 <2.4 <0.2 <0.2 0.44 <0.80 Mg 71.9 67.8 13.4 58.7 25.3 36.8 Mn 7.64 5.85 1.31 5.44 2.35 3.55 Mo <0.07 <0.01 <0.02 <0.03 <0.02 <0.01 Na <1.7 1282 <0.80 <0.94 3.60 4.63 NH^ <1.0 1.0 -l -l -l -l Ni 0.43 0.27 0.07 0.30 0.22 0.05 Pb <0.37 <0.06 <0.02 <0.02 <0.01 <0.04 Sb ■ <0.38 <0.06 <0.02 <0.02 <0.02 <0.03 Se <0.33 <0.05 0.05 <0.05 <0.06 <0.08 Si 13.7 16.8 0.98 9.26 51.2 14.1 Sn <0. 10 <0.02 <0.04 <0.06 <0.05 <0.04 S0 4 2153 2253 755 2179 1383 1595 V <0.20 <0.02 <0.08 <0.12 <0.06 <0.05 Zn 11.5 6.78 2.31 7.87 4.59 0.88 x not determined. 2 average of three replicates. sample taken after 147 days of equilibration. 4 sample taken after 150 days of leaching. 5 relative to normal hydrogen electrode. -101- Table A-4. Chemical composition of extracts generated from samples of the WSSL (concentrations in mg/L) . Extraction procedures ASTM-A 2 ASTM-B 2 EP 2 CAP 2 LTE 3 PH 2.43 3.88 2.81 2.48 2.24 Eh (mV)" +636 +448 +631 +613 +855 EC (dSm -1 ) 5.6 6.4 2.1 5.4 4.6 Al 63.5 37.0 11.2 70.1 27.4 As 0.42 0.17 0.23 0.86 0.58 Ba <0.02 <0.01 <0.01 <0.01 0.002 Be 0.04 0.03 <0.01 0.03 0.02 Ca 188 155 31.3 169 60.4 Cd 0.14 0.16 0.13 0.94 0.64 CI 5.2 2.2 <1.0 1.5 <1.0 Cr <0.11 <0.08 <0.02 <0.05 <0.02 Cu <0.29 <0.03 0.14 Si 5 0.58 F 2.0 <0.5 <0.5 <0.5 -l Fe 14 11 965 264 1274 971 K <11.3 <2.4 0.26 0.31 0.60 Mg 89.3 98.6 15.6 84.5 32.5 Mn 19.3 19.2 2.66 16.3 6.42 Mo <0.07 <0.01 <0.03 <0.03 <0.02 Na <27 1184 <0.94 <0.94 <0.94 NH., <1.0 <1.0 -l -l -l Ni 2.28 2.01 0.28 1.65 0.91 Pb <0.37 <0.06 <0.02 <0.02 0.01 Sb 0.38 <0.06 0.10 <0.02 <0.02 Se <0.33 <0.05 0.05 0.05 <0.06 Si 11.2 16.8 0.08 2.62 55.2 Sn <0.10 <0.02 <0.06 <0.06 <0.05 SO,, 5135 2670 516 5934 3001 V 0.28 <0.08 0.09 0.83 <0.08 Zn 7.67 8.09 1.25 6.57 2.99 *not determined. 2 average of three replicates. 3 sample taken after 147 days of equilibration. relative to normal hydrogen electrode. 5 could not be determined due to spectral interferences. -102- Table A-5. Chemical ci Dmposition of extracts generated from samples of DMNS (concentrations in mg/L) . Extraction procedures ASTM-A 2 ASTM-B 2 EP 2 CAP 2 LTE 3 pH 6.94 5.4 1 5.19 6.14 7.56 Eh (mV) 1 * +535 +476 +503 +497 +545 EC (dSm- 1 ) 1.0 7.0 1.3 1.6 0.8 Alkalinity 107 -1 -1 -1 -1 Al <0.88 <0.11 <0.08 <0.08 <0.07 As <0.25 <0.07 <0.04 <0.04 <0.07 Ba 0.04 0.26 0.06 0.06 0.02 Be <0.01 <0.01 <0.01 <0.01 <0.01 Ca 165 1238 197 323 144 Cd <0.03 <0.01 <0.01 <0.01 <0.01 CI 2.9 3.1 <1.0 2.3 2.1 Cr <0.06 <0.01 <0.02 <0.02 <0.02 Cu <0.29 <0.03 <0.01 <0.01 <0.01 F <0.05 <0.05 <0.05 <0.05 -1 Fe 0.07 <0. 19 <0.07 <0.07 <0.05 K <11.3 12.7 3.24 10.1 6.30 Mg 37.6 113 19.8 46.2 25.5 Mn 2.09 24.4 3.93 6.14 0.15 Mo <0.07 <0.01 <0.03 <0.03 <0.02 Na <27 1155 1.82 7.06 3.94 NH„ <1.0 1.1 -1 -1 -1 Ni <0.06 0.52 0.07 0.14 <0.05 Pb <0.37 <0.06 <0.02 <0.02 <0.01 Sb <0.38 <0.06 <0.02 <0.02 <0.02 Se <0.33 0.05 <0.05 <0.05 <0.06 Si 3.07 14.8 0.31 2.96 3.45 Sn <0. 10 0.06 <0.06 <0.06 <0.05 S0 4 509 440 86 4 74 427 V <0.02 <0.02 <0. 12 <0.12 <0.06 Zn <0.20 0.27 0.26 <0.07 0.33 *not determined. 2 average of three replicates. 3 sample taken after 147 days of equilibration. ''relative to normal hydrogen electrode. -103- Table A-6. Chemical composition of extracts generated from samples of DMOS (concentrations in mg/L) . Extraction proc edures ASTM-A 2 ASTM-B 2 EP 2 CAP 2 LTE 3 pH 7.57 5.64 5.07 6.24 7.77 Eh (mV) k +494 +495 +577 +491 +536 EC (dSm -1 ) 0.6 7.3 1.5 1.5 0.57 Alkalinity 67 -l -l -l 80 Al <0.88 <0.11 <0.08 <0.08 <0.07 As <0.25 <0.07 <0.04 <0.04 <0.07 Ba 0.04 0.62 0.18 0.11 0.04 Be <0.01 <0.01 <0.01 <0.01 <0.01 Ca 96.9 1320 259 271 67.0 Cd <0.03 <0.01 <0.01 <0.01 <0.01 CI 3.2 3.6 <1.0 2.4 4.1 Cr <0.06 <0.01 <0.02 <0.02 <0.02 Cu <0.29 <0.03 <0.01 <0.01 <0.01 F 0.7 0.5 <0.5 <0.5 -i Fe <0.06 1.44 <0.07 <0.07 <0.05 K <11.3 8.1 2.33 4.97 2.59 Mg 26.4 177 40.9 58.0 21.5 Mn <0.05 7.06 1.38 0.67 <0.01 Mo <0.07 <0.01 <0.03 <0.03 0.05 Na <27 1182 2.13 9.56 5.53 NHi, <1.0 <1.0 -l -l -l Ni <0.06 0.10 0.03 <0.02 <0.05 Pb <0.37 <0.06 0.02 <0.02 <0.01 Sb <0.38 <0.06 0.03 0.04 <0.02 Se <0.33 <0.05 <0.05 <0.05 <0.06 Si 4.44 19.3 2.76 7.39 7.25 Sn <0.10 0.07 0.11 0.08 <0.05 SC% 315 253 58.3 291 245 V <0.20 <0.02 <0.12 <0. 12 <0.06 Zn 0.18 0.13 0.33 <0.07 0.11 ! not determined. 2 average of three replicates. 3 sample taken after 147 days of equilibration. ''relative to normal hydrogen. -104- Table A-7. Chemical composition of extracts generated from samples of KG (concentrations in mg/L) . Extraction proc edures ASTM-A 2 ASTM-B 2 EP 2 CAP 2 LTE 3 PH 8.10 4.86 5.00 6.17 9.02 Eh (mV) 4 +516 +533 +594 +484 +446 EC (dSm -1 ) 0.6 0.5 0.3 1.2 0.6 Alkalinity 35 -l -l -l 88 Al <0.88 2.87 <0.08 <0.08 <0.07 As <0.25 <0.07 <0.04 <0.04 <0.07 Ba 0.02 0.20 0.03 0.06 0.004 Be <0.01 <0.01 <0.01 <0.01 <0.01 Ca 78.8 327 33.3 197 40.8 Cd <0.03 0.01 <0.01 0.01 <0.01 CI 2.8 2.6 1.0 2.4 3.6 Cr <0.06 0.11 <0.02 <0.02 <0.02 Cu <0.29 1.53 0.11 0.02 <0.01 F <0.5 <0.5 <0.5 <0.5 -l Fe 0.33 2.53 <0.07 <0.07 <0.05 K <11.3 12.3 1.51 5.64 4.48 Mg 16.3 187 6.72 59.1 40.8 Mn <0.05 1.65 0.10 0.44 <0.01 Mo <0.07 <0.01 <0.03 <0.03 0.12 Na <27 1192 1.51 5.58 4.00 NHi, <1.0 <1.0 -l -l -l Ni <0.06 3.00 0.40 1.18 <0.05 Pb <0.37 <0.06 <0.02 <0.02 <0.01 Sb <0.38 <0.06 <0.02 <0.02 <0.02 Se <0.33 <0.05 <0.05 <0.05 0.06 Si 4.19 39.6 1.41 9.82 7.06 Sn 0.10 0.03 <0.06 <0.06 <0.05 SO,, 251 266 85 267 314 V <0.20 <0.02 <0.12 <0. 12 <0.06 Zn 0.09 0.70 0.39 0.09 0.42 *not determined. 2 average of three replicates. 3 sample taken after 147 days of equilibration. ^relative to normal hydrogen electrode. -105- APPENDIX-B INTERELEMENT INTERFERENCES IN INDUCTIVELY COUPLED ARGON PLASMA (ICAP) APPLIED TO HIGH-IRON SOLUTIONS W. R. Roy and I. G. Krapac Argon supported inductively coupled plasma (ICAP) is among the state-of- the-art analytical methods for multielemental analyses. Quantitative simultaneous elemental determinations can be quickly made from aqueous solutions. These capabilities include measurements of constituents such as arsenic or boron, which are not easily performed by conventional atomic absorption or emission techniques at trace levels. Complex solutions such as the high-iron laboratory extracts of the coal solid wastes yielded spectral interferences that greatly hindered the determination of certain trace elements. Iron has many emission lines, many of which lie close to the primary or international spectral lines of other elements. This situation is illustrated in figure B-l. The ICAP used in this study (Jarrell-Ash 975 ICAP AtomComp) uses the 259.99 nm wavelength for Fe and the 238.89 nm wavelength for Co; however, there are two iron emission lines at 238.863 nm and 238.940 nm. The spectral band pass of the exit slit is typically 0.03 nm. This means that the detector views a wavelength region of approximately 0.03 nm on either side of the analytical wavelength. In other words, the detector cannot resolve signals emanating from the cobalt line from those coming from the near-by iron lines as suggested by figure B-l. As the iron in solution increases, so will its addition to the cobalt signal, producing a greater apparent concentration of Co. While Co was the element most susceptible to spectral interferences by iron, several other elements were also influenced by the presence of iron (table B-l), aluminum, and silicon. The magnitude of the apparent concentrations depends on the concentration of the interfering element and specific wavelengths studied. Ediger et al. (1980) found that apparent concentrations of Cr ranged from 0.1 to 0.8 mg/L using a 1000 mg/L Fe solution at wavelengths ranging from 205.55 to 284.32 nm for the determination of Cr. Spectral interferences are not unique to ICAP but are common problems in any spectroscopic technique. To overcome this problem in ICAP spectro- scopy, empirical correction factors are derived from single-element solutions at different concentrations as outlined by Marciello and Ward (1978). These factors are used, via computer software, to subtract out artifact emissions generated by major aqueous species that enhance the emission spectra of other constituents at trace levels. These correc- tion factors vary from instrument to instrument and many such factors have been reported and used in generating trace constituent data. -106- Table B-l . Interelement interferences generated by iron at a concentration of 100 mg/L Fe. Element Wave length Apparent Typical effected (run) concentration (mg/L) detection limit 1 (mg/L) Co 238.89 Al 308.22 Mn 257.01 B 249.77 Sn 189.99 Ba 455.4 Cu 324.75 Sb 206.53 Pb 220.35 Zn 206.20 V 292.4 Cr 267.72 Ni 231.60 Be 313.04 2.51 0.77 0.46 0.42 0.24 0.11 0.11 0.06 0.06 0.05 0.03 0.03 0.03 0.02 0.003 0.03 0.007 0.003 0.03 0.001 0.003 0.01 0.01 0.04 0.02 0.01 0.02 0.003 1 twice the standard deviation of 10 determinations of an acidified solution of de ionized water. -107- Analytical cobalt line (238.892 nm) (238.86 nm) | i (238.94 nm) |-< — Interfering iron lines 238.86 nm 238.92 nm Bandpass (0.06 - 0.08 nm) Figure B-1 Relationship between the spectral bandpass, analytical cobalt emission line, and two interfering iron lines. -108- In the course of the project, correction factors were derived and tested. The apparent concentration of some elements, such as, including cobalt, were also reduced by adjusting the position of the band pass to partially exclude the iron lines relative to the analytical line. Comparisons of corrected concentrations with analytical results from conventional atomic absorption indicated that these empirical correction factors were capable of eliminating the interferences for most of the trace elements. It was found, however, that determinations of V, Mn, B, and Co at concentrations less than 1 mg/L were still unreliable in some solutions containing Fe in concentrations exceeding 100 mg/L. Consequently, data for these elements were frequently omitted in the analytical results reported in this study. Refining the analytical capabilities of ICAP for trace elements in the presence of iron is a current area of study by the authors. It is recommended that determin- ations of trace constituents in complex solutions be routinely checked by independent methods and/or by different sample preparation procedures . REFERENCES Ediger, R. D., W. Stelzer, T. J. Hanson, and A. R. Knott, 1980, The selection of optimum ICP instrument parameters: Atomic Spectro- scopy, ICP Application Study 8, 17 p. Marciello, L. , and A. F. Ward, 1978, Technical Aid Note 3: Interelement corrections for spectral line interferences: Jarrell-Ash Plasma Newsletter, v. 1, n. 2, p. 12-13. -109- APPENDIX-C COMPILATION OF CHEMICAL ANALYSES OF ACUTE BIOASSAY TEST SOLUTIONS AND CORRELATIONS BETWEEN CONSTITUENT CONCENTRATION AND MORTALITY -110- Table C-l. Chemical composition of acute bioassay test solutions generated from OBG waste and the correlations between the concentrations of various chemical constituents and mortality in Daphnia magna . Range of concentration Recommended water in test solutions quality levels r 2 (mg/L) (mg/L) a Change b r c pH. 4.9 - 7.1 6.5 - 9.0 d 0.707 -0.84 (0.001) pH f 5.0 - 7.7 6.5 - 9.0 d 0.097 -0.93 (0.001) Al <0.06 - 0.87 1.0 0.069 0.75 (0.001) As <0.04 0.05 B <0.04 - 0.22 25 <0.001 0.77 (0.001) Ba <0.001 - 0.006 2.5 <0.001 0.18 Be <0.001 - 0.006 0.055 0.001 0.03 Ca 15.5 - 49.3 16 0.010 0.76 (0.001) Cd <0.08 0.001 0.011 0.04 Cr <0.01 0.25 0.012 -0.31 Cu <0.004 0.05 0.008 0.21 Fe <0.02 0.25 Mg 6.6 - 8.69 87 0.001 0.51 (0.001) Mn <0.005 - 0.250 0.1 0.004 0.77 (0.001) Mo <0.01 7 Ni <0.01 - 0.03 0.01 <0.001 0.42 (0.01) Pb <0.02 0.05 <0.001 -0.15 Sb <0.02 0.02 Se <0.04 0.25 0.008 0.04 Si 0.90 - 4.97 0.017 0.85 (0.001) Sn <0.02 0.013 -0.15 S0 4 86 - 234 250 0.84 (0.01) Zn <0.08 - 1.09 0.1 0.002 0.28 (0.05) MATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated, contribution to multiple regression of chemical composition on mortality. c linear regression value (statistical significance). from Quality Criteria for Water. (U.S. EPA, 1976). -Ill- Table C-2. Chemical composition of acute bioassay test solutions generated from OBG waste and the correlations between the concentrations of various chemical constituents and mortality in Lepomis cyanellus . Range of concentration Re commend ( ;d water in test solutions (mg/L) quality levels (mg/L) a r 2 Change b r c vn. 4.8 - 6.8 6.5 - 9 .0 d 0.413 -0.64 (0.001) P H f 4.8 - 7.6 6.5 - 9 .Qd 0.001 -0.74 (0.001) Al <0.06 - 3.16 1.0 0.331 0.92 (0.001) As <0.04 0.05 0.006 -0.03 B 0.05 - 0.38 25 <0.001 0.71 (0.001) Ba <0.001 - 0.005 2.5 0.018 0.05 Be <0.001 - 0.004 0.055 0.030 0.17 Ca 23.1 - 46.3 16 0.002 0.67 (0.001) Cd <0.003 0.001 Cr <0.02 0.25 <0.002 -0.12 Cu <0.004 0.05 0.018 0.35 (0.05) Fe <0.02 - 0.09 0.25 0.058 0.36 (0.01) Mg 6.66 - 9.68 87 0.022 0.57 (0.001) Mn 0.09 - 0.331 0.1 0.016 0.73 (0.001) Mo <0.02 7 0.009 0.04 Ni 0.01 - 0.04 0.01 0.002 0.51 (0.001) Pb <0.06 0.05 Sb <0.04 0.2 0.002 -0.03 Se <0.04 0.25 0.003 0.12 Si 1.69 - 7.72 0.003 0.77 (0.001) Sn <0.02 0.014 -0.13 V Zn 102 <0.1 0.25 - 267 250 0.96 (0.001) - 2.62 0.1 0.011 0.15 MATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated, contribution to multiple regression of chemical composition on mortality, linear regression value (statistical significance), from Quality Criteria for Water. (U.S. EPA, 1976). -112- Table C-3. Chemical composition of acute bioassay test solutions generated from OBG waste and the correlations between the concentrations of various chemical constituents and mortality in Pimephales promelas . Range of concentration in test solutions (mg/L) Recommended water quality levels (mg/L) a r* Change b P H f Al As B Ba Be Ca Cd Cr Cu Fe Mg Mn Mo Ni Pb Sb Se Si Sn SC\ V Zn 5.0 5.0 <0.06 <0.04 <0.04 <0.004 <0.001 27.0 <0.08 <0.01 <0.004 <0.04 7.44 0.106 <0.01 <0.01 <0.02 <0.04 <0.04 1.67 <0.02 91 <0.1 <0.01 6.7 7.5 0.96 0.17 - 48.6 0.299 0.03 - 173 6.5 - 9.0 C 6.5 - 9.0 C 1.0 0.05 25 2.5 0.055 16 0.001 0.25 0.05 0.25 87 0.1 7 0.01 0.05 0.2 0.25 250 0.673 0.115 0.058 <0.001 0.006 0.013 0.003 0.014 0.005 0.051 0.012 0.005 0.011 -0.82 (0.001) -0.93 (0.001) 0.63 (0.001) 0.52 (0.01) 0.26 0.58 (0.001) 0.00 0.26 0.26 -0.02 0.69 (0.001) 0.78 (0.001) 0.75 (0.001) 1.33 0.1 0.008 0.54 "^lATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated contribution to multiple regression of chemical composition on mortality. linear regression value (statistical significance). from Quality Criteria for Water. (U.S. EPA, 1976). -113- Table C-4. Chemical composition of acute bioassay test solutions generated from OBG waste and the correlations between the concentrations of various chemical constituents and mortality in Physa anatina . Range of concentration in test solutions (mg/L) Recommended water quality levels (mg/L) a r 2 Change b r c P H £ 4.6 - 4.8 6.5 - 9.0 d 0.118 -0.34 (0.05) P H f 4.8 - 5.1 6.5 - 9.0 d 0.023 -0.32 (0.05) Al 0.69 - 2.73 1.0 0.182 0.66 (0.001) As <0.06 0.05 B 0.27 - 0.47 25 0.208 0.39 (0.01) Ba <0.02 2.5 0.028 0.46 (0.01) Be <0.02 0.055 Ca 61.0 - 73.1 16 0.033 0.26 Cd <0.01 0.001 0.007 0.48 (0.01) Cr <0.02 0.25 0.116 0.36 (0.05) Cu <0.01 - 0.04 0.05 0.04 1 -0.06 Fe <0.02 - 0.08 0.25 0.085 0.31 (0.05) Mg 9.20 - 10.5 87 <0.005 0.37 (0.05) Mn 0.3 - 0.4 0.1 0.017 0.47 (0.01) Mo <0.02 7 0.033 0.20 Ni 0.03 - 0.05 0.01 <0.001 0.47 (0.01) Pb <0.02 0.05 Sb <0.02 0.02 0.027 -0.16 Se <0.05 0.25 <0.001 0.21 Si 5.74 0.006 0.44 (0.01) Sn <0.02 S0 4 245 - 315 250 V <0.03 Zn 0.52 - 1.40 0.1 0.013 0.51 (0.001) flATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated contribution to multiple regression of chemical composition on mortality, linear regression value (statistical significance), from Quality Criteria for Water. (U.S. EPA, 1976). -114- Table C-5. Chemical composition of acute bioassay test solutions generated from WSNG waste and the correlations between the concentrations of various chemical constituents and mortality in Lepomis cyanellus . Range of concentration in test solutions (mg/L) Recommended water quality levels (mg/L) a r 2 Change b r c P^ 4.0 - 5.2 6.5 - 9 .0d 0.580 -0.76 (0.001) P H f 3.9 - 5.6 6.5 - 9 .0 d <0.001 -0.82 (0.001) Al <0.04 - 1.20 1.0 0.179 0.93 (0.001) As <0.06 0.05 0.069 -0.26 B <0.08 25 Ba <0.01 2.5 0.018 -0.03 Be <0.02 0.055 0.010 0.26 Ca 13.2 - 16.5 16 0.002 0.72 (0.001) Cd <0.01 0.001 0.014 0.26 Cr <0.02 0.25 Cu <0.02 0.05 0.031 0.54 (0.01) Fe 0.38 - 1.10 0.25 0.047 0.90 (0.001) Mg 7.71 - 8.33 87 0.027 0.73 (0.001) Mn 0.11 - 0. 17 0.1 0.001 0.76 (0.001) Mo <0.02 7 Ni 0.04 - 0.06 0.01 0.015 0.72 (0.001) Pb <0.04 0.05 Sb <0.04 0.02 Se <0.04 0.25 Si 1.30 - 2.08 <0.001 0.81 (0.001) Sn <0.02 S0 4 88 - 128 250 0.83 (0.05) V <0.04 Zn 0.07 - 1.33 0.1 0.003 0.03 MATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated, contribution to multiple regression of chemical composition on mortality. linear regression value (statistical significance), from Quality Criteria for Water. (U.S. EPA, 1976). -115- Table C-6. Chemical composition of acute bioassay test solutions generated from WSNG waste and the correlations between the concentrations of various chemical constituents and mortality in Pimephales promelas . Range of concentration in test solutions (mg/L) Recommended water quality levels (mg/L) a r 2 Change rC PH £ 5.0 - 5 .7 6.5 - 9.0 d 0.219 -0.47 (0.001) P H f 5.3 - 7 .1 6.5 - 9.0 d 0.122 -0.80 (0.001) Al <0.04 - .36 1.0 0.001 0.55 (0.001) As <0.06 0.05 B <0.08 25 0.018 0.20 Ba <0.01 2.5 Be <0.02 0.055 Ca 12.1 - 15 .6 16 0.011 0.23 Cd <0.01 0.001 Cr <0.02 0.25 Cu <0.03 0.05 0.047 0.51 (0.001) Fe <0.08 0.25 0.130 0.44 (0.001) Mg 6.31 - 8, .54 87 0.105 -0.01 Mn 0.07 - 0, ,16 0.1 0.010 0.38 (0.01) Mo <0.03 7 Ni <0.02 - 0. ,05 0.01 0.021 0.10 Pb <0.03 0.05 0.090 -0.30 Sb <0.05 0.2 Se <0.05 0.25 Si 0.99 - 1. ,50 0.010 -0.05 Sn <0.03 0.013 0.18 SO,, 83 - 92 250 V <0.03 Zn 0.07 - 1. ,07 0.1 0.093 0.45 (0.001) MATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated, contribution to multiple regression of chemical composition on mortality, linear regression value (statistical significance), from Quality Criteria for Water. (U.S. EPA, 1976). -116- Table C-7. Chemical composition of acute bioassay test solutions generated from WSNG waste and the correlations between the concentrations of various chemical constituents and mortality in Physa anatina . Range of concentration in test solutions (mg/L) Recommended water quality levels (mg/L) 3 Change PH ± pH f Al As B Ba Be Ca Cd Cr Cu Fe Mg Mn Mo Ni Pb Sb Se Si Sn SO4 V Zn 3.5 4.3 <0.04 - <0.06 <0.07 - 0.002 - <0.003 - 16.9 <0.06 <0.01 <0.006 - 0.26 - 7.45 - 0.13 - <0.009 0.04 - <0.03 <0.02 <0.06 1.54 - <0.03 104 <0.06 0.11 4.3 6.0 1.63 0.13 0.005 0.009 27.3 0.029 1.63 9.24 0.24 0.08 2.90 - 141 6.5 - 9.0 6.5 - 9.0° 1.0 0.05 25 2.5 0.055 16 0.001 0.25 0.05 0.25 87 0.1 7 0.01 0.05 0.2 0.25 250 0.267 0.015 0.242 0.005 0.022 0.198 0.015 0.002 0.018 0.013 0.003 0.001 0.001 0.130 <0.001 0.84 0.1 0.011 -0.52 (0.001) -0.66 (0.001) 0.93 (0.001) 0.64 (0.001) 0.62 (0.001) 0.70 (0.001) 0.68 (0.001) -0.14 0.67 (0.001) 0.87 (0.001) 0.64 (0.001) 0.68 (0.001) 0.42 (0.05) -0.34 0.71 (0.001) 0.36 (0.05) 0.15 MATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated, contribution to multiple regression of chemical composition on mortality, linear regression value (statistical significance), from Quality Criteria for Water (U.S. EPA, 1976). -117- Table C-8. Chemical composition of acute bioassay test solutions generated from WSOG waste and the correlations between the concentrations of various chemical constituents and mortality in Daphnia magna . Range of concentration in test solutions (mg/L) Recommended water quality levels (mg/L) a Change pH. P H f Al As B Ba Be Ca Cd Cr Cu Fe Mg Mn Mo Ni Pb Sb Se Si Sn S0„ V Zn 4.9 - 7.0 4.9 - 7.8 <0.06 - 0.18 <0.04 <0.04 <0.001 <0.001 7.70 - 9.16 <0.01 <0.01 <0.004 <0.02 - 10.1 6.30 - 6.84 <0.005 - 0.042 <0.01 <0.01 <0.02 <0.02 <0.04 0.40 - 40 <0.02 52 - 85 <0.1 <0.01 - 0.37 6.5 - 9.0 C 6.5 - 9.0 C 1.0 0.05 25 2.5 0.055 16 0.001 0.25 0.05 0.25 87 0.1 7 0.01 0.05 0.2 0.25 250 0.672 0.080 <0.001 -0.82 (0.001) -0.94 (0.001) 0.41 (0.01) <0.001 0.04 0.020 -0.28 (0.05) 0.002 0.26 (0.05) 0.1 0.029 0.83 (0.001) 0.013 0.09 <0.001 0.32 (0.05) 0.006 -0.03 0.086 -0.15 0.004 -0.15 0.002 0.38 (0.01) 0.73 (0.01) <0.001 0.01 0.015 0.06 ^lATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated, contribution to multiple regression of chemical composition on mortality. linear regression value (statistical significance). from Quality Criteria for Water (U.S. EPA, 1976). ■118- Table C-9. Chemical composition of acute bioassay test solutions generated from WSOG waste and the correlations between the concentrations of various chemical constituents and mortality in Lepomis cyanellus . Range of concentration Recommendi ad water in test solutions quality Levels r 2 (mg/L) (mg/L) a Change b r c pHi 4.6 5.9 6.5 - 9 .0 d 0.556 -0.75 (0.001) P H f 4.5 5.5 6.5 - 9 .0 d 0.014 -0.63 (0.001) Al <0.06 - 0.60 1.0 0.059 0.71 (0.001) As <0.06 0.05 0.035 0.39 (0.05) B <0.08 25 0.022 0.28 Ba <0.002 2.5 0.044 0.39 (0.05) Be <0.003 0.055 0.001 0.60 (0.001) Ca 8.33 - 9.18 16 0.007 0.70 (0.001) Cd <0.006 0.001 Cr <0.01 0.25 Cu <0.006 - 0.017 0.05 0.066 0.42 (0.05) Fe 3.92 - 11.3 0.25 0.004 0.87 (0.001) Mg 6.12 - 6.82 87 0.008 0.09 Mn 0.011 - 0.037 0.1 0.035 0.67 (0.001) Mo <0.009 7 Ni <0.02 0.01 0.083 0.40 (0.05) Pb <0.03 - 0.08 0.05 0.024 0.49 (0.001) Sb <0.04 0.2 Se <0.06 0.25 Si 0.25 - 0.54 0.006 0.75 (0.001) Sn <0.04 S0 4 73 - 101 250 V <0.06 0.015 0.28 Zn <0.01 - 0.15 0.1 <0.001 0.15 MATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated, contribution to multiple regression of chemical composition on mortality, linear regression value (statistical significance), from Quality Criteria for Water. (U.S. EPA, 1976). -119- Table C-10. Chemical composition of acute bioassay test solutions generated from WSOG waste and the correlations between the concentrations of various chemical constituents and mortality in Pimephales promelas . Range of concentration in test solutions (mg/L) Recommended water quality levels (mg/L) a r 2 Change b r c PH £ 4.7 - 5.4 6.5 - 9 .o d 0.349 -0.59 (0.01) P H f 4.9 - 6.7 6.5 - 9 .o d 0.161 -0.87 (0.001) Al <0.06 1.0 0.058 0.34 As <0.06 0.05 B <0.07 25 Ba <0.004 2.5 <0.001 -0.28 Be <0.003 0.055 Ca 7.99 - 9.70 16 0.010 0.37 Cd <0.006 0.001 Cr <0.01 0.25 Cu <0.006 0.05 <0.001 -0.19 Fe <0.02 - 0.08 0.25 0.012 -0.14 Mg 5.75 - 7.31 87 0.015 -0.28 Mn <0.008 0.1 Mo <0.009 7 Ni <0.01 0.01 Pb <0.03 - 0.04 0.05 0.105 0.62 (0.01) Sb <0.02 0.2 Se <0.06 0.25 Si 0.20 - 0.38 0.171 0.33 Sn <0.03 SO,, 69 77 250 V <0.06 Zn 0.03 - 0.79 0.1 0.068 0.07 MATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated, contribution to multiple regression of chemical composition on mortality, linear regression value (statistical significance). from Quality Criteria for Water. (U.S. EPA, 1976). -120- Table Oil. Chemical composition of acute bioassay test solutions generated from WSOG waste and the correlations between the concentrations of various chemical constituents and mortality in Physa anatina , Range of concentration in test solutions (mg/L) Recommended water quality levels (mg/L) a r 2 Change^ r c ?\ 3.4 - 4.9 6.5 - 9 • d 0.147 -0.48 (0.001) pH f 3.9 - 6.2 6.5 - 9 . d 0.018 -0.74 (0.001) Al <0.06 - 1.13 1.0 0.364 0.87 (0.001) As <0.06 0.05 0.002 -0.02 B <0.07 25 Ba <0.003 - 0.009 2.5 <0.001 0.38 (0.01) Be <0.001 - 0.009 0.055 0.040 0.35 (0.05) Ca 12.3 - 19.8 16 0.003 0.73 (0.001) Cd <0.006 0.001 Cr <0.01 - 0.02 0.25 0.060 0.34 (0.05) Cu <0.006 - 0.022 0.05 <0.001 0.50 (0.001) Fe 4.15 - 18.5 0.25 0.002 0.72 (0.001) Mg 6.69 - 7.28 87 0.005 0.70 (0.001) Mn 0.013 - 0.065 0.1 0.019 0.49 (0.001) Mo <0.009 7 Ni <0.02 0.01 0.022 0.24 Pb <0.03 - 0.08 0.05 <0.001 0.27 Sb <0.02 0.2 0.008 -0.07 Se <0.06 0.25 Si 0.34 - 0.88 <0.001 0.78 (0.001) Sn <0.03 S0 4 83 - 161 250 0.83 (0.001) V <0.06 Zn <0.01 - 2.06 0.1 0.147 0.53 (0.001) MATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated, contribution to multiple regression of chemical composition on mortality, linear regression value (statistical significance), from Quality Criteria for Water. (U.S. EPA, 1976). -121- Table C-12. Chemical composition of acute bioassay test solutions generated from WSSL waste and the correlations between the concentrations of various chemical constituents and mortality in Daphnia magna . Range of concentration in test solutions (mg/L) Recommended water quality levels r 2 (mg/L) a Change b ph. pH f Al As B Ba Be Ca Cd Cr Cu Fe Mg Mn Mo Ni Pb Sb Se Si Sn SO;, V Zn 5.2 5.9 6.5 - 9.0 d 5.9 7.0 6.5 - 9.0 d <0.06 1.0 <0.06 0.05 <0.07 25 <0.002 - 0.004 2.5 <0.003 0.055 7.90 - 8.62 16 <0.006 - 0.015 0.001 <0.01 0.25 <0.006 0.05 <0.08 0.25 1.65 - 2.03 87 0.046 - 0.068 0.1 <0.009 7 <0.01 - 0.02 0.01 <0.03 0.05 <0.02 0.2 <0.06 0.25 0.21 - 0.52 <0.03 46 74 250 <0.06 0.01 - 0.10 0.1 0.252 0.147 0.008 0.034 0.036 0.258 0.073 0.040 <0.001 0.054 0.001 -0.51 (0.001) -0.78 (0.001) 0.10 -0.17 0.44 (0.01) 0.52 (0.001) 0.26 0.22 0.51 (0.001) 0.31 (0.05) 0.35 (0.05) 0.011 0.22 "MATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated, contribution to multiple regression of chemical composition on mortality, linear regression value (statistical significance), from Quality Criteria for Water. (U.S. EPA, 1976). -122- r 2 Change b r c 0.451 -0.67 (0.001) <0.001 -0.76 (0.001) 0.364 0.93 (0.001) 0.039 -0.13 <0.001 -0.25 0.043 -0.14 <0.001 0.30 Table C-13. Chemical composition of acute bioassay test solutions generated from WSSL waste and the correlations between the concentrations of various chemical constituents and mortality in Lepomis cyanellus . Range of concentration Recommended water in test solutions quality levels (mg/L) (mg/L) a pH. 3.9 - 5. 1 6.5 - 9.0 d l pH f 3.9 - 5.8 6.5 - 9.0 d Al <0.06 - 0.33 1.0 As <0.04 0.05 B <0.04 25 Ba <0.001 - 0.002 2.5 Be <0.001 0.055 Ca 8.52 - 9.10 16 Cd <0.08 0.001 Cr <0.03 0.25 Cu <0.03 0.05 Fe 0.10 - 0.28 0.25 Mg 6.82 - 7.12 87 Mn 0.05 - 0.09 0.1 Mo <0.02 7 Ni <0.03 0.01 <0.001 0.15 Pb <0.04 0.05 Sb <0.03 0.2 Se <0.04 0.25 <0.001 -0.17 Si 0.47 - 0.74 0.005 0.73 (0.001) Sn <0.02 S0 4 73 - 89 250 V <0.03 Zn <0.01 - 1.51 0.1 0.010 -0.11 MATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated, contribution to multiple regression of chemical composition on mortality, linear regression value (statistical significance), from Quality Criteria for Water. (U.S. EPA, 1976). 0.017 0.51 (0.01) 0.021 0.43 (0.05) <0.001 0.36 (0.05) 0.046 0.45 (0.05) -123- Table C-14. Chemical composition of acute bioassay test solutions generated from WSSL waste and the correlations between the concentrations of various chemical constituents and mortality in Pimephales promelas . Range of concentration in test solutions (mg/L) Recommended water quality levels (mg/L)a r 2 Change b r c p R i 4.9 - 5.9 6.5 - 9 .o d 0.348 -0.59 (0.001) pH f 5.5 7.0 6.5 - 9 .o d 0.317 -0.89 (0.001) Al <0.06 1.0 As <0.06 0.05 0.038 -0.16 B <0.07 25 Ba <0.002 2.5 <0.001 -0.02 Be <0.003 0.055 0.018 -0.04 Ca 8.25 - 9.08 16 0.026 -0.08 Cd <0.06 0.001 Cr <0.02 0.25 0.101 -0.16 Cu <0.006 - 0.009 0.05 0.032 0.22 Fe <0.02 - 0.32 0.25 <0.001 -0.31 Mg 6.46 - 7.44 87 <0.001 -0.24 Mn 0.050 - 0.072 0. 1 0.063 0.45 (0.05) Mo <0.02 7 Ni <0.01 - 0.02 0.01 0.028 0.06 Pb <0.04 0.05 Sb <0.03 0.2 Se <0.06 0.25 Si 0.31 - 0.57 0.015 0.68 (0.001) Sn <0.03 S0 4 70 76 250 V <0.06 Zn 0.02 - 0.91 0.1 <0.001 -0.06 T1ATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated, contribution to multiple regression of chemical composition on mortality, c linear regression value (statistical significance). from Quality Criteria for Water. (U.S. EPA, 1976). -124- Table C-15. Chemical composition of acute bioassay test solutions generated from WSSL waste and the correlations between the concentrations of various chemical constituents and mortality in Physa anatina. Range of concentration Recommended water in test solutions quality levels r 2 (mg/L) (mg/L) a Change b r c pH. 3.1 - 3.7 6.5 - 9.0 d 0.204 -0.45 (0.001) pH f 3.2 - 5.0 6.5 - 9.0 d 0.002 -0.69 (0.001) Al <0.06 - 1.01 1.0 0.161 0.90 (0.001) As <0.04 0.05 0.037 -0.17 B <0.04 25 0.047 0.27 (0.05) Ba 0.001 - 0.007 2.5 0.040 0.68 (0.001) Be <0.001 0.055 Ca 12.4 - 23.6 16 * 0.029 0.86 (0.001) Cd <0.01 0.001 Cr <0.01 0.25 Cu <0.03 0.05 Fe 0.21 - 3.66 0.25 Mg 6.25 - 28.2 87 Mn 0.09 - 0.245 0.1 Mo <0.02 7 Ni 0.01 - 0.04 0.01 Pb <0.04 0.05 Sb <0.03 0.2 Se <0.04 0.25 Si 0.71-2.11 Sn <0.02 SO., 77 - 160 250 V <0. 10 Zn <0.01 - 1.68 0.1 MATE values cited from Cleland and Kingsbury (1977) unless otherwise indicated, contribution to multiple regression of chemical composition on mortality, linear regression value (statistical significance), from Quality Criteria for Water. (U.S. EPA, 1976). 0.025 0.50 (0.001) 0.222 0.72 (0.001) 0.001 0.32 (0.05) 0.033 0.25 (0.05) 0.002 0.02 0.019 0.74 (0.001) 0.003 -0.03 0.001 0.85 (0.001) 0.90 (0.001) 0.007 -0.07 0.081 -0.17