PRELIMINARY ASSESSMENT OF SUSPECTED CARCINOGENS IN DRINKING WATER Appendices) INTERIM REPORT TO CONGRESS U S. ENVIRONMENTAL PROTECTION AGtl^ WASHINGTON, DC 20460 JUNE 1975 \ ------- ------- TABLE OF CONTENTS Page I. Inventory of Organics Presently Identified in Drinking Water 1 II. National Organics Reconnaissance Survey 12 III. Organic Chemicals Found in Industrial Effluents 101 IV. Monitoring for Radiation in Drinking Water 123 V. Analysis of Inorganic Chemicals in Water Samples 129 VI. Preliminary Results of Pilot Plants to Remove Water Contaminants 147 VII. Health Effects Caused by Exposure to Drinking Water Contaminants . . 199 ------- APPENDIX I INVENTORY OF ORGANICS PRESENTLY IDENTIFIED IN DRINKING WATER Prepared By Judith L. Mullaney Robert G. Tardiff Water Supply Research Laboratory National Environmental Research Center Office of Research and Development Cincinnati, Ohio ------- INVENTORY OF ORGANICS PRESENTLY IDENTIFIED IN DRINKING WATER The following list of 187 compounds was compiled from an exhaustive search of the chemical literature and from EPA reports generated from the Agency's analytical activities. These compounds were identified from only a handful of public water supplies and do not constitute a defin- itive list of all compounds in all supplies. Because of the restrictive nature of the analytic systems employed to generate these identities, the list also is not inclusive of all compounds present in the water samples analyzed. These identifications represent the result of single or dupli- cate "grab" samples and, consequently, cannot be used to conclude contin- uous occurrence. Likewise, fluctuations in concentrations with time cannot be determined unequivocally from these same samples. The terminology used in the list is not uniform because caution was taken to use the terminology employed by the investigator, regardless of the nomenclature system. For compounds identified by Water Supply Research Laboratory analysts, the chemical abstract names were assigned and used on the list. The concentrations listed are to be considered minimum ranges. The values represent those reported by the analysts; however, in most cases, the values reflect concentrations in the extracted samples with extra- polation to the volume of water employed for the extraction. Since, for most quantification, the recovery data were not generated for the various extraction steps, the values must be considered minimum concentrations in the tap water samples analyzed. This list of organics identified from potable water is being contin- uously updated, and information concerning the chemical properties and toxicity of these agents is being assembled and evaluated. Appendix VII(a) provides additional information about these compounds. ------- ORGANIC COMPOUNDS IDENTIFIED IN DRINKING WATER CO IN THE UNITED STATES (March 15, 1975) Compound 1. acenaphthene 2. acenaphthylene 3. acetaldehyde 4. acetic acid 5. acetone 6. acetophenone 7. acetylene dichloride 8. aldrin 9. atrazine 10. desethyl atrazine 11. barbital 12. behenic acid, methyl ester 13. benzaldehyde 14. benzene 15. benzene sulfonic acid 16. benzoic acid 17. benzopyrene 18. benzothiazole 19. benzothiophene 20. benzyl butyl phthalate 21. bladex 22. borneol 23. bromobenzene 24. bromochlorobenzene 25. bromodichloromethane Range of Minimum Concentrations (yg/D 5.0 0.51 <5.0 0.8 <5.0 References 5 5 24,30 23 1,2,7,20,24,30 1 1 15 13,24 24 14 14 25,31 1,2,7,23,26,28,31 15 5 12 3 5 14,24 25 7,12,19 1,2,7 1,2,7,12 6,20,22,23,24,26,27,28,30,31 ------- 26. bromoform 27. bromoform butanal 28. bromophenyl phenyl ether 29. butyl benzene 30. butyl bromide 31. camphor 32. e-caprolactam 33. carbon dioxide 34. carbpn disulfide 35. carbon tetrachloride 36. chlordan(e) 37. chlordene 38. chlorobenzene 39. 1,2-bis-chloroethoxy ethane 40. chloroethoxy ether 41. bis-2-chloroethyl ether 42. 2-chloroethyl methyl ether 43. chloroform 44. chlorohydroxybenzophenone 45. bis-chloroisopropyl ether 46. chloromethyl ether 47. chloromethyl ethyl ether 48. m-chloronitrobenzene 49.,1-chloropropene 50. 3-chloropyridine 51. o-cresol 52. crotonaldehyde 53. cyanogen chloride 54. cyclopheptanone 55. DDE 56. DDT 57. decane 58. dibromobenzene 59. dibromochloromethane 60. dibromodichloroethane 61. di-t-butyl-p-benzoquinone 0.57-<5.0 <5.0 5 0.07-0.42 1.0-133 0.18-1.58 0.04 0.33 0.23 1,6,24,26,28 23 1,7 1,2,7 23 7,19 14 26 24,30 1,2,23,24,28 15,25 25 1,2,7,23 3 3 1,6,7,12,24,33 12,14 1,2,7,20,21,22,24,27,28,30,31 6 1,6,24,33 7 7 1,7,12 27 1,7 5 31 30 29 15 15,16 23,24 23,24 6,20,23,24,27,28,31 24 24 ------- 62. dibutyl phthalate 63. 1,3-dichlorobenzene 64. 1,4-dichlorobenzene 65. dichldrodifluoroethane 66. 1,2-dichloroethane 67. 1,l-dichloro-2-hexanone 68. 2,4-dichlorophenol 69. dichloropropane 70. 1,3-dichloropropene 71. dieldrin 72. di-(2-ethylhexyl) adipate 73. diethyl benzene 74. diethyl phthalate 75. di(2-ethyl hexyl) phthalate 76. dihexyl phthalate 77. dihydrocarvone 78. di-isobutyl carbinol 79. di-isobutyl phthalate 80. 1,2-dimethoxy benzene 81. 1,3-dimethylnaphthalene 82. 2,4-dimethyl phenol 83. dimethyl phthalate 84. dimethyl sulfoxide 85. 4,6-dinitro-2-aminophenol 86. 2,6-dinitrotoluene 87. dioctyl adipate 88. diphenylhydrazine 89. dipropyl phthalate 90. docosane 91. n-dodecane 92. eicosane 93. endrin 94. ethanol 95. ethyl amine 96. ethyl benzene 0.19 0.01 1.0 8.0 1 36 0.07-8.0 0.003-0.31 1 0.03 0.31 0.03 0.14 0.59 0.27 1.0-3.0 1 0.14 0.01 0.004-0.008 14,23,24 25 1,2,7,23 24 1,2,3,24,27,28 31 5 27 27 15,24,32 25,31 31 14,24 14,24 24 24 12 24 1,7,19 1 5 24,25 1 5 1,7 31 31 24 14 24 14 15,24,25 22,24,30,31 23 1,2,6,7 ------- 97. 2-ethyl-n-hexane - 12 98. cis-2-ethyl-4-methyl-l,3-dioxolane - 9,10 99. trans-2-ethyl-4-methyl-l,3-dioxolane - 9,10 100. o-ethyltoluene 0.04 24 101. m-ethyltoluene 0.01-0.05 25 102. p-ethyltoluene 0.03 24 103. geosmin - 4 104. heptachlor - 15 105. heptachlor epoxide - 15 106. 1,2,3,4,5,7,7-heptachloronorbornene 0.06 24 107. hexachlorobenzene - 1 108. hexachloro-1,3-butadiene 0.06 15 109. hexachlorocyclohexane - 15,16 110. hexachloroethane 4,4 6,24 111. hexachlorophene 0.01 11 112. hexadecane - 14 113. 2-hydroxyadiponitrile - 4 114. indene - 5 115. isoborneol - 7,11 116. isodecane 5.0 23 117. isophorone 2.9 24 118. l-isopropenyl-4-isopropylbenzene - 1 119. isopropyl benzene - 2,7,12 120. limonene 0.03 24 121. p-meth-l-en-8-01 - 1,19 122. methane - 34 123. methanol - 24 124. 2-methoxy biphenyl - 1 125. o-methoxy phenol - 117,19 126. methyl benzoate - 24 127. methyl benzothiazole - 3 128. methyl biphenyl - 1 129. 3-methyl butanal - 24,29 130. methyl chloride - 1,2,7 131. methylene chloride <5 2,20,26,30,31 ------- 132. methyl ethyl benzene <1 31 133. methyl ethyl ketone - 23,24,25 134. 2-methyl-5-ethyl-pyrid1ne - 12 135. methylindene - 12 136. methyl methacrylate <1.0 31 137. methyl naphthalene - 12 138. methyl palmitate - 14 139. methyl phenyl carbinol - 12 140. 2-methylpropanal - 24,30 141. methyl stearate - 14 142. methyl tetracosanoate - 14 143. naphthalene ' 1.0 5,23 144. nitroanisole - 1,11 145. nitrobenzene - 1,7 146. nonane 0.03-10.0 23,24 147. octadecane - 3 148. octane - 14 149. octyl chloride - 23 150. pentachlorobiphenyl - 14 151. pentachlorophenol 0.06 18 1/52. pentachlorophenyl methyl ether - 25 153. pentadecane 0.02 24 154. pentane - 3 155. pentanol 1.0 23 156. phenyl benzoate - 14 157. phthalic anhydride - 14 158. piperidene - 23 159. propanol 1.0 23,30 160. propazine - 25 161. propylamine - 23 162. propylbenzene <5 26 163. simazine - 25 164. 1,1,3,3-tetrachloroacetone - 14 165. tetrachlorobiphenyl - 14 166. 1,1,1,2-tetrachloroethane 0.11 23,24,25 167. tetrachloroethylene 0.41-<5.0 24,27,28 ------- 00 168. tetradecane 169. tetramethyl benzene 170. thiomethylbenzothiazole 171. toluene 172. trichlorobenzene 173. trichlorobiphenyl 174. 1,1,2-trichloroethane 175. 1,1,2-trichloroethylene 176. trichlorofluoromethane 177. 2',4,6-trichlorophenol 178. n-tridecane 179. trimethyl benzene 180. 3,5,5-trimethyl-bicyclo (4,1,0) heptene-2-one 181. trimethyl-trioxo-hexadydro-triazine 182. triphenyl phosphate 183. n-undecane 184. vinyl benzene 185. o-xylene 186. m-xylene 187. p-xylene 1.0-<5.0 1.0 0.12 <5.0 0.02 <5.0 0.05-<5.0 <5.0 24 31 3 1,2,3,6,7,23,24,26 23 14 1,7,24,28 24,27 28 14 24 26,31 29 24 24 24 1,2,6,7,31 6,25,26,31 26 26 ------- REFERENCES 1. Industrial Pollution of the Lower Mississippi River in Louisiana. U. S. Environmental Protection Agency, Region VI, Dallas, Texas. Surveillance and Analysis Division, April 1972. 2. Progress Report: Identification of Hazardous Materials, Lower Mississippi River Basin. U. S. Department of Interior, Federal Water Quality Administration, Lower Mississippi River Basin Field Station, October 1970. 3. Burnham, A. K., Calder, G. V., Fritz, J. S., Junk, G. A., Svec, H. J. and Vick, R. Trace organics in water: their isolation and identifi- cation. Journal American Water Works Assn. 65(11): 722-25, 1973. Iowa State University, Ames, Iowa. 4. Deinzer, Max. Informal memorandum - Recovery from Merrimac River, Lawrence, Massachusetts. WSRL, NERC-Cinti., Dec. 1972. 5. Burnham, A. K., Calder, G. V., Fritz, J. S., Junk, G. A., Svec, H. J. and Willis, R. Identification and estimation of neutral organic contaminants in potable water. Anal. Chem. 44(1) :139-41, 1972 6. Kleopfer, Robert D. and Fairless, Billy J. Characterization of or- ganic components in municipal water supply. Environ. Sci. and Tech. 6_:1036, Nov. 1972. 7- Friloux, James (Acting'Chief). Petrochemical wastes as a water pollution problem in the Lower Mississippi River. Lower Mississippi Basin Office, Water Quality Office, EPA, Baton Rouge, Louisiana, Oct. 1971. (Submitted to Senate Subcommittee on air and water pollution, New Orleans, Louisiana - April 5, 1971.) 8. Tardiff, Robert G. and Deinzer, M. Toxicity of organic .compounds in drinking water. Proceedings of 15th Water Quality Conference, Feb. 7-8, 1973, University of Illinois, pp.23-27. 9. Finger, James H. Chemical Services Branch, Region IV, EPA, Surveil- lance and Analysis Division, March 16, 1973. Correspondence to W. Bowman Crum, Jr. of Water Pollution Control Division, South Carolina Pollution Control Authority, Columbia, S. Carolina 29211. 10. Finger, James H. Chemical Services Branch, Region IV, EPA Surveil- lance and Analyses Division, March 16, 1973. Correspondence (to) Thos. C. Kurimcak, S. Carolina State Board of Health, Columbia, S. Carolina. ------- 11. Buhler, Donald R., Rasmusson, M. E. and Nakahue, H. S. Occurence of hexachlorophene and pentachlorophenol in sewage and water. Envir. Sci. and Tech. 7(10):929-34. Oct. 1973. 12. Miller, S. S. (Mg. Ed.). Are you drinking biorefractories, too? Env. Sci and Tech. 7:14, 1973 13. Kleopfer, Robert D., Kansas (Region VII). Correspondence (to) Dr. L. E. Harris, NERC-Cincinnati, Sept. 19, 1973. EPA-Mass Spec- trometer Users' Group Newsletter #5, Sept. 1973. 14. Deinzer, M., Melton, R., Mitchell, D., Kopfler, F. and Coleman, E. Trace organic contaminants in drinking water; their concentration by reverse osmosis. Presented to Division of Environmental Chem- istry, A. C. S., Los Angeles, Calif., March 1974. 15. Schafer, M. L., Peeler, J. T., Gardner, W. S., Campbell, J. E. Pesticides in drinking water: water from the Mississippi and Missouri Rivers. Env. Sci. Tech. 3(12):1261, 1969. 16. West, I. Pesticides as contaminants. Arch. Environ. Health 9:626, 1964. 17. Weibel, S. R., Weidner, R. B., Cohen., J. M. and Christiansen, A. G. Pesticides and other contaminants in rainfall and runoff. JAWWA 58(8):1075, 1966. 18. Young, Clarence L. California Department of Health. Memo March 22, 1974 (to) Henry Ongerth and Dr. Alice Ottoboni. Cellon treated wood - pentachlorophenol - reservoir covers. 19. Alford, Ann L. Environmental Applications of Advanced Instrumental Analysis: Assistance Project, FY-72. (EPA 660/2-73-013) NERC, Corvallis, Oregon. Washington, D. C., G. P. 0., 1973. 20. Melton, R. (Task 05) Application of mass spectrometry and NMR spectroscopy to identification of organics in drinking water NERC- Cincinnati Quarterly Report for April-June 1974. 21. Kleopfer, Robert D. Kansas (Region VII). Correspondence (to) Dr. W. L. Budde, NERC-Cinti. (Halogenated methanes.) February 12, 22. Bellar, T A., Lichtenberg, J. J and Kroner, R. C. The occurrence ™,n°,n9r?°5™ 1n finisned drinking waters. MDQRL, NERC-Cincinnati JAWHA 66:739, 1974. ' 10 ------- 23. Scheiman, M. A., Saunders, R. A., and Saalfeld, F. E. Organic contaminants in the District of Columbia water supply. Chemistry Division, Naval Research Laboratory, Washington, D. C., 1974. (Submitted to J. of Biomedical Mass Spectrometry.) 24. New Orleans Area Water Supply Study. (Draft Analytical Report) Lower Mississippi River Facility, Slidell, Louisiana, November 1974. 25. Garrison, Arthur W. Technical Assistance Summary: Analysis of New Orleans Drinking Water. EPA Region VI, Task TA 75-03, November 1974. 26. Saunders, R. A., Blackly, C. H., Kovacina, T. A., Lamontagne, R. A., Swinnerton, J. W., Saalfeld, F. E. Identification of volatile organic contaminants in Washington, D. C. municipal water. Naval Research Laboratory, Washington, D. C. 20375. 27. Dowty, Betty; Carlisle, Douglas; and Laseter, John L. Halogenated hydrocarbons in New Orleans water and blood plasma. Science 187 (4171):75-77, Jan. 10, 1975. 28. Melton, R. G. Task 005. Chemical characterization of organics in tap water and tap water concentrates. WSRL, NERC-Cincinnati Quarterly Report, October-November 1974. 29. Lee, Ramon G. EPA Region III. Correspondence to R. W. Ludlow, Jr., Department Health and Mental Hygiene, Baltimore, Md. about low levels of organic compounds analyzed in Annapolis water supply. Jan. 21, 1975. 30. Melton, R. G. and Coleman, Emile. Internal (EPA) memo to Earl McFarren. GC-MS analysis of White House water. (D. C. Water Supply.) Feb. 4, 1975. 31. Snyder, Daniel J., III. EPA Regional Administrator Water Supply Analysis. Press Release delivered at Allegheny Board of Health, January 30, 1975. 32. Thomas, R. F. Identification of organophosphorus compound in water. Environmental Protection Agency, Mass Spectrometer Users' Group Newsletter #14, February 1975, Cincinnati, Ohio 45268. 33. Dressman, R. C. and McFarren, E. F. Detection and measurement of bis(2-chloro-)ethers and dieldrin by gas chromatography. Presented at the 2nd Annual Water Quality Technology Conference of the American Water Works Association, Dallas, December 1-4, 1974. 34. Nordell, E. Water Treatment for Industrial and Other Uses. New York, Reinhold Publishing Corp., 1961. 11 ------- APPENDIX II NATIONAL ORGANICS RECONNAISSANCE SURVEY Prepared by James M. Symons Water Supply Research Laboratory National Environmental Research Center Office of Research and Development Cincinnati, Ohio ------- APPENDIX II NATIONAL ORGANICS RECONNAISSANCE SURVEY Table of Contents Page A. Objectives 18 B. Selection of Cities 18 C. Procedure 18 1. Engineering Evaluation of Treatment Facilities 18 2. Sampling 20 a. Selected Organic Compounds 20 1) Trihalomethanes, Carbon Tetrachloride, 1,2-Dichloroethane 20 2) Polychlorinated biphenyls, Haloethers and Organophosphate pesticides 21 3) Vinyl chloride 21 b. General Organic Parameters 21 1) Non-volatile Total Organic Carbon, Ultra- violet Absorption, Fluorescence 21 c. Comprehensive Organics Analyses 22 1) Organics Purged from Sample 22 2) Organics Extracted from Sample with Solvent 22 3) Organics Adsorbed on Activated Carbon from Sample 22 13 ------- d. Constituents in Drinking Water Regulations ..... 24 1) Inorganics .................. 24 2) Organics - Carbon Adsorbables (CCE-m) ..... 24 3) Pesticides (chlorinated hydrocarbons) and Herbicides ................ 24 3. Analytic Methods .................... 25 a. Selected Organic Compounds ............. 25 1) Chloroform, Bromodichloromethane, Dibromo- chloromethane, Bromoform, Carbon Tetra- chloride, 1 ,2-Dichloroethane ......... 25 2) Polychlorinated biphenyls ........... 29 3) Bis-(2-chloroethyl) ether, Bis-(2-chloro- isopropyl) ether ............... 29 4) Vinyl chloride ................ 29 5) Organophosphate Pesticides .......... 30 b. General Organic Parameters ............. 30 1) Non-volatile Total Organic Carbon ....... 30 2) Ultraviolet Absorption ... ......... 30 3) Fluorescence ..... ............ 30 c. Comprehensive Organic Analyses ........... 30 1) Organics Purged from Sample .......... 31 2) Organics Extracted from Sample with Solvent .................... 36 3) Organics Adsorbed on Activated Carbon from Sample .................... 38 d. Constituents in Drinking Water Regulations ..... 40 1 ) Inorganics .................. 40 2) Organics - Carbon Adsorbable (CCE-m) ...... 40 14 ------- 3) Pesticides (chlorinated hydrocarbons), 2,4-D and Silvex 40 4. Quality Control 42 D. Results 42 1. Source and Treatment Information 44 2. Data from 80 Location Survey 44 a. Raw Water Data 44 b. Finished Water Data 46 1) Organics 46 2) Inorganics 46 3. Confirmation Samples 46 a. Quantitative 46 b. Qualitative 69 4. Comprehensive 5-Location Study 69 a. Groundwater, Miami, Florida 69 1) Selected Compound Analysis 69 2) Organics Purged from Sample 70 3) Organics Extracted from Sample with Solvent 70 4) Organics Adsorbed from Sample by Activated Carbon 70 b. Uncontaminated Upland Water, Seattle, Washington- • 73 1) Selected Compound Analysis 73 2) Organics Purged from Sample 73 3) Organics Extracted from Sample with Solvent- • 73 4) Organics Adsorbed from Sample by Activated Carbon 74 15 ------- c, Raw Water Contaminated by Agricultural Runoff, Ottumwa, Iowa 74 1) Selected. Compound Analysis 74 2) Organics Purged from Sample 75 3) Organics Extracted from Sample with Solvent . . 75 4) Organics Adsorbed from Sample by Activated Carbon 76 d. Raw Water Contaminated by Municipal Discharges, Philadelphia, Pennsylvania 76 1) Selected Compound Analysis 76 2) Organics Purged from Sample 78 3) Organics Extracted from Sample by Solvent ... 78 4) Organics Adsorbed from Sample on Activated Carbon 78 e. Raw Water Contaminated with Industrial Discharges, Cincinnati, Ohio 79 1) Selected Compound Analysis 79 2) Organics Purged from Sample 79 3) Organics Extracted from Sample by Solvent ... 79 4) Organics Adsorbed from the Sample on Activated Carbon 79 E. Discussion 82 1. Are Trihalomethanes Formed by Chlorination and, If So, How Widespread Is Their Occurrence? 82 a. Trihalomethanes 82 b. 1,2-Dichloroethane and Carbon Tetrachloride .... 82 c. Non-Volatile Total Organic Carbon 82 16 ------- Page 2. Influence of Source Type and Treatment Practice on Tribalomethane Formation 84 a. Source Influence 86 b. Treatment Influence 87 1) Chlorination Practice 37 2) Filtration Practice 88 3) Use of Activated Carbon 88 A. Powder 88 B. Granular 88 c. Section Summary 91 3. Alternate Indicators of Organic Contaminant Levels ... 92 4. Organics Found in the 5-Location Study 94 5. Significance of Findings 94 F. Acknowledgements 98 G. References 99 17 ------- NATIONAL ORGANICS RECONNAISSANCE SURVEY A. OBJECTIVES The National Organics Reconnaissance Survey has three major objec- tives. One, is to determine the extent of the presence of the four trihalomethanes, chloroform, bromodichloromethane, dibromochloromethane, and bromoform in finished water, and to determine whether or not these compounds are created by chlorination. A second objective is to deter- mine what effect raw water source, and other water treatment practices have on the formation of these compounds, if they are formed by chlorina- tion. The third objective is to characterize, as completely as possible using existing analytic technology, the organic content of finished drinking water produced from raw water sources representing the major categories in use in the United States today. B. SELECTION OF CITIES For the study of the formation of chlorination by-products, 80 water supplies were chosen to participate in the NORS in consultation with State water supply officials. These 80 supplies were geographically dis- tributed, some being in each of the U.S. EPA's 10 Regions. The supplies were chosen to represent as wide a variety of raw water sources and treatment techniques as possible. Table 1 lists the names of the 80 supplies chosen. Ten of the 80 cities below were chosen as sites for a more compre- hensive survey of the organic content of the finished water. These locations were chosen to represent five major categories of raw water sources. These were: 1) ground water; 2) uncontaminated upland water; 3) raw water contaminated with agricultural runoff; 4) raw water contami- nated with municipal waste; and 5) raw water sources contaminated with industrial discharges. Table 2 lists these ten cities by category. C. PROCEDURE 1. Engineering Evaluation of Treatment Facilities At each of the 80 sites chosen for study, engineers from the U.S. EPA Regional Office visited the water treatment plant and evaluated the facilities. They collected basic information on the raw water source and treatment facilities, which are enclosed in this report. In addition to this information these engineers also determined the dosage of various water treatment chemicals used and their points of application. 18 ------- TABLE 1 1. Lawrence, Massachusetts 41. 2. Waterbury, Connecticut 42. 3. Boston, Massachusetts (MDC) 43. 4. Newport, Rhode Island 44. 5. New York, New York 6. San Juan, Puerto Rico 45. 7. Passaic Valley Water 46. Commission, New Jersey 47. 8. Tom's River, New Jersey 48. 9. Buffalo, New York 49. 10. Rhinebeck, New York 50. 11. Philadelphia, Pennsylvania 51. 12. Wilmington Suburban, 52. Delaware 53. 13. Newark, Delaware (Artesian 54. Water Co.) 55. 14. Washington, District of 56. Columbia 57. 15. Baltimore, Maryland 58. 16. South Pittsburgh, 59. Pennsylvania 60. 17. Strasburg, Pennsylvania 61. 18. Fairfax County Water 62. Authority, Virginia 63. 19. Hopewell, Virginia 64. 20. Huntington, West Virginia 65. 21. Wheeling, West Virginia 66. 22. Miami, Florida 67. 23. Jacksonville, Florida 68. 24. Atlanta, Georgia 69. 25. Owensboro, Kentucky 70. 26. Greenville, Mississippi 27. Chattanooga, Tennessee 71. (Tennessee American Water Company) 72. 28. Memphis, Tennessee 73. 29. Nashville, Tennessee 30. Charleston, South Carolina 74. 31. Cincinnati, Ohio 32. Chicago, Illinois 75. 33. Clinton, Illinois 76. 34. Indianapolis, Indiana 77. 35. Whiting, Indiana 78. 36. Detroit, Michigan 79. 37. Mt. Clemens, Michigan 80. 38. St. Paul, Minnesota 39. Cleveland, Ohio 40. Columbus, Ohio Dayton, Ohio Indiana Hill, Ohio Piqua, Ohio Youngstown (Mahoning Valley San. Dist.) Milwaukee, Wisconsin Oshkosh, Wisconsin Terrebonne Parish, Louisiana Camden, Arkansas Logansport, Louisiana Albuquerque, New Mexico Oklahoma City, Oklahoma Brownsville, Texas Dallas, Texas San Antonio, Texas Ottumwa, Iowa Clarinda, Iowa Davenport, Iowa Topeka, Kansas Cape Girardeau, Missouri Kansas City, Missouri St. Louis County, Missouri Lincoln, Nebraska Grand Forks, North Dakota Denver, Colorado Pueblo, Colorado Huron, South Dakota Salt Lake City, Utah Phoenix, Arizona Tucson, Arizona California Water Project at Coalinga, California Contra Costa County Water District, California Dos Palos, California Los Angeles, California (Owens Aqueduct) San Diego, California (Colorado River Aqueduct) San Francisco, California Seattle, Washington Douglas, Alaska Idaho Falls, Idaho Corvallis, Oregon Illwaco, Washington 19 ------- TABLE 2 Ground Water 1) Miami, Florida 2) Tucson, Arizona Uncontaminated Upland Water 1) Seattle, Washington 2) New York, New York Contamination by Agricultural Runoff 1) Ottumwa, Iowa 2) Grand Forks, North Dakota Contamination by Municipal Waste 1) Philadelphia, Pennsylvania 2) Terrebonne Parish, Louisiana Contamination by Industrial Discharges 1) Cincinnati, Ohio 2) Lawrence, Massachusetts 1 = First Series, sampled in early 1975. 2 = Second Series, to be sampled in the future. 2. Sampling a. Selected Organic Compounds 1) Trihalomethanes. Carbon Tetrachloride, 1,2-Dichloroethane Because the six compounds chosen for study were known to be volatile, a sampling procedure was chosen that would provide for minimum loss of the six compounds from the water to the atmosphere while the sample was in shipment or awaiting analysis by the technique of volatile organic analysis (VOA) (see Section C(3)(2)(l) for analytic technique). The containers chosen were glass 50-ml "Hypo-Vials"* sealed with Teflon faced "Tuf-Bond" discs, both available from Pierce Chemical Co., *Mention of commercial products does not constitute endorsement by U.S. EPA. 20 ------- Rockford, 111. Prior to use, the glass vials were capped with aluminum foil and muffled at 400°C for at least one hour to destroy or remove any organic matter interfering with analysis. The bottles were packed, aluminum foil still in place, along with sufficient discs and aluminum seals (to secure the discs in place), labels and re-usable ice packs in an insulated container, and shipped to the appropriate regional office for sampling. Sufficient materials were provided for taking three raw- and three finished-water samples. In the field the vials were filled bubble-free, to overflowing so that a convex meniscus formed at the top. The excess water was displaced as the disc was carefully placed, teflon side down, on the opening of the vial. The aluminum seal was then placed over the disc and the neck of the vial and crimped into place. A sample taken and sealed in this man- ner was completely headspace-free at the time of sampling. Usually a small bubble would form during shipping and storage, however. The samples were labeled appropriately, repacked with the frozen ice packs in the original insulated container and returned via air mail to the Water Supply Research Laboratory in Cincinnati. After receipt at the laboratory, the samples were refrigerated until analyzed. Samples were collected from the 80 locations during the period late January to end of March 1975. 2) Polychlorinated biphenyls, Haloethers, Organophosphate Pesticides Samples were collected in glass gallon jugs that had been detergent washed, tap water rinsed and muffled at 400°C for 15 minutes in an ultra high temperature oven. Caps were teflon lined. Samples were received over a period of one month, late January to late February 1975 from the First Series of the comprehensive analyses locations (Table 2) and were refrigerated until all could be extracted at the same time. 3) Vinyl chloride Samples for vinyl chloride in raw and finished water were collected using the same procedure described in Section C(2)(a)(l) during the period from the end of January through the end. of February 1975 from the First Series of the comprehensive analysis locations (Table 2). b. General Organic Parameters 1) Non-volatile Total Organic Carbon, Ultraviolet Absorption, Fluorescence One of the sealed bottles of both raw and finished water described in in Section C(2)(a)(1), collected from all 80 locations, was used as the sample for these three parameters. 21 ------- c. Comprehensive Organic Analyses 1) Organics Purged from Sample All samples were collected from a potable water tap in one predeter- mined water plant of each study city (First Series - Table 2). With the exception of Ottumwa, Iowa, the samples for comprehensive volatile organ- ics analyses were taken from the same tap as the samples for other organic analyses in the NORS. Samples were collected between the last of January through the last of February 1975. Prior to sampling, the tap water was allowed to run at a maximum discharge rate for 15 minutes. During sam- pling, the discharge rate was adjusted to avoid agitation of the sample. All samples were collected in glass serum bottles previously muffled at 55C°C for 4 hours, were capped with teflon lined discs, and were sealed with aluminum caps, as described in Section C(2)(a)(l). The vials con- taining samples collected for comprehensive volatile organic analyses were filled completely so that no air would be present; whereas, those for head gas analyses were filled to within 1/4 inch of the disc to allow the escape of volatiles into the head space. Samples were stored and shipped at 4°C and were analyzed 24 to 168 hours after collection. 2) Organics Extracted from Sample with Solvent Samples were collected in glass gallon jugs that had been detergent washed, tap water rinsed and.muffled at 400°C for 15 minutes in an ultra high temperature oven. Caps were teflon lined. Samples were received over a period of one month, late January to late February 1975 from the First Series of the comprehensive analyses locations (Table 2) and were refrigerated until all could be extracted at the same time. 3) Organics Adsorbed on Activated Carbon from Sample A low flow CAM sampling train was used in the First Series of the comprehensive analyses locations (Table 2). Each unit consisted of two 3" diameter pyrex glass columns packed with Filtrasorb 300 granular acti- vated carbon, a teflon-stainless rotameter for flow rate control, and a volume measuring device to count the liters that'passed through the carbon columns (see Figure 1). The end plates, fittings and valves were stainless steel. The gaskets and tubing that contacted the water sampled were teflon or stainless steel. Prior to use in the field the pyrex glass columns were detergent washed, then muffled in an oven at 400°C for 15 minutes to render them organic free. The units were then placed into operation by connecting them to a finished water tap at the site sampled and flushing the fines from the activated carbon columns with twenty liters of finished water. The units were then operated, with continuous flow, 24 hours a day for seven days at a rate of approximately 600 ml/min. The time of sam- pling and flow rate were selected to result in the passage of at least 6000 liters of finished water through the two columns. Because of 22 ------- LEGEND: 1. TEFLON TUBING 2. STAINLESS STEEL AND TEFLON ROTAMETER 3. STAINLESS STEEL TUBING 4. STAINLESS STEEL VALVE 5. TEFLON GASKETS, STAINLESS STEEL SCREENS AND END PLATES 6. 18" LONG x 3" DIAMETER PYREX COLUMN PACKED WITH FILTRASORB 300 ACTIVATED CARBON 7. VOLUME MEASUREMENT CONTROL 8. COUNTER J_S V \ FIGURE 1. CARBON ADSORPTION MONITORING UNIT 23 ------- difficulties with this procedure the 5 locations were sampled in early April 1975, rather than in February, when the other samples were taken. d. Constituents in Drinking Water Regulations 1) Inorganics Four one-quart plastic cubitainers of water were collected at the same place, and at approximately the same time, so as to represent essen- tially one sample. Each was identified by writing the same serial number on the container. To assist the analyst, each container was also identi- fied by writing on the outside the preservative added, i.e., no preserva- tive, HNOo, HgCl2 or NaOH. The amount of preservative added to each quart cubitainer and the analyses carried out on each of the particularly preserved samples is as follows: 1. Trace metals - 1-1/2 ml of concentrated nitric acid. 2. Nitrates, and methylene blue active substances - 1 ml of a 20,000 mg/1 solution of mercury (2.21 g HgCl2 per 100 ml). 3. Cyanide - 1-1/2 ml of 2 N sodium hydroxide. 4. Turbidity, color, pH, chloride, sulfate, fluoride, specific conductance, and total dissolved solids - no preservative added. These samples were collected from all 80 locations from the last week in January through March 1975. 2) Organics - Carbon Adsorbable (CCE-m) The sampler and sampling techniques described in Reference 1 were used to collect samples for carbon-chloroform extract (CCE-m). These samples were collected at the First Series of locations listed in Table 2 from the last week in January through the last week in February 1975. 3) Pesticides (chlorinated hydrocarbons) and Herbicides These analyses were performed on the sample referred to in Section C(2)(a)(2). 24 ------- 3. Analytic Methods a. Selected Organic Compounds 1) Chloroform*, Bromodlchloromethane*, DlbromochTorome^thajie^ Bromoform*, Carbon Tetrachloride**, 1,2-Dichloroethane*** Part I. Routine Analysis. The sample concentration procedure chosen for the initial step of identification and measurement of the six volatile halogenated organics was essentially that of Bellar and Lichtenberg.2 In this procedure, the sample is purged with an inert gas that is passed, in series, through an adsorbant material that traps and concentrates the organic materials of interest. The organics are then desorbed from the trapping material by heating under a gas flow and transferred thusly to the first few millimeters of a cold gas chromatography (GC) column. Separation (chromatography) is then carried out with temperature programming. During this survey, only single column GC was routinely performed, mostly because of the shortness of time for completion of the NORS. A high level of confidence that proper identifications were made was at- tained by use of the Hall Electrolytic Conductivity Detector operated in the specific halogen mode. Further assurance of proper identifications was given by supplementary analysis of 9 each, raw- and finished-duplicate water pairs (from selected locations) on a second column using a micro- coulometric detector operated in the oxidative halogen mode. Finally, the qualitative results of analysis of 15 of the finished water samples were confirmed by GC/MS analysis (see Part II of this Section). Apparatus. The glass purging device and stainless steel traps used in the analyses were fabricated exactly according to Bellar and Lichtenberg.2 The adsorbant material used in the trap was Tenax-GC, 60/80 mesh (Applied Science, State College, Pa., or Alltech Associates, Arlington Heights, 111.) The chromatograph used for analysis was a Varian Model 2100 with one inlet modified to the general configuration of Bellar and Lichtenberg's desorber Number 1. The column used for separation of the six compounds was 12 ft x 2 mm I.D. glass, packed with, Tenax GC, 60/80 mesh. The column effluent was connected via a stainless steel transfer line to a Tracor Model 310 Hall Electrolytic Conductivity Detector (Tracer, Inc., Austin, Texas) for detection and measurement of the compounds. This detector was chosen as the most suitable for the immediate needs of the survey.3 *Selected as possible chlorination by-products. **Se1ected because of known effect on health. ***Selected because presence in previously sampled finished waters. 25 ------- Reagents. Blank water and..water used for dilution of standards was prepared by purging distilled water with helium until no interfering peaks could be detected by use of the complete analytical procedure. Stock standards were prepared with dilutions of 95% ethanol of the test compounds. The appropriate final aqueous dilution was made by 1-10 yg/1 injection of an appropriate stock standard directly through the valve on the 5-ml sampling syringe (see description below) into a blank water contained therein. Procedure. The sealed water sample as received from the field, was heated to 25°C in a water bath. Just prior to the actual analysis, the entire disc-seal combination cap was removed with a "Dekapitator" (Pierce Chemical Co.). Duplicate aliquots from the sample were each taken as follows: A glass 5-ml Luer-Lok syringe (plunger removed) was fitted at the tip with a closed Luer-Lok one-way brass stopcock. The water sample was poured into the back of the barrel of the syringe until the barrel was completely full. The plunger was then quickly inserted into the barrel in such a way as to eliminate air bubbles. The valve was opened momentarily. The plunger was depressed to the 5 ml mark to expel excess sample, whereupon the valve was again closed. Only one of these aliquots was routinely analyzed; the duplicate was simply stored in this configura- tion until the success of the first analysis was assured. The syringe assembly containing the aliquot to be analyzed was con- nected to the Luer-Lok needle that was inserted into the sample inlet of the purging device (the needle was never withdrawn from the septum). At the time of analysis the valve was opened and the sample was expelled from the syringe by depressing the plunger. After this, the valve was closed until purging was complete. After purging, the water (to be discarded) was removed by reversing the above procedure. The technique of purging the sample and desorption of the trap con- tents onto the GC, column were carried out exactly as described by Bellar and Lichtenberg. Purging was for 11 minutes with a helium gas flow of 20 ml per minute. Desorption was for three minutes at 180°C with a flow of helium through the trap onto the GC column of 20 ml per minute (this was in addition to the carrier gas flow). At this time, the GC column was at room temperature. Separation of the compounds was accomplished by first quickly heating the column to 95°C, following with a 15-minute hold, then programming at 2°C per minute to a final temperature of 180°C with a helium carrier flow of 20 ml per minute. Conditions for operation of the detector were those recommended by the manufacturer for optimum performance in the halogen mode. Compounds were identifided according to retention time (measured from beginning of the hold at 95°C) and quantified by comparison of peak heights relative to standards prepared at similar concentration. 26 ------- Retention data and the range of minimum quantifiable concentrations (MQC) encountered for the six compounds during the survey are summarized in Table 3. TABLE 3 CHROMATOGRAPHIC RETENTION AND SENSITIVITY DATA Typical Minimum Quantifiable Concentration (MQC)**, yg/1 Compound CHC13 (CH2C1)2 CC14 CHBrCl2 CHBr2Cl CHBr3 Retention 20 25 27 31 41 49 Time (min.) .3 .8 .7* .8 .2 .7 range obs. 0 0 1 0 0 1 duri .1 - .2 - .0 - .2 - .4 - .0 - ng survey 0.2 0.4 2.0 0.8 2.0 4.0 Retention times given were typical. They varied slightly with aging of the columns and significantly with installation of a replacement column. MQC was not constant throughout the study because of various changes in normal operating parameters. No attempt was made to standardize the MQC; operational parameters were simply adjusted to the optimum for any given day. Part II, Confirmation Analysis. As noted above, to add confidence to the routine analysis for the six chosen volatile halogen containing organics, replicate samples from selected locations were subjected to reanalysis for quantisation on a second GC-Detector system and for quali- tative analysis with a GC/MS system. Table 4 shows the sampling locations of these confirmation samples. Quantitative Analysis The quantitative analysis was similar to Reference 2. The following details describe the specific procedure. analysis. Storage. All samples were stored at 4°C until just prior to *Broad peak not completely resolved from (CH2C1)2. **2% scale deflection. 27 ------- Quantitative Confirmation 1. Waterbury, Connecticut 7. Passaic Valley Water Commission, New Jersey 16. South Pittsburgh, Pennsylvania 30. Charleston, South Carolina 51. Oklahoma City, Oklahoma 60. Kansas City, Missouri 65. Pueblo, Colorado 71. Contra Costa County Water District, California 79. Corvallis, Oregon TABLE 4 Qualitative Confirmation 11. Philadelphia, Pennsylvania 21. Wheeling, West Virginia 22. Miami, Florida 30. Charleston, South Carolina 31. Cincinnati, Ohio 41. Dayton, Ohio 51. Oklahoma City, Oklahoma 55. Ottumwa, Iowa 58. Topeka, Kansas 60. Kansas City, Missouri 66. Huron, South Dakota 71. Contra Costa County Water District, California 72. Dos Palos, California 76. Seattle, Washington 79. Corvallis, Oregon Extraction. Five ml of each sample was purged for 11 minutes with nitrogen flowing at 20 ml/min. The purging device was maintained at 19°C. The sample was introduced into the purging device at 4°C. Therefore, as the sample was purged it warmed up to 19°C at an unknown rate. Concentration. The sample was concentrated using a trap packed with 18 cm of Davison silica gel, grade 15, 35-60 mesh. Desorption took place for 4.0 minutes at 200°C. 28 ------- Analytic Procedure. chromatograph equipped with a specific mode, oxidative) was An Infotronics Model 2400 gas Dohrman microcoulometric detector (halide used to perform the analyses. A stainless steel column packed with Porasil-C coated with Carbowax- 400, 100/120 mesh, 6' long, 0.1 inch I.D. was used to perform the separa- tions. Nitrogen flowing at 50 ml/minute was employed as the carrier gas. The column was programmed over the following conditions: 1) Desorb into column for four minutes at <30°C; 2) heat column to 50°C and hold one minute; and 3) program column to 175°C at 8°/minute. Using the above mentioned conditions the limit of detection for the materials of interest were: chloroform, 0.05 yg/1; bromodichloromethane, 0.1 yg/1; dibromochloromethane, 0.1 yg/1; bromoform, ^5 yg/1; 1,2- dichloroethane, 0.1 yg/1; and carbon tetrachloride, 0.05 yg/1. Methylene chloride was routinely detected; the limit of detection was 0.05 yg/1. Other unknown organohalides were detected; unfortunately their concentra- tions were below the limit of detection for GC/MS identification. By calculating relative retention times it was found that the same unknown organohalides were present in many of the water supplies tested. Qualitative Analysis - GC/MS A Varian aerograph 1400 gas chromatograph with a Finnigan 1015C quadrupole mass spectrometer controlled by a Systems Industries 150 data acquisition system was used to perform the analyses. A glass column packed with Porasil-C coated with Carbowax-400, 100/120 mesh, 6' long x 2 mm I.D. was used to perform the separations. Helium at 30 ml/min was used as the carrier gas. The column was programmed under the following conditions: 1) Desorb into the column for 4 minutes at <30°C; 2) hold at <30°C for one minute; 3) heat column to 100°C and hold for three minutes; and 4) program to 200°C at 8°/min. Mass range scan Integration time Samples/AMU Total Run 20-350 12 1 30 minutes 2) Polychlorinated biphenyls See reference 4. Arochlors 1221, 1232, 1242, 1248, 1245, 1260 and 1016 were sought. 3) Bis(2-chloroethyl) ether and Bis-(2-ch1oroisopropy1) ether See reference 5. 4) Vinyl chloride Vinyl chloride was analyzed using a modified version of Seller's and Lichtenberg's procedure.* Samples were collected in and purged from 29 ------- 70 ml septum sealed vials. This technique was employed to gain greater sensitivity from purging a larger sample and to eliminate losses to the headspace in the sample container. A microcoulometric detection system was employed. A chromasorb 101 column was operated isothermally at 100°C. 5) Organophosphate Pesticides See reference 7. Phosdrin, Thimet, Diazinon, Disulfoton, Dimethoate, Ronnel, Merphos, Malathion, Methyl Parathion, Parathion, DEF, Ethion, Trithion, EPN and Guthion were sought. b. General Organic Parameters c 1) Non-volatile Total Organic Carbon Non-volatile total organic carbon (NVTOC) is determined on an instru- ment made by Phase Separations Ltd., United Kingdom. Samples are acidi- fied with nitric acid, purged with nitrogen gas for about 10 minutes to remove carbon dioxide, then pumped into the instrument at a constant rate of 0.6 ml/minute for about 10 minutes. After water and ammonia are re- moved the non-volatile organic carbon is thermally oxidized to carbon dioxide (002) at 920°C with copper oxide as a catalyst, then reduced to methane (Cfy) at 450°C with nickel in a hydrogen atmosphere. The methane is analyzed with a flame ionization detector. 2) Ultraviolet Absorption See reference 8. 3) Fluorescence The Rapid Fluorometric Method (RFM) as described by Sylvia^ and a fluorescence emission scan was performed. In this latter determination, the excitation and emission slit widths are 12 nm and 16 nm, respectively. The aqueous sample is excited at 310 nm and the fluorescence emission recorded between 370 nm and 580 nm. c. Comprehensive Organic Analyses In an attempt to determine as broad a range of organic compounds as possible in the samples collected from the First Series of the Comprehen- sive Locations (Table II), three different techniques of concentrating the organics were used. Because in all three cases the separation tech- niques involved the use of gas chromatography, only those organics in the water that can be volatilized and passed through the gas chromatograph were determined. This means that an undefined number of organic compounds that were originally in the sample, but non-volatile under the temperature of gas chromatographic conditions, were not determined. Techniques such as high pressure liquid chromatography and others would be needed to be applied to determine organic compounds with these properties. 30 ------- Although three different concentrating techniques were used, they were not mutually exclusive. This means that certain organic compounds originally in the water would be determined by all three techniques. In general, however, one new techniques was designed to determine the lower boiling point (more volatile) organic compounds, that were not too solu- ble in water, while the other two techniques were used to determine or- ganic compounds with higher boiling points. The concentration technique used to determine the lower boiling point organics begins by purging these organics from the liquid sample using helium. The higher boiling point organics were determined, in general, by liquid-liquid extraction with ethyl ether, and by adsorption onto granular activated carbon fol- lowed by desorption with chloroform. Details of all three procedures are contained in the three sub-sections that follow. 1) Organics Purged from Sample Types of volatile organic analyses. Analysis for volatile organics is accomplished by the comparative analysis of three types of samples. These three types include: (a) head gas analysis in which some of the volatiles are allowed to escape into the head space above the water sample, and the gas is removed from the serum bottle and injected directly into a gas chromatography/mass spectrometry (GO/MS) system; (b) direct aqueous injection in which a small aliquot of the water sample is injected directly into a GC/MS system; and (c) active stripping of the organics in which a carrier gas removes the organics from the sample. The compounds then are adsorbed on a porous polymer medium, subsequently desorbed, separated by chromatographic techniques and analyzed with appropriate detectors. Although the method appears to emphasize the more volatile compounds, the ability to identify all "volatile" compounds is not within the scope of the method. Volatility is a chemical characteristic of a relative nature. The compounds amenable to the technique described below are those whose volatility is quite high, whose water solubility is quite low, and whose selective adsorptivity to the trapping medium is relatively high. Consequently, some volatile compounds may not be recovered by this technique. Apparatus Purging Apparatus. The method of Bellar and Lichtenberg6 was applied to the purging of volatile organics from tap water samples. Two modifications were made to the original 5-ml purging device: a scale-up to 140 ml and to 500 ml. The 5-ml instrument was used for quantisation with the gas chromatograph and flame ionization detector. The 140-ml device was employed for quantitative assessment using the gas chromato- graph with the mass spectrometer as detector in order to increase the sensitivity of detection. The 500-ml device was utilized for qualitative analysis only. Samples analyzed in the 500-ml instrument were dechlori- nated prior to analysis. The actual design of the modified purgers is presented below. 31 ------- 140-ml_Purger. This device was built by the Paxton Uoods Glass Shop, Cincinnati, Ohio. The main difference between this device and that described by Bellar and Lichtenberg6 is the capacity -- the original capacity was 5 ml; whereas, the modified version has a capacity of 140 ml. The device has the appearance of a 140-ml gas washing bottle with: 1. A 29/42 ground-glass joint on the top. 2. A 20-mm medium fritted filter disc on the end of the gas tube to disperse the helium gas, an additional 5-mm (I.D.) sample port on the top of the male 29/42 joint. 3. A 6-mm (O.D.) by 9-mm (high) silicone rubber cylindrical in- jection septum fitted inside the injection port. 4. A 10-gauge and 762-mm long stainless steel hypodermic needle to penetrate the cylindrical rubber septum. 5. A stainless steel stopcock with male-female Luer-Lock fittings on the 10-gauge needle. 6. A water jacket surrounding the sample container for temperature control. 7. A 1/4-inch (O.D.) glass tubing on helium inlet and outlet ports. 8. A foam trap on the helium outlet trap. The overall height of the device is approximately 27 cm. 500-mlBurger. The responsibility for the design and con- struction of this device is the same as for the 140-ml device. The 500- ml device is virtually identical to the 140-ml device, except for the higher sample capacity of 500 ml. In addition, the overall height of this device is approximately 44 cm. Specific modification includes a 3-mm (I.D.) by 6-mm (O.D.) by 42.5-cm (long) teflon tubing that was attached to the 10-gauge needle tip to prevent splashing during sample introduction. Trapping Apparatus. The compounds stripped from the water were adsorbed onto a porous polymer, Tenax GC of 60/80 mesh. The size of the adsorbing column was adjusted to complement the size of the 500-ml stripping device. The trap for the 5-ml and 140-ml device is described by Bellar and Lichtenberg6, and modifications of the trap for the 500-ml purger are as follows: 32 ------- 1. The stem is fabricated of 1/4-inch stainless steel tubing. 2. The length of the stem from 1/4-inch female swagelock fitting of the body assembly to the stem tip (trap inlet) is approximately 29 mm. 3. The stem assembly is made from Swagelock part number B-QC6-S-400 and body assembly from Swagelock part number B-QC4-B-400. Desorption Apparatus. Desorption of organics from the three traps was accomplished by heat and the passage of helium gas as described by Bellar and Lichtenberg.6 Three desorption units were utilized to accommodate the three trapping devices. The desorption unit used for quantisation with the flame ionization detector and the unit with the 5-ml trap are identical, respectively, to "desorber 1" and "desorber 2" described by Bellar and Lichtenberg.° The third desorption unit was employed with the 500-ml trap. This unit is composed of Swagelock part B-QC6-B-600 and has a total length of 26 cm. Mass Spectrometry. When the mass spectrometer was employed as a detector, the following chromatographic conditions were established. The chromatographic instrument, the Finnigan 9000, was equipped with one of three columns: (a) ten-foot column packed with Chromosorb 101, (b) ten- foot column packed with Tenax GC, and (c) five-foot column packed with Chromosorb 101. All adsorbants were of 60/80 mesh. Mass spectra were obtained on a Finnigan 1015D quadrupole instrument operating in the electron impact mode, and data were acquired and ana- lyzed with the Systems Industries 150 computer system. Using graphic software, data (i.e., reconstructed gas chromatograms and mass spectra) were outputted on Tektronix 4010 crt data terminal. Operating parameters for the mass spectrometer and the data acquisition system are described below: Mass Spectrometer. The mass spectrometer was operated in the following mode: 1. ionization potential = 70 eV 2. emission current = 500 ya 3. ion energy = 4 V 4. repeller potential = 6 V 5. lens potential = 100 V 6. analyzer temperature = 70 degrees C 7. continuous dynode electron multiplier detector = 2.0 KV 33 ------- 8. analyzer pressure = 5 x 10~6 Torr 9. output preamplifier = 10-7 amperes/V 10. mass range = 10 to 250 amu 11. daily calibrations according to manufacturer's specifications Gas Chromatography. With the flame ionization detector, a Perkin-Elmer model 900 was utilized. Samples were analyzed on two different columns: (a) a six-foot column packed with Chromosorb 101 and (b) a six-foot column packed with Tenax GC. The former allows separation of compounds that elute early; whereas, the latter favors shorter retention of the compounds along with improved peak symmetry for later eluting compounds. Standards of chloroform, bromodichloro- methane, dibromochloromethane, and of compounds identified from the mass spectrometric analyses and capable of yielding uncontaminated peaks with the flame ionization detector were analyzed daily. Data Acquisition System Data acquisition parameters were varied only as to the type of sample analyzed, and not from study city to study city. Four sets of data acquisition parameters were used: (a) one for qualitative head gas analy- ses and direct aqueous injection samples, (b) a second for head gas analy- ses of vinyl chloride, (c) a third for the 500-ml purged sample, and (d) another for all quantitative 140-ml purged samples. 1. For head gas analyses and direct aqueous injections: a. software program = IFSS b. mass range = 26-27, 29-31, 41-64, 72-78, 82-102, 112-133, 146-150 and 239 amu c. maximum repeat count = 4 d. integration time = 17 msec e. repeat count before checking lower threshold = 4 f. lower threshold = 4 g. upper threshold = 1 2. For head gas analyses of vinyl chloride: a. software program = IFSS b. mass range = 27, 61-64, 83, 85 amu 34 ------- c. maximum repeat count = 8 d. integration time = 68 msec e. repeat count before checking lower threshold = 8 f. lower threshold = 4 g. upper threshold = 1 3. For the 500-ml purged samples: a. software program = IFSS b. mass range = 14-16, 19-27, 29-31, 33-240 amu c. maximum repeat count = 4 d. integration time = 3 msec e. repeat count before checking lower threshold = 4 f. lower threshold = 4 g. upper threshold = 4 4. For quantisation using 140-ml purged samples: a. software program = IFSS b. mass range = 26-27, 29-31, 41-102, 112-133, 146-150, 166-177, 239 amu c. maximum repeat count = 4 d. integration time = 4 msec e. repeat count before checking lower threshold = 4 f. lower threshold = 4 g. upper threshold = 1 Reagents. Water low in organic carbon was prepared by purging Mi Hi pore Super Q~pre-distilled water with helium at a rate of 60 ml per minute for 38 hours at 95 degrees C. (Organic-free water was impossible to obtain.) This water was used for blanks and for the preparation of standards. Potassium ferrocyanide was used to eliminate chlorine and chloramines in 500-ml samples to be purged at 95 degrees C. 35 ------- Procedure Purging. Blanks, 140-ml samples, and 500-ml samples were trans- ferred in the following manner: a. inversion of serum bottle, b. penetration of the septum with a 10-gauge hypodermic needle connected to the introduction port of the appropriate purging device, c. penetration of the same septum with a second hypo- dermic needle (20-gauge and 6 inches in length) connected to a helium supply, d. application of gas (helium at a flow rate of 20 ml/min) pressure to force the sample out of the bottle. Organic standards used in the 5-ml and 140-ml purging devices were prepared by a procedure previously described.6 Additional information about the procedures for purging, adsorption, desorption and chromatographic and spectral analyses will be presented in the December 1975 report. 2) Organics Extracted from Sample with Solvent After measuring the pH of the gallon sample, three liters were trans- ferred to a six-liter separatory funnel. Fifty milliliters of ethyl ether were added, and the mixture was shaken for one minute. The sample was then extracted three times with 75 ml portions of methylene chloride, and the extracts were combined in a 300-ml erlenmeyer flask. The pur- pose of the ethyl ether is to improve the extraction efficiency of the more polar compounds like phenols and acids. The combined extract was poured through two inches of anhydrous sodium sulfate in a 19-mm I.D. glass column. As an added precaution, the anhydrous sodium sulfate was prerinsed with 100-ml methylene chloride to remove any impurities. The dried extract was collected in a 500-ml Kuderna-Danish (K-D) flask fitted with a 10-ml ampule graduated in 0.1 ml increments. After the combined extract had filtered through the sodium sulfate, the sodium sulfate was rinsed with 50 ml of acetone. This was done for two reasons: to rinse any residual sample components from the sodium sulfate, and to introduce a nonchlorinated solvent into the sample for GC/MS injection. The pH of the water layer was then adjusted to 2.0 using concentrated HC1 and the above steps repeated. In the first step, it was not necessary to add the ethyl ether a second time. 36 ------- When the second extraction was completed, the pH of the water layer was adjusted to 12.0 using a saturated NaOH solution. Again, the extrac- tion and drying steps were repeated, ignoring the addition of ethyl ether. The three sample extracts were now contained in three K-D flasks: the neutral compounds extracted from a solution of approximately pH 7, the acid compounds extracted from a solution of pM 2, and the basic compounds extracted from a solution of pH 12. The reagent blank was in a separate K-D flask. A Snyder column was fitted to each K-D flask, and the extracts were concentrated on a steam bath to approximately 5 ml. After concentration, the methylene chloride (BP = 39.8°C) was completely removed and the sample was contained in acetone (BP = 56.1°C). The acetone was used because one or two microliters of methylene chloride will cause an excessive increase in the pressure in the mass spectrometer and automatically shut down the system, whereas up to 8 microliters of acetone will not cause this un- desirable situation. The extracts were further concentrated in the ampule to 100 yl in a warm water bath under stream of clean, dry nitrogen with repeated rinsing of the inside of the ampule. Five micro!iter injections were made into the GC/MS. The GC column used in this study is 6 ft by 2 mm I.D., packed with Supelcoport (80/100 mesh) coated with 1.5% OV-17 and 1.95% QF-1. The initial column temperature was 60°C, which was held for 1.5 minutes, then the temperature was programmed at 3° per minute to a final temperature of 220°C which was held for 15 minutes. The total run time was approximately 35 minutes. The sample run was set up as follows: System 150 is on select mode: Cont Calibrate?: No Title: Enter appropriate title Calibration file name: Cal File name: Enter appropriate file name Mass range: 33-450 Integration time: 8 Samples/AMU: 1 Threshold: RT GC Atten: 7 Fast scan opt?: 37 ------- MS range setting?: H Max run time: 35 Delay between scans (sec)?: 3) Organics Adsorbed on Activated Carbon from Sample CAM Carbon Processing. On removal from the sampling sites, the CAM carbon cylinders were drained of excess water, sealed and shipped by com- mercial air carrier to the processing laboratory. The columns were stored at 4°C until carbon processing could be initiated. Columns were opened in a special activated carbon handling room de- signed to minimize the potential for contamination. The activated carbon was transferred to Pyrex glass dishes and dried at 35-38°C for 48 hours under a gentle flow of clean air in a mechanical convection oven. The oven air inlet was equipped with an activated carbon filter to prevent atmospheric contamination. The dried activated carbon was transferred to 220-ml Soxhlet extrac- tors and extracted for 48 hours with chloroform. The chloroform extracts were filtered through solvent-extracted glass fiber filters to remove activated carbon fines and then vacuum concentrated at temperatures not exceeding 27°C in rotary evaporators to final volumes of 30-60 ml. The concentrated extracts were transferred quantitatively to 10-ml ampules, several ampules being required to accommodate each extract. The ampules were purged with dry, clean nitrogen and sealed while the contents were held at -50°C in a cold bath. The filled ampules were maintained under refrigeration (4°C) until shipment to the analytical laboratory by air mail. Gas Chromatography - Mass Spectrometry. Gas chromatography was per- formed using a Varian 1400GC with a flame ionization detector. Carbon chloroform extracts (CCE's) were received in sealed glass ampules from the R. S. Kerr Environmental Research Laboratory. After each CCE volume was measured it was concentrated in a Kuderna-Danish apparatus to about 8 ml. Concentration to a final volume of 6 ml was achieved by blowing a gentle stream of nitrogen over the surface of the extract at room tempera- ture. Since 6,000 liters of water were passed through each filter, the organics in each 6-ml extract are 1 million times more concentrated than in the original water sample. However, the percent adsorption on carbon, percent desorption into the solvent, and percent loss on concen- tration of the solvent are unknown and vary with each individual compound. Therefore, the quantitation of each compound is only approximate and the quantity of each chemical reported can be considered as its minimum concentration. Concentrated extracts were analyzed with a computerized combined gas chromatograph-mass spectrometer (GC-MS) system. A Finnigan 1015 38 ------- quadrupole mass spectrometer was operated in the electron impact mode and data was acquired using a System Industries 150 computer interface. A Varian 1400 gas chromatograph was interfaced directly to the mass spectrometer with a 9-inch stainless steel capillary tube. The gas chromatograph contained a 30-meter by 0.4-mm I.D. glass capillary column (No. 646) coated with Supelco SP-2100 at the Southeast Environmental Research Laboratory. Optimized gas chromatographic conditions included multiple tempera- ture and carrier gas (helium) flow programming. Injection of 0.4 yl of each sample was made with the GC oven door open, the column at room temperature (about 30°C), and the MS pressure at 1.5 x ID'5 torr. The GC oven door was closed 5 minutes after injection and the temperature slowly increased to about 50° over the next 6 minutes. At 11 minutes after injection the oven temperature controller was set at 60°. Two minutes later temperature programming at 2°/min was started. Twenty- three minutes after injection (80°C) the temperature program rate was increased to 6°/min and carrier gas flow was increased to produce a MS pressure of 2.8 x 10~5 torr (previously determined to correspond to a helium flow of 2 cc/min at room temperature). Thirty-three minutes after injection (140°C) the temperature program rate was increased to 10°/min. The final temperature of 250°C was maintained for 20 minutes. Computer-controlled collection of mass spectral data was begun im- mediately after sample injection. To prevent filament damage as solvent entered the MS, the ionization current was shut off 2.5 minutes after injection and turned on again 3.5 minutes after injection. Electron energy was maintained at 70 eV and filament current at 400 ya. A mass spectrum from m/e 41 to 350 was acquired approximately every 2.5 seconds by the PDP-8/e computer. At the end of data acquisition a computer-reconstructed gas chromato- gram was plotted. Sample spectra were then chosen and plotted after appropriate background spectra were subtracted. Spectral matching was performed using the EPA computerized Mass Spectral Search System at the National Institutes of Health in Washington, D.C. Tentative identi- fications of compounds were based on these spectral matches and on inter- pretation of the mass spectra. To confirm these identifications, mass spectra and gas chromatographic retention times of mixtures of standards (when available) were compared with those of sample components. The retention times of these components were calculated relative to camphor because it was present in the CCE blank and therefore in all samples. Camphor also served as the internal standard used for all the standard mixtures. Concentrations were calculated with a computer program that compared the total ion current (TIC) summation of sample component mass spectra with the TIC summation of a known amount of that compound in the standard solution. When a standard was not available, a standard compound of 39 ------- similar molecular structure was used to estimate the quantity of the tentatively-identified sample component. Procej^i__BJLan!ks_. The foregoing discussion of preparation and analytical methods has been concerned with the processing of actual samples. However, to assure that components identified were actually derived from the original samples and were not artifacts,Contaminants, or inherent components deriving from the sampling method itself, the sampling media, commercial solvents, or the sample preparations, it was necessary to process blank samples taken through all stages of the operations in parallel with the actual samples, including washing the sampling activated carbon with activated carbon treated water to remove any water soluble materials. As a consequence of this processing of blanks through the analytical stage, no components could be accepted as deriving from the finished water samples unless these components were not present at a significant level in the blanks relative to the samples. d. Constituents in Drinking Water Regulations 1) Inorganics Analytical methods to determine compliance with the requirements of the regulations shall be those specified in the current (13th) Edition of Standard Methods for the Examination of Water and Wastewater (SMEWW), published by the American Public Health Association,10 and/or Methods for Chemical Analysis of Water and Wastewater (MCAWW), U.S. Environmental Protection Agency ,~1974 J1 except for the following which are either not in the current editions, or are undergoing extensive revision. Arsenic and Selenium. The atomic absorption spectrophotometer method is preferable to the wet chemical procedures in the present edi- tion of SMEWW as these will conserve time and effort in analysis and produce improved sensitivity, see reference 12. This procedure will also appear in the 14th Edition of SMEWW and the 1974 Edition of MCAWW. Cyanide. See reference 13. Mercury. See reference 14. This procedure will appear in the 14th Edition of SMEWW and is the same as that .appearing in MCAWWJ1 2) Organics - Carbon Adsorbable (CCE-m) See reference 1. 3) Pesticides (chlorinated hydrocarbons), 2,4-D and Si 1 vex See references 15 and 16. Table 5 lists all the chlorinated hydro- carbons sought. 40 ------- TABLE 5 Organochlorine Pesticides a BHC PCNB Lindane Dichloran Heptachlor Aldrin Heptachlor Epoxide Endosulfan p,p' DDE Dieldrin Captan Endrin DDT p,p' DDD Mi rex Methoxychlor Tech. Chlordane Toxaphene 41 ------- 4. Quality Control Accuracy To test the accuracy of the method as used by Water Supply Research Laboratory during the survey, a pair of "unknown" standard mixtures was prepared by another EPA laboratory in the following manner: Two different stock solutions each containing all of the compounds of interest were prepared by injecting a known volume of each material into a volumetric flask containing 90 ml of methyl alcohol. After all of the compounds were injected into the flask the mixture was diluted to volume (100.0 ml) and mixed by inverting. Two hundred microliters of the stock solution was then dosed into 1.0 liter of super-Q water and mixed by inverting two times. One-half of the dosed water was then trans- ferred into a 500 ml separatory funnel. Several 60-ml vials were then filled with the mixture and promptly sealed with Teflon septums. The samples were stored at 4°C until delivery to Water Supply Research Laboratory. The blank (Sample D-4) contained only super-Q water. The calculated concentrations of the dosed mixtures, D-2 and D-3, are listed with the analytical results in Table 6. Analysis by the respective laboratories was exactly as described in the Section C(3)(a)(l), for the determination of the six halogenated organic compounds: chloroform, bromodichloromethane, dibromochloro- methane, bromoform, carbon tetrachloride, and 1,2-dichloromethane. Precision To test variability of results during a typical day of analysis, two series of 5 replicate samples were prepared as ten discrete samples in the same manner as standards were prepared throughout the survey. One series was at low concentrations and the other at high concentrations. All of the samples were analyzed exactly as described in Section C (3)(a)(l) for the determination of the six halogenated organic compounds, chloroform, bromodichloromethane, dibromochloromethane, bromoform, carbon tetrachloride, 1,2-dichloroethane. Spiked concentrations and relative standard deviations (a/XAV) are listed in Table 7. D. RESULTS At this time (April 1975), all of the results of the National Organ- ics Reconnaissance Survey are not complete. Work is continuing on several facets of the Survey. For this interim report, all of the results avail- able at the present time will be presented, summarized, and discussed The December 1975 report will contain all of the data. 42 ------- D-3 (Conf. Lab.) D-3 (Conf. Lab.) TABLE 6 DETERMINATION OF ACCURACY Concentration (yg/1) Sample D-2 (True Value) D-2a (Prim. Lab.) D-2b (Prim. Lab.) D-2 (Conf. Lab.) D-2 (Conf. Lab.) D-3 (True Value) D-3a (Prim. Lab.) D-3b (Prim. Lab.) Chloroform 74.6 63 65 59.6 46 46 1,2- Dichloro- ethane 10.1 9 10 5.0 6 5 Carbon tetra- chloride 9.5 9 8 6.4 5 6 Bromo- dichloro- methane 39.6 39 40 23.8 22 23 Dibromo- chloro- methane 23.8 23 23 19.0 14 18 Bromo- form 40.4 40 38 23.2 18 24 D-4b (Blank-Prim. Lab.) 0.2 NF NF NF NF ------- TABLE 7 SPIKED CONCENTRATIONS AND RELATIVE STANDARD DEVIATIONS Compound Cone, (yig/1) Rel. a(%) Cone, (yg/1) Rel. o Chloroform 1 ,2-Dichloroethane Carbon Tetrachloride Bromodi chl oromethane Di bromochl oromethane Bromoform 2. 1. 2. 2. 2. 4. 6 5 14 5 10 20 18 * * 20 30 30 7 * * 7 13 12 *Not determined at high concentrations. 1. Source and Treatment Information At the time of the preparation of this report, engineering data were available on the water supplies of 59 locations. Table 8 shows the per- centages of these 59 locations that used the different categories of sources studied in this investigation. Table 8 also shows the treatment practices of these 59 locations. When all the data are in, a study popu- lation from 25-30 million is expected. Because a major objective of this study was to determine the effect of disinfection practices on the formation of the 4 trihalomethanes, Table 9 shows the distribution of the prechlorination dosages used at the 42 locations where prechlorination was practiced. In 82% of these locations the prechlorination dose was between 1 and 6 mg/1. Table 10 shows the distribution of the concentration of chlorine residual, both free and combined. In general, rather low residuals were present in the finished waters.studied, and at 20% of the locations there was less than 0.4 mg/1 of either free or combined residual. 2. 80 Location Study a. Raw Hater Data The data summarized in Table 11 shows that the six selected compounds measured in the raw water at the 80 locations were mostly absent or present in very low concentrations. One location was receiving water prechlorinated by others and this water did contain some chloroform, bromodichloromethane and dibromochloromethane. Non-volatile total organic carbon determinations were made on each sample but were not reported as they were considered unreliable because of the presence of suspended solids in the samples. This was also true of the ultraviolet absorp- tion and fluorescence data. 44 ------- TABLE 8 SUMMARY OF ENGINEERING DATA (All Percentages are of 59 Locations) Source Ground 24% Lake or Reservoir 37% River 39% Mixed 0% Treatment Prechlorination 71% Filtration 73% Polyelectrolyte 20% Powdered Activated Carbon 22% Granular Activated Carbon 9% Softening Precipitative 17% Zeolite 3% Taste and Odor Control Practiced 37% Note: One location was pre-ozonated and another used ozonation as the only treatment. 45 ------- TABLE 9 PRECHLORINATION DOSAGES (All Percentages of 42 Locations) 0-1 mg/1 10% 1-2 mg/1 29% 2-3 mg/1 8% 3-4 mg/1 17% 4-5 mg/1 14% 5-6 mg/1 14% 6-7 mg/1 2% 7-8 mg/1 0% 8-9 mg/1 2% >10 mg/1 2% Unknown 2% b. Finished Water Data 1) Organics Table 12 summarizes all.of the data on finished water quality from the 80 locations. The ultraviolet absorption and fluorescence data were not presented at this time as their significance, if any, are not currently known. The range of each measurement is noted at the end of the Table. To show the central tendency of the data, Table 13 presents the frequency distribution of the concentrations of six selected organic compounds measured in all 80 locations as well as the concentration of finished water non-volatile total organic carbon. Each of these seven parameters is not evenly distributed over the range but is biased toward the low concentration end of the range. Therefore, high concen- trations of these parameters were a somewhat unusual occurrence in this study. 2) Inorganics Table 14 contains the concentrations of the inorganic substances in the Interim Primary Drinking Water Regulations. Very few locations ex- ceeded the limits. 3. Confirmation Samples a. Quantitative The data presented in Tables 15a and 15b show good quantitative con- firmation of the routine analysis of the six selected compounds in the 46 ------- TABLE 10 CHLORINE RESIDUAL (All Percentages of 56 Locations) Combined Residual - mg/1 0-0.4 63% 0.4-0.8 20% 0.8-1.2 4% 1.2-1.6 2% 1.6-2.0 5% 2.0-2.4 2% 2.4-2.8 4% Free Residual - mg/1 0-0.4 43% 0.4-0.8 20% 0.8-1.2 5% 1.2-1.6 17% 1.6-2.0 4% 2.0-2.4 7% 2.4-2.8 4% Free and Combined Residual Each 0-0.4 mg/1 20% 47 ------- TABLE 11 RAW WATER DATA Bromo- Dibromo 1,2- Carbon dichloro- chloro Bromo Dichloro- Tetra- Chloroform methane methane form ethane chloride yg/1 yg/1 yg/1 yg/1 yg/1 yg/1 Non-Volatile Total* Organic Carbon mg/1 1. Lawrence, Massachusetts <0.1 <0.2 NF NF NF NF 2. Waterbury, Connecticut — NF NF NF NF NF NF 3. Boston, Mass. (Metropolitan Dist. Comn.) NF NF NF NF NF NF 4. Newport, Rhode Island (Plant #1) — NF NF NF NF NF NF 5. New York, New York NF NF NF NF NF NF 6. San Juan, P.R. (Sergio Cuevas Plant) - <0.2 NF NF NF NF NF 7. Little Falls, N.J., Passaic Valley Water Comn. 0.3 NF NF NF <0.2 <2 •8. Toms River, New Jersey 0.4 NF NF NF NF NF 9. Buffalo, New York NF NF NF NF NF NF 10. Rhinebeck, New York 0.3 NF NF NF NF NF 11. Philadelphia, Pa. (Torresdale Plant) 0.2 NF NF NF 3 NF 12. Stanton, ''Delaware, Wilmington Suburban - 0.3 <0.4 NF NF NF NF 13. Newark, Delaware, Artesian Water Company 0.2 NF NF NF NF NF 14. Washington, D.C., Wash. Aqueduct (Dalecarlia Plant <0.2 NF NF NF NF NF 15. Baltimore, Maryland NF NF NF NF NF NF 16. South Pittsburgh, Pa., West. Penn. Water Co., (Hays Mine Plant) — 0.3 NF NF NF NF NF 17. Strasburg, Pennsylvania NF NF NF NF NF NF 18. Annnandale, Va., Fairfax County Water Authority (New Lorton Plant) - <0.2 <0.4 NF NF NF NF 19. Hopewell, Virginia 3.7 2.2 2.1 4.6 3.0 2.0 3.6 <0.05 2.6 3.5 2.6 2.8 3.6 1.8 1.8 0.9 0.2 4.7 (Postponed) ------- Table 11 (Continued) 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. =-30. D31. 32. 33. 34. 35. 36. 37a. 37b. 38. 39. 40. 41. 42. 43. Huntington, West Virginia Wheeling, West Virginia Miami, Florida (Preston Plant) Jacksonville, Florida r Atlanta, Georgia (Chattahoochee Plant) Owensboro, Kentucky Greenville, Mississippi Chattanooga, Tenn., Tenn. American Water Co. Memphis, Tennessee Nashville, Tennessee Charleston, S.C. (Stoney Plant) — Cincinnati, Ohio Chicago, Illinois (South Plant) — Clinton, Illinois Indianapolis, Indiana (White River Plant) — - — Whiting, Indiana ' — Detroit, Michigan (Park Plant) Mt. Clemens, Michigan ' Mt. Clemens, Michigan --' St. Paul, Minnesota Cleveland, Ohio (Division Plant) -- Columbus, Ohio (Dublin Plant) Dayton, Ohio (Ottawa Plant) Indian Hill, Ohio Piqua, Ohio Chloroform uq/1 1 0.2 NF NF <0.2 NF 0.3 0.9 <0.2 <0.1 <0.2 0.5 <0.2/0.4 <0.2 0.1 16 <0.2 NF 0.9 <0.2 NF 0.1 NF <0.2 NF Bromo- dichloro- me thane uq/1 NF NF NF NF NF NF NF NF NF NF NF NF NF/0.5 NF NF 11 NF NF NF NF NF NF NF NF NF Dibromo chloro methane uq/1 NF NF NF NF NF NF NF NF NF NF NF NF NF/NF NF NF 3 NF NF NF NF NF NF NF NF NF Bromo form ug/1 NF NF NF NF NF NF NF NF NF NF NF NF NF/NF NF NF NF NF NF NF NF NF NF NF NF NF 1,2- Dichloro- ethane uq/1 <"<0.3 <0.3 <0.2 NF <0.3 NF NF NF NF NF NF NF NF/NF NF <0.3 NF NF NF <0.2 NF NF NF NF NF NF Carbon Tetra- chloride ug/1 4 NF <2 NF NF NF NF NF NF NF NF 2 NF/NF NF NF NF NF NF NF NF NF NF NF NF NF Non-Volatile Total* Organic Carbon mg/1 2.2 3.2 9.8 2.4 1.3 1.7 3.3 1.1 0.2 1.2 11.4 2.3' 1.9/1.7 7.7 5.1 2.0 2.6 2.0 6.7 (After re- 7.9 placement of 2.2 granular act carbon) 6.8 0.9 0.8 6.0 ------- Table 11 (Continued) Chloroform Bromo- dichloro- methane ug/1 Dibromo chloro methane ug/1 Bromo form ug/1 1,2- Dichloro- ethane ug/1 Carbon Tetra- chloride Non-Volatile Total* Organic Carbon 44. Youngstown, Ohio (Mahoning Valley San. Dist.) NF NF 45. Milwaukee, Wisconsin (Howard Ave. Plant) - <0.2 NF 46. Oshkosh, Wisconsin NF NF 47. Houma, La., Terrebonne Parish Water Works #1 NF NF 48. Camden, Arkansas NF NF 49. Logansport, Louisiana 0.7 NF 50. Albuquerque, New Mexico NF NF 51. Oklahoma City, Okla. (Hefner Plant) NF NF ^ 52. Brownsville, Texas (Plant #2) NF NF » 53. Dallas, Texas (Bachman Plant) <0.1 NF 54. San Antonio, Texas ~ NF NF 55a. Ottumwa, Iowa <0.2 NF 55b. Ottumwa, Iowa NF NF 56. Clarinda, Iowa - <0.2 NF 57. Davenport, Iowa 0.4 NF 58. Topeka, Kansas 0.4 0.8 59. Cape Girardeau, Mo., Mo. Utilities Co. 0.2 NF 60. Kansas City, Missouri NF NF 61. St. Louis, Missouri, St. Louis County Wat. Co., (Central Plant) NF NF 62. Lincoln, Nebraska NF NF 63. Grand Forks, North Dakota NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF 0.2 NF 0.3 NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF NF 4.7 2.4 4.5 5.4 3.1 5.3 <0.05 3.6 4.7 3.4 0.5 4.1 4.9 3.5 6.5 3.4 4.5 3.4 3.4 1.4 9.2 (2/17/75 sample) (4/7/75 sample) ------- Table 11 (Continued) Bromo- dichloro- Chloroform methane Dibromo chloro methane ug/1 Bromo form 1,2- Dichloro- ethane Carbon Tetra- chloride ug/1 Non-Volatile Total* Organic Carbon "ig/1 64. Denver, Colorado (North Side Marston Plant) — <0.2 NF NF NF NF NF 65. Pueblo, Colorado (Gardner Plant) <0.2 NF NF NF NF NF 66. Huron, South Dakota NF NF NF NF NF NF 67. Salt Lake City, Utah (Parleys Plant) 0.2 ' NF NF NF 0.4 NF 68. Tuscon, Arizona (Plant #1) <0.1 NF NF NF NF NF 69. Phoenix, Arizona (Verde Plant) <0.2 NF NF NF NF NF 70. Coalinga, Calif. (Coalinga Mun. Water Plant) <0.2 NF NF NF NF NF 71. Concord, Cal., Contra Costa Cnty. Wat. Dist., (Bellman Plant) - 0.3 0.3 NF NF NF NF 72. Dos Palos, California NF NF NF NF NF NF 73. Los 'Angeles, California (Owen's Aqueduct) - - — <0.1 NF NF NF NF NF 74. San Diego, California (Miramar Plant) NF NF NF NF NF NF 75. San Francisco, Calif. (San Andreas Plant) NF NF NF NF NF NF 76. Seattle, Washington (Cedar River) - <0.2 NF NF NF NF NF 77. Douglas, Alaska NF NF NF NF NF NF 78. Idaho Falls, Idaho <0.2 NF NF NF NF NF 79. Cprvallis, Oregon (Taylor Plant) — NF NF NF NF NF NF 80. Ilwaco, Washington 0.1 NF NF NF NF NF Range <0.1 - 0.9 <0.2-0.8 NF NF <0.2-3 -2-4 2.0 1.8 19.2 1.2 <0.05 1.0 3.7 3.4 4.4 1.2 2.9 1.3 0.9 (End of Dist. System) 3.4 0.5 1.0 7.5 <0.05-19.2 NF - None Found * - May be low because of incomplete combustion of particulates in raw water. ------- TABLE 11 (Cont'd.) SUMMARY OF RAW WATER ANALYSIS Nothing found Chloroform Bromodi chloromethane Di bromochloromethane Bromoform 1,2-Dichloroethane Carbon Tetrachloride Number of Locations 30 45 6 0 0 11 4 Range, yg/1 <0.1 - 0.9 <0.2 - <0.8 <0.2 - 3 <2 - 4 52 ------- Table 12 FINISHED WATER DATA Bromo- Dibromo dichloro- chloro Bromo Chloroform methane methane form yg/1 yg/1 yg/1 yg/1 1,2- Carbon Dichloro- Tetra- ethane chloride yg/1 yg/1 Non-Volatile Total Organic Carbon mg/1 1. Lawrence, Massachusetts 91 9 0.6 NF NF NF 2. Waterbury, Connecticut 93 10 0.6 <1 <0.2 <2 3. Boston, Mass. (Metropolitan Dist. Comn.) 4 0.8 NF NF NF NF 4. Newport, Rhode Island (Plant #1) --- 103 42 13 1 NF NF 5. New York, New York -- 22 7 0.9 NF NF NF 6. San Juan, P.R. (Sergio Cuevas Plant) 47 29 16 2 NF NF 7. Little Falls, N.J., Passaic Valley Valley Water Comn. 59 16 2 NF <0.2 <2 8. Toms River, New Jersey —- 0.6 <0.8 3 NF NF NF 9. Buffalo, New York 10 10 4 NF <0.2 NF 10. Rhinebeck, New York 49 11 1 NF 2 NF 11. Philadelphia, Pa. (Torresdale Plant) 86 9 5 NF 6 NF 12. Stanton, Delaware, Wilmington Suburban 23 11 3 NF <0.4 <2 13. Newark, Delaware, Artesian Water Company 0.5 0.5 1 <1 <0.2 NF 14. Washington, D.C., Wash. Aqueduct (Dalecarlia Plant) 41 8 2 NF <0.3 NF 15. Baltimore, Maryland 32 11 2 NF NF NF 16. South Pittsburgh, Pa., West. Penn. Water Co., (Hays Mine •Plant ) 8 2 0.4 NF NF NF 17. Strasburg, Pennsylvania - — — <0.1 NF NF NF NF NF 18. Annandale, Va., Fairfax County Water Authority (New Lorton Plant) 67 6 <0.6 NF NF NF 19. Hopewell, Virginia 1.6 2.9 2.0 4.1 2.5 2.0 1.9 <0.05 1.7 1.6 1.7 1.8 0.2 1.2 1.2 0.8 0.05 2.7 (Postponed) ------- Table 12 (Continued) Chloroform Bromo- dichloro- methane ug/1 Dibromo chloro methane ug/1 Bromo form 1,2- Dichloro- ethane Carbon Tetra- chloride Non-Volatile Total Organic Carbon mg/1 20. Huntington, West Virginia 23 16 21. Wheeling, West Virginia 72 28 22. Miami, Florida (Preston Plant) 311 78 23. Jacksonville, Florida 9 4 24. Atlanta, Georgia (Chattahoochee Plant) 36 10 25. Owensboro, Kentucky 13 20 26. Greenville, Mississippi 17 • 6 27. Chattanooga, Tenn., Tenn. American Water Co. — 30 9 28. Memphis, Tennessee 0.9 2 29. Nashville, Tennessee - 16 5 30. Charleston, S.C. (Stoney Plant) 195 9 31. Cincinnati, Ohio 45 13 32. Chicago, Illinois (South Plant) 15 10 33. Clinton, Illinois —- 4 0.5 34. Indianapolis, Indiana (White River Plant) 31 8 35. Whiting, Indiana 0.5 0.: 36. Detroit, Michigan (Park Plant) 12 9 37a. Mt. Clemens, Michigan 11 6 37b. Mt. Clemens, Michigan 6 3 38. St. Paul, Minnesota - 44 7 39. Cleveland, Ohio (Division Plant) — 18 9 40. Columbus, Ohio (Dublin Plant) 134 8 41. Dayton, Ohio (Ottawa Plant) 8 8 42. Indian Hill, Ohio - 5 7 43. Piqua, Ohio 131 13 44. Youngstown, Ohio (Mahoning Valley San. Dist. 80 5 5 17 35 2 2 17 3 0.7 1 <0.4 0.8 4 4 NF <2 NF 3 2 2 <2 4 <0.4 11 11 3 NF NF 3 NF NF 3 NF NF NF 0.8 NF NF NF NF NF NF NF NF NF NF NF 4 NF NF NF <0.4 <0.4 <.2 NF NF NF <0.2 <0.4 NF NF NF <0.4 <0.4 NF NF NF 0.4 <0.4 NF NF NF NF <0.2 NF <0.2 NF 3 NF NF NF NF NF NF NF NF NF NF <2 NF NF 2 NF NF NF NF NF NF NF <2 NF NF NF 1.0 1.8 5.4 2.3 0.9 2.0 4.0 0.6 0.2 0.8 4.1, 1.1 1.5 6.7 2.6 1.5 1.4 (After re- 4.4 placement of 1.8 granular act. carbon) 2.3 0.7 0.9 4.2 3.1 ------- Table 12 (Continued) Chloroform ug/1 Bromo- dichloro- methane pg/1 Dibromo chloro methane yg/1 Bromo form yg/1 1,2- Dichloro- ethane yg/1 Carbon Tetra- chloride pq/1 Non-Volatile Total Organic Carbon 45. Milwaukee, Wisconsin (Howard Ave. Plant 9 7 3 NF <0.2 NF 46. Oshkosh, Wisconsin 26 4 <0.4 NF <0.2 NF 47. Hourna, La., Terrebonne Parish Water Works #1 134 32 8 <1 0.2 NF 48. Camden, Arkansas 40 19 7 NF NF NF 49. Logansport, Louisiana 28 39 24 3 NF NF 50. Albuquerque, New Mexico 0.4 1 23 NF NF 51. Oklahoma City, Okla. (Hefner Plant) 44 28 " 20 6 <0.4 <2 52. Brownsville, Texas (Plant #2) 12 37 100 92 NF NF 53. Dallas, Texas (Bachman Plant) 18 4 <2 NF NF NF 54. San Antonio, Texas 0.2 0.9 3 3 NF NF 55a. Ottumwa, Iowa 0.8 NF NF NF NF NF 55b. Ottumwa, Iowa 1 NF NF NF NF NF 56. Clarinda, Iowa 48 19 4 NF NF NF 57. Davenport, Iowa 88 8 <0.6 NF <0.4 NF 58. Topeka, Kansas 88 38 19 5 NF 3 59. Cape Girardeau, Mo., Mo. Utilities Co. 116 21 2 NF 0.3 2 60. Kansas City, Missouri 24 8 2 NF NF NF 61. St. Louis, Missouri, St. Louis Wat. Co., (Central Plant) — 55 13 3 <1 0.4 NF 62. Lincoln, Nebraska 4 6 4 <2 NF NF 63. Grand Forks, North Dakota 3 1 NF NF NF NF 64. Denver, Colorado (North Side Karston Plant) 14 10 3 NF NF NF 65. Pueblo, Colorado (Gardner Plant) --- 2 2 <2 NF NF NF 1.7 3.3 3.2 1.5 3.5 <0.05 2.9 0.5 2.3 2.4 3.0 4.4 2.2 3.6 1.9 2.6 1.4 5.2 1.7 1.6 (2/17/75 sample) (4/7/75 sample) ------- Table 12 (Continued) Bromo- Dibromo dichloro- chloro Bromo Chloroform methane methane form yg/1 yg/1 yg/1 yg/1 1,2- Carbon Dichloro- Tetra- Non-Volatile Total ethane chloride Organic Carbon yg/1 ug/1 mg/1 66. Huron, South Dakota 309 116 67. Salt Lake City, Utah (Parleys Plant) 20 14 68. Tuscon, Arizona (Plant #1) — <0.2 <0.8 69. Phoenix, Arizona (Verde Plant) 9 '' 15 70. Coalinga, Calif. (Coalinga Mun. Water Plant) 16 17 71. Concord, Cal., Contra Costa Cnty. Wat. Dist., (Bollman Plant) 31 18 72. Dos Palos, California 61 53 73. Los Angeles, California (Owen's Aqueduct) — 32 6 74. San Diego, California (Miramar Plant) 52 30 75. San Francisco, Calif. (San Andreas Plant) -- - 41 15 76. Seattle, Washington (Cedar River) -- 15 0.9 77. Douglas, Alaska 40 0.8 78. Idaho Falls, Idaho 2 3 79. Corvallis, Oregon (Taylor Plant) --- 26 -3 80. Ilwaco, Washington - 167 35 Range <0.1-311 NF-116 49 2 17 15 6 34 3 19 4 NF <0.4 3 NF 5 8 NF 13 <4 NF 3 <0.8 NF NF NF NF NF NF-100 NF-92 NF NF 12.2 NF NF 0.9 NF NF <0.05 NF NF 1.0 NF NF 2.4 NF NF 1.9 NF NF 2.9 NF NF 1.3 NF NF 2.8 NF NF 1.6 NF NF 0.9 (End of Dst. System) NF NF 2.8 NF NF 0.3 NF NF 0.4 NF NF 3.1 NF-6 NF-3 <0.05-12.2 NF - None Found ------- TABLE 13 FREQUENCY DISTRIBUTION OF TRIHALOMETHANES FINISHED WATER Bromodichloro- Chloroform methane Dibromo- chloromethane Bromoform of < 10 _ ol < 10 _ Concentration Range, yg/1 NF 0-1 1.1-5 6-10 11-15 16-20 21-25 26-30 31-40 41-50 51-75 76-100 101-150 151-200 201-250 251-300 301-350 % in Range 0 11.3 8.8 8.8 3.8 7.5 5.0 5.0 8.8 10.0 7.5 7.5 2.5 0 0 2.5 Upper Cone. 0 11.3 20.1 28.9 37.7 45.2 50.2 55.2 64.0 74.0 81.5 89.0 97.5 97.5 97.5 100.0 % in Range 2.5 13.8 13.8 32.5 11.3 . 3.8 1.2 5.0 6.3 1.2 1.2 1.2 Upper Cone. 2.5 16.3 30.1 62.6 73.9 82.7 83.9 88.9 95.2 96.4 97.6 98.8 " in Range 11.3 20.0 43.7 5.0 7.5 6.4 1.2 0 2.5 .1-2 0 1-2 . Upper Cone. 11.3 31.3 75.0 80.0 87.5 93.9 95.1 95.1 97.6 98.8 98.8 100.0 % in Range 68.8 20.0 6.3 2.5 1.2 0 0 0 0 0 0 1.2 Upper Cone. 63.8 88.8 95.1 97.6 98.8 98.8 98.8 98.8 98.8 98.8 98.8 100.0 NF = None found. 57 ------- TABLE 13 (Cont'd.) FREQUENCY DISTRIBUTION OF 1,2-DICHLOROETHANE AND CARBON TETRACHLORIDE FINISHED WATER 1,2-Dichloroethane Carbon Tetrachloride Concentration % of % 5 Concentration % of % < yg/1 Total Concentration pg/1 Total Concentration NF <0.2 0.2 <0.3 0.3 <0.4 0.4 2 6 67.5 12.4 1.3 1.3 1.3 11.2 2.4 1.3 1.3 67.5 79.9 81.2 82.5 83.8 95.0 9.7.4 98.7 100.0 NF 87.5 87.5 <2 7.5 95.0 2 2.5 97.5 3 2.5 100.0 NF = None found. 53 ------- TABLE 13 (Cont'd.) FREQUENCY DISTRIBUTION OF NON-VOLATILE TOTAL ORGANIC CARBON FINISHED WATER Concentration % _ Range % in Upper mg/1 <0.05 0.05-0.5 0.6-1.0 1.1-1.5 1.6-2.0 2.1-2.5 2.6-3.0 3.1-3.5 3.6-4.0 4.1-4.5 4.6-5.0 5.1-5.5 5.6-6.0 6.1-6.5 6.6-7.0 7.1-9.0 9.1-11.0 11.1-13.0 Range 4.9 6.2 12.3 13.6 21.0 8.6 12.3 7.4 2.5 6.2 0 2.5 0 0 1.3 0 0 1.2 Concentration 4.9 11.1 23.4 37.0 58.0 66.6 78.9 86.3 88.8 95.0 95.0 97.5 97.5 97.5 98.3 93.8 98.8 100.0 59 ------- Nitrate TABLE 14 Summary of Inorganic Analysis All data in mg/1 Barium Arsenic Selenium Fluoride Cyanide Chromium Silver Lead Cadmium Mercury 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. Lawrence, Massachusetts Waterbury, Connecticut Boston, Mass. (Metropolitan Dist. Comn.) Newport, Rhode Island (Plant #1) New York, New York San Juan, P.R. (Sergio Cuevas Plant) Little Falls, N.J., Passaic Valley Water Comn. Toms River, New Jersey Buffalo, New York Rhinebeck, New York Philadelphia, Pa. (Torresdale Plant) - Stanton, Delaware, Wilmington Suburban Newark, Delaware, Artesian Water Company Washington, D.C., Wash. Aque- duct (Dalecarlia Plant) — Baltimore, Maryland South Pittsburgh, Pa., West. Penn. Water Co. , (Hays Mine Plant) Strasburg, Pennsylvania Annandale, Va., Fairfax County Water Authority (New Lorton Plant) 1. 2. <1 . 5. 1. <1 . 2. 12. 1. 2, 4. 7. NES 4. 3. 3. 12. 2. <.2 !02 <.2 .03 <.2 <.05 <.2 <.05 .06 <.2 <.05 <.05 <.05 <.05 <.05 <.2 .08 <.05 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 .1 .9 .1 1.3 <. ] .5 .1 <.l 1.5* <.l .9 1.1 1.3 1.0 1.1 1.4 .1 .8 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 e.02 <.02 <.02 <.02 <.02 <.02 <.005 <.01 <.005 ^.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.005 <.005 <.005 <.005 <.005 .020 .020 <.005 <.005 .025 .014 .040 .005 .008 .018 <.005 '.010 <.002 <.002 «.002 <.002 «.002 <.002 <.002 <.002 <.002 <.002 <.002 <.002 .003 <.002 <.002 <.002 <.002 <.002 <.0005 <.OOC5 <.0005 <.0005 .0006 .0015 <.0005 <.0005 <.0005 <.0005 <.0005 .0022 .0026* <.0005 <.0005 <.0005 <.0005 <.0005 CTl O ------- 'able 14 (Continued) Nitrate AsN03 Barium Arsenic Selenium Fluoride Cyanide Chromium Silver Lead Cadmium Mercury ,005 ,005 ,005 .005 ,005 ,005 .005 <.05 <.005 .005 .005 .005 .005 .005 .005 .005 <.005 <.005 9. Hopewell, Virginia !0. Huntington, West Virginia — 2. <.2 <.005 <.005 II. Wheeling, West Virginia 3. <.2 <.005 <.005 12. Miami, Florida (Preston Plant) <1. .04 13. Jacksonville, Florida <1. <.05 !4. Atlanta, Georgia (Chatta- hoochee Plant) 1. <.05 15. Owensboro, Kentucky <1. .04 16. Greenville, Mississippi <1. .04 17. Chattanooga, Tenn., Tenn. American Water Co. 1. <.05 !8. Memphis, Tennessee <1. <.05 !9. Nashville, Tennessee 2. <.2 ,!0. Charleston, S.C. (Stoney Plant) - !1. Cincinnati, Ohio 2. !2. Chicago, Illinois (South Plant) <1. <.05 <.005 !3. Clinton, Illinois <1. .15 .013 !4. Indianapolis, Indiana (White River Plant) - 5. .03 <.005 55. Whiting, Indiana 2. <.2 <.005 J6. Detroit, Michigan (Park Plant) — - 1. <.2 <.005 J7. Mt. Clemens, Michigan <1. .06 <.005 38. St. Paul, Minnesota 2. <.2 <.005 39. Cleveland, Ohio (Division Plant) - <1. <-2 10. Columbus, Ohio (Dublin plant) — 15. <.2 <.005 <.005 11. Dayton, Ohio (Ottawa Plant)— 5. .04 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 .1 1.4 1.1 .7 .7 1.3 .5 1.0 <.l 1.3 1.0 1.0 1.9* .1 1.1 1.3 1.3 .1 1.4 .2 <.02 <.02 .02 .02 .02 .02 .02 .02 .02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.005 <.005 .005 .005 .005 .005 .005 .005 .005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 .070* <.002 =.0005 <.005 <.002 <.0005 <.005 .016 <.005 .012 .029 <.005 <.005 <.005 .010 <.002 <.002 .00095 .00185 .025 <.002 <.0005 <.005 «.002 <.0005 <.005 <.002 <.0005' .010 <.002 .0028* .020 <.002 <.0005 .010 <.002 <.0005 <.013 <.002 <.0005 <.002 <.002 <.002 <.002 <.002 <.002 <.002 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 .018 <.002 <.0005 .020 <.005 <.002 <.002 <.0005 <.0005 ------- Table 14 (Continued) Nitrate AsN03 Barium Arsenic Selenium Fluoride Cyanide Chromium Silver Lead Cadmium Mercury 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55a. 56. 57. 58. 59. 60. 61. 62. 63. .'•"('. Indian Hill, Ohio Piqua, Ohio Youngstown, Ohio (Mahoning Valley San. Dist) Milwaukee, Wisconsin (Howard Ave. Plant) Oshkosh, Wisconsin Houma, La., Terrebonne Parrish Water Works #1 — Camden, Arkansas Logansport, Louisiana Albuquerque, New Mexico Oklahoma City, Okla. (Hefner Plant) Brownsville, Texas (Plant #2) Dallas, Texas (Bachman Plant) - — r San Antonio, Texas Ottumwa, Iowa Clarinda, Iowa Davenport, Iowa Topeka, Kansas Cape Girardeau, Mo., Mo. Utilities Co. Kansas City, Missouri St.. Louis, Missouri, St. Louis County Wat. Co. (Central Plant) Lincoln, Nebraska Grand Forks, North Dakota — Denver, Colorado (North Side Marston Plant) 5. 8. 1. 1. 2. < i . 1. < i . < i . < 1 . 2. 4. 8. 1. < 1 . 1. 5. 3. 3. <1 . 2. <1. <.05 .03 <.05 .04 .02 .1 <.05 < .2 .04 <.05 < .2 < .2 .07 <.05 <.05 < .2 .08 .03 .01 <.2 < .2 <.2 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 .01 <.005 <.005 <.005 <.005 <.005 <.005 <.005- <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 1.4 .2 1.0 .8 1.0 .1 1.1 .1 1.3 .5 1.0 0.1 1.3 .3 1.1 1.2 1.1 .1 1.0 1.1 1.3 1.3 <.02 <.02 <.02 <.02 '.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.on <.02 <.02 <.02 <.02 <.02 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.01 .016 <.01 <.005 <.01 .008 <.01 .024 <.01 <.005 <.01 .026 <.01 .013 <.01 .026 <.01 <.005 <.01 .017 <.01 .014 <.01 <.005 •=.01 .023 <.01 .025 <.01 .012 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.002 <.002 <.002 <.002 <.002 <.002 .<.002 <.002 <.002 <.002 <.002 <.002 <.002 <.005 <.002 <.002 <.002 <.002 <.002 <.002 <.002 <.002 <.0005 <.00p5 <.0005 <.00b5 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 NES <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 en ro ------- Table 14 (Continued) Nitrate AsN03 Barium Arsenic Selenium Fluoride Cyanide Chromium Silver Lead Cadmium Mercury 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. Pueblo, Colorado (Gardner Plant) - <1. Huron, South Dakota <1. Salt Lake City, Utah (Parleys Plant) <1. Tuscon, Arizona (Plant #1) 3. Phoenix, Arizona (Verde Plant) - — <1. Coalinga, Calif. (Coalinga Mun. Water Plant) — - 2. Concord, Cal . , Contra Costa Cnty. Wat. Dist., (Bollman Plant) — 2. Dos Palos, California 3. Los Angeles, California (Owen's Aqueduct) <1 . San Diego, California (Miramar Plant) <1. San Francisco, Calif. (San Andreas Plant) <1 . Seattle, Washington (Cedar River) <1. Douglas, Alaska <1. Idaho Falls, Idaho <1. Corvallis, Oregon (Taylor Plant) — - <1. Ilwaco, Washington <1 . <.| <.05 <.2 <.05 <.05 .08 .08 <.2 .04 <.2 .03 <.2 .085 <.2 <.2 <.005 <.005 <.005 <.005 .005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 : <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 .6 1.6* .1 .8 .4 <.l 1.2 <.l .7 .4 1.0 1.1 <.l .2 1.2 <.l <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.02 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 .020 <.01 .017 <.01 .018 <.01 <.005 <.01 .013 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 <.005 <.01 .011 <.01 .024 <.01 .018 < < < < < < < < < < < < < < .004 .003 .002 .002 .002 .002 .002 .002 .002 .002 .002 .002 .002 .002 .002 .002 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 <.0005 CTl U> ------- Location TABLE 15a CONFIRMATION ANALYSIS DATA OF RAW WATER (All data in Pg/l) Bromodichloro- Dibromochloro- Carbon 1,2 Chloroform methane methane Bromoform Tetrachloride Dichloroethane 1. 2. 16. 30. 51. 60. Waterbury, Connecticut Passaic Valley, New Jersey South Pittsburgh, Pennsylvania Charleston, South Carolina Oklahoma City, Oklahoma Kansas City, Missouri * NF** (0.2) (0.1) 0.3 (0.9) (0.4) 0.3 (1.2) (0.4) <0.2 NF (0.5) (NF) NF (NF) (NF) NF (NF) (NF) NF (0.2) (NF) NF (NF) (NF) NF NF (NF) (NF) NF (NF) (NF) NF (NF) (NF) NF (NF) (NF) NF (NF) (NF) NF NF (NF) (NF) NF (NF) (NF) NF (NF) (NF) NF (NF) (NF) NF (NF) (NF) NF NF (NF) (NF) NF (NF) (NF) NF (NF) (NF) <2 (0.5) (0.2) NF (0.1) (NF) NF NF (1.5) (NF) NF (NF) (NF) NF (0.31) HJ.03) <0.2 (0.7) (0.3) NF ' (0.1) (NF) NF NF (NF) (NF) NF (NF) (NF) ------- CT) cn TABLE 15a (Cont'd.) Bromodichloro- Dibromochloro- Carbon 1,2- 65. 71. 79. Location Pueblo, Colorado Contra Costa, California Con/all is, Oregon Chloroform <0.2 (NF) (NF) 0.3 (NF) (NF) NF (0.5) (0.2) methane NF (NF) (NF) <0.3 (NF) (NF) NF (NF) (NF) methane NF (NF) (NF) NF (NF) (NF) NF (NF) (NF) Bromoform NF (NF) (NF) NF (NF) (NF) NF (NF) (NF) Tetrachloride NF (NF) (NF) NF (NF) (NF) NF (0.1) (0.1) Dichloroethane NF (NF) (NF) NF (NF) (NF) NF (NF) (NF) NF = None found. **The first listing of data is from the primary analysis performed by the method in section C(3)(a)(l), Part I. The data in parentheses () are from the quantitative confirmation analysis performed by the method in section C(3)(a)(l), Part II. ------- cr> Location TABLE 15b CONFIRMATION ANALYSIS DATA OF FINISHED WATER (All Data in ug/1) Bromodichloro- Dibromochloro- Carbon Chloroform methane methane Bromoform Tetrachloride 1,2- Dichloroethane 1. 2. 11. 16. 21. 22. 30. 31. Waterbury, Connecticut Passaic Valley, New Jersey Philadelphia, Pennsylvania*** South Pittsburgh, Pennsylvania Wheeling, West Virginia Miami*** Florida Charleston, South Carolina Cincinnati , Ohio*** 93** (61.6) (61.2) 59 (51.1) (35.9) 86* 8 (10.7) (9.2) 72 311* 195 45* 10 (5.5) (5.6) 16 (9.6) (7.4) 9* 2 (1.5) (1.5) 28 78* 9 13* 0.6 (NF) (NF) 2 (0.3) (NF) 5* 0.4 (0.2) (NF) 17 35* 0.8 4* <1 (NF) (NF) NF (NF) (NF) NF* NF (NF) (NF) NF 3* 0.8 NF* <2 (0.5) (0.5) <2 (0.4) (0.3) NF* NF (0.3) (0.1) NF NF* NF <2* <0.2 (NF) (NF) <0.2 (NF) (NF) 6* NF (0.3) (0.2) <0.4 <0.2* NF <0.4* ------- TABLE 15b (Cont'd.) Bromodichloro- Dibromochloro- Carbon 1,2- Location Chloroform methane methane Bromoform Tetrachloride Dichloroethane 41. 51. 55. 58. 60. 65. 66. 71. 72. Dayton, Ohio Oklahoma City, Oklahoma Ottumwa, Iowa*** Topeka, Kansas Kansas City, Missouri Sample Lost Pueblo, Colorado Huron, South Dakota Contra Costa, California Dos Palos, 8 44 (40.1) (45.5) 0.8* 88 24 (23.1) - ( ) 2 (1.9) (1.8) 309 31 (25.1) (26.7) 61 8 28 (23.8) (28.5) NF* 38 8 (5.6) ( ) 2 (0.8) (0.8) 116 18 (11.9) (12.4) 53 11 20 (13.5) (10.9) NF* 19 2 (0.52) ( ) <2 (NF) (NF) 49 6 (2.3) (2.1) 34 4 6 (NF) ( ) NF 5 NF (NF) ( ) NF (NF) (NF) 8 <1 (NF) (NF) 7 <2 <2 (0.6) (0.8) NF* 3 NF (0.3) ( ) NF (0.3) (0.3) NF NF (NF) (NF) NF <0.2 <0.4 (NF) (NF) NF NF NF (NF) ( ) NF (NF) (NF) NF NF (NF) (NF) NF California ------- Location TABLE 15b (Cont'd.) Bromodichloro- Dibromochloro- Chloroform methane methane Carbon 1,2- Bromoform Tetrachloride Dichloroethane 76. 79. Seattle, Washington*** 'Con/all is, Oregon 15* 26 (23.0) (18.6) 0.9* 3 (1-6) (1.2) NF* NF (NF) (NF) NF NF (NF) (NF) NF NF (0.2) (0.1) NF NF (NF) (NF) CTi OO NF = None found. indicates a positive qualitative gas chromatography-mass spectrometry determination performed either by the method in section C(3)(a)(l), Part II or C(3)(c)(l). **The first listing of data is from the primary analysis performed by the method in section C(3)(a)(l), Part I. The data in parentheses () are from the quantitative confirmation analysis performed by the method in section C(3)(a)(1), Part II. ***GC/MS data available for these supplies only at this time. ------- raw and finished waters in the 80 locations. Because of the increased sensitivity of the method described in Section C(3)(a)(l), Part II, analysis by that technique often produced a low measurable concentration where the routine method did not find the compound. This is not an in- consistency. The differences between the concentrations of the routine" and confirmation analyses in a few cases is not considered to be significant. b. Qualitative The data in Table 15 shows that the compounds quantified by the routine analysis were the correct compounds. In no case did the routine analysis ever quantify a given compound and have a negative confirma- tion by gas chronatography-mass spectrometry (GC/MS). In few cases, because one of the GC/MS methods used a larger sample for purging, this technique would detect the presence of a compound when none was found by the routine procedure. This is not an inconsistency, and as noted above, the reverse did not occur. 4. Comprehensive 5-Location Organic Study Three types of samples were collected from each of the First Series (Table 2) of locations for a comprehensive organic analysis. Work is still continuing on all of these 15 samples, therefore the results pre- sented in the following tables must be considered preliminary. The com- plete analysis of these samples will be presented in the December report. a. Groundwater, Miami, Florida The Carbon-Chloroform Extract (CCE-m) concentration was 0.9 mg/1. 1) Selected Compound Analysis TABLE 16 » Organochlorine Pesticides Organophosphate Pesticides Polychlorinated Biphenyls Herbicides Haloethers Vinyl chloride - Raw - Finished 2 ng/1 Dieldrin None Found None Found None Found None Found 1.2 yg/1 5.6 ug/1* Sample collected 1/20/75. *This value includes a trace amount of cyanogen chloride. is higher than the raw value is not known at this time. The reason it 69 ------- 2) Qrganics Purged from Grab Sample See Table 17, next page. 3) Organics Extracted from Sample with Solvent TABLE 18 Compounds Detected* Approximate Concentration,** yg/1 Bromoform 0.2 Hexachloroethane 0.07 Di-n - octyl adipate 20.0 Nicotine 3.3 Sample collected 1/20/75 *List incomplete as samples are still being analyzed. **Concentrations are probably accurate to within a factor of ten; with di-n-octyl adipate and nicotine, authentic samples were available and the concentrations of these are probably accurate to within ±50%. 4) Organics Adsorbed from Sample on Activated Carbon See Table 19. All of these data were somewhat surprising initially, as ground water has traditionally been thought of as low in contaminants. These results may not apply to all ground waters, however, but may only be representative of areas with relatively high ground water tables and relatively shallow wells. 70 ------- TABLE 17 Results reflect a single grab sample taken on January 20, 1975 in Miami, Florida Compounds Found** 1. acetaldehyde 2. acetone 3. acetylenebromide 4. acetylenechloride 5. acetylenedichloride *6. benzene 7. bromoform 8. bromomethane 9. carbon disulfide 10. carbon tetrachloride *11. chlorobenzene 12. chloroethane 13. chloroform 14. chloromethane 15. cyanogen chloride 16. dibromochloromethane 17. m-dichlorobenzene 18. o-dichlorobenzene *19. p-dichlorobenzene 20. dichlorobromomethane 21. 1,1 dichloroethane 22. 1,2 dichloroethane *23. 1,1 dichloroethylene (vinylidene chloride) *24. cis-1,2 dichloroethylene *25. trans 1,2 dichloroethylene 26. dichloromethane 27. methanol 28. 3-methyl butanal 29. 2-methyl butyl nitrile 30. 2-methyl propanal 31. 2-methyl propyl nitrile *32. toluene 33. 1,1,2 trichloroethane *34. trichloroethylene *35. vinyl chloride *Selected for future quantification. **List incomplete as analysis is continuing. Any ambiguities in nomenclature will be corrected in the December 1975 report by using the systematic name as well as the common name. 71 ------- TABLE 19 ORGANICS ADSORBED ON ACTIVATED CARBON FROM MIAMI, FLORIDA SAMPLE Approximate Concentration *1. *2. *3. *4. *5. *6. 7. *8. *9. *10. *n. *12. *13. *14. 15. *16. *17. 18. *19. 20. 21. *22. Compounds Found bromodi chl oromethane bromoform camphor chlorobenzene chl orodi bromomethane p-chlorotoluene cymeme isomer 2,6-di-t-butyl-p-benzoquinone di-n-butyl phthalate m-di chlorobenzene p-di chl orobenzene o-di chl orobenzene diethyl phthalate di-(2-ethylhexyl) phthalate di-n-propyl phthalate hexachloroethane n-propyl benzene n-propyl cycl ohexanone tetrachl oroethyl ene 1 ,1 ,3,3-tetrachloro-2-propanone tetramethyl benzene isomer • tri-n-butyl phosphate in yg/liter 4.5 1.5 0.5 1 15 1.5 0.1 0.1 5 0.5 0.5 1 1 30 0.5 0.5 0.05 0.2 0.1 0.2 0.2 0.5 Confirmed by comparison of MS and RRT with standard. 72 ------- b. Uncontanrinated Upland Water, Seattle, Washington The carbon chloroform extract (CCE-m) of this water was 0.1 mg/1. 1) Selected Compound Analysis TABLE 20 Organochlorine Pesticides 1 ng/1 Dieldrin Organophosphate Pesticides None Found Polychlorinated Biphenyls None Found Herbicides None Found Haloethers None Found Vinyl Chloride - Raw None Found - Finished None Found Sample collected 1/27/75 2) Organics Purged from Grab Sample TABLE 21 RESULTS REFLECT A SINGLE GRAB SAMPLE TAKEN ON JANUARY 27, 1975, IN SEATTLE, WASHINGTON Compounds Found* 1. acetaldehyde 2. acetone 3. 2-butanone 4. chloroform 5. dibromochloromethane 6. dichlorobromomethane 7. dichloromethane 8. ethanol 9. methanol 10. methyl acetate 11. methyl ether 12. methyl formate 13. 2-methyl propanal *List incomplete as analysis is continuing. Any ambiguities in nomenclature will be corrected in the December 1975 report by using the systematic name as well as the common name. 3) Organics Extracted from Sample by Solvent None found (sample collected 1/27/75). 73 ------- 4) Organics Adsorbed from Sample on Activated Carbon TABLE 22 ORGANICS ABSORBED ON ACTIVATED CARBON FROM SEATTLE, WASHINGTON SAMPLE Compounds Found *1. acetaldehyde *2. acetone *3. bromodichloromethane *4. camphor *5. chloral (trichloroacetaldehyde) *6. di-n-butyl phthalate *7. diethyl phthalate 8. p-ethyltoluene 9. B-santalene Approximate Concen- tration in yg/1 liter 0.1 1 0.1 0.5 3.5 0.01 0.01 0.05 0.01 *Confirmed by comparison of MS and RRT with standard. c. Raw Water Contaminated by Agricultural Runoff. Ottumwa, Iowa The carbon-chloroform extract (CCE-m) concentration of this water was 0.7 mg/1. 1) Selected Compound Analysis TABLE 23 Organochlorine Pesticides Organophosphate Pesticides Polychlorinated Biphenyls Herbicides Haloethers Vinyl Chloride - Raw - Finished Sample collected 2/25/75. 2ng/l Dieldrin None Found None Found None Found None Found None Found None Found 74 ------- 2) Organics Purged From Grab Sample TABLE 24 Results reflect a single grab sample taken on February 2, 1975 in Ottumwa, Iowa Compounds Found** 1. acetaldehyde 10. dichlorobromomethane 2. acetone 11. dichloromethane *3. benzene 12. dimethyl disulfide 4. 2-butanone 13. ethanol 5. carbon tetrachloride 14. 3-methyl butanal 6. chloroform 15. 3-methyl-2-butanone 7. chloromethane 16. 2-methyl propanal 8. cyanogenchloride *17. toluene 9. dibromochloromethane 18. 1,1,1 trichloroethane *19. trichloroethylene *Selected for future quantification. **List incomplete as analysis is continuing. Any ambiguities in nomenclature will be corrected in the December 1975 report by using the systematic name as well as the common name. 3) Organics Extracted from Sample with Solvent TABLE 25 Compounds Found** Approximate Concentration, yg/1* Benzoic Acid 15 Phenylacetic Acid 4 Sample collected 2/17/75. *List incomplete as analysis is continuing. **Concentrations are probably accurate to within a factor of ten; with benzoic acid authentic samples were available and the con- centrations of this are probably accurate to with ±50%. 75 ------- 4) Organics Adsorbed from Sample on Activated Carbon TABLE 26 Compounds Found *1. atrazine *2. camphor *3. chloropicrin (trichloronitromethane) *4. cyclohexanone *5. di-n-butyl phthalate 6. 3-methyl-3-pentanal 7. n-pentanal *8. 2-pentanone *9. a-terpeneol 10. tetramethyltetrahydrofuran Approximate Concentration _ in yg/liter _ 0.1 o.i o.i i 0.5 0.1 0.5 0.5 *Confirmed by comparison of MS and RRT with standard. d. Raw Water Contaminated by Municipal Dishcarges, Philadelphia, Pennsylvania The carbon chloroform extract (CCE-m) concentration of this water was 0.4 mg/1. 1) Selected Compound Analysis TABLE 27 Organochlorine Pesticides Organophosphate Pesticides Polychlorinated Biphenyls Herbicides Haloethers Resample 3/31/75 Vinyl Chloride - Raw - Finished None Found None Found None Found None Found 0.4 yg/1 Bis-2 (chloroethyl)ether* 0.5 yg/1 Bis-2 (chloroethyl)ether None Found 0.27 yg/1** Sample collected 2/3/75 ^Confirmed qualitatively by mass spectrometer. **This value represents a combination of vinyl chloride and cyanogen chloride. Mass spectrometer analysis indicates a greater amount of cyano- gen chloride than vinyl chloride. The reason the finished water value is higher than the raw water value is not known at this time. 76 ------- Table 28 Results reflect a single grab sample taken on February 3, 1975, in Philadelphia, Pennsylvania. Compounds Found** 1. acetaldehyde 2. acetone 3. acetylenechloride 4. acetylene dichloride *5. benzene 6. bromoform 7. 2-butanone 8. carbon tetrachloride *9. chlorobenzene 10. chloroethane 11. chloroform 12. chloromethane 13. cyanogenchloride 14. dibromochloromethane 15. m-dichlorobenzene 16. o-dichlorobenzene *17. p-dichlorobenzene 18. dichlorobromomethane 19. 1,2 dichloroethane 20. 1,1 dichloroethylene *21. cis,l, 2 dichloroethylene 22. dichloromethane 23. dimethoxymethane 24. ethanol 25. ethyl ether 26. methanol 27. 3-methyl butanal 28. 2-methyl butyl nitrile 29. methyl ether 30. 2-methyl propanal 31. 2-methyl propyl nitrile 32. nitromethane *33. tetrachloroethylene *34. toluene *35. trichloroethylene *36. vinyl chloride *Selected for future quantification **List incomplete as analysis is continuing. Any ambiguities in nomenclature will be corrected in the December 1975 report by using the systematic name as well as the common name. 77 ------- 2) Organics Purged from Grab Sample See Table 28 on next page. 3) Organics Extracted from Sample by Solvent TABLE 29 Compound Found* Approximate Concentration,** yg/1 1,2-Bis(2-chloroethoxy)ethane 0.03 Sample collected 2/3/75 *List incomplete as analysis is continuing. **Concentration is probably accurate to within a factor of ten. * 4) Organics Adsorbed from Sample on Activated Carbon TABLE 30 Compounds Found Approximate Concentration in ug/1 *1. acetaldehyde 0.1 *2. acetophenone 1 *3. bromodichloromethane 1 4. t-butyltoluene 0.01 *5. chloral (trichloroacetaldehyde) 5 *6. chlorodibromomethane 0.5 *7. di-n-butyl phthalate 0.05 *8. diethyl phthalate 0.01 *9. di-(2-ethylhexyl) phthalate . 0.5 10. 1,1,3,3-tetrachloro-2-propanone 1 Confirmed by comparison of MS and RRT with standard. 78 ------- e. Raw Water Contaminated with Industrial Discharges. Cincinnati., OFTo The carbon chloroform extract (CCE-m) concentration of this water was 0.7 mg/1. 1) Selected Compound Analysis TABLE 31 Organochlorine Pesticides 1 ng/1 Dieldrin Organophosphate Pesticides None Found Polychlorinated Biphenyls None Found Herbicides None Found Haloethers None Found Vinyl Chloride - Raw None Found - Finished None Found Sample collected 2/11/75. 2) Organics Purged from Grab Sample See Table 32 on next page. 3) Organics Extracted from Sample by Solvent TABLE 33 Compounds Found** Approximate Concentrations,* in yg/1 Dibromochloromethane 0.05 Isophorone 0.02 Trimethyl isocyanurate 0.02 Sample collected 2/11/75. Concentrations are probably accurate to within a factor of ten. **List incomplete as analysis is continuing. 4) Organics Adsorbed from the Sample on Activated Carbon See Table 34. 79 ------- Table 32 Results reflect a single grab sample taken on February 11, 1975, in Cincinnati, Ohio. Compounds Found** 1. acetaldehyde 2. acetone 3. acetylenechloride 4. acetylene dichloride *5. benzene 6. bromoform 7. 2-butanone 8. carbon disulfide 9. carbon tetrachloride *10. chlorobenzene 11. chloroethane 12. chloroform 13. chloromethane 14. cyanogenchloride 15. dibromochloromethane 16. m-dichlorobenzene 17. o-dichlorobenzene *18. p-dichlorobenzene 19. dichlorobromomethane 20. 1,2 dichloroethane 21. 1,1 dichloroethylene *22. cis, 1,2 dichloroethylene 23. dichloromethane ~ 24. ethanol 25. ethyl ether 26. methanol 27. 3-methyl butanal 28. 2-methyl butyl nitrile 29. methyl ether 30. 2-methyl propanal 31. 2-methyl propyl nitrile 32. nitromethane ***33. nitrotrichloromethane (chloropicrin) *34. tetrachloroethylene *35. toluene *36. trichloroethylene *Selected for future quantification. **List incomplete as analysis is continuing. Any ambiguities in nomenclature will be corrected in the December 1975 report by using the systematic name as well as the common name. ***Alternate for future quantification. 80 ------- TABLE 34 ORGANICS ADSORBED ON ACTIVATED CARBON FROM CINCINNATI, OHIO SAMPLE Approximate Concentration Compounds Found in ug/1iter *1. bromodichloromethane 1 *2. camphor 0.1 *3. chloral (trichloroacetaldehyde) 2 *4. chlorodibromomethane 0.5 *5. diethyl malonate 0.01 *6. diethyl phthalate 0.1 *7. lindane (Y BHC) 0.01 *8. n-propylbenzene 0.01 *9. tetrachloroethylene 0.1 10. 1,1,3,3-tetrachloro-2-propanone 0.5 *li: tri-n-butyl phosphate 0.05 *12. 1,3,5-trimethyl-2,4,6-trioxo- hexa-hydrotriazine 0.5 Confirmed by comparison of MS and RRT with standard. 81 ------- E. DISCUSSION 1. Are Trihalomethanes Formed by Chiorination and If So, How Widespread is Their Occurrence? a. Trihalomethanes The first objective of the national Organics Reconnaissance Survey was to determine the extent of chlorination by-products in finished drinking water as reported by Rookl? and tfellar, Lichtenberg and Kroner.18 To meet this objective, raw and finished water from 80 locations, repre- senting a wide variety of raw water sources and water treatment practices, were sampled for the four trihalomethanes -- chloroform, bromodichloro- methane, dibromochloromethane, and bromoform. In general, these four compounds were absent from the raw waters tested or were present in concentrations of less than 1 yg/1. Therefore, the presence of any of these four compounds in the finished water was concluded to be caused by chlorination practices. None of the systems investigated did not disinfect, but one system practiced ozonation as the only treatment the water received. All of the finished waters tested contained some chloroform although the system de- scribed above only contained 0.1 ug/1. Although a number of finished waters did not contain bromodichloromethane, dibromochloromethane and bromoform, the presence of these compounds were concluded to be wide- spread throughout the finished waters of the nation. Although the range of concentrations found for each of the four tri- halomethanes was wide for the type of systems surveyed, the concentrations of each of the compounds was not evenly distributed throughout the range but were grouped toward the lower end of the range. Note: Many ground water supplies in the United States do not chlorinate and therefore wo'uld not contain any trihalomethane, but none of these supplies were included in the Survey. Based on Figure 2, the theoretical finished water with the median concentration (one-half of the data above and below) of each compound, would contain about 21 ug/1 of chlbrofrom, 6 yg/1 of bromodi- chloromethane, 1.2 yg/1 of dibromochloromethane, and bromoform below the detection limit of the analytic method used. Therefore, although the presence of these compounds was widespread, in many'of the finished waters tested in this survey their concentrations were fairly low. Although most of the finished waters had concentrations of the four trihalomethanes declining in the same order as those in the theoretical "median" water described above, this was not true in all cases. The reasons for concentrations of the heavier compounds being greater than the lighter ones in some finished water are not known. Rookl7 has postu- lated that if bromide was present in a water, the chlorine will oxidize the bromide to bromine and the heavier bromo-compounds would be formed. Whether this phenomenon occurred in some of the finished waters surveyed is not known. 82 ------- 300 f I I 100 50 en a. 01 o o o UJ UJ o E 10 1.0 0.5 -D O D 0.1 i i i i i 2 5 10 20 30 40 50 60 70 80 90 95 98 99 PERCENT EQUAL TO OR LESS THAN GIVEN CONCENTRATION FIGURE 2. FREQUENCY DISTRIBUTION OF TRIHALOMETHANE DATA 83 ------- b. 1,2-Dichloroethane and Carbon Tetrachloride Analysis was also made of all samples for 1,2-Dichloroethane and carbon tetrachloride because they had been found in other drinking waters previously and had potential health significance. In this Survey, these two compounds were mostly absent from finished water. In about one-third of the cases where these compounds were present in the finished water, they were also present in the raw water, indicating they were environ- mental contaminants and were not created during water treatment. The cause for the appearance of these compounds in the finished water when they were absent from the raw water is not known at this time. c. Non-Volatile Total Organic Carbon In addition to studying the six specific compounds discussed above, an attempt was made to investigate the general organic level in finished drinking waters by measuring the non-volatile total organic carbon (NVTOC) concentration in all 80 locations. The range of these data was from less than 0.05 mg/1 to 12.2 mg/1, but again, the data were grouped toward the lower end of the range, see Figure 3. The median NVTOC con- centration (one-half of the data above and below) was 1.5 mg/1. 2. Influence of Source Type and Treatment Practice on Trihalomethane Formation The second objective of'the Survey was to determine, if possible, the influence of type of source and treatment practiced on the formation of chlorination by-products. An initial examination of the data indi- cated that the dominant factor influencing the creation of chlorination by-products was the general organic level of the water, provided suffi- cient chlorine was added to satisfy the chlorine demand. To test this hypothesis, the total trihalomethane concentration was first calculated for each finished water. This was done by dividing each of the four concentrations by the appropriate molecular weight and adding the quotients together. This yielded a total trihalomethane (TTHM) con- centration in yMoles/liter.* These data were then plotted against the NVTOC concentration of the finished water. The TTHM data was divided into NVTOC cells in ascending order, each cell having a range of 0.5 mg/1 NVTOC. The average TTHM concentration was then calculated for each cell and plotted against the appropriate NVTOC concentration. This analysis is appropriate based on the assumption that each cell is sufficiently large and heterogeneous with respect to the other variables that their influence is damped out by the averaging process. During this analysis, the raw water NVTOC concentration was consid- ered to be a better measure of the level of precursor available to react *Note: 1 yM/1 TTHM = 119 yg/1 chloroform if only chloroform was present. 84 ------- T—r 5.0 g H- oc ULI O O O z o CO cc o o o cc o 2.0 1.0 01 0.5 O i O 0.3 0.2 I 5 10 20 30 40 50 60 70 80 90 95 % EQUAL TO OR LESS THAN GIVEN CONCENTRATION 98 99 FIGURE 3. FREQUENCY DISTRIBUTION OF NON-VOLATILE TOTAL ORGANIC CARBON DATA 85 ------- with the chlorine than the finished water NVTOC, particularly in situa- tions where pre-chlorination is practiced, but the raw water NVTOC data contained a negative error because of the incomplete combustion of sus- pended material in the analytic procedure and could not be used. Analy- sis of the data showed, however, that finished water NVTOC could be used as an indicator of precursor level because raw- and finished-NVTOC con- centrations are proportional to one another.* The good correlation in Figure 4 shows that because most finished waters contain a residual, meaning an excess of one of the reactants is present, the concentration of the product (TTHM) is related to the con- centration of the other reactants (unknown precursors) and further that the NVTOC concentration is a reasonable indication of their concentrations. All of the data were then divided into four NVTOC concentration cells, 0-1 mg/1, 1-2 mg/1, 2-3 mg/1, and greater than 3 mg/1 to eliminate the influence of that variable and then sorted so that like source types and treatment practices were in the same cells. a. Source Influence In the NVTOC 0-1 mg/1 cell, ground water sources had lower average TTHM concentrations than surface waters. Considering all NVTOC cells not much difference existed between the various types of surface water. TABLE 30 SOURCE INFLUENCE NVTOC Range 0-1 mg/1 Avg. TTHM Cone. n uM/1 n 1-2 mg/1 Avg. TTHM Cone. yM/1 n 2-3 mg/1 Avg. TTHM Cone. yM/1 n >3 mg/1 Avg. TTHM Cone. yM/1 All Locations Ground Water River Water 18 0.15 20 0.35 9 0.07 1 -0.32 7 0.25 10 0.47 Lake and Reservoir Water 2 0.21 0.21 10 0.56 1 0.11 5 0.60 4 0.61 10 2 2 1.07 1.65 0.98 0.90 *Based on these data, coagulation and filtration removed about 30% of the raw NVTOC on the average, but this percentage is probably low. 86 ------- b. Treatment Influence 1) Chiorination Practice TABLE 31 CHLORINATION PRACTICE INFLUENCE NVTOC Range All Locations Prechl ori nation No Prechl ori nation n 18 10 8 0-1 mg/1 Avg. TTHM Cone. yM/1 0.15 0.23 0.05 n 20 17 3 1-2 mg/1 Avg. TTHM Cone. yM/1 0.35 0.36 0.28 n 10 8 2 2-3 mg/1 Avg. TTHM Cone. yM/1 0.56 0.58 0.48 >3 mg/1 Avg. TTHM n 10 7 3 Cone. yM/1 1.07 1.33 0.45 Little or no Free Residual 8 0.10 5 0.15 5 0.40 5 0.71 Little or No Combined Residual 7 0.21 11 0.34 3 0.70 3 1.68 >0.4 mg/1 Free Residual In all NVTOC cell locations where prechlorination was practiced higher average TTHM concentrations resulted than where no preclorination was practiced. An attempt was made to relate prechlorine dose to average TTHM production, but the number of locations in each cell was too small to produce meaningful data. The trend of average TTHM production was generally higher as prechlorine dose increased, but the data were quite variable. The data on chlorine residual indicated that finished waters that did not contain much free chlorine residual had lower TTHM concen- trations than systems that had higher free chlorine residuals. The^two locations using ozone had very low concentrations of TTHM. In Whiting, Indiana pre-ozonation is used following pre-chlorination. Whether or not the reduction in TTHM concentration following ozonation is caused by simple stripping or reaction of the ozone with the trihalomethanes is not known at this time. In the other installation, Strasburg, Pennsylvania, not only was ozonation the only treatment, but also the NVTOC concentra- tion was only 0.05 mg/1. Both of these factors may have contributed to the low TTHM concentration. 87 ------- 2) Filtration Practice All of the locations that practice filtration were sorted into NVTOC concentration cells and then re-sorted based on the use of polyelectrolyte either as a coagulant or filter-aid. Surface water was the raw water source for 90% of these plants, so that variable is essentially removed. This was to determine whether or not polyelectrolyte could aid as a pre- cursor for TTHM formation. In the study group, the polyelectrolyte dose varied from 0.02 mg/1 to 3.94 mg/1 (1.27 mg/1 in the raw water, plus 2.67 mg/1 on the filters) on the days of sampling. At two locations the dose was unknown. Table 32 shows that the use of polyelectrolyte does not enhance TTHM formation. TABLE 32 INFLUENCE OF FILTRATION PRACTICE NVTOC Range 0-1 mg/1 Avg. TTHM Cone. n yM/1 n 1-2 mg/1 Avg. TTHM Cone. UM/1 n 2-3 mg/1 Avg. TTHM Cone. nM/1 n >3 mg/1 Avg. TTHM Cone. yM/1 All 18 0.1.5 20 0.35 10 0.56 10 1.07 All Filter Plants 10 0.23 13 0.38 7 0.61 10 1.07 With Polyelectro- lytes 4, 0.26 4 0.42 2 0.81 2 1.28 Without Poly- electrolytes 6 0.21 14 0.37 5 0.53 8 1.01 3) Use of Activated Carbon A. Powdered Of the treatment plants using powdered activated carbon the dosage varied from 0.6 mg/1 to 6.5 mg/1. All of these plants were surface water plants. Table 33 shows that, in NVTOC concentration cells where sufficient numbers exist for comparison purposes, locations where powdered activated carbon was used had average TTHM concentrations similar to those locations without powdered activated carbon. Either powdered activated carbon cannot remove trihalomethane precursors or the dosages used were insufficient to accomplish this. 88 ------- TABLE 33 . INFLUENCE OF POWDERED ACTIVATED CARBON NVTOC Range All All Filter Plants n 18 10 0-1 mg/1 Avg. TTHM Cone. yM/1 0.15 0.23 n 20 18 1-2 mg/1 Avg. TTHM Cone. yM/1 0.35 0.38 n 10 7 2-3 mg/1 Avg. TTHM Cone. pM/1 0.56 0.61 n 10 10 >3mg/l Avg. TTHM Cone. yM/1 1.07 1.07 With Powdered Activated Carbon 2 0.31 5 0.42 5 0.58 3 0.45 Without Powdered Activated Carbon 7 0.20 11 0.35 5 0.58 5 1,35 B. Granular Only six water treatment plants used granular activated carbon as a combination filtration/adsorption media, and this number is too small to make an analysis as above. All treat surface water, pre-chlorinate, and all but one had >0.4 mg/1 free residual in the finished water, so some of the variables noted above were eliminated. Because all of the locations originally sampled were using granular activated carbon that had been in place for at least several months, the activated carbon was exhausted for NVTOC removal. This is shown in Table 34; the average NVTOC removal at these locations was not much higher than equal to or greater than 30 percent NVTOC removal previously reported for all coagulation-filtration plants. Therefore the TTHM concentration in these finished waters being higher than the TTHM concentration in the theoreti- cal "median" finished water for the entire survey in 5 out of 6 locations is not surprising. This is also true when the data are examined on a "TTHM production per unit of NVTOC" basis. Shortly after the Survey samples were taken at one of these locations, the granular activated carbon was removed and replaced with virgin lignite- base material. This location was resampled in an effort to evaluate the performance of fresh granular activated carbon. The data in Table 34 show a marked improvement in all three of the parameters listed indi- cating the effectiveness of fresh granular activated carbon for treatment. Another attempt was made to evaluate the performance of granular activated carbon for treating a variety of waters by monitoring the activated carbon (CAM) units installed in the five locations where the samples of organics that could be adsorbed on activated carbon from 89 ------- Table 34 SUMMARY OF GRANULAR ACTIVATED CARBON PLANTS Location 20 37 1 47 43 57 Avg. Theo. Median Water* Fresh Gran. Act. Carbon Finished Water NVTOC Cone. mg/1 1.0 1.4 1.6 3.2 4.2 4.4 1.5 1.4 % Removal of NVTOC >55 >30 >56 141 >30 I32 >41 • >79 TTHM Concentration pM/1 0.31 0.41 0.82 1.36 1.19 0.79 0.22 0.08 TTHM/ Fin. NVTOC yM/mg 0.31 0.10 0.51 0.43 0.28 0.18 0.30 0.15 0.06 *See Figures 2 and 3 for median concentration. 90 ------- finished waters were being collected. The samplers were 3-foot columns of coal-based granular activated carbon operated downflow at an approach velocity of 3.2 gallons per minute/square foot, and finished water was passed through them for seven days. The empty bed contact time was about 7 minutes. Table 35 shows that fresh granular activated carbon produced low NVTOC concentrations at first in all locations except Miami where the load was so heavy that a longer contact time would be needed to produce a lower NVTOC concentration. c. Section Summary To test the hypothesis that the use of surface water as a source, pre-chlorination, and the presence of greater than 0.4 mg/1 free chlorine residual enhances the formation of trihalomethanes, the data were sorted on that basis. Out of the entire survey 28 locations met these three criteria. TABLE 35 PERFORMANCE OF FRESH COAL-BASED GRANULAR ACTIVATED CARBON SAMPLERS TREATING FINISHED WATER NVTOC Concentration - mg/1 Influe Location Day Influent to Sampler Effluent from Sampler Miami, 0 Florida 7 Seattle, 0 Washington 7 Ottumwa, 0 Iowa 7 Philadelphia, 0 Pennsylvania 7 Cincinnati, 0 Ohio 7 8.1 7.1 1.9 0.8 3.6 3.4 2.0 1.9 1.2 1.6 1.3 3.5 1.9* 0.05 1.6* 0.9 0.3 0.5 0.1 0.1 NVTOC Removed 84% 51% 0%* 94% 56%* 73% 85% 74% 92% 94% *Data Suspect. Of these, 10 finished waters had an NVTOC concentration less than the Survey median concentration of 1.5 mg/1, the remainder being equal to 91 ------- or greater than the median. Of those with a finished water NVTOC concen- tration below the median concentration, 80% had a TTHM concentration above the median TTHM concentration. While of those with a finished water NVTOC concentration equal to or greater than the median concentra- tion, only 11% had TTHM concentrations below the median TTHM concentration. While this indicates the general validity of the proposed hypothesis, a rigorous multiple regression analysis of the data would be helpful. This analysis will be included in the December 1975 report. 3. Alternate Indicators of Organic Contaminant Levels Because various organic contaminants vary in toxicity, specific organic compounds should be monitored in finished waters. This is the recommended procedure for monitoring organochlorine pesticides, for example. Except for a few specific examples, this approach is beyond the capabilities of most water utilities and to some degree even is beyond the capabilities of researchers, given the current state of organic analy- sis. All specific organic compounds present in water cannot now be identified and quantified. In the absence of measuring for specific organic compounds, the next best alternative is to measure some organic parameter that includes a large number of organic compounds and assume that the level of this parameter is proportional to. the level of toxicity of the water. On this basis carbon chloroform extract (CCE-m) was included in the Interim Primary Drinking Water Regulation. In the National Organics Reconnaissance Survey non-volatile total organic carbon was the parameter chosen to represent the concentration of organics in the water. Figure 4 shows NVTOC to be generally propor- tional to trihalomethane formation, so a measure such as this is probably useful, but little else is known about NVTOC. In an effort to find an easier analytic procedure for monitoring the organic level in water, three other measurements were made on each raw and finished water in addition to NVTOC concentration. These were ultraviolet absorption (UV), emission fluorescence scan (EmFC), and the Rapid Fluorometric Method (RFM). An attempt was made to correlate these parameters, even though different organics absorb ultraviolet to dif- fering degrees and some different organics fluoresce to differing degrees. Therefore, although the a priori judgment was that those three parameters might not correlate with NVTOC concentrations because they would be heavily influenced by the types of organics present in the water, the hypothesis that different waters would be sufficiently similar to make these procedures useful was tested. Just as particulates in some raw waters interfered with the NVTOC measurement, the resultant turbidity interfered with the UV, EmFS, and RFM measurements. Plots of NVTOC concentration versus each parameter 92 ------- (NO.) = NUMBER OF LOCATIONS IN NVTOC CELL NOTE: 1 ^MOLE/LITER TTHM = 119 /iG/L CHLOROFORM IF IT WAS ALL CHLOROFORM FIGURE 4. CORRELATION OF TOTAL TRIHALOMETHANE CONCENTRATION WITH NON-VOLATILE TOTAL ORGANIC CARBON I 234 FINISHED WATER NVTOC - mg/l 93 ------- for finished water, Figures 5, 6, and 7 show a wide scatter of data. On Figure 5 a band 1 mg/1 of NVTOC wide includes 39 data points, while only excluding 28 data points, up to an NVTOC concentration of 3.5 mg/1, but the overall correlation is not very good. The two fluorescence tech- niques correlated well with each other but not with NVTOC concentration. 4. Organics Found in the 5-Location Study Because the qualitative results are incomplete and the quantitative results are absent, these data cannot be discussed, except to note that thus far the upland water and the water contaminated by agricultural runoff have had the fewest organics identified from them. 5. Significance of Findings Most water treatment plants are not designed to remove soluble organic compounds from raw water, and disinfection creates some compounds that were not originally present in the raw water. Therefore, the finding that all finished waters in the Survey contained one type of organic compound or another should not be surprising. The presence of an organic compound in a finished water is not significant, however, unless its con- centration is such that it poses a health hazard. The data contained in Appendix II, therefore, must be combined with that in Appendix VII, "Health Effects Caused by Exposure to Drinking Water Contaminants" before any significance can be attached to the data contained herein. If a health hazard is found to exist with any contaminant, then the treatment information contained in Appendix VI must be applied. 94 ------- .18 .16 .14 Z I O .12 cc O .10 3 .08 CC u. .04 .02 FIGURE 5 CORRELATION BETWEEN ULTRA-VIOLET ABSORBANCE AND NON-VOLATILE TOTAL ORGANIC CARBON IN FINISHED WATER 234 FINISHED WATER NVTOC, mg/l 95 ------- 21 19 17 15 z Q O iu s 13 11 FIGURE 6 CORRELATION OF RAPID FLUORMETRIC METHOD AND NON-VOLATILE TOTAL ORGANIC CARBON IN FINISHED WATER I 234 FINISHED WATER NVTOC, mg/l ------- 1900 1700 1500 [2 1300 z tn o 1100 Z iu Ul cc o D ill 900 Z s ui 700 500 300 100 I FIGURE 7 CORRELATION OF EMISSION FLUORESCENCE SCAN AND NON-VOLATILE TOTAL CARBON IN FINISHED WATER I 234 FINISHED WATER NVTOC, mg/l 97 ------- F- ACKNOWLEDGMENTS The following organizational units of EPA contributed to the timely conduct of the National Organics Reconnaissance Survey and to the prepara^ tion of this report. Water Supply Research Laboratory, ORD Methods Development and Quality Assurance Laboratory, ORD Southeast Environmental Research Laboratory, ORD R. S. Kerr Environmental Research Laboratory, ORD Office of Environmental Sciences, ORD Office of Monitoring Systems, ORD National Field Investigations Center - Cincinnati, OE Water Supply Division, OWHM Regional Water Supply Representatives The EPA personnel whose cooperation and dedication in accomplishing the many activities that resulted in this report are to be commended for their outstanding performance within the difficult constraints involved. Special commendation is due Mrs. Maura M. Lilly who typed the entire first draft of this appendix in three working days. 98 ------- G. REFERENCES 1. Buelow, R. W., J. K. Carswell and J. M. Symons, "An Improved Method for Determining Organics by Activated Carbon Adsorption and Solvent Extraction (Parts I and II)," Journal American Water Works Associa- tion^ 65_, 57-72, 195-199 (Jan.-March 1973). 2. Bellar, T. A. and J. J. Lichtenberg, "The Determination of Volatile Organic Compounds at the yg/l Level in Water by Gas Chromatography," USEPA Report, NERC, Cincinnati, EPA-670/4-74-009, Nov. 1974. 3. Stevens, A. A. and J. M. Symons, "Analytical Considerations for Halogenated Organic Removal Studies," USEPA Water Supply Research Laboratory, NERC-Cincinnati, Dec. 1974, Prepublication copy. 4. "Method for Polychlorinated Biphenyls in Industrial Effluents," National Pollution Discharge Elimination System, Appendix A, Methods Development and Quality Assurance Research Laboratory, Cincinnati, Ohio, November 28, 1973. 5. Dressman, R. C. and E. F. McFarren, "Detection and Measurement of Bis-(2-chloro) Ethers and Dieldrin by Gas Chromatography," Second Annual Water Quality Conference of American Water Works Association, Dallas, Texas, December 2-4, 1974. 6. Bellar, T. A. and J. J. Lichtenberg, "Determining Volatile Organics at the yg/1 Level in Water by Gas Chromatography," JAWWA, 66^ 739-744 (December 1974). 7. "Method for Organophosphorous Pesticides in Industrial Effluents," National Pollution Discharge Elimination System, Appendix A, Methods Development and Quality Assurance Research Laboratory, Cincinnati, Ohio, November 28, 1973. 8. Dobbs, R. A., R. H. Wise and R. B. Dean, "The Use of Ultraviolet Absorbance for Monitoring the Total Organic Carbon Content of Water and Wastewater," Water Research, 6., 1173-80 (1972). 9. Sylvia, A., "Detection and Measurement of Microorganics in. Drinking Water," Jour. New Eng. Water Works Assn.. 87_, No. 2 (June 1973). 10. Standard Methods for the Examination of Water and Wastewater, 13th Edition, American Public Health Association, New York, New York. 11. "Methods for Chemical Analysis of Water and Wastewater," U. S. Environmental Protection Agency, 1974. 12. Caldwell, J. S., R. J. Lishka, and E. F. McFarren, "Evaluation of Low-Cost Arsenic and Selenium Determination at Microgram-per-Liter Levels," JAWWA, 63_, 731 (1973). 99 ------- 13. "Cyanide in Water and Wastewater, Industrial Method," No. 119-71W, April 1972, Technicon Instrument Corp., Tarrytown, New York. 14. Kopp, J. F., M. C. Longbottom, and L. B. Lobring, "Cold Vapor Method for Determining Mercury," JAWWA, 64_, 20 (1972). 15. Organochlorine Pesticides in Water, 1974 Annual Book of ASTM Standards, American Society for Testing Materials. 16. Georlitz, D. T., and W. L. Lower, Determination of Phenoxy Acid Herbicides in Water by Electron Capture and Micro Coulometric Gas Chromatography, Geological Survey Water Supply Paper 1817C, U. S. Government Printing Office. 17. Rook, J. J., "Formation of Haloforms During Chlorination of Natural Waters," Water Treatment and Examination, 23. Part 2, 234-243 (1974). 18. Bellar, T. A., J. J. Lichtenberg, and R. C. Kroner, "The Occurrence of Organohalides in Chlorinated Drinking Water," JAWWA, 66, 703-706, (December 1974). 100 ------- APPENDIX III ORGANIC CHEMICALS FOUND-IN INDUSTRIAL EFFLUENTS Southeast Environmental Research Laboratory Athens, Georgia and Office of Research and Development Environmental Protection Agency Washington, D.C. ------- ORGANIC CHEMICALS IN INDUSTRIAL EFFLUENTS The compositions of industrial effluents are being systematically studied at the Southeast Environmental Research Laboratory. In addition, short-term studies for special purposes have been conducted at the request of Regional and other offices. Table 1 is a composite list of substances and their sources as of mid-1973. Compounds in textile mill effluents identified since 1973 are listed in Table 2. In general the lists of compounds already found in drinking water appear to have more in common with the lists of compounds occurring in industrial wastes than the list of compounds occurring in domestic sewage. Of those substances identified as suspect carcinogens, two, chloroform and bis (2-chloroethyl) ether appear in industrial wastes and have not been shown to occur in domestic sewage. It should be mentioned, however, that there is the possibility that these compounds are formed during the chlorination of drinking water. With the presently available information, it would appear that the organic substances occurring in drinking water are for the large part of industrial origin. Where special studies have been undertaken to identify specific compounds causing problems, such as taste and odor, in water supplies, the results have led to the conclusion that the causative agents were of industrial origin. It should be kept in mind, however, that the analyses of drinking water, municipal wastewaters and industrial effluents is continuing and the final results may present a somewhat different picture. 102 ------- Table 1 ORGANIC CHEMICALS FOUND IN INDUSTRIAL WASTES Compound Sample source 6,8,11,13-Abietatetraen-18- oic acid 13-Abieten-18-oic acid Abietic acid Acenaphthalene Acenaphthene Acetophenone Acetosyringone Acetovanillone Paper mill's raw waste and trickling filter effluent Paper mill's raw waste and trickling filter effluent Paper mill's raw waste and lagoon Petrochemical plant's five-day lagoon effluent Petrochemical plant's five-day lagoon effluent Wood preserving plant's lagoon effluent Wood preserving plant's settling pond Pesticide plant's raw effluent Chlorinated paraffin plant's lagoon Petrochemical plant's five-day lagoon effluent Gulf coast paper mill's settling pond Gulf coast paper mill's settling pond Paper mill's raw waste and lagoon 103 AWBERC LIBRARY U.S. EPA ------- Table 1 (Continued) 2-Acetylthiophene Acrylonitrile Adipic acid Adiponitrile Aldrin m-Anethole .o-Anethole p-Anethole Anthraquinone Anteisomargan'c acid Anteisopentadecanoic acid Arachidic acid Arachidonic acid Behenic acid Benzaldehyde Benzyl alcohol 2-Benzothiazole Paper mill's raw waste Acrylic fiber plant's settling pond Nylon plant's raw waste Nylon plant's raw waste Pesticide plant's raw effluent Paper mill's raw waste Paper mill's raw waste Paper mill's raw waste Wood preserving plant's settling pond Paper mill's raw waste and five-day lagoon Paper mill's five-day lagoon Paper mill's raw waste Paper mill's five-day lagoon Paper mill's raw effluent and five- day lagoon Paper mill's raw waste Petrochemical plant's five-day lagoon effluent Latex accelerators and thickeners plant's holding pond Synthetic rubber plant's aerated lagoon 104 ------- Table 1 (Continued) Biphenyl Borneol 1-Butanol 2-Butoxyethanol n-Butylisothiocyanate Camphor Caproic acid Carbazole Chlordane Chlordene o-Chlorobenzoic acid bis-(2-Chloroethoxy) methane bis-2-Chloroethyl ether bis-2-Chloroisopropyl ether trans-Communic acid River below textile finishing plant Paper mill's raw waste and trickling filter effluent Petrochemical (alcohols) plant's raw effluent Petrochemical plant's five-day lagoon effluent Latex accelerators and thickeners plant's holding pond Paper mill's raw waste and trickling filter effluent Gulf coast paper mill's settling pond Nylon plant's raw waste Wood preserving plant's settling pond Pesticide plant's raw effluent Pesticide plant's raw waste Chlorinated paraffin plant's lagoon Synthetic rubber plant's treated waste Synthetic rubber plant's treated waste Glycol plant's thickening and sedimentation.pond Paper mill's raw waste and trickling filter effluent 105 ------- Table 1 (Continued) o-Cresol o-Cresol m-Cresol p-Cresol Cumene (isopropylbenzene) Cyclohexanol 1,5-Cyclooctadiene p-Cymene n Decane 1-Decanol Dehydroabietic acid Diacetone alcohol Wood preserving plant's settling pond Petrorefinery's eight-hour lagoon effluent Wood preserving plant's settling pond Paper mill's raw waste and lagoon Petrochemical plant's five-day lagoon effluent Nylon plant's raw waste *•**• Petrochemical plant's five-day lagoon effluent Paper mill's raw waste and trickling filter effluent Pesticide plant's raw waste Polyolefin plant's lagoon Petrochemical (alcohols) plant's raw effluent Wood preserving plant's settling pond Paper mill's raw waste and trick- ling filter effluent Gulf coast paper mill's settling pond Tall oil refinery's settling pond Petrochemical plant's five-day lagoon effluent 106 ------- Table 1 (Continued) 4,4'-Diamino-dicyclohexyl methane Dibenzofuran 2,3-Dibromo-l-propanol Dibromopropene isomer Dibutyl amine Dieldrin N,N-Diethylformamide Diethyl phthalate 3,4-Di hydroxyacetophenone (pungenin) 3,5-Dimethoxy-4-hydroxy- acetophenone 2,4-Dimethyldiphenylsulfone Nylon and polyester plant's effluent after neutralization and sedimentation Wood preserving plant's settling pond Wood preserving plant's lagoon effluent Nylon plant's settling pond Acrylic fibers plant's settling pond Acrylic fibers plant's settling pond Latex accelerators and thickeners plant's raw effluent Anaerobic lagoon of yarn finishing mill Pesticide plant's raw effluent Latex accelerators and thickeners plant's raw effluent Synthetic rubber plant's settling pond Paper mill's trickling filter effluent Paper mill's raw effluent and five-day lagoon Nylon plant's settling pond Acrylic fibers plant's settling pond 107 ------- Table 1 (Continued) Dimethyl furan isomer 2,6-Dimethyl naphthalene Dimethyl naphthalene isomer Dimethyl phthalate Dimethyl pyridine isomer Dimethyl quincline isomers Dimethyl sulfone Dimethyl sulfoxide 10,12-Dimethyl tridecanoic acid 4,6-Dinitro-o-cresol (2-methyl-4,6-dinitro-phenol 2,4-Dinitrotoluene 2,6-Dinitrotoluene 3,4-Dinitrotoluene Petrochemical plant's five-day lagoon effluent Petrochemical plant's five-day lagoon effluent Pesticide plant's raw effluent Plastic (PVA) plant's settling pond Synthetic rubber plant's settling pond Wood preserving plant's settling pond Wood preserving plant's settling pond Paper mill's raw waste and trickling filter effluent Paper mill's raw waste and trickling filter effluent Paper mill's five-day lagoon Specialty chemical plant's effluent Explosives (DNT) plant's raw waste and settling pond effluent Explosives (DNT) plant's raw waste and settling pond effluent TNT plant's raw effluent Explosives (DNT) plant's raw waste and settling pond effluent 108 ------- Table 1 (Continued) Diphenylene sulfide Diphenyl ether 3,3-Diphenylpropanol 2,6-Di-t-butyl-p-benzo-quinone p-Dithiane Dodecane Eicosane (C20) Endrin Ethyl carbamate 2-Ethyl-l-hexanol Ethylidenecyclopentane Wood preserving plant's settling pond Pesticide plant's raw effluent Petrochemical plant's five-day lagoon effluent Surface drainage from closed waste treatment system of particle board plant Synthetic rubber plant's treated waste Petrorefinery's lagoon effluent after activated sludge treatment Petrorefinery's eight-hour lagoon effluent Paper mill's raw effluent Petrorefinery's lagoon effluent after activated sludge treatment Pesticide plant's raw effluent Paper mill's trickling filter and aerated lagoon Gulf coast paper mill's settling pond Laboratory sewage Plastic (PVA) plant's settling pond River below textile finishing plant Paper mill's raw waste 109 ------- Table 1 (Continued) Ethyl isothiocyanate Ethyl naphthalene isomer M m-Ethyl phenol Ethyl phenylacetate o-Ethyl toluene Eugenol Fenchyl alcohol Fenchone Fluoranthene Fluorene 2-Formylthiophene Furfural Guaiacol Latex accelerators & thickeners plant's raw effluent Petrochemical plant's five-day lagoon effluent Pesticide plant's raw effluent Paper mill's raw waste and lagoon Resin plant's lime treated holding pond effluent Petrochemical plant's five-day lagoon effluent Paper mill's raw waste and lagoon Paper mill's raw waste and trick- ling filter effluent Paper mill's raw waste and trick- ling filter effluent Wood preserving plant's settling pond Wood preserving plant's settling pond Petrochemical plant's five-day lagoon effluent Paper mill's raw waste Paper mill's raw waste Synthetic rubber plant's settling pond Gulf coast paper mill's settling pond 110 ------- Table 1 (Continued) Guaiacol Heneicosane (C2l) Heptachlor Heptachloronorbornene isomers Heptadecane Hexachlor epoxide Hexachlorobenzene Hexachlorobutadiene Hexachlorocyclopentad iene Hexachloronorbornadiene isomers Hexadecane Paper mill's raw waste and trick- ling filter effluent Petrorefinery's lagoon effluent after activated sludge treat- ment Pesticide plant's raw waste Pesticide plant's raw effluent Nylon plant's settling pond Petrorefinery's eight-hour lagoon effluent Petrorefinery's lagoon effluent after activated sludge treatment Pesticide plant's raw waste Chlorinated solvents plant's raw effluent Pesticide plant's raw effluent Pesticide plant's raw waste Pesticide plant's raw effluent Nylon plant's settling pond Petrorefinery's eight-hour lagoon effluent Petrorefinery's lagoon effluent after activated sludge treatment Paper mill's raw waste 111 ------- Table 1 (Continued) Hexadecane Hexadieneal 1-Hexanol Homovanillie acid p-Hydroxyacetophenone p-Hydroxybenzaldehyde o-Hydroxybenzoic acid Hydroxybiphenyl isomer 4-Hydroxy-3 methoxypropio- phenone p-Hydroxythiophe'nol Indan Indene Isodrin Isoeugenol Isopalmitic acid Isopentyl alcohol Isooctyl phthalate Isopimaric acid Petrochemical plant's five-day lagoon effluent Pesticide plant's raw effluent Petrochemical (alcohols) plant's raw effluent Paper mill's raw waste and five-day lagoon Paper mill's raw waste and lagoon Paper mill's raw waste and lagoon Paper mill's raw waste Pesticide plant's raw effluent Paper mill's raw effluent Paper mill's raw waste Petrochemical plant's five-day lagoon effluent Petrochemical plant's five-day lagoon effluent Pesticide plant's raw effluent Paper mill's raw waste and lagoon Paper mill's five-day lagoon Laboratory sewage Nylon plant's raw waste Paper mill's raw waste and trickling filter effluent 112 ------- Table 1 (Continued) Jasmone Lignoceric acid Limonene Linoleic acid Mandelic acid Margaric acid 2-Mercaptobenzothiazole ii alpha-Methyl benzyl alcohol Methyl biphenyl isomer Methyl 3,4-Dimethoxybenzyl ether 2-Methyl-4-ethyl dioxolane Methyl ethyl naphthalene isomer 1-Methyl indene 3-Methyl indene 1-Methyl naphthalene Pesticide plant's raw effluent Paper mill's raw waste Paper mill's raw waste and trick- ling filter effluent Paper mill's raw waste and lagoon Paper mill's raw waste Paper mill's raw waste Synthetic rubber plant's aerated lagoon Paper mill's raw waste and lagoon Petrochemical plant's five-day lagoon effluent Petrochemical plant's five-day lagoon effluent Paper mill's raw waste Fiberglass plant's effluent Petrochemical plant's five-day lagoon effluent Petrochemical plant's five-day lagoon effluent Petrochemical plant's five-day lagoon effluent River below textile finishing plant 113 ------- Table 1 (Continued) 1-Methyl naphthalene 2-Methyl naphthalene Methyl naphthalene isomer Methyl naphthalene isomers 13-Methyl pentadecanoic acid Methyl phenanthrene Methyl quinoline isomers o-Methylstyrene beta-Methylstyrene Methyl trisulfide Myristic acid Naphthalene Petrorefinery's eight-hour lagoon effluent Petrochemical plant's five-day 0 lagoon effluent Synthetic rubber plant's settling pond Petrorefinery's eight-hour lagoon effluent Petrochemical plant's five-day lagoon effluent Wood preserving plant's lagoon "effluent Pesticide plant's raw effluent Paper mill's five-day lagoon Wood preserving plant's lagoon effluent Wood preserving plant's settling pond Petrochemical plant's five-day effluent Petrochemical plant's five-day lagoon effluent Paper mill's raw waste Paper mill's raw waste Nylon plant's settling pond Surface drainage from closed treat- ment of system of particle board plant 114 ------- Table 1 (Continued) Naphthalene 2-Naphthoic acid Neoabietic acid Nitrobenzene 2-Nitro-p-cresol o-Nitrophenol o-Nitrotoluene m-Nitrotoluene p-Nitrotoluene Nonachlor Nonadecane Nonylphenol Petrochemical plant's five-day lagoon effluent Pesticide plant's raw waste Wood preserving plant's settling pond Paper mill's raw waste Chemical company's lagoon after steam stripping Chemical company's lagoon after steam stripping Chemical company's lagoon after steam stripping Paper mill's five-day lagoon TNT plant's raw effluent DNT plant's raw effluent DNT plant's raw effluent Chemical company's lagoon after steam stripping DNT plant's raw effluent Pesticide plant's raw effluent Petrorefinery's lagoon effluent after activated sludge treatment Petrorefinery's eight-hour lagoon effluent Anaerobic lagoon of yarn finishing mill 115 ------- Table 1 (Continued) Nonylphenol Norcamphor beta-Ocimene 1-Octanol Octachlorocyclopentene Octadecane Oleic acid Octylphenol Palmitic acid Palmitoleic acid Pentachlorocyclopentadiene i somers Pentachloronorbornadiene isomer River below textile finishing plant Paper mill's raw waste Paper mill's raw waste Petrochemical (alcohols) plant's raw effluent Pesticide plant's raw effluent Petrorefinery's eight-hour lagoon effluent Nylon plant's settling pond Tall oil refinery's settling pond Paper mill's raw waste and trickling filter effluent River below textile finishing plant Textile chemical plant's raw effluent Tall oil refinery's settling pond Paper mill's raw waste and trickling filter effluent Gulf coast paper mill's settling pond Paper mill's five-day lagoon Pesticide plant's raw effluent Pesticide plant's raw effluent 116 ------- Table 1 (Continued) Pentachloronorbornene isomer Pentachloronorbornadiene epoxide isomer Pentachlorophenol Pentadecane Pentadecanoic acid Phenanthrene Phenol Pesticide plant's raw effluent Pesticide plant's raw waste Pesticide plant's raw waste Latex accelerators and thickeners plant's holding pond Wood preserving plant's raw effluent Resin plant's lime treated holding pond efiluent Synthetic rubber plant's aerated lagoon Wood preserving plant's lagoon effluent Petrorefinery's eight-hour lagoon effluent Petrorefinery's lagoon effluent after activated sludge treatment Paper mill's raw waste Petrochemical plant's five-day lagoon effluent Paper mill's lagoon Wood preserving plant's lagoon effluent Wood preserving plant's settling pond Laboratory sewage 117" ------- Table 1 (Continued) Phenol Phenyl ether o-Phenylphenol Pimaric acid beta-Pinene Pinene isomer Polychlorinated biphenyls (Arochlor 1254) 2-Propionylthiophene 4-n-Propylphenol Pyrene Qu incline Sandaracopimeric acid Petrorefinery's eight-hour lagoon effluent Wood preserving plant's settling pond Petrochemical plant's five-day lagoon effluent Paper mill's raw waste Nylon plant's settling pond River below textile finishing plant Paper mill's raw waste and trick- ling filter effluent Gulf coast paper mill's settling pond Paper mill's raw waste Gulf coast paper mill's settling pond Nylon plant's raw waste Paper mill's raw waste Paper mill's raw waste and lagoon Wood preserving plant's settling pond Wood preserving plant's settling pond Paper mill's raw waste and lagoon 118 ------- Table 1 (Continued) Stearic acid Styrene Syringaldehyde Terpinene-4-ol alpha-Terpineol Terpineol isomer Terpinolene 1,1,2,2-Tetrachloroethane Tetrachlorophenol isomer Tetradecane Textile chemical plant's raw effluent Gulf coast paper mill's settling pond Petrochemical plant's five-day lagoon effluent Synthetic rubber plant's settling pond Gulf coast paper mill's settling pond Paper mill's lagoon Paper mill's raw waste Nylon plant's settling pond Paper mill's raw waste and trick- ling filter effluent Petrochemical plant's five-day lagoon effluent Gulf coast paper mill's settling pond Paper mill's raw waste Chlorinated solvents plant's raw effluent Wood preserving plant's raw effluent Petrorefinery's lagoon effluent after activated sludge treat- ment Petrorefinery's eight-hour lagoon effluent 119 ------- Table 1 (Continued) Tetramethylbenzene isomer 2,2'-Thiodiethanol (Thiodiglycol) Toluic acid Trichlorobenzene isomer Tri chlorocyclopentene isomers 1,1,2-Trichloroethane Trichloroguaiacol n-Tridecane Triethyl urea 3,4,5-Trimethoxyaceto- phenone 2,4,6-Trimethylpyridine 2,4,6-Tri ni trotoluene n-Undecane Pesticide plant's raw waste Synthetic rubber plant's treated waste Chlorinated paraffin plant's lagoon River below textile finishing plant Textile chemical plant's raw effluent Pesticide plant's raw effluent Chlorinated solvents plant's raw effluent Paper mill's raw waste Petrorefinery's eight-hour lagoon effluent Petrorefinery's lagoon effluent after activated sludge treat- ment Paper mill's raw waste Latex accelerators & thickeners plant's raw effluent Paper mill's raw waste and trick- ling filter effluent Wood preserving plant's settling pond TNT plant's raw effluent Paper mill's raw waste 120 ------- Table 1 (Continued) n-Undecane Valeric acid Vanillin Veratraldehyde o-Xylene m-Xylene p-Xylene 2,5-Xylenol 3,4-Xylenol 3,5-Xylenol Petrorefinery's eight-hour lagoon effluent Polyolefin plant's lagoon Petrorefinery's lagoon effluent after activated sludge treat- ment Nylon plant's raw waste Paper mill's raw waste and trick- ling filter effluent Gulf coast paper mill's settling pond Paper mill's raw waste & lagoon Syntehtic resin plant's settling pond Petrochemical plant's five-day lagoon effluent Petrochemical plant's five-day lagoon effluent Petrochemical plant's five-day lagoon effluent Wood preserving plant's settling pond Wood preserving plant's settling pond Wood preserving plant's settling pond 121 ------- Table 2 ORGANIC COMPOUNDS IN TEXTILE EFFLUENTS Compound 1,2,4-trichlorobenzene benzole acid (methyl ester) p-nonylphenol p-tert -butylphenol di-n-butyl phthalate methyl isobutyl ketone acetophenone chlorobenzene p-dichlorobenzene toluene ethyl benzene naphthalene 1-methyl naphtha 1ene dodecane 2-methylpyrrolidone 1,3,5-trimethylbenzene cymene tridecane tetradecane chloroform tetrachloroethylene styrene o-phenylphenol biphenyl diphenyl oxide ethylene dichloride benzophenone n-butanol 122 ------- APPENDIX IV MONITORING FOR RADIATION IN DRINKING WATER Office of Radiation Programs Environmental Protection Agency Washington, D.C. ------- MONITORING FOR RADIATION IN DRINKING WATER This appendix focuses on the Environmental Radiation Ambient Monitoring System. Radium-226 and methods of removing it from water supplies are the subject of section C(4) of Appendix VI. The Environmental Radiation Ambient Monitoring System (ERAMS), which began in July 1973, was developed from previously operating radiation monitoring networks to form a single monitoring system more responsive to current and projected sources of environmental radiation. The ERAMS Drinking Water Component is an expansion of the previous Tritium Surveillance System which was operated by the Office of Radiation Programs from 1970 through June 1973. The Drinking Water Com- ponent consists of 77 quarterly drinking water samples taken from major population centers and selected nuclear facility environs. Tri- tium is analyzed on a quarterly basis with grab samples. Tritium, a long-lived (half-life of 12.3 years) isotope of hydrogen (hydrogen-3), is produced in nuclear power production and nuclear weapons testing, and naturally by cosmic radiation. Because it is chemically similar to hydrogen, tritium readily enters the body as water and is incorporated into living tissue. Table 1 presents the tritium concentrations in drinking water at the Drinking Water Component stations for 1974. The average tritium concentration was 0.3 nCi/liter. The radiation dose to individuals may be calculated from the formula: H (mrem/year) = 0.1C (nCi/liter) where H is the dose equivalent rate and C represents the tritium con- centration in body water in nCi/liter (nCi = 10~y curie). Assuming that the concentration of tritium in all water taken into the body is equal to that found in the drinking water, and that the specific activity of tritium in the body is essentially the same as that in the drinking water, then the radiation dose to individuals may be estimated. The highest individual concentration of tritium observed in drinking water was 6.8 nCi/liter during 1974. This corresponds to a dose of 0.7 mrem/year (0.007 rem/year). The average tritium concentration during 1974 was 0.3 mrem/year. The calculated health effects to the U. S. population may be estimated by using a risk factor of 7 x 10-4 health effects per person-rem. Therefore, the calculated number of potential health effects in the U. S. population would be 4.5 based upon a constant intake at the average concentration. 124 ------- Table 1 ERAMS Drinking Water Component, 1974 Tritium concentration3 (nCi/liter +_ 2a)° Location Jan-Mar April-June July-Sept Oct-Dec Ala: Dothan 0 000 Montgomery 0 .20 0 Muscle Shoals--- 0 .3 .3 .2 Alaska: Anchorage NS 0 .5 .4 Fairbanks .5 .5 .5 .3 Ark: Little Rock 0 000 Calif: Berkeley .2 .2 .2 0 Los Angeles 0 000 C. Z: Ancon .5 000 Colo: Denver .5 .5 .4 .6 Platteville .9 1.0 .9 .6 Conn: Hartford 0 0 .2 .2 Del: Wilmington .3 0 .3 .3 D. C: Washington 0 .200 Fla: Miami 0 000 Tampa 0 000 Ga: Baxley—- NS 0 NS 0 Savannah-— 3.1+0.3 6.8 + 0.3 3.0 2.9 Hawaii: Honolulu 0 000 Idaho: Boise .3 0 NS .2 125 ------- Table 1 (Continued) Tritium concentration9 (nCi/liter +_ Location Jan-Mar April-June July-Sept Oct-Dec Idaho: Idaho Falls .3 .3 .6 .3 4? Ill: Chicago— 1.0 .6 0 .2 Morris 0 000 Iowa: Cedar Rapids-— NS NS .3 .5 Kans: Topeka 0 0 .30 La: New Orleans .2 0 .3 .3 Maine: Augusta .2 0 0 .2 Md: Baltimore 0 NS .3 .5 Conowingo 0 0 .3 .3 Mass: Lawrence 0 .2 .2 0 Rowe .3 0 NS A Mich: Detroit —- .4 .4 .4 .2 Grand Rapids--- .3 0 .3 .2 Minn: Minneapolis .4 .3 .5 .5 Red Wing 0 000 Miss: Jackson 0 0 0 .2 Mo: Jefferson City- 0 .40 0 Mont: Helena .3 .5 .4 .4 Nebr: Lincoln .2 .2 .2 0 Nev: Las Vegas .8 .7 . .6 .7 N. H: Concord 0 .2 .2 .3 N. J: Trenton 0 NS .20* 126 ------- Table 1 (Continued) N. J: N. Mex: N. Y: N. C: N. Dak: Ohio: Okla: Oreg: Pa: P.R: R.I.: S.C.: Locati on Al hanv— -———--_. Rnffaln East Liverpool- Painesville Oklahoma City-- narri sourg Tritium concentration3 (nCi /liter + Jan-Mar 0 .5 0 .3 .3 .6 0 0 .5 0 .4 0 NS 0 0 0 0 .4 0 .2 .3 n April -June July-Sept NS 0 NS .5 .3 0 .2 .2 NS .3 .6 .5 .7 .3 - 0 .2 .5 .7 .3 .2 . .3 .4 .3 .3 NS NS 0 .2 0 0 0 .2- .2 .3 .2 .3 0 0 0 0 .2 .3 0 -4 2a)b ' Oct-Dec 0 0 .3 .5 0 .7 .2 .2 .4 .2 .3 .5 NS 0 .3 .7 .3 .3 0 0 A .3 127 ------- Table 1 (Continued) Tritium concentration9 (nCi/liter +_ Location Jan-Mar April-June July-Sept Oct-Dec S. C.: Hartsville 0 000 Seneca .2 .4 .3 .3 Tenn: Chattanooga .5 .6 .4 0 Knoxville .4 .400 Tex: Austin 0 000 Va: Doswell— - 0 0 0 .2 Lynchburg 0 .2 .2 .2 Norfolk—- .2 0 0 .2 Wash: Richland— NS .5 .4 .5 Seattle .2 0 0 .4 Wise: Genoa 0 0 NS 0 Madison 0 000 Average 0.2 0.3 0.3 0.2 aThe minimum detection limit for all samples was 0.20 nCi/liter. All values equal to or less than 0.20 nCi/liter before rounding have been reported as zero. bThe 2o error for all samples is 0.20 nCi/liter unless otherwise noted. NS, no sample. 128 ------- APPENDIX V ANALYSIS OF INORGANIC CHEMICALS IN WATER SUPPLIES Water Supply Research Laboratory National Environmental Research Center Office of Research and Development Cincinnati, Ohio ------- APPENDIX V ANALYSIS OF INORGANIC CHEMICALS IN WATER SUPPLIES Table of Contents Page A. Interstate Carrier Water Supplies 131 B. Community Water Supply Survey 131 C. Special Studies 131 D. National Organic Reconnaissance Survey Sampling. . . . 132 E. Asbestos Studies 135 130 ------- ANALYSIS OF INORGANIC CHEMICALS IN WATER SUPPLIES A. INTERSTATE CARRIER WATER SUPPLIES For many years the federal government has exercised a regulatory function over the water supplies that provide the water to the watering points of carriers in interstate commerce. If water is loaded aboard a train, ship, plane, or bus, the regulation of the actual watering point is conducted by the Food and Drug Administration, but the regulation of the water systems that supply the water is done by the Environmental Protection Agency. For these interstate carrier supplies, the state agency controlling community water supplies makes an annual report on the quality of each supply. Besides the summary on the numerous bacteriological samples, data are provided from the most recent chemical"analyses on the consti- tuents limited by the Drinking Water Standards. At about three-year intervals, a joint survey is made by the state agency and the EPA Re- gional Office of each of these 700 or so supplies. At the time of the joint survey, a water sample is collected and sent to the Water Supply Research Laboratory in Cincinnati for analyses of the chemicals limited by the standards. Tabulation of these data is made periodically, the latest being Chemical Analyses of Interstate Carrier Water Supply Systems, October 1973.4 Table II is the summary from this report. B^ COMMUNITY WATER SUPPLY SURVEY Water samples are collected at the water plant for chemical analy- ses in the interstate state carrier surveillance and by most state agen- cies. Evidence has been developed that for some constituents the water quality is degraded in distribution. This has been recognized for bac- teriological sampling but the effect of the distribution system and household plumbing was not determined by the chemical sampling. The first comprehensive set of data on water quality at the consumer's tap was reported in 1970.1 A comparison of the results of this study with the 1962 Drinking Water Standards and the American Water Works Associa- tion's water quality goals are shown in Table I. C. SPECIAL STUDIES EPA has conducted some studies in water systems where the water is particularly corrosive to the distribution system and plumbing. Results of two of these studies have been reported2 and indicate that a signif- icant number of homes have lead concentrations exceeding the limits. This is most noticeable in the first water drawn in the morning. Human body burden studies are being conducted to see if these high morning concentrations may lead to a health effect. Table III presents some of the water data. 131 ------- A cooperative study now underway will obtain data on the inorganic chemical content of drinking water at a representative set of homes in the U. S. Water samples are being collected at the homes of persons included in the current series of the National Health Examination Survey. Because of the interest of the National Heart and Lung Institute and EPA in the suggested association of heart disease mortality and soft drinking water, this detailed analysis of drinking water quality and health examination results is underway. The study is designed for health effects research but it will also provide data on water quality for the chemicals limited by the drinking water standards as well as 86 additional chemicals for which we have little information on occur- rence in drinking water. D. NATIONAL ORGANICS RECONNAISSANCE SURVEY SAMPLING To round out the analyses of NORS and to possibly provide some in- sight to causes for the developing of the chlorine reaction products, samples were analyzed for the inorganic chemicals proposed to be limited by the new drinking water standards. The results of these analyses are tabulated in Appendix II. Analyses have been completed in all but three of the water systems included in NORS. The results were as expected from previous surveys but because sampl-es were collected at the water plant or well head the pickup of metals in distribution would not be noted. Three supplies exceeded the flouride limits which is comparable with the results of the Community Water Supply StudyJ The three samples exceeding the fluoride limit were collected from water supplies adding fluoride in an attempt to provide an optimum con- centration. A larger study of 286 water systems in Wisconsin that fluo- ridated was conducted in 1968-1970.3 The findings from this study showed that only 40% of the systems that consistently fluoridated pro- duced a water with a fluoride concentration within the range specified in the drinking water standards. These data show that additional sur- veillance and operator training in methods of good fluoridation practice may be necessary on a national scale. The one sample exceeding the lead limit was collected from the Huntington, West Virginia, water supply. In any large set of water samples at least one percent exceeds the lead limit which reflects the use of lead pipe and solder for copper pipes. Lead would be of concern if it were consistently over the limit at a sampling point. Two supplies exceeded the new mercury limit, the Artesian Water Company of New Castle County, Delaware, and the Tennessee American Water Company of Chattanooga. Mercury was detected at the Chattanooga Supply in 1970 .also but at half the concentration found in this survey. The Artesian Water Company uses wells and lower mercury concentrations were found in the past. 132 ------- TafcHe I Community Water Supply Study 2595 DISTRIBUTION SAMPLES FROM 969 PUBLIC WATER SUPPLY SYSTEMS Recommended Standards A. B. S. Arsenic Boron Chloride Color Copper CCE* Cyanide Fluoride Iron Manganese Nitrates Ra-226 Sr-90 Su-lfate Dissolved Solids Turbidity Zinc CAE* Limit mg/1 .05 .01 1.0 250.0 15.un 1.0 .2 .01 Varies 0.3 0.05 45.0 3 pCi/1 10 pCi/1 250.0 500.0 1-5. un 5.0 — Maximum Concentration .41 .10 3.28 1950.0 49.0 8.35 .56 .008 4.40 26.0 1.32 127.0 135.9 2.0 770.0 • 2760.0 53.0 13.0 .81 Percent Exceeding 0.0 .4 .8 1.2 .7 1.6 1.2 0.0 4.6 8.6 8.2 2.1 .6 0.0 1.8 8.5 2.4 .3 — Mandatory Standards Arsenic Barium* Boron Cadmi urn Chromium Col i forms Cyanide Fluoride Gross Beta Lead Selenium Silver .05 1.0 5.0 .01 .05 1/100 ml. 0.2 Varies 1000 pCi/1 .05 .01 .05 .10 1.55 3.28 .011 .08 TNTC .008 4.40 154.0 .64 .07 .026 .2 .1 0.0 .2 ' .2 8.8 0.0 2.1 0.0 1.4 .4 0.0 AWWA Goal mg/1 .20 PHS PHS — 3.un .2 .04 PHS PHS .05 .01 KHS PHS PHS — 200.0 0.1 1.0 .10 AWWA PHS PHS PHS PHS PHS 0.0 PHS PHS 100 pCi/1 PHS PHS PHS Goals Percent Exceedinc .2 .4 .8 — 9.9 15.5 25.5 0.0 4.6 44.5 31.0 2.1 .6 0.0 — 48.7 90.6 4.4 26.6 Goals .2 .1 0.0 .2 .2 11.7 0.0 2.1 <.l 1.4 .4 0.0 *These constituents were evaluated only on selected samples. remainder were assumed not to exceed the limits or goals. The 133 ------- TABLE II u> ;UBSTANCE ABS Arsenic Barium Cadmium Chloride Chromium Copper CCE CAE*** Cyanide Fluoride** Iron Lead Manganese Mercury*** Nitrate Selenium Silver Sulfate IDS Zinc EXTENT OF NON-COMPLIANCE WITH 1962 USPHS DRINKING WATER STANDARDS ANALYSES* REPORTED 282 501 405 541 641 535 555 47 19 237 633 652 544 623 389 582 344 411 592 575 523 SAMPLES FAILING DWS RECOMMENDED LIMIT NO. 17 2 6 65 46 27 75 1 1.8 2.7 0.4 12.7 10.0 7.4 4.6 13.0 0.2 SAMPLES FAILING DWS MANDATORY LIMIT NO"! % *Total sampling points - 702 **DWS varies with temperature and not flagged *,**Proposed for 1973 Federal Drinking Water Standards SAMPLES WHOSE MINIMUM DETECTABLE LIMIT IS HIGHER THAN THE DWS 0.4 1 6 0.2 1.5 N0_. 1 12 5 1 8 8 32 1 3 0.4 2.2 0.9 0. 1. 1, 8, 0.3 0.7 ------- TABLE III Percent of Homes with a Sample Exceeding the DWS Boston Seattle Cd 0 7% Cr 0 Cu 19% 24% Fe 9% 76% Pb 65% 24% Mn - 5% Zn 0 10% E. ASBESTOS STUDIES Because of the potential health effect of asbestos fibers in drink- ing water, the U. S. Environmental Protection Agency has conducted and is currently conducting several studies in an attempt to determine how widespread the problem of asbestos contamination is. This section sum- marizes the work of EPA's Office of Research and Development and its early findings relative to this issue. 1. Duluth Study The possibility of asbestos contamination of drinking water at Duluth was discussed by Mrs. Arlene Lehto of Duluth at an International Joint Commission hearing held in Duluth in December 1972. After this, in March 1973 the U. S. EPA National Water Quality Laboratory in Duluth began monitoring the Duluth water supply for amphibole mass by x-ray diffraction. The presence of amphibole fibers was indicated by electron micrographs. This analytical work is continuing. One report was pub- lished by Cook et al.5 They indicated that the total content of amphi- bole minerals in the Duluth water supply averaged 0.19 mg/1 from March 1973 to January 1974. In June 1973 the U. S. EPA announced that the drinking water of Duluth and North Shore Lake Superior Communities contained asbestiform fibers. 135 ------- An extensive sampling program was undertaken by Region V in the summer of 1973 in order to learn about the extent of the asbestos con- tamination problem in Western Lake Superior. Fairless reported on the results of this study.6 Fairless indicated that in Western Lake Superior, particulate matter from the Silver Bay area was carried by lake currents to the Duluth area and then along the southern shore of the lake (north- ern Wisconsin and Michigan). The trend for results of both amphibole mass and asbestiform fiber analysis is that the values are lightest at Beaver Bay, Minnesota, decreasing from there to Duluth, and then to Ashland, Wisconsin and Marquette, Michigan. In 1974 a pilot plant was operated at Duluth's Lakewood Pumping Station for fiber removal research. From May through September Lake Superior water that was pumped into the distribution system at Duluth was analyzed for amphibole mass and asbestiform fibers. Analytical data are shown in Figures 1 and 2. Because of the state of the art in EM analysis for asbestiform fibers in water, a laboratory can be expect- ed to be internally consistent on fiber count from sample to sample, but comparison of results between laboratories is usually not possible. Because no standard method yet exists, some laboratories may count con- sistently higher than others or vice versa. Pilot plant results for raw water at "the Lakewood Intake (Duluth's drinking water) showed amphibole and chrysotile fiber counts typically in the range of 0.5 x 106 f/1 to 2 x 106 f/1, with some samples either higher or lower than that range. The results of EPA work on waters of western Lake Superior have established firmly the existence of asbestiform fibers. Studies of ways to reduce the fiber content of drinking water are described else- where in this report, Appendix VI. 2. Asbestos-Cement Pipe Studies a. Field Studies In an effort to study the effect of waters of various corrosive- ness on asbestos-cement pipe several systems utilizing A/C pipe were selected for study. In each case, samples were taken of the source and at two places in the distribution system. These will be followed up by analyzing samples from the same locations every month for at least nine months, so as to cover any seasonal variations. Initially three sites were selected. When these are completed or as time permits, others of high pH and hardness will also be selected for study. b. Controlled Pipe-Loop Study The objective of this study is to determine the influence of water velocity, aggressiveness of water and elapsed time on the erosion. 136 ------- (0 10 9 : 8 - § DC 7 LU CO cc LLJ eg 5 LL 4 i= 3 CO LLJ CO 3 2 -o AMPHIBOLE -• CHYRSOTILE 5 15 25 APRIL 15 25 MAY 5 15 25 JUNE 5 15 25 JULY 5 15 25 AUGUST 5 15 25 SEPTEMBER FIGURE 1. ONTARIO RESEARCH FOUNDATION ASBESTIFORM FIBER COUNTS RAW WATER AT DULUTH LAKEWOOD INTAKE 1974 ------- OJ CO o SUSPENDED SOLIDS • AMPHIBOLE 5 15 25 APRIL 5 15 25 JUNE 5 15 25 JULY 5 15 25 AUGUST 5 15 25 SEPTEMBER FIGURE 2. ENVIRONMENTAL PROTECTION AGENCY NATIONAL WATER QUALITY LABORATORY AMPHIBOLE MASS CONCENTRATION RAW WATER AT DULUTH LAKEWOOD INTAKE - 1974 ------- of asbestos fibers from asbestos-cement pipe. The influence of tapping the pipe wall will also be studied. To conduct this study, a "pipe-loop" was constructed of 100 ft of 4-inch and 6-inch diameter asbestos cement pipe. Water is pumped through both pipes at approximately 150 gpm, producing a velocity of 3.8 ft/sec in the 4-inch pipe and 1.7 ft/sec in the 6-inch pipe. As the water en- tered the pipe test section, it is filtered through a 1 ym pore diameter filter. The pH and hardness of the water are adjusted and maintained at any desired level. Water circulates continuously through the pipe loop. Each day the water passes through an equivalent of 62 miles of 4-inch pipe and 28 miles of 6-inch pipe. Once each week water is diverted from the exit end of each pipe specimen and 300-500 liters passes through a 0.45 ym pore Mi Hi pore filter. Just prior to sampling, water entering the pipe loop is divert- ed through large 0.45 ym pore Mi Hi pore filters to assure that during sampling the water entering the pipe loop is nearly particle-free. Therefore any fibers appearing at the exit end resulted from the test length of pipe. After sampling, the Mi Hi pore filters are subjected to EM analysis using the technique cited in Reference 7. The present plan is to study nine combinations of hardness and pH ranging from a low hardness, low pH water (hardness - 20 mg/1 as CaC03, pH = 5.5) to a high hardness, high pH water (hardness - 400 mg/1 as CaC03, pH = 9.5). Both the hardness and the pH will be varied between the extreme limits in three steps. The current test involves the use of the low hardness, low pH water. This test has been under way for about two months. Figure 3 shows the results to date. Because of the large volume of water passed through the sampling filters, the test is much more sensitive than the routine analysis for asbestos fibers. This is why such low fiber count can be reported with some degree of reliability. 3. Finished Water at Various Locations in U. S. In the process of attempting to develop a procedure for the routine analysis of asbestos in water the Water Supply Research Laboratory routinely selected some Interstate Carrier (mostly) Water samples (finished water) received in our laboratory for chemical analysis. The developed procedure has now, or soon will be, published in the Proceed- ings of the Water Quality Technology Conference, AWWA?. 139 ------- The water supplies analysed and the results obtained were as follows: City Fibers/1 x IP6 Duluth, Minnesota 1.1 to 4.8 A BDL to 0.4 C Abilene, Texas BDL Cincinnati, Ohio NSS Cheyenne, Wyoming NSS Columbia, South Carolina 0.13 C Cairo, Illinois NSS Anchorage, Alaska 0.07 A Jackson, Mississippi (2 grids) 0.25 to 0.7 C Ashland, Kentucky BDL Pittsburgh, California NSS N. Troy, Vermont (2 grids) 0.98 to 2.2 C Enosburg, Vermont Q.-£& r Brattleboro, Vermont 0.11 C Eden, Vermont (Spring) 0.38 C St. Louis, Missouri NSS Seattle, Washington 1.812 A (Tolt Pipe Line) 2.464 C Seattle, Washington BDL (Cedar River System) NSS C Elizabeth, New Jersey BDL Amarillo, Texas , 0.09 A Boulder, Colorado BDL 140 ------- City Fibers/1 x IP6 Glens Falls, New York BDL Jonesboro, Arkansas NSS New Haven, Connecticut NSS Clarksville, Tennessee 0.09 C Jersey City, New Jersey 0.16 C Erie, Pennsylvania 0.07 C Newport, R. I. (2 grids) 0.04 to 1.0 C Little Rock, Arkansas 0.27 C Charlottesville, Virginia NSS Skidaway Island, Ga. (2 grids) 1.74 to 2.03 C Jericho - Underbill, Vermont NSS .Crystal Springs, Vermont NSS Niagara Falls, New York NSS Rochester, New York BDL Buffalo, New York 0.13 C San Francisco, California 1.54 C Nashville, Tennessee (2 grids) 0.43 to 0.80 C South Pittsburgh, Pa. 0.21 C Independence, Missouri (2 grids) 0.36 to 0.58 C Montgomery, Ala. (2 grids) BDL to 0.12 C Ft. Lauderdale, Florida NSS Indianapolis, Indiana 0.18 C Kansas City, Missouri 0.07 C 141 ------- City Fibers/1 x 10s Springfield, Missouri 0.30 C Melbourne, Florida NSS Tulsa, Oklahoma BDL Wilmington. Delaware 0.29 C Bethlehem, Pa NSS Fairbanks, Alaska BDL Elmira, New York NSS Muskogee, Oklahoma BDL Richmond, Harrington, Vt. • NSS Quarry Hill, Vermont NSS Tuscaloosa, Alabama 0.45 C Birmingham, Alabama BDL Topeka, Kansas NSS Greenville, S. Carolina NSS Yuma, Arizona 0.12 C Dayton, Ohio NSS Washington, D. C. NSS Sacramento, California NSS Miami, Florida BDL San Juan, Puerto Rico NSS Chattanooga, Tennessee 0.13 C BDL - Below detection limit NSS - Not statistically significant (less than 5 fibers in 20 fields) A - amphibole C - chrysotile 142 ------- NET CHRYSOTILE FIBERS OUTLET MINUS INLET -B 4-INCH ASBESTOS CEMENT PIPE -A 6-INCH ASBESTOS CEMENT PIPE 10,000 oc HI •5 1,000 cc Ol 0. w cc W CO UD CO I 100 10 pH 5,5 HARDNESS 20 mg/l AS CaCO3 1 MILLION GALLONS PASSES THROUGH EACH PIPE IN 4.6 DAYS. (-1,000 CONTAMINATION SUSPECTED) I 1 1 L 10 15 20 25 30 FIGURE 3. CUMULATIVE VOLUME OF WATER THROUGH SYSTEM - GALLONS x 10* 143 ------- As can be noted, of 63 supplies analyzed, only nine supplies (14%) had counts over 0.5 x 106 fibers per liter and of these, only five (8%) were over 1.0 by 106, namely, Duluth, Minn.; North Troy, Vermont; Seattle (Tolt) Washington, Skidaway Isl, Ga., and San Francisco, California. Eleven (18%) had fiber counts below the detection limits of the method. 144 ------- ACKNOWLEDGEMENTS The first four sections of Appendix V were written by Gunter Craun and Leland McCabe. The fifth section on asbestos was prepared by E. McFarren, R. Lishka, J. Millette, G. Logsdon, R. Buelow, J. Agee, J. Symons, P. Cook, G. Glass, B. Fairless. ------- REFERENCES 1. McCabe, L. J., Symons, J. M., Lee, R. D., and Robeck, G. G. Survey of Community Water Supply Systems, Journal Am. Water Works Assoc. Vol 62(11), 670-687, November 1970. 2. Craun, G. F., and McCabe, L. J., "Overview of Problems Associated with Inorganic Contaminants in Drinking Water." Proceedings National Symposium on the State of America's Drinking Water, Chapel Hill, North Carolina (In Press). 3. Hertsch, F. F. and Maddox, F. D., Fluoridation Practice in Wisconsin, Journal Am. Water Works Assoc., Vol 63, 778-782, 1971. 4. USEPA Report, Chemical Analysis of Interstate Carrier Water Supply Systems, October 1973. 5. Cook, Philip M., Glass, G. E., and Tucker, J. H., "Asbestiform Am- phibole Minerals: Detection and Measurement of High Concentrations in Municipal Water Supplies," Science. 185, 853-855 (September 6, 1974). 6, Fairless, B., "Asbestos Fiber Concentrations in the Drinking Water of Communities Using the "Western Arm of Lake Superior as a Potable Water Source," U. S. Environmental Protection Agency, Region V, Surveillance and Analysis Laboratory, Chicago, Illinois, 17 pp. Mimeo 7. McFarren, E. F., Millette, J. R. and Lishka, R. J., "Asbestos Anal- ysis by Electron Microscope," In Proceedings AWWA Water Quality Conference, Dallas, Texas, December 1974 (In Press) 146 ------- APPENDIX VI PRELIMINARY RESULTS OF PILOT PLANTS TO REMOVE WATER CONTAMINANTS Prepared by 0. Thomas Love, Jr. J. Keith Carswell Alan A. Stevens Thomas J. Sorg Gary S. Logsdon Compiled By James M. Symons Water Supply Research Laboratory National Environmental Research Center Office of Research and Development Cincinnati, Ohio ------- ' APPENDIX VI PRELIMINARY RESULTS OF PILOT PLANTS TO REMOVE WATER CONTAMINANTS Table of Contents Page A. Introduction 151 B. Treatment for the Removal of Organic Contaminants .... 151 1. Specific Organic Compounds 151 a. Literature Reports 151 b. Naphthalene 151 c. Bis-(2-chloroethyl) ether and Bis-(2-chlorois- opropyl) ether 152 d. Chloroform, Bromodichloromethane, Dibromo- chloromethane, and Bromoform 152 e. 1,2-Dichloroethane and Carbon Tetrachloride • • 152 f. Trihalomethane Precursors 152 1) Description of Pilot Plant 152 2) Chiorination Experiments 155 3) Ozonation Experiments 155 4) Controlled Bench-scale Experiments .... 158 2. General Organics , 160 a. Pilot Plant 160 b. Column Studies 160 1) Upflow-Counter-current 160 2) Upflow-cocurrent 163 3. Future Plans 163 a. Pilot Plant Studies 163 b. Controlled Storage Studies 166 c. Column Studies 166 4. Acknowledgments 166 5. References 166 148 ------- Page C. Treatment for the Removal of Trace Metal Contaminants . . 167 1. Introduction 167 2. Research Program 167 a. Jar Tests 167 1) Procedure 167 2) Analytic Methods 168 b. Pilot Plant 168 1) Description and Operation 168 2) Analytic Methods 171 3. Results 171 a. Jar Tests *•• 171 1) Previous work 171 2) Barium 171 b. Pilot Plant Studies 172 1) Mercury 172 2) Cadmium 173 3) Arsenic 173 4) Selenium 174 c. Summary of results 174 4. Radium-226 177 a. Introduction 177 b. Results 177 c. Discussion 177 d. Future Plant 178 5. Acknowledgments 178 6. References 179 D. Treatment for the Removal of Asbestiform Fibers 179 1. Introduction 179 2. ' Scope of Study ' 180 149 ------- Page 3. Experimental Methods and Equipment 180 a. Equipment 180 b. Analytical Methods 181 4. Results 182 a. Raw Water Quality 182 b. Asbestiform Fiber Removal by Filtration 183 5. Discussion 186 a. Asbestiform Fiber Removal 186 b. Efforts to Develop Rapid Detection Methods • • • 195 6. Future Research 195 7. Conclusions 196 8. Acknowledgments 197 9. References 197 150 ------- PRELIMINARY RESULTS OF PILOT PLANTS TO REMOVE WATER CONTAMINANTS A. INTRODUCTION This report will summarize the in-house and out-of-house research conducted by the Standards Attainment Branch of the Water Supply Research Laboratory and their predecessors concerning the treatment technology required for the removal of specific contaminants present in raw water. Because many of these projects are on-going, future plans will also be included. The report will be divided into three general contaminant areas -- organics, inorganics, and asbestos fibers. B. TREATMENT FOR THE REMOVAL OF ORGANIC CONTAMINANTS 1. Specific Organic Compounds a. Literature Reports In 1965, Robeck, Dostal, Cohen and Kreissl demonstrated that coal-base granular activated carbon (GAC) partially exhausted for carbon-chloroform extract (CCE-hf) removal, could reduce the concentra- tion of dieldrin, lindane, 2,4,5-T, DDT, and parathion dosed into river water. In the same year Dostal, Pierson, Hager, and Robeck2 showed that seven compounds, listed below, present in the Kanawha River water after aeration could be reduced to below detectable concentrations by fresh (2-day old) GAC. These compounds were bis-2(2-chloroethyl) ether, 2-ethylhexanol, bis-(2-chloroisopropyl) ether, a-methylbenzyl alcohol, acetophenone, isophorone and tetralin. Forty days later, however, all of these compounds with the excep- tion of acetophenone, had broken through the GAC beds to a depth equal to an actual contact time of 3.8 minutes. Providing an additional 1.8 minutes of actual contact time did remove these seven compounds at this time, although another organic, ethyl benzene, was penetrating the GAC to a depth equal to 7.5 minutes of actual contact time -- almost twice that provided in conventionally operated GAC beds. b. Naphthalene About 7 months ago, a coal-base GAC column 28 in. deep was constructed, and Cincinnati tap water spiked with approximately 30 yg/1 of naphthalene passed down through it at a rate of 2 gpm/ft*. After 7 months of operation, the NVTOC front has penetrated to the extent that the 50% removal point is approximately 20 in. down the column, whereas the 50% removal point for naphthalene is only approximately 2 in. down the column. This test is continuing. 151 ------- c. Bis-(2-ch1oroethy1) ether and Bis-(2-chloroisopropy1; ether In recent studies, the effluent from a mini-sampler operating on Evansville, Indiana, finished water was analyzed and fresh coal-base GAC was shown to remove all detectable bis-(2-chloroethyl) ether and bis-(2-chloroisopropyl) ether. No information is available on how long GAC would continue to remove this material, however. d. Chloroform, Bromodichloromethane, Dibromochloromethane and Bromoform To investigate the ability of GAC to remove chloroform and the other three trihalomethanes two 28-in. deep glass columns were con- structed. One contains a coal-base GAC and the other a lignite-base GAC. The columns were arranged such that Cincinnati tap water flowed down through the columns at a rate of about 2 gpm/sq. ft. Figure 1 shows that after 4 weeks of operation the ability of the coal-based GAC to remove chloroform was seriously restricted. The trend of the data from the lignite-base GAC would indicate that its ability to remove chloroform was somewhat greater than the coal-base material. Figure 2 shows a similar result for bromodichloromethane. The dibromochloromethane concentration applied to these two columns varied between none found and 4 ug/1. None has yet appeared in the effluent from either column. N_o bromoform was found in the applied water during this study. These two columns were started at different times, however, and therefore received different general organic loads. Whether or not this influenced the trihalomethane removal patterns is not known at this time. e. 1,2-Dichloroethane and Carbon Tetrachloride No 1,2-Dichloroethane was found in the Cincinnati tap water during the study. Carbon tetrachloride appeared occasionally at concen- trations from <0.2 to 5.6 yg/1 in the Water applied to the two GAC col- umns, but none ever appeared in the effluent from either column. f. Trihalomethane Precursors 1) Description of Pilot Plant A pilot water treatment plant was constructed of stain- less steel, Teflon and glass in order to minimize contamination from structure materials during experimentation on the formation and removal of trihalomethanes. The pilot plant uses untreated Ohio River water as 152 ------- CJ1 co COAL-BASE LIGNITE-BASE APPLIED CHLOROFORM RANGE, 34-72 /xg/l 345 WEEKS IN OPERATION FIGURE 1. REMOVAL OF CHLOROFORM FROM TAP WATER WITH GRANULAR ACTIVATED CARBON ------- 100 80 o D 60 in c: -. I- S" Z L Art O 40 cc III o. 20 I •a- BROMODICHLOROMETHANE RANGE IN APPLIED WATER 8-20 COAL-BASE LIGNITE-BASE I 345 WEEKS IN OPERATION 6 8 FIGURE 2. REMOVAL OF BROMODICHLOROMETHANE FROM TAP WATER WITH GRANULAR ACTIVATED CARBON ------- a source, made available through the assistance and cooperation of the Cincinnati Water Works. Following conventional pretreatment with alum (without predisinfection), the settled water flow has been divided between: 1) a dual media (sand/coal) filter (A), 2) a coal-base gran- ular activated carbon filter (B), and 3) a dual-media filter followed by a coal-base granular activated carbon filter (CD). The filtered water is then either ozonated, or chlorinated, or both. Flow, headless, turbidity, temperature and pH are monitored daily. Figure 3 is a sche- matic diagram of the pilot plant. Samples for trihalomethane and non- volatile total organic carbon analysis are collected with zero head- space in muffled, glass vials. 2) Chlorination Experiments The pilot plant has been running continuously and the first experiments have focused on eliminating the haliform reaction through removal of the precursors with GAC. Using chlorine dosages of 2-3 mg/1 and contact times of 30 minutes and 4 days, studies have shown that after 3 to 4 weeks of operation, sufficient materials are being passed through the GAC beds to produce measurable amounts of chloro- form (See Table 1). This experiment is continuing. 3) Ozonation Experiments The purpose of this portion of the organics research project is to determine whether post-ozonation can be used to oxidize trihalomethane precursors to compounds that will not react during post- chlorination. A small (1.5 in diameter) glass contact chamber is used to provide about 5 minutes of contact time between the pilot plant filter effluents and an ozone-oxygen gas mixture. Filter A, B or CD effluents are applied to the top of the contactor, while the ozone- oxygen output from a pilot-plant scale ozone generator is applied at the bottom. Both the gas flowrate and the ozone concentration can be varied. Batches of ozonated effluent are collected for reaction with various chlorine concentrations. Initially, disinfection-level (0.5-0.7 mg Og/1 H20 ozone doses, followed by rather high (8 mg C12/1 H20) post chlorination doses were applied to the filter effluents. When very low (<1 yg/1) chloro- form concentrations were produced after a one-half hour chlorine contact period, it was decided to dose and store (at 25°C) effluent samples for a longer time period to follow trihalomethane development. Also, chlorinated effluent samples (without ozonation) wo.uld be stored as a control to better show any changes produced by the ozone. 155 ------- COAGULANT FEED PUMP PERISTALTIC TYPE RAPID MIX - STATIC MIXER THEO. RET. TIME - <1 MIN. en MIXER RAW WATER STORAGE THEORETICAL RETENTION TIME - 66 HRS., VOL. - 1500 L. RAW WATER PUMP DIAPHRAM TYPE CHLORINE CONTACTOR VOL. - 3.2 L. THEO. RET. TIME - 30 MIN. ACTUAL RET. TIME - 29 MIN. OZONE CONTACTOR VOL. - 0.8 L. THEO. RET. TIME - 5 MIN. ACTUAL RET. TIME -6 MIN. TR. VOL. - 30 L. THEO. RET. TIME - 75 MIN. ACTUAL RET. TIME - 12 MIN. SEDIMENTATION BASIN VOL. - 270 L. THEO. RET. TIME -675 ACTUAL RET. TIME - 205 MIN. CHLORINATED EFFLUENT FILTERS VOL. - 1.9 L. THEO. RET. TIME - 20 MIN. ACTUAL RET. TIME -- 4 MIN. LOADING - 2 GPM/FT2 FLOW - 90-95 ML/MIN. FILTERS A & 10" SAND (ES - 0.38, UC - 1.3) 20" COAL (1% ANTHRAFILT. ES - 1.2, UC - 1.5) FILTERS B & D LfJ flX •' OZONE EFFLUENT STATIC MIXER 30" CARBON (CALGON FILTRASORB 200 ES - 0.6, UC - 1.6) FIGURE 3. STAINLESS STEEL PILOT WATER TREATMENT PLANT ------- TABLE 1. COMPARISON OF MIXED MEDIA AND GRANULAR ACTIVATED CARBON FOR REMOVING CHLOROFORM PRECURSORS CHLOROFORM CONCENTRATIONS, NON-VOLATILE TOTAL ORGANIC CARBON (NVTOC) CONCENTRATIONS, mg/l SYSTEM MIXED MEDIA (SAND/COAL) (A) GRANULAR ACTIVATED CARBON (B) MIXED MEDIA FOLLOWED BY GRANULAR ACTIVATED CARBON (CD) 1ST WEEK CI2 CONTACT TIME 30 MIN. 2 NVTOC = 1.02 <0.2 NVTOC = <0.05 <0.2 NVTOC = <0.05 2ND WEEK CI2 CONT. TIME 30 MIN. 4 DAY 2 13 NVTOC = 0.87 <0.2 <0.2 NVTOC = <0.05 <0.2 <0.2 NVTOC = <0.05 3RD WEEK CI2 CONT. TIME 30 MIN. 4 DAY 3 13 NVTOC = 0.60 <0.2 0.5 NVTOC = <0.05 0.2 0.7 NVTOC = <0.05 4TH WEEK CI2 CONT. TIME 30 MIN. 4 DAY 2 16 NVTOC =1.16 <0.2 2 NVTOC = 0.65 <0.2 2 NVTOC = 0.59 5TH WEEK CI2 CONT. TIME 30 MIN. 4 DAY 1 10 NVTOC = 0.70 0.2 1 NVTOC = 0.35 <0.2 1 NVTOC = 0.10 On ------- The results of this first storage study show that a dis- infection-level ozone dose had no apparent effect on the trihalomethane concentrations produced in Filter B and CD effluents after 6 days of storage. In Filter A effluent, ozonation appeared to cause an increase in chloroform concentration and a decrease in bromodichloromethane con- centration. See Table 2. This latter situation will receive further study. Other future studies will investigate the effect of higher ozone doses and/or longer ozone contact periods. 4) Controlled Bench-Scale Experiments In an effort to understand the mechanism of trihalo- methane formation and the factors that influence it, experiments are being conducted under controlled conditions in sealed containers changing one variable at a time. At the start of the experiment several containers are prepared in replicate. Periodically over a 7-day period a container is opened and the trihalomethanes measured. Thus the in- fluence of the variable under study on the rate and extent of trihalo- methane formation can be determined. One bench-scale study investigated the formation of tri- halomethanes during chlorination of raw and different types of treated water from the Ohio River. This study was conducted in sealed con- tainers at constant pH and 25° Celcius. Some preliminary observations were: 1. When adequate chlorine is added to satisfy chlorine demand for the duration of the experiment, chlorination of raw water yields approximately seven times as much chloroform as does chlorination of the effluent of the dual-media pilot plant filter (A) and approximately 80 times as much as does chlorination of the effluent of the fresh GAC filter (B) (207yg/l, 32 vg/1, and 2.7 yg/1, respectively, in 7+ days). The reason the chloroform production is so low in filter effluent compared to the raw water is not known at this time, but in future studies settled water will be included in the series to determine at what step in treatment the precursors are removed. 2. Of the trihalomethanes under study, chloroform is formed in the highest concentration with bromodichloromethane and dibromochloro- methane following in approximate rations of 100:15:1. 3. With the waters tested, those with a higher chlorine demand gave a greater trihalomethane concentration upon chlorination, but trihalomethane formation accounted for only about 3% of the chlorine consumed. Therefore other chlorination by-products are being formed, both organic and inorganic. 158 ------- Sample TABLE 2 SUMMARY OF OZONATION EXPERIMENT Bromo- Dibromo- Chlorine dichloro- chloro- Contact Time Chloroform methane methane Bromoform A + C12 A + C12 A + 0, + C17 O L. A + 03 + C12 B + C12 B + C12 B + 03 + C12 B + 03 + C12 CD + C12 CD + C12 CD + 0~ + C19 3 2 CD + 03 + C12 1/2 hour 6 days 1/2 hour 6 days 1/2 hour 6 days 1/2 hour 6 days 1/2 hour 6 days 1/2 hour 6 days 4 6 NF 15 0.3 2 NF 3 0.2 2 0.2 2 NF 14 NF 8 NF 3 NF 3 NF 3 NF 4 NF 4 NF 3 NF <] NF 2 NF <-1 NF NF NF NF NF NF NF NF NF NF NF NF NF NF = None Found All Trihalomethane Concentrations in yg/1 159 ------- 4. In all waters tested, trihalomethane production continued as long as a measurable chlorine residual was present, but at a decreasing rate. The initial rate of chloroform formation in the raw river water was about 10 pg/l/hr for the first six hours. The rate of formation was very low, however, for GAC effluent throughout the 7+ days. 2. General Organics The monitoring of specific organics may be beyond the capability of most water utilities for some time to come. Therefore, treatment for the removal of specific organics may be accomplished by providing treat- ment that will produce a water with a very low concentration of a gen- eral organic parameter such as non-volatile total organic carbon, although monitoring for specific organics is the only method of assuring their removal. a. Pilot Plant In addition to the study of the elimination of trihalomethane precursors, NVTOC concentrations are being measured at various stages of the pilot plant in an attempt to determine under what conditions very low NVTOC concentration water can be produced for extended periods of time. Figure 4 shows the average relative concentrations of non- volatile total organic carbon at various stages of treatment for the first three weeks of operation and operational weeks three to six. Figure 5 compares the percent removal of NVTOC by coal-base GAC, both with and without prefiltration of the influent water. The presence of carryover floe in the influent to filter B is not interfering with NVTOC removal. b. Column Studies 1) Upf 1 ow-counter-current The objective of this study is to determine if low con- centrations of NVTOC can be continuously maintained in the effluent of a GAC filter by periodic removal of a portion of the exhausted GAC, replacing it with fresh GAC. A small (1.5 in diameter) glass column has been placed in operation. Cincinnati tap water (approximately 80 ml/min) is applied to the bottom of the column, flows upward tlvough the GAC bed, and overflows from the top of the column to a vo'jme measuring device. When the effluent NVTOC limit is exceeded one-half of the GAC bed is removed (as a slurry) from the bottom of t.ie column. and an equal quantity of fresh GAC is added at the top. 160 ------- cr> •To 1 ,\J 0.75 LU O V) LU > 0.5 £r _i LU CC 0.25 0 — CC LU <£ 5 5 < CC AVERAGE NVTOC IN RAW WATER = 2.5 mg/l 0 LU j H LU W *o: O LU LU h- H < < 3 _l Z) a < o o CC LU H > 0 CC LU LU CC H uj < < h- 52 =! 01 « U- •^ Q Si uu z Q °~ O m "J CQ uu ^^ ^^ ^^ ^^ Lbt _ ~^ ^ 5 u. 0 j i CC LU ^ S s CC AVERAGE NVTOC IN RAW WATER = 3.3 mg/l CC LU H ^ 5 LU CC LU —I Si Q CC LU H LU Q: < £ 2 . S'th si § en LU ^ S CC Q Q LU O r( °3 LU CC X LU 1 5 ^ 0-3 WEEKS 3-6 WEEKS TIME IN OPERATION t FIGURE 4. RELATIVE NON-VOLATILE TOTAL ORGANIC CARBON REMOVAL DURING WATER TREATMENT ------- 100 IJ- o o z g o D Q UJ cc 80 60 LU u cc HI Q. 40 20 FILTER CD FILTER B NVTOC RANGE IN APPLIED WATER 0.8 - 1.6 mg/l 10 15 20 DAYS IN SERVICE 25 30 35 40 FIGURE 5. FILTRATION/ADSORPTION VS. POST ADSORPTION FOR NON-VOLATILE TOTAL ORGANIC CARBON REMOVAL ------- For the initial test series, an 8-inch bed of GAC was chosen. Preliminary observations, after 20 days of operation, include: 1. The effluent NVTOC limit can be maintained for only 2-3 days before GAC removal and addition is necessary. 2. Exhaustion of the GAC bed appears to occur much more rapidly than in similar downflow beds, indicating possible wall effects and/or flow channelization within the column. This study will continue, with future efforts directed toward deeper GAC beds and/or larger diameter columns. Thus far a 16-inch deep bed is performing more reliably. 2) Upflow - Co-current In an attempt to compare the performance of various types of GAC, six 3-inch diameter, 18-inch long columns were filled with six different types of GAC produced by three different manufacturers. Cin- cinnati tap water was continuously passed upflow through each column at a rate of about 3 gpm/sq ft for 32 weeks. The influent and effluent concentrations of NVTOC and'carbon-chloroform extract (CCE-m) were moni- tored weekly during this time. Data from each column were averaged for several four-week intervals and plotted in Figures 6 and 7. All of the data fell within the envelopes shown, indicating that the type of GAC had little influence on performance. These data also show that the life of the GAC in these columns was fairly short for the removal of these two parameters. References 3 and 4 contain additional information on the performance pf activated carbon. 3. Future Plans a. Pilot Plant Studies Current chlorination and ozonation studies will continue to determine how the aging of the GAC columns will alter the trihalomethane formation patterns. When these experiments are completed they will be repeated using lignite-base GAC in the columns. Possible future studies will cover the influence of powdered activated carbon, pre-disinfection with chlorine or ozone, addition of chlorine, just prior to filtration, and the use of chlorine dioxide. Also studies on removal of raw water contaminants will be conducted. 163 ------- CD -p. 100 80 o o 60 UJ DC I- z 111 oc UJ Q. 40 20 0-4 4-8 ALL DATA POINTS WITHIN ENVELOPE. _L _L 8-12 12-16 16-20 WEEKS IN OPERATION I 20-24 I 24-32 FIGURE 6. NON-VOLATILE TOTAL ORGANIC CARBON REMOVAL FOR SIX TYPES OF GRANULAR ACTIVATED CARBON ------- 100 80 UJ u o O 5 UJ cc 8 H- Z UJ O CC UJ CL ALL DATA POINTS WITHIN ENVELOPE. 60 40 20 _L 1 1 1 0-4 4-8 8-12 12-16 WEEKS !N OPERATION 16-20 20-24 FIGURE 7. CARBON CHLOROFORM EXTRACT (CCE-m) REMOVAL FOR SIX TYPES OF GRANULAR-ACTIVATED CARBON 165 ------- b. Controlled Storage Studies Possible future controlled storage studies will include experiments on chlorination of specific tribalomethane precursors and the influence of temperature and form and concentration of chlorine residual on the formation of trihalomethane. Also other chlorination by-products will be studied to follow-up on the finding that nitrome- thane becomes chloropicrin and m-xylene becomes chlorxylene upon chlor- ination. c. Column Studies Work will continue studying the ability of columns of GAC operated upflow, downflow, co-, and counter-current to effectively remove organics as measured by general organic parameters. Because the column test described in the Naphthalene Section B(l)(b) was so successful, it will be repeated using other compounds. Possible candidates are: benzene, bis-(2-chlorethyl) ether, carbon tetrachloride, phthalic anhydride, beta-chloroethylmethyl ether, octa- decane, DDT, dieldrin and aklrin. 5. References 1. Robeck, G.G., Dostal, K.A., Cohen, J. and Kreissl, J.F., "Effectiveness of Water Treatment Processes in Pesticide Removal," Journal American Water Works Association, 57, 2, 181-199 (February 1965). 2. Dostal, K.A., Pierson, R.C.., Hager, D.G. and Robeck, G.G., "Carbon Bed Design Criteria Study at Nitor, West Virginia," JAWWA. 57, 5, 663-674 (May 1965). 3. Activated Carbon in Water Treatment, University of Reading Conference, April 3-5, 1973. The Water Research Association, Medmenham, Marlow, Buckinghamshire, SL7, 2HD, United Kingdom. 4. Love, O.T., Jr., Carswell, O.K., Stevens, A.A., and Symons, J.M., "Evaluation of Activated Carbon as a Drinking Water Treatment Unit Process," Water Supply Research Laboratory, United States Environmental Pro- tection Agency, Cincinnati, Ohio, March 1975, 17 pp Mimeo. 166 ------- C. TREATMENT FOR THE REMOVAL OF TRACE METAL CONTAMINANTS 1. Introduction The Federal Proposed Interim Drinking Water Regulations (IDWR)1 established limits for a number of substances in water including various trace metals. For many years, these limits were rarely exceeded and, therefore, knowledge on treatment methods to remove these contami- nants from water was of only minor concern. In recent years, however, the awareness of trace metal contamination has increased for various reasons, including improved analytical procedures and more frequent and comprehensive surveillance activities. As a result, this awareness has stimulated the interest and concern for information and knowledge on the removal of trace metals from water by conventional treatment methods. In anticipation and response to the need for this information, the Water Supply Research Laboratory, U. S. Environmental Protection Agency developed a research program on the removal of trace inorganic substan- ces in water by conventional treatment processes. The trace metals in the IDWR that have been, or are presently being studied are mercury, barium, arsenic, selenium, cadmium and radium*. Chromium and lead will be studied in the near future. 2. Research Program The WSRL research program consists of two phases: (1) experiments in the laboratory using jar test apparatus and (2) pilot plant tests using a 2 gpm water treatment pilot plant. a. Jar Tests 1) Procedure The laboratory jar test methods have been described in detail by Logsdon and Symons^. The waters used in the jar test work were raw Ohio River water; raw well water from Glendale, Ohio; Cincinnati tap water; and a Midwestern groundwater containing barium. Except for the barium-laden water, the test waters were dosed with the contaminant to be studied, given 2 minutes of rapid mix after addition of the treat- ment chemical and 20 minutes of slow mix for the coagulation test, or they were given 3 minutes of rapid mix and 30 minutes of slow mix for softening. One hour of settling was used for all tests. Analyses were made for pH, turbidity, alkalinity, and in some cases, hardness, as well as for contaminant concentration. *Note: The radium research will be treated separately. 167 ------- 2) Analytic Methods Two methods were used for metal analyses. An atomic absorption spectrophotometer was used for analysis of non-radioactive contaminants in portions of the mercury, barium, and arsenic work, and occasionally in the selenium experiments. These methods have been described by Kopp et al.3, McFarren^, and Caldwell, et al.5 In some experiments, radiotracers were used with stable carriers. Radioactivity was measured using a shielded Nal (TI) crystal and single-channel analyzer. When radiotracers were used, the initial contaminant concen- tration was determined by adding the radibisotope, plus a known volume of stock carrier solution, to the water being treated and calculating the initial metal concentration. The removal percentage was taken as equal to the percentage of reduction of radioactivity. b. Pilot Plant 1) Description and Operation The WSRL pilot plant is capable of treating in parallel two 2 gpm flows of water. The plant has been designed to operate in a number of different configurations, but has been run primarily in a con- ventional manner for the metal removal studies with rapid mixing, floc- culation, sedimentation, and rapid granular filtration. The test waters used so far, Cincinnati tap water and raw well water fronr Glendale, Ohio, were stored in a 7500-gallon stain- less steel tank. This water was pumped to a constant-head tank that had an overflow line back to the storage tank. Water to be treated was pumped at 2.1 gpm through two rapid mix tanks having a theoretical detention time of about 2 minutes each. In the first mixing tank, the contaminants and, if required, soda ash for pH control, were added. The coagulant, Ferrifloc, alum, or lime for softening experiments, was intro- duced into the second mixing tank. After the rapid mix, the water was flocculated for one hour in a mechanically mixed flocculation basin, and then settled for about 6.5 hours (theoretical detention times). Except for excess lime softening experiments, the settled water was then filtered through either one or two parallel filter columns (4 1/4-inch diameter) at a rate of 4 gpm/sq ft. The filters were: (1) a dual-media filter containing 20 inches of No. 1 1/4 Anthrafilt over 12 in. of 0.4mm effective size Muscatine sand and (2) a granular activated carbon filter containing 30 inches of lignite-base, 0.8-0.9 mm effective size, granular activated carbon. When excess lime softening tests were run, the settled water was recarbonated to pH about 9.6 and settled (6.5 hours theoretical detention time) in a second sedimentation basin before being filtered. Figures 1 and 2 are schematic diagrams- of the pilot plant. 168 ------- PROPELLOR MIXERS RAW WATER CONSTANT HEAD TANK 01 ELEVATION VIEW ,OVERFLOW KU? RAPID MIX TANKS 2.3 MIN. FLOCCULATION BASIN, 1.0 HR. m 111 i n HI U M FILTER 0.10 SQ. FT. SEDIMENTATION BASIN, 6 HR. MOYNO PUMP METER (POSITIVE DISPLACEMENT) EXCESS FLOW P MOYNO PUMP METER (POSITIVE DISPLACEMENT) V) cc 111 l- LLJ 5 CN THEORETICAL DETENTION TIMES FOR 2.0 GPM STAINLESS STEEL HOLDING TANK, 7500 GAL. FIGURE 1 ------- CHEMICAL CHEMICAL ADDITION ADDITION TEMP. pH TURBIDITY >H RAW WATER RAW WATER CONST. HEAD TANK TEMP. pH TURBIDITY H. TEMP. pH TURBIDITY XED MIX r i i .j t F V n V B J A X MECHANICAL LOCCULATIONS ERT. PADDLES SETTLING LINE FOR TWO-STAGE SOFTENING 1 * f 1 1 1 1 f J W| "* 1 i FILTERS TOTAL FLOW METERS LINE TO STUDY DIRECT - FILTRATION Jfc XHC— — X IN-LINE MIXERS FIGURE 2. PILOT PLANT SCHEMATIC PLAN VIEW CHEMICAL ADDITION CHEMICAL ADDITION ------- Initially, the plant ran continuously for 40 hours (^ 5000 gallons treated water), but later the time was lengthened to about 100 hours (* 12,000 gallons treated water). Grab samples of raw, settled, and filtered water were obtained periodically in 1-liter cubi- tainers and preserved with 1.5 ml of concentrated nitric acid prior to analysis. 2) Analytic Methods Mercury analysis was done by the fTameless atomic absorp- tion method.-3 Arsenic and cadmium analysis was done on a Perkin Elmer Model 403 Atomic Absorption Spectrophotometer equipped with a graphite furnace and a Perkin Elmer Model HGA-2000 controller. Selenium analysis was done using the method of Caldwell et al.5 3. Results a. Jar Tests 1) Previous Work Jar test studies have been completed on mercury, barium, arsenic, selenium and cadmium. Pilot plant tests, on the other hand, have been only partially completed for mercury, arsenic, selenium, and cadmium. Because the results of most of the jar test experiments have been presented in detail by Logsdon and Symons2>6 an(j by Logsdon, Sorg and Symons', they will not be repeated. 2) Barium The jar test work on barium removal has been completed, but the pilot plant studies have not begun. The laboratory experiments used a midwestern ground water containing 7 to 8 mg/1 of barium. Coagu- lation with aluminum sulfate and ferric sulfate was expected to remove barium effectively because the producton of rather insoluble barium sulfate was anticipated. However, the anticipated results were not achieved; removals did not exceed 30 percent with either type of coagu- lant. A possible explanation for the poor removal was supersaturation of barium sulfate. A series of two-stage coagulation experiments were carried out using 100 mg/1 of coagulant initially and a 20 mg/1 dose for the second stage. These studies produced higher barium removals, giving support to the suggestion that barium sulfate was supersatu- rated after the first stage of coagulation. Unfortunately, such treat- ment would not be very practical because of the higher cost required of two-stage treatment. 171 ------- Barium removal by lime softening was also studied. In the pH range of 10-11, barium removals exceeded 90 percent; the maximum removal was near 98 percent. Data from a full-scale ion-exchange soft- ening plant also showed a 98 percent barium reduction when the initial barium content was about 11 mg/1. Finally, laboratory tests showed that 400 mg/liter of ActiveX resin could remove about 80 percent of the barium, but that powdered activated carbon was not effective for barium removal. b. Pilot Plant Studies The test waters used in the pilot plant studies to date were Cincinnati tap water and well water from Glendale, Ohio. The treatment methods used were alum coagulation, Ferrifloc coagulation, and lime softening. 1) Mercury Two types of mercury were selected for study: mercuric chloride (inorganic form) and methyl mercury chloride (organic form). Pilot plant test results on inorganics using spiked Cincinnati tap water generally agreed with the jar test data. For raw water concentrations of 4.0-7.5 yg/1, removals for settled water ranged from 24 to 70 percent and for filtered water (dual-media) 47 to 80 percent when the water was coagulated with 15-72 mg/1 of Ferrifloc. When alum (22 to 62 mg/1) was used as a coagulant, removals were less than 10 percent for both settled and filtered water._ Organic mercury removal by activated carbon in combina- tion with Ferrifloc was studied using Cincinnati tap water spiked with 3.7 to 5.6 yg/1 of mercury. The treatment consisted of adding 4.5 to 5.6 mg/1 of powdered carbon in the first rapid mix tank and 28-35 mg/1 of Ferrifloc in the second rapid mix tank. The mercury removal results ranged from 5 to 32 percent for settled water; 8 to 38 percent for dual media filtered water and; 98 to 100 percent for granular activated carbon filtered water. Lime softening pilot plant tests using raw well water have been only partially completed. Preliminary test data, however, have shown removals for both types of mercury to be significantly higher than those achieved in the jar test experiments. Inorganic mercury removals were 10-40 percent higher than in the jar tests with the settled water, ranging from 45 to 63 percent and filtered water (dual media) from 69 to 90 percent. When organic mercury was studied, early test results indicated that some mercury was being removed, as compared to no removal in the jar tests. Additional lime softening work will be carried out to determine the reason for the differences. 172 ------- 2) Cadmium Pilot plant studies have been completed on cadmium using Cincinnati tap water and the raw well water. In all cases, the results were in agreement with the jar test data. The tests showed that cadmium removal was pH dependent for both alum and Ferrifloc coagulation using Cincinnati tap water spiked with 0.028 - 0.032 mg/1 of cadmium. For example, when Ferrifloc was used as the coagulant, removals for the settled water was 20-26 per- cent at pH 6.8 and 70-80 percent at pH 8.3. Removals for dual media filtration was about 5-10 percent higher than the settled water. Lime softening at pH 9.5 and 11.3 was also studied. Cad- mium removals at both pH values were over 95 percent for the settled water, dual media filtered water, and granular carbon filtered water. An eight-week series of direct filtration tests were also carried out using two granular activated carbon filters and Cin- cinnati tap water spiked with 0.028-0.032 mg/1 of cadmium. Each test run lasted about 100 hours and the filters were not backwashed between runs. Cadmium removals ranged from 7 to 30 percent for the filter containing 30 inches of Filtrasorb 100 and 17 to 54 percent for the filter containing 30 inches of exhausted Filtrasorb 200. 3) Arsenic Two forms of arsenic have been studied, Arsenic III (arsenite) and Arsenic V (arsenate). Because As(III) would probably be found in ground water, the behavior of As(III) was studied using only the raw well water. Pilot plant tests on As(V) however, were carried out using both Cincinnati tap water and the raw well water. In all cases, the arsenic concentration was near 0.40 mg/1 and results compared very closely with the jar test data. Arsenic III removal by lime softening was investigated at pH 9.5 and 11.3. At pH 9.5, removal for the settled water was 10 percent and for the filtered water 24 percent (GAC) and 26 percent (dual-media). When the water was softened to 11.3, settled water removal was 63 percent and filtered water removal 72 percent for both filters. Although pilot plant tests have not been run to confirm it, laboratory jar tests showed that when As(III) is oxidized to As(V) using chlorine, As(III) behaves like As(V). Because higher removals were obtained on As(V) under all treatment processes studied, arsenite should, therefore, be oxidized to arsenate before removal is attempted. 173 ------- Removal of As(V) was studied in the pilot plant using alum, Ferrifloc and lime. Arsenic removals using -Cincinnati tap water and Ferrifloc were excellent; settled water removals were 91-94 percent and filtered water (dual media) removals were greater than 98 percent. When alum was used as the coagulant, removals were somewhat less; settled water removals ranged from 75 to 86 percent and filtered water (dual media) removals ranged from 85 to 96 percent. Softening tests on the raw well water were also investi- gated at pH 9.5 and 11.3. At pH 9.5, the test data showed an As(V) removal of 49 percent for the settled water and 53 percent for the filtered waters. At pH 11.3, As(V) removals were above 98 percent, for both settled and filtered waters. 4) Selenium The behavior of two forms of selenium has been studied, selenium IV (selenite) and selenium VI (selenate). Although the jar test studies have been completed, only limited pilot plant work has been carried out. Because selenite has been identified as a problem in some ground waters, SE(IV) was investigated primarily with raw well water. Removals of 0.1 mg/1 of SE(IV) by lime softening in the laboratory did not exceed 40 to 50 percent and generally were lower. Coagulation studies with alum and Ferrifloc were also undertaken with well water and a surface water. The results of these laboratory experiments found that Ferrifloc produced higher removals than alum and that removals for both coagulants decreases as the pH increases from 6 to 8. Re- movals ranged from about 80 to 20 percent with Ferrifloc and were 10 percent or less with alum when 25 mg/1 of coagulant was used. The removal of selenate (VI) was uniformly poor for all jar test and pilot plant studies. Selenate removal by coagulation with iron or alum (up to 100 mg/1 of coagulant), by softening from pH 9 to 10.8 or by treatment with up to 100 mg/1 of powdered activated carbon was less than 10 percent for initial selenium concentrations of 0.1 mg/1. Although conventional treatment experiments were unsuccessful in removing selenate, a short reverse osmosis test showed that this techni- que could remove it and merits further investigations. During a two- hour reverse osmosis test run on Cincinnati tap water spiked with 0.1 mg/1 of selenium VI, over 97 percent of the selenium was removed by the R.O. unit operating at 1.5 liters/minute. a. Summary of Results^ Table I summarizes all of the jar test and pilot plant data collected thus far. These studies are continuing. 174 ------- TABLE I SUMMARY OF RESULTS OF TREATMENT PROCESSES TO REMOVE TRACE METALS FROM DRINKING WATER Coagulation Softening Absorption Trace Metal Mercury (0)* CH3HgCl Mercury (I), HgCl2 Barium Alum Poor** Poor Poor Ferric Sulfate Jar Test Data Poor Fair Poor Lime pH 9.5-10 Poor Fair Good Zeolite field Selenium(I) oc Selenium(I) Se+6 Arsenic(I), As+3 Arsenic(I), As+5 Cadmium(I) Mercury (0) CH3HgCl Poor pH<7 Poor Poor Good tot very good pH<8 Poor to Fair - Fair to good pH<7 Poor Fair to goodt Good tot very good pH<8 Good tot very good pH>7.5 Pilot Plant - Poor Poor Poor Good Very goodt Data - Excess lime pH 10.6-11 Poor Goodt Good to very goodt data - very Fair Poor Goodt Very goodt Very goodt - Activated Carbon Goodt Good Poor goodt Poor Poor Poor Poor ~ - 175 ------- Table 1 (continued) Coagulation Softening Absorption Trace Metal Mercury (I) HgCl2 Barium Selenium(I) Se+4 Selenium(I), Se+6 Arsenic(I), As+3 Arsenic(I), As+5 Cadmium(I) Alum Poor - _ - ^ Good tot very good Poor to fair Ferric Sulfate Fair - _ Poor — Veryt good Good pH 8.4 Lime pH 9.5-10 - Poor to Fair - Poor Fair Very goodt Excess lime Activated pH 10.6-11 Carbon Good - _ _ - Good Very Goodt Poor to Fair Very goodt * - (0) = Organic; (I) = Inorganic ** - Key - Poor=0-30% removal; Fair=30-60% removal; Good=60-90% removal; Very good = above 90% removal. t - Best treatment techniques. - Not yet tested. 176 ------- 4. Radium-226 a. Introduction Radium-226 is found in some waters of the USA in excess of the 1962 Public Health Service Drinking Water Standards limit of 3 pCi/1. If the new Federal Drinking Water Regulations set an even lower limit for Ra-226, additional water sources would need treatment to meet the standard. Radium-226 is usually found in groundwaters, because it is a geochemical contaminant. It is associated with certain aquifers such as St. Peter sandstone in the upper Mississippi Valley and Cheyenne sand- stone in Colorado and New Mexico8. Radium-226 may be found in surface waters contaminated by radium-bearing springs. Other sources of con- tamination are leachates from tailings of uranium milling plants^ and from the phosphate rock mining and milling industry in Florida. Because Ra-226 more frequently is found in ground waters, treatment processes used for ground water are of interest. Some basic information on removal of radium was contained in Straub's report.8 b. Results A recent study of water treatment plants in Iowa has shown results similar to those reported by Straub.8 The Radiochemistry and Nuclear Engineering Facility (R&NEF), a part of the EPA's Office of Radiation Programs has contracts with the states of Iowa and Illinois for studying radium removal by water plants. The following results have been obtained in the Iowa study to date and are presented in Table II. c. Discussion On the basis of Iowa data, Ra-226 removals of about 75% could be anticipated for lime softening. If greater removal is needed, ion exchange or reverse osmosis treatment would be necessary. In either case, the practice of blending raw and treated water to obtain a less corrosive water and save capital costs by reducing plant size could result in a plant effluent having an excessive concentration of Ra-226. In such a case, corrosion control would have to be accomplished by methods other than raw water blending, and softening 100% of plant flow would increase costs at existing softening plants that now bypass some raw water. 177 ------- RADIUM REMOVAL BY WATER TREATMENT PROCESSES Radium pCi/1 % Treatment Technique raw finished reduction Greensand for iron removal 6.9 6.7 7% Iron removal - aeration and pressure filtration 16 12 25% Lime-soda softening Iron removal followed by ion exchange softening , f \ Reverse osmosis 6.1 9.3 49 5.7 6.7 12 14 0.9 2.3 1.9 0.3 0.2 0.5 0.6 85% 75% 96% 95% 97% 96% 96% d. Future Plans Radium removal -research contracts are continuing under the management of R&NEF with water supply engineering input from WSRL. In addition, WSRL has under consideration a grant application for develop- ment of detailed construction and operating cost data and estimates for water treatment plants built and operated primarily for radium removal. 5. Acknowledgments The following personnel contributed to this report. Water Supply Research Laboratory Radiochemistry and Nuclear Engineering Facility. NERC- Maura M. Lilly Cinc1nnati Thomas J. Sorg William Brinck Kenneth L. Kropp Richard Engelmann Iowa Department of Environmental Bradford L. Smith Quality Jeffrey Klieve „ , c .,. , , Raymond J. Lishka R' J" Schlickelman James S. Caldwell Gary S. Logsdon 178 ------- 6. References 1. Interim Primary Drinking Water Standards, Federal Register, Volume 40, No. 51, Part II, p. 11190-11198, March 14, 1975. 2. Logsdon, 6.S. and J. M. Symons, Journal American Water Works Association. 65., 554 (1973"T 3. Kopp, J.F., M. C. Longbottom and L. B. Lobring. JAWWA. 64, 20 (1972). 4. McFarren, E. F., JAWWA, £4, 28 (1972). 5. Caldwell, J.S., R.J. Lishka and E. F. McFarren, JAWWA. 65, 731 (1973). 6. Logsdon, G.S. and J.M. Symons, Removal of Trace Inorganics by Drinking Water Treatment Unit Processes. Water-1973. American Institute of Chemical Engineers Symposium Series, 70, 136 367-377, (1974). 7. Logsdon, G.S., T.J. Sorg, and J.M. Symons, Removal of Heavy Metals by Conventional Treatment, Proceedings 16th Water Quality Conference - Trace Metals in Water Supplies: Occurrance, Significance and Control, University of Illinois Bulletin, 71_, 108, 111-133 (April 29, 1974). 8. Straub, C.P., Radium-226 and Water Supplies: Cost- Benefit-Risk Appraisal, Unpublished Report, 1973. 9. Tsivoglou, E.G. and O'Connell, R.L., Waste Guide for the Uranium Milling Industry, DHEW, USPHS, DWSPC, RATSEC, Technical Report, W62-12. D. TREATMENT FOR REMOVAL OF ASBESTIFORM FIBERS 1. Introduction The presence of asbestiform fibers in the drinking water of communi- ties using western Lake Superior as a water source was made known in the summer of 1973. In the fall and early winter of that year an interagency agreement for studies of the problem was formulated and signed by the U. S. Environmental Protection Agency and the U. S. Army Corps of Engineers. Under this agreement EPA funded the pilot plant research on 179 ------- asbestiform fiber removal while the Corps of Engineers funded a study of alternative water sources and sites for construction of a filtration plant or plants for the Duluth-Cloquet-Superior area.* The Corps also managed the contract for the entire study, while EPA provided technical assistance on the filtration. The contractor was Black and Veatch, of Kansas City. The pilot plant research was conducted at the Lakewood Pumping Station in Duluth, with the assistance of the Department of Water and Gas of the City of Duluth. Pilot plant operations were conducted in the period from April through September 1974. In this time a.total of 227 granular media and 228 diatomaceous earth (DE) filter runs were conduc- ted. 2. Scope of Study There were two principal objectives in the research. First, the pilot plants were to be operated .in such a way that data needed for engineering design and cost estimates could be obtained. The results and conclusions related to design and cost factors are being presented by Robinson et al.' A paper on DE filtration optimization is being presented by Baumann.2 The second objective of the study was to obtain information on the removal of asbestiform fibers. That information is presented in this paper. In order to learn how to reduce the asbestiform fiber count by filtration, a number of variables were studied. Those common to both granular and DE filtration were filtration rate, seasonal conditions, and raw water turbidity. Other important variables in the granular filtration study were filtration with and without sedimentation, dual media vs. mixed (tri) media, doses and combinations of inorganic salts and organic polymers, .single-stage vs. multi-stage flash mixing, and flash mix chambers vs. in-line mixers. Variables important in the DE study were one-step vs. two-step precoating, vacuum vs. pressure filtra- tion, DE conditioning with alum or polymers, and body feed doses. 3. Experimental Methods and Equipment a. Equipment The apparatus used in the research has been described in the *Duluth-Superior Urban Study, Interim Report on Water Supply for the Duluth-Superior-Cloquet Area, A Joint Study by the U. S. Army Corps of Engineers, St. Paul District, and the U. S. Environmental Protection Agency (March, 1975). 180 ------- o EPA report on the project. Two types of filters, granular media and DE, were used. All units were situated in Lakewood Pump Station. Raw- water for all units was drawn from the wet well at the pump station. Total water flow through individual filter systems generally ranged from 10 to 20 gpm. Two granular filters were employed. Both units were Water Boy package plants with 4.0 square feet of filter surface. Equipment variations with these units included use of dual media, mixed media, no settling, tube settlers, single-stage rapid mix and two-stage rapid mix with propeller mixers, two-stage and three-stage rapid mix with in- line mixers, alum or ferric chloride as the primary coagulant, anionic, cationic, and non-ionic polymers, and .filtration rates from 2 to 7 gpm/sf. Two kinds of DE filter systems were employed. Pressure fil- tration was carried out with an Erdlator filter. In this unit water was not coagulated and settled, contrary to U. S. Army practice, but the clear Lake Superior water merely passed through the pretreatment portions of the Erdlator on its way to Jthe pressure filter. The Erdla- tor had two pressure vessels, each containing six cylindrical septa. Total filter surface area for one pressure vessel was 10.0 square feet. After the filter septum was precoated, body feed could be added dry or in slurry form. The gravity, or vacuum DE filter unit consisted of an open rectangular tank with flat septa. The driving force for filtration was the difference between atmospheric pressure and the pressure at the pump intake on the effluent side of the filter. Filter surface was also 10.0 square feet on this unit. Body feed could be added dry or in slurry form. Both kinds of DE filters were operated in various ways in order to evaluate conditioning of DE with alum, cationic polymer and anionic polymer. On some runs a cationic polymer was added to the raw water. Single-step vs. two-step precoat was studied. Conditioned DE was used in precoat situations as well as for body feed. Various grades of DE, from fine to coarse, were evaluated. b. Analytical Methods. Most of the analytical procedures were done in accordance with Standard Methods4. In addition to laboratory turbidity measurements on grab samples, continuous flow turbidity data were obtained with both 90° scatter and 15° forward scatter instruments. Grab samples were obtained for the analyses, including asbestos. Since there is no 181 ------- standard method for asbestiform fibers in water, analytical methods were different for each laboratory used. Three analytical laboratories were involved in this study. The National Water Quality Laboratory of EPA in Duluth ana- lyzed raw and filtered samples for suspended solids and amphibole mass. The x-ray diffraction technique for amphibole mass has been published by Cook. Although this method measured only amphibole mass irrespective of shape (by definition fibers have a length:width ratio of 3:1 or greater), and did not measure chrysotile, the availability of amphibole mass data within a few days of sample collection made this method a valuable tool. Transmission electron microscope analysis of water samples was done at the Ontario Research Foundation (ORF) and at the University of Minnesota at Duluth (UMD). The ORF analytical method has been pub- lished6. ORF obtained size information on each fiber (length and width) and confirmed that all fibers were amphibole or chrysotile by electron diffraction. Electron diffraction was used to identify a portion, but not all, of the fibers counted by UMD.7 4. Results The results of all pilot plant filter runs and analyses are presented in the EPA filtration report and appendices. The data pre- sented in this paper relate principally to the problems of asbestiform fiber removal by filtration. Relevant raw water data are also presented in order to place the filtration results in proper, perspective. a. Raw Water Quality Water quality parameters of greatest interest in this study were turbidity, asbestiform fiber count, and amphibole mass. Other data on pH, alkalinity, hardness, temperature, and suspended solids can be found in the EPA report. Most turbidity measurements.were made with a Hach 2100A labora- tory turbidimeter. When a comparison was made between a Monitek in-line turbidimeter and the Hach 2100A, it was found that although the numeri- cal readings differed for the two instruments (15° forward scatter vs. 90° scatter), the trends of turbidity variation were quite similar. Both instruments showed rising or declining turbidities at the same time. These data were presented in the EPA report. Turbidity of the raw water at Lakewood changed very little during most of the five months of pilot plant operation. Except for a period extending from 2300 hours on June 6 to 0700 hours on June 15, 182 ------- 1974, and other briefer periods, the turbidity of the raw water from the Lakewood Intake ranged from 0.35 toa 1.0 TU. During the period begin- ning on June 6, the raw water turbfdity ranged from 1.6 to 6.3 TU and averaged 2.7 TU. Other periods of raw water turbidity in excess of 1.0 TU were relatively short, ranging from a period of 2 hours to one of 29 hours, with the raw water turbidity seldom exceeding 1.5 TU during these periods. The turbidity of the raw water transported from the Cloquet Pipeline Intake was not as low as that from the Lakewood Intake, but it also varied only over a slight range. The turbidity of the Cloquet raw water tested was never below 2.0 TU nor above 4.0 TU. Asbestiform fiber count and amphibole mass concentrations showed much greater variation than raw water turbidity. Fiber counts and amphibole mass concentrations are plotted vs. time in Figures 1 and 2 to show the time variation of these parameters. It should be mentioned that there were no violent storms during the pilot plant operation, and in the fall and winter of 1974, amphibole mass concentrations exceeding 1.0 mg/1 were measured during a storm. The pilot plant was shut down two or three months before the worst water conditions (high turbidity.and fiber count) occurred because funds for conducting the study were limited. During the May-September period of operation, amphibole and chrysotile fiber counts were frequently in the 0.5 to 1.5 x 106 f/1 range, and amphibole mass often ranged between 0.05 and 0.2 mg/1. There were extremes both above and below these values. b. Asbestiform Fiber Removal by Filtration It is apparent from the portions of this paper that deal with the scope of the study and the equipment used that there were numerous variations in experimental conditions. In order to simplify data analysis, amphibole mass and asbestiform fiber removal results are pre- sented for treatment categories that specify such variables as: filter media, filtration rate, use of sedimentation, inorganic coagulant and polymer, and type of rapid mixing for granular filtration; and number of layers of precoat, precoat conditioning, body feed conditioning, and polymer feed to raw water for DE filtration. Tables 1, 2 and 3 show summarized results for dual media, mixed media, and diatomite filtra- tion. The treatment data are given in terms of the number of filter- ed samples submitted for analysis and the number of samples with a 183 ------- 10 9 CO ° 8 CC 7 LLJ < 6 CO cc LU eg 5 LL CC .00 O ••£> U. CO 111 CG CO < __ o AMPHIBOLE CHYRSOTILE 5 15 25 APRIL 15 25 MAY 15 25 JUNE 5 15 25 JULY 5 15 25 AUGUST 5 15 25 SEPTEMBER FIGURE 1. ONTARIO RESEARCH FOUNDATION ASBESTIFORM FIBER COUNTS RAW WATER AT DULUTH LAKEWOOD INTAKE - 1974 ------- o SUSPENDED SOLIDS • AMPHIBOLE 5 15 25 APRIL 5 15 25 AUGUST 5 15 25 SEPTEMBER FIGURE 2. ENVIRONMENTAL PROTECTION AGENCY NATIONAL WATER QUALITY LABORATORY AMPHiBOLE MASS CONCENTRATION RAW WATER AT DULUTH LAKEWOOD INTAKE - 1974 ------- result equal to or less than 0.04 x 106 fibers/liter (f/1) (fiber data from ORF), or equal to or less than 0.005 mg/1 in the case of amphibole mass data. The amphibole mass detection limit varied according to the volume of the water sample filtered for analysis. Waters which had a greater tendency to clog membrane filters had higher detection limits. If the detection limit was above 0.005 mg/1, it became impossible to say whether the amphibole mass in a treated water exceeded 0.005 mg/1. Samples clouded in this uncertainty were not tabulated in the results presented herein. For example, in Table 1, treatment category filtra- tion without sedimentation, alum and nonionic polymer, ^4 gpm/sf, ten samples are listed as having been analyzed for asbestiform fibers, while only five are listed as having been analyzed for amphibole mass. The other five amphibole mass samples had a detection limit that exceeded 0.005 mg/1. Tables 1, 2 and 3 show that the more successful variations of filtration, whether dual media, mixed media, or diatomite, produced effluents having amphibole mass concentrations and amphibole fiber counts near the detection limits of the analytical methods employed. Chrysotile fiber count in filtered water generally exceeded 0.04 x 10b f/1 for dual media filtration tests and for DE filtration tests not employing polymer conditioning. Mixed media filter runs employing alurn and nonionic polymer or alum, anionic polymer and another polymer, and diatomite runs employing A-23 conditioning of DE or Catfloc B condi- tioning of raw water did have some runs with effluent chrysotile counts <0.04 x 106. 5. Discussion a. Asbestiform Fiber Removal The initial objective of the filtration research at Lakewood Pumping Station was the removal of amphibole asbestiform fibers and turbidity-causing suspended matter. According to the pilot plant research contract, the principal criterion for successful treatment will be the economical attainment of virtually complete removal of asbestos- like fibers as defined by optical and electron microscope analysis using the best current state of the art. A secondary criterion shall be the production of water having a turbidity of not more than one turbidity unit.8 The fibers referred to in the contract were expected to be prin- cipally amphibole. In conversations between the WSRL and EPA Region V (Chicago)9-, which was then heavily involved with contracts for analysis of asbesti- form fiber content of water samples, the Ontario Research Foundation was determined to be one of the laboratories that could satisfactorily 186 ------- CO TABLE 1 AMPHIBOLE MASS AND SUMMARY OF ASBESTIFORM FIBER REMOVAL BY DUAL MEDIA FILTRATION Treatment Technique NWQL Amphibole Mass Number of Samples <.005 Analyzed Ing/1 Ontario Research Foundation Amphibole Fibers Chrysotile Fibers Number of Samples Number of Samples <.04x106 <0.04xl06 Analyzed^ ~ f/J Analyzed ~" f/1 Filtration w/o Sedimentation Alum & Nonionic Polymer (985N) 2 gpm/sq ft 33 Alum & Nonionic Polymer (N-17 or 985N) M gpm/sq ft 54' Alum & Nonionic Polymer (985N) 6-8 gpm/sq ft - Fed 3 & Cationic Polymer (C-31) 4 gpm/sq ft - 323 0 10 9 10 0 323 0 222 1 Filtration w/ Sedimentation Tube Settlers 4 gpm/sq ft Alum & Nonionic Polymer (985N) FeCl3 & Nonionic Polymer (985N) 12 12 2 1 12 2 12 2 12 2 1 0 ------- TABLE 2 AMPHIBOLE MASS AND SUMMARY OF ASBESTIFORM FIBER REMOVAL BY MIXED MEDIA FILTRATION Treatment Technique NWQL Amphibole Mass Number of Samples , 1-005 Analyzed nig/1 Ontario Research Foundation Amphibole Fibers Chrysotile Fibers Number of Samples Number of Samples £.04x10° <0.04x10° Analyzed f/1 Analyzed f/1 CO 00 Chemicals Added to Mixing Chamber Alum & Nonionic Polymer 4 gpm/sq ft (985N) -. Chemicals Added to Two Flash Mixers Alum & Nonionic Polymer 4 gpm/sq ft (985N) Alum & Nonionic Polymer 2 gpm/sq ft (985N) Alum & Anionic & Cationic Polymer (A-23 & Catfloc B or C-31) 4 gpm/sq ft Alum & Anionic & Nonionic Polymers (A-23 & 985N) 4 gpm/sq ft In-Line Mixers Alum & Nonionic Polymer 4gpm/sq ft (985N) 0 9 1 9 1 9 1 9 1 9 1 5 0 0 ------- Table 2 (continued) Treatment Technique NWQL Amphibole Mass Number of Samples <.005 Analyzed mg/1 Ontario Research Foundation Amphibole Fibers Chrysotile Fibers Number of Samples, Number of Samples <.04xlOb <0.04xl06 Analyzed f/1 Analyzed ILL CO Alum & Nonionic & Anionic Polymers (985N & A-23) 4 gpm/sq ft Alum & Anionic & Cationic Polymers (A-23 & C-31) 4 gpm/sq ft Alum & Nonionic Polymer 6 gpm/sq ft (985N) Alum & Cationic Polymer 4 gpm/sq ft (C-31) 1 2 1 1 2 1 2 0 0 ------- UD O TABLE 3 SUMMARY OF AMPHIBOLE MASS AND ASBESTIFORM FIBER REMOVAL BY MIXED MEDIA FILTRATION NWQL Amphibole Mass Ontario Research Foundation Amphibole Fibers Chrysotile Fibers Treatment Technique Pressure Filtration Two-Step Precoat 1 gpra/sq ft Anionic Polymer (A-23) to 2nd Step of Precoat, Alum & Soda Ash to Body Feed Alum & Soda Ash to 2nd Step of Precoat Alum & Soda Ash to 2nd Step of Precoat and to Body Feed Cationic Polymer to Raw Water (Catfloc B) 1 1 Ull IM 1— | \J f Analyzed 1 6 3 5 ^f\AlllhS 1 V- wS <.005 mg/1 1 6 3 5 I 1 VMI|I>S V» ( VI Analyzed 1 6 3 5 <. 04x10° f/1 1 3 3 5 iiuritwiui v/ i Analyzed 1 6 3 5, <,04xl06 f/1 1 1 0 2 Alum & Soda Ash to 2nd Step of Precoat. Cationic Polymer (Catfloc B) to Raw Water Vacuum Filtration One Step Precoat 1 gpm/sq ft Anionic Polymer (A-23) to Precoat ------- Table 3 (continued) Treatment Technique NWQL Amphibole Mass Number of Samples £.005 Analyzed mg/1 Ontario Research Foundation Amphibole Fibers Chrysotile Fibers Number of Samples Number of Samples <_.04xl06 i.04x!06 Analyzed f/1 Analyzed f/1 Vacuum Filtration Two Step Precoat 1 gpm/sq ft Anionic Polymer (A-23) to Total Precoat Anionic Polymer (A-23) to 2nd Step of Precoat 1 0 Anionic Polymer (A-23) to 2nd Step of Precoat, Alum & Soda Ash to Body Feed 5 4 Alum & Soda Ash to Second Step of Precoat 2 2 ' Alum & Soda Ash to 2nd Step of Precoat and to Body Feed 2 2 Cationic Polymer (Catfloc B) to Raw Water 1 1 1 1 5 2 3 3 1 1 5 0 2 3 1 1 5 2 3 3 0 1 1 0 0 0 Alum & Soda Ash to 2nd Step of Precoat Cationic Polymer (Catfloc B) to Raw Water 0 ------- analyze water samples for asbestiform fibers. Thus, ORF became a sub- contractor for the filtration research conducted by Black and Veatch. One factor which must be considered when interpreting EM fiber analysis data is the meaning of the detection limit. When ORF found zero fibers in ten fields, the reported result was below detectable limits (BDL), not zero. Depending on the circumstances of the individual analysis, such as sample volume initially filtered, ORF reported a number of BDL limits from 0.02 x 106 f/1 to 0.07 x 106 f/1. Most of the time the detection limit was reported as 0.02 x 106 f/1 although early in the work 0.04 x 10^ f/1 was frequently reported. McFarren'O uses an intermediate category, not statistically significant (NSS) between BDL and reportable fiber counts. The NSS finding is applied to observation of less than 5 fibers in 20 fields. This would correspond to about two fibers in 10 fields for ORF. The rationale for use of NSS is that fiber counts become less reliable as fewer fibers are found. The standard deviation varies as l//n , where n is the number of fibers foundJ' Thus the standard deviation is 10% when 100 fibers are found, and 100% when 1 fiber is found. For the EM work done by ORF on this project, the finding of two fibers in 10 fields usually represented 0.04 x 106 f/1. Since the goal of the research was the "virtually complete removal," 0.04 x 10^ f/1 and lower were considered not statistically significant, and the research goal was considered to have been attained when filtered water fiber counts were 0.04 x 10^ f/1 or lower. It is apparent from the data in Tables 1-3 that amphibole asbestiform fibers could be readily removed by filtration. Additional evidence to confirm the efficacy of filtration is found in the amphibole mass data. Many of the filter runs that were sampled contained 0.005 mg/1 or less in the filter effluent. Based on the amphibole mass con- centration in the raw water, this represented amphibole mass reductions of ten-fold to forty-fold or more. Some of the variables considered in the research are to be found in Tables 1-3. Table 1 contains fiber removal data for dual media granular filters only. There is nothing in Table 1 that indicates that sedimentation before filtration was beneficial for amphibole or chryso- tile fiber removal. For the treatment of clear Lake Superior water, direct filtration performed as well as filtration with sedimentation. Ferric chloride appears to be effective for fiber removal, but alum and nonionic polymer were used in most tests because that was the combination of treatment chemicals that gave the desired combination 192 ------- of very low effluent turbidity and longer filter runs. This is explained in more detail by Robinson et al. Research results for mixed media filtration are summarized in Table 2. A comparison of Tables 1 and 2 shows that mixed media filtra- tion after two-stage flash mix was more effective for chrysotile removal than dual media filtration after one-stage flash mix, when alum and a nonionic polymer were used. With the one-stage flash mix arrangement, polymer was added at the flocculation chamber. Since two variables were changed at once, it is difficult, if not impossible, to decide which affected fiber removal more. Another variable studied in the mixed media system was three- stage rapid mix. The purpose of the triple mix was to add and mix sequentially three conditioning chemicals, anionic polymer, alum, and cationic or nonionic polymer, with the objective of establishing, at different times in the treatment process, environments in which positive, and then negative, surface charges predominated. Unfortunately, two chemicals were added to a common barrel (mixing chamber) in the propeller type flash mixed system, so valid data were obtained only for in-line mixers. The results of three-stage rapid mixing are encouraging for both amphibole and chrysotile removal. Diatomite filtration for asbestiform fiber removal may appear to be less successful than granular filtration, but this is not the case. More operational variations were tried with diatomite, and some were not successful. Some successful techniques were found, however, and these are found in Table 3. Effective filtration techniques for removal of amphibole mass and fibers were the following: a. alum conditioning of both precoat and body feed; b. precoat conditioning with anionic polymer and body feed conditioning with alum; c. conditioning of the raw water with Catfloc B and in some cases, alum conditioning of the precoat also. Diatomite filtration techniques most effective for removal of both chrysotile and amphibole involved the following: d. conditioning of precoat with anionic polymer and conditioning of body feed with alum; e. conditioning of raw water with Catfloc B. 193 ------- The treatment scheme in case d involved both negative (anionic polymer) and positive (alum coated DE) charges. A system in which Catfloc B was added to raw water before filtration through unconditioned DE would also involve both positive (Catfloc B) and negative charge systems, since diatomite ordinarily has a negative surface charge.2 However, plain diatomite probably would not be as electronegative as diatomite coated with an anionic polymer. Experiments with anionic polymers were conducted because of a fundamental difference in chrysotile and amphibole. Parks^2 summarized the work of numerous investigators in an article on the isoelectric point or zero point of charge of complex oxide minerals in water. The zero point of charge, or pH at which there is no net charge on the par- ticle, is in the pH 10-11 range for chrysotile, but it is pH 5 for cummingtonite (an amphibole). In the 7-8 pH range used in filtration tests at Duluth, chrysotile would have a positive surface charge while cummingtonite, like clays and most bacteria, would be negative. It follows that in order to overcome the surface charges of both amphibole and chrysotile, it would be necessary to use treatment chemicals carry- ing positive and negative surface charges, respectively. The treat- ment chemicals should be introducted to the water separately so that the coagulation is not confined to reaction between only the treatment chemicals. It would be logical to ask why amphibole and chrysotile fibers do not coagulate themselves since they are of opposite surface charge. A probable answer is that there are so few present, even when the con- centration is 106 f/1. For example, if a liter of water contained 10^ chrysotile fibers with 0.04 ym diameter and 1 ym length, the total vol- ume of fibers would be ^10"^ 1 or 0.001 microliter. Also, 10^ amphibole with 0.2 ym diameter and 1 ym length would occupy a volume of 3 x 10~° 1 or 0.03 microliters. The addition of polymers to raw water in this work should have resulted in molecular concentrations on the order of 10^ to 10^ molecules per liter, depending upon the dose and molecular weight of the polymer, and assuming that every polymer molecule was a separate entity (the same assumption made for asbestiform fibers). It is obvious that polymer molecules very greatly outnumber (by a factor of 108 to 10^°) asbestos particles, so the chances for a polymer-fiber collision would be much better than for a fiber-fiber collision. Thus polymer conditioning is needed for fiber removal, and since surface charges differ with type, different polymers are needed to remove amphibole and chrysotile. 194 ------- One other factor that may be related to fiber removal or, conversely, to the ability of fibers to pass through filters, is fiber size. For all filters, the typical size of chrysotile fibers in the effluent was smaller than the typical size in the raw water. This size relationship was also true for amphiboles in DE filtrate. Because only 25 amphibole fibers were observed and sized for granular media tests, this small sample was more subject to distortion by a typical fibers and was not very suitable for a chi-square analysis. The factor of particle size is probably less important than surface charge, since particles in the water, both before and after filtration are in the size range (M ym) suggested by Yao13 to have minimum removal efficiency by granular filtration. Filtration results can be summarized briefly. The methods found more effective than others for removal of both amphibole and chrysotile were mixed media filtration employing alum and an anionic polymer and two-stage flash mixing; and pressure diatomite filtration employing Catfloc B added to the raw water and no conditioning of the DE. Methods showing potential for further research are three-stage flash mixing with sequential addition of anionic polymer, alum, and cationic or nonionic polymer, followed by flocculation and filtration; and pressure DE filtration with anionic polymer conditioning of the precoat and alum conditioning of the body feed. b. Efforts to Develop Rapid Detection Methods A limited effort to learn about rapid detection of asbestiform fibers was made in this research, but the principal objective was to learn about fiber removal by filtration. Other efforts to develop a rapid fiber detection method are underway, sponsored by EPA and other Federal agencies. One method being investigated involves placing a water sample in a laser light beam measuring scattered light from inci- dent angles of about 10° to 135°, and relating variations of light intensity and incidence angle to the types of particles present in water. One of the goals of these efforts is to provide a technique that is practical for monitoring both amphibole and chrysotile asbesti- form fibers in water at filtration plants. Such a technique should be rapid enough to permit a plant operator to make meaningful changes in the treatment process in order to hold fiber content of the filtered water to a minimum. Until a more rapid method is available, water filtration plants on Lake Superior should use the x-ray diffraction method, along with occasional EM analyses. 6. Future Research Information developed in this pilot plant research permits a number of questions to be formulated for future study. Among the ideas that 195 ------- could be investigated are the following: a. Ways to improve chrysotile removal by anionic polymer conditioning of DE or by the use of three-stage mixing and combination of three conditioning chemicals in granular filtration. b. Effect of high algal counts on filter performance. c. Fiber removal during times of highest amphibole mass and fiber count. d. Verification of POPO optimization of diatomite filtration, e. Additional filtration experiments at 5 to 6 gpm/ft with granular filters. f. Effect of mixing intensity on filtration, and a compari- son of back-mixing vs. in-line mixing. g. Further laboratory development, followed by pilot plant tests, of an operator's method for monitoring the presence of asbesti- form fibers in water. A number of the suggestions for future work represent an extension of past work into promising study areas. Additional research is needed to increase the knowledge of the water treatment profession on the topic of asbestiform fiber removal by filtration. 7- Conclusions a. No discernible tie was evident between the Duluth raw water turbidities and the asbestiform fiber levels. b. At finished water turbidities of less than 0.1 TU, the amphibole fiber count and mass determinations were usually below the detection limits of the analytical method used. c. A general association was indicated between the NWQL amphibole mass concentration and the ORF amphibole fiber counts in the Duluth raw water. d. No relationship was observed between the counts of the amphibole and the chrysotile fibers in the Duluth raw water. e. Based on achieving BDL or near it, 32 of 34 MM-2 (granular) runs and-21 of 23 MM-1 runs were successful for amphibole 196 ------- removal. Only 8 of 34 MM-2 runs and 2 of 23 MM-1 runs were successful for chrysotile removal. Alum and nonionic polymer worked best in granu- lar filters. f. Amphibole fiber removal accomplished by the tri-media filter exceeded that accomplished by the dual media filters and the DE filters. g. For the pressure DE tests, 19 of 27 were successful for amphibole removal, but only 4 of 27 were successful for chrysotile removal. Vacuum DE filtration (BIF) was not found suitable for treat- ing the raw water being tested. h. A medium grade precoat and a fine grade body feed were most effective in turbidity and asbestiform fiber removal by DE filtration. i. For the Duluth raw water, two treatment conditions, alum coated or plain precoat, with a cationic polymer introduced to the raw water, and an anionic polymer added to the precoat and alum coated body feed were most effective in turbidity and asbestiform fiber re- moval filtration. 8. Acknowledgments The following EPA personnel contributed to this report: E. McFarren R. Lishka J. Millette G. Logsdon J. Symons M. Lilly M. Lubratovich P. Cook G. Glass B. Fairless 9. References 1. Robinson, O.K., et al., Direct Filtration of Lake Superior Water for Asbestiform Solids Removal. Presented at Annual Conference, American Water Works Association, Minneapolis, Minnesota (1975). 197 ------- 2. Baumann, E.R., Diatomite Filter for Asbestiform Fiber Removal from Water. Presented at Annual Conference, American Water Works Association, Minneapolis, Minnesota (1975). 3. Direct Filtration of Lake Superior Water for Asbestiform Fiber Removal. By Black and Veatch, Consulting Engineers, U.S. Environmental Protection Agency, National Environmental Research Center - Cincinnati, Water Supply Research Laboratory (April 1975). Summary Report #EPA 670/2-75-050a and six appendices #EPA 670/2-75-050b to EPA 670/2-75-050g. 4. Standard Methods for the Examination of Water and Wastewater. APHA, AWWA, WPCF, New York (13th Edition 1971). 5. Cook, Philip M., Semi-quantitative Determination of Asbestiform Amphibole Mineral Concentrations in Western Lake Superior Water Samples. Proc. 23rd Annual Conference on Applications of x-ray Analysis, Denver, Colorado (1974). 6. Chatfield, E.J. and Pullen, H., Measuring Asbestos in the Environment. Canadian Research and Development, 7:6:23 (Nov.-Dec. 19741". 7. Carter, Robert E., Private Communication to Mr. 0. J. Schmidt. 8. Contract No. DACW 37-74-C-0079, Department of the Army, St. Paul District, Corps of Engineers (February 1974). 9. Fairless, William. Memorandum from U. S. EPA Region V to Water Supply Research Laboratory. 10. McFarren, Earl F., et a!., Asbestos Analysis by Electron Microscope. Presented at Second Annual Water Quality Technol- ogy Conference, AWWA, Dallas, Texas (1974). 11.. Chatfield, E.J., Private communication to, Mr. 0. J. Schmidt. 12. Parks, George A., Aqueous Surface Chemistry of Oxides and Complex Oxide Minerals. Symposium on Equilibrium Concepts in Natural Water Systems, Proc. 151st Meeting Am. Chem. Soc. Pittsburgh, Pa. (1966). published as Equilibrium Concepts in Natural Water System, W. Stumm, ed., (1967), p. "121. 13. Yao, K., et al., Water and Waste Water Filtration: Concepts and Applications. Environmental Science and Technology, 5:11:1105 (November 1971). 198 ------- APPENDIX VII HEALTH EFFECTS CAUSED BY EXPOSURE TO DRINKING WATER CONTAMINANTS Organics Section Prepared by Robert G. Tardiff Inorganics Section Prepared by Gunther Craun Lei and McCabe Water Supply Research Laboratory National Environmental Research Center Office of Research and Development Cincinnati, Ohio ------- HEALTH EFFECTS CAUSED BY EXPOSURE TO DRINKING WATER CONTAMINANTS The following sections on the toxicity of organics and inorganics found in drinking water reflect the views of the respective authors of those sections and not necessarily the views of the Environmental Protection Agency. These reports provide necessary preliminary informa- tion with which to assess the health effects of these contaminants and will be carefully reviewed along with other investigations, such as that of the Science Advisory Board, in future discussions. A. TOXICITY OF ORGANICS PRESENT IN DRINKING WATER 1. Introduction Over the years, the occurrence of organic materials in all tap water has been acknowledged almost universally. Until relatively recently, data describing such occurrence has been almost exclusively the result of gross measurements such as carbon-chloroform-extracts and non-volatile-total organic-carbon. The advent and application of more sophisticated analytical tools, such as the mass spectrometer, has led to the conclusive identification of some of the organic components of drinking water. Appendix I .is the most recent compilation of compounds that have been found in various potable supplies. Recent estimates by the Water Supply Research Laboratory of E.P.A. indicate that of all the compounds present the identified compounds may account for no more than 10 percent by weight. The compounds in Appendix I are not all unique to drinking water. Concurrent exposure by various segments of the U.S. population exists via some foods, ambient air, occupational environment, and/or household products (e.g., over-the-counter medications, cleaning solutions, and cosmetics). For some compounds, particularly some of those suspected of being by-products of chlorination of tap water (e.g., dibromochloro- methane and bromodichloromethane), man's exposure is restricted solely to potable water and to foods processed with that water. Many factors enter into the hazard/safety evaluation of organics in drinking water. Among them is determination of the toxicity of the materials to which man is exposed. Toxicity data include a broad range of biological parameters, a few of which are listed below: 1. the amount of material required for the production of acute illness arid mortality; 2. the ways in which a compound is handled metabolically by the body; 200 ------- 3. the types of diseases and specific organs affected from repeated exposures for a part or all of the lifespan; 4. the reversability or irreversability of the lesions; 5. the particular groups of the populations that might be at greater risk to intoxication; and 6. the factors, both endogenous and exogenous, that alter the toxicity of foreign compounds and/or compromise the organism's ability to respond to insults from foreign compounds. The central objective from the investigations of such questions is the identification of what will occur in man through the utilization of predictive experimental animal models. Well designed and closely con- trolled experimentation can yield information valuable in protecting man against exposure to hazardous doses of a chemical or mixture of compounds Epidemiologic surveillance can monitor body burdens and health status as a function of exposure levels and durations of exposure to insure against the possibility of incorrect extrapolations and to guard against the un- expected sensitivity in population subsets. 2. Acute Toxicity Data on acute doses required for intoxication serve, first, as a yardstick against which to compare one compound with another and, second, as a starting point in the design of repeated exposure and metabolism studies. The comparative evaluations of acute toxicity have been forma- lized into a rating system (1) which is described in Table 1. The compounds lifted in Appendix I underwent a literature search to find data on acute toxicity and to categorize the relative toxicities according to the rating system of Gleason et. al. (1). Table 2 displays the results of this evaluation. Most of the compounds for which data are available are in the categories "moderate" and "very" toxic. For 30 percent of all the compounds, no acute toxicity data were available from which to assign a rating. While individual compounds are usually rated for their acute toxic potential, mixtures of these agents can be similarly classified. Tardiff and Deinzer (2) reported that extracts of organics from drinking water were tested in mice and found to have LDsg values that classified the mixtures as "very" toxic. The mixtures represented approximately 30 per- cent of the organics originally present in the tap water samples used. It should be remembered that acute toxicity measurements for these contaminants are based upon doses that are far greater than those en- countered from drinking water. Acute toxicity does not necessarily bear any relationship to chronic toxicity which is more relevant to low-level 201 ------- TABLE 1 CLASSIFICATION SYSTEM FOR ACUTE TOXICITY OF CHEMICALS (1) * Toxicity Rating Probable Lethal Dose for Man or Class or LD5Q for Experimental Animals 6 - Super Toxic less than 5 mg/kg 5 - Extremely Toxic 5 to 50 mg/kg 4 - Very Toxic 50 to 500 mg/kg 3 - Moderately Toxic 500 to 5000 mg/kg 2 - Slightly Toxic 5 to 15 gin/kg 1 - Practically Non-toxic greater than 15 gm/kg TABLE 2 ACUTE TOXICITY RATINGS OF COMPOUNDS IDENTIFIED IN DRINKING WATER Toxicity Rating Number of Compounds 6 - Super Toxic 1 5 - Extremely Toxic 7 4 - Very Toxic , 47 3 - Moderately Toxic 62 2 - Slightly Toxic 11 1 - Practically Non-toxic 3 Unknown ". 56 202 ------- human exposure to organic chemicals in drinking water. The following section discusses chronic toxicity. 3. Chronic Toxici ty/Carcinogenici ty Exposure to repeated small quantities of environmental chemicals suggests a greater possibility of chronic, rather than acute, intoxica- tion. One of the more serious irreversible expressions of chronic toxicity is carcinogenesis. Because of the nature of the disease, chemically-induced carcinogenesis is considered one of the more dread toxic properties. However, an entire spectrum of chronic—but non- neoplastic--diseases can be equally serious personal and societal trage- dies. Attention must be focused on all forms of chronic illness whose etiology is environmental agents. The determination that a compound at ambient concentrations is or is not a tumorgenic risk to man is a relatively difficult task as ac- knowledged by scientists of the National Cancer Institute. The observa- tion of a neoplastic response in an experimental species from exposure to a chemical invokes many questions. Perhaps one of the most significant questions concerns whether the animal model in which the carcinogenic expression was observed is predictive of the same response in man. Thus, of itself, a chemical may be a carcinogen in an experimental species (e.g., the mouse); however, the same chemical may not necessarily be a carcinogenic hazard to man. It must be emphasized that such a model can be validated by specific and definitive studies, but that such studies may not have been performed at the time the neoplastic response is dis- covered. In an effort to take into account all factors that enter into the evaluation of a compound's carcinogenic property, operational definitions were generated by the Water Supply Research Laboratory of E.P.A. with assistance from the National Cancer Institute. Those definitions are listed in Table 3. The definitions reflect the necessity to make reli- able and accurate judgments about the agents. Thus, relatively few compounds meet the criteria for "positive" carcinogen as exemplified by the brevity of the list of occupational carcinogens (OSHA list of 14 compounds). However, many more compounds are classified as "suspect" carcinogens because of the lack of sufficient and"appropriate informa- tion from which to definitely predict or acknowledge the hazard to man. (Acknowledgment of the effect via human data is never a goal with respect to cancer but may be a reality because of accidents or misjudgments.) Preceding considerations were related only to qualitative aspects of carcinogenesis: Is a compound a carcinogen or not? Is it a carcin- ogenic hazard to man or not? Such a consideration excludes the concept of potency;-namely, how potent is one compound vs. another in the induction of tumors. Stated differently, potency involves how much of a compound and how long an exposure are required to develop tumors in a defined population. For some time, oncologists have spoken of "strong" and 203 ------- TABLE 3 CLASSIFICATION SYSTEM AND CRITERIA FOR THE DEFINITION OF CARCINOGENIC PROPERTIES OF CHEMICALS* Class #1 - Positive or Recognized Carcinogen Criteria: a. On an acceptable list of human carcinogens (e.g., the OSHA list) b. Strong experimental evidence - many species and strains, etc. c. Strong evidence or strong suspicion as to cause and effect in man Class #2 - Suspect. Possible or Potential Carcinogen Criteria: a. Structure similar to proven carcinogen b. Positive response in one species c. Mutagencity data d. No epidemiologic evidence e. Either not tested or tests inadequate Class #3 - Unknown Carcinogenic Potential Criteria: a. Tests limited in time b. Tests limited in dose schedule c. Insufficient number of animals d. Route of administration not relevant e. Improper species and/or strain used f. Dose schedule not relevant: strong overlay of toxicity g. Role of contaminants h. Not tested & no structure-activity suspicion Class #4 - Negative or Non-carcinogenic Criteria: a. Repeated tests in many species and strains b. Adequate protocols c. Confirmed in several laboratories d. Established non-carcinogenic in the absence of contaminants e. Strong epidemiologic evidence that it is non-carcinogenic in man * System developed in collaboration with the National Cancer Institute. 204 ------- "weak" carcinogens implying a difference in dose to obtain the same effects (e.g., 50 percent tumor formation). If, for example, a compound at environmental levels requires 100 years of exposure to induce tumor formation in man, this compound may be regarded as a relatively small cancer hazard to society as compared to one which requires only a decade to obtain a similar response/ Although the regulatory agencies will wish to exercise control over all carcinogenic substances within their respective jurisdictions, consideration of potency may assist in estab- lishing priorities for control measures and for allocation of resources. The recognition of the tumor-inductive property of chemicals lecf,to a cursory examination of biological literature to determine the evidence both positive and negative, for carcinogenic responses induced by the chemicals identified in tap water (Appendix I). The results of this evaluation are listed in Table 4. The "positive" or known carcinogens are aldrin, benzene, benzopyrene, carbon tetrachloride, DDT, 2,4-dichlor- ophenol5 dieldrin, and 2,4-dimethyl phenol. The "negative" or non-carcinogens are acetic acid, acetone, barbital, benzoic acid, ethanol, and methanol. Of the 187 compounds, there was no data and insufficient structure-activity information to make a judgment of 137 chemicals (i.e., over 70 percent of those found to have been present in tap water and to which some humans were exposed). It must be concluded that although more information must be obtained on "suspect" carcinogens, a great deal more experimental evidence must be learned about the chronic toxicity of a substantial number of compounds. Con- tinued search for additional chronic toxicity data may yield additional pertinent information on these compounds. A few studies (3-5) have been reported in which organic mixtures extracted from drinking water were administered repeatedly to determine carcinogenic potential. The results indicated that, in mice, carbon extracts elicited neoplastic responses when injected but not when in- gested (3,4). In another investigation (5), injections of carbon extract of organics from drinking water failed to induce tumor formation. TABLE 4 CARCINOGENICITT CLASSIFICATION OF COMPOUNDS IDENTIFIED IN DRINKING WATER Class 1 - Positive 2 - Suspect 3 - Unknown 4 - Negative Number of 8 35 138 6 Compounds 205 - ------- 4. Ongoing Research The Water Supply Research Laboratory of E.P.A. is actively engaged in investigating the toxicity of organics in drinking water for the purpose of identifying hazards and risks to man's health via this mode of exposure and of determining, if no hazard exists, the magnitude of the margin of safety from environmental exposures. Toxicologic experimentation on the organics in drinking water is guided by Principles for Evaluating Chemicals in the Environment (6). A two-pronged approach is being used to investigate the organics in tap water. The first studies the biological effects of individual com- pounds. The second is aimed at the elucidation of the toxic properties of mixtures of organics which are present in tap water. Several classes of compounds identified in potable water are under active investigation with regard to their toxicity in experimental ani- mals. These classes include the chlorinated ethers, the chlorinated and brominated benzenes, and the halogenated methanes. Investigations are designed (1) to determine the most significant animal model through studies of comparative metabolism and (2) to uncover pathologic changes resulting from varying levels of repeatedly administered compounds in appropriate experimental species. The chloro-ethers of immediate interest are bis(2-chloroethyl) ether and bis(2-chloroisopropyl) ether. The metabolism of these agents is being studied in several species including sub-human primates in order to determine the species that most closely resembles man in its metabolism so that additional toxicity studies may be performed in a species that is more predictive of man's response. Base-line data are being obtained on the effects from single and relatively short-term repeated exposures in one classical model. A specialized model is being utilized to determine possible carcinogenic potential. Investigations have been designed to establish any mutagenic activity that might be of concern to man. Because very little is known of the toxicity of these compounds (although they are chemically related to a potent toxicant and carcinogen), a broad scope of experimentation is required on these com- pounds. Halogen-substituted benzenes demonstrate a relatively long biologi- cal half-life that suggests accumulation in the body with repeated expo- sures with consequent chronic toxicity. Because of evidence suggesting the acute alteration of zenobiotic metabolism, these compounds are being studied to determine their potential interaction with other foreign com- pounds to alter toxicity (e.g., synergistic responses). The entire homologous series of chlorine- and bromine-substituted benzenes are under investigation. 206 ------- FIGURE 1. PROTOCOL FOR TOXICITY SCREENING OF ORGANIC CONCENTRATES FROM DRINKING WATER CONCENTRATES i r^. nff*^^^^^^^^^^^^ •-050 ** MICE i^H LOW HIGH BACTERIAL MUTAGENIC ASSAYS- — , 4 POSITIVE FR ACTIONS -^ POSITIVE *«» &^ -^NEGAT,VE ^^ NEGATIVE ^^^^^^ MAMMALIAN CELL TRANSFORMATION ASSAY «=— — -^_ POSITIVE FRACTIONS- _ * ^^^^^^^^ V POSITIVE B V ^-T**rHFMirAI r.HARAP.TFRI7ATI ro 0 ^j LIST OF COMPONENTS & CONCENTRATIONS IN n •NEGATIVE NEGATIVE SUSPECTED OF BEING HIGHLY TOXIC EVALUATION. TERATOLOGICAL ASSAY POSITIVE FRACTIONS POSITIVE 'SUSPECTED OF BEING NON-TOXIC RETEST IN POSITIVE SCREEN(S) SPECIFIC & SPECIALIZED TOXICITY TESTS rr V ^HAZARD EVALUATION NEGATIVE NEGATIVE ------- Halogenated methanes (dibromochloromethane and bromodichloromethane) are possible chlorination by-products for which there is presently no toxicity information. A broad spectrum of experimentation is planned including comparative metabolism and comparative toxicity with special emphasis on chronic toxicoses. The investigation of the toxicity of mixtures of organics from tap water is described schematically in Figure 1. Extracts or concentrates of organics from municipal water supplies can be screened with biological systems to determine what types of toxicity problems to investigate further and to establish which water supplies may have the greater poten- tial for adverse health effects. Presently extracts are being obtained from five U.S. cities that represent the major types of water sources for drinking water. These extracts will be subjected to the four screening systems identified in Figure 1. The LD5Q is utilized as a reference for comparison with the toxicity of known compounds and with the toxicity of other concentrates. The in vitro systems are used predominantly to predict possible mutagenic and carcinogenic expressions in vivo. The teratology assay is performed in whole animals and indicates the ability to induce birth deformities. Positive responses in any of the screening assays initiate an attempt to isolate the effects in subfractions of the extracts. By iso- lating a few fractions with biological activity, it is then more feasible to identify the constituents within the active fractions rather than in the entire concentrate. Chemical identification of components requires a reconfirmation of the pure agent in the positive screens. Subsequent to reconfirmation, the active compounds are subjected to more definitive investigations for ultimate evaluation of impact on man. Throughout these investigations, substantial efforts are expended in coordination and collaboration with scientists and administrators of other federal agencies such as the National Cancer Institute, the Food and Drug Administration, and the National Institute of Environmental Health Sciences. Through such interactions, governmental resources are maximally utilized for the benefit of the citizens. 5. References (1) Gleason, Marion N., Gosselin, Robert E., Hodge, Harold C., and Smith, Roger P. Clinical Toxicology of Commercial Products: Acute Poisoning. 3d ed. Baltimore, WilTTani? and Wilkins, Co., 1969. 208 ------- (2) Tardiff, Robert G. and Deinzer, M. "Toxicity of organic compounds in drinking water." In: Proceedings of 15th Water Quality Conference, Feb. 7-8, 1973, University of Illinois, pp. 23-37. (3) Hueper, W. C. and Payne, W. W. "Carcinogenic effects of raw and finished water' supplies." Amer. J. Clin. Path. 39(5):475 May 1963. (4) Hueper, W. C. and Ruchoft, C. C. "Carcinogenic studies on adsorbates of industrially polluted raw and finished water supplies." Arch. Ind. Hyg. Occup. Med. 9:488-95, 1954. (5) Dunham, Lucia J., O'Gara, Roger W., and Taylor, Floyd B. "Studies on pollutants from processed water: collection from three stations and biologic testing for toxicity and carcino- genesis." Amer. J. Public Health 57(12):2178-85, December 1967. (6) Principles for Evaluating Chemicals in the Environment. Washington, D.C., National Academy of Sciences, 1975. 209 ------- B. TOXICITY OF INORGANIC CHEMICALS PRESENT IN DRINKING WATER 1. Introduction Because of health effects concerns the concentration of several inorganic chemicals are limited in drinking water. Limits for arsenic, barium, cadmium, chromium, cyanide, fluoride, lead, mercury, nitrate, selenium, and silver have been proposed and published in the Federal Register (1) under provisions of P.I. 93-523: Safe Drinking Hater Act. In the 27-year period 1946-1973, there were 405 waterborne disease outbreaks but only 10 of these outbreaks were related to inorganic chem- ical poisonings. Deaths were more likely to be associated with these chemical-caused outbreaks; seven deaths occurred as well as 210 cases of illness (2). Cancer has not been attributed to have been caused by con- tamination of drinking water with inorganic chemicals in this country. None of the inorganic chemicals have been limited in drinking water because the chemical was considered to be a carcinogen but for several of the chemicals (arsenic, cadmium, chromium, nitrate, and selenium) consideration was given to data concerning carcinogenic effects. Beryl- lium and nickel are not limited in drinking water but are two additional metals that should be considered for carcinogenic effects. 2. Arsenic In certain parts of the world the high levels of arsenic found in drinking water have been associated with a high rate of arsenicism and skin cancer in the population (3). Tseng et. al. (12) reported a geo- graphical correlation in Taiwan between levels of arsenic exposure in well water and the frequencies of skin cancer, hyperpigmentation, Kera- tosis, and a peripheral vascular disorder (Blackfoot disease). A dose- response relationship was seen between the occurrence of skin lesions, including cancer, and the arsenic content of the water. No excessive occurrence of other cancers has been reported in areas where the water contains arsenic. The available studies consistently point to a causal relationship between skin cancer and heavy exposure to inorganic arsenic in drugs, in drinking water with a high arsenic content, or in the occu- pational environment. Adequate oral studies on arsenic trioxide in the mouse and on lead arsenate, calcium arsenate, sodium arsenate, arsenic trioxide and sodium arsenite in the rat gave negative results. It should be noted that OSHA has formally proposed a new limit for inorganic arsenic of 4 yg/m3; the previous limit suggested by NIOSH was 50 yg/m3 (5). The 4 yg/m3 limit represents an-arsenic intake of 40 yg 210 ------- per work week. Extrapolating to water exposure the 40 pg/week of arsenic would represent a 4 yg/liter intake for water. Applying a safety factor of 100, the comparable drinking water standard would be 0.04 ug/1. Arsenic has usually been considered a geochemical contaminant and high concentrations have been noted in ground water in selected areas of the southwest and northwest of the country. Water supplies exceeding the limit of 0.05 mg per liter are located in California, Oregon, and Nevada. Arsenic was related to five, or half of the inorganic chemical- caused water poisonings in the past 27 years, and the reasons for these poisonings are most varied. One outbreak of ten cases and three deaths resulted when an arsenical weed killer was dumped into a well in West Virginia. These are the only murders that have been noted in the review of waterborne outbreaks. Two outbreaks concerned the back-siphonage of arsenic compounds into water supplies, and there were five cases of ill- ness and four deaths resulting. Recently, a well was drilled at a new factory site in Minnesota and people working there became ill (10). Arsenic was detected in their blood and investigation revealed arsenic in the well water of 11.8 - 21 nig/liter. The site had been used to mix .grasshopper bait many years before and it is likely that some unused pesticide had been buried where the well was drilled. Two girls in a Nevada family became ill and, after some difficulty of diagnosis, it was determined that they had arsenic poisoning. The well at the family ranch varied between 0.5 - 2.75 mg/liter of arsenic from natural causes. Health effects research planned for arsenic includes a study of body burden in areas where arsenic is high in drinking water. The mutageni- city of arsenic will be determined by use of cultured mammalian cells. 3. Beryl!ium Bone and lung cancers have been produced experimentally in animals and 20 malignant tumors have been recorded among the 735 cases of beryl- lium disease; however, the available evidence was not considered suffi- cient to positively incriminate beryllium as a carcinogen in humans (6). Beryllium is classified as an experimental carcinogen by the American Conference of Government Industrial Hygienists (7). They define an ex- perimental carcinogen as industrial substances found to be capable of inducing tumors under experimental conditions in animals and have estab- lished a TLV of 0.002 mg/m3 of air. Beryllium will be tested for mutagenicity in a cultured mammalian cell test system. 4. Cadmium Several studies suggest that occupational exposure to cadmium oxide may increase the risk of prostate cancer in man but the size of the groups studied was considered small (3). It was recently reported that there was an increased risk of death due to malignant neoplasms in a 211 ------- study of 283 cadmium smelter workers (5). No data are available to suggest that non-occupational exposure to cadmium constitutes a carcino- genic hazard. Studies of rats and mice shov/ed that a level of 5 mg/1 cadmium acetate given in drinking water until death did not significantly increase the incidence of tumors (3). The estimated intake of cadmium from drinking water is 3 \\g per day (1). Health effects research currently being conducted and planned is in regard to the possible role of cadmium in hypertension and cardiovascular disease. The mutagenicity of cadmium will be tested in a cultured mam- malian cell test system. The relative bioavailability of cadmium in water as opposed to cadmium in foodstuffs is also being determined. 5. Chromium There is an excessive risk of lung cancer among workers in the chromate-producing industry (3,4). It is likely that exposure to one or more chromium compounds is responsible, but the identity of this or these is not known. There is no evidence that non-occupational exposure to chromium constitutes a cancer hazard. The NAS reports that rw harmful effects on the health of man are known to have resulted from the presence of chromium in public drinking water at current concentrations (4). Studies of rats and mice she-wed that a level of 5 mg/1 chromic acetate given in drinking water until death did not significantly increase the incidence of tumors at various sites as compared with controls (3). The estimated intake of chromium from drinking water is 5 ug per day (1). No health effects research is planned other than testing the muta- genicity of chromium in a cultured mammalian cell test system. 6. Nickel There has been an excessive risk of cancers of the nasal sinus and lung among nickel refinery workers and it is probable that nickel in some form is carcinogenic (3,4). There is no evidence to suggest that non- occupational exposure to nickel constitutes a cancer hazard (3). The estimated intake of nickel from drinking water is 10 yg per day (1). It is planned to produce a criteria document recommending a drinking water standard for nickel. The mutagenicity of nickel will be determined by use of a cultured mammalian cell test.system. 7. Nitrate Nitrate concentrations in drinking water have been limited because of the possibility of developing methemoglobinemia in infants who were fed water high in nitrate. A few community water systems exceed the nitrate limit but in many rural areas the farm wells have a very high concentration of nitrate. It has been hypothesized that in high concen- trations the nitrogen might combine with amines in contaminated water or 212 ------- in the gastrointestinal tract to form nitrosamines, a recognized carcino- Qen. The development of nitrosamines has been demonstrated experimentally using much higher concentrations of nitrates or nitrites than are known to occur in water. It has been pointed out that a few counties of Texas that had nitrate-in-ground-water problems had higher cancer rates but a suitable data base for an epidemiological study was not available. The production of nitrosamines in cured meat is being researched by other agencies. The concentrations of nitrate and nitrite are greater when these chemicals are used as a preservative of food than drinking water concentrations. 8. Selenium In 1962 the drinking water limit was lowered to 0.01 mg per liter primarily out of concern for possible carcinogenic properties of the element. Since that time evidence has been developed indicating that selenium could both cause and prevent cancer. Several animal studies showed that tumors were developed from exposure to selenium. In the North Central and Rocky Mountain Regions of the country there are areas that are geochemically rich in selenium. Forage crops and plants in these areas often contain more than 100 parts per million of selenium. Cows, sheep, and horses in these areas may die from consuming enough selenium in forages to develop selenium toxicity. Research has shown that grain from selenium-rich areas had a higher selenium content and when used as poultry feed, it promoted the growth of chickens and tur- keys. It was proposed that selenium be used as an additive to animal feed. The Commissioner of the Food and Drug Administration reviewed the carcinogenic problem of selenium last year (11). He concluded that selenium could be safely used as an additive to swine, turkey, and chicken feed because of its nutritive value and lack of health hazard when used at prescribed concentrations. The inadequacy of the toxi- cological studies that produced tumors was reviewed. Research is being conducted on the comparative availability of selenium from food and water so that a drinking water limit can be estab- lished with consideration given to intake from food. A study is planned to determine the human body burden in areas where selenium is high in drinking water. Mutagenic screening tests will also be conducted. 9. Consequences Apparently, the inhalation exposure to fumes or dust in the indus- trial setting produces a very different biological effect that the in- gestion exposure from food and water. An increased risk of developing cancer is not expected from consuming water contaminated with beryllium, cadmium, chromium, or nickel. There are other health effects that re- quire limiting the concentration of these elements in drinking water. 213 ------- Arsenic has been demonstrated to be a carcinogen in Drinking water but may not present as serious a problem as indicated from industrial inhalation exposure. Epidemiological research should be conducted to see if a lower drinking water limit is necessary. More information is needed on the formation of nitrosamines and on-going research should provide this. Selenium apparently does not present a cancer problem. 10. References (1) Train, R. E. Primary Drinking Water, Proposed Interim Standards. (2) McCabe, L. J. Problem of Trace Metals in Hater Supplies - An Overview. Proceedings IGth Water Quality Conference, University of Illinois (1974). (3) IARC Monographs on the Evaluation of Carcinogenic Risk of Chemicals to Man, Some Inorganic and Organometallic Compounds, Volume 2, International Agency for Research on Cancer, Lyon (1973). (4) National Academy of Sciences, Medical and Biologic Effects of Environmental Pollutants, Chromium (ISBN 0-309-02217-7) 1974; Nickel (ISBN 0-309-02314-9) 1975. (5) Toxic Materials News, Vol. 2(7), p. 53 (1975). (6) Preliminary Air Pollution Survey of Beryllium and its Compounds, USDHEW, Raleigh, North Carolina ("1969). (7) ACGIH, P.O. Box 1937, Cincinnati, Ohio. (8) McCabe, L. J. et_. a\_., "Survey of .Community Water Supply Systems." JAWWA, £2(11), 670 (1970). (9) Kopp, J. F., and Kroner, R. C., Trace Metals in Waters of the U.S., Cincinnati, Ohio. (10) Feinglass, E. J. Arsenic Intoxication from Well Water in the United States. New England J. Med. 288, 828(1973); Federal Register 40(51) 11990-11998 (March 14, 1975). (11) Schmidt, A. M. Selenium in Animal Feed, Federal Register 39(5) 1355-1358 (January 8, 1974). (12) Tseng, W. P., Chu, H. M., How, S. W., Forg, J. M., Lin, C. S., and Yels. Prevalance of Skin Cancer in an Endemic Area of Chronic Arsenicism in Taiwan, J. Nat. Cancer Inst. 40, 454-463 (1968) 214 A U.S. GOVERNMENT PRINTING OFFICE 1975- 210-810/13 ------- |