MONITORING TO DETECT PREVIOUSLY UNRECOGNIZED POLLUTANTS IN SURFACE WATER July 1977 Office of Toxic Substances Environmental Protection Agency Washington, D.C. 20460 ------- EPA-560/7-77-001 MONITORING TO DETECT PREVIOUSLY UNRECOGNIZED POLLUTANTS IN SURFACE WATERS INSTITUTE FOR ENVIRONMENTAL STUDIES University of Illinois at Urbana-Champaign B. B. Ewing Principal E. S. K. Chian Investigators J. C. Cook J. C. Means F. B. DeWalle R. Milberg C. A. Evans E. G. Perkins P. K. Hopke J. D. Sherwood J. H. Kim W. H. Wadlin Final Report Contract No. 68-01-3234 July 1977 Project Officer—Vincent J. DeCarlo Prepared for Office of Toxic Substances U.S. Environmental Protection Agency Hashington, D.C. 20460 Document is available to the public through the National Technical Information Service, Springfield, Virginia 22151 ------- NOTICE This report has been reviewed by the Office of Toxic Substances, EPA, and approved for publication. Aprroval does not signify that the contents necessarily reflect the views and policies of the Environmental Protection Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use. n ------- CONTENTS LIST OF ILLUSTRATIONS iv LIST OF TABLES v ACKNOWLEDGMENTS vii INTRODUCTION 1 Background and Objectives 1 Organization 1 Sample Collection and Analysis Schedule 3 SITE SELECTION, SAMPLE COLLECTION, AND GROSS ANALYSIS Site Selection 6 Sample Collection 29 Gross Analysis 32 SAMPLE PREPARATION Stripping Procedures 35 Liquid-liquid Extraction 37 Sorptive Extraction Technique 39 INORGANIC ANALYSIS 42 Spark-source Mass Spectral Analysis 44 Instrumental Neutron Activation Analysis 46 Energy-dispersive X-ray Fluorescence Analysis 49 ORGANIC ANALYSIS Identification of Organics 52 Quantitation of Organics 54 Summary of the Organic Compounds Found 63 REFERENCES 75 ------- ILLUSTRATIONS 1. Industrialized Areas Sampled 2 2. Sample Collection and Analysis Schedule 4 3. Sites Sampled in the Chicago Metropolitan Area 8 4. Sites Sampled in the Illinois River Basin 9 5. Sites Sampled in the Delaware River Basin 11 6. Sites Sampled in the Hudson River Basin 13 7. Sites Sampled on the Upper and Middle Mississippi River 15 8. Sites Sampled on the Lower Mississippi River 16 9. Sites Sampled in the Houston Area 17 10. Sites Sampled in Alabama 18 11. Sites Sampled in the Ohio River Basin 20 12. Sites Sampled in the Great Lakes and Their Tributaries 22 13. Sites Sampled in the Tennessee River Basin 23 14. Sites Sampled in the Greater Los Angeles Area 25 15. Sites Sampled in the San Francisco Bay Area 26 16. Sites Sampled on the Willamette River and in the Greater Portland Area 27 17. Sites Sampled in the Seattle-Tacoma Area 28 18. Rate of Water Inflow into a Sample Bottle Suspended at Constant Depths 30 19. Rate of Water Inflow into a Sample Bottle Released at Different Heights 31 20. Procedure for Stripping Volatile Organic Compounds 36 21. Procedure for the Extraction of the Less-volatile Organics 38 22. Flow Diagram of Experimental Setup for Sorptive Studies 41 23. Schematic Diagram of Volatile Organic Elution System 55 24. Gas-chromatographic Trace of a Representative System Blank 59 IV ------- TABLES 1. Sampling Sites in the Chicago Area and the Illinois River Basin 7 2. Sampling Sites in the Delaware River Basin 10 3. Sampling Sites in the Hudson River Basin 12 4. Sampling Sites in the Mississippi River Basin, in Alabama, and in Texas 14 5. Sampling Sites in the Ohio River Basin 19 6. Sampling Sites in the Great Lakes and the Tennessee River Basin 21 7. Sampling Sites on the West Coast 24 8. Location of Gross Analysis Data for All Samples 33 9. Summary of Gross Analysis Data 34 10. Directory of Inorganic Analyses 43 11. Gas Chromatograph Column Conditions 56 12. Extraction Efficiencies of Selected Amines 61 13. Relative Retention Times and Relative Response Factors for Selected Amines on 3% OV-17 62 14. Relative Retention Times and Relative Response Factors for Selected Amines on 10% Apiezon L/2% KOH 62 15. Acid-extractable Compounds and Their Frequency of Occurrence 65 16. Base-extractable Compounds and Their Frequency of Occurrence 69 17. Volatile Compounds and Their Frequency of Occurrence 72 ------- ACKNOWLEDGMENTS The Institute for Environmental Studies and the project team members express their appreciation to Dr. Larry Keith, formerly of the USEPA laboratory in Athens, Georgia, to Dr. John McGuire of the Athens laboratory, and to Dr. Billy Fairless of the USEPA Region V office for valuable advice on the refine- ment of sample preparation and analysis techniques. Gratitude is also expressed to the many local, state, and federal officials whose cooperation greatly facilitated the process of locating sampling sites and collecting samples. ------- 1, INTRODUCTION This report summarizes the activities and accomplishments of a research project conducted to detect previously unrecognized pollutants in surface waters. The work was supported by the U. S. Environmental Protection Agency under Contract No. 68-01-3234. BACKGROUND AND OBJECTIVES The heavy concentration of industry in certain areas of the United States has caused increasing concern about the introduction of contaminants into our surface waters, especially since some constituents of industrial waste discharges have proven to be carcinogenic or toxic at trace levels. With the proliferation of new chemical substances it is likely that some potentially harmful pollutants in our surface waters have gone undetected. At the same time as waste discharges to our waterways have increased, we have come to rely more heavily on those water resources not only for industrial but also for municipal water supplies. It was therefore deemed vital to detect contaminants wherever they may be present. The purpose of the present study was to undertake a sampling and ana- lytical survey to determine, insofar as possible, the identities and semi quan- titative concentrations of organic compounds and inorganic elements present in the waterways around industrial centers in the United States. A total of 204 water samples were collected from fourteen heavily industrialized river basins. These areas and the number of samples taken from each are indicated in Figure 1. Each sample was analyzed using state-of-the- art techniques for detecting trace contaminants. ORGANIZATION This interdisciplinary project, conducted at the University of Illinois at Urbana-Champaign, was administered by the Institute for Environmental Studies. ------- CHATTANOOGA MEMPHIS —v^/x"~ Encircled numbers indicate quantity of samples to be collected in each area. ro Figure 1. Industrialized areas sampled. ------- It involved the personnel and facilities of five university units. The pro- ject was directed by the co-principal investigators, Professor E. S. K. Chian, Department of Civil Engineering, and Professor B. B. Ewing, Director of the Institute. The participating units and principal research personnel were: Department of Civil Engineering Site Selection—E. S. K. Chian and F. B. DeWalle Sample Collection—F. B. DeWalle Sample Preparation—E. S. K. Chian and J. H. Kim Department of Chemistry Identification of Organics—K. L. Rinehart, J. C. Cook, and R. Milberg Department of Food Science Quantification of Organics—E. G. Perkins and J. C. Means Materials Research Laboratory Inorganic Analysis (SSMS)—C. A. Evans and W. H. Wadlin Institute for Environmental Studies Inorganic Analysis (INAA and XRF)—P. K. Hopke and J. D. Sherwood Editing and Publication—T. W. Knecht SAMPLE COLLECTION AND ANALYSIS SCHEDULE The research contract was executed for initiation on July 1, 1975, for an 18-month period to terminate December 31, 1976. The first six samples were collected in the Chicago area in August 1975. The initial six months of the project were devoted to staffing, development of methods, and the analysis of these six samples. The sampling rate was then increased so that all sampling would be completed by September 1976 and was adjusted as necessary to keep sample collections only moderately in advance of the analysis process. The sampling schedule for the entire project is shown in Figure 2. All samples were numbered in the chronological sequence of collection. ------- 220 200 180 160 (/> £ 140 5> 120 .a E 100 80 60 40 20 0 E - Extractable Organic Analyses VOA - Volatile Organic Analyses SSMS - Spark-Source Mass Spectral Analyses INAA - Instrumental Neutron Activation Analyses Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan 1975 1976 Date Figure 2. Sample collection and analysis schedule. ------- Inorganic analyses by x-ray fluorescence and spark-source mass spec- trometry were completed at a rate of approximately 20 samples per month, closely following the delivery of the samples to the University of Illinois campus. It was intended that instrumental neutron activation analysis be performed at the same rate, but after processing 28 samples the schedule was delayed for about three months because of an equipment failure. Analysis for volatile organics was delayed a few months after pro- cessing the initial six samples so that the methodology could be revised to improve detection limits. Thereafter, the rate of analysis was increased to about 30 samples per month until the backlog was reduced, after which it followed the sampling schedule closely. The preparation and analysis of extractable organics generally followed the collection schedule throughout the last 12 months of the project. The organic and inorganic analysis schedules are also shown in Figure 2. ------- 2, SITE SELECTION, SAMPLE COLLECTION, AND GROSS ANALYSIS SITE SELECTION E. S. K. Chian F. B. DeWalle During the study, water samples were collected from 204 sites across the continental United States. The areas sampled and the number of samples collected in each are indicated in Figure 1. The sites were chosen in such a way that the concentration of organic and inorganic contaminants in the col- lected water samples would be affected by industrial pollutants and so that all principal types of industry would be represented. So that the impact of indus- trial discharges on water quality could be assessed, a limited number of samples were generally taken upstream from industrial sources in the less polluted reaches of rivers while the majority of the samples were collected from the waterways near major industrial areas. Several downstream sites were also sampled to indicate the extent to which contaminant concentrations were atten- uated. The samples were generally taken at sampling locations established by state or federal water pollution regulatory agencies. The sites are listed in Tables 1 through 7 along with the coordinates and description of their locations. The maps presented in Figures 3 through 17 show the general position of each site along the waterways sampled. Ninety-one of the sites were located along major rivers such as the Hudson, the Delaware, the Mississippi, the Ohio, and the Tennessee. Fifty-seven samples were collected in tidal areas and estuaries, such as the Hudson River estuary, the Delaware River estuary, Mobile Bay, Galveston Bay, Los Angeles Harbor, San Francisco Bay, and Puget Sound. Twelve sites were located in manmade canals and three in major lakes. Since industrial wastewater is often treated at municipal sewage treatment plants, four samples were taken from effluent discharge structures. ------- Table 1 Sampling Sites in the Chicago Area and the Illinois River Basin Sample Number Waterway Station Latitude Longitude Nearest Bridge, Point, River or Highway Nearest Town Remarks WSW Sewage Treatment Plant 41.48.51 87.46.11 Chicago Central Water Wks Calumet-Sag Channel Highway 83 Bridge Calumet-Sag Channel Ashland Avenue Bridge Chicago Sanitary Lockport Powerhouse & Ship Canal Chicago Sanitary & Ship Canal 41.53.45 41.41.53 41.39.22 41.34.08 87.36.20 87.56.12 87.39.39 88.04.41 Highway 83 Bridge 41.42,02 87.56.22 Pershing & Austin Roads Lake Shore Drive & Ohio St. U.S. Highway 83 Ashland Avenue U.S. Highway 83 Stickney, IL Chicago, IL Lemont, IL Blue Island, IL Lockport, IL Lemont, IL Final effluent after sec. sedimentation & chlorination Final tap water Midstream Midstream From sideline of large water tunnel. Midstream 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Chicago Sanitary & Ship Canal — — Indiana Harbor Cnl . Calumet River Calumet-Sag Channel Calumet-Sag Channel — Chicago Sanitary & Ship Canal Calumet River Calumet-Sag Channel Des Plaines River Illinois River Illinois River Illinois River Illinois River Illinois River Illinois River Illinois River Illinois River Chicago Central Water Wks Chicago Central Water Wks Lockport Powerhouse North Side Sewage Treatment Plant West Side Sewage Treatment Plant Calumet Sewage Treatment Plant South West Filtration Plant Indiana Harbor Calumet River Ashland Avenue Bridge Highway 83 Bridge Chicago Central Water Wks Lockport Powerhouse Highway 41 Bridge Ashland Avenue Bridge Des Plaines River Dresden Island Lock & Dam Utica, Illinois Hennepin, Illinois Peoria Water Works Peoria Water Works Pekin, Illinois Havana, Illinois Meredosia, Illinois Hardin, Illinois 41.53.45 41.53.45 41.34.08 42.01.11 41.48.51 41.39.36 41.47.10 41.39.19 41.39.36 41.39.22 41.41.53 41.53.45 41.34.08 41.43.37 41.39.22 41.25.25 41.23.53 41.19.29 41.15.00 40.43.30 40.43.30 40.34.25 40.18.00 39.50.00 39.10.00 87.36.20 87.36.20 88.04.41 87.42.42 87.46.11 87.44.23 87.32.00 87.27.34 87.44.23 87.39.39 87.56.12 87.36.20 88.04.41 87.42.30 87.39.39 88.11.35 88.16.45 89.02.00 89.23.00 89.33.10 89.33.10 89.39.15 90.04.00 90.34.00 90.37.00 Lake Shore Drive & Ohio St. Lake Shore Drive 8 Ohio St. -.- Howard and McCormick Blvd. Pershing and Austin Roads 130th St. & Lawrence Ave. South Shore Dr. & Chattenham Dickey Road 130th Street Bridge Ashland Avenue U.S. Highway 83 Lake Shore Drive » Ohio St. ... U.S. Highway 83 Ashland Avenue U.S. Highway 55 — U. S. Highway 178 U. S. Highway 26 — — U. S. Highway 9 U. S. Hiahway 97 U. S. Highway 104 U. S. Highway 100 Chicago, IL Chicago, IL Lockport, IL Lincolnwood, IL Stickney, IL Chicago, IL Chicago, IL E. Chicago, IN Chicago, IL Blue Island.'IL Lemont, IL Chicago, IL Lockport, IL Chicago, IL Blue Island, IL El wood, IL Dresden, IL Utica, IL Hennepin, IL Peoria, IL Peoria, IL Pekin, IL Havana, IL Meredosia, IL Hardin, IL Untreated L. Michigan V'tr. Final tap water after chlorination Tunnel water (field extracted & stripped) Final effluent after chlorination Final effluent after chlorination Final effluent after chlorination Final tap water after chlorination Midstream Midstream Midstream Midstream Final tap water (XAD, carbon extractions) Tunnel water (XAS, carbon extractions) Near mouth, midstream Midstream Midstream Midstream Midstream Midstream Untreated river water Finished water Midstream Midstream Midstream Midstream ------- LAKE MICHIGAN Sampling locations shown within concentric circles Figure 3. Sites sampled in the Chicago metropolitan area. ------- ((Sampling Locations 1 .inch = 38 miles Figure 4. Sites sampled in the Illinois River basin. ------- Table 2 Sampling Sites in the Delaware River Basin Sample Number Waterway 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 Delaware River Delaware River Delaware River Delaware River Delaware River Delaware River Delaware River Delaware River Delaware River Delaware River Delaware River Delaware River Delaware River Delaware River Delaware River Delaware River Delaware River Schuylkill River Schuylkill River Schuylkill River Delaware River Delaware River Lehigh River Lehigh River Lehigh River Lehigh River Delaware River Delaware River Delaware River Delaware River Station St. John Reedy Island D. Memorial Bridge Marcus Hook Paulsboro Navy Yard B. Franklin Bridge Five Mile Point Torresdale Range Bristol D. Memorial Bridge Mouth Mouth Mouth Mouth Margaretville St. John Mouth Queens Lane Queens Lane Torresdale Torresdale Allentown Hill to Hill Bethlehem Easton Easton Trenton Trenton Frenchtown Latitude 39. 39. 18. 30. 39.42. 39. 39. 39. 39. 39. 40. 40. 39. 38. 38. 39. 39. 42. 39, 39. 39. 39. 40. 40. 47. .50. 52. 57. 53. .02. 05. 42. ,49. .58. 14 46 35 55 54 39 10 40 00 13 35 60 53 03.03 .10. 50 22.42 ,18. 53. ,58. 58. 02. 02. 40.37. 40. 40. 40. 40. 40. 40. 40. 36. 37. .41. .42. .13. ,13. ,31. 14 24 00 00 24 24 22 58 11 13 43 17 17 40 Longitude 75.22.57 75.33.12 75.32.13 75.25.48 75.15.53 75.11.45 75.08.10 75.04.35 74.59.20 74.51.12 75.32.13 75.01.40 75.07.42 75.10.00 75.16.24 74.32.18 75.22.57 75.11.45 75.11.05 75.11.05 74.59.40 74.59.40 75.28.57 75.22.40 75.20.11 75.12.32 75.11.48 74.46.44 74 . 46 . 44 75.04.00 Nearest Point, River Bridge, or Highway Bombay Hook Pt. — U.S. Highway 295 Blue Ball Avenue Little Tinicum Island West Horseshoe Range U.S. Highway 676 Frankford Creek Pennypack Creek Otter Creek U.S. Highway 295 Cape Henlopen Brandywine Shoal Fourteen Foot Bank Elbow of Cross Ledge McGregor Mnt. on U.S. Hwy 30 Bombay Hook Point Reserve Basin U.S. Hwy 1 to U.S. Hwy 76 U.S. Hwy 1 to U.S. Hwy 76 City Water Treatment Plant City Water Treatment Plant Tilghman Street Wyandelle Ave. , P.S. Hwy 378 Freemonsburg Bridge S. Delaware St., P.S. Hwy 611 N. Delaware Drive, City WTP City WTP, P.S. Hwy 29 City WTP, P.S. Hwy 29 N.J.S. Hwy 12 Nearest Town Woodland Beach, DE Port Penn, DE Pigeon Point, DE Marcus Hook, PA Paulsboro, NJ Philadelphia, PA Philadelphia, PA Brides burg, PA Torresdale, PA Bristol , PA Pigeon Point, DE Lewes, DE Fowlers Beach, DE Big Stone Beach, DE Pickering Beach, DE Margaretville, DE Woodland Beach, DE Philadelphia, PA Philadelphia, PA Philadelphia, PA Torresdale, PA Torresdale, PA Allentown, PA Bethlehem, PA Bethlehem, PA Easton, PA Easton, PA Trenton, NJ Trenton, NJ Frenchtown, NJ Remarks mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid mid channel ; channel; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel ; channel; channel ; channel ; channel ; channel ; low slack tide low slack tide low slack tide low slack tide low si ack tide low slack tide low slack tide low slack tide low slack tide low slack tide high slack tide high slack tide high slack tide hiqh slack tide high slack tide high slack tide high slack tide low slack tide non tidal finished water high slack tide finished water non tidal non tidal non tidal non tidal non tidal non tidal finished water non tidal ------- 11 LEHIGH RIVER 57( 61 SCHUYLKILL RIVER 60 Figure 5. Sites sampled in the Delaware River basin. ------- Table 3 Sampling Sites in the Hudson River Basin Sample Number 62 63 64 65 66 57 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 Waterway Raritan Bay Raritan Bay Arthur Kill Arthur Kill Arthur Kill Arthur Kill Arthur Kill Newark Bay Hudson River Hudson River Hudson River Hudson River Hudson River Hudson River Hudson River Hudson River Hudson River Hudson River Hudson River Hudson River Hudson River Mohawk River Mohawk River Passaic River Hackensack River Hudson River Hudson River Hudson River Station Perth Amboy Perth Amboy Perth Amboy Sewaren Tufts Point Tremley Point Port Elizabeth Newark Bayonne Narrows Lower Bay Beacon Poughkeepsie Poughkeepsie Kingston Catskill Glenmont Waterford Thomson Glens Falls Corinth Schenectady Waterford Mouth Mouth Fort Lee Piermont lona Latitude 40.29.12 40.29.46 40.30.44 40.33.05 40.34.42 40.36.17 40.38.47 40.39.17 40.39.11 40.36.20 40.32.10 41.30.18 41.44.05 41.44.05 41.55.40 42.12.36 42.35.43 42.47.50 43.07.36 43.18.20 43.14.53 42.49.07 42.49.07 40.43.54 40.43.39 40.50.37 41.02.34 41.18.51 Lonai tude 74.14.21 74.16.52 74.15.34 74.15.00 74.13.00 74.12.08 74.10.42 74.08.47 74.03.43 74.02.45 74.01.35 74.59.21 73.56.15 73.56.15 73.57.44 73.51.12 73.45.43 73.40.33 73.35.16 73.36.58 73.49.49 73.56.59 73.56.59 74.07.04 74.05.57 73.58.03 73.53.48 73.59.08 Nearest Point, River Bridge, or Highway Hard Point Bent CRR of N.J. RRB at Sandy Pt. Ferry Dock Smith Creek Fresh Kills Pralls Island North of Shooter's Range - CRR of N.J. RRB, Bergen Pt. Robbins Reef Verrazano Bridge, U.S. 278 Romer Shoal Main Street City Water Treatment Plant City Water Treatment Plant N.Y.S. Hwy 30 N.Y.S. Hwy 385 Elect. Power & Light Co. N.Y.S. Hwy 32 U.S. Hwy 4 N.Y.S. Hwy 32 N.Y.S. Hwy 9N fl.Y.S. Hwy 50 N.Y.S. Hwy 32 Lincoln Hwy, U.S. Hwy 1/9 Lincoln Hwy, U.S. Hwy 1/9 N.Y.S. Hwy 505 Continental Can Dock Bear Mountain Bridge Nearest Town Tottenville, NY Perth Amboy, NO Perth Amboy, NJ Sewaren, NJ Chrome, NJ Graselli, NJ Port Elizabeth, NJ Newark, NJ Bayonne, NJ Rosebank, NY Sandy Hook, NJ Beacon, MY Poughkeepsie, NY Poughkeepsie, NY Kingston, NY Catskill , NY Glenmont, NY Waterford, NY Thomson, NY Glens Falls, NY Corinth, NY Schenectady, NY Waterford, NY Newark, NJ Jersey City, NJ Fort Lee, NJ Piermont, MY Ft. Montgomery, NY Remarks mid channel; low slack tide mid channel; low slack tide mid channel; low slack tide mid channel; low slack tide mid channel; low slack tide mid channel; low slack tide mid channel; low slack tide mid channel; low slack tide mid channel; low slack tide mid channel; low slack tide mid channel; low slack tide shore sample; low slack tide shore sample; low slack tide finished water shore sample; low slack tide shore sample; low slack tide shore sample; low slack tide mid channel; non tidal mid channel; non tidal mid channel; non tidal mid channel; non tidal shore samole; non tidal mid channel; non tidal mid channel; low slack tide mid channel; low slack tide shore sample; low slack tide shore sample; low slack tide shore sample; low slack tide ro ------- 13 ARTHUR 68. KILL RARITAN RIVER RARITAN BAY LOWER BAY MOHAWK RIVER 82 HUDSON RIVER 76 NEW YORK AREA HUDSON AREA Figure 6. Sites sampled in the Hudson River basin. ------- Table 4 Sampling Sites in the Mississippi River Basin, in Alabama, and in Texas Sample Number 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 Waterway Mississippi River " 11 11 11 11 11 " 11 Wisconsin River 11 " Buffalo Bayou Houston Ship Channel 11 11 " 11 Galveston Bay " Mississippi River " 11 11 11 11 11 11 Mobile Bay " Mobile River Black Warrior River " 11 11 Mississippi River " " " " " Illinois River Mississippi River 11 11 Station State Highway 6 Fourth St., Minneapolis Wabash St., Minneapolis St. Paul Intake St. Paul Finished Water Lock and Dam 3 Reads Landing Weaver Bottom Lower Weaver Bottom Wausau Nekoosa Bridgeport Shepard Dr. Morgan Point Lynchburg Ferry Tuckers Bayou North Shaver Road Turning Basin Pelican Island Red Fish Bay Head of Passes Port Sulphur Lul ing Lutcher New Orleans Finished Water Plaquemine St. Francisville Dauphin Island Fowl River Pt. McDuffie Island Demopolis Tuscaloosa Bankhead L.D. Atwood Ferry Vicksburg Memphis Ensley Plantation St. Louis Finished Water Lock and Dam 26 Highway 100 Jefferson Barracks Lock and Oam 14 Lock and Dam 16 Latitude 46.32.40 44.51.12 44.56.40 45.07.33 45.00.10 44.36.40 44.24.45 44.12.29 44.12.26 44.56.55 44.17.46 43.00.00 29.45.30 29.40.24 29.45.39 29.44.30 29.43.24 29.44.54 29.21.54 29.29.37 29.09.08 29.28.39 29.56.19 30.01.55 29.57.03 29.57.55 30.17.38 30.45.30 30.06.35 30.29.30 30.39.25 32.32.30 33.06.05 33.27.36 33.35.12 32.19.36 35.12.42 35.03.50 38.42.06 38.42.06 38.53.48 39.09.24 38.29.10 41.32.36 41.27.24 Longitude 93.57.09 93.00.35 93.05.19 93.16.36 93.10.50 92.36.42 92.06.47 91.47.45 91.47.43 89.37.34 89.53.97 91.03.00 95.22.36 94.58.42 95.04.25 95.11.18 95.13.12 95.17.12 94.47.46 94.51.52 89.15.06 89.41.21 90.21.40 90.41.45 90.08.17 90.07.40 91.13.59 91.23.45 38.02.11 88.01.06 88.01.55 87.49.30 87.39.12 87.21.12 87.06.48 90.53.49 90.04.18 90.10.45 90.15.00 90.15.00 90.14.36 90.41.36 90.16.28 90.24.30 91.00.00 Nearest Point, River Bridge, or Highway M.S. Hwy 6 4th Ave. and County Hwy 24 Wabash St. Talraadge Lane Roselawn U.S. Hwy 63 U.S. Hwy 61 U.S. Hwy 61 U.S. Hwy 61 W.S. Hwy 52 Above Munic. STP U.S. Hwy 18 — Main Street T.S. Hwy 134 Tidal Road N. Shaver Rd., County 526 75th Street T.S. Hwy 87 Eagle Point — M.S. Hwy 23 Ferry Crossing M.S. Hwy 18 & 44 Ferry Crossing M.S. Hwy 18 & 44 Eagle/Spruce Eagle/Spruce Ferry Crossing, M.S. Hwy 1 & 75 Ferry Crossing, M.S. Hwy 10 Middle Ground Fowl River Point McDuffie Island U.S. Hwy 43/SL.SF RRB U.S. Hwy 11 and 43 Lock and Dam Atwood Ferry Bridge U.S. Hwy 80 Loosahatchie River Arvid Power Line Crossing Chain of Rocks Br.U.S. 270 Chain of Rocks Br.U.S. 270 U.S. Hwy 67 I.S. Hwy 100 U.S. Hwy 50 I.S. Hwy 92 U.S. Hwy 80 Nearest Town Remarks Crosby, MN Inver Grove Heights, MM St. Paul , MN Fridley, MN Maplewood, MN Red Wing, MN Reeds Landing, MM Weaver, MN Weaver, MN Uausau, WI Nekoosa, WI Bridgeport, WI Houston, TX Morgan Point, TX Lynchburg, TX Deer Park, TX Pasadena Gardens, TX Magnolia Park, TX Galveston, TX San Leon, TX Venice, LA Port Sulphur, LA Lul ing, LA Lutcher, LA New Orleans, LA New Orleans, LA Plaquemine, LA St. Francisville, LA Dauphin Island, AL Fowl River, AL Mobile, AL Demopolis, AL Tuscaloosa, AL Fosters, AL Birmingham, AL Vicksburg, MS Memphis, TN Ensley Plantation, TN St. Louis, MO St. Louis, MO Alton, IL Hardin, IL Mehlville, MO Muscatine, IA Davenport, I A Midchannel Midchannel Midchannel Shore Roseville Midchannel Midchannel Midchannel Near shore Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Shore Shore Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel Shore liore Midchannel Midchannel Midchannel Midchannel Midchannel ------- 15 90 MINNEAPOLIS WISCONSIN RIVER MADISON MISSOURI RIVER, Figure 7. Sites sampled on the upper and middle Mississippi River. ------- 16 ARKANSAS RIVER MISSISSIPPI RIVER RLEANS Figure 8. Sites sampled on the lower Mississippi River. ------- GULF OF MEXICO Figure 9. Sites sampled in the Houston area. ------- 18 MULBERRY FORK/ 123) VALLEY CREEK TOMBIGBEE RIVER LOCUST FORK xFIVE MILE CREEK 'VILLAGE CREEK 'BIRMINGHAM TOMBIGBEE RIVER Figure 10. Sites sampled in Alabama. ------- Table 5 Sampling Sites in the Ohio River Basin Sample Number Waterway 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 Ohio River Tennessee River Wabash River Ohio River Ohio River Ohio River Ohio River Kanawha River Ohio River Ohio River Ohio River Ohio River Ohio River Ohio River Ohio River Monongahela River Monongahela River Allegheny River Monongahela River Monongahela River Allegheny River Beaver River Beaver River Ohio River Ohio River Ohio River Ohio River Station Joppa Calvert City New Harmony Evansville WTP Evansville WTP Cannelton LD Louisville WTP Winfield Markland UD Cincinnati WTP Cincinnati WTP Huntington WTP Huntington WTP Belleville LD Joppa Point Marion Charlerni Freeport LD S. Pittsburgh WTP S. Pittsburgh WTP Oakmont WTP Beaver Falls WTP Beaver Falls WTP South Heights E. Liverpool WTP E. Liverpool WTP Wheeling WTP Latitude 37.12,00 37.02.16 38.07.55 37.58.20 37.58.20 37.53.53 38.16.52 38.31.32 38.46.29 39.04.11 39.04.11 38.25.57 38.25.57 39.07.07 37.12.00 39.43.57 40.08.30 40.42.41 40.24.36 40.24.36 40.31.51 40.45.48 40.45.48 40.34.12 40.38.20 40.38.20 40.06.54 Longitude 88.51.00 88.31.46 87.56.25 87.34.35 87.34.35 86.42.20 85.42.08 81.54.40 84.57.52 84.25.57 84.25.57 82.25.57 82.25.57 81.44.32 88.51.00 79.54.42 79.53.35 79.34.59 79.57.15 79.57.15 79.50.12 80.18.55 80.18.55 80.13.47 80.31.15 80.31 .15 80.42.21 Nearest Point, River Bridge, or Highway Joppa Steam Plant G.R. Clark Br., K.S. Hwy 60 U.S. Hwy 460 Inland Marina Y.C. Inland Marina Y.C. Lock and Dam Falls City Boat Company Lock and Dam Lock and Dam South of U.S. Hwy 275 South of U.S. Hwy 275 Tristate Materials Corp. Tri state Materials Corp. Lock and Dam Joppa Steam Plant Upstream from Lock & Dam 8 Belle Vernon Hwy Bridge Above Lock & Dam 5 Bedes Run Bedes Run Twelve Mile Island U.S. Hwy 18 U.S. Hwy 18 Duquesne Light Co. Intake Mill Creek Mill Creek U.S. Army Base Nearest Town Joppa, IL Paducah, KY New Harmony, IN Evansville, IN Evansyille, IN Cannelton, IN Louisville, KY Winfield, WV Markland Cincinnati, OH Cincinnati, OH Huntington, WV Huntington, WV Belleville, IL Joppa, IL Point Marion, PA Charleroi, PA Freeport, PA Pittsburgh, PA Pittsburgh, PA Oakmont, PA Beaver Falls, PA Beaver Falls, PA South Heights, PA E. Liverpool , OH E. Liverpool , OH Wheeling, WV Remarks Midstream Midstream Midstream Raw water Finished water Midstream Raw water Midstream Midstream Raw water Finished water Raw water Finished water Midstream Midstream Midstream Midstream Midstream Finished water Raw water Raw water Finished water Raw water Midstream Finished water Raw water Raw water ------- WABASH RIVER 137 TENNESSEE RIVER BEAVER RIVER ALLEGHENY RIVER KANAWHA RIVER 150 MONONGAHELA RIVER ro CD Figure 11. Sites sampled in the Ohio River basin, ------- Table 6 Sampling Sites in the Great Lakes and the Tennessee River Basin Sample Number 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 Waterway St. Lawrence Seaway Black River Oswego River Genessee River Niagara River Lake Erie Fields Brook Cuyahoga River Maumee River Detroit River Detroit River St. Clair River St. Clair River Grand River Saginaw River Lake Michigan St. Mary's River St. Louis River Lake Superior Fox River Milwaukee River Indiana Harbor Canal French Broad Holston River Hiwassee River Ocoee River Chattanooga Creek Tennessee River Station Cape Vincent Dexter Oswego Rochester Fort Niagara Buffalo Ashtabula Cleveland Toledo Maple Beach Detroit Port Huron Algonac Grand Haven Bay City Cecil Bay Brush Point Duluth Beaver Bay Green Bay Milwaukee Chicago Asheville, N.C. Church Hill, Tenn. Brittsville, Tenn. Ducktown, Tenn. Chattanooga, Tenn. Paducah, Ky. Latitude 44.07.58 44.00.15 43.27.23 43.13.59 43.14.14 42.52.47 41.53.28 41.29.15 41.41.35 42.03.20 42.16.21 43.00.11 42.37.15 43.03.35 43.38.10 43.45.35 46.28.46 46.44.58 47.16.00 44.32.12 43.01.29 41.39.19 35.36.32 36.31.00 35.22.03 35.00.13 35.01.08 37.02.16 Longitude 76.20.40 76.02.39 76.30.35 77.37.06 79.03.20 78.54.45 80.47.52 81.41.11 83.28.09 83.11.35 33.06.32 82.25.06 82.31.00 86.14.36 83.50.42 84.45.00 84.26.58 92.06.02 91.16.42 88.00.21 87.54.01 87.27.34 82.34.43 82.43.22 84.54.35 84.24.22 85.19.35 88.31.46 Nearest Point, River Bridge, or Highway Ferry to Alexandria N.Y. S. Hwy 180 U.S. Hwy 104 Turning Basin, Rattlesnake Pt. St. Catheriens Boat Club Middle Reefs Crib Intake Riverside Yacht Club W. 3rd St. near U.S. 71 & 90 U.S. Hwy 65 Gage Lee Rd. & Rockwood Drive Ft. Gratiot Light Ferry to Ualpole Isl . Corps of Engineers Boatyard Corps of Engrs. Field Office One mile north of shore Point aux Pins West Gate Basin, U.S. 535 Near munic. ramp, Pellet Isl. Green Bay Yacht Club U.S. 794, Evinrude Exp. St. Dickey Place, Cty Hwy 912 USGS Gaging Station, Rm* 144 Patterson Hill Br., Rm 131.5 I.S. Hwy 58 at Chickamonga L. Rogers Bridge L & N RR Bridge Ashland Oil Terminal Nearest Town Cape Vincent, NY Dexter, NY Oswego, NY Rochester, NY Youngstown, NY Buffalo, NY Ashtabula, OH Cleveland, OH Toledo, OH Gibraltar, MI Detroit, MI Port Huron, MI Algonac, MI Grand Haven, MI Bay City, MI Mackinaw City, MI Sault Ste. Marie, Duluth, MN Beaver Bay, HI Green Bay, WI Milwaukee, WI East Chicago, IN Asheville, NC Church Hill, TN Brittsville, TN Ducktown, TN Chattanooga, TN Paducah, KY Remarks Midstream Midstream Midstream Midstream Midstream Raw water Midstream Midstream Midstream 1/6 from shore 1/6 from shore Midstream Midstream Midstream Midstream Midstream MI Midstream Midstream Near shore Midstream Midstream Midstream Midstream Midstream Midstream Midstream Midstream Midstream *Rm = Rivermile ------- Figure 12. Sites sampled in the Great Lakes and their tributaries. ------- MISSISSIPPI RIVER OHIO RIVER CLINCH RIVER CHATTANOOGA CREEK OCOEE RIVER HOLSTON RIVER WATAUGA RIVER NOLICHUCKY RIVER FRENCH BROAD RIVER ro CO Figure 13. Sites sampled in the Tennessee River basin. ------- Table 7 Sampling Sites on the West Coast Sample Number Waterway 190 Burbank Western Wash 191 Los Angeles River 192 Los Angeles River 193 Los Angeles Harbor 194 Dominguez Channel 195 Ballona Creek 196 San Pablo Strait 197 San Pablo Bay 198 Carquinez Strait 199 Carquinez Strait 200 Willamette River 201 Willamette River 202 Willamette River 203 Commencement Bay 204 Duwamish River Station Glendale South Gate Long Beach Los Angeles Carson Playa Del Rey San Pablo Point San Pablo Valona Port Costa Portland Oregon City Wheatland Ferry Taconia Seattle Latitude 34.09.39 33.57.10 33.46.02 33.45.00 33.48.22 33.58.03 37.59.04 38.01.47 38.03.38 38.02.50 45.34.28 45.21.54 45.05.06 47.14.13 47.34.00 Longitude 118.18.14 118.10.20 118.12.16 118.16.14 118.13.37 118.19.09 122.25.43 122.22.19 122.15.41 122.10.18 122.37.49 122.36.03 123.00.55 122.30.58 122.21.10 Nearest Point, River Bridge, or Highway Stanton Ave. Firestone Blvd. Ocean Blvd. Vincent Thomas Bridge Sepulveda Ave. Lincoln Blvd. San Pablo Point Point Pinole Davis Point Port Costa SPS RR Bridge Sportscraft Marina Wheatland Ferry Commercial Street Spokane Street Nearest Town Glendale, CA South Gate, CA Long Beach, CA Los Angeles, CA Carson, CA Playa Del Rey, CA San Pablo, CA San Pablo, CA Valona, CA Port Costa, CA Portland, OR Oregon City, OR Uheatland, OR Tacoma, WA Seattle, WA Remarks Midchannel Midchannel Midchannel Midchannel Midchannel Midchannel "R2"* "5"* "5"* "5"* "5"* Shore Shore Shore Midchannel ------- 25 BUR BANK WESTERN WASH 190 LOS ANGELES RIVER RIO. HONDO SANTA MONICA BAY SANTA ANNA RIVER Figure 14. Sites sampled in the Greater Los Angeles area. ------- 26 Figure 15. Sites sampled in the San Francisco Bay area, ------- 27 COLUMBIA RIVER OREGON CITY 200%. PORTLAND Figure 16. Sites sampled on the Willamette River and in the Greater Portland area. ------- 28 GREEN RIVER PAYALLUP RIVER Figure 17. Sites sampled in the Seattle-Tacoma area, ------- 29 Most of the samples from major rivers and canals were collected in midstream or at the location of greatest river depth. In some instances, shore samples were taken when the established sampling sites were located on the shore instead of in the middle of the waterway. Midstream samples were generally taken from a boat or from bridges and spillways. The sequence of sampling generally followed the flow of the river. Estuarine samples were generally taken during low slack tide, which at a given location is the time just before the outgoing tide reverses to an incoming tide. No flow occurs in the estuary at that location and time, and collected pollutants therefore generally reflect the discharge at the location. Since the low slack tide starts at the mouth of the estuary and subsequently travels upstream, samples were collected in a corresponding sequence. SAMPLE COLLECTION The waterway samples were collected with a 3.8-liter (1-gallon) glass bottle clipped into a metal frame. Depth-integrated samples were obtained by dropping the bottle from a heightof 63 centimeters above the surface and allow- ing it to fall freely through the water. When it reached the bottom, the bottle was pulled up rapidly. During ascent, the remaining air in the bottle expanded and left the bottle, preventing additional water from entering. Prior to sampling, tests were conducted to study the water inflow rate versus time. The sampling bottle was suspended at two different depths, 1.5 meters and 3.5 meters, and allowed to fill. As Figure 18 shows, the inflow rate is nearly linear with time at both depths. Further experiments were conducted to determine the rate at which such a bottle fills when lowered through the water column. It was found that releasing the bottle at the water surface and allowing it to fall freely through the water resulted in a disproportionately large amount of water entering the bottle in the first meter below the surface. Allowing the bottle to drop from a heightof 63 centimeters above the surface, however, minimized this nonlinearity, as shown in Figure 19. ------- 30 10 - 20 - 30 o 01 10 O) ,10 - 20 - 30 1.0 2.0 3.0 Water Volume in Glass Bottle (liters) Figure 18. Rate of water inflow into a sample bottle suspended at constant depths of (a) 1.5 m and (b) 3.5 m. ------- o. O) o 0> o 0 1.0 2.0 3.0 4.0 Water Volume In Glass Bottle (I) 0.5 1.0 1.5 2.0 I I O Sample Bottle Released at Water Surface A Sample Bottle Released 15cm Above Water Surface D Sample Bottle Released 32cm Above Water Surface V Sample Bottle Released 63cm Above Water Surface Figure 19. Rate of water inflow into a sample bottle released at different heights above the water surface and allowed to fall freely through the water column. CO ------- 32 The sample was divided among eight different storage containers: 1. a 3.8-liter (1-gallon) glass bottle for analysis of extractable organics 2. a 3.8-liter (1-gallon) glass bottle for a reserve sample 3. a 1.9-liter (0.5-gallon) polyethylene container for inorganic analysis 4. a 1.9-liter (0.5-gallon) polyethylene container for gross analysis 5. four 120-ml vials for analysis of volatile organics Both the sampling bottle and the storage containers were thoroughly cleaned prior to sample collection. The glass bottles were new or baked at 350° C overnight to remove any traces of organics, while the polyethylene containers were rinsed first with nitric acid and then with distilled water to remove any heavy metals which might be attached to the container wall. Immediately prior to filling, the containers were rinsed three times with portions of the sample. The sample was poured into the eight containers in rotation, each being only partially filled at each pass to insure uniform division of the sample. The sampling process was repeated until all storage bottles were filled. The samples for inorganic analysis were stabilized by acidifying them with three ml per liter of ultrapure nitric acid supplied by the U. S. Bureau of Standards. The containers were closed with teflon-lined caps, refrigerated immediately, and transported to the analytical laboratories by surface courier or air freight. GROSS ANALYSIS In addition to advanced analyses for trace organic and inorganic con- taminants, each sample was subjected toan analysis for gross pollutants for comparison with data collected previously at the same sampling sites. The para- meters examined were: 1. total chemical oxygen demand (COD) 2. turbidity ------- 33 3. conductivity 4. pH 5. color 6. oxidation reduction potential (ORP) 7. suspended solids 8. volatile suspended solids The results of these analyses have been presented in the quarterly reports. Table 8 gives the location of the results for each sample. TABLE 8 Location of Gross Analysis Data for All Samples Sample Number 1-6 7-31 32-61 62-89 90-134 135-173 174-204 Report Number PR 1 PR 2 PR 3 PR 3 PR 4 PR 5 PR 5 Table Number 2 3 4 5 3 5 6 Page 13 10 11 12 12 16 17 ------- 34 To indicate the range of water quality exhibited by the 204 water samples collected, the results of the gross analyses are summarized in Table 9. Mean values are presented only for the COD and suspended solids measure- ments because the other parameters are not linear functions of concentration and mean values would therefore have no significance. Table 9 Summary of Gross Analysis Data Parameter PH Turbidity, JTU Total COD, mg/1 Suspended Solids, mg/1 Color (absorbance at 400 nm) ORP, +mv Minimum 5.21 0.1 1.3 0 .000 100 Mean Maximum 9.80 92.5 33.0 78.5 31.1 194 .690 458 The lowest COD was encountered in the Delaware River at Torresdale and the highest at Burbank Western Wash near Glendale, California. The sus- pended solids concentration was lowest in the Hudson River at Corinth and highest in the Mississippi'River at New Orleans. Turbidity, on the other hand, was lowest in the water flowing out of Lake Superior through St. Mary's River and highest in the Houston Ship Channel near Shaver Road. Color was maximum in the Wisconsin River at Nekoosa, Wisconsin. The pH varied from 5.21 in the Hudson River at Bayonne, N.J., to 9.80 in the Saginaw River at Bay City, Michigan. The ORP was minimum in the Ohio River at the Belleville Lock and Dam and maximum at the mouth of the Schuylkill River at Philadelphia. ------- 3, SAMPLE PREPARATION E. S. K. Chian J. H. Kim Samples to be analyzed for volatile organic contaminants were prepared by a stripping procedure; those for the less-volatile organics were prepared by liquid-liquid extraction techniques. They were then forwarded to the labora- tories which performed the quantification and identification procedures described in Chapter 5. STRIPPING PROCEDURES The volatile organic compounds were stripped from the incoming samples by a technique similar to that described by Chian and Kuo (1975) and illustrated in Figure 20. The compounds were stripped from a 120-ml sample at a temperature of 60° C by passing nitrogen through the sample at a rate of 200 ml per minute. The compounds were adsorbed in a Tenax GC trap, which was then sealed in a glass tube and transmitted to the appropriate laboratories for analysis. All glassware was baked at 450° C overnight prior to use. The stripping efficiency of this method was in the range from 12 to 100 percent, depending on the initial concentrations and the physical properties of the compounds present. The stripping procedure of Bellar and Lichtenberg (1974) was used initially for this project, the sample being held at 98° C during the stripping operation. It was found, however, that at that temperature excessive moisture accumulated in the Tenax traps, interfering with later analysis by gas chromato- graphy-mass spectrometry (GC/MS). Stripping tests conducted at various temperatures indicated that the moisture content in the trap decreased with temperature. Below 65° C, however, the stripping efficiency for compounds with a relatively high boiling point and those that are less polar (such as hexane and toluene) de- creased appreciably. Recovery of polar compounds was optimum at about 65° C. The temperature ultimately selected, 60° C, represented the best compromise between these variables. Problems with background contamination from the Tenax traps and from the Carbowax 1500 GC columns provided another impetus for changing 35 ------- 36 120-ml sample in 120 ml glass bottle sealed with Teflon-lined septum and aluminum seal* Transfer to 1-liter stripping vessel Strip at 60 with 200 ml/min N2 gas for 20 minutes. Collect sample on TenaxGCtrap. [Seal Tenax GC trap in glass tube for later analysis *Kopfler e-t al. , 1976 Figure 20. Procedure for stripping volatile organic compounds, ------- 37 the stripping procedure. Initially, the volume of sample stripped was 5 ml. Increasing it to 125 ml made it possible to exceed the background contamination level by a significantly greater margin. The level of background contamination from the stripping flask, Carbowax 1500 GC column, and the Tenax traps was mea- sured at less than the 0.03 ppb level. A number of other tests of the stripping procedure were also conducted using prepared samples containing selected concentrations of 12 model compounds. Specifically, the following parameters were examined: 1. techniques for transferring the samples from the storage bottles to the stripping flask 2. reproducibility of the stripping procedure 3. variation in stripping efficiency (recovery rate) with stripping gas flow rate 4. variation in stripping efficiency with stripping time 5. variation in stripping efficiency with stripping flask size 6. variation in stripping efficiency with the concentration of compounds in the original sample 7. the effects of storing the Tenax traps for periods of 5 and 28 days in sealed glass tubes after stripping The results, presented in Progress Report Number 3, were used in establishing the stripping procedure ultimately adopted. LIQUID-LIQUID EXTRACTION The less-volatile organic compounds were extracted from the water sam- ples using the procedure shown in Figure 21. Each sample was first spiked with camphor, which served as an internal standard. The pH was adjusted to approxi- mately 12, and the sample was extracted using nanograde chloroform. The solvent ------- 38 3.8-1iter water sample in 1-gallon glass container Spike with 15 yg of camphor from 30 yl of methanol solution. Stir for 15 minutes. Adjust pH to M2 with NaOH pellets. Extract three times with 200 ml of chloroform in 100/50/50 ml portions under mechanical agitation using Teflon-coated magnetic bar. Base Extract Acid Extract 160 to 185 ml of solvent extract concentrated to 2 ml with Kuderna- Danish (K-D) evaporator. Concentrate to 0.4 ml with Micro- K-D evaporator. Adjust aqueous layer pH to with concentrated HC1. Spike with 40 yg of 2-ethyl hexanoic acid from 80 yl of methanol solution. Store in vials with Teflon-coated septum in 100/100/200 yl portions in refrigerator. Extract three times with 200 ml of chloroform in 100/50/50 ml portions. Concentrate to 5 ml with K-D evaporator. Concentrate to dryness wi th Micro-K-D evaporator followed by helium blowing, then dilute back to 0.4 ml with methylene chloride. DAM (diazomethane) treatment in methylene chloride. Fix final volume to 0.4 ml in vials for analysis. Figure 21. Procedure for extraction of the less-volatile organics. ------- TECHNICAL REPORT DATA (Please read Instructions on the reverse before completing) . REPORT NO. EPA DCO/7 2. 3. RECIPIENT'S ACCESSION-NO. 4. TITLE AND SUBTITLE Monitoring to Detect Previously Unrecognized Pollutants in Surface Waters 5. REPORT DATE July 1977 6. PERFORMING ORGANIZATION CODE 7. AUTHOR(S) 8. PERFORMING ORGANIZATION REPORT NO. B. B. Ewing, E. S. K. Chian, J. C. Cook, C. A. Evans, P. K. Hooke. and E. G. Perkins 9. PERFORMING ORGANIZATION NAME AND ADDRESS Institute for Environmental Studies 408 S. Goodwin Avenue University of Illinois at Urbana-Champaign Urbana, Illinois 61801 10. PROGRAM ELEMENT NO. 11. CONTRACT/GRANT NO. EPA 68-01-3234 12. SPONSORING AGENCY NAME AND ADDRESS United States Environmental Office of Toxic Substances Washington, D. C. 20460 Protection Agency 13. TYPE OF REPORT AND PERIOD COVERED Final 14. SPONSORING AGENCY CODE 15. SUPPLEMENTARY NOTES 16. ABSTRACT Samples of surface waters were collected from 204 sites near heavily industrialized areas across the United States. The samples were analyzed for all contaminants present at concentrations greater than one part per billion. Each water sample was preconcentrated for analysis of organics in three fractions: volatile organics by nitrogen-gas stripping and the less-volatile organics by extraction with chloroform under both basic and acidic conditions. Organic constituents were identified by gas chromatography/mass spectrometry and quantified by gas- chromatographic techniques. Inorganic constituents were determined by spark- source mass spectrometry, energy-dispersive x-ray fluorescence analysis, and instrumental neutron activation analysis. For comparison with previous data from the same sites, the samples were also analyzed for total chemical oxygen demand, turbidity, conductivity, pH, color, oxidation-reduction potential, suspended solids, and volatile suspended solids. Results of the inorganic analyses were presented in previous quarterly progress reports. Final results of the organic analyses are presented in the appendix to this report. 17. KEY WORDS AND DOCUMENT ANALYSIS DESCRIPTORS b.lDENTIFIERS/OPEN ENDED TERMS c. COSATI Field/Group Water quality Water pollution Water analysis Water chemistry Surface waters Trace contaminants Organic compounds Volatile organic compounds Industrial pollution Nationwide survey 8h 7b, 7c, 7e 18. DISTRIBUTION STATEMENT Unrestricted Available from National Technical Infor- mation Sprvirp. Springfield. VA 22151 19. SECURITY CLASS (ThisReport)' Unclassified 21. NO. OF PAGES 86 20. SECURITY CLASS (This page) Unclassified 22. PRICE EPA Form 2220-1 (9-73) ------- 39 and aqueous layers were then separated. The solvent was concentrated to 0.4 ml in evaporators, and the resulting base extract was stored in three vials, one containing 200 pi and two containing 100 yl each. The latter two vials were transferred to the appropriate laboratories for analysis. The remaining aqueous layer was then adjusted to a pH of between 2 and 3 and spiked with 2-ethyl hexanoic acid as an internal standard. It was extracted with chloroform and the solvent was concentrated to 5 ml in an evaporator. The chloroform was then exchanged with a different solvent, methylene chloride, by evaporating the sample to dryness and then diluting it to a 0.4-ml volume with the latter compound. The resulting acid extract was methylated with diazomethane. The final volume was adjusted to 0.4 ml and the sample was transmitted to the analytical laboratories in the same volumes as for the base extracts. In the process used initially for preparing the acid extracts, the chloroform was not exchanged with methylene chloride. A high level of sample contamination was observed, however. Because the level of impurities in the nanograde chloroform used was found to be less than 0.01 ppb, the contamination of the sample was attributed to impurities in the diazomethane used as a methyl - ating agent. Further tests revealed that the interfering contaminants were not present in the diazomethane but were formed as side-reaction products between the diazomethane and the chloroform solvent. This problem did not arise when methylene chloride was used as the solvent. The recovery of organic compounds was better when the sample was extracted with chloroform, however. Therefore, chloroform was used for the initial extraction but replaced with methylene chloride prior to the methylation step. SORPTIVE EXTRACTION TECHNIQUE Adsorption onto a sorptive medium was evaluated as a possible alternative to the liquid-liquid extraction process for preconcentrating the less-volatile organic compounds. XAD-4 resin (Rohm and Haas, Philadelphia, Pennsylvania) and six types of activated charcoal were evaluated for their effectiveness as sorptive media. Of these, the XAD-4 resin and Widco activated carbon (Widco Chemical Corp, N.Y.) exhibited the lowest levels of background contamination and were therefore chosen for further testing. ------- 40 Samples number 18 and 19 were used to compare the effectiveness of the sorptive technique with the liquid-liquid extraction method. These samples were processed both by the normal procedure in the laboratory and by the sorp- tive method at the sample collection site using the experimental setup shown in Figure 22. Several different methods for processing the resulting resin and carbon samples were evaluated. The analytical procedures applied to sample 18 were described in Progress Report Number 2. Based on the results, the process was modified somewhat for sample number 19, as described in Progress Report Number 4. The activated carbon was found to be better for sampling neutral organics than was the XAD-4 resin,.as indicated by the number of peaks observed. For acidic organics, the number of peaks for the two sorptive materials were comparable, but the carbon column exhibited some impurities. The data indicate that preconcentration by the use of sorptive materials in columns is superior to liquid-liquid extraction for neutral organic compounds, 0 while the two methods produce comparable results for acidic organics. ------- 41 HC1 Spi ked Sample Neu- tral Acti- vated Car- bon or Resin Col- umn Acid i fy to pH 2 Meter ing Pumping Bypass Valves •M- Acidic Column Vacuum Seal Effluent Col lector Amount of Sample Used: Spiking Compound and Amount: Flow Rate: Period of Experiment: 200 1iters 0.253 m9 °f camphor and 1.51 mg of 2-ethyl hexanoic acid dissolved in 3.5 mg CH OH per 200-liter sample ^60 ml/min 4 to 5 days Figure 22. Flow diagram of experimental setup for sorptive studies. ------- 4, INORGANIC ANALYSIS Inorganic multielemental analysis was performed by three separate and independent techniques: spark-source mass spectrometry (SSMS), instrumental neutron activation analysis (INAA), and energy-dispersive x-ray fluorescence (XRF) analysis. The results of the analyses have been presented in the five quarterly progress reports. Table 10 lists the location of the SSMS data for all samples. The SSMS technique is capable of detecting or establishing detection limits for approximately 80 elements in the type of water sample analyzed for this project. For 44 of these elements the semi quantitative determination is confirmed by either a multiply charged or second isotopic spectral line. For the remaining 36 elements the analysis is based on a single spectral line. For 22 of those 36 elements the INAA and XRF methods are able either to provide a quantitative confirmation of the SSMS estimates (for Sc, Co, Ni, Se, Sb, and Hg) or to establish a better detection limit than SSMS (for Mo, Ru, Ag, Cd, Cs, Ce, Sm, Eu, Tb, Yb, La, Hf, Ta, Re, Ir, and U). The combined use of these techniques provided two significant benefits. The number of unconfirmed anal- yses was reduced from 36 to 14, and confirmed analyses were obtained on six of the seven environmentally important elements: Pb, Tl, Cd, Hg, Se, and As. (Three were confirmed by INAA and three by XRF). Only the Be determination depended on a single analytical technique. The analyses by the three different techniques agreed within the limits of detection and experimental error for almost all elements. INAA determinations (44 elements) are quantitative within the stated limits, as are the XRF determinations. SSMS determinations are semiquantitative; the true concentration is expected to be within the range from one-third to three times the stated value. SSMS values obtained for the alkali metals and alkaline- earth metals appear to be somewhat higher than the INAA and XRF results. It can be concluded that the three complementary techniques yield a complete elemental analysis of this type of water sample. Limitations assoc- iated with one method are usually compensated by high sensitivities of the other methods. 42 ------- 43 Table 10 Directory of Inorganic Analyses Sample Number 1 - 6 7 8, 9 10, 11 12 13 14 15 - 17 18 - 31 32 - 38 39 - 41 42 43 44 - 46 47 48, 49 50 - 61 62-73 74 - 84 85, 86 87 - 89 90 - 102 103, 104 105 - 107 108 - 110 111 - 117 118 - 120 121 - 134 135 - 204 Progress SSMS PR1 PR2 PR2 PR2 PR2 PR2 PR2 PR2 PR2 PR3 PR3 PR3 PR3 PR3 PR3 PR3 PR3 PR3 PR3 PR3 PR4 PR4 PR4 PR4 PR4 PR4 PR4 PR4 PR5 Report INAA PR1 PR2 PR5f PR2 PR5f PR2 PR5 PR2 PR51" PR5 PR4 PR5 PR5 PR5 PR4 PR5 PR4 PR5 PR4 PR5 PR5 PR4 PR5 PR4 PR5 PR4 PR5 •PR4 PR5 Number XRF PR! PR2 PR2.5* PR2 PR2,5 PR2 PR2,5 PR2 PR2,5 PR3.5 PR3,4 PR3.5 PR5 PR3.5 PR3,4 PR3.5 PR3,4 PR3,5 PR3,4 PR3,5 PR3,5 PR4 PR5 PR4 PR5 PR4 PR5 PR4 PR5 *Where two progress report numbers are listed, complete XRF data are presented in both reports. ^Partial INAA results for these samples were presented in Progress Report No. 2; the complete data were presented in Progress Report No. 5 ------- 44 SPARK-SOURCE MASS SPECTRAL ANALYSIS C. A, Evans W. H. Wadlin Most of the samples analyzed by SSMS were composited,consisting of from two to five individual samples. Samples analyzed as composites were grouped according to geographic origin and expected composition. For example, composite XXIII comprised three consecutive samples which represent the upper Hudson River, and composite LXV comprised waters from three tributaries to Lake Erie. Of the eighty composites, fourteen consisted of only one component sample. Also, the first six samples were analyzed individually and were not assigned composite numbers. In addition to these 86 samples which constitute the reported SSMS results, an additional 62 samples consisting of standards, blanks, replicates, and research samples were processed during the course of the project. Procedure The samples were evaporated in the presence of a suitable matrix ma- terial and the residues formed into electrodes. A laboratory study indicated that the loss of trace elements by volatilization during the bulk evaporation process was insignificant when a silver matrix was used. A total volume of 100ml of sample per composite was nominally used. There was no advantage to using more sample, since the detection limits for most elements were limited by the appearance of organic interferences in the spectra. Thus, detection limits become lower as the sample size is increased up to the point at which the electrodes are so heavily loaded with sample that enough organics are pre- sent to appear in the longest exposures, interfering with element identification. For saltwater samples, only 25 ml could.be used because of interferences from inorganic molecular ions. The samples were doped with 50 yg of yttrium as an internal standard and evaporated nearly to dryness in Pyrex evaporating dishes with 1.0 g of matrix material at 80°C. The drying was completed in an oven at 105°C. Silver powder was used as the matrix material in preference to graphite, as it resulted in fewer matrix interferences and higher sensitivity. The residue was then transferred to a plastic ball mill for mixing. Since the dried residues ------- 45 were frequently quite hygroscopic, it was found necessary to make the transfer quickly while the material was still warm and to store it immediately in a dessicator. After mixing, the powder was pressed into electrodes in a poly- ethylene slug. The electrodes were mounted in the AEI MS-7 mass spectrometer for analysis using photographic detection. The samples were presparked for the equivalent of a 30 nC exposure at 30 pulses per second. This process removed surface contamination and the abnormally high initial sensitivity caused by the adsorption of trace elements contained in the sample on the surface of matrix particles. Although it was considered possible that some elements could be lost by thermal vaporization from the electrodes during the sparking process, tests conducted using a sample containing model elements indicated that selective volatilization was not a problem. In obtaining the analytical exposures the pulse repetition rate was kept as low as was practical for obtaining the exposure in a reasonable amount of time. The maximum pulse repetition rate used was 100 pulses per second regardless of the time required to obtain the exposure. Higher rates caused sufficient heating of the electrode bulk to drive organic materials out from the interior, giving rise to severe interferences and unacceptably high detec- tion limits. Three exposures per decade of exposure magnitude were obtained over the range from .001 to 300 nC. Concentrations of trace elements were determined by comparing the exposures required for their spectral lines to be just detectable with the equivalent exposure for the internal standard. Elemental sensitivities were assumed to be equal. The results were presented in the appendices of the five quarterly progress reports. If an element was confirmed by the pre- sence of either a multiply charged ion or multiple isotopes of the correct relative intensities, the value for that element was given without a prefix symbol. If the value was prefixed by the symbol "<" (meaning "less than or equal to") it indicates that a line corresponding to the +1 ion of that element was observed, but its presence could not be confirmed by the procedures above. ------- 46 That is, either the element was present at the concentration stated or there was an interference. The symbol "*<" (meaning "definitely less than") was applied if no lines were observed which could be attributed to this element or if there was a definite, known interference. In cases where no line was observed, the number given was calculated on the basis of what the concentration would be had the element been just detectable in the longest exposure. In the case of a known interference, the value given was the concentration at which the element would have had to be present to appear with an intensity equal to that observed for the interference. Discussion of Results Considering that the data were taken with three exposures per decade and that there are some differences in elemental sensitivity, the results reported should generally be accurate to within one-third to three times the actual concentration. Detection limits are generally in the range from 0.2 to 2 yg/1 for freshwater samples. For saltwater samples, inorganic molecular interferences and a general loss of sensitivity raise the detection limits to the range from 20 to 200 yg/1. Comparison of SSMS results with the INAA and XRF results for these samples shows general agreement within the expected range of errors given above. Notable exceptions are the concentrations determined for K, Mg, and Ca. The SSMS results for K and Mg are consistently higher by about a factor of 3 and for Ca are higher by a factor of 5 to 10. Analysis of standards indicates that the error is in the SSMS determination and results from unusually large differ- ences in sensitivity for these elements, a factor not compensated for in the calculations. INSTRUMENTAL NEUTRON ACTIVATION ANALYSIS P. K. Hopke J. D. Sherwood Procedure The sensitivity of instrumental neutron activation analysis (INAA) is governed by the number of neutrons with which a sample is irradiated, the delay time between irradiation and the start of the count, and the counting time ------- 47 intervals. Other investigators (Clemente and Mastinu, 1974; Salbu et al., 1975) have used various methods of applying INAA to obtain determinations of trace elements in water samples. In some studies where a small number of samples were processed, long irradiation intervals (up to 72 hours) were used. For the large number of samples involved in the present project, however, long intervals become economically unfeasible because of reactor costs. Therefore, a relatively short interval was used, compensated for by using larger sample volumes, counting the irradiated samples for comparatively longer intervals, and using a more efficient gamma counting system. The following procedures were used for freshwater and saltwater samples through number 94. Freshwater Samples 1. The quartz ampoules used to hold the samples during irradiation were pretreated in a bath of dilute nitric acid at 85°C for four hours to remove any contaminants 2. A 15-ml sample was loaded into each ampoule, which was then heat-sealed 3. The sealed ampoules were tested to insure that they would not break from thermal stress during irradiation 4. The samples were irradiated for 30 minutes at 500 kW 5. After a decay period of three days, the samples were transferred to standard polyethylene counting vials. The ampoules were rinsed with two 1-ml portions of dilute hydrochloric acid to recover any adsorbed molecular or ionic species and the rinse was added to the contents of the counting vial 6. The emitted radiation was counted for 1,000 seconds using a 10% Ge(Li) detector in conjunction with a 4096-channel analyzer system 7. The samples were repackaged and irradiated for eight hours at 1.5 MW 8. After a decay period of 14 days, the samples were transferred to counting vials as before and the emitted radiation was counted for 4,000 seconds ------- 48 Saltwater Samples 1. The ampoules were pretreated as for freshwater samples 2. Each ampoule was loaded with a 10-ml sample and heat-sealed 3. The ampoules were tested for thermal stress resistance 4. The samples were irradiated for eight hours at 1.5 MW 5. After a decay period of seven to ten days the samples and acid rinse were transferred to polyethylene counting vials 6. The emitted radiation was counted for 1,000 seconds 7. After an additional 7-day decay period, the sample radiation was counted for 20,000 seconds Data obtained from the multichannel analyzer system were transferred to magnetic tape for computerized peak analysis. Two computer programs were written which combine peak identification and quantitative calculations in a single computer run. The results have been presented in the five quarterly reports. Table 10 lists the location of INAA data for all samples. A preconcentration procedure was adopted for freshwater samples above number 94 in order to achieve a higher rate of sample analysis. Pre- concentration permits the use of larger sample volumes with a proportionate decrease in the required counting time. The samples were preconcentrated by evaporation in the presence of AVICEL, a microcrystalline cellulose, which acts as an adsorption medium for the inorganic constituents. A 500-ml sample is evaporated with 100 mg of AVICEL at 50°C in a polyethylene beaker and quanti- tatively transferred to an irradiation container. In addition to faster sample analysis, this technique provides lower detection limits and reduces the hazards of handling the samples after irradiation. The disadvantages are the increased possibilities of contamination and errors resulting from loss of sample during the transfer. ------- 49 ENERGY-DISPERSIVE X-RAY FLUORESCENCE ANALYSIS P. K. Hopke J. D. Sherwood Samples analyzed by energy-dispersive x-ray fluorescence were precon- centrated by three methods: 1. Precipitation with ammoniurn-1-pyrrolidine dithiocarbomate (APDC) at a pH of 4 and filtration through a 25-mm 0.2 y-pore Nuclepore filter 2. Precipitation of cyanide complexes at a pH of 12.0 followed by filtration twice through Reeve Angel SB-2 anion-exchange paper 3. Filtration twice through Reeve Angel SA-2 cation-exchange filter paper at a pH of 2.0 Each sample was divided into two aliquots, each of which was analyzed separately. For samples through number 89, one sample was preconcentrated by the first (APDC) method while the other portion was preconcentrated by one of the other two methods. For later samples, both portions were preconcentrated by the APDC technique, since it was found to yield the most consistent recovery rate for the elements analyzed in both freshwater and saltwater samples. The basic procedure used for analyzing samples by the XRF method was as follows: 1. A 1% w/v APDC solution was prepared daily and filtered through a 0.2 y Nuclepore filter. 2. The water sample was shaken sufficiently to resuspend particulate matter. 3. Two 50-ml aliquots were removed and the pH was adjusted to 4.0. 4. 5 ml of APDC solution was added and the precipitation process was allowed to develop for approximately 15 minutes.. ------- 50 5. The precipitate was filtered through a 25-mm, 0.2 y Nuclepore filter. 6. After the filter dried, it was mounted between 0.00010-in. Mylar film on a polyethylene XRF sample cup. 7. The sample was positioned under the source exciter system. 4 8. Each sample was counted for 10 seconds for each secondary target (Mo and Dy). 9. The accumulated spectrum was transferred to magnetic tape for processing. 10. Treating the filter as a thin sample, the results were calculated by a method identical to that used by Bonner, Bazan, and Camp (1975), except that all of the material was assumed to be on the top of the filter. With the Nuclepore filters used in this study it is believed that there was very little penetration of precipitate into the filter. To determine the net area of each peak in the spectra produced, it was necessary to subtract the background level from the region of interest. For the early samples through number 89, a background-stripping computer pro- gram similar to that of Bonner, Bazan, and Camp (1975) was used. Discrepancies between the results of XRF analyses and those obtained by the INAA and SSMS techniques were traced to problems with this program. Consequently, for samples after number 89, the background was subtracted manually, resulting in much greater consistency between the three techniques. The results of the XRF analyses have been presented in the quarterly reports. The location of the XRF data for all samples is listed in Table 10. ------- 5, ORGANIC ANALYSIS Samples prepared for the determination of organic constituents were divided and sent to two separate laboratories for analysis. For ex- tractable organics, one 100-pl aliquot was sent to the gas chromatography- mass spectrometry (GC/MS) laboratory in the Department of Chemistry for the identification of constituents, and another 100-yl aliquot was analyzed by gas chromatography (GC) in the Department of Food Science to establish the quantity of each contaminant present. Similarly, separate Tenax traps were prepared and sent to the two laboratories for the analysis of volatile organics. Identical gas chromatographic operating parameters were used in the two labora- tories, making it possible to use the relative retention times for each compound to correlate the identification and quantification of each peak observed. In some samples a comparatively large number of compounds were detected at levels below 1 ppb. Because identifying and quantifying all of these com- pounds would have been excessively time consuming, it was determined in consul- tation with the sponsor that only those organic substances present at or above the 1 ppb level would be reported. Preliminary results of the organic analyses were presented in the quarterly reports. Upon completion of sample analysis, the data were reexamined and corrected or adjusted as necessary. The final results are presented in the appendix of this report. In particular, the quantisations were revised on the basis of the relative GC response factors determined for selected compounds, with the result that some compounds formerly determined to be present at concen- trations below 1 ppb are now known to be at a higher level and are therefore in- cluded in the updated listings. Conversely, some compounds previously listed have been found to be present at concentrations less than 1 ppb and have hence been deleted. In addition, the relative retention times have been rechecked, resulting in corrections to some of the identifications. Compounds which could not be identified were not listed in the previous reports, but their presence is indicated in the final results presented here. 51 ------- 52 IDENTIFICATION OF ORGANICS J. C. Cook, Jr. R. M. Milberg All of the samples processed by the organic identification laboratory were analyzed on a Varian-MAT 311A combined gas chromatagraph-mass spectrometer with a Varian-Aerograph 2700 gas chromatagraph using a 2-stage Watson-Biemann sample enricher. The system is of all-glass construction from column inlet to source inlet. Procedures The Tenax traps for analysis of volatile organic compounds were received from the sample preparation laboratory in sealed glass tubes. The traps were removed from the tubes, connected to the gas chromatagraph, and flushed for two minutes at ambient temperature with helium gas. A tubular furnace with an in- terior temperature of 250°C was then slipped over the trap and allowed to pre- heat the trap for two minutes, after which the volatiles were flushed from the trap onto the GC column for four minutes with a helium flow rate of 40 ml per minute. The column was held at ambient temperature during this period. The outside diameter of the glass column used was 6 mm, the inside diameter was 2 mm, and the length was 12 ft. The column was packed with 0.2% Carbowax 1500 on 60/80-mesh Carbopack C (Supelco, Inc., Bellefonte, Pa.). After the volatiles were flushed onto the column, the temperature of the column was programmed to increase at a rate of 8°C per minute from 30°C to 200°C. It was held at the latter temperature for the remainder of the run. For the analysis of base and acid extracts, 1 yl of the sample was in- jected onto a glass column of the same dimensions as above but packed with 3% OV-17 on 80/100-mesh Gas Chrom-Q (Applied Science, State College, Pa.). The column temperature was held at 50°C during injection and was then programmed to increase to 300°C at a rate of 8°C per minute. The mass spectrometer was scanned continuously from an m/e of 33 to an m/e of 350 for volatiles and from m/e 33 to m/e 600 for extractable organics. The scan rate was 2.3 seconds per mass decade, and the ionization potential was ------- 53 70 eV. Data were acquired on a Varian-MAT SSI00 data system and stored on a disk cartridge. Identification Mass spectra were identified by inspection, manual searches, and computer searches using the ADP-Cyphernetics Mass Spectral Search System (MSSS). The usefulness of the MSSS was limited by the fact that the spectra for many compounds found in the samples were not in the computer file, nor were there any compounds of similar type. Also, spectra for most of the compounds in the file were obtained by direct-probe mass spectrometry under ideal conditions with the result that many of the spectra were different from those obtained by the GC/MS technique at the 1 ppb level. The volatile compounds were the easiest to identify because of their low molecular weights and simple spectra and because many of them were halogen- ated, giving excellent isotope cluster patterns. Compounds in the acid extracts were the most difficult to identify because of the large number of peaks and the fact that the methyl ester spectra for many of the compounds were not present in the MSSS file. A computer program was developed to calculate the relative retention times of the observed peaks. These relative retention times provided a second confirmation of the identifications and, as discussed above, were used to correlate the identifications of the compounds with their quantisations. Results . All compounds identified in the samples were listed along with their relative retention times in the quarterly reports. The final adjusted data for those compounds present at concentrations of 1 ppb or greater are presented in appendix B of this report. ------- 54 QUANTITATION OF ORGANICS E. G. Perkins J. C. Means The work of the quantitation laboratory focused on several areas: (1) quantitative analysis of the stripped volatile, acid extractable, and base extractable organic compounds contained in the 204 surface water samples col- lected, (2) the optimization of the chromatographic conditions used to separate the organic constituents of each sample fraction, (3) the investigation of con- ditions affecting the purity of blanks, (4) the investigation of parameters related to the selection of internal standards for the acid and base extract- able fractions, and (5) the investigation of parameters related to the sepa- ration and quantition of selected amines. Quantitative Analysis of Purgeable and Extractable Organics As mentioned previously, the gas chromatographic conditions used in the quantitation laboratory and in the identification laboratory were coordi- nated during all stages of the project. Initially, base and acid extract samples were run routinely on a Hewlett-Packard 5830A programmable gas chroma- tograph. One-microliter samples were injected onto a 12-foot by 1/4-inch (2 mm ID) all-glass column packed with 3% OV-17 on Gas-Chrom Q (60-80 mesh). Other pertinent instrument conditions are given in Table 11. These conditions corresponded exactly to those used by the mass spectrometry laboratory and the sample preparation team, making it possible to compare relative retention times. Peaks exceeding the 1 ppb level in water ('v-lO ng/yl in extracts) were identified by peak integration values. Once these peaks were identified, quantitation was accomplished by converting the integration units to ng/yl and then calculating the concentration in the original water samples on the basis of the known extraction efficiencies in the sample preparation step. ------- VENT TO ATMOSPHERE TUBE FURNACE CONTAINING TENAX TRAP AUX GAS CYLINDER GAS CHROMATOGRAPH CARRIER GAS CYLINDER 01 01 Figure 23. Schematic diagram of volatile organic elution system. ------- 56 Table 11 Gas Chromatograph Column Conditions* Initial temperature 35° C Initial time 5.0 min. Programming rate 10° C/min Final temperature 300° C Injector temperature 275° C Detector temperature 350° C Carrier gas flow 40 ml/min *12 ft. x 1/4 in. (2 mm ID) all-glass column packed with 3% OV-17 on Gas-Chrom Q (60-80 mesh). After two months, all quantisations were performed on a new Hewlett- Packard-Model 5711 gas Chromatograph equipped with a Model 3380A reporting integrator. This instrument was used routinely to quantitate the balance of the samples collected. The procedure was modified slightly to make the best use of this new instrument and to shorten analysis times. The initial five- minute isothermal hold at 35° C was reduced to four minutes and the initial temperature was increased to 50° C. These changes reduced the tailing of the solvent peak and generally improved the characteristics of the total chroma- togram. The temperature programming rate was decreased from 10° C/min to 8° C/min to improve resolution of peak clusters in the chromatograms. All other conditions remained unchanged. Volatile samples were chromatographed on a 12-foot, all-glass column (2 mm ID) packed with 0.4% Carbowax 1500 on Carbopack C (Supelco, Inc., Bellefonte, Pa.). This column was determined to have resolution, capacity, and thermal stability (bleed) characteristics superior to the other column packings traditionally used for volatiles (e.g., Porapacks, Chromasorbs, etc.). For quantitation purposes, the Tenax trap was heated to 250° C and stripped for six minutes with a nitrogen carrier gas flow of 40 ml/min onto the Carbowax column held at 30° C (Figure 23). After purging, the trap was isolated from the column by changing the position of the valve. The column temperature was programmed to increase from 30° C to 200° C at 8° C/min and then held at 200° C for an addi- tional 6 minutes. The injector and detector temperatures were maintained at 250° C. ------- 57 Because of early problems encountered in the other laboratories with contamination from the gas valving system, the possibility of using a high- temperature valve when analyzing the stripped samples was investigated. Initially, the traps were heated to 250° C while the valve remained at 30 to 70° C. This temperature difference affected the purity of the stripping blanks and could have been a potential source of cross-contamination of vola- tiles from one run to the next. A new Valvco two-position, six-port valve which can be heated to 200° C was therefore substituted, making it possible to heat both the Tenax GC trap and the valve during analysis. The new valve had other advantages as well in that the carrier flow to the GC column was not interrupted and the column was never exposed to the atmosphere. Also, traps could be swept with carrier gas before connecting them to the GC column, again eliminating exposure of the column to the atmosphere. Finally, the valve material was stable at high temperatures, which helped to eliminate some of the peaks in the stripping blanks believed to originate in the Teflon gaskets of the original valves used. The only difference between the procedures used by the quantisation and identification laboratories was in the method of heating the traps. In the quantisation laboratory, a combustion-tube furnace comrnercially available from A. H. Thomas was used rather than a hand-made tube heater. Evaluation of the tube furnace showed that it had several advantages: (1) the dimensions of the furnace accommodated the entire length of the trap in the heated zone; (2) the temperature of the furnace at a given potentiometer setting was highly reproducible; (3) the temperature variation along the length of the heated zone was negligible; (4) traps could be inserted and removed for cooling in seconds; and (5) the internal temperature of room-temperature traps inserted into the furnace rose to 250° C in two minutes or less. Increasing the quantity of water sample stripped from 5 ml to 125 ml also simplified the analysis of the volatile samples by eliminating many of the significant contamination problems encountered in the early stages of this project. None of the fluorinated hydrocarbon species believed to have come ------- 58 from the sampling valve were identified in any of the samples at the equiv- alent 1 ppb level or above after increasing the volume of sample stripped. The quality of the system blanks, which include bleed peaks from the Tenax GC trap matrix, was also improved. Figure 24 is a reproduction of a typical system blank. All of the peaks observed in the blank are well below the 1 ppb level. The first two peaks are methanol and ethanol. These substances are observed in every volatile sample by gas chromatography but are not seen in the GC/MS runs because they are vented out with the water. Peaks 3 through 5 are acetone, benzene, and toluene, respectively. Peak 6 is due to column bleed. Quantisation of the volatile organics was accomplished by determining individual relative response factors for each of the commonly occurring vola- tile substances observed in the samples collected. The estimated concentrations of these compounds were corrected for these factors prior to the preparation of this report. Selection of Internal Standards for the Acid- and Base-Extractable Fractions To increase the accuracy of the quantitative data obtained on the acid- and base-extractable organics and to maintain consistent control of laboratory extraction procedures, internal standards were sought for both the acid and base extraction steps. The criteria used in selection were (1) gas chromato- graphic retention time, (2) purity, (3) mass spectral characteristics, (4) extraction efficiency at pH ^12 or pH^2 with chloroform, (5) chemical and bio- logical stability, and (6) occurrence in surface water samples. Since it was known that hydrocarbon-type materials extract readily at basic pH, the cyclic hydrocarbon camphor was selected for the base extraction step. Likewise, fatty-acid-type compounds were known to be characteristic of the pH 2 extractables. Therefore, a branched-chain Cg fatty acid (2-ethyl hexanoic acid) was selected for the acid internal standard. Both camphor and 2-ethyl hexanoic acid extraction efficiencies were determined on each group of samples processed. The average extraction efficiencies observed were 67% and 69%, respectively. These factors were used in correcting the quantitative data for the corresponding fractions. The 2-ethyl hexanoic acid served the added function of providing a check on the efficiency of the methylation step as well. ------- CJ1 ID Figure 24. Gas-chromatographic trace of a representative system blank. ------- 60 Since both camphor and napthalene, another compound considered as an internal standard, had previously been reported as constituents of certain surface water samples, an investigation was begun to find an alternative com- pound which would not be expected in nature and which would meet the criteria listed above. Several brominated compounds were evaluated: Bromobenzene Bromocyclohexane Bromoheptane Bromopentane Bromononane Bromodecane Bromotetradecane p-Bromo-Anisole a-Bromo-Toluene a-Bromo-p-Xylene A number of the compounds were determined to be unsatisfactory based on the selection criteria above. The last three compounds in the list all had retention times in a good range, were of high purity, gave distinctive mass spectra, and had extraction efficiencies in a satisfactory range. When these compounds were tested for chemical stability in chloroform and surface water, however, the bromo compounds decomposed or reacted with compounds in the water. Other brominated compounds were considered, but none was found which met the criteria. Investigation of Gas Chromatographic Methods for Monitoring Amines Pure samples of six selected amines were obtained for analysis: Aniline Benzidine 3-Naphthylamine o-Tolidine o-Toluidine Phenyl Hydrazine The initial work was directed toward determining whether the ex- traction and separation techniques being used for this project would be able to detect these amines if they were present in any of the water samples. A liter of ultrapure water was spiked with a pure amine compound at a level of 100 ppm, 1 ppm, or 50 ppb. Samples were prepared in triplicate. Once the amine was dissolved completely, the pH was adjusted to between 11 ------- 61 and 12. The sample was then extracted with a total of 200 ml of chloroform in three portions (100, 50, and 50 ml). The amount of the amine recovered in the chloroform was then determined either by direct weighing of the residue after removal of the solvent or by quantisation of the amine by gas chromatography. The extraction efficiencies determined are reported in Table 12. Table 12 Extraction Efficiencies of Selected Amines Ami ne Aniline Benzidine 3-Naphthyl amine 0-Tolidine Phenyl Hydrazine o-Toluidine 100 ppm 85.3 99.9 47.7 82.2 73.2 87.5 % Recovery* 1 ppm 83.7 99.9 46.8 83.1 75.1 87.2 50 ppb 84.0 99.8 46.7 82.7 74.8 86.8 *Average of triplicate determinations. Next, the relative retention times of these six amines were determined against the retention time of camphor using the gas chromatographic conditions routinely used for this project. Relative response factors for each amine (integrator counts/ng of amine divided by integrator counts/ng of camphor) were also determined using solutions of known concentration on the OV-17 column used for this project. The values are reported in Table 13. Five other gas chromatographic column packings for the analysis of amines were evaluated using the six selected amines. Chromasorb 103, 4% Carbo- wax 20M +0.8% KOH on Carbopack B, 4% Carbowax 20M + 1% Polypropyleneimine on Carbopack B, and 0.1% SP-1000 on Carbopack C columns were tested. In each case, the retention times of all of the amines being studied were too long or the compounds were not eluted at all. Relative retention times of the six amines selected for study were determined against the retention time of camphor using 10% Apiezon L/2% KOH on 80/100 mesh Chromasorb WAW. Relative response factors were also determined using solutions of known concentration. The results are presented in Table 14. ------- 62 Table 13 Relative Retention Times and Relative Response Factors For Selected Amines on 3% OV-17 Amine Aniline Benzidine B-Naphthylamine o-Tolidine Phenyl Hydrazine o-Toluidine Relative Retention Time1 0.77 2.48 1.72 2.65 1.38 0.93 Relative Response Factor2 0.96 0.75 1.45 1.39 0.36 0.96 2 Calculated relative to the retention time of camphor (10.84 min). Calculated relative to the response per nanogram of camphor (1650/ng), Table 14 Relative Retention Times and Relative Response Factors For Selected Amines on 10% Apiezon 1/2% KOH Amine Aniline Benzidine 3-Naphthylamine o-Tolidine o-Toluidine Phenyl Hydrazine Relative Retention Time1 0.746 2.425 1.638 3.180 0.919 1.500 Relative Response Factor2 0.695 0.693 0.655 0.715 0.761 0.0143 Calculated relative to the retention time of camphor (14.19 min). Calculated relative to the response per nanogram of camphor (3409/ng) 3Coloration of sample and multiple peaks in chromatogram suggest some chemical and/or thermal degradation. ------- 63 Both the OV-17 column and the Apiezon L/KOH column gave very satis- factory separation of the amines selected. The response per nanogram of material injected on column, however, was significantly higher using the deactivated Apiezon packing, indicating that some of the amine material was adsorbed to the OV-17 column packing. In survey studies such as the one just completed, a good approach may be to use the deactivated packings for quanti- tation and the OV-17 packing for mass spectrometry, since the bleed character- istics of the latter are more favorable than those of Apiezon L. A major effort was directed toward evaluating the potential of a nitrogen-specific flame ionization detector (NFID) for amine analysis. A prototype conversion kit NFID detector was installed in the Hewlett-Packard 5710 gas chromatograph. The detector specifications state that a response discrimination factor of 5000 to 1 for nitrogen-containing vs. nonnitrogen- containing compounds can be achieved. This selectivity makes the NFID detector system ideal for screening water sample extracts (which may contain hundreds of organic compounds) for those that contain nitrogen. In practice, the NFID device was variable in response and had to be tuned and checked with standards frequently. When the device was operating properly, however, 0.1 ng of an amine (aniline) could be detected reproducibly at a relatively insensitive attenuation. New devices which are easier to maintain and which are more sensitive are avail- able and should be evaluated. Studies of the recovery of selected amines on ion exchange resins using purified XAD-4 and Biorex 70 resins were initiated. The selective recovery of amines on ion-exchange resins and subsequent elution showed some promise as a technique for the analysis of amines as an alternative to ex- traction. Limited access to liquid chromatographic facilities, however, prevented any detailed studies in this area. SUMMARY OF THE ORGANIC COMPOUNDS FOUND J. C. Means A comprehensive listing of all of the organic compounds identified in the 204 samples collected and analyzed during the project is presented in Tables 15 through 17. ------- 64 In the acid-extractable fraction (Table 15) 110 compounds were identified. These compounds generally fell into the classes of alcohols, fatty acid methyl esters, phthalate esters, polycyclic and polyunsaturated hydrocarbons, hydrocarbons, substituted phenolics, and halogenated hydro- carbons. The compounds appearing the most frequently in the acid extracts were: methyl palmitate (183), methyl stearate (165), diethyl hexyl phthalate (132), C,r terpineol (56), and methyl myristate (47). Of these compounds, only the phthalate ester is a synthetic organic compound. The others are believed to be products of the decay of natural materials. The majority of the com- pounds identified, however, were of synthetic origin and many have been iden- tified as toxic or carcinogenic. In the base extractable fraction (Table 16), 89 compounds were iden- tified. These compounds fell into the general classes of: phthalate esters, hydrocarbons, halogenated hydrocarbons, and polycyclic and polyunsaturated hydrocarbons. The compounds appearing the most frequently in the base ex- tracts were: diethyl hexyl phthalate (132), dibutyl phthalate (84), C15 terpineol (55) and C-.Q terpineol (36). Of these, the two phthalate esters are widely used synthetic organics while the terpineols are believed to be natural products. Many of the compounds identified in the base extracts were of synthetic origin and many of these have been identified as toxic. Eighty-one purgeable organic compounds were identified in the 204 samples collected (Table 17). The majority of the compounds were halogenated hydrocarbons from C-, to Cg. The compounds appearing the most frequently in the purgeable fraction were: chloroform (178), trichloroethylene (88), tetra- chloroethylene (77), 1,2 dichloroethane (53), benzene (40), acetone (33), dichloromethane (32), toluene (31), and bromo-dichloromethane (24). It is significant that chloroform and many of the other chlorinated and brominated hydrocarbons appeared in almost every sample but at levels below 1 ppb. These compounds are now suspected carcinogens or are known to be toxic. Their wide- spread occurrence in surface waters emphasizes the need for further study of the origin and impact of these highly mobile substances in the environment. ------- 65 Table 15 List of Acid-Extractable Compounds Found in 204 Water Samples and Their Frequency of Occurrence Compound Name Cc Alcohol 0 Cj Alcohol Cg Alcohol Cg Alcohol C,Q Alcohol C-j-j Alcohol C12 Alcohol C, ~ Alcohol C14 Alcohol C,5 Alcohol C16 Alcohol C-iy Alcohol C. Alcohol '19 Alcohol C2Q Alcohol C2, Alcohol C22 Alcohol C23 Alcohol Alcohol Alcohol Alcohol 74 "26 Alkyl Benzene (C-,gH, 2) Benzoanthrene Butyl benzyl Phthalate Butyl Phthalyl Butyl Glycolate Caffeine CvHvC17 ' x - 4, z - 5 Isomers A y L CgH,,- Isomer C10H10 Isomer C15H24 Frequency .1 2 3 5 6 7 7 8 9 9 9 9 9 6 7 6 6 4 3 1 1 3 1 3 23 1 12 1 1 1 ------- 66 Table 15, cont. Compound Name Frequency C, ,Hin Isomer 1 ID I u ci6Hio (pyrene) 3 ^21^2^2 Methyl Pimarate Isomer 8 Dibutyl Phthalate 15 Di chlorinated Hydrocarbon C - 5 2 Dichlorobutane 11 Dichloroheptane 1 Di ethyl Hexyl Phthalate 132 Diisobutyl Phthalate 2 Dioctyl Adi pate 6 Dioctyl Phthalate 1 Diphenyl Dulfone 1 Fatty Acid Methyl Ester C - 10 1 Fatty Acid Methyl Ester C - 12 7 Fatty Acid Methyl Ester C - 13 1 Fatty Acid Methyl Ester C = 14 3 Fatty Acid Methyl Ester C - 14 15 Fatty Acid Methyl Ester C - 15 7 Fatty Acid Methyl Ester C - 16 11 Fatty Acid Methyl Ester C - 17 2 Fatty Acid Methyl Ester C - 18 5 Fatty Acid Methyl Ester C - 19 2 Fatty Acid Methyl Ester C - 20 14 Fatty Acid Methyl Ester C - 21 1 Fatty Acid -Methyl Ester C - 22 18 Fatty Acid Methyl Ester C - 23 1 Fatty Acid Methyl Ester C - 24 11 Fatty Acid Methyl Ester C - 26 3 Hexachlorobenzene 2 Hexachlorobutadiene 2 Hexachlorobutene 2 Hexachloroethane 1 Hexachloro-hexafluoropentane 1 ------- 67 Table 15, cont. Compound Name Frequency Hydrocarbon C - 8 3 Hydrocarbon C - 10 4 Hydrocarbon > 12 3 Hydrocarbon C - 14 4 C, 5 Hydrocarbon 2 Hydrocarbon C - 16 7 C-|g Hydrocarbon 2 C, 7 Hydrocarbon 2 Hydrocarbon C - 18 11 Hydrocarbon C - 20 7 Hydrocarbon C - 22 10 Hydrocarbon C - 24 4 Hydrocarbon C - 26 5 Hydrocarbon C - 28 4 Hydrocarbon C - 30 5 Methoxy Carbonyl Benzophenone 1 Methyl Arachidate 2 Methyl-2(4-chlorophenoxy) Butanoate 13 Methyl Dehydroabietate 11 Methyl Dichlorophenoxy Acetate 4 Methyl Dichlorophenyl Ether 1 Methyl-2, 2-Dichloro-3-Methyl Butanoate 1 Methyl Laurate 7 Methyl Myristate 47 Methyl Naphthoate 1 Methyl Palmitate 183 Methyl Pentachlorophenyl Ether 12 Methyl Pentachlorophenoxy Ether 2 Methyl Stearate 165 Methyl Tetrachlorophenyl Ether 1 2-Methyl Thiobenzothiazole 1 Methyl-Trichlorophenoxy Acetate 1 Methyl Trichlorophenyl Ether 5 Pentachloroanisole 10 ------- 68 Table 15 , cont. Compound Name Frequency Pentachlorobutadiene 2 Pentachlorobutene 3 Pentachloroethane 3 Tetrachlorobutadienne 3 C,5 Terpene 37 CIQ Terpineol 27 C-|5 Terpineol 56 Tetrachloroanisole 4 Trichloroanisole 3 Trichloroheptane 1 Trichlorohexane 3 Trichloropentane 4 Unidentified Phthalate 2 ------- 69 Table 16 List of Base-Extractable Compounds Found in 204 Water Samples and Their Frequency of Occurrence Compound Name Frequency Alkyl Acid Ester (R + R1 - 8) 3 Alkyl Phenyl Ether 1 Anthracene 1 Atrazine 5 Benzothiazole 2 Biphenyl 1 Bromopropyl Benzene 1 Butyl benzyl Phthalate 2 Butyl Phthalyl Butyl Glycolate 17 Caffeine 9 C-|Q Camphenol 2 CXH Clz, x - 4, z - 5 (series of isomers) 5 C5H10C12 1 CgH-|6 Isomer 1 ci6Hio (pyrene) 7 C18H12 3 C22 H14 ] Chloro-Nitrobenzene 3 Chloroprene Dimer 1 Chloroprene 1 Dibromo-chloroethane 1 Dibromoethane 1 Di butyl Nonanedioate 1 Dibutyl Phthalate 84 Dichlorobenzene 9 Dichlorobutane 19 Dicyclohexyl Phthalate 1 Diethoxyethane 4 N,N-Di ethyl aniline 1 ------- 70 Table 16, cont. Compound Name Di ethyl Hexyl Phthalate Diethyl Phthalate Diisobutyl Nonanedioate Diisobutyl Phthalate Dimethyl Biphenyl Dimethyl Naphthalene Dimethyl Styrene Dinitrotoluene Dioctyl Adi pate Diphenyl Benzene Diphenyl Ethane Diterpene C,gH,g Hexachlorobenzene Hexachlorobutadiene Hexachlorobutene Hexachloroethane Hydrocarbon C-8 Hydrocarbon C ^ 10 Hydrocarbon C - 12 Hydrocarbon C - 14 C-.,- Hydrocarbon C,g Hydrocarbon Hydrocarbon C - 16 Cig Hydrocarbon Hydrocarbon C - 18 Hydrocarbon C - 20 Hydrocarbon Hydrocarbon Hydrocarbon Hydrocarbon Hydrocarbon Hydrocarbon Hydrocarbon C - 22 C - 24 C - 26 C - 28 C - 29 C - 30 C - 31 Frequency 132 6 1 5 1 2 1 2 2 1 1 4 1 2 2 2 1 1 5 10 1 2 16 3 8 16 13 11 7 7 1 7 1 ------- 71 Table 16, cont. Compound Name Hydrocarbon C - 32 Hydrocarbon C - 33 Hydrocarbon C - 34 Hydrocarbon C - 35 Hydroxy-borneol Indole Methyl Acenaphthene Methyl Naphthalene 2-Methyl Thiobenzothiazole Nitro toluene Pentachlorethane Pentachlorobutadiene Pentachlorobutene Phenanthrene Terpene C^ Terpineol C,Q Terpineol C,5 Tetrachlorobutadiene Tetrachloroethane Tetrach1oroethy1ene Tri chloroethylene Tritepene C,5H-. Unidentified Phthalate Xylene Frequency 1 1 1 1 1 1 1 2 1 3 2 3 2 1 1 36 55 3 2 4 1 3 6 1 ------- 72 Table 17 List of Volatile Compounds Found in 204 Water Samples and Their Frequency of Occurrence Compound Name Frequency Acetone 33 Acetophenone 6 Benzaldehyde 1 Benzene 40 Bromobenzene 9 1-Bromo-l-Chloroethane 2 l-Bromo-2-Chloroethane 1 l-Bromo-2-Chloropropane 1 Bromo-Dichloroethane 2 Bromo-dichloromethane 24 Bromo-Trichloropropane 1 Butadiene 1 Butanal 1 Butane 3 Butene 5 C5HgO or CgH120 11 C5H1Q0 or C6H120 2 C8H17 ] Carbon Tetrachloride 6 Chlorobenzene 11 l-Chloro-2-Bromoe.thane 1 l-Chloro-2-Bromopropane 1 Chloroform 178 Chloroprene 1 Cyclohexane 13 Cyclopentane 1 Dibromo-Chloromethane 10 Dibromoethane 2 Dichlorobenzene 23 ------- 73 Table 17, cont. Compound Name 1,2-Dichloroethane Dichloroethylene Di chloro-lodomethane Dichloromethane 1,2-Dichloropropane Die thy 1 Ether Diisopropyl Ether Dimethoxymethane Dimethyl Sulfide Dimethyl Disulfide Dime thy Iformamide Dioxane Dioxolane Ethanethiol Ethyl Acetate Ethyl benzene Ethyl Methyl Dioxolane Fluoro-dichloro-bromomethane Freon Furfural Heptene Hexane Hexanol Hexene Isomers Methacrylonitrile Methyl-t-Butyl Ketone Methyl Ethyl Ketone Methyl-isobutyl Ketone Methyl Methacrylate 2-Methyl Propanal Methylal 4 Methyl-2-Ethyl-l,3-Dioxolane Methyl -tetrahydrofyran Neopentane Frequency 53 19 1 32 8 9 2 10 1 5 1 4 1 1 1 5 5 1 1 1 1 14 1 5 1 1 1 2 1 1 1 2 2 1 ------- 74 Table 17, cont. Compound Name Frequency Nonene 1 Pen tane 18 Pentene Isomer 4 C,g Terpene 2 1,1,2,2-Tetrachloroethane 12 Tetrachloroethylene 77 Tetrahydrofuran 29 Tetrahydropyran 4 Toluene 31 Tribromomethane 5 Trichlorobutane 1 Trichloroethane 11 1,1,1-Trichloroethane 18 Trichloroethylene 88 Trichloropropane 6 Trichloro-fluoromethane 11 Trichloro-trifluoroethane 8 ------- 75 REFERENCES Bellar, T. A., and Lichtenberg, J. J. 1974. The determination of volatile organic compounds at the yg/1 level in water by gas chromatography. USEPA Report No. EPA-670/4-74-009. Bonner, N. A.; Bazan, S.; and Camp, D. C. 1975. Trace element analysis using x-ray fluorescence. Chemical Instrumentation 6:1-36. Chian, E. S. K., and Kuo, P. K. 1975. Fundamental study on the post treatment of RO permeates from army wastewater. First annual report to the U. S. Army Medical R&D Command. Contract No. DAMD 17-75-C-500b. Clemente, G. F., and Mastinu, G. G. 1974. Instrumental method for the determination of trace elements in water samples by neutron activation analysis. J. Radioanal. Chem. 20:707-14. Kopfler, F. C. et al. 1976. GC/MS determination of volatiles for the national organics reconnaissance survey of drinking water. In Identification and Analysis of Organic Pollutants in Water, ed. L. H. Keith. Ann Arbor, Mich.: Ann Arbor Science Publishers. Salbu, B.; Steinnes, E.; and Pappas, A. C. 1975. Multielement neutron activation analysis of fresh water using Ge(Li) spectrometry. Analytical Chemistry 47:1011-16. ------- |